Electronic Devices: and Amplifier Circuits
Electronic Devices: and Amplifier Circuits
Electronic Devices: and Amplifier Circuits
Second Edition
Steven T. Karris
WI T AB H MATL
OM PUT ING
Electronic Devices
and Amplifier Circuits
with MATLAB Computing
Second Edition
Students and working professionals will find Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition, to be a concise and easy-to-learn text. It provides complete, clear, and detailed explanations of the state-of-the-art electronic devices and integrated circuits. All topics are illustrated with many real-world examples.
This text includes the following chapters and appendices: Basic Electronic Concepts and Signals Introduction to Semiconductor Electronics - Diodes Bipolar Junction Transistors Field Effect Transistors and PNPN Devices Operational Amplifiers Integrated Circuits Pulse Circuits and Waveforms Generators Frequency Characteristics of Single-Stage and Cascaded Amplifiers Tuned Amplifiers Sinusoidal Oscillators Introduction to MATLAB Introduction to Simulink PID Controllers Compensated Attenuators The Substitution, Reduction, and Millers Theorems Each chapter contains numerous practical applications supplemented with detailed instructions for using MATLAB to plot the characteristics of non-linear devices and to obtain quick solutions.
Steven T. Karris is the founder and president of Orchard Publications, has undergraduate and graduate degrees in electrical engineering, and is a registered professional engineer in California and Florida. He has more than 35 years of professional engineering experience and more than 30 years of teaching experience as an adjunct professor, most recently at UC Berkeley, California. products and the publication of MATLAB and His area of interest is in The MathWorks, Inc. Simulink based texts.
Second Edition
Steven T. Karris
Electronic Devices and Amplifier Circuits with MATLABComputing, Second Edition Copyright 2008 Orchard Publications. All rights reserved. Printed in the United States of America. No part of this publication may be reproduced or distributed in any form or by any means, or stored in a data base or retrieval system, without the prior written permission of the publisher. Direct all inquiries to Orchard Publications, [email protected] Product and corporate names are trademarks or registered trademarks of The MathWorks, Inc. They are used only for identification and explanation, without intent to infringe.
Disclaimer
The author has made every effort to make this text as complete and accurate as possible, but no warranty is implied. The author and publisher shall have neither liability nor responsibility to any person or entity with respect to any loss or damages arising from the information contained in this text.
Preface
This text is an undergraduate level textbook presenting a thorough discussion of stateofthe art electronic devices. It is selfcontained; it begins with an introduction to solid state semiconductor devices. The prerequisites for this text are first year calculus and physics, and a twosemester course in circuit analysis including the fundamental theorems and the Laplace transformation. No previous knowledge of MATLABis required; the material in Appendix A and the inexpensive MATLAB Student Version is all the reader needs to get going. Our discussions are based on a PC with Windows XP platforms but if you have another platform such as Macintosh, please refer to the appropriate sections of the MATLABs User Guide which also contains instructions for installation. Additional information including purchasing may be obtained from The MathWorks, Inc., 3 Apple Hill Drive, Natick, MA 01760-2098. Phone: 508 6477000, Fax: 508 6477001, e mail: [email protected] and web site http://www.mathworks.com.This text can also be used without MATLAB. This is our fourth electrical and computer engineering-based text with MATLAB applications. My associates, contributors, and I have a mission to produce substance and yet inexpensive texts for the average reader. Our first three texts* are very popular with students and working professionals seeking to enhance their knowledge and prepare for the professional engineering examination. The author and contributors make no claim to originality of content or of treatment, but have taken care to present definitions, statements of physical laws, theorems, and problems. Chapter 1 is an introduction to the nature of small signals used in electronic devices, amplifiers, definitions of decibels, bandwidth, poles and zeros, stability, transfer functions, and Bode plots. Chapter 2 is an introduction to solid state electronics beginning with simple explanations of electron and hole movement. This chapter provides a thorough discussion on the junction diode and its volt-ampere characteristics. In most cases, the non-linear characteristics are plotted with simple MATLAB scripts. The discussion concludes with diode applications, the Zener, Schottky, tunnel, and varactor diodes, and optoelectronics devices. Chapters 3 and 4 are devoted to bipolar junction transistors and FETs respectively, and many examples with detailed solutions are provided. Chapter 5 is a long chapter on op amps. Many op amp circuits are presented and their applications are well illustrated. The highlight of this text is Chapter 6 on integrated devices used in logic circuits. The internal construction and operation of the TTL, NMOS, PMOS, CMOS, ECL, and the biCMOS families
* These are Circuit Analysis I, ISBN 9780970951120, Circuit Analysis II, ISBN 978097095115 1, and Signals and Systems, 9781934404119.
of those devices are fully discussed. Moreover, the interpretation of the most important parameters listed in the manufacturers data sheets are explained in detail. Chapter 7 is an introduction to pulse circuits and waveform generators. There, we discuss the 555 Timer, the astable, monostable, and bistable multivibrators, and the Schmitt trigger. Chapter 8 discusses to the frequency characteristic of single-stage and cascade amplifiers, and Chapter 9 is devoted to tuned amplifiers. Sinusoidal oscillators are introduced in Chapter 10. This is the second edition of this title, and includes several Simulink models. Also, two new appendices have been added. Appendix A, is an introduction to MATLAB. Appendix B is an introduction to Simulink, Appendix C is an introduction to Proportional-Integral-Derivative (PID) controllers, Appendix D describes uncompensated and compensated networks, and Appendix D discusses the substitution, reduction, and Millers theorems. The author wishes to express his gratitude to the staff of The MathWorks, the developers of MATLAB and Simulink for the encouragement and unlimited support they have provided me with during the production of this text. A companion to this text, Digital Circuit Analysis and Design with Simulink Modeling and Introduction to CPLDs and FPGAs, ISBN 9781934404058 is recommended as a companion. This text is devoted strictly on Boolean logic, combinational and sequential circuits as interconnected logic gates and flipflops, an introduction to static and dynamic memory devices. and other related topics. Although every effort was made to correct possible typographical errors and erroneous references to figures and tables, some may have been overlooked. Our experience is that the best proofreader is the reader. Accordingly, the author will appreciate it very much if any such errors are brought to his attention so that corrections can be made for the next edition. We will be grateful to readers who direct these to our attention at [email protected]. Thank you. Orchard Publications Fremont, California 945384741 United States of America www.orchardpublications.com [email protected]
MATLAB Computing Pages 17, 114, 115, 119, 127, 128, 131 Simulink Modeling Pages 134, 135
Introduction to Semiconductor Electronics Diodes 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 2.13 2.14 2.15 2.16 Electrons and Holes ............................................................................................... 21 Junction Diode ...................................................................................................... 24 Graphical Analysis of Circuits with NonLinear Devices ..................................... 29 Piecewise Linear Approximations....................................................................... 213 Low Frequency AC Circuits with Junction Diodes ............................................ 215 Junction Diode Applications in AC Circuits...................................................... 219 Peak Rectifier Circuits ........................................................................................ 230 Clipper Circuits ................................................................................................... 232 DC Restorer Circuits........................................................................................... 235 Voltage Doubler Circuits .................................................................................... 236 Diode Applications in Amplitude Modulation (AM) Detection Circuits ......... 237 Diode Applications in Frequency Modulation (FM) Detection Circuits........... 237 Zener Diodes ....................................................................................................... 238 Schottky Diode ................................................................................................... 245 Tunnel Diode ...................................................................................................... 245 Varactor .............................................................................................................. 248 i
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
MATLAB Computing Pages 227, 228, 229, 261, 270 Simulink Modeling Pages 219, 222, 225, 230
Bipolar Junction Transistors 3.1 3.2 3.3 Introduction ....................................................................................................... 31 NPN Transistor Operation ................................................................................ 33 Bipolar Junction Transistor as an Amplifier ...................................................... 33 3.3.1 Equivalent Circuit Models NPN Transistors........................................ 36 3.3.2 Equivalent Circuit Models PNP Transistors ........................................ 37 3.3.3 Effect of Temperature on the i C v BE Characteristics...................... 310 3.3.4 Collector Output Resistance Early Voltage ....................................... 311 Transistor Amplifier Circuit Biasing ................................................................ 318 Fixed Bias ......................................................................................................... 321 SelfBias ........................................................................................................... 325 Amplifier Classes and Operation ..................................................................... 328 3.7.1 Class A Amplifier Operation................................................................. 331 3.7.2 Class B Amplifier Operation ................................................................. 334 3.7.3 Class AB Amplifier Operation .............................................................. 335 3.7.4 Class C Amplifier Operation................................................................. 337 Graphical Analysis ........................................................................................... 338 Power Relations in the Basic Transistor Amplifier .......................................... 342 PiecewiseLinear Analysis of the Transistor Amplifier ................................... 344 Incremental Linear Models .............................................................................. 349 Transconductance............................................................................................ 354 HighFrequency Models for Transistors.......................................................... 355 The Darlington Connection ............................................................................ 360 Transistor Networks......................................................................................... 361 3.15.1 hEquivalent Circuit for the CommonBase Transistor..................... 361 3.15.2 TEquivalent Circuit for the CommonBase Transistor .................... 364 3.15.3 hEquivalent Circuit for the CommonEmitter Transistor ................ 365 3.15.4 TEquivalent Circuit for the CommonEmitter Transistor................ 370 3.15.5 hEquivalent Circuit for the CommonCollector Transistor ............. 371 3.15.6 TEquivalent Circuit for the CommonCollector Transistor............. 376
ii
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
3.16 Transistor Cutoff and Saturation Regions..................................................... 377 3.16.1 Cutoff Region ....................................................................................378 3.16.2 Active Region....................................................................................378 3.16.3 Saturation Region .............................................................................378 3.17 The EbersMoll Transistor Model................................................................. 381 3.18 Schottky Diode Clamp...................................................................................385 3.19 Transistor Specifications................................................................................385 3.20 Summary ........................................................................................................387 3.21 Exercises.........................................................................................................391 3.22 Solutions to EndofChapter Exercises ........................................................397 MATLAB Computing Pages 313, 339, 3113
Field Effect Transistors and PNPN Devices 4.1 Junction Field Effect Transistor (JFET) ............................................................. 41 4.2 Metal Oxide Semiconductor Field Effect Transistor (MOSFET) ..................... 46 4.2.1 NChannel MOSFET in the Enhancement Mode................................... 48 4.2.2 NChannel MOSFET in the Depletion Mode ...................................... 412 4.2.3 PChannel MOSFET in the Enhancement Mode ................................. 414 4.2.4 PChannel MOSFET in the Depletion Mode ....................................... 417 4.2.5 Voltage Gain ......................................................................................... 417 4.3 Complementary MOS (CMOS)....................................................................... 419 4.3.1 CMOS CommonSource Amplifier ..................................................... 420 4.3.2 CMOS CommonGate Amplifier......................................................... 420 4.3.3 CMOS CommonDrain (Source Follower) Amplifier......................... 420 4.4 Metal Semiconductor FET (MESFET)............................................................ 421 4.5 Unijunction Transistor ..................................................................................... 422 4.6 Diac.................................................................................................................. 423 4.7 Silicon Controlled Rectifier (SCR).................................................................. 424 4.7.1 SCR as an Electronic Switch ................................................................ 427 4.7.2 SCR in the Generation of Sawtooth Waveforms .................................. 428 4.8 Triac ............................................................................................................... 437 4.9 Shockley Diode.............................................................................................. 438 4.10 Other PNPN Devices...................................................................................... 440 4.11 The Future of Transistors ............................................................................... 441 4.12 Summary ........................................................................................................ 442 4.13 Exercises ........................................................................................................ 445 4.14 Solutions to EndofChapter Exercises ........................................................ 447 MATLAB Computing Pages 45, 410
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
iii
Operational Amplifiers 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 5.10 5.11 5.12 5.13 5.14 5.15 5.16 5.17 5.18 5.19 5.20 5.21 5.22 5.23 5.24 Operational Amplifier.......................................................................................51 An Overview of the Op Amp ...........................................................................51 Op Amp in the Inverting Mode........................................................................52 Op Amp in the NonInverting Mode ..............................................................55 Active Filters.....................................................................................................58 Analysis of Op Amp Circuits ..........................................................................511 Input and Output Resistances ........................................................................522 Op Amp Open Loop Gain ..............................................................................525 Op Amp Closed Loop Gain ............................................................................526 Transresistance Amplifier ...............................................................................529 Closed Loop Transfer Function ......................................................................530 Op Amp Integrator .........................................................................................531 Op Amp Differentiator ...................................................................................535 Summing and Averaging Op Amp Circuits ....................................................537 Differential Input Op Amp..............................................................................539 Instrumentation Amplifiers .............................................................................542 Offset Nulling ..................................................................................................544 External Frequency Compensation .................................................................545 Slew Rate .........................................................................................................545 Circuits with Op Amps and Non-Linear Devices ...........................................546 Comparators ....................................................................................................550 Wien Bridge Oscillator....................................................................................550 DigitaltoAnalog Converters ........................................................................552 AnalogtoDigital Converters ........................................................................556 5.24.1 Flash AnalogtoDigital Converter ....................................................557 5.24.2 Successive Approximation AnalogtoDigital Converter..................558 5.24.3 DualSlope AnalogtoDigital Converter..........................................559 Quantization, Quantization Error, Accuracy, and Resolution.......................561 Op Amps in Analog Computers .....................................................................563 Summary .........................................................................................................567 Exercises..........................................................................................................571 Solutions to EndofChapter Exercises .........................................................578
6
iv
Integrated Circuits 6.1 Basic Logic Gates.............................................................................................. 61 6.2 Positive and Negative Logic ..............................................................................61
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12
6.13
Inverter ............................................................................................................. 62 AND Gate......................................................................................................... 66 OR Gate............................................................................................................ 68 NAND Gate ..................................................................................................... 69 NOR Gate....................................................................................................... 614 Exclusive OR (XOR) and Exclusive NOR (XNOR) Gates ........................... 615 Fan-In, Fan-Out, TTL Unit Load, Sourcing Current, and Sinking Current . 617 Data Sheets ..................................................................................................... 620 Emitter Coupled Logic (ECL)......................................................................... 624 NMOS Logic Gates......................................................................................... 628 6.12.1 NMOS Inverter................................................................................... 631 6.12.2 NMOS NAND Gate ........................................................................... 631 6.12.3 NMOS NOR Gate .............................................................................. 632 CMOS Logic Gates ........................................................................................ 632 6.13.1 CMOS Inverter ................................................................................... 633 6.13.2 CMOS NAND Gate ........................................................................... 634 6.13.3 The CMOS NOR Gate ....................................................................... 635 Buffers, Tri-State Devices, and Data Buses ................................................... 635 Present and Future Technologies................................................................... 639 Summary......................................................................................................... 643 Exercises ......................................................................................................... 646 Solutions to EndofChapter Exercises ......................................................... 649
Pulse Circuits and Waveform Generators 7.1 7.2 7.3 7.4 7.5 Astable (Free-Running) Multivibrators ........................................................... 71 555 Timer ......................................................................................................... 72 Astable Multivibrator with 555 Timer ............................................................. 73 Monostable Multivibrators ............................................................................. 714 Bistable Multivibrators (FlipFlops)............................................................... 719 7.5.1 FixedBias Flip-Flop............................................................................ 719 7.5.2 SelfBias FlipFlop .............................................................................. 722 7.5.3 Triggering Signals for FlipFlops ......................................................... 728 7.5.4 Present Technology Bistable Multivibrators ....................................... 730 The Schmitt Trigger ....................................................................................... 730 Summary......................................................................................................... 733 Exercises ......................................................................................................... 734 Solutions to EndofChapter Exercises ......................................................... 737
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Frequency Characteristics of SingleStage and Cascaded Amplifiers 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10 Properties of Signal Waveforms........................................................................81 The Transistor Amplifier at Low Frequencies..................................................85 The Transistor Amplifier at High Frequencies.................................................89 Combined Low- and HighFrequency Characteristics...................................814 Frequency Characteristics of Cascaded Amplifiers ........................................815 Overall Characteristics of Multistage Amplifiers ...........................................827 Amplification and Power Gain in Three or More Cascaded Amplifiers ........832 Summary .........................................................................................................834 Exercises..........................................................................................................836 Solutions to EndofChapter Exercises..........................................................839
Tuned Amplifiers 9.1 Introduction to Tuned Circuits.......................................................................... 91 9.2 Single-tuned Transistor Amplifier...................................................................... 98 9.3 Cascaded Tuned Amplifiers ............................................................................. 914 9.3.1 Synchronously Tuned Amplifiers ........................................................... 914 9.3.2 StaggerTuned Amplifiers...................................................................... 918 9.3.3 Three or More Tuned Amplifiers Connected in Cascade...................... 927 9.4 Summary........................................................................................................... 928 9.5 Exercises ........................................................................................................... 930 9.6 Solutions to EndofChapter Exercises ........................................................... 931 MATLAB Computing Page 918
10
Sinusoidal Oscillators 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 Introduction to Oscillators ...........................................................................101 Sinusoidal Oscillators ...................................................................................101 RC Oscillator................................................................................................104 LC Oscillators...............................................................................................105 The Armstrong Oscillator ............................................................................106 The Hartley Oscillator .................................................................................107 The Colpitts Oscillator.................................................................................107 Crystal Oscillators ........................................................................................108 The Pierce Oscillator..................................................................................1010
vi
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
10.10 Summary .................................................................................................... 1012 10.11 Exercises ..................................................................................................... 1014 10.12 Solutions to EndofChapter Exercises..................................................... 1015
Introduction to MATLAB A.1 A.2 A.3 A.4 A.5 A.6 A.7 A.8 A.9 A.10 A.11 MATLAB and Simulink......................................................................... A1 Command Window....................................................................................... A1 Roots of Polynomials..................................................................................... A3 Polynomial Construction from Known Roots............................................... A4 Evaluation of a Polynomial at Specified Values............................................ A6 Rational Polynomials..................................................................................... A8 Using MATLAB to Make Plots .................................................................. A10 Subplots....................................................................................................... A18 Multiplication, Division and Exponentiation ............................................. A18 Script and Function Files ............................................................................ A26 Display Formats........................................................................................... A31
MATLAB Computing Pages A3 through A8, A10, A13, A14, A16, A17, A21, A22, A24, A27
Introduction to Simulink B.1 Simulink and its Relation to MATLAB..............................................................B1 B.2 Simulink Demos................................................................................................B20 MATLAB Computing Page B4 Simulink Modeling Pages B7, B12, B14, B18
ProportionalIntegralDerivative (PID) Controller C.1 Description and Components of a Typical PID................................................. C1 C.2 The Simulink PID Blocks .................................................................................. C2 Simulink Modeling Pages C2, C3
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
vii
Substitution, Reduction, and Millers Theorems E.1 The Substitution Theorem ..................................................................................E1 E.2 The Reduction Theorem .....................................................................................E6 E.3 Millers Theorem ...............................................................................................E10
References Index
R1 IN1
viii
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Chapter 1
Basic Electronic Concepts and Signals
lectronics may be defined as the science and technology of electronic devices and systems. Electronic devices are primarily nonlinear devices such as diodes and transistors and in general integrated circuits (ICs) in which small signals (voltages and currents) are applied to them. Of course, electronic systems may include resistors, capacitors and inductors as well. Because resistors, capacitors and inductors existed long ago before the advent of semiconductor diodes and transistors, these devices are thought of as electrical devices and the systems that consist of these devices are generally said to be electrical rather than electronic systems. As we know, with todays technology, ICs are becoming smaller and smaller and thus the modern IC technology is referred to as microelectronics.
We will now define the average value of a waveform. Consider the waveform shown in Figure 1.2. The average value of f ( t ) in the interval a t b is
f ( t ) dt a Area = ---------------- = ----------------------Period ba a b
f ( t ) ave
(1.1)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
11
t a Period b
(1.2)
for all time t and for all integers n . The constant T is the period and it is the smallest value of time which separates recurring values of the waveform. An alternating waveform is any periodic time function whose average value over a period is zero. Of course, all sinusoids are alternating waveforms. Others are shown in Figure 1.3.
t T T
t T
The effective (or RMS) value of a periodic current waveform i ( t ) denoted as I eff is the current that produces heat in a given resistor R at the same average rate as a direct (constant) current I dc , that is,
Average Power = P ave = RI eff = RI dc
2 2
(1.3)
(1.4)
and
1 P ave = -T
T 0
1 p ( t ) dt = -T
Ri
dt
R = --T
dt
(1.5)
12
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Amplifiers
Equating (1.3) with (1.5) we obtain or or
I RMS = I eff =
2 R T 2 RI eff = --- i dt T 0
2 1 T 2 I eff = -- i dt T 0
(1.6)
Ave ( i )
2
1 T 2 -- i dt = T 0
(1.7)
where RMS stands for Root Mean Squared, that is, the effective value I eff or I RMS value of a current is computed as the square root of the mean (average) of the square of the current. Warning 1: In general, Ave ( i ) ( i ave ) . Ave ( i ) implies that the current i must first be squared and the average of the squared value is to be computed. On the other hand, ( i ave ) implies that the average value of the current must first be found and then the average must be squared. Warning 2: In general, P ave V ave I ave . If v ( t ) = V p cos t and i ( t ) = I p cos ( t + ) for example, V ave = 0 and I ave = 0 , it follows that P ave = 0 also. However,
1 T 1 T P ave = -- p dt = -- vi dt 0 T 0 T 0
2 2 2 2
In introductory electrical engineering books it is shown* that if the peak (maximum) value of a current of a sinusoidal waveform is I p , then
I RMS = I p 2 = 0.707I p
(1.8)
1.2 Amplifiers
An amplifier is an electronic circuit which increases the magnitude of the input signal. The symbol of a typical amplifier is a triangle as shown in Figure 1.4.
v in
v out
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
13
or
G v = V out V in
1.3 Decibels
The ratio of any two values of the same quantity (power, voltage or current) can be expressed in decibels (dB). For instance, we say that an amplifier has 10 dB power gain, or a transmission line has a power loss of 7 dB (or gain 7 dB ). If the gain (or loss) is 0 dB , the output is equal to the input. We should remember that a negative voltage or current gain G v or G i indicates that there is a 180 phase difference between the input and the output waveforms. For instance, if an op amp has a gain of 100 (dimensionless number), it means that the output is 180 outof phase with the input. For this reason we use absolute values of power, voltage and current when these are expressed in dB terms to avoid misinterpretation of gain or loss. By definition, P out dB = 10 log --------P in Therefore,
10 dB represents a power ratio of 10 10n dB represents a power ratio of 10
n
(1.9)
14
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
From these, we can estimate other values. For instance, 4 dB = 3 dB + 1 dB which is equivalent to a power ratio of approximately 2 1.25 = 2.5 . Likewise, 27 dB = 20 dB + 7 dB and this is equivalent to a power ratio of approximately 100 5 = 500 . Since y = log x = 2 log x and P = V Z = I Z , if we let Z = 1 the dB values for voltage and current ratios become V out dB v = 10 log --------V in and I out dB i = 10 log ------I in
2 2 2 2
(1.10)
(1.11)
2 2 = 0.707 and these two points are known as the 3dB 2 2 = 0.707 the power is 1 2 , that is, it is
down or half power points. They derive their name from the fact that since power
p = v R = i R , for R = 1 and for v or i =
halved.
1 0.707 Bandwidth 1 Figure 1.5. Definition of bandwidth 2 V out
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
15
(1.12)
where V out ( ) is known as the magnitude response and e j ( ) is known as the phase response. These two responses together constitute the frequency response of a circuit. Example 1.1 Derive and sketch the magnitude and phase responses of the RC lowpass filter shown in Figure 1.6.
v in
R C
v out
Solution:
(1.13)
or
(1.14)
(1.15)
and the phase angle (sometimes called argument and abbreviated as arg) is
16
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(1.16)
To sketch the magnitude, we let assume the values 0 , 1 RC , and . Then, as 0 , V out V in 1 for = 1 RC , V out V in = 1 ( 2 ) = 0.707 and as , V out V in 0 To sketch the phase response, we use (1.16). Then, as , tan ( ) 90 as 0 , tan 0 0 for = 1 RC , tan 1 45 as , tan ( ) 90
1 1 1 1
The magnitude and phase responses of the RC lowpass filter are shown in Figure 1.7.
Vout Vin 1 0.707 1/RC 0 45 1/RC 90 90 45 1/RC
Figure 1.7. Magnitude and phase responses for the lowpass filter of Figure 1.6
We can use MATLAB* to plot the magnitude and phase angle for this low-pass filter using relation (1.13) expressed in MATLAB script below and plot the magnitude and phase angle in semilog scale.
w=1:10:100000; RC=1; xf=1./(1+j.*w.*RC);... mag=20.*log10(abs(xf)); phase=angle(xf).*180./pi;... subplot(121); semilogx(w,mag); subplot(122); semilogx(w,phase) * For an introduction to MATLAB, please refer to Appendix A
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
17
10
10
(1.17)
then two common frequency intervals are (1) the octave for which 2 = 2 1 and (2) the decade for which 2 = 10 1 . Now, let us consider a circuit whose gain is given as
G ( )v = C
k
(1.18)
where C is a constant and k is a nonzero positive integer. Taking the common log of (1.18) and multiplying by 20 we obtain
20 log
10
{ G ( )v }
(1.19)
We observe that (1.19) represents an equation of a straight line with abscissa log 10 , slope of
20k , and { G ( ) v } intercept at 20 log 1010 C = cons tan t . We can choose the slope to be either 20k dB decade or 6k dB octave . Thus, if k = 1 , the slope becomes 20 dB decade as
18
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Transfer Function
G ( )V 0 dB
20
slope = 20 dB/decade
log10
Then, any line parallel to this slope will represent a drop of 20 dB decade . We observe also that if the exponent k in (1.18) is changed to 2 , the slope will be 40 dB decade . We can now approximate the magnitude and phase responses of the lowpass filter of Example 1.1 with asymptotic lines as shown in Figure 1.9.
dB 0 10 45 20 30 0.1 1 10 log 100 90 20 dB/decade () 0.1
10
100
log
Figure 1.9. Magnitude and phase responses for the lowpass filter of Figure 1.6.
* A continuoustime signal is a function that is defined over a continuous range of time. A linear system consists of linear devices and may include independent and dependent voltage and current sources. For details, refer to Circuit Analysis I with MATLAB Applications, ISBN 9780970951120. A timeinvariant system is a linear system in which the parameters do not vary with time.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
19
Figure 1.10. Inputoutput block diagram for linear, timeinvariant continuoustime system
We will assume that initially no energy is stored in the system. The inputoutput relationship can be described by the differential equation of
d d d b m -------- v out ( t ) + b m 1 --------------- v out ( t ) + b m 2 --------------- v out ( t ) + + b 0 v out ( t ) = m m1 m2 dt dt dt d d d a n ------ v in ( t ) + a n 1 -------------- v in ( t ) + a n 2 -------------- v in ( t ) + + a 0 v in ( t ) n n1 n2 dt dt dt
n n1 n2 m m1 m2
(1.20)
For practically all electric networks, m n and the integer m denotes the order of the system. Taking the Laplace transform* of both sides of (1.20) we obtain
( bm s + bm 1 s
n m m1
+ bm 2 s
m2 n2
+ + b 0 )V out ( s ) = + + a 0 )V in ( s )
( an s + an 1 s
n1
+ an 2s
where N ( s ) and D ( s ) are the numerator and denominator polynomials respectively. The transfer function G ( s ) is defined as
V out ( s ) N(s) G ( s ) = ------------------ = ----------D(s) V in ( s )
(1.21)
Example 1.2 Derive the transfer function G ( s ) of the network in Figure 1.11.
* The Laplace transform and its applications to electric circuit analysis is discussed in detail in Circuit Analysis II, ISBN 9780970951151.
110
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
+
v in ( t )
0.5 H C
1F
+
R 1 v out ( t )
Solution: The given circuit is in the t domain .* The transfer function G ( s ) exists only in the s domain and thus we redraw the circuit in the s domain as shown in Figure 1.12. +
V in ( s ) 0.5s
+
1s 1 V out ( s )
For relatively simple circuits such as that of Figure 1.12, we can readily obtain the transfer function with application of the voltage division expression. Thus, parallel combination of the capacitor and resistor yields
1 1s1 ----------------- = ---------s+1 1s+1
or
(1.22)
* For brevity, we will denote the time domain as t domain Henceforth, the complex frequency, i.e., s = + j , will be referred to as the s domain .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
111
(1.23)
The coefficients a k and b k for k = 0, 1, 2, , n are real numbers and, for the present discussion, we have assumed that the highest power of N ( s ) is less than the highest power of D ( s ) , i.e., m < n . In this case, F ( s ) is a proper rational function. If m n , F ( s ) is an improper rational function. It is very convenient to make the coefficient a n of s in (12.2) unity; to do this, we rewrite it as
m m1 m2 1---- ( b m s + b m 1 s + bm 2 s + + b1 s + b0 ) an N(s) F ( s ) = ----------- = --------------------------------------------------------------------------------------------------------------------------------D(s) a1 a0 n an 1 n 1 an 2 n 2 s + ----------- s + ----------- s + + ---- s + ---an an an an n
(1.24)
The roots of the numerator are called the zeros of F ( s ) , and are found by letting N ( s ) = 0 in (1.24). The roots of the denominator are called the poles* of F ( s ) and are found by letting D ( s ) = 0 . However, in most engineering applications we are interested in the nature of the poles.
1.8 Stability
In general, a system is said to be stable if a finite input produces a finite output. We can predict the stability of a system from its impulse response h ( t ) . In terms of the impulse response, 1. A system is stable if the impulse response h ( t ) goes to zero after some time as shown in Figure 1.13. 2. A system is marginally stable if the impulse response h ( t ) reaches a certain nonzero value but never goes to zero as shown in Figure 1.14. 3. A system is unstable if the impulse response h ( t ) reaches infinity after a certain time as shown in Figure 1.15.
* The zeros and poles can be distinct (different from one another), complex conjugates, repeated, of a combination of these. For details refer to Circuit Analysis II with MATLAB Applications, ISBN 978097095115 1. For a detailed discussion on the impulse response, please refer to Signals and Systems with MATLAB Computing and Simulink Modeling, Fourth Edition, ISBN 9781934404119.
112
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Stability
Voltage
Time
We can plot the poles and zeros of a transfer function G ( s ) on the complex frequency plane of the complex variable s = + j . A system is stable only when all poles lie on the lefthand half plane. It is marginally stable when one or more poles lie on the j axis, and unstable when one or more poles lie on the righthand halfplane. However, the location of the zeros in the s plane is Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
113
The bode(sys,w) function displays both magnitude and phase. If we want to display the magnitude only, we can use the bodemag(sys,w) function. MATLAB requires that we express the numerator and denominator of G ( s ) as polynomials of s in descending powers. Example 1.3 The transfer function of a system is
3 ( s 1 ) ( s + 2s + 5 ) G ( s ) = -------------------------------------------------2 ( s + 2 ) ( s + 6s + 25 )
2
a. is this system stable? b. use the MATLAB bode(sys,w) function to plot the magnitude of this transfer function. Solution: a. Let us use the MATLAB solve('eqn1','eqn2',...,'eqnN') function to find the roots of the quadratic factors.
syms s; equ1=solve('s^2+2*s+50'), equ2=solve('s^2+6*s+250')
[-1-2*i] [-3-4*i]
114
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Stability
3 + j4 1 + j2 2 1 j 2 3 j 4 1
Figure 1.16. Poles and zeros of the transfer function of Example 1.3
From Figure 1.16 we observe that all poles, denoted as , lie on the lefthand halfplane and thus the system is stable. The location of the zeros, denoted as , is immaterial. b. We use the MATLAB expand(s) symbolic function to express the numerator and denominator of G ( s ) in polynomial form
syms s; n=expand((s1)*(s^2+2*s+5)), d=expand((s+2)*(s^2+6*s+25))
For this example we are interested in the magnitude only so we will use the script
num=3*[1 1 3 5]; den=[1 8 37 50]; sys=tf(num,den);... w=logspace(0,2,100); bodemag(sys,w); grid
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
115
5 Magnitude (dB)
-5
-10 0 10
10
10
Example 1.4 It is known that a voltage amplifier has a frequency response of a lowpass filter, a DC gain of 80 dB , attenuation of 20 dB per decade, and the 3 dB cutoff frequency occurs at 10 KHz . Determine the gain (in dB ) at the frequencies 1 KHz , 10 KHz , 100KHz , 1 MHz , 10 MHz , and 100 MHz . Solution: Using the given data we construct the asymptotic magnitude response shown in Figure 1.18 from which we obtain the data shown on the table below.
80 60 40 20 0 dB 20 dB decade
Hz
4
10
10
10
10
10
10
116
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A voc v in
v out
A voc = ---------
v out v in
i out = 0
Figure 1.19. Circuit model for voltage amplifier where A voc denotes the open circuit voltage gain
The ideal characteristics for the circuit of Figure 1.19 are R in and R out 0 . Example 1.5 For the voltage amplifier of Figure 1.20, find the overall voltage gain A v = v load v s . Then, use MATLAB to plot the magnitude of A v for the range 10 10 . From the plot, estimate the
3 dB cutoff frequency.
Rs vs 1 K R out
3 8
2 cos t mV
v in
R in C
100
10 K
+ A voc v in
R load 5 K
v load
0.1 nF A voc = 20
* Readers who have a copy of Circuit Analysis I, ISBN 9780970951120, are encouraged to review Chapter 4 on equivalent circuits of operational amplifiers. Refer also to Chapter 5, this text.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
117
10
4
V in ( s )
10
10
10
s 20V in ( s )
5 10
V load ( s )
10
(1.25)
Also,
10 5 10 V load ( s ) = ------------------------------- 20V in ( s ) = --------------------- V in ( s ) = 19.61V in ( s ) 2 3 3 5.1 10 10 + 5 10
3
(1.26)
(1.27)
The plot is shown in Figure 1.22 and we observe that the cutoff frequency occurs at about 22 dB where
f C 10 2 1.59 MHz
7
118
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
5 3 10
10
10
10
10
10
Figure 1.22. Bode plot for the voltage amplifier of Example 1.5
A isc i in
Figure 1.23. Circuit model for current amplifier where A isc denotes the short circuit current gain
The ideal characteristics for the circuit of Figure 1.23 are R in 0 and R out .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
119
i out i load
R out V load
is
A isc i in
R load
(1.28)
Also,
(1.29)
(1.30)
or
In Sections 1.9 and 1.10 we presented the voltage and current amplifier equivalent circuits also known as circuit models. Two more circuit models are the transresistance and transconductance equivalent circuits and there are introduced in Exercises 1.4 and 1.5 respectively.
120
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
f ( t ) ave
for all time t and for all integers n . The constant T is the period and it is the smallest value of time which separates recurring values of the waveform. An alternating waveform is any periodic time function whose average value over a period is zero. The effective (or RMS) value of a periodic current waveform i ( t ) denoted as I eff is the current that produces heat in a given resistor R at the same average rate as a direct (constant) current I dc and it is found from the expression
I RMS = I eff = 1 T 2 -- i dt = T 0
Ave ( i )
where RMS stands for Root Mean Squared, that is, the effective value I eff or I RMS value of a current is computed as the square root of the mean (average) of the square of the current. If the peak (maximum) value of a current of a sinusoidal waveform is I p , then
I RMS = I p ( 2 ) = 0.707I p
An amplifier is an electronic circuit which increases the magnitude of the input signal. An electronic (or electric) circuit which produces an output that is smaller than the input is called an attenuator. A resistive voltage divider is a typical attenuator. An amplifier can be classified as a voltage, current or power amplifier. The gain of an amplifier is the ratio of the output to the input. Thus, for a voltage amplifier
Voltage Gain = Output Voltage ---------------------------------------Input Voltage
or Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
121
The current gain G i and power gain G p are defined similarly. The ratio of any two values of the same quantity (power, voltage or current) can be expressed in decibels (dB). By definition, dB = 10 log P out P in The dB values for voltage and current ratios are dB v = 20 log V out V in dB i = 20 log I out I in The bandwidth is BW = 2 1 where 1 and 2 are the cutoff frequencies. At these frequencies, V out = power points. The lowpass and highpass filters have only one cutoff frequency whereas bandpass and bandstop filters have two. We may think that lowpass and highpass filters have also two cutoff frequencies where in the case of the lowpass filter the second cutoff frequency is at = 0 while in a highpass filter it is at = . We also recall also that the output of circuit is dependent upon the frequency when the input is a sinusoidal voltage. In general form, the output voltage is expressed as
V out ( ) = V out ( ) e
j ( )
2 2 = 0.707 and these two points are known as the 3dB down or half
where V out ( ) is referred to as the magnitude response and e j ( ) is referred to as the phase response. These two responses together constitute the frequency response of a circuit. The magnitude and phase responses of a circuit are often shown with asymptotic lines as approximations and these are referred to as Bode plots. Two frequencies 1 and 2 are said to be separated by an octave if 2 = 2 1 and separated by a decade if 2 = 10 1 . The transfer function of a system is defined as
V out ( s ) N(s) G ( s ) = ------------------ = ----------D(s) V in ( s )
122
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Summary
where the numerator N ( s ) and denominator D ( s ) are as shown in the expression
( an s + an 1 s + an 2 s + + a0 ) N(s) V out ( s ) = ---------------------------------------------------------------------------------------------------------------- V in ( s ) = ----------- V in ( s ) m m1 m2 D(s) ( bm s + bm 1 s + bm 2 s + + b0 )
n n1 n2
In the expression
m m1 m2 1 ---- ( b s + b m 1 s + bm 2 s + + b1 s + b0 ) an m N(s) F ( s ) = ----------- = --------------------------------------------------------------------------------------------------------------------------------D(s) a1 a0 n an 1 n 1 an 2 n 2 s + ----------- s + ----------- s + + ---- s + ---an an an an
where m < n , the roots of the numerator are called the zeros of F ( s ) , and are found by letting N ( s ) = 0 . The roots of the denominator are called the poles of F ( s ) and are found by letting D(s) = 0 . The zeros and poles can be real and distinct, or repeated, or complex conjugates, or combinations of real and complex conjugates. However, in most engineering applications we are interested in the nature of the poles. A system is said to be stable if a finite input produces a finite output. We can predict the stability of a system from its impulse response h ( t ) . Stability can easily be determined from the transfer function G ( s ) on the complex frequency plane of the complex variable s = + j . A system is stable only when all poles lie on the left hand halfplane. It is marginally stable when one or more poles lie on the j axis, and unstable when one or more poles lie on the righthand halfplane. However, the location of the zeros in the s plane is immaterial. We can use the MATLAB function bode(sys) to draw the Bode plot of a system where sys = tf(num,den) creates a continuoustime transfer function sys with numerator num and denominator den, and tf creates a transfer function. With this function, the frequency range and number of points are chosen automatically. The function bode(sys,{wmin,wmax}) draws the Bode plot for frequencies between wmin and wmax (in radians/second) and the function bode(sys,w) uses the usersupplied vector w of frequencies, in radians/second, at which the Bode response is to be evaluated. To generate logarithmically spaced frequency vectors, we use the command logspace(first_exponent,last_exponent, number_of_values). The bode(sys,w) function displays both magnitude and phase. If we want to display the magnitude only, we can use the bodemag(sys,w) function. Amplifiers are often represented by equivalent circuits also known as circuit models. The common types are the voltage amplifier, the current amplifier, the transresistance amplifier, and the transconductance amplifier. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
123
+
v in ( t )
+
C
1F R
+
1 v out ( t )
3. A system has poles at 4 , 2 + j , 2 j , and zeros at 1 , 3 + j2 , and 3 j2 . Derive the transfer function of this system given that G ( ) = 10 . 4. The circuit model shown below is known as a transresistance amplifier and the ideal characteristics for this amplifier are R in 0 and R out 0 .
i in v in
R in R out
i out v out
out R in = ---------
R in i in
i in
i out = 0
With a voltage source v s in series with resistance R s connected on the input side and a load resistance R load connected to the output, the circuit is as shown below.
RS vS
1 K
i in
R out R in C
10 K 0.1 nF 100
2 cos t mV
v in
+
Rm im
R load
5 K
+
v load
Find the overall voltage gain A v = v load v s if R m = 100 . Then, use MATLAB to plot the magnitude of A v for the range 10 10 . From the plot, estimate the 3 dB cutoff frequency.
3 8
124
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
5. The circuit model shown below is known as a transconductance amplifier and the ideal characteristics for this amplifier are R in and R out .
i out v in
R in R out
v out
G m = -------
i out v in
v out = 0
G m v in
With a voltage source v s in series with resistance R s connected on the input side and a load resistance R load connected to the output, the circuit is as shown below.
i out vS
RS
v in
R in
G m v in
R out
R load v load
Derive an expression for the overall voltage gain A v = v load v s 6. The circuit shown below is an R-C high-pass filter.
C R
a. Derive the transfer function b. Use MATLAB to plot the magnitude and phase angle for this high-pass filter. Hint: Use the procedure in Example 1. 7. For the low-pass filter circuit in Example 1 show that the output across the capacitor at high frequencies, i.e., for 1 RC , approximates an integrator, that is, the voltage across the capacitor is the integral of the input voltage. 8. For the high-pass filter circuit in Exercise 6 above 1 show that the output across the resistor at low frequencies, i.e., for 1 RC , approximates a differentiator, that is, the voltage across the resistor is the derivative of the input voltage.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
125
Chapter 1 Basic Electronic Concepts and Signals 1.13 Solutions to EndofChapter Exercises
Dear Reader: The remaining pages on this chapter contain solutions to all endofchapter exercises. You must, for your benefit, make an honest effort to solve these exercises without first looking at the solutions that follow. It is recommended that first you go through and solve those you feel that you know. For your solutions that you are uncertain, look over your procedures for inconsistencies and computational errors, review the chapter, and try again. Refer to the solutions as a last resort and rework those problems at a later date. You should follow this practice with all endofchapter exercises in this book.
126
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
or
2 2 2 V out j RC - j RC + R C - RC ( j + RC ) G ( j ) = ---------- = ----------------------- = ---------------------------------------- = -------------------------------------2 2 2 2 2 2 V in 1 + j RC 1+ R C 1+ R C 2 2 2 1 ) = RC 1 + R C atan ( 1 ( RC ) - = -------------------------------------------- atan ( 1 ( RC ) ) ------------------------------------------------------------------------------------------2 2 2 2 2 2 1+ R C 1 + 1 ( R C )
(1)
We can obtain a quick sketch for the magnitude G ( j ) versus by evaluating (2) at = 0 , = 1 RC , and . Thus, As 0 ,
G(j) 0
For = 1 RC , and as ,
G ( j ) = 1 2 = 0.707 G(j) 1
We will use the MATLAB script below to plot G ( j ) versus radian frequency . This is shown on the plot below where, for convenience, we let RC = 1 .
w=0:0.02:100; RC=1; magGs=1./sqrt(1+1./(w.*RC).^2); semilogx(w,magGs); grid
We can also obtain a quick sketch for the phase angle, i.e., = arg { G ( j ) } versus , by evaluating (3) at = 0 , = 1 RC , = 1 RC , , and . Thus, as 0 ,
atan 0 0
For = 1 RC , For = 1 RC ,
= atan 1 = 45 = atan ( 1 ) = 45
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
127
Half-power point
10
As ,
= atan ( ) = 90
and as ,
= atan ( ) = 90
We will use the MATLAB script below to plot the phase angle versus radian frequency . This is shown on the plot below where, for convenience, we let RC = 1 .
w=8:0.02:8; RC=1; argGs=atan(1./(w.*RC)).*180./pi; plot(w,argGs); grid
100
50
-50
-100 -8
-6
-4
128
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
+
V out ( s )
1s
To determine the value of the constant K we divide all terms of G ( s ) by s3 and we obtain
) G ( s ) = K ( 1 + 7 s + 19 s + 13 s --------------------------------------------------------------------2 3 1 + 8 s + 21 s + 20 s
2 3
and as s , G ( s ) K . It is given that G ( ) = 10 , then K = 10 and the final form of the transfer function is
10 ( s + 7s + 19s + 13 ) 10 ( s + 1 ) ( s + 6s + 13 ) G ( s ) = --------------------------------------------------------- = -------------------------------------------------------3 2 2 s + 8s + 21s + 20 ( s + 4 ) ( s + 4s + 5 ) num=10*[1 7 19 13]; den=[1 8 21 20]; w=logspace(0,2,100); bode(num,den,w);grid
3 2 2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
129
10
10
Frequency (rad/sec)
10
VS ( s )
10
+
10
4
+ +
10
10
V in ( s )
s 100I in ( s )
V load ( s )
5 10
3
10
Also,
5 10 5 10 V load ( s ) = ------------------------------- 100I in ( s ) = --------------------- I in ( s ) = 98I in ( s ) (1) 2 3 3 10 + 5 10 5.1 10
3 5
where
130
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Then,
or
10
10
10
10
10
10
Frequency (rad/sec)
This plot shows a high attenuation of the source voltage v s and thus the transresistance circuit model should not be used as a voltage amplifier.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
131
v in
R in
G m v in
R out
R load v load
Also,
6.
C R
or
132
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
10
0 0 10
10
7.
v in
R i C
v out
v in i = --------------------------R + 1 jC
1 t v out ------- v in dt RC 0
We can illustrate this with a Simulink* model as follows: The transfer function of the low-pass filter above is readily derived from the voltage division expression, i.e.,
1 sC V ( s ) out = ----------------------- V ( s ) in R + 1 sC
from which
* For an introduction to Simulink, please refer to Appendix B
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
133
and for RC = 1 ,
1 G ( s ) = ---------s+1
After the execution command is issued, the Scope block displays the input and output waveforms below.
8.
C
i
v in i = --------------------------R + 1 jC
and for 1 RC or R 1 C
134
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
or Also, Thus,
We can illustrate this with a Simulink model as follows: The transfer function of the high-pass filter above is readily derived from the voltage division expression, i.e.,
R V ( s ) out = ----------------------- V ( s ) in R + 1 sC
from which
V ( s ) out RCs 1 G ( s ) = ------------------ = ------------------- = ------------------RCs + 1 RCs + 1 V ( s ) in
and for RC = 1 ,
s G ( s ) = ---------s+1
After the execution command is issued, the Scope block displays the input and output waveforms below.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
135
136
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Chapter 2
Introduction to Semiconductor Electronics Diodes
his chapter begins with an introduction to semiconductor electronics. The electron and hole movement is explained and illustrated in simple terms. The Ntype and Ptype semiconductors are discussed and majority and minority carriers are defined. The junction diode, its characteristics, and applications are discussed. The chapter concludes with the introduction of other types of diodes, i.e., Zener diodes, tunnel diodes, and others.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
21
free electron
hole
Figure 2.2. Free electron and the created hole in a partial silicon crystal
When a free electron approaches a hole it is attracted and captured by that hole. Then, that free electron becomes once again a bound electron and this action is called recombination. Accordingly, in a silicon crystal that has been thermally agitated we have two types of current movement; the free electron movement and the hole movement. The movement of holes can be best illustrated with the arrangement in Figure 2.3.
free electron
1 2
hole 3
Figure 2.3 shows that a hole exists in position 1. Let us now suppose that the bound electron in position 2 is attracted by the hole in position 1. A new hole has now been created in position 2 and thus we say that the hole has moved from position 1 to position 2. Next, the hole in position
22
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
f ree electron hole Figure 2.4. Free electron and hole movement when an external voltage is applied
We should keep in mind that holes are just vacancies and not positive charges although they move the same way as positive charges. We should also remember that in both Ntype and Ptype materials, current flow in the external circuit consists of electrons moving out of the negative terminal of the battery and into the positive terminal of the battery. Hole flow, on the other hand, only exists within the material itself. Doping is a process where impurity atoms* which are atoms with five valence electrons such as phosphorous, arsenic, and antimony, or atoms with three valence electrons such as boron, aluminum, and gallium, are added to melted silicon. The silicon is first melted to break down its original crystal structure and then impurity atoms are added. The newly formed compound then can be either an Ntype semiconductor or a Ptype semiconductor depending on the impurity atoms that were added as shown in Figure 2.5.
Si
free electron
Si
hole
Si
As
Si
Si = Silicon As = Arsenic Ga = Gallium
Si
Ga
Si
Si
N type semiconductor
Si
P type semiconductor
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
23
P N
Physical Structure
B
Depletion Region
Symbol
We would think that if we join the N and P materials together by one of the processes mentioned earlier, all the holes and electrons would pair up. This does not happen. Instead the electrons in the N material diffuse (move or spread out) across the junction into the P material and fill some
24
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
P N
Figure 2.8. Forwardbiased junction diode
When a junction diode is forwardbiased, conventional current will flow in the direction of the arrow on the diode symbol. If we reverse the voltage source terminals as shown in Figure 2.9, a reversebiased PN junction is formed.
P N
Figure 2.9. Reversebiased junction diode
When a junction diode is reversebiased, ideally no current will flows through the diode.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
25
vD
With reference to Figure 2.11 we see that when v D > 0 , ideally i D , and when v D < 0 , ideally
i D 0 . However, the actual i D v D relationship in a forwardbiased junction diode is the non-
linear relation
iD = Ir [ e
( qv D nkT )
1]
(2.1)
where i D and v D are as shown in Figure 2.10, I r is the reverse current, that is, the current which would flow through the diode if the polarity of v D is reversed, q is charge of an electron, that is,
coulomb , the coefficient n varies from 1 to 2 depending on the current level and the nature or the recombination near the junction, k = Boltzmann's constant , that is, q = 1.6 10
* It is immaterial whether we use the electron current flow or the conventional current flow. The equations for the voltage current relationships are the same as proved in Circuit Analysis I with MATLAB Applications, ISBN 97809709511 20.
19
26
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
joule Kelvin , and T is the absolute temperature in degrees Kelvin, that is,
o
T = 273 + temperature in C . It is convenient to combine q , k , and T in (2.1) into one variable V T known as thermal voltage where V T = kT q
(2.2)
1]
iD = Ir [ e
( v D nV T )
(2.3)
19
= kT q = 1.38 10
23
300 1.6 10
26mV
(2.4)
We will use the MATLAB script below to plot the instantaneous current i D versus the instantaneous voltage v D for the interval 0 v D 10 v , with n = 1 , and temperature at 27 C .
vD=0: 0.001: 1; iR=10^(15); n=1; VT=26*10^(3);... iD=iR.*(exp(vD./(n.*VT))1); plot(vD,iD); axis([0 1 0 0.005]);... xlabel('Diode voltage vD, volts'); ylabel('Diode current iD, amps');... title('iDvD characteristics for a forwardbiased junction diode'); grid
5 x 10
-3
0.1
0.2
0.3
0.8
0.9
The curve of Figure 2.12 shows that in a junction diode made with silicon and an impurity, conventional current will flow in the direction of the arrow of the diode as long as the voltage drop v D across the diode is about 0.65 volt or greater. We also observe that at v D = 0.7 V , the current through the diode is i D 0.5 mA . When a junction diode is reversebiased, as shown in Figure 2.13, a very small current will flow
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
27
Commercially available diodes are provided with a given rating (volts, watts) by the manufacturer, and if these ratings are exceeded, the diode will burnout in either the forwardbiased or the reversebiased direction. The maximum amount of average current that can be permitted to flow in the forward direction is referred to as the maximum average forward current and it is specified at a special temperature, usually 25 C . If this rating is exceeded, structure breakdown can occur. The maximum peak current that can be permitted to flow in the forward direction in the form of recurring pulses is referred to as the peak forward current. The maximum current permitted to flow in the forward direction in the form of nonrecurring pulses is referred to as the maximum surge current. Current should not equal this value for more than a few milliseconds. The maximum reversebias voltage that may be applied to a diode without causing junction breakdown is referred to as the Peak Reverse Voltage (PRV) and it is the most important rating. All of the above ratings are subject to change with temperature variations. If, for example, the operating temperature is above that stated for the ratings, the ratings must be decreased. There are many types of diodes varying in size from the size of a pinhead used in subminiature circuitry, to large 250ampere diodes used in highpower circuits. A typical diode is identified as XNYYYY where X denotes the number of semiconductor junctions (1 for diodes, 2 for transistors, and 3 a tetrode which has three junctions), N identifies the device as a semiconductor, and YYYY is an identification number. For instance, 1N4148 is a semiconductor diode and 2N3904 is a transistor. We will discuss transistors in Chapter 3. The N side of a typical junction diode has a black band as shown in Figure 2.14.
28
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Diodes are used in various applications where it is desired to have electric current flow in one direction but to be blocked in the opposite direction as shown in Figure 2.15.
VD R 1.5 V ID V out 1.5 V ID = 0 V out = 0
In the circuit of Figure 2.15(a) the diode is forward biased, so current flows, and thus V out = 1.5 V D = 1.5 0.7 = 0.8 V . In the circuit of Figure 2.15(b) the diode is reversebiased, no current flows, and thus V out = 0 .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
29
V TH
+
1V
1 K
+
D
vD
iD
0.1
0.2
0.3
0.8
0.9
Solution: The current i D through the diode is also the current through the resistor. Then, by KVL
vR + vD = 1 V Ri D = v D + 1 1 1 i D = --- v D + --R R
(2.5)
We observe that (2.5) is an equation of a straight line and two points of this straight line can be obtained by first letting v D = 0 , then i D = 0 . We obtain the straight line shown in Figure 2.18 which is plotted on the same graph as the given diode i v characteristics. This line is referred to as a load line.
210
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Figure 2.18. Curves for determining voltage and current in the diode of Example 2.1
The intersection of the nonlinear curve and the load line yields the voltage and the current of the diode where we find that v D 0.68V and i D 0.32 mA . Check: Since this is a series circuit, i R = 0.32 mA also. Therefore, the voltage drop v R across the resistor is
v R = 1 K 0.32 mA = 0.32 V . Then, by KVL v R + v D = 0.32 + 0.68 = 1 V
The relation of (2.3) gives us the current when the voltage is known. Quite often, we want to find the voltage when the current is known. To do that we rewrite (2.3) as
iD + Ir = Ir e
( v D nV T )
(2.6)
or
iD ( v nV ) ---- = e D T Ir
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
211
(2.7)
(2.8)
Example 2.2 Derive an expression for the voltage change v = V 2 V 1 corresponding to a current change
i = I 2 I 1 .
1]
iD + Ir = Ir e
( v D nV T )
and since i D I r
iD Ir e
( v D nV T )
I D1 I r e I D2 I r e
V nV
V D1 nV T
V D2 nV T
D2 T I D2 I r e ( V V ) nV T ------- ----------------------- = e D2 D1 V D1 nV T I D1 I e
212
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(2.9)
Example 2.3 Experiments have shown that the reverse current I r increases by about 15% per 1 C rise, and it is known that for a certain diode I r = 10 Solution:
Ir
52 C 14
A at 27 C . Compute I r at 52 C . = 10
14
= 10
14
( 1 + 0.15 )
( 52 27 )C
( 1.15 )
25
3.3 10
13
This represents about 97% increase in reverse current when the temperature rises from 27 C to 52 C .
Using the approximation with the straight lines shown in Figure 2.19, we can now represent a typical junction diode with the equivalent circuit shown in Figure 2.20.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
213
rD
B ID
( a ) Practical diode
In Figure 2.20(b), the diode represents an ideal diode whose i v characteristics are shown in Figure 2.21, the horizontal solid line in Figure 2.19 represents the small voltage source V D in Figure 2.20(b), and the reciprocal of the slope of the line in Figure 2.19 is represented by the resistance r D shown in Figure 2.20(b). For convenience, these representations are also illustrated in Figure 2.21.
rD
VD 0V 0.65 V slope = 1 r D
Example 2.4 In the circuit of Figure 2.22(a) the diodes are identical and the piecewise linear i v characteristics are shown in Figure 2.22(b). Find the voltage V out .
i ( ma )
2 K 1V 2V 3V
2 K
AV
v (V)
(a)
(b)
Figure 2.22. Circuit and piecewise linear i v characteristics for Example 2.4
Solution: In Figure 2.23 we have replaced the diodes by their piecewise linear equivalents and have combined the two parallel resistors. Also, for each branch we have combined the diode voltage
V D = 0.7 V with the applied voltages, and r D = 1 slope = 50 10
3
A V = 20 .
214
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
V out
Let us follow the procedure below to find out if we can arrive to a valid answer. By Kirchoffs Current Law (KCL)
I1 + I2 + I3 = I4 V out 0.3 V out 1.3 V out 2.3 V out ------------------------ + ----------------------- + ------------------------ = ----------20 20 1000 20 15 50V out + 65 50V out + 115 50V out V out ------------------------------------------------------------------------------------------------------------------------- = 0 1000 151 V out = 195
Check:
We see that the current I 1 cannot be negative, that is, it cannot flow on the opposite direction of the one shown. Also, the current I 2 is zero. Therefore, we must conclude that only the diode on the right side conducts and by the voltage division expression
V out = ----------------------- 1.3 1.3 V 1000 20 + 1000
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
215
iD ( t ) Q t
vD ( t ) t Figure 2.24. Junction diode biased at point Q and changes in i D corresponding to changes in v D
Figure 2.24 shows how changes in v D ( t ) result in changes in i D ( t ) . We can derive an expression that relates v D ( t ) and i D ( t ) in a junction diode. The current I D produced by the bias DC voltage V D is
ID = Ir e
V D nV T
(2.10)
(2.11)
216
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
= Ir e
V D nV T v D nV T
(2.12)
If v D nV T 0.1 we can use Maclaurins series* expansion on (2.12) using the relation
f (0) n f ( 0 ) 2 f ( x ) = f ( 0 ) + f ( 0 )x + ------------- x + + ---------------- x n! 2!
(n)
(2.13)
(2.14)
We must remember that the approximation in (2.14) is a smallsignal approximation and should be used only when v D nVT 0.1 . The ratio I D nV T in (2.14) is denoted as g D and is referred to as incremental conductance. Its reciprocal nV T I D is denoted as r D is called incremental resistance. Example 2.5 In the circuit of Figure 2.25, the current through the diode is I D = 1 mA when V D = 0.7 V , and it is known that n = 2 . Find the DC voltage V out and the AC voltage v out at 27 C where
V T = 26 mV .
R 5mV AC 5 K
v out
5V DC
Solution: Replacing the diode with the piecewise linear equivalent we obtain the circuit of Figure 2.26.
* For a detailed discussion on Taylor and Maclaurins series refer to Numerical Analysis Using MATLAB and Excel, ISBN 9781934404034.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
217
5V DC
iD rD
v T out
Figure 2.26. The piecewise linear equivalent of the circuit in Example 2.5
We apply the superposition principle* for this example. We fist consider the DC voltage source acting alone, by suppressing (shorting out) the AC voltage source to find V out . Then, we consider the AC voltage source acting alone by suppressing (shorting out) the DC voltage source to find v out . The total output voltage will be the sum of these two, that is, v T out = V out + v out . With the DC voltage source acting alone and with the assumption that r D 5 K , the current
I D is 5 0.7 I D = ---------------- = 0.86mA 3 5 10
v = 0.7V
Therefore,
With the AC voltage source acting alone the value of the incremental resistance r D is the reciprocal of (2.14) and thus
3 nV T 2 26 10 r D = --------- = ------------------------------- = 60.5 3 ID 0.86 10
Then, Therefore
* Generally, the superposition principle applies only to linear circuits. However, it is also applicable for this example since it is applied to a piecewise linear equivalent circuit.
218
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
vD
1 .5
v in
iD
v out
1 0 .5 0 0 .5
(a)
(b)
(c)
In the halfwave rectifier circuit of Figure 2.27(b), the diode is forwardbiased during the positive halfcycle from 0 to of the input voltage v in and so current flows through the diode and resistor where it develops an output voltage drop v out = v in v D . For instance, if the maximum value of v in is 10 V volts, the maximum value of v out will be v out = 10 0.7 = 9.3 V . The diode is reverse biased during the negative half cycle from to 2 so no current flows, and thus v out = 0 V . The waveform in Figure 2.27(c) can be simulated* with the Saturation block in the model in Figure 2.28.
Figure 2.28. Simulink model for creating halfrectification waveform with a saturation block
The Scope block in Figure 2.28 displays the halfrectification waveform shown in Figure 2.29.
* This and other simple Simulink models presented in this text could be parts of a large model. For a brief introduction to Simulink please refer to Appendix B in this text, and for a complete coverage refer to Introduction to Simulink with Engineering Applications, Second Edition, ISBN 9781934404096.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
219
Figure 2.29. Halfrectification waveform displayed by the Scope block in Figure 2.28
Example 2.6 Design a DC voltmeter* that will have a 10 volt fullscale using a milliammeter with 1 milliampere fullscale and internal resistance 20 , a junction diode, and an external resistor R whose value must be found. The input is an AC voltage with a value of 63 volts peaktopeak. Assume that the diode is ideal. Solution: Typically, a voltmeter is a modified milliammeter where an external resistor R V is connected in series with the milliammeter as shown in Figure 2.30 where
I = current through circuit R M = internal resis tan ce of milliameter R V = external resistor in series with RM V M = voltmeter full scale reading IM
+
mA
RM VM
RV
Because the available input voltage is AC and we want to read DC (average) values, we insert a junction diode is series as shown in Figure 2.31, and we need to find the value of R V so that the
* For a detailed discussion on electronic instruments refer to Circuit Analysis I with MATLAB Applications, ISBN 978 0970951120.
220
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
RM 20 VM
RV = ?
1 mA
Since the diode does not conduct during the negative half cycle, it follows that
Vp p 2 31.5 I peak = I p = --------------------- = -----------------20 + R V RM + RV
and
31.5 - -----------------I p sin t dt sin t dt 20 + R V 0 Area - = ---------------------------- = ----------------------------------------- = 31.5 ( cos t ) 0 = --------------------------31.5 0 --------------------------------I ave = ---------------Period 2 2 ( 20 + R V )2 ( 20 + R V )
31.5 3 20 + R V = --------- 10 R V = 10 K
Of course, we can label our voltmeter for any value such as 50 V , 100 V , and so on and can use any AC waveform with different peaktopeak values. Figure 2.32(a) shows a halfwave rectifier circuit consisting of a transformer,* a junction diode, and a load resistor, and Figure 2.32(b) shown the waveforms of the input voltage v in and the load voltage v load .
* For a detailed discussion on transformers, refer to Circuit Analysis II with MATLAB Applications, ISBN 9780 970951151.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
221
vD
+ vs
iD + v load v in v load
(b)
R load
Figure 2.32. Halfwave rectifier circuit and input and output waveforms
When using diodes in rectifier circuits we must calculate: a. the maximum current that the diode will allow without being damaged, and b. The Peak Inverse Voltage (PIV) that the diode can withstand without reaching the reverse biased breakdown region. If the applied voltage is v S = V p sin t , then PIV = V p but in practice we must use diodes whose reversebiased breakdown voltages 70% or greater than the value of V p . As shown in Figure 2.32(b), the diode begins conducting sometime after the input voltage shown by the dotted curve rises to about 0.7 V . We derive the angle by which the solid curve lags the dotted curve with reference to Figure 2.33* as follows.
v = v in v out
Figure 2.33. Waveforms for the derivation of conduction angle for a halfwave rectifier
* These waveforms can be created in a Simulink model with a Sine Wave block, an Abs block, a Dead Zone block, and displayed in a Scope block.
222
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(2.15)
We also observe that the conduction terminates at the angle ( ) and therefore the entire conduction angle is (2.16) ( ) = ( 2 ) We can also find the average value of the waveform of v out . We begin with the definition of the average value, that is,
1 Area V out ( ave ) = ---------------- = ----2 Period
( )
Generally, the angle is small and thus cos 1 and ( 2 ) . Therefore, the last relation above reduces to
Vp - -----V out ( ave ) ----- v 2
(2.17)
Figure 2.34 shows a fullwave bridge rectifier circuit with input the sinusoid v in ( t ) = A sin t as shown in Figure 2.35, and the output of that circuit is v out ( t ) = A sin t as shown in Figure 2.36.
A
+
v in ( t )
D C
+
v out ( t )
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
223
2 A
Figure 2.35. Input waveform for the circuit in Figure 2.34 A v out 2
0 0
From Figure 2.34 we see that during the positive half cycle conventional current flows from the voltage source to Point A, then to Point B, it goes through the resistor from Point B to Point C, and through Point D returns to the negative terminal of the voltage source. During the negative half cycle the lower terminal of the voltage source becomes the positive terminal, current flows from Point D to Point B, it goes through the resistor from Point B to Point C, and through Point A returns to the upper (now negative) terminal of the voltage source. We observe that during both the positive and negative halfcycles the current enters the right terminal of the resistor, and thus v out is the same for both halfcycles as shown in the output waveform of Figure 2.36. Figure 2.37 shows the input and output waveforms of the fullwave bridge rectifier on the same graph. It is to be noted that the difference in amplitude between v in and v out is denoted as 2 v because in a fullwave bridge rectifier circuit there are two diodes in the conduction path instead of one as shown in Figure 2.33 for the halfwave rectifier. Accordingly, v out lags v in by the angle
= sin ( 2 v v in ) . Also, the output is zero for an angle 2 = 2 sin ( 2 v v in ) centered
1 1
Figure 2.37. Input and output waveforms for the fullwave bridge rectifier of Figure 2.34
FF
224
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
vS
(a)
v load vin
v load
vin v D
+
The waveform in Figure 2.36 can be simulated with the Saturation block in the model in Figure 2.39.
Figure 2.39. Simulink model for creating fullrectification waveform with an Abs block
The Scope block in Figure 2.39 displays the halfrectification waveform shown in Figure 2.40.
Figure 2.40. Fullrectification waveform displayed by the Scope block in Figure 2.39
The output voltages v out from the half and full wave rectifiers are often called pulsating DC voltages. These voltages can be smoothedout with the use of electric filters as illustrated with the following example.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
225
(2.18)
where A is the amplitude and this waveform appears across the resistor R of the fullwave rectifier in Figure 2.41 where the inductor and capacitor form a filter to smooth out the pulsating DC. +
v in ( t )
R L
+
vR ( t )
5H
+
C R load
10 F
v load
2 K
Compute and sketch the voltage v load assuming that v in ( t ) = 120 V RMS operating at the fundamental frequency f = 60 Hz . Solution: We replace the given circuit by its phasor equivalent as shown in Figure 2.42 where the inductive reactance is X L = j n L and the capacitive reactance is X C = 1 j n C . +
V in
R Z1
+
VR
j5 n 10 j n
5
+
Z2
V load 2 10
3
226
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(2.19)
We will now compute the components of V load for n = 0, 2, and 4 and using superposition we will add these three terms. We were given that
2A 4A cos 2t cos 4t v R ( t ) = ------ ------ ----------------- + ----------------- + 3 15
(2.20)
and so for n = 0 ,
v R ( t ) n = 0 = 2A -----
and since for this DC case the inductor is just a short and the capacitor an open,
V load
n=0
2A = V R = -----
(2.21)
4A = ------ 0 3
We will use MATLAB to find the magnitude and phase angle of the bracketed expression above.
bracket1=(2*10^8)/((j*5*754)*(10^5+2*10^3*j*754)+2*10^8);... abs(bracket1), angle(bracket1)*180/pi
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
227
(2.22)
n=2
4A = -------- 0 15
We will again use MATLAB to find the magnitude and phase angle of the bracketed expression above.
bracket2=(2*10^8)/((j*5*1508)*(10^5+2*10^3*j*1508)+2*10^8);... abs(bracket2), angle(bracket2)*180/pi
(2.23)
n=4
(2.24)
228
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(2.25)
This is the filtered output voltage across the 2 K resistor. Let us plot (2.25) using MATLAB with A = 20 and = 377 .
t=0: 1: 2000; vL=20.*(2./pi0.0154.*cos(2.*377.*t176.1.*pi./180)... 0.00076.*cos(4.*377.*t178.1.*pi./180)); plot(t,vL); axis([0 2000 0 20]); hold on;... plot(t,(40/pi0.0154)); axis([0 2000 0 20]); xlabel('Time (sec)');... ylabel('Load voltage (vL)'); title('Filtered output voltage for Example 2.7'); grid
The plot is shown in Figure 2.43 and we observe that the ripple is approximately 0.5 V about the average (DC) value.
20 Filtered output voltage for Example 2.7
15
10
500
1500
2000
The waveform in Figure 2.43 can be simulated with the Simulink model in Figure 2.44. The value A = 20 must be entered at the MATLAB command prompt.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
229
Figure 2.45. Fullrectification waveform displayed by the Scope block in Figure 2.44
230
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v in
(a)
v out
The peak rectifier circuit shown in Figure 2.46(a) will behave like an AC to DC converter if we add a resistor across the capacitor as shown in Figure 2.47(a). Then, assuming that the diode is ideal and that the time constant = RC is much greater than the discharge interval, the voltage across the resistorcapacitor combination will be as shown in Figure 2.47(b).
V pp ( ripple )
iD + v in
(a)
v out iR iC +
C
v in = V p sin t t
v out
T ( period ) (b)
From the waveforms of Figure 2.47(b) se see that peaktopeak ripple voltage V pp ( ripple ) can be made sufficiently small by choosing the time constant = RC much larger that the period T , that is, RC T . Also, from Figure 2.47(b) we see that the average output voltage V out ( ave ) when the diode conducts is
1 V out ( ave ) = V p -- V pp ( ripple ) 2
(2.26)
Next, we need to find an expression for V pp ( ripple ) when the diode does not conduct. We recall from basic circuit theory* that in an simple RC circuit the capacitor voltage v C discharges as the decaying exponential
v C = v out = V p e
( t RC )
(2.27)
( t RC )
or
v out = V p V pp ( ripple ) = V p e
(2.28)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
231
x x = 1 x + ---- ---- + 2! 3! e
x
1x T ) = V p 1 1 ------- RC
= Vp ( 1 e
( t RC )
or
V pp ( ripple ) = V p ------RC
(2.29)
and from (2.29) we observe that for V pp ( ripple ) to be small, we must make RC T
v out
vin
+ v out
vin
Figure 2.48. Circuit where v out = v in when diode does not conduct and v out 0.7 V when v in 0.7 V
Figure 2.49 shows a clipper circuit and its inputoutput characteristics where the diode does not conduct for v in > 0.6 V but it conducts for v in 0.7 V and thus v out 0.7 V .
232
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Clipper Circuits
R
vin
vin
Figure 2.49. Circuit where v out = v in if diode does not conduct and v out 0.7 V if v in 0.7 V
Figure 2.50 shows a clipper circuit with two diodes in parallel with opposite polarities. This circuit is effectively a combination of the clipper circuits shown in Figures 2.48 and 2.49 and both diodes are not conducting when 0.7 v in 0.7 and for this interval v out = v in . Outside this interval, one or the other diode conducts. This circuit is often referred to as a hard limiter.
R 0.7 V
v out
vin
+ v out 0.7 V
vin
Figure 2.50. Circuit where v out = v in when 0.7 v in 0.7 and v out 0.7 V outside this interval
Occasionally, it is desirable to raise the limit level to a value other than v out 0.7 V . This can be accomplished by placing a DC source in series with the diode as shown in Figure 2.51.
VA + 0.7 V VA
v out
+ v in
VA
+ + v out
v in
Figure 2.51. Circuit where v out = v in if diode does not conduct and v out VA + 0.7 V if it conducts
Example 2.8 In the circuit of Figure 2.52, all three resistors have the same value, VA = V B with the polarities shown, the diodes are identical with v D = 0.7 , and r D R . Derive expressions for and sketch the
v out v in characteristics.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
233
+ v in
+
D1 VA
+
R
D2 VB
v + out
R
I D1
I D2
Solution: Diode D 1 conducts when v in > VA and under this condition diode D 2 does not conduct. Then,
v in ( VA + v D ) v in VA 0.7 I D1 = ------------------------------------ = -------------------------------R+R 2R
(2.30)
and
( v in VA 0.7 ) v out1 = 0.7 + VA + RI D1 = 0.7 + VA + R ------------------------------------- = 0.5 ( v in + VA + 0.7 ) 2R
(2.31)
Diode D 2 conducts when v in < V B and under this condition diode D 1 does not conduct. Then, in accordance with the passive sign convention* we obtain
v in ( V B v D ) v in V B + 0.7 I D2 = ------------------------------------ = --------------------------------R+R 2R
(2.32)
and
v in V B + 0.7 v out2 = 0.7 V B R I D1 = 0.7 V B R --------------------------------- = 0.5 ( v in V B 0.7 ) 2R
(2.33)
For range V B < v in < V A , v out = v in , and the v out v in characteristics are shown in Figure 2.53.
v out v out1 v in v out2 Figure 2.53. The v out v in characteristics for the circuit of Example 2.8 * Please refer to Circuit Analysis I with MATLAB Applications, ISBN 9780970951120.
234
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
+ iD
+ v out
v in
v in ( V )
t
Solution: From the circuit of Figure 2.54 we observe that the diode conducts only when v in < 0 and thus the capacitor charges to the negative peak of the input which for this example is 5 V , and the capacitor voltage is v C = 5 V with polarity as shown in Figure 2.54. We observe that there is no path for the capacitor to discharge and therefore v C = 5 V always. Since v out = v in + v C , when
v in = 5 V , v out = 5 + 5 = 0 and the output has shifted upwards to zero volts as shown in Figure 2.56. When the input voltage is positive, v in = 5 V , and v out = 5 + 5 = 10 V . The waveform
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
235
v out ( V )
+ + D2 v D1 C v out = 2 v in
2
v in = V p sin t
With reference to the circuits of Figures 2.46 and 2.54, we observe that the circuit of Figure 2.57 is a combination of a DC restorer circuit consisting of the capacitor C 1 and diode D 1 and a peak rectifier consisting of C 2 and diode D 2 . During the positive halfcycle of the input waveform diode D 1 conducts, the peak value V p of the input v in appears across the capacitor C 1 of and thus v D1 = 0 . During the negative halfcycle of the input waveform diode D 2 conducts, capacitor C 2 is being charged to a voltage which is the sum of v in and v C1 and thus the output voltage is v out = v in + v C1 = V p sin t V p sin t = 2 V p sin t = 2v in as shown in Figure 2.58.
v out
0
Vp 2V p Figure 2.58. Output waveform for the voltage doubler circuit of Figure 2.57
236
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
C2 R2
+ v out
The operation of an AM audio frequency signal is shown in Figure 2.60 where the magnitude of v m ( t ) is always less than V C ; hence the envelope of the modulated wave never drops to zero. The waveform of the envelope of the modulated signal is the same as the waveform of the modulating signal v m ( t ) . Thus if the information in the modulated carrier is to be preserved, the waveform of the envelope must be preserved. The information can be recovered by recovering the waveform of the envelope; the peak rectifier circuit is used for this purpose. The peak rectifier is also known as diode detector, and in this application it is frequently called an envelope detector. v m . The diode removes the lower envelope and the upper envelope. In the circuit of Figure 2.59 capacitor C 2 and resistor R 2 form a highpass filter that removes the carrier frequency.
vm ( t )
Modulating signal Carrier Envelope
vC ( t )
Figure 2.60. Input and output signals for the circuit of Figure 2.59
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
237
These diodes are called Zener diodes and they are designed to operate in the breakdown region. The Zener diode is always connected as a reversebiased diode and its voltage rating V Z and the maximum power it can absorb are given by the manufacturer. The Zener diode symbol is shown in Figure 2.63.
+ IZ
VZ
Zener diodes from small voltage ratings of 5 volts to large voltage ratings of 250 volts are commercially available and we can calculate the maximum current that a Zener diode can withstand from its voltage and power ratings. For instance, a 2 watt, 25 volt Zener diode can withstand a maximum current of 80 milliamperes.
238
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Zener Diodes
One important application of the Zener diode is in voltage regulation. A voltage regulator is a circuit or device that holds an output voltage constant during output load variations. Example 2.10 below illustrates how a Zener diode accomplishes this. Example 2.10 For the circuits (a) and (b) in Figure 2.64 it is required that the output voltage v out remains constant at 25 volts regardless of the value variations of the load resistor R L . Calculate the output voltage v out for both circuits and then show how a 25 volt Zener diode can be used with these circuits to keep the output voltage constant at 25 volts.
RS 20 K VS R load1 v out1 40 V 20 K RS VS
20 K R load2 40 V
v out2 60 K (b)
Solution: For the circuit of Figure 2.64(a) the output voltage v out1 is found by application of the voltage division expression. Thus,
20 K load1 v out1 = --------------------------- VS = ------------------------------------- 40 V = 20 V 20 K + 20 K RS + R load1 R
Likewise, for the circuit of Figure 2.64(b) the output voltage v out2 is
60 K L v out2 = --------------------------- VS = ------------------------------------- 40 V = 30 V 20 K + 60 K RS + R load2 R
These calculations show that as the load R load varies from 20 K to 60 K the output voltage varies from 20 volts to 30 volts and this variation indicates a poor voltage regulation. To improve the regulation we connect a 25 volt Zener diode in parallel with the load as shown in Figure 2.65 and the output voltages v out1 and v out2 will be the same, that is, 25 volts.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
239
20 K R load1 40 V 20 K
+
IZ
+
V Z = 25 V VS
20 K R load2 40 V 60 K
+
IZ
V Z = 25 V
(a)
(b)
Figure 2.65. Circuits for Example 2.10 with Zener diodes to improve regulation
In Example 2.10 above we did not take into consideration the manufacturers specifications. Example 2.11 below is a more realistic example and shows that a Zener diode may or may not conduct depending on the value of the load resistor. The Zener diode model is very similar to the junction diode model and it is shown in Figure 2.66(a). Of course, the Zener diode is operating at a different region where V Z0 and r Z are as shown in Figure 2.66(b).
VZ0
i
0 Slope = 1 r Z
+
VZ
IZ
rZ
+ VZ0
VZ = VZ0 + rZ IZ
(a)
(b)
Figure 2.66. The Zener diode model and the definitions of V Z0 and r Z
Thus, V Z0 is v axis intercept of the straight line with slope 1 r Z and r Z = v i . Like the incremental resistance r D of a junction diode, the incremental resistance r Z of a Zener diode is typically about 20 and it is specified by the manufacturer. Example 2.11 For the circuit of Figure 2.67, the manufacturers datasheet for the Zener diode shows that V Z = 10 V when I Z = 6 mA . The datasheet provides also the i v characteristics but the value of V Z0 is not given. The current I ZBD in Figure 2.68 represents the current where the Zener diode begins entering the breakdown region.
240
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Zener Diodes
RS VS
+ 24 V
1 K IZ
+
VZ R load
+
V load
I ZBD = 0.3 mA
Compute: a. The output voltage V load if the load resistor R load is disconnected. b. The output voltage V load if the load resistor R load is connected and adjusted to 1 K . c. The output voltage V load if the load resistor R load is connected and adjusted to 200 . d. The minimum value of the load resistor R load for which the Zener diode will be conducting in the breakdown region. Solution: a. From Figure 2.66(a),
VZ = VZ0 + rZ IZ
(2.34)
and replacing the Zener diode by its equivalent we obtain the circuit shown in Figure 2.69 where the slope is
1 r Z = i v = 0.04 r Z = v i = 1 0.04 = 25
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
241
+
VS
1 K 24 V
+ V Z0 rZ IZ
+ V load
Figure 2.69. Zener diode replaced by its equivalent circuit and the load is disconnected
VZ0 = VZ rZ IZ = 10 25 6 10
= 8.5 V
Thus b.
= 8.8 V
When the load resistor is connected and adjusted to 1 K the circuit is as shown in Figure 2.70.
RS
+
VS
1 K
VZ0
24 V
IZ
+ I load R load rZ 1 K
+ V load
Figure 2.70. Circuit of Example 2.11 with load resistor connected and adjusted to 1 K
We will assume that the Zener diode still operates well within the breakdown region so that the addition of the load resistor does not change the output voltage significantly. Then,
V load 8.8 I load ------------- = ------- = 8.8 mA 3 R load 10
242
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Zener Diodes
V load = VZ0 + rZ I Z = 8.5 + 25 6.4 10 '
3
= 8.5 V
c. When the load resistor is connected and adjusted to 200 the circuit is as shown in Figure 2.71.
RS
VS
1 K 24 V
VZ0
+ rZ
+ I load
R load
V load
IZ
200
Figure 2.71. Circuit of Example 2.11 with load resistor connected and adjusted to 200
We will again assume that the Zener diode still operates well within the breakdown region so that the addition of the load resistor does not change the output voltage significantly. Then,
V load 8.8 I load ------------ = -------- = 44 mA R load 200
But this is almost three times higher than the current of IZ = 15.2 mA found in (a). Therefore, we conclude that the Zener diode does not conduct and, with the Zener diode branch being an open circuit, by the voltage division expression we obtain
R load 200 V load = ------------------------ V S = -------------------------- 24 = 4 V 1000 + 200 R S + R load
and since this voltage is less than 10 V the Zener diode does not conduct. d. From the graph of Figure 2.68 it is reasonable to assume that when I ZBD = 0.3 mA , then
V ZBD 9.7 V . Then, the current I RS through the resistor which would allow the Zener diode
to conduct will be
V S V ZBD 24 9.7 I RS = -------------------------- = ------------------ = 14.3 mA 3 RS 10
This is the total current supplied by the V S source and since I ZBD = 0.3 mA , the current
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
243
Therefore,
Zener diodes can also be used as limiters. Figure 2.72 shows how a single Zener diode can be used to limit one side of a sinusoidal waveform to V Z while limiting the other side to approximately zero.
v in +
VZ
+ v out
VZ
v out
v in
0.7 V
Figure 2.73 shows a circuit with two opposing Zener diodes limiting the input waveform to V Z on both polarities.
v in + v in + v out v out
VZ
VZ
244
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(a)
(b)
The junction of a doped semiconductor usually n type with a special metal electrode can produce a very fast switching diode which is mainly used in high (up to 5 MHz) frequency circuits or high speed digital circuits. When the diode is forwardbiased, the electrons move from the n type material to the metal and give up their energy quickly. Since there are no minority carriers (holes) the conduction stops quickly and changes to reversebias. Schottky diodes find wide application as rectifiers for high frequency signals and also are used in the design of galliun arsenide circuits. The forward voltage drop of a conducting Schottky diode is typically 0.3 to 0.5 volt compared to the 0.6 to 0.8 found in silicon junction diodes.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
245
3 1
The portion of the characteristic curve between points 2 and 3 is the region of negative resistance and an explanation of why a tunnel diode has a region of negative resistance is best understood by using energy levels discussed in semiconductor physics texts. The negative resistance characteristic makes the tunnel diode useful in oscillators and microwave amplifiers. The unijunction transistor, discussed in Chapter 4, has similar oscillator characteristics. A typical tunnel diode oscillator is shown in Figure 2.76.
R1 R2 Tunnel
diode
+
DC bias
+
R3 L C
v out
i DC bias point v
v out
Because the negative resistance region of the tunnel diode is its most important characteristic, it is desirable that the socalled peaktovalley ratio I P V P must be quite high. Typical values of
I P V P are 3.5 for silicon, 6 for germanium, and 15 for gallium arsenide (GaAs), and the corre-
sponding values of V P are 65, 55, and 150 mV respectively. For this reason, silicon is not used in the manufacturing of tunnel diodes. The voltage at which the current begins to rise again is denoted as V V and typical values for silicon, germanium, and GaAs are 420, 350, and 500 mV respectively. A variation of the tunnel diode is the backward diode that is used as a rectifier in which forward (p side negative) direction occurs without the usual voltage offset of a conventional junction diode. The reverse direction corresponds to the conventional forward direction so that the reverse breakdown voltage is typically 0.65 V . The backward diode is also used in circuits with small amplitude waveforms.
246
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Figure 2.77. Circuit and tunnel diode i v characteristics for Example 2.12
Solution: Let the voltage across the tunnel diode be V TD and the voltage across the resistor be V R . Then,
V TD + V R = 0.6 V
or
V TD + 20i = 0.6 V
(2.35)
When i = 0 , (2.35) reduces to V TD = 0.6 and when V TD = 0 , i = 0.6 20 = 30 mA , and with these values we draw the load line shown in Figure 2.78. We observe that when V TD 0.11 V ,
i 25 mA , when V TD 0.29 V , i 16 mA , and when V TD 0.52 V , i 4 mA .
i ( mA ) 30 20 10 0 v (V)
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Figure 2.78. Graphical solution for the determination of the current i for the circuit of Figure 2.77
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
247
A varactor diode uses a PN junction in reverse bias and has a structure such that the capacitance of the diode varies with the reverse voltage. A voltage controlled capacitance is useful in tuning applications. Typical capacitance values are small, in the order of picofarads. Presently, varactors are replacing the old variable capacitor tuning circuits as in television tuners. Figure 2.80 shows a typical varactor tuning circuit where the DC control voltage V C changes the capacitance of the varactor to form a resonant circuit.
R
+ +
vin
VC
v out
248
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Optoelectronic Devices
The LED operating voltage is small, about 1.6 volts forward bias and generally about 10 milliamperes. The life expectancy of the LED is very long, over 100,000 hours of operation. For this reason, LEDs are used widely as power on indicators of digital voltmeters, frequency counters, etc. and as displays for pocket calculators where they form sevensegment displays, as shown in Figure 2.82.
a f g e d c b
In Figure 2.82 the anodes are internally connected. When a negative voltage is applied to the proper cathodes, a number is formed. For example, if negative voltage is applied to all cathodes the number 8 is produced, and if a negative voltage is changed and applied to all cathodes except LED b and e the number 5 is displayed. Sevensegment displays are also available in commoncathode form, in which all cathodes are at the same potential. When replacing LED displays, one must ensure the replacement display is the same type as the faulty display. Since both types look alike, the manufacturers number should be checked. Laser diodes are LEDs specifically designed to produce coherent light with a narrow bandwidth and are suitable for CD players and optical communications. Another optoelectronic device in common use today is the photodiode. Unlike the LED, which produces light, the photodiode receives light and converts it to electrical signals. Basically, the photodiode is a lightcontrolled variable resistor. In total darkness, it has a relatively high resistance and therefore conducts little current. However, when the PN junction is exposed to an external light source, internal resistance decreases and current flow increases. The photodiode is operated with reversebias and conducts current in direct proportion to the intensity of the light
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
249
Switching the light source on or off changes the conduction level of the photodiode and varying the light intensity controls the amount of conduction. Photodiodes respond quickly to changes in light intensity, and for this reason are extremely useful in digital applications such as photographic light meters and optical scanning equipment. A phototransistor is another optoelectronic device that conducts current when exposed to light. We will discuss phototransistor in the next chapter. An older device that uses light in a way similar to the photodiode is the photoconductive cell, or photocell, shown with its schematic symbol in Figure 2.84. Like the photodiode, the photocell is a lightcontrolled variable resistor. However, a typical lighttodark resistance ratio for a photocell is 1: 1000. This means that its resistance could range from 1000 ohms in the light to 1000 kilohms in the dark, or from 2000 ohms in the light to 2000 kilohms in the dark, and so forth. Of course, other ratios are also available. Photocells are used in various types of control and timing circuits as, for example, the automatic street light controllers in most cities. The symbol for a photocell is shown in Figure 2.84.
A solar cell is another device that converts light energy into electrical energy. A solar cell acts much like a battery when exposed to light and produces about 0.45 V volt across its terminals, with current capacity determined by its size. As with batteries, solar cells may be connected in series or parallel to produce higher voltages and currents. The device is finding widespread application in communications satellites and solarpowered homes. The symbol for a solar cell is shown in Figure 2.85.
250
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Optoelectronic Devices
Figure 2.86 shows an optical coupler, also referred to as an optoisolator. The latter name is derived from the fact that it provides isolation between the input and output. It consists of an LED and a photodiode and each of these devices are isolated from each other.
LED
Photodiode
Isolation between the input and output is desirable because it reduces electromagnetic interference. Their most important application is in fiber optic communications links.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
251
1]
where i D is the current through the diode, v D is the voltage drop across the diode, I r is the reverse current, that is, the current which would flow through the diode if the polarity of v D is reversed, q is charge of an electron, that is, q = 1.6 10 coulomb , the coefficient n varies from 1 to 2 depending on the current level and the nature or the recombination near the junction, k = Boltzmann's constant , that is, k = 1.38 10
23 19
252
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Summary
nient to combine q , k , and T into one variable V T known as thermal voltage where
V T = kT q and thus iD = Ir [ e
( v D nV T )
1]
and at T = 300 K , V T 26mV In a junction diode made with silicon and an impurity, conventional current will flow in the direction of the arrow of the diode as long as the voltage drop v D across the diode is about 0.65 volt or greater. In a typical junction diode at v D = 0.7 V , the current through the diode is
i D 1 mA .
When a junction diode is reversebiased, a very small current will flow and if the applied voltage exceeds a certain value the diode will reach its avalanche or Zener region. Commercially available diodes are provided with a given rating (volts, watts) by the manufacturer, and if these ratings are exceeded, the diode will burnout in either the forwardbiased or the reverse biased direction. The maximum reversebias voltage that may be applied to a diode without causing junction breakdown is referred to as the Peak Reverse Voltage (PRV) and it is the most important rating. The analysis of electronic circuits containing diodes can be performed by graphical methods but it is greatly simplified with the use of diode models where we approximate the diode forwardbiased characteristics with two straight lines representing the piecewise linear equivalent circuit. When used with AC circuits of low frequencies, diodes, usually with 1.8 n 2.0 are biased to operate at some point in the neighborhood of the relatively linear region of the i v characteristics where 0.65 v D 0.8 V . A bias point, denoted as Q , is established at the intersection of a load line and the linear region. The current i D ( t ) can be found from the relation
ID i D ( t ) I D + --------- v D ( t ) nV T
provided that v D nVT 0.1 . The ratio I D nV T , denoted as g D , is referred to as incremental conductance and its reciprocal
nV T I D , denoted as r D , is called incremental resistance.
Junction diodes are extensively used in halfwave and fullwave rectifiers to convert AC voltages to pulsating DC voltages. A filter is used at the output to obtain a relatively constant output with a small ripple.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
253
254
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Summary
capacitance values are small, in the order of picofarads. Varactors have now replaced the old variable capacitor tuning circuits as in television tuners. Optoelectronic devices either produce light or use light in their operation. A LightEmitting Diode (LED) is an optoelectronic device that, when forward biased, produces visible light. The light may be red, green, or amber, depending upon the material used to make the diode. The life expectancy of the LED is very long, over 100,000 hours of operation. For this reason, LEDs are used widely as power on indicators of digital voltmeters, frequency counters, etc. and as displays for pocket calculators where they form sevensegment displays. Laser diodes are LEDs specifically designed to produce coherent light with a narrow bandwidth and are suitable for CD players and optical communications. A photodiode is another optoelectronic device which receives light and converts it to electrical signals. Photodiodes respond quickly to changes in light intensity, and for this reason are extremely useful in digital applications such as photographic light meters and optical scanning equipment. A photocell is an older device that uses light in a way similar to the photodiode. Photocells are used in various types of control and timing circuits as, for example, the automatic street light controllers in most cities. A solar cell is another device that converts light energy into electrical energy. A solar cell acts much like a battery when exposed to light and as with batteries, solar cells may be connected in series or parallel to produce higher voltages and currents. Solar cells find widespread application in communications satellites and solarpowered homes. An optical coupler, sometimes referred to as an optoisolator, consists of an LED and a photodiode and each of these devices are isolated from each other. Isolation between the input and output is desirable because it reduces electromagnetic interference. Their most important application is in fiber optic communications links.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
255
log scale. 3. Suggest an experiment that will enable one to compute the numerical value of n . 4. It is known that for a certain diode I r = 10 conclusion can we draw from the result? 5. When a junction diode operates at the reversebias region v < 0 and before avalanche occurs, the current I r is almost constant. Assuming that v V T we can show that i D I r and for this reason I r is often referred to as the saturation current and whereas in the forwardbiased region I r has a typical value of 10 biased region is 10
9 14 13
increases by about 15% per 1 C rise. At what temperature will I r double in value? What
A to 10
15
In the circuit below, the applied voltage V S is not sufficient to drive the diode into the avalanche region, and it is known that V R = 0.5 V at 20 C .
VS
Ir
500 K
VR
Find: a. V R at 0 C b. V R at 40 C 6. For the circuit below, find the value of V out assuming that all three diodes are ideal. Compare your answer with that of Example 2.4. Make any reasonable assumptions.
256
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
1 K 1V 2V 3V i ( ma )
V out
0
v (V)
7. For the circuit shown below, the diodes are identical and it is known that at V D = 0.65 ,
I D = 0.5 mA . It is also known that the voltage across each diode changes by 0.1 V per decade
8. For the circuit shown below, the diodes are identical and when V S = 12 V exactly,
V out = 2.8 V . Assuming that n = 1.5 and T = 27 C find:
a. The percent change in V out when V S changes by 10 % . b. The percent change in V out when a load resistor V load = 2 K is connected across the four diodes.
R = 1 K V S = 12 V 10% V out
Vs = 12V
= 2.8 V
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
257
v load + v load v in
conduction angle
(a)
(b)
11. For the fullwave bridge rectifier shown below it is given that v in ( peak ) = 17 V ,
v D = 0.7 V , R = 200 , and the diode resistance r D R . Find:
a. the diode peak current I peak b. the average value of v load c. the minimum theoretical Peak Inverse Voltage (PIV)
258
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
+
v in ( t )
R
v out ( t )
+
v out ( t ) v in ( t )
12. For the peak rectifier shown below, find the value of the capacitor so that the peaktopeak ripple will be 1 volt .
R
+
C
v in
v out
13. A circuit and its input waveform are shown below. Compute and sketch the waveform for the output v out .
C
+ v in +
iD
v in ( V ) t
+ v out
5
14. The nominal value of a Zener diode is V Z = 12 V at I Z = 10 mA , and r Z = 30 . a. Find V Z if I Z = 5 mA b. Find V Z if I Z = 20 mA c. Find V Z0 of the Zener diode model
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
259
IV VP VV VF
v ( mV )
Shown below is a resistor R , referred to as a tunnel resistor, placed in parallel with the tunnel diode. Determine the value of this resistor so that this parallel circuit will exhibit no negative resistance region.
Tunnel diode
16. A typical solar cell for converting the energy of sunlight into electrical energy in the form of heat is shown in the Figure (a) below. The terminal voltage characteristics are shown in Figure (b) below. a. Determine the approximate operating point that yields the maximum power. What is the value of the maximum power output? b. What should the value of the resistor be to absorb maximum power from the solar cell?
Light i (mA) 40
260
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
0.2
0.8
We observe that when n = 2 , the diode begins to conduct at approximately 7.8 V . 2. Let I D1 = I r e
V D1 nV T
and I D2 = I r e
V D2 nV T
. Then,
e
V D1 nV T
I D2 I D1 = e
V D2 nV T
= e
( V D2 V D1 ) nV T
or
For convenience, we let V D2 V D1 = y and I D2 I D1 = x and we use the following MATLAB script to plot y = 2.3nV T log 10x on semilog scale.
x=1: 10: 10^6; y=2.3.*1.*26.*10.^(3).*log10(x); semilogx(x,y);... xlabel('x=ID2/ID1'); ylabel('y=VD2VD1'); title('Plot for Exercise 2'); grid
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
261
0.3 y=VD2-VD1
0.2
0.1
0 0 10
10
10 x=ID2/ID1
10
3. This can be done with a variable power supply V S , 1 K resistor, a DC voltmeter V , a DC ammeter A and the diode D whose value of n is to be found, connected as shown below. We assume that the voltmeter internal resistance is very high and thus all current produced by V s flows through the diode.
A
1 K D
Vs
VD ID
We can now adjust the variable power supply V S to obtain two pairs i v , say
I D2, V D2
I D1, V D1
and
and use the relation V D2 V D1 = 2.3nVT log 10( I D2 I D1 ) to find n . However, for bet-
ter accuracy, we can adjust V S to obtain the values of several pairs, plot these on semilog paper and find the best straight line that fits these values. 4.
10
13
( 1 + 0.15 ) 1.15
( x 50 )
= 2 10
13
( x 50 ) x
= 2
= 2.167 10
262
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Therefore, we can conclude that the reverse current I r doubles for every 5 C rise in temperature. 5. At 20 C ,
Ir
20 C
and since the reverse current doubles for every 10 C rise in temperature, we have: a. and b. and
VR VR Ir Ir
40 C
= 2 2 1 A = 4 A
6
40 C
= 4 10
500 10 = 2V
0 C
0 C
= 0.25 10
500 10 = 0.125 V
6. We assume that diode D 3 and the resistor are connected first. Then, V out = 3 V and stays at this value because diodes D 1 and D 2 , after being connected to the circuit, are reversebiased.
D1
1V
D2
2V
D3
3V
1 K V out
i ( ma )
v (V)
The value of V out = 3 V is unrealistic when compared with the realistic value which we found in Example 2.4. This is because weve assumed ideal diodes, a physical impossibility.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
263
Since the diodes are identical, for the output voltage to be V out = 3 V , the voltage drop a c r o s s e a c h d i o d e m u s t b e 3 V 4 = 0.75 V , a n d s i n c e t h e c h a n g e i s V D = 0.75 0.65 = 0.1 V , the current is 0.5 10 = 5 mA . Then,
V s V out 12 3 R = ---------------------- = ------------------- = 1.8 K 3 ID 5 10
8.
a.
R = 1 K V s = 12 V 10% V out
Vs = 12V
= 2.8 V
10
The incremental resistance r D per diode is found from r D = nV T I D with n = 1.5 . Thus, for all four diodes
4 1.5 26 10 4r D = --------------------------------------------- = 17 3 9.2 10
3
Therefore, and
V' out = 2.8 0.0167 V 2.8167 2.8 %V out = ----------------------------- 100 = 59.6 % 2.8
264
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Therefore, the current through the four diodes decreases by 1.4 mA and the decrease in voltage across all four diodes will be
V out = 1.4 mA 17 = 23.8 mV
9.
i D ( mA )
1
vD ( V )
0.7
From analytic geometry, the equation of a straight line in an x y plane is y = mx + b where m is the slope and b is the y intercept. For this exercise the equation of the straight line is
i D = mv D + b (1)
where
3 ID 10 1 m = ---- = --------- = --------------------------------------- = 0.02 r D nV T 1.92 26 10 3
Next, we find the i D intercept using the given data that when v D = 0.7 V , i D = 1 mA . Then, by substitution into (2) we obtain
10
3
= 0.02 0.7 + b
from which b = 0.013 and thus the equation of the straight line becomes
i D = 0.02v D 0.013 (3)
Now, from (3) above we find that the v D axis intercept is 0 = 0.02vD 0.013 or v D = 0.65 V .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
265
v load
v in
conduction angle (a) (b)
a. It is given that the transformer is a step down type with ratio 10:1 ; therefore, the secondary voltage is 12 V RMS or v in = 12 2 peak = 17 V peak . Then,
v in v D I peak = -----------------------r D + R load
c.
d.
From (2.17)
V p v 17 0.7 V load ( ave ) ----- ------ = ----- ------ = 5.06 V 2 2
e.
266
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v out ( t )
+
v out ( t ) v in ( t )
and since r D R b.
( )
1 ( V p sin 2v ) d = -- ( V p cos 2v )
or
Generally, the angle is small and thus cos 1 and ( 2 ) . Therefore, the last relation above reduces to
2V p 2 17 V out ( ave ) --------- 2v = -------------- 2 0.7 = 9.42 V
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
267
+
C
v in
v out
13.
C
+ v in +
iD
v in ( V ) t
+ v out
5
From the circuit above we observe that the diode conducts only when v in > 0 and thus the capacitor charges to the positive peak of the input which for this example is 5 V , and the capacitor voltage is v C = 5 V with polarity shown. Since v out = v in v C , when v in = 5 V ,
v out = 5 5 = 0 and the output has shifted upwards to zero volts as shown below. When the
input voltage is negative, v in = 5 V , and v out = 5 5 = 10 V . The waveform for the output voltage is as shown below.
v out ( V )
t
10
= VZ
I Z = 10mA
r Z I Z = 12 30 ( 10 5 ) 10
= 11.85 V
268
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
= VZ
I Z = 10mA
r Z I Z = 12 30 ( 10 20 ) 10
= 12.3 V
c. From (2.34) VZ = VZ0 + rZ IZ where VZ and IZ are the nominal values. Then
VZ0 = VZ rZ IZ = 12 30 10 10
3
= 11.7 V
15.
i ( mA )
IP
IV VP VV VF
v ( mV )
Tunnel diode
and thus
di D 1 ------- = 0.03 (1) dv
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
269
it follows that and (2) above will be satisfied if From (1) and (3)
and thus the maximum value of the resistor should be R 33.33 . With this value, the slope of the total current i versus the voltage v across the parallel circuit will be zero. For a positive slope greater than zero we should choose a resistor with a smaller value. Let us plot the total current i versus the voltage v with the values R 1 33.33 and R 2 = 25 . We will use the following MATLAB script for the plots.
v=51:0.1:349; m=0.03; b=11.5; id=m*v+b;... R1=33.33; i1=id+v/R1; R2=25; i2=id+v/R2; plot(v,id,v,i1,v,i2); grid
and upon execution of this script we obtain the plot shown below.
15 i for R=25 Ohms
10 i (mA)
i without resistor
0 50
100
150
200 v (mV)
250
300
350
270
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
a. By inspection, the operating point that will yield maximum power is denoted as Q where
P max = 400 40 = 16 mw
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
271
272
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Chapter 3
Bipolar Junction Transistors
his is a long chapter devoted to bipolar junction transistors. The NPN and PNP transistors are defined and their application as amplifiers is well illustrated with numerous examples. The small and large signal equivalent circuits along with the hparameter and Tequivalent circuits are presented, and the EbersMoll model is discussed in detail.
3.1 Introduction
Transistors are three terminal devices that can be formed with the combination of two separate PN junction materials into one block as shown in Figure 3.1.
N P P N P N N P
E Emitter
B Base C
C Collector
E Emitter
B Base C
C Collector
As shown in Figure 3.1, an NPN transistor is formed with two PN junctions with the Ptype material at the center, whereas a PNP transistor is formed with two PN junctions with the Ntype material at the center. The three terminals of a transistor, whether it is an NPN or PNP transistor, are identified as the emitter, the base, and the collector. Can a transistor be used just as a diode? The answer is yes, and Figure 3.2 shows several possible configurations and most integrated circuits employ transistors to operate as diodes.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
31
E Cathode
B Anode
E Anode
B Cathode
B Anode
C Cathode
B Cathode
C Anode
Transistors are used either as amplifiers or more commonly as electronic switches. We will discuss these topics on the next section. Briefly, a typical NPN transistor will act as a closed switch when the voltage V BE between its base and emitter terminals is greater than 0.7 V but no greater than 5 V to avoid possible damage. The transistor will act as an open switch when the voltage V BE is less than 0.6 V . Figure 3.3 shows an NPN transistor used as an electronic switch to perform the operation of inversion, that is, the transistor inverts (changes) an input of 5 V to an output of 0 V when it behaves like a closed switch as in Figure 3.3, and it inverts an input of 0 V to an output of 5 V when it behaves like an open switch as in Figure 3.4.
V CC RC V in
= 5V
V CC RC
= 5V
= V BE = 5 V
V in B
C V out E
= V CE = 0 V
Like junction diodes, most transistors are made of silicon. Gallium Arsenide (GaAs) technology has been under development for several years and its advantage over silicon is its speed, about six times faster than silicon, and lower power consumption. The disadvantages of GaAs over silicon is that arsenic, being a deadly poison, requires very special manufacturing processes and, in addition, it requires special handling since it is extremely brittle. For these reasons, GaAs is much more expensive than silicon and it is usually used only in superfast computers.
32
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
= 5V
V CC RC C
= 5V
= V BE = 0 V
V in B
C V out E
= V CE = 5 V
E
PNP Transistor
Figure 3.5. Biased NPN and PNP Transistors for proper operation
The bias voltage sources are V BB for the base voltage and V CC for the collector voltage. Typical values for V BB are about 1 V or less, and for V CC about 10 V to 12 V . The difference in these bias voltages is necessary to cause current flow from the collector to the emitter in an NPN transistor and from the emitter to collector in a PNP transistor.
33
iB
B
iC
E iE
iC iE
iB
E PNP Transistor
NPN Transistor
For any transistor, NPN or PNP, the three currents are related as iB + iC = iE We recall from Chapter 2, equation (2.3), that i D = I r [ e the collector current is iC = Ir e
v BE V T 12 15 ( v D nV T )
v BE is the basetoemitter voltage, and V T 26 mV at T = 300 K . A very useful parameter in transistors is the commonemitter gain , a constant whose value typically ranges from 75 to 300. Its value is specified by the manufacturer. Please refer to Section 3.19. The base current i B is much smaller than the collector current i C and these two currents are related in terms of the constant as iB = iC and with (3.2) we obtain v BE V T i B = ( I r )e From (3.1) and (3.3) we obtain +1 i E = i B + i C = i C + i C = ----------- i C and with (3.2) + 1 v V i E = ----------- I r e BE T (3.3) (3.4)
(3.5)
(3.6)
Another important parameter in transistors is the commonbase current gain denoted as and it
* It is customary to denote instantaneous voltages and currents with lower case letters and the bias voltages and currents with upper case letters.
34
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Another lesser known ratio is the commoncollector current gain ratio denoted as and it is defined as the ratio of the change in the emitter current to the change in the base current. Thus, di C = ------di E di C = ------di B di E = ------di B and their relationships are = ----------+1 Example 3.1 A transistor manufacturer produces transistors whose values vary from 0.992 to 0.995. Find the range corresponding to this range. Solution: For = 0.992 (3.10) yields 0.992 = ----------- = --------------------- = 124 1 1 0.992 and for = 0.995 , = ----------1 = +1 (3.14) (3.11) (3.12) (3.13)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
35
(3.15) (3.16)
i C = i B i E = i B + i C = ( I r )e
B
v BE V T
+ i B
C
(3.17)
iB v BE
iC
( I r )e
v BE V T
i C = i B
iE
Figure 3.7. NPN transistor equivalent circuit model for relations (3.15), (3.16), and (3.17)
If we know I r , , i B , and the operating temperature, we can find the other parameters for the circuit model in Figure 3.7. Example 3.2 For a given NPN transistor, I r = 2 10
14
numerical values of the parameters shown in Figure 3.7. The collector bias voltage V CC (not
* Dependent sources are discussed in detail in Chapters 1 through 4, Circuit Analysis I with MATLAB Applications, ISBN 9780970951120.
36
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v BE V T
= 10
16 v BE V T
10 = 5 10 - = 5 10 ----------------- 16 10 10
v BE V T = ln ( 5 10 ) v BE = V T ( ln 5 + 10 ln 10 ) = 26 10 ( 1.61 + 23.03 ) 0.64 V The collector bias voltage V CC is used for proper transistor operation and its value is not required for the above calculations.
3
iE ( I r )e
v EB V T
v EB
i C = i B iC
C
iB
Figure 3.8. PNP transistor equivalent circuit model for relations (3.15), (3.16), and (3.17)
For easy reference we summarize the currentvoltage relationships for both NPN and PNP transistors in the active mode in Table 3.1.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
37
iE = iB + iC
v BE
iC = ----------- = ---iE +1
v EB
iC = ----------- = ---iB 1
iE = iB + iC I r --------V i E = -- e T iB = iC i E = ( + 1 )i B
v EB
I r --------V i B = -- e T iB = iC iB = iE ( + 1 ) iB = ( 1 ) iE
v BE
iC = Ir e
v BE -------VT
I r --------I r --------V V i E = -- e T i B = -- e T iB = iC iB = iC
iC = Ir e
v EB -------VT
iC = iB iC = iE
VT
iC = iB iC = iE
VT
i E = ( + 1 )i B i B = i E ( + 1 ) iB = ( 1 ) iE
= 26 mV at T = 27 C
= 26 mV at T = 27 C
v BE = V T [ ln ( ) ln ( I r ) + ln ( i B ) ]
v BE = V T [ ln ( ) ln ( I r ) + ln ( i B ) ]
The relations in Table 3.1 are very useful in establishing voltage and current levels at various points on an NPN or PNP transistor. Example 3.3 An NPN transistor with = 150 is to operate in the common (grounded) base configuration. A DC power supply at V S = 12 V is available and with two external resistors, one connected between the collector and V CC and the other between the emitter and V EE , we want to keep the collector current I C * at 1.6 mA and the collector voltage V C at 4 V . Find the values of the resistors given that when V BE = 0.7 V , I C = 1.2 mA . The circuit operates at T = 27 C . Solution: Since the transistor is to operate at the common base configuration, after connecting the resistors and the bias voltages, our circuit is as shown in Figure 3.9.
IC RC V CC 12 V
= 1.6 mA
C VC E
IE RE 12 V V EE
= 4V
B IB
38
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
+ V C = V CC
We are given that for V BE = 0.7 V , I C = 1.2 mA and we need to find V BE at I C = 1.6 mA . We find V BE from the ratio
BE T ( V 0.7 ) V T 1.6 mA = I r e ------------------------------------- = e BE 0.7 V T 1.2 mA Ir e
1.6 ln ------ = ( V BE 0.7 ) V T 1.2 1.6 3 V BE = 0.7 + 26 10 ln ------ = 0.708 1.2 Next, V BE = V B V E
and since the base is grounded, V B = 0 , and V E = V BE = 0.708 as shown in Figure 3.10.
RC IC V CC C E RE IE
= 1.6 mA
VC
12 V
= 4V
IB
VE B
= 0.708 V
12 V
V EE
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
39
v BE
Characteristics
As with diodes, the baseemitter voltage v BE decreases approximately 2 mV for each 1 C rise in temperature when the emitter current i E remains constant. Example 3.4 For the PNP transistor circuit in Figure 3.11, at T = 27 C the emitter to base voltage v BE = 0.7 V and the emitter current I E is held constant at all temperature changes. Find the changes in emitter voltage V E and collector voltage V C if the temperature rises to T = 50 C .
RC
= 5 K C
IC VC
RE IE
= 10 K
V CC 12 V
B IB
VB
VE
V EE
Solution: Since the base is grounded, v EB = V E and since this voltage decreases by 2 mV for each 1 C temperature rise, the change in V E is V E = ( 50 27 ) C ( 2 mV 1 C ) = 46 mV and at T = 50 C ,
VE
There is no change in the collector voltage V C because the emitter current I E is held constant, and since I C = I E , the voltage V C remains unchanged.
310
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
The Bipolar Junction Transistor as an Amplifier 3.3.4 Collector Output Resistance Early Voltage
From basic circuit analysis theory, we recall that a current source has a parallel resistance attached to it and it is referred to as output resistance. Ideally, this resistance should be infinite and we can connect any passive load to the current source. However, most integrated circuits use transistors as loads instead of resistors R C , and when active loads (transistors) are used, we should consider the finite output resistance that is in parallel with the collector. This resistance is in the order of 100 K or greater. This output resistance looking into the collector is defined as v CE r out = ----------i C or (3.18)
vBE = cons tan t
V IC
A r out = -----
(3.19)
where V A is the Early voltage supplied by the manufacturer, and I C is the DC collector current. Before we consider the next example, let us illustrate a transistor in the commonemitter mode with the resistive circuit shown in Figure 3.12.
V CC RC C B E VC
When the potentiometer resistance is decreased (the wiper moves upwards) the current through the collector resistor R C increases and the voltage drop across R C increases. This voltage drop subtracts from the supply voltage V CC and the larger the voltage drop, the smaller the voltage V C at the collector. Conversely, when the potentiometer resistance is increased (the wiper moves downwards) the current through the collector resistor R C decreases and the voltage drop across R C decreases. This voltage drop subtracts from the supply voltage V CC and the smaller the voltage drop, the larger the voltage V C at the collector. We recall also that a resistor serves as a current limiter and it develops a voltage drop when current flows through it. A transistor is a currentin, currentout device. We supply current to the base of the transistor and current appears at its collector. The current into the base of the transistor is in the order of a few microamps while the current at the collector is in the order of a few milElectronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
311
vC iC
EE
iB vS vS
RB B
vC
v BE
For the circuit of Figure 3.13, for the basetoemitter voltage interval 0 v BE 0.65 , no current flows into the base. But when the basetoemitter voltage is v BE 0.7 V , a small current flows into the base. A further increase in the supply voltage v S has no effect on v BE which remains fairly constant at 0.7 V , but the base current continues to increase causing an increase in the collector current i C . This, in turn causes the voltage drop across the resistor R C to increase and thus the collector voltage v C decreases. and for this reason the output voltage v C appears as
180 outofphase with the input (supply) voltage v S . As v S decreases, less current flows into
the base and the collector current decreases also causing the voltage drop across the resistor R C to decrease, and consequently the collector voltage v C increases. Example 3.5 The datasheet for the NPN transistor in Figure 3.14 indicates that the Early voltage is V A = 80 . The base to emitter voltage V BE is held constant at 0.7 V and when V CC is adjusted so that
V CE
to 10 V and plot I C versus V CE . Do you expect a linear relationship between I C and V CE ? Solution: Relations (3.18) and (3.19) are applicable here. Thus, VA 80 r out = ------ = ------ = 100 K IC 0.8
312
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
RC IB VS RB B C IC E V CE V CC
V BE
and
v CE 5 i C = ----------- = 10 v CE r out
This expression shows that there is a linear relationship between I C and V CE . The MATLAB script below performs all computations and plots I C versus V CE . The plot is shown in Figure 3.15.
vCE=1: 10; vA=80; iC1=0.8*10^(3); r0=vA/iC1; iC=zeros(10,3);... iC(:,1)=vCE'; iC(:,2)=(vCE./r0)'; iC(:,3)=(iC1+vCE./r0)'; fprintf(' \n');... disp('vCE delta iC new iC');... disp('-------------------------');... fprintf('%2.0f \t %2.2e \t %2.2e\n',iC');... plot(vCE,iC(:,3)); xlabel('vCE (V)'); ylabel('iC (A)'); grid;... title('iC vs vCE for Example 3.5')
vCE delta iC new iC ------------------------1 1.00e-005 8.10e-004 2 2.00e-005 8.20e-004 3 3.00e-005 8.30e-004 4 4.00e-005 8.40e-004 5 5.00e-005 8.50e-004 6 6.00e-005 8.60e-004 7 7.00e-005 8.70e-004 8 8.00e-005 8.80e-004 9 9.00e-005 8.90e-004 10 1.00e-004 9.00e-004 Generally, the i C versus v CE relation is nonlinear. It is almost linear when a transistor operates in the active region, and nonlinear when it operates in the cutoff and saturation regions. Table 3.2 shows the three modes of operation in a bipolar transistor and the forward or reversebiasing of the emitterbase and collectorbase junctions.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
313
5 6 vCE (V)
10
TABLE 3.2 Region of operation for bipolar transistors Region of Operation EmitterBase junction Collectorbase junction
Active Saturation Cutoff Forward Forward Reverse Reverse Forward Reverse
Example 3.6 For the circuit in Figure 3.16, = 120 and V BE = 0.7 . Find V E , I E , I C , V C , and determine whether this circuit with the indicated values operates in the active, saturation, or cutoff mode. Solution: From the given circuit, by observation V E = V B V BE = 5 0.7 = 4.3 V and VE 4.3 I E = ------ = --------------------- = 1.72 mA 3 RE 2.5 10 = ----------- = 120 = 0.992 -------+1 121
314
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
RC RB VS B IB C
V CC
V BE VB
= 5V
IE
RE 2.5 K
VE
and Then,
Finally, we find the collector voltage V C as V C = V CC I C R C = 10 1.71 10 4 10 = 3.16 V We observe that the transistor is an NPN type and for the active mode operation the basecollector junction PN must be reversebiased. It is not because V C < V B and thus we conclude that with the given values the transistor is in saturation mode. Example 3.7 For a PNP transistor circuit with the base grounded, it is given that V EE = 12 V , V CC = 12 V , R E = 5 K , R C = 3 K ,and = 150 . Find V E , I E , I C , V C , and I B . Is the circuit operating in the active mode? Solution: The circuit is as shown in Figure 3.17. With V EE = 12 V it is reasonable to assume that the emitterbase junction is forwardbiased and since the base is grounded, we have V E = V EB = 0.7 V and
V EE V E 12 0.7 I E = --------------------- = ----------------- = 2.3 mA 3 RE 5 10
3 3
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
315
5 K
IE V EE
VE
12 V
IC VC C
3 K V CC 12 V
RC
= 2.28 mA
3
V C = R C I C V CC = 3 10 2.28 10 12 = 5.16 V
With this value of V C , the collectorbase junction is reversebiased and the PNP transistor is in the active mode. The base current is I B = I E I C = 2.3 10 2.28 10 Example 3.8 For the circuit in Figure 3.18, it is known that = 120 . Find I B , I E , I C , V B , V C , and V E . Is the transistor operating in the active mode? Solution: To simplify the part of the circuit to the left of the base, we apply Thevenins theorem at points x and y as shown in Figure 3.19, and denoting the Thevenin equivalent voltage and resistance as V TH and R TH respectively, we find that R2 30 V TH = V xy = ------------------ V S = ----------------- 12 = 4 V 60 + 30 R1 + R2
3 3
= 20 A
316
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
RC
8 K
IC VC IE RE
5 K
C
R1 60 K 30 K 12 V IB B V BE R2 VB E
12 V
V CC
VS
VE
x
V TH
R TH 20 K
4V
and
R1 R2 60 30 R TH = ------------------ = ----------------- = 20 K R1 + R2 60 + 30
RC
8 K IC V CC
C
R TH I B B V TH 4V 20 K VC V BE I E E VB RE VE 5 K 12 V
Figure 3.20. The circuit in Figure 3.18 after application of Thevenins theorem
Application of KVL around the left part of the circuit of Figure 3.20 yields R TH I B + V BE + R E I E = V TH ( 20 10 )I B + 0.7 + ( 5 10 )I E = 4 Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
3 3
317
From Table 3.1, I E = ( + 1 )I B . Then, 4I B + ( + 1 )I B = 0.66 10 125I B = 0.66 10 I B = 5.28 10 and Next, and Also, and
6 3 3
A = 5.28 A
6
I E = ( + 1 )I B = 121 5.28 10
3
= 0.639 mA
3
V E = R E I E = 5 10 0.639 10
= 3.2 V
= 0.634 mA
3
V C = V CC R C I C = 12 8 10 0.634 10
= 6.93 V
Since this is an NPN transistor and V C > V B , the basecollector PN junction is reversebiased and thus the transistor is in active mode.
318
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
+
Vp 0
Vp
v in = Vp sin t
Figure 3.21. The basic NPN transistor amplifier biased with a resistive network
With the transistor properly biased, direct current flows continuously, with or without an input signal, throughout the entire circuit. The direct current flowing through the circuit develops more than just base bias; it also develops the collector voltage V C as it flows from V CC through resistor R C and, as we can see on the output graph, the output signal starts at the V C level and either increases or decreases. These DC voltages and currents that exist in the circuit before the application of a signal are known as quiescent voltages and currents (the quiescent state of the circuit). The DC quiescent point is the DC bias point Q with coordinates ( V BE, I B ) . We will discuss the Q point in detail in a later section. The collector resistor R C is placed in the circuit to keep the full effect of the collector supply voltage off the collector. This permits the collector voltage V C to change with an input signal, which in turn allows the transistor to amplify voltage. Without R C in the circuit, the voltage on the collector would always be equal to V CC . The coupling capacitor C is used to pass the ac input signal and block the dc voltage from the preceding circuit. This prevents DC in the circuitry on the left of the coupling capacitor from affecting the bias on the transistor. The coupling capacitor also blocks the bias of the transistor from reaching the input signal source. The input to the amplifier is a sine wave that varies a few millivolts above and below zero. It is introduced into the circuit by the coupling capacitor and is applied between the base and emitter. As the input signal goes positive, the voltage across the baseemitter junction becomes more posiElectronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
319
RB
+
V CC VC 0V v out
+
Vp 0
Vp
v in = V p sin t
When the PNP input signal goes positive, it opposes the forward bias of the transistor. This action cancels some of the negative voltage across the emitterbase junction, which reduces the current through the transistor. Therefore, the voltage across R C decreases, and the voltage
320
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Fixed Bias
across the transistor increases. Since V CC is negative, the voltage on the collector V C goes in a negative direction toward V CC . Thus, the output is a negative alternation of voltage that varies at the same rate as the sine wave input, but it is opposite in polarity and has a much larger amplitude. During the negative alternation of the input signal, the transistor current increases because the input voltage aids the forward bias. Therefore, the voltage across R C increases, and consequently, the voltage across the transistor decreases or goes in a positive direction. This action results in a positive output voltage, which has the same characteristics as the input except that it has been amplified and the polarity is reversed. In summary, the input signals in the preceding circuits were amplified because the small change in base current caused a large change in collector current. And, by placing resistor R C in series with the collector, voltage amplification was achieved.
R1
RC
C
B R2 E RE
Following the procedure of Example 3.8 we can simplify the circuit in Figure 3.23 with the use of Thevenins theorem to the circuit in Figure 3.24.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
321
RC C IB B V BE RE IE IC
+
V BB
RB
Figure 3.24. The circuit in Figure 3.23 after application of Thevenins theorem
With reference to Figures 3.23 and 3.24 we obtain the following relations: R2 V BB = ------------------ V CC R1 + R2 R1 R2 R B = -----------------R1 + R2 (3.20) (3.21)
It was stated earlier that it is imperative to keep variations in temperature and changes in values to a minimum and this can be achieved by making the emitter current I E fairly constant. Therefore, let us now derive an expression for I E . From Figure 3.24 with application of KVL we obtain R B I B + V BE + R E I E = V BB and from Table 3.1 Then, or 1 I B = ----------- I E +1 1 R B ----------- I E + V BE + R E I E = V BB + 1 V BB V BE V BB 0.7 I E = ---------------------------------------- = ---------------------------------------RE + RB ( + 1 ) RE + RB ( + 1 ) (3.22)
From expression (3.22) we see that the emitter current I E will be fairly constant if V BB 0.7 and R E R B ( + 1 ) . Accordingly, R B must be small and since R B = R 1 R 2 ( R 1 + R 2 ) , we should use small resistance values for R 1 and R 2 . Designers recommend that the sum of R 1 and R 2 is
322
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Fixed Bias
such that the current through them (assuming that the base current is zero) is about half of the emitter current I E . It is also recommended that V BB , V CB or V CB , and the product R C I C each be close to onethird of the value of V CC . It is often said that the emitter resistor R E provides a negative feedback action which stabilizes the bias current. To understand how this is done, let us assume that the emitter current I E increases. In this case, the voltages R E I E and V E will also increase. But the base voltage being determined by the voltage division provided by resistors R 1 and R 2 will remain relatively constant and according to KVL, the base to emitter voltage V BE will decrease. And since IC = Ir e
V BE V T
, both the collector current I C and emitter current I E will decrease. Obviously,
this is a contradiction to our original assumption (increase in I E ) and thus we say that the emitter resistor R E provides a negative feedback action. Let us consider the transistor circuit in Figure 3.23 which is repeated as Figure 3.25 for convenience.
RC C IR IB R2 B RB B IB V BE RE E IE V CC
+
RC C V CC
R1
0
E RE IE
+
V BB
Circuit designers maintain that if we specify the V CC voltage, the I E current, and the value of , we can determine the appropriate values of the four resistors for proper biasing by letting the values of V BB , V CB or V CB , and the product R C I C each be close to onethird of the value of V CC , and assuming that the base current is negligible, the current I R through resistors R 1 and R 2 is I R = 0.5I E . For convenience, we will denote the sum of R 1 and R 2 as R eq . Example 3.9 For the transistor circuit of Figure 3.26 = 120 . Find the values of the four resistors for appropriate fixed biasing. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
323
R1
RC
15 V
C
B R2 E RE
IE
Solution:
The given circuit and its Thevenin equivalent are shown in Figure 3.27. The Thevenin equivalent is shown just to indicate the value of V BB which will be used in the calculations.
RC C IR IB R2 B RB B IB V BE RE E IE V CC
15 V
R1
RC C IC
V CC
V CC
0
E RE IE
+
V BB
Figure 3.27. The transistor circuit in Example 3.9 and its Thevenin equivalent
As stated above, it is suggested that the values of V BB , V CB or V CB , and the product R C I C each be close to onethird of the value of V CC , and assuming that the base current is negligible, the current I R through resistors R 1 and R 2 is I R = 0.5I E . Then, 1 1 V BB = -- V CC = -- 15 = 5 V 3 3 and with reference to the Thevenin equivalent circuit above, since I B 0 by KVL V BE + R E I E = V BB
324
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
SelfBias
V BB V BE 5 0.7 R E = ------------------------- = ------------------ = 4.3 K 3 IE 1 10 Also, R eq I R = ( R 1 + R 2 )I R = V CC V CC V CC 15 R eq = ( R 1 + R 2 ) = --------- = ----------- = ----------------------- = 300 K 3 IR 0.5I E 0.5 10 To find R eq = ( R 1 + R 2 ) we use the Thevenin equivalent voltage expression, that is, R2 R2 R2 V TH = V BB = 5 V = ------------------ V CC = ------------------ 15 V = ------------------- 15 V R1 + R2 R1 + R2 300 K R2 5 V------------------- = ----------- = 1 -300 K 15 V 3 and thus and R 2 = 100 K R 1 = 300 K R 2 = 300 K 100 K = 200 K 1 R C I C = -- V CC 3 or ( 1 3 )V CC ( 1 3 )V CC ( 1 3 )V CC 5 R C = ------------------------ = ------------------------ = --------------------------------- = ---------------------------------------------------- = 4.96 K 3 IC I E ( ( + 1 ) )I E ( 120 121 ) 1 10
3.6 SelfBias
The fixed bias arrangement discussed in the previous section is thermally unstable. If the temperature of the transistor rises for any reason (due to a rise in ambient temperature or due to current flow through it), the collector current will increase. This increase in current also causes the DC quiescent point to move away from its desired position (level). This reaction to temperature is undesirable because it affects amplifier gain (the number of times of amplification) and could result in distortion, as we will see later in this chapter. A better method of biasing, known as self bias is obtained by inserting the bias resistor directly between the base and collector, as shown in Figure 3.28.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
325
RC IB RB B E
+ IC = IE
IC
IE
By tying the collector to the base in this manner, feedback voltage can be fed from the collector to the base to develop forward bias. Now, if an increase of temperature causes an increase in collector current, the collector voltage V C will fall because of the increase of voltage produced across the collector resistor R L . This drop in V C will be fed back to the base and will result in a decrease in the base current. The decrease in base current will oppose the original increase in collector current and tend to stabilize it. The exact opposite effect is produced when the collector current decreases. From Figure 3.28, R C I E + R B I B + V BE = V CC 1 R C I E + R B ----------- I E + V BE = V CC +1 V CC V BE I E = ---------------------------------------RC + RB ( + 1 ) collector and base resistors such that R B ( + 1 ) R C . Example 3.10 For the transistor circuit in Figure 3.29 = 120 and we want the collector voltage to vary in accordance with V C = 3 sin t V + V BE and I E = 1 mA . Find the values of R C and R B to meet these specifications. (3.23)
From (3.23) we see that to maintain the emitter current I E fairly constant, we should choose the
326
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
SelfBias
15 V V CC
RC
RB B
Solution:
+ IC = IE
IC
IE
Figure 3.30. Circuit for Example 3.10 with assigned current directions
By inspection, 1 V C = R B I B + V BE = R B ----------- I E + 0.7 = 3 sin t + 0.7 = 3 + 0.7 + 1 or Also, by inspection 3 R B = ---------------------------------------------- = 363 K 3 1 10 ( 1 121 ) V C = V CC R C I E V CC V C 15 3.7 R C = ---------------------- = ------------------ = 11.3 K 3 IE 1 10
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
327
328
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
RC C
RC C B R tr
v out
B R tr
=
E RE
R load
v out
= 0
E RE
R load
v out
V EE
V EE
(a)
(b)
Figure 3.31. Simple amplifier with resistive load
(c)
Application of the voltage division expression for the circuit in Figure 3.31(b) yields the maximum voltage across the load resistor, that is,
load v out ( max ) = ------------------------ V CC
R C + R load
(3.24)
and the corresponding maximum current is V CC i out ( max ) = -----------------------R C + R load (3.25)
Obviously, we want the power delivered to the load resistor to be maximum; therefore, we must make the collector resistor R C much smaller than the load resistor R load , that is, R C R load to maximize the current through the load resistor. For convenience, we choose R C to be onetenth of the value of R load , that is, R C = 0.1 R load (3.26) Next, by application of Thevenins theorem for the circuit of Figure 3.31(c) where the load resistor is disconnected, we redraw the circuit as shown in Figure 3.32.
R TH RC V CC RE I V EE V TH V TH R load
v out
From Figure 3.32, Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
329
R E V CC R C V EE = V CC R C I = -------------------------------------RC + RE RC RE R TH = ------------------RC + RE
and thus
R TH + R load
From (3.29) we see that for v out ( min ) to be close to V EE , the Thevenin resistance R TH shown in (3.28) and (3.29) should be minimized. We already have minimized R C to be R C = 0.1 R load , so let us make R E onetenth of R C , that is, R E = 0.1R C , or R E = 0.01R load (3.30)
Example 3.11 Let us suppose that the circuit of Figure 3.33 is to be used as the output stage of an audio system whose load resistance is R load = 8 and the supply voltages are V CC = +24V and
V EE = 24V . Taken into account the recommendations discussed above for sizing the emitter
and collector resistors, find the power absorbed by the combination of the collector resistor, the transistor, and the emitter resistor when v out = 0 V and the amplifier is on.
RC C V CC
v in
B E RE
R load
v out
V EE
330
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Class A
Class
Class AB
Class
Figure 3.34. Output signals for Class A, Class B, Class AB, and Class C amplifiers
We should remember that the circuits presented in our subsequent discussion are only the output stages of an amplifier to provide the necessary drive to the load.
331
RB
+
V CC v out
+
Vp 0
Vp
R E1
i1 i load v in2
+
V EE
v in1
+ R load v out
R E2
i2 +
V EE
From the circuit in Figure 3.36, it is evident that we can control the currents i 1 and i 2 by varying the input voltages v in1 and v in2 . It is convenient to set the quiescent current, denoted as I Q , to onehalf the maximum current drawn by the load. Then, we can adjust the currents i 1 and i 2 to be equal and opposite. Then,
* The expression pushpull (or double ended) derives its name from the fact that one of the two transistors pushes (sources) current into the load during the positive cycle while the other pulls (sinks) current from the load during the negative cycle.
332
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
The currents i 1 and i 2 in each transistor vary from 0 to 2I Q ; therefore, the load current is within the range 2I Q i load 2I Q . Example 3.12 In the circuit in Figure 3.37, + V EE = 24 V , V EE = 24 V , and R load = 8 . Compute the power absorbed by the circuit if we want to apply up to 24 V to the 8 load. R E1
i1 i load v in2 +
24 V
V EE
v in1
R load +
8
v out
R E2
24 V
i2 +
V EE
Solution: The maximum load current is and the quiescent current is Then, each transistor will absorb
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
333
I bias
R C1
v D1 v in v D2
+ v BE1 v1
R E1
i load
R E2 R load
v EB2 +
I bias R C2
v2
i1
v out
+ V CC
When v in goes positive, the upper transistor conducts and the lower transistor is cutoff. Then, the input to the base of the upper transistor is v in + v D1 and the voltage at the emitter terminal of the upper transistor is v 1 = v in + v D1 v BE1 , and since v BE1 = v D1 , we find that v 1 = v in for v in > 0 (3.33)
When v in goes negative, the upper transistor is cutoff and the lower transistor conducts. then the input to the base of the lower transistor is v in v D2 and the voltage at the emitter terminal of the lower transistor is v 2 = v EB2 + v in v D2 , and since v EB2 = v D2 , we find that
334
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Therefore, when v in = 0 , v 1 = 0 , and v 2 = 0 also, and the current and the output of both transistors including the quiescent currents I bias will be zero also, and the power absorbed by the circuit will be zero. Accordingly, it appears that this arrangement has perfect efficiency. However, this ideal condition is never achieved because no two diodes or two transistors are exactly identical. There exists a range where both transistors are cutoff when the input signal changes polarity and this results in crossover distortion as shown in Figure 3.39. This distortion is due to the nonlinearities in transistor devices where the output does not vary linearly with the input.
Crossover distortion
v in
v out
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
335
R C1 I bias
V CC
+ v in
v D1 v D2 v D3 v D4
+ v BE1 v1
A R1 R adj R2 B
+ + v BE1
v1 i load
R E1
i load + R load v out
R E1 v AB R E2
v EB2 + v2 i1
v in
R E2
v2 v EB2 + i 1
R load
v out
I bias
R C2
I bias
(a)
V + CC
V + CC
(b)
Figure 3.40. Two possible arrangements for the output stage in a typical Class AB amplifier
For small output signals that require currents in the range 2 I bias < i load < 2 I bias both transistors will conduct and will behave as Class A amplifier. But for larger signals, one transistor will conduct and supply the current required by the load, while the other will be cutoff. In other words, for large signals the circuit of Figure 3.40(a) acts like a Class B amplifier and hence the name Class AB amplifier. For sinusoidal signals into the load R load , the RMS voltage will be
D v out ( RMS ) = ------------ R load
2v RE
(3.38)
2v D = -------- R load 2 RE
(3.39)
For maximum power, we should make R E as small as possible, so let R E = 0.5 , and let R load = 8 and v D = 0.5 V . Then,
336
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
If we let V CC = 24 V and I bias = 1 A , the total power absorbed will be P total = 24 1 + 20 = 44 w For the circuit in Figure 3.40(b), the voltage v AB can be adjusted to any desired value by selecting appropriate values for resistors R 1 and R 2 . The adjustable resistor R adj is set to a position to yield the desired value of the quiescent current I bias . From the above discussion, we have seen that the Class AB amplifier maintains current flow at all times so that the output devices can begin operation nearly instantly without the crossover distortion in Class B amplifiers. However, complete current is not allowed to flow at any one time thus avoiding much of the inefficiency of the Class A amplifier. Class AB designs are about 50 percent efficient (half of the power supply is power is turned into output to drive speakers) compared to Class A designs at 20 percent efficiency. Class AB amplifiers are the most commonly used amplifier designs due to their attractive combination of good efficiency and highquality output (low distortion and high linearity close to but not equal to Class A amplifiers).
v in
Load v BE iC + iL
iC
In Class C amplifier operation, the transistor in Figure 3.41 behaves as a openclosed switch. The load can be thought of as an antenna. During the positive halfcycle the transistor behaves like a closed switch, current i L flows through the inductor and creates a magnetic field, and at the same time the capacitor discharges and thus the two currents i C + i L flow through the emitter to the Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
iL
V CC
337
iC
+ V CC
v in
V BB
i B v BE
+
v CE
iE
Figure 3.42. Circuit showing the variables used in the graphs in Figures 3.43 and 3.44
We start with a plot of i B versus v BE to determine the point where the curve i B = ( I r )e and the equation of the straight line intersect V BB v BE i B = -----------------------RB (3.41)
v BE nV T
(3.40)
The equation of (3.41) was obtained with the AC source v in shorted out. This equation can be expressed as V BB 1 i B = ------ v BE + --------(3.42) RB RB We recognize (3.42) as the equation of a straight line of the form y = mx + b with slope 1 R B . This equation and the curve of equation (3.40) are shown in Figure 3.43.
338
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Graphical Analysis
1 x 10
-6
0.8
Base current iB, amps
V BB R B Slope = 1 R B ( V BE, I B ) IB V BE V BB
0.6
0.4
0.2
0.2
0.4
0.6
0.8
Figure 3.43. Plot showing the intersection of Equations (3.40) and (3.42)
We used the following MATLAB script to plot the curve of equation (3.40).
vBE=0: 0.01: 1; iR=10^(15); beta = 100; n=1; VT=27*10^(3);... iB=(iR./beta).*exp(vBE./(n.*VT)); plot(vBE,iB); axis([0 1 0 10^(6)]);... xlabel('BaseEmitter voltage vBE, volts'); ylabel('Base current iB, amps');... title('iBvBE characteristics for the circuit of Figure 3.42'); grid
From Figure 3.43 we obtain the values of V BE and I B on the v BE and i B axes respectively. Next, we refer to the family of curves of the collector current i C versus collectoremitter voltage v CE for different values of i B as shown in Figure 3.44. where the straight line with slope 1 R C is derived from the relation (3.43) v CE = V CC R C i C
V CC
-------RC
iC
Slope = 1 R C
i Bn i B4 i B3 iB = IB i B2 i B1
V CE V CC
IC
v CE
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
339
As shown in Figure 3.44, this straight line, commonly known as load line, and the curve i B = I B intersect at point Q whose coordinates are the DC bias values V CE and I C . Obviously, the value of the collector resistor R C must be chosen such that the load line is neither a nearly horizontal nor a nearly vertical line. Example 3.13 For the circuit in Figure 3.45, the input voltage v i is a sinusoidal waveform. Using the i B versus v BE and i C versus v CE curves shown in Figure 3.46, sketch the waveforms for v BE , i B , v CE , and iC .
RC RB
iC
+ V CC
v in
V BB
i B v BE
+
v CE
iE
Solution: Let v in = V p sin t . We draw three parallel lines with slope 1 R B , one corresponding to input v in = 0 , the second at v in = V p , and the third at v in = V p as shown in Figure 3.47. The input voltage v in is superimposed on the DC bias voltage V BB .
* We observe that (3.40) describes an equation of a straight line of the form y = mx + b where the slope is m = 1 RC
and the ordinate axis intercept is b = VCC RC .
340
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Graphical Analysis
iB iC
Slope = 1 R B V CC Slope = 1 R C
V BB RB
--------
-------RC
IB
Q v BE
V BE V BB
IC
iB = IB
V CE
V CC
v CE
iB
V BB
-------RB
Slope = 1 R B
IB
iB
v BE
V BE
v BE
V BB
v in
Figure 3.47. Graphical representation of v BE and i B when the input voltage v in is a sinusoid
Figure 3.48 is the graphical representation of v CE and i C when the input voltage v in is a sinusoid.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
341
-------RC
iC iB = IB
IC
V CE
V CC
v CE
v CE
Figure 3.48. Graphical representation of v CE and i C when the input voltage v in is a sinusoid
342
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
iC
+ V CC
v in
V BB
i B v BE
+
v CE
iE
The power drawn from the collector supply at each instant is p CC = V CC i C = V CC I C + V CC i c (3.47)
Since V CC is constant and i c has zero average value, the average value of the term V CC i c is zero.* Therefore, the average power drawn from the collector supply is P CC = V CC I C (3.48)
and if there is negligible or no distortion, the current I C and power P CC are both independent of the signal amplitude. The power absorbed by the load at each instant is P LOAD = R LOAD i C = R LOAD ( I C + i c ) = R LOAD I C + 2R LOAD I C i c + R LOAD i c
2 2 2 2
(3.49)
The term 2R LOAD I C i c is zero since 2R LOAD I C is constant and the average of i c is zero; hence the average power absorbed by the load is P LOAD = R LOAD I C + R LOAD ( i c ) ave = R LOAD I C + R LOAD ( i c RMS ) The power absorbed by the transistor at each instant is p C = v CE i C = ( V CC R LOAD i C )i C = V CC i C R LOAD i C = p CC p LOAD Therefore, the average power absorbed by the transistor is P C = P CC P LOAD (3.52)
2 2 2 2 2
(3.50)
(3.51)
* By definition Average = Area Period = i dt T and thus if the value of the integral is zero, the average is also 0 zero.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
343
Chapter 3 Bipolar Junction Transistors 3.10 PiecewiseLinear Analysis of the Transistor Amplifier
The circuit shown in Figure 3.50 is a model to represent the transistor where the two ideal diodes are included to remind us of the two PN junctions in the transistor.
I CO
iB
iC
i E iE
In Figure 3.50, I CO is the current into the reversebiased collector with i E = 0 . By KCL, i C = i E + I CO and since i E = i C + i B i C = ( i C + i B ) + I CO i C i C = i B + I CO ( 1 )i C = i B + I CO or From Table 3.1
----------- = 1
(3.53) (3.54)
1 i C = ---------------- i B + ---------------- I CO (1 ) (1 )
(3.55)
Also,
----------1
+1 = +1
1--------- = + 1
1
and by substitution into (3.55) i C = i B + ( + 1 )I CO Thus, the equivalent circuit in Figure 3.50 may be redrawn as shown in Figure 3.51. (3.56)
344
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
iB
i B
iC
iE
E
The transistor models shown in Figures 3.50 and 3.51 are essentially ideal models. An improved transistor model is shown in Figure 3.52 where for silicon type of transistors V D 0.7 V and r b , referred to as the base spreading resistance, is included to account for the small voltage drop in the base of the transistor. ( + 1 )I CO
B
rb
iB iE
E
i B
iC
VD
Analogous to the base resistance r b are the emitter diffusion resistance defined as v E r e = -------i E and the collector resistance v C r c = -------i C (3.58)
i E = cons tan t
(3.57)
i C = cons tan t
Typical values for r b are between 50 to 250 , for r e are between 10 to 25 , and r c is very high, in excess of 1 M . The model in Figure 3.52 is for an NPN transistors. It applies also to PNP transistors provided that the diodes, voltage polarities, and current directions are reversed. The transistor amplifier in Figure 3.53 can be analyzed by piecewise linear methods with the aid of the transistor model in Figure 3.52.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
345
iS
+
v BE
v CE
RC V CC
vS
V1
For convenience, we neglect the small effects of r b , V D , and ( + 1 )I CO in Figure 3.52 and thus the model is now as shown in Figure 3.54.
B
iB
i B
iC
iE
E
Figure 3.54. The piecewise linear model in Figure 3.52 where r b , V D , and ( + 1 )I CO are neglected
When the transistor in the amplifier in Figure 3.53 is replaced by the piecewise linear model in Figure 3.54, we obtain the circuit shown in Figure 3.55(a). iC iS
RS
+
CC = -------
B R1
i B
IB DE E
iC
DC C RC V CC
V RC
v CE
vS
V1
iE
(a)
is
= IB
(b)
CS i S = ----- I B
Figure 3.55. Piecewise linear model for the circuit in figure 3.53 and its current transfer characteristics
The current transfer characteristics are constructed by determining the points at which the diodes change from the conducting to the nonconducting state. The current I CS represents the collector saturation current. Let us now suppose that the source current i s reaches a value that causes the reverse voltage across the collector diode D C to become zero and the emitter diode D E is conducting and allows
346
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
RC
(3.59)
Next, let us now suppose that the source current i S becomes negative and reaches a value that causes the collector diode D C to become reversebiased and the current in the emitter diode D E to reach a zero value. When this occurs, there is no current in either diode, and the base terminal is at ground potential. This condition establishes the lefthand break on the transfer characteristics and its value is determined as follows: i B = i C = i B Therefore, with the base terminal at ground potential and i B = 0 , (3.59) reduces to iS = IB Example 3.14 A DC power supply with a transistor regulator is shown in Figure 3.56. The resistor R 1 provides a suitable current to sustain the breakdown condition of the Zener diode. Any change in supply voltage causes a compensating change in the voltage drop across the transistor from collector to emitter and the load voltage v load is thereby held constant in spite of changes in the input voltage or load resistance R load . The transistor parameters are = 100 , r b = 75 , V BE = 0.7 V and the current I CO into the reversebiased collector is negligible. Find the values of the load voltage v load , the collecttoemitter voltage v CE , and the power P C absorbed by the transistor. (3.61) (3.60)
But this equation is true only if = 1 . For 1 , this equation is satisfied only if i B = 0 .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
347
R1
+
R load + 100
v load
C VZ
VC
= 35 V
= 25 V
Solution: We replace the transistor with its piecewise linear model as shown in Figure 3.57. iC
+
V BE = 0.7 V iB + iE 75
= ( + 1 ) iB v load
R1 VZ
rb
35 V
i BR load
+
= 25 V
100
Figure 3.57. The piecewise linear model for the transistor in Figure 3.55
= r b i B + ( + 1 )R load i B
V Z V BE 25 0.7 24.3 i B = ---------------------------------------- = ----------------------------------- = -------------- = 2.4mA 75 + 101 100 10175 r b + ( + 1 )R load
= 24.24V
We observe that the load voltage is independent of the supply voltage of 35 V . This occurs because the collector can be represented as an ideal current source i B . However, if the supply voltage falls below the Zener voltage of 25 V , the collector base junction will no longer be reversebiased, and the voltage regulation action of the transistor will fail. Also, when this occurs, the breakdown state will not be sustained in the Zener diode. The collector current is i C = i B = 100 2.4 = 240 mA and the collectortoemitter voltage is
348
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
If i B and v CE are changed in small increments, the resulting increment in v BE can be expressed as v BE v BE v BE dv BE = ----------- di B + ----------- dv CE (3.64) i B v CE and the increment in i C can be written as i C i C i C di C = ------- di B + ----------- dv CE i B v CE (3.65)
The partial derivative in the first term of (3.64) has the dimensions of resistance and it is denoted as r n , and that in the second term is a dimensionless voltage ratio denoted as . It is also convenient to denote these derivatives in lower case letters with lower case subscripts. Then, (3.64) is expressed as v be = r n i b + v ce (3.66) Likewise, the partial derivative in the first term of (3.65) is a dimensionless current ratio denoted as , and that in the second term has the dimensions of conductance and it is denoted as g o . Then, (3.65) is expressed as
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
349
The relations of (3.66) and (3.67) along with i e = i b + i c suggest the circuit shown in Figure 3.58 known as the hybrid incremental network model for the transistor. It is referred to as hybrid model because of the mixed set of voltages and currents as indicated by the expressions of (3.66) and (3.67). ib ic
B
v be
rn v ce
E
i b ie
go
v ce
Figure 3.58. The hybrid incremental model for a transistor in the commonemitter configuration
The input resistance r n is the slope of the input voltage and current characteristics and it accounts for the voltage drop across the baseemitter junction. Likewise, the output conductance g o is the slope of the output current and voltage characteristics. The voltage amplification factor is related to the input characteristics caused by a change in v CE , and the current amplification factor is related to the output characteristics caused by a change in i B . Typical values for the parameters of relations (3.66) and (3.67) are r n = 2 K , = 5 10 , = 100 , and g o = 2 10
5 4 1
source v ce in Figure 3.58 can be replaced by a short circuit, and thus the model reduces to that shown in Figure 3.59.
B
v be i b rn
i b
E
i c v ce go
ie
Figure 3.59. The hybrid incremental model for the transistor with v ce = 0
350
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
ib rn is
Rs
ic v ce
E
i b ie
go
R load
v ce
Now, we let R eq represent the parallel combination r o = 1 g o and R load . Then, v ce = R eq i b and v ce = R eq i b (3.68) (3.69)
Hence, the voltage of the v ce is proportional to the current flowing through the source, and from this fact we can replace the voltage source with a resistance R eq , and thus the transistor incremental model in Figure 3.60 can be redrawn as shown in Figure 3.61.
B C
ib rn is
Rs
ic
R load
R eq
E
i b
go
R eq
v ce
Figure 3.61. The circuit in Figure 3.60 with the voltage source v ce replaced by the resistance R eq
The negative resistance R eq is always much smaller than r n and thus the net input resistance to the transistor is always positive. Therefore, the negative resistance R eq can be replaced by a short circuit, and assuming that the base current i b is unaffected by this assumption, the voltages and currents in the collector side of the circuit are not affected.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
351
The parameters r n , , , and g o are normally denoted by the h (hybrid) parameters* as r n = h 11 = h ie , = h 12 = h re , = h 21 = h fe , and g o = h 22 = h oe . These designations along with the additional notations v be = v 1 , i b = i 1 , v ce = v 2 , and i c = i 2 , provide a symmetrical form for the relations of (3.66)and (3.67) as follows: v 1 = h 11 i 1 + h 12 v 2 i 2 = h 21 i 1 + h 22 v 2 or v 1 = h ie i 1 + h re v 2 i 2 = h fe i 1 + h oe v 2 (3.72)
(3.73)
In (3.73) the subscript i denotes the input impedance with the output shortcircuited, the sub-
* For a detailed discussion of the z , y , h , and g parameters refer to Circuit Analysis II with MATLAB Applications, ISBN 978097095115-1.
352
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Parameter
h 11 h 12 h 21 h 22 Example 3.15
h ib h rb h fb h ob
h ie h 11 ( 1 + h 21 ) h re h 11 h 22 ( 1 + h 21 ) h 12 h fe h 21 ( 1 + h 21 ) h oe h 22 ( 1 + h 21 )
h ic h 11 ( 1 + h 21 ) h rc 1 h fc 1 ( 1 + h 21 ) h oc h 22 ( 1 + h 21 )
For the amplifier circuit in Figure 3.62 it is known that r n = h 11 = 2 K , = h 21 = 100 , = h 12 = 5 10 , and g o = h 22 = 2 10 A c = i load i s . i load is
Rs R1 2V 10 K
+ +
4 5
v CE
RC
2 K
+
16 V
V CC
vs
Solution: The incremental model of this transistor amplifier is shown in Figure 3.63 where
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
353
ib
B Rs R1
i load rn C
R load
10 K
is
R eq
E
i b
go
v ce
2 K R eq
Figure 3.63. The incremental model for the transistor circuit in Figure 3.62
and thus the magnitude of the equivalent resistance reflected into the input part of the circuit is R eq = 5 10 100 1.293 10 = 64.65 = 0.06465 K and since this is much smaller than r n = h 11 = 2 K , it can be neglected. The current gain A c can be found from the relation (3.71). Then, ro i load R1 10 50 A c = --------- = --------------------- --------------------------- = ----- 100 ----- = 80 ( R 1 + r n ) ( r o + R load ) 12 52 is
4 3
3.12 Transconductance
Another useful parameter used in small signal analysis at high frequencies is the transconductance, denoted as g m , and defined as di C g m = ----------dv BE (3.74)
iC = IC
and as a reminder, we denote timevarying quantities with lower case letters and lower case subscripts. Thus, the transconductance g m is the slope at point Q on the i C versus v BE characteristics at i C = I C as shown in Figure 3.64. An approximate value for the transconductance at room temperature is g m 40I C . This relation is derived as follows: From (3.2)
354
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(3.75) (3.76)
iC = IC
1- v V = I r ------ e BE T VT
iC
IC Q Slope = V BE
ic
gm
v BE
v BE
Figure 3.64. The transconductance g m defined
(3.77) (3.78)
g m 40i C
and thus we see that the transconductance is proportional to the collector current i C . Therefore, a transistor can be viewed as an amplifier with a transconductance of 40 millimhos for each milliampere of collector current.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
355
ib r' b r be
v be
v' be i b
E (a)
ic ro
v ce
ib v be r' b r be
v' be g m v' be
E
ic ro
v ce
ie
ie
(b)
In Figure 3.65(a) the input impedance r n is separated into two parts; one part, denoted as r' b , accounts for the ohmic base spreading resistance; the other part, denoted as r be , is a nonlinear resistance and accounts for the voltage drop v' be across the emitter junction. Thus, r n = r' b + r be From Figure 3.65(a) and thus v' be i b = -------r be i b = ----- v' be = g m v' be * r be (3.79) (3.80) (3.81)
and thus the model in Figure 3.65(a) can also be represented as that in Figure 3.65(b). From (3.76) or g m = r be r be = ----- = ---------gm 40i C (3.82) (3.83)
The incremental models we have discussed thus far are valid only for frequencies of about 2 MHz or less. For higher frequencies, the effects of junction capacitances must be taken into account. Figure 3.66 shows a model for the transistor at high frequencies referred to as hybrid model.
v ' be i di c = ------------dv ' be
= gm
v CE = cons tan t
v ce = 0
356
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
ib v be r' b
v' be
1
C2
ic ro
r be
C1
v ce C
g m v' be
E
ie
Figure 3.66. The hybrid model for the transistor at high frequencies
The capacitor C 1 represents the capacitance that exists across the forwardbiased emitter junction while the capacitor C 2 represents a much smaller capacitance that exists across the reverse biased collector junction. At low frequencies the capacitors act as open circuits and thus do not affect the transistor performance. At high frequencies, however, the capacitors present a relatively low impedance and thereby reduce the amplitude of the signal voltage v' be . This reduction in v' be causes in turn a reduction in the strength of the controlled source g m v' be and a reduction in the collector current i c . We can derive some useful relations by determining the shortcircuit collector current i c when a sinusoidal input current is applied between the base and emitter terminals. With the output shortcircuited, capacitor C 2 is in parallel with C 1 between Node 1 and ground, and the equivalent impedance is r be 1 j ( C 1 + C 2 ) r be Z be = r be || 1 j ( C 1 + C 2 ) = j ( C 1 + C 2 ) = ------------------------------------------------ = -------------------------------------------r be + 1 j ( C 1 + C 2 ) r be j ( C 1 + C 2 ) + 1 However, C 1 is typically 100 times as large as C 2 and it can be neglected. Then, r be Z be = --------------------------r be j C 1 + 1 and denoting the phasor* quantities with bolded capical letters, we obtain r be V ' be = Z be I b = --------------------------- I b r be j C 1 + 1 and g m r be I c = g m V ' be = --------------------------- I b r be j C 1 + 1 (3.86) (3.85)
* Phasors are rotating vectors and are used to represent voltages and currents in complex form. Impedances and admittances are complex quantities but not phasors. For a detailed discussion, refer to Circuit Analysis I, ISBN 978-0-9709511-2-0.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
357
Thus, if the amplitude of the input current is held constant as the frequency is increased, the amplitude of the collector current decreases and approaches zero at very high frequencies. The coefficient of I b in (3.86) must be a dimensionless constant; therefore the quantity 1 r be C 1 has the dimensions of frequency. Also, it is customary to represent C 1 as C e , and it is helpful to define a new symbol as 1 -----------r be C e By substitution of (3.88) into (3.87) we obtain I c = ------------------------- I b j + 1 and the magnitute of the collector current is I c = ---------------------------------- I b 2 ( ) + 1 (3.90) (3.89) (3.88)
Therefore, at the frequency = , the magnitude of the collector current I c is reduced to 1 2 times its lowfrequency value. Thus, serves as a useful measure of the band of frequencies over which the shortcircuit current amplification remains reasonably constant and nearly equal to its lowfrequency value. For this reason, is referred to as the cutoff frequency. The frequency at which the current amplification is unity, that is, the frequency at which I c = I b , is found from (3.90) as = ( ) + 1
2 2
(3.91)
As we know, a typical value for is 100 ; therefore, in (3.91) the term ( ) is much larger than unity and . Letting = T , we obtain T f T f (3.92)
The relation of (3.92) is referred to as the current gainbandwidth product for the transistor and
358
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
or
Example 3.16 The specifications for a certain transistor state that = 100 , the current gainbandwidth product is f T = 200 MHz , and the collector current is i C = 1 mA . Using the hybrid model of Figure 3.66, find: a. the transconductance g m b. the baseemitter capacitance C e c. the baseemitter resistance r be d. the cutoff frequency Solution: a. From (3.78) g m 40i C = 40 millimhos b. From (3.94)
3 gm 40 10 ----------------------------- 32 pF C e = ---------- = 8 2f T 2 2 10
and this indicates that at this frequency the current amplification is still large.
* The figure of merit is useful in comparing different devices for their overall performance.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
359
i C1
iT i C2
2 2
i E1 = i B2
The combined T and T are evaluated as follows: i E1 = i B2 = i C1 + i B1 i C1 = 1 i E1 = 1 i B2 i B2 = 1 i B2 + i B1 i B1 i B2 = -------------1 1 Also, i E2 = i C2 + i B2 = 2 i E2 + i B2 i B2 i E2 = -------------1 2 Then, 1 i B1 2 i B2 i T = i C1 + i C2 = 1 i E1 + 2 i E2 = 1 i B2 + 2 i E2 = -------------- + -------------1 1 1 2 From (3.98) and by substitution into (3.101) 2 i B2 1 ( 1 1 )i B2 2 i B2 i T = -------------------------------- + -------------- = 1 i B2 + -------------1 1 1 2 1 2 Also, from (3.100) and by substitution into (3.103) i B2 = ( 1 2 )i E2 (3.103) (3.104) i B1 = ( 1 1 )i B2 (3.101) (3.102) (3.95) (3.96) (3.97) (3.98) (3.99) (3.100)
360
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Transistor Networks
1 ( 1 1 )i B2 2 i B2 2 ( 1 2 )i E2 i T = -------------------------------- + -------------- = 1 ( 1 2 )i E2 + -------------------------------1 1 1 2 1 2 = ( 1 ( 1 2 )i E2 + 2 i E2 ) Let the overall value be denoted as T ; then, T = i T i E2 and dividing (3.105) by i E2 we obtain iT T = ----- = 1 ( 1 2 ) + 2 = 1 + 2 1 2 (3.106) i E2 From Table 3.1, iC = ----------- = ---(3.107)
1 iB
(3.105)
and with (3.102), (3.104), (3.105), and (3.107) iT 1 2 1 ( 1 2 )i E2 + 2 i E2 1 ( 1 2 ) + 2 T = ------ = --------------------------------------------------- = --------------------------------------- = ------------------- + --------------------------------------i B1 ( 1 1 ) ( 1 2 )i E2 ( 1 1 ) ( 1 2 ) ( 1 1 ) ( 1 1 ) ( 1 2 ) 1 2 2 1 1 1 = ------------------- + ------------------- ------------------- = ------------------- + ----------------------------- ------------------- = 1 + ( 1 2 ) 1 ( 1 1 ) ( 1 1 ) ( 1 2 ) ( 1 1 ) ( 1 1 ) 1 ( 1 2 ) and since 1 Also, since 1 1 2 T 1 + 1 2 T 1 2 (3.108) (3.109)
ic ib
C
ie v cb v eb re v cb
(b)
ic g0 v cb
i e
(a)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
361
v eb
v cb
Typical values for the hparameter equivalent circuit for the commonbase transistor are: h ib = 30 h rb = 5 10
4
h fb = 0.98
h ob = 6 10
As indicated above, the input resistance h ib = r e has a small value, typically in the 25 to 35 . This can be shown as follows: From (3.2) iC = Ir e or iC = IC + ic = Ir e and since IC = Ir e
v BE V T
(3.110)
v BE V T
( v BE + v be ) V T
= Ir e
v be V T
(3.111) (3.112)
v BE V T
(3.113)
Retaining only the first two terms of this series, by substitution into (3.113) we obtain v be IC i C = I C 1 + ------ = I C + ------ v be VT VT (3.114)
362
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Transistor Networks
and the signal component of i C is Also, and thus IC i c = ------ v be VT ic IC IE i e = --- = ---------- v be = ------ v be V T VT v be VT r e = ------ = -----ie IE r in = r e = 26 the input, and a load resistor R C at the output as shown in Figure 3.70. Rs vs h ib h rb v cb ie h fb i e h ob v cb
RC
(3.115)
(3.116)
To find the overall voltage gain, we connect a voltage source v s with its internal resistance R s at
v eb
v out
Figure 3.70. Circuit for the computation of voltage gain in a common basetransistor amplifier
or r o = 2 M , and since it is
4
age source h rb v cb can also be neglected (short circuit). With these observations, the circuit of Figure 3.70 is simplified to that in Figure 3.71. i in R s vs h ib = r e ie h fb i e = i e i out
RC
v out
Figure 3.71. Simplified circuit for the computation of voltage gain in a common base transistor
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
363
(3.121)
(3.122)
Example 3.17 An NPN transistor is connected in a common base configuration with V T = 26 mV , i E = 1 mA , = 120 , R S = 2 K , and R C = 5 K . Find the voltage and current gains, and input and output resistances. Solution: = ----------- = 120 = 0.992 -------+1 121 h ib = r e = V T i E = 26 mV 1 mA = 26
3 R C 0.992 5 10 A V = --------------- = ----------------------------------------- = 2.45 3 Rs + re ( 2 + 0.026 ) 10
A i = = 0.992 R i = r e = 26 R o = R C = 5 K
364
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Transistor Networks
ie
Rs
i e rb ib rc
ic
RL
re
vs
Typical numerical values for the 2N525 transistor are as follows: = 0.978 r c = 1.67 M r e = 12.5 r b = 840 and since re rc rb rc
RL < rc
and with these assumptions the input and output resistances and voltage and current gains for the Tequivalent model for the commonbase transistor are: Input resis tan ce = r in = r e + r b ( 1 ) re + rb ( 1 ) + RS Output resis tan ce = r out = r e -------------------------------------------re + rb + RS RL Voltage gain = A v = -------------------------------re + rb ( 1 ) Current gain = A i = (3.123) (3.124) (3.125) (3.126)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
365
ic ib
B v be C
v be r n v ce
ib
A
ic i b ie
(b)
v ce
ib rn
A
ic i b g0
v ce
ie
v ce
re
E
(a)
(c)
The equivalent circuit in Figure 3.73(b), is same as in Figure 3.58 with the addition of the emitter resistor
re .
or r 0 = 40 K , and since it is
4
and
the dependent voltage source v ce can also be neglected. Therefore, for simplicity and compactness, we can represent the circuit of Figure 3.73(b) as that in Figure 3.73(c). The hparameter equivalent circuit in Figure 3.73(b) is shown in Figure 3.74 where h ie = r n ib v be h ie h re v ce h fe i b h oe h oe = g 0 h fe = ic v ce h re =
Typical values for the hparameter equivalent circuit for the commonemitter transistor are: h ie = 1.5 K h re = 3.4 10
4
h fe = 80
h oe = 27 10
Figure 3.75 shows the hparameter equivalent circuit in Figure 3.74 with a signal source v S and its internal resistance R S connected on the left side, and a load resistor R L connected on the right side. Using the circuit in Figure 3.75 we can compute the exact voltage and current gains, and input and output resistances.
366
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Transistor Networks
RS
v be
h ie h re v ce h fe i e h oe
vS
v ce R L
v out
Figure 3.75. Circuit for exact computation in voltage and current gains and input and output resistances
or r o = 40 K ,
4
and the
voltage source h re v ce can also be neglected. Therefore, to compute the input and output resistances and overall voltage and current gains to a fairly accurate values, we can use the simplified circuit in Figure 3.73(c), we connect a voltage source v S with its internal resistance R S at the input, and a load resistor R C at the output as shown in Figure 3.76 where an additional resistor R E is connected in series with the emitter resistor r e to increase the input resistance, and R' S represents the series combination of R S and r n .*
B RS
ib re ie i b
ic
RC
vS
v be
E RE
v ce
From the equivalent circuit in Figure 3.76, v be ( r e + R E )i e ( re + RE ) ( ib + ic ) ( r e + R E ) ( i b + i b ) r in = ------ = ------------------------- = ----------------------------------------- = -------------------------------------------- = ( r e + R E ) ( + 1 ) ib ib ib ib or and since 1 r in = ( r e + R E ) ( + 1 ) r in ( r e + R E ) (3.127) (3.128)
The resistor R E is referred to as the emitter degeneration resistance because it causes negative feedback. That is, if the collector current i c increases, the emitter current i e will also increase since
* Typically, r n r e + RE .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
367
Substitution of (3.133) and (3.134) into (3.130) yields v out v b ( re + RE ) ( + 1 ) v out R C - A V = -------- = -------- ---- = --------------------------------------- ---------------------------------------------------vb vS ( re + RE ) ( + 1 ) ( re + RE ) ( + 1 ) + RS vS ( + 1 ) RC RC A V = ---------------------------------------------------- ------------------------------------( + 1 ) ( re + RE ) + RS ( re + RE ) + RS and the minus () sign indicates that the output is 180 outofphase with the input. (3.135)
368
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Transistor Networks
From (3.134) we observe that the introduction of the external resistor R E results in reduction of the overall gain. The current gain is i out i b A i = ------- = ---------- = ib ib Example 3.18 An NPN transistor is connected in a common emitter configuration with V T = 26 mV , i E = 1 mA , = 120 , R S = 2 K , and R C = 5 K . a. Find the voltage and current gains, and input and output resistances if R E = 0 . b. Find the voltage and current gains, and input and output resistances if R E = 200 . c. Find the maximum value of the applied signal v S so that v be or v b under the conditions of (a) and (b) will not exceed 5 mV . Solution: a. From (3.116) h ie = r e = V T i E = 26 mV 1 mA = 26 From (3.128) and from (3.129) r out = R C = 5 K The voltage and current gains are found from (3.135) and (3.136). Thus, with R E = 0 , the voltage gain is 3 RC 120 5 10 A V ------------------- = -------------------------------------------- = 117.2 r e + R s 120 26 + 2 10 3 and the current gain is b. A i = = 120 r in ( r e + R E ) = 120 ( 26 + 0 ) = 3.12 K (3.136)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
369
We observe that for an increase of 88.5% in input resistance, the maximum value of the applied signal v S decreases by 51.8% .
B RS
ib rb re
E
vS
v be
As with the T-equivalent model for the common-base transistor, we assume that re rc rb rc
370
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Transistor Networks
RL < rc
and with these assumptions the input and output resistances and voltage and current gains for the Tequivalent model for the commonemitter transistor are: re Input resis tan ce = r in = r b + -----------1 r c + R S Output resis tan ce = r out = r c ( 1 ) + r e -------------------------re + rb + RS RL Voltage gain = A v = -------------------------------re + rb ( 1 ) Current gain = A i = (3.137) (3.138) (3.139) (3.140)
RC C B
ic ie
RL
ib re go
E
i b ie
RL
ic
ib v in = v b
vb v out
v ec
RE
(a)
(b)
From the equivalent circuit in Figure 3.78(b) we can draw the hparameter equivalent circuit shown in Figure 3.79 where h ic = r e h oc = g o h fc = h rc =
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
371
Typical values for the hparameter equivalent circuit for the commoncollector transistor are: h ic = 1.5 K h rc = 1 h fc = 45 h oc = 27 10
6
Figure 3.80 shows the hparameter equivalent circuit in Figure 3.79 with a signal source v s and its internal resistance R S connected on the left side, and a load resistor R L connected on the right side. RS vS v bc h ic h rc v ec h fc i b v ec
h oc
RL
v out
Figure 3.80. Circuit for the computation of voltage gain in a commoncollector transistor amplifier
To find the input and output resistances and overall voltage and current gains, we denote the conductance h oc = g o with its reciprocal r o in the circuit in Figure 3.78(b), we connect a voltage source v s with its internal resistance R s at the input, and the circuit is as shown in Figure 3.81. ib B C ic i b RS re ie E vb i out vS r0 v out R
L
372
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Transistor Networks
vb ( r e + r o || R L )i e ( r e + r o || R L ) ( i b + i c ) r in = ---- = ---------------------------------- = -------------------------------------------------ib ib ib ( r e + r o || R L ) ( i b + i b ) = ------------------------------------------------------ = ( r e + r o || R L ) ( + 1 ) ib and since 1 r in ( r e + r o || R L ) Also, since r o R L and r e R L r in R L (3.142)
(3.141)
Relation (3.142) indicates that the input resistance r in can be controlled by choosing an appropriate value for the load resistor R L . The output resistance r out cannot be determined by inspection; therefore we will remove the load resistor R L and the voltage source v S , and we will connect a test voltage source v x across the emitter to ground terminals, that is, across the resistor r o and we will find the output resistance from the relation of the equivalent circuit derive it from the relation r out = v x i x as shown in Figure 3.82 where the current i b is shown as ( 1 )i e .*
RS ( 1 ) ie B
ib re
E
ie
i b ix vx
r0
Figure 3.82. Equivalent circuit for the computation of the output resistance
From Figure 3.82 we obverse that the voltage drop across R S is equal to the sum of the voltage drops across r e and r 0 , that is, R S ( 1 )i e = r e i e + v x v x = R S ( 1 )i e r e i e
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
373
374
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Transistor Networks
r o || R L v out RL -------- = -------------------------- ---------------r e + r o || R L r e + R L vb and thus v out v b RL v out RL R2 L -------- ---- ---------------- --------------------- = ----------------------------------------------------- A v = -------- = vb vS re + RL RS + RL ( re + RL ) ( RS + RL ) vS (3.148) (3.147)
Relation (3.148) reveals that the voltage gain of the emitterfollower is less than unity. The current gain is i out A i = ------ib where by the current division expression ro ro i out = ---------------- i e = ---------------- ( + 1 )i b ro + RL ro + RL and thus r o i out ( + 1 )r o A i = ------- = --------------------- ---------------ro + RL ro + RL ib Ai Example 3.19 Figure 3.83 shows the equivalent circuit of a typical emitterfollower and it is given that = 80 . Find the input and output resistances and voltage and current gains.
RS 5 K B
Since r o R L , (3.149)
ib
10 E 30 K
re
vb vS
i b i c i out ie
RL 2 K
r0
v out
Figure 3.83. Emitterfollower transistor amplifier equivalent circuit for Example 3.19
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
375
This is the voltage gain with a load resistor connected to the circuit. With this resistor disconnected, the input resistance as given by (3.141) where 1 , is reduced to r in ( r e + r o ) and (3.146) becomes r in ( re + ro ) vb ---- = ------------------ ----------------------------------R S + r in R S + ( r e + r o ) vS Also, (3.147) becomes v out ro -------- = -------------vb re + ro Then, the voltage gain with the load resistor disconnected is Av
RL
(3.150) (3.151)
(3.152)
(3.153)
and since r e r o this relation reduces to Av With the given values Av and from (3.149) A i 80
RL RL
r o = -------------------R S + r o
3
(3.154)
80 30 10 = ------------------------------------------------------- = 0.998 3 3 5 10 + 80 30 10
376
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
ie re rc i b
RL
rb ic
vS
v out
As with the T-equivalent models for the common-base and common-emitter transistors, we assume that re rc rb rc
RL < rc
and with these assumptions the input and output resistances and voltage and current gains for the Tequivalent model for the commoncollector transistor amplifier are:
L Input resis tan ce = r in = ----------1
Output resis tan ce = r out = r e + ( r b + R S ) ( 1 ) Voltage gain = Av 1 Current gain = A i Table 3.4 summarizes the three possible transistor amplifier configurations.
TABLE 3.4 Phase, input and output resistances, and voltage and current gains for transistor amplifiers CommonBase CommonEmitter Common Collector Input/Output Phase 0 180 0 Input Resistance r in Output Resistance r out Voltage Gain A v Current Gain A i Low Equal to R C Depends on ratio R C R S Close to unity Moderate High High High High Low Close to unity Large
377
Let us consider the transistor circuit in Figure 3.85. We will refer to it in our subsequent discussion to define the cutoff, active, and saturation regions.
RC
iC v CE
+
V CC
iB
+ vS
RB
v BE
iE
and if v CB 0.7 V , the transistor will be in the active mode. However, if v CB < 0.7 V , the transistor will be in the saturation mode which is discussed in the next subsection.
378
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
We can also find the maximum value of the applied signal voltage v S that will keep the transistor in the active mode from the relation v S max = R B I BM + 0.7 (3.161)
If we increase the base current above I BM , there will be a corresponding increase in I CM and since v C = V CC R C I C , v C will decrease and if it falls below the value of v BE , the collectorbase junction will become forwardbiased and if it reaches a value of 0.6 V , any further increase in the base current will result in a very small increase in the collector current, and whereas = di C di B in the active mode, we can see that this relation does not hold when the transistor is in the saturation mode. In this case is referred to as the current gain at saturation and it is denoted as sat . When a transistor is deeply into saturation, the collector to emitter voltage is denoted as v CE sat and its value is approximately 0.2 V , that is, v CE sat 0.2 V and the corresponding collector current i C sat is found from the relation V CC v CE sat i C sat = -----------------------------RC Example 3.20 For the transistor circuit in Figure 3.86 it is known that it is in the active mode and > 50 . Find v E , i E , v C , i C , and i B . Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications (3.163) (3.162)
379
iB vS
+
C B + +
iC
12 V
vC vE
+
RB
v B v BE E 8 V RE 3 K i E
Solution:
We do not know whether the transistor operates in the cutoff, active, or saturation region, but since v B = 8 V , it is safe to assume that it operates either in the active or saturation region. Assuming active mode of operation, we obtain v E = v B v BE = 8 0.7 = 7.3 V vE 7.3 i E = ----- = ---------------- = 2.43 mA 3 RE 3 10 and since i C i E , let us assume that i C 2.3 mA Then, in active mode of operation v C = V CC R C i C = 12 5 10 2.3 10
3 3
= 0.5 V
and since this value is much less that v B = 8 V , we conclude that the transistor is deeply into saturation. Then, in saturation v C = v CE sat + v E = 0.2 + 7.3 = 7.5 V V CC v C 12 7.5 i C = --------------------- = ------------------ = 0.90 mA 3 3 5 10 5 10 and i B = i E i C = 2.43 0.90 = 1.53 mA
and this is indeed a very large base current. We now can find the value of at saturation. iC sat = --- = 0.90 = 0.59 --------iB 1.53 This value also indicates that the transistor is deeply into saturation.
380
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
These current gains are related to one another by F, R F, R = ------------------ F, R + 1 F, R F, R = ------------------1 F, R (3.165)
By substituting these relationships into (3.164), we can show that the EbersMoll model satisfies Kirchhoffs current law, that is, IE = IC + IB (3.166)
If the emitterbase junction is forward biased and the collectorbase junction is reverse biased, then I F I R and the bipolar transistor is said to be in the forwardactive mode of operation. In this case, the terminal currents are given by IC = IF IF I E = ----F IF I B = ---F (3.167)
Figure 3.87 shows how an NPN and a PNP transistors are biased.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
381
IC
VC
+
C B IB VB
IC
VC
IB VE
E
+
IE
IE E + VE
NPN Transistor
PNP Transistor
Figure 3.87. Biased bipolar transistors showing the direction of conventional current flow
To bias an NPN bipolar transistor into the forward active region, we should ensure that V BE > 4V T and V C V B . In this case, I C , I E , and I B are positive and are flowing in the directions indicated. To bias a PNP bipolar transistor into the forwardactive region, we should ensure that V EB > 4V T and V C V B . In this case, I C , I E , and I B are positive and are flowing in the directions indicated. By using (3.166) and (3.167), we can express the terminal currents in terms of one another as IC = F IB = F IE IC I E = ----- = ( F + 1 )I B F IC IE I B = ---- = -------------F F + 1 (3.168)
If the collectorbase junction is forward biased, and the emitterbase junction is reverse biased, then I R I F and the transistor is said to be in the reverseactive mode of operation in which the collector and emitters effectively reverse their roles. In this case the terminal currents are given by IC = IR R IE = IR IB = IR R (3.169)
By using (3.166) and (3.169), we can express the terminal currents in terms of one another as IE = R IB = R IC I C = ( R + 1 )I B = I E R (3.170)
If both junctions are forward biased simultaneously, the transistor is said to be in saturation. Let I S be the saturation current and V T the thermal voltage. Then for the NPN bipolar transistor biased as shown in Figure 3.87, the forward current component, I F , is given by IF = IS ( e
V BE V T ( VB VE ) VT
1 ) = IS ( e
1)
(3.171)
1 ) = IS ( e
( VB VC ) VT
1)
(3.172)
382
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
IS ( V V ) V 1 ) ----- ( e B C T 1 ) R (3.173)
If V BE > 4V T and V C V B , the transistor is biased deeply into the forwardactive region and (3.172) reduces to IC IS ( e
( VB VE ) VT
IS (V V ) V I E ----- ( e B E T ) F
IS (V V ) V I B ---- ( e B E T ) F
(3.174)
If V BC > 4V T and V E > V B , the transistor is biased deeply into the reverseactive region and (3.173) reduces to IS (V V ) V I C ----- ( e B C T ) R IE IS ( e
( VB VC ) VT
IS ( V V ) V I B ----- ( e B C T ) R
(3.175)
If V BE > 4V T and V BC > 4V T , the transistor is deeply saturated and (3.173) can be expressed as IC IS ( e
( VB VE ) VT
IS (V V ) V ) ----- ( e B C T ) R (3.176)
If V BE < 4 V T and V BC < 4 V T , the transistor is deeply into the cutoff region and (3.173) can be expressed as IS IS IS IS I E ---I B ---- ----I C ----(3.177) R F F R For the PNP transistor shown in Figure 3.86, the forward current component, I F , is given by IF = IS ( e
V EB V T
1 ) = IS ( e
( V E V EB ) V T
1)
(3.178)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
383
1 ) = IS ( e
( VC VB ) VT
1)
(3.179)
From the EbersMoll model equations we can derive relationships for small signal parameters. First, we will find the incremental resistance seen looking into the base terminal of an NPN transistor with the emitter voltage held fixed. From (3.174), that is, IS ( V V ) V I B ---- ( e B E T ) F (3.180)
we obtain this incremental base resistance by differentiating I B with respect to V B , which yields 1 I B 1 1- I S V B --------- = ------ ---- ( e ( VB VE ) VT ) r b = --------- = VT F V B I B IB VT r b = -----IB (3.181)
or
Next, we shall compute the incremental resistance seen looking into the emitter terminal with the base voltage held constant. From (3.174), that is, IS ( V V ) V I E ----- ( e B E T ) F (3.182)
we obtain this incremental emitter resistance by differentiating I E with respect to V E , which yields 1 IS I E 1 V E 1 ----- ( e ( VB VE ) VT ) r e = --------------- = --------= ------ V E V - F ( IE ) T IE or re = VT IE (3.183) Now, we shall define the incremental transconductance gain of the NPN transistor as the partial derivative of the collector current with respect to the base voltage. From (3.174) IC IS ( e
( VB VE ) VT
(3.184)
384
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(3.185)
(3.186)
(3.187)
we can also express these smallsignal parameters in terms of one another as F r b = ----- = ( F + 1 )r e gm rb F r e = ----- = -------------gm F + 1 F F g m = ---- = ----rb re (3.188)
B E NPN Transistor
B E PNP Transistor
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
385
atV CE = 10 V I C = 1 mA , ,
at V CE = 10 V , I C = 1 mA ,
386
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
12
A to 10
15
A as in junction
diodes, v BE is the basetoemitter voltage, and V T 26 mV at T = 300 K . A very useful parameter in transistors is the commonemitter gain , a constant whose value ranges from 75 to 300. Its value is specified by the manufacturer. The base current i B is much smaller than the collector current i C and these two currents are related in terms of the constant as iB = iC Another important parameter in transistors is the commonbase current gain denoted as and relates the collector and emitter currents as i C = i E The parameters and are related as Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
387
v BE = cons tan t
where V A is the Early voltage supplied by the manufacturer, and I C is the DC collector current. Two common methods of biasing a transistor are the fixed bias method and the selfbias method. There are four classes of amplifier operations: Class A, Class AB, Class B, and Class C. Class A amplifiers operate entirely in the active region and thus the output if a faithful reproduction of the input signal with some amplification. In a Class A amplifier the efficiency is very low. Class A amplifiers are used for audio and frequency amplification. Class AB amplifiers are biased so that the collector current is cutoff for a portion of one cycle of the input and so the efficiency is higher than that of a Class A amplifier. Class AB amplifiers are normally used as pushpull amplifiers to alleviate the crossover distortion of Class B amplifiers. Class B amplifiers are biased so that the collector current is cutoff during onehalf of the input signal. Therefore, the efficiency in a Class B amplifier is higher than that of a Class AB amplifier. Class B amplifiers are used in amplifiers requiring high power output. Class C amplifiers are biased so that the collector current flows for less than onehalf cycle of the input signal. Class C amplifiers have the highest efficiency and are used for radio frequency amplification in transmitters. The operation of a simple transistor circuit can also be described graphically using i B versus v BE and i C versus v CE curves. The average power drawn from the collector supply is P CC = V CC I C
388
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Summary
The average power absorbed by the load is P LOAD = R LOAD I C + R LOAD ( i c RMS ) The average power absorbed by the transistor is P C = P CC P L The piecewiselinear analysis is a practical method of analyzing transistor amplifiers. The incremental model for the transistor provides a better approximation than the piecewise linear approximation. In the incremental model for the transistor the input resistance r n is the slope of the input voltage and current characteristics and it accounts for the voltage drop across the baseemitter junction. Likewise, the output conductance g o is the slope of the output current and voltage characteristics. The voltage amplification factor is related to the input characteristics caused by a change in v CE , and the current amplification factor is related to the output characteristics caused by a change in i B . Transistor hybrid parameters provide us with a means to evaluate voltages, currents, and power in devices that are connected externally to the transistor. The parameters r n , , , and g o are normally denoted by the h (hybrid) parameters as r n = h 11 = h ie , = h 12 = h re , = h 21 = h fe , and g o = h 22 = h oe. These designations along with the additional notations v be = v 1 , i b = i 1 , v ce = v 2 , and i c = i 2 , provide a symmetrical form for the relations of (3.62)and (3.63) as follows: v 1 = h 11 i 1 + h 12 v 2 i 2 = h 21 i 1 + h 22 v 2 In the incremental model for the transistor the current amplification A c is defined as i A c = --c is The transconductance, denoted as g m , and defined as di C g m = ----------dv BE iC = -----VT
2 2
iC = IC
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
389
390
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
= 6 K
IC
RE
= 8 K
IE
V CC
VC B 10 V
IB
VE 10 V
V EE
4. For the PNP transistor circuit below, it is known that R B = 130 K , V E = 2 V and
V EB
= 5 K
IC
RE IE
= 10 K
V CC
12 V
VC VB B VE RB IB
12 V
V EE
5. Find the output resistance r out for a bipolar junction transistor whose Early voltage is V A = 75 V and the DC collector current is a. I C = 0.1 mA b. I C = 1 mA c. I C = 5 mA Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
391
12 V
V CC
VS
VE
7. For the circuit below, = 150 . Find I E , V E , I C , V C , I B , and determine whether this circuit with the indicated values operates in the active, saturation, or cutoff mode. Hint: The base to emitter junction of an NPN transistor is considered to be reverse biased if VB = VE = 0
RC
5 K
10 V
V CC
IB B V BE VE
IC E
VC
3 K
RE
IE
8. For the circuit below, = 110 . Find the highest voltage to which the base can be set so that the transistor will be in the active mode. Hint: The collector base junction will still be reversebiased if we make V B = V C .
392
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
RC C B IB VS RB VB VE V BE E IE RE 3 K IC VC 5 K
12 V
V CC
9. For an NPN transistor circuit = 100 , V CC = 11.3 V , V B = 3.5 V and with the reverse biased collector base junction set at V CB = 1.8 V we want the collector current to be I C = 0.8 mA . What should the values of R C and R E be to achieve this value? 10. For a PNP transistor circuit with = 120 , V EE = 12 V , V B = 0 V , V CC = 12 V , and R E = 3 K , what would the largest value of R C be so that the transistor operates at the active mode? 11. For a PNP transistor circuit with = 150 , V EE = 12 V , V B = 0 V , V CC = 12 V , I E = 0.8 mA , and V CB = 3 V , what should the values of R C and R E be so that the transistor operates at the active mode? 12. For the circuit below, it is known that = 120 for both transistors. Find all indicated voltages and currents. Are both the transistors operating in the active mode? What is the total power absorbed by this circuit?
R E2 R C1 R1 60 K C1 8 K I B2 10 A I C1 V B2 V C2 V
C1
5 K I E2 C2 I C2 R C2 8 K
12 V
V CC
E2 V EB2 B2
V E2
B1 I B1 30 K
V BE1 E1 V E1
V B1 R2
I E1 R E1 5 K
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
393
20 V
+ V CC
iC iS
RS R 1 2 M 5 K
vS
RE + V BB 10 V
a. Sketch a family of these curves for i B = 2.5, 5.0, 7.5, and 10 A . Construct a load line, and indicate the current and the voltage at which the load line intersects the axes. b. Find the quiescent collector current I C and collectortoemitter voltage V CE . c. Determine graphically and plot i C versus i S for 15 < i S < 15 A . 14. For the PNP transistor circuit below, = 70 and the output voltampere characteristic curves are approximately horizontal lines.
RC
15 V
iC iS
RS
V CC
R1
vS
15 V
+
V BB
RE
a. Find the values of R 1 and R C required to locate the quiescent operating point Q at i C = 1 mA and v EC = 5 V .
394
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
b. Sketch the load line on the i C versus v EC coordinates. Show the current and voltage at which the load line intersects the axes, and indicate the quiescent point on this line. It is not necessary to draw the collector characteristics. c. If the input signal is a sinusoidal current, approximately what is the greatest amplitude that the signal i s can have without waveform distortion at the output? 15. For the transistor model shown below, it is known that = 80 , i c = 5 mA , r' b = 50 , and r o = 30 K .
B
ib v be r' b r be
v' be ro g m v' be
E
ic
v ce
ie
a. Find the transconductance g m and the baseemitter resistance r be . b. Repeat part (a) for = 60 , i c = 1 mA , r' b = 50 , and r o = 30 K . 16. For the transistor circuit below, i C = 1 mA and the hybrid parameters are r n = 2.5 K , = 2 10 , = 100 , and r o = 100 K . a. Find the values of R B and R C required for a quiescent point Q at i C = 1 mA and v CE = 5 V .
RC 20 V
+
4
V CC
is
Rs
+
RB V BB RE
vs
20 V
b. Using the hybrid representation for the transistor, draw an incremental model. c. Find the current amplification A c . Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
395
5 K 10 V
V CC
V EE
18. The equations describing the h parameters can be used to represent the network shown below. This network is a transistor equivalent circuit for the commonemitter configuration and the h parameters given are typical values for such a circuit. Compute the voltage and current gains for this network if a voltage source of v 1 = cos t mV in series with 800 is connected at the input (left side), and a 5 K load is connected at the output (right side). i1
+
i2
+ +
h 11 h 12 v 2 h 22 h 21 i 1
h 11 = 1.2 K h 12 = 2 10 h 21 = 50
4
v1
v2
h 22 = 50 10 6
396
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(1)
and as we know from Chapter 2 at T = 27 C , V T = 26 mV . From the given data 1.2 10 Division of (1) by (2) yields
3
= Ir e
0.7 V T
(2)
i C = 1.2 10 e ln ( i C ) = ln ( 1.2 10 e
3 ( v BE 0.7 ) V T
3 ( v BE 0.7 ) V T
) = ln ( 1.2 10 ) + ( v BE 0.7 ) V T
3
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
397
= 1.188 10
= 1.188 mA
3 3 3 iC 15 I r = --------------- = 1.188 103- = 1.188 10 - = 1.188 10 11 = 2.42 10 A ------------------------------------------------------------------------------------v BE V T 26.92 0.7 ( 26 10 ) e 4.911 10 e e
3.
RC
= 6 K
IC
E VE
RE
= 8 K
IE 10 V V EE
V CC
VC B 10 V
IB
0.72 V
V E + R E I E = V EE V EE + V E 10 0.72 I E = --------------------- = --------------------- = 1.16 mA 3 RE 8 10 From Table 3.1, I B = I E ( + 1 ) and with = 115 we find that
5 I B = 1.16 mA = 10 A = 10 A --------------------115 + 1
I C = I E I B = 1.16 10 10
= 1.15 10
= 1.15 mA
By application of KVL on the collector (left) side of the circuit above, we obtain
VC
= V CC R C I C = 10 6 10 1.15 10
= 3.1 V
398
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
= 5 K
IC
= 10 K
V CC
12 V
From Table 3.1, = I C I E , = I C I B , and from the circuit above, V C = R C I C 12 , and V B = V E V BE = 2 0.7 = 1.3 V . Therefore, we need to find I E , I B , and I C . R E I E + V E = 12 V 12 2 I E = ---------------- = 1 mA 10 K VB 1.3 V I B = ------ = ------------------- = 0.01 mA RB 130 K I C = I E I B = 1 0.01 = 0.99 mA IC = ----- = 0.99 = 0.99 --------IE 1
I C = 0.1 mA
b.
I C = 1 mA
c.
I C = 5 mA
75 = ------------------ = 5 K 3 5 10
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
399
VS
V BE VB E 3.7 V VE
Since V B = 3.7 V it is reasonable to assume that the baseemitter junction is forward biased and thus V BE = 0.7 V . Then, V E = V B V BE = 3.7 0.7 = 3 V and VE 3 I E = ------ = ------------------ = 1 mA 3 RE 3 10 It is given that = 110 . Then, = ----------- = 110 = 0.991 -------+1 111 and I C = I E = 0.991 1 = 0.991 mA The collector voltage is V C = V CC R C I C = 12 5 10 0.991 10
3 3
= 7.05V
This is an NPN transistor and the collector (N) to base (P) must be reversebiased for the transistor to operate in the active mode. Since V C > V B , the transistor is indeed operating in the active mode and our assumption that the baseemitter junction is forwardbiased, is correct. Finally, I B = I E I C = 1 0.991 = 9 A
3100
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
10 V
IB B V BE VE
IC E
VC
3 K
RE
IE
Since the base is grounded, V B = 0 , the emitterbase junction does not conduct, V BE = 0 , IB = 0 , IE = 0 , a n d VE = 0 . T h e c o l l e c t o r c u r r e n t IC i s a l s o z e r o b e c a u s e I C = I E I B = 0 and thus V C = V CC R C I C = 10V . Under those conditions the transistor behaves like an open switch and thus it is operating in the cutoff mode. 8.
RC C B IB VS RB VB VE V BE E IE RE 3 K IC VC 5 K 12 V V CC
The transistor will be in the active mode for V B > 0.7V and for V B V C . Since = 110 , = ----------- = 110 = 0.991 -------+1 111 and V B = V BE + V E = V BE + R E I E V B 0.7 V B V BE I E = ---------------------- = ------------------3 RE 3 10 Then, Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
3101
3102
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
IE
VE
A PNP transistor will still be in the active mode if V C = V B . For this exercise the base is grounded so V B = 0 and we can let V C 0 also. Then, V C = R C I C V CC = R C I E 12 0 or where and from V E = V EB = 0.7 = V EE R E I E = 12 R E I E we obtain -----------------I E = 12 0.7 = 3.77 mA (3) 3 3 10 12 R C = -------- (1) I E 120 = ----------- = -------- = 0.992 (2) +1 121
By substitution of (2) and (3) into (1) we obtain 12 12 R C = -------- = ---------------------------------------------- = 3.21 K 3 I E 0.992 3.77 10
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
3103
IB B
VE
VC C RC 12 V IC V CC
3104
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
V E2
B1 I B1 30 K
V C1 V BE1 E1 I E1 V E1 R E1 5 K
V B1 R2
The left part in the circuit above is the same as that in Example 3.8 where we applied Thevenins theorem. Therefore the given circuit is redrawn as shown below.
8 K R C1 C1 R TH V TH 4V 20 K B1 I B1 I' C1 10 A I C1 V C1 5 K E2 I E2 C2 I C2 R C2 8 K V E2 V CC
R E2 V I B2EB2 B2 V C2
12 V
V BE1 E1 V B2
V B1 I E1
V E1
R E1 5 K
Application of KVL around the left part of the circuit yields R TH I B1 + V BE1 + R E1 I E1 = V TH ( 20 10 )I B1 + 0.7 + ( 5 10 )I E1 = 4
3 3.3 4I B1 + I E1 = ---------------- = 0.66 10 3 5 10 3 3
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
3105
A = 5.28 A
6
I E1 = ( + 1 )I B1 = 121 5.28 10
3
= 0.639 mA
3
V E1 = R E1 I E1 = 5 10 0.639 10
= 3.2 V
= 0.634 mA
= 7.09 V
Since this is an NPN transistor and V C1 > V B1 , the basecollector PN junction is reverse biased and thus the transistor is in active mode. The emitter voltage V E2 of the PNP transistor is
V E2
and the emitter current I E2 is V CC V E2 12 7.79 I E2 = ------------------------ = --------------------- = 0.842 mA 3 R E2 5 10 It is given that = 120 . Then, I C2 = I E2 = ----------- I E = 0.992 0.842 = 0.835 mA +1 and V C2 = R C2 I C2 = 8 10 0.835 10
3 3
= 6.68 V
3106
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
+ V CC
iC iS
RS R 1 2 M 5 K
vS
RE + V BB 10 V
a. With = 100 , i C = i B = 100i B and for i B = 2.5, 5.0, 7.5, and 10 A , we obtain i C = 0.25, 0.50, 0.75, and 1.00 mA and these are shown on the plot below.
1.00 0.75 0.50 0.25
i C ( mA )
10
15
20
v CE ( V )
The load line is obtained from the relation v CE + R C i C = V CC = 20 V Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
3107
Then, for the range 15 < i s < 15 A , we obtain the corresponding collector current values as shown on the table below. i s ( A ) i C ( mA )
15 0 10 0 5 0 0 5 5 10 10 10 15 10
15
10
10
15
3108
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
15 V
iC iS
RS
V CC
R1
vS
15 V
+
V BB
RE
b.
v EC ( V )
The slope of the load line is i C2 i C1 0 1m = --------------------------- = -------------- = 0.1 v EC2 v EC1 15 5 The i C intercept, denoted as i CQ , is found from the straight line equation i C = mv EC + i CQ or i CQ = i C mv EC
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
3109
ib v be r' b r be
v' be ro g m v' be
E
ic
v ce
ie
3
a. From (3.78), g m 40i c = 40 5 10 and from (3.83) 80r be = ----- = -------- = 400 gm 200 b. g m 40i c = 40 10 and r be = ----- = 60 = 1.5 K ----gm 40
3
= 200 millimhos
= 40 millimhos
3110
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
V CC
is
Rs
+
RB V BB RE
vs a.
20 V
R C i C + v CE = V CC V CC v CE 20 5 R C = ------------------------ = -------------- = 15 K 3 iC 10 I B = I C = 10 100 = 10 A V BB 20 R B = --------- = --------- = 2 M 5 IB 10 b. The incremental model circuit in Figure 3.63 is applicable here and it is shown below with the given parameters and the values obtained in part (a).
B RB 2 M
3
rn
2.5 K
iB
100 i B
iC
R eq
C
RC 15 K 100 K
is
Rs
R eq 26
i b
E
ro
v CE
c. From (3.67) ro ic RB A c = --- = --------------------- --------------------( RB + rn ) ( ro + RC ) is or 2 10 10 A c = ------------------------------ 100 ------------------------ 87 6 5 2.0025 10 1.15 10
6 5
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
3111
V CC
Let us assume that the transistor is in the active mode. Then, V EB = 0.7 V
IB
Also,
VE IE
= 10 ( 2 K ) I E
where the numerical value of V B in the last expression above is in mA , but since weve assumed that the transistor is in the active mode, its value should be just a few microamps. Therefore, I E 4.65 mA and this value implies that the transistor operates in the saturation mode. In this case, V CE sat 0.2 V , and thus V C = V E V CE sat = 0.7 + V B 0.2 = V B + 0.5 V The collector current is V C ( 10 ) V B + 0.5 + 10 I C sat = -------------------------- = -------------------------------- = 0.2V B + 2.1 mA (3) 5 K 5 K and with I E = I B + I C and (1), (2), and (3), we obtain
3112
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
I C sat = 0.2V B + 2.1 3.03 mA Also, I C sat 3.03 sat = ---------- = ------------ 15.5 0.196 IB v 1 = h 11 i 1 + h 12 v 2 (1) i 2 = h 21 i 1 + h 22 v 2 (2) With the voltage source v 1 = cos t mV in series with 800 connected at the input and a
5 K load connected at the output the network is as shown below.
800 1200
+ v1
1 0 mV
4
i1
i2
+
5000 50 i 1 50
v2
10 v 2
10 6
v2 6 50i 1 + 50 10 v 2 = i 2 = ----------5000 or 2 10 i 1 + 2 10 v 2 = 10 50i 1 + 250 10 v 2 = 0 We write the two equations above in matrix form and use MATLAB for the solution.
A=[2*10^3 2*10^(4); 50 250*10^(6)]; B=[10^(3) 0]'; X=A\B;... fprintf(' \n'); fprintf('i1 = %5.2e A \t',X(1)); fprintf('v2 = %5.2e V',X(2))
6 3 4 3
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
3113
v2 = -1.02e-001 V
3114
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Chapter 4
Field Effect Transistors and PNPN Devices
his chapter begins with a discussion of Field Effect Transistors (FETs), characteristics, and applications. Other PNPN devices, the fourlayer diode, the silicon controlled rectifier (SCR), the silicon controlled switch (SCS), and the triac are introduced with some of their applications. The chapter includes also a brief discussion on unijunction transistors, and diacs.
P P
Figure 4.1. Pictorial representation and output voltampere characteristics for a typical JFET
The lower terminal in the N material is called the source, and the upper terminal is called the drain; the two regions of P material, which are usually connected together externally, are called gates. PN junctions exist between the P and N materials, and in normal operation the voltage applied to the gates biases these junctions in the reverse direction. A potential barrier exists across the junctions, and the electrons carrying the current i D in the N material are forced to flow through the channel between the two gates. If the voltage applied to the gates is changed, the width of the transition region at the junction changes; thus the width of the channel changes, resulting in a change in the resistance between source and drain. In this way the current in the Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
41
Depletion Region
The width of the depletion region depends on the applied bias. Forwardbiasing, where a voltage source is connected with the plus (+) on the P side and the minus () on the N side, makes the depletion region narrower by repelling the holes and free electrons toward the junction, and if the bias is about 0.65 V , the free electrons will cross the junction and join with the holes. Thus, the forward biasing causes the depletion region to decrease resulting in a low resistance at the junction and a relatively large current across it. If a reversebiasing is applied, where a voltage source is connected with the plus (+) on the N side and the minus () on the P side, the free electrons move towards the right side while the holes move towards the left side causing the depletion region to widen. A high resistance between the terminals is developed, and only a small current flows between the terminals.
42
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
V DD
In Figure 4.3, let us adjust the voltage source at the gate so that V GG = 0 , that is, the gate is grounded, and set the voltage at the drain to V DD = 10 V . Let us assume that under those conditions the draintosource current through the channel is I DS = 10 mA .Therefore, we conclude that the draintosource resistance R DS is R DS = V DD I DS = 10 V 10 mA or R DS = 1 K . Next, let us adjust the voltage source at the gate so that V GG = 2 V , and maintain the voltage at the drain to V DD = 10 V . This reversebias condition cause the depletion region to expand and that reduces the effective crosssectional area of the channel. Let us assume that under those conditions the draintosource current through the channel is I DS = 0.1 mA .Therefore, we conclude that the draintosource resistance R DS is now R DS = 10 V 0.1 mA or R DS = 100 K . Now, let us adjust the voltage source at the gate so that V GG = 4 V , and maintain the voltage at the drain to V DD = 10 V . This reversebias condition cause the depletion region to expand even more and that reduces further the effective crosssectional area of the channel. and we assume that under those conditions the drain to source current through the channel is I DS = 0.01 mA .Thus, we conclude that the drain to source resistance R DS is now
R DS = 10 V 0.01 mA or R DS = 1 M , and therefore, we observe that the drain to source
resistance R DS can be controlled by the voltage applied at the gate. The voltage at the gate which causes the draintosource resistance R DS to become infinite, is referred to as the pinchoff voltage and it is denoted as V P . In other words, when V GG = V P , no current flows through the channel.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
43
D D G G P N
S N Channel JFET
S P Channel JFET
Like in bipolar transistors, one important parameter in FETs is its transconductance g m defined as the ratio of the change in current i DS to the change of voltage v GS which produced it. In other words,
i DS g m = ----------v GS
(4.1)
v DS = cons tan t
Example 4.1 Figure 4.5 shows a commonsource Nchannel JFET amplifier circuit and Table 4.1 shows several values of the current i DS corresponding to the voltage v GS .
RD
D G
1 K
+
15 V
V DD
v in
v out i DS
S
44
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
a. Find the output voltage v out if the input signal is v in = 2.4 V . b. Find the output voltage v out if the input signal is v in = 1.8 V . c. Is this an inverting or a noninverting amplifier? d. Find the transconductance using the results of (a) and (b). e. Plot i DS versus v GS and indicate how the transconductance can be calculated from this plot. Solution: a.
RD D G 1 K 15 V
V DD
v in = v GS
v out i DS
S
(4.2)
but we do not know all values of i DS for the interval 0 i DS 35 mA . Therefore, let us use the following MATLAB script to plot a suitable curve for this interval.
vGS=[4 3 2 1 0]; iDS=[0 2 8 20 35];... % These are the data in Table 4.1 curve=polyfit(vGS,iDS,4);... % Fits the data to a polynomial of fourth degree vGSaxis=4.0:0.1:0.0;... % Creates horizontal (vGS) axis polcurve=polyval(curve,vGSaxis);... % Computes the polynomial for vGS axis values plot(vGSaxis,polcurve);... % Plots the fourth degree polynomial xlabel('vGS (volts)'); ylabel('iDS (milliamps)'); title('Plot for Example 4.1'); grid
The generated plot is shown in Figure 4.7. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
45
From the plot of Figure 4.7 we see that for v in = v GS = 2.4 V , i DS 5 mA . Therefore,
v out
v GS = 2.4 V
= V DD R D i DS = 15 10 5 10
= 10 V
(4.3)
b. From the plot of Figure 4.7 we see that for v in = v GS = 1.8 V , i DS 10 mA . Therefore,
v out
v GS = 1.8 V
3
= V DD R D i DS = 15 10 10 10
= 5V
(4.4)
c.
The results of (a) and (b) indicate that an increase in the input voltage results in a decrease of the output voltage. Therefore, we conclude that the given JFET circuit is an inverting amplifier.
3 i DS 1 ( 5 10 ) 10 g m = ------------ = ------------------------------------ = 0.0083 2.4 ( 1.8 ) v GS
d. e.
(4.5)
The plot in Figure 4.7 shows the slope of g m and it is calculated as in (4.5).
46
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
w OX
Drain ( D )
n+
n+
Substrate ( B )
In Figure 4.8, heavily doped Ntype regions, indicated by n + , are diffused into a Ptype substrate or base. An n conducting channel may be formed and exist with the gate voltage V G = 0 . A negative gate voltage will then drive electrons out of the channel, increasing the resistance from source to drain. This is termed depletionmode operation. The JFET also operates in this manner. Conversely, if no channel exists with V G = 0 , one can be formed by applying positive voltage V G and attracting electrons to a thin surface layer. This is termed enhancementmode operation. The enhancement mode MOSFET has a lightly doped channel and uses forward bias to enhance the current carriers in the channel. A MOSFET can be constructed that will operate in either mode depending upon what type of bias is applied, thus allowing a greater range of input signals. The symbols for the four basic variations of the MOSFET are shown in Figure 4.9.
D B G S n channel G S p channel D D B G D B
Depletion MOSFET
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
47
Chapter 4 Field Effect Transistors and PNPN Devices 4.2.1 The NChannel MOSFET in the Enhancement Mode
Figure 4.10 shows the draintosource current i DS versus the draintosource voltage v DS for different values of gatetosource voltage v GS .
i D S ( mA ) 2.5 2.0 1.5 1.0 0.5 0 5 10 vGS = 5 V vG S = 4 V vG S = 3 V vG S = 2 V vG S = 1 V vD S ( V )
The voltage at which the channel is closed is known as the pinchoff voltage V P , and the minimum voltage required to form a conducting channel between the drain and source is referred to as the threshold voltage and it is denoted as V T . Typical values for V T are 2 to 4 volts for high voltage devices with thicker gate oxides, and 1 to 2 volts for lower voltage devices with thinner gate oxides. In terms of this, the drain current i D in a MOSFET can be written approximately as
W 1 2 i D = ------------- ( v GS V T )v DS -- v DS Lw OX 2 for v DS v GS V T
(4.6)
where L is the channel length as shown in Figure 4.8, W is the width of the structure perpendicular to the paper in Figure 4.8, w OX is the oxide thickness, is its dielectric constant, and is the mobility of carriers in the channel. The values for the channel length and width depend on the voltage v GS and typical values are 2 m L 10 m and 100 m W 100 m . It is convenient to represent the quantity w OX as k n *, and typical values of k n are around
20 A V . Then (4.6) is expressed as W 1 2 i D = k n ---- ( v GS V T )v DS -- v DS L 2 for v DS vGS V T
2
(4.7)
* The subscript n is used here as a reminder that relation (4.7) applies to nchannel MOSFETs.
48
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(4.8)
where v DS v GS V T defines the socalled saturation or pentode region. When v GS < V T , the MOSFET is said to be in the cutoff region. The quadratic (triode) and saturation regions are as shown in Figure 4.11.
i D S ( mA ) Triode region Saturation region
v DS v GS V T v DS v GS V T v DS = v GS V T
0 vD S ( V )
Figure 4.11. Triode and saturation regions for an enhancement type NMOS where v GS > V T
As shown in Figure 4.11, the triode region starts as a linear function where the draintosource resistance r DS is linear but it changes to a curve because the resistance changes with changes in the draintosource voltage v DS . The saturation region starts where for any further increases in
v DS there is no increase in the drain current i D .
Figures 4.10 and 4.11 reveal that for small values of v DS , say 0.05 v DS 0.15 V , relation (4.7) can be written as
W i D k n ---- [ ( v GS V T )v DS ] L for v DS v GS V T
(4.9)
Thus, in that range of v DS the MOSFET behaves as a linear resistor, usually denoted as r DS , and its value can be found from
v DS L r DS -------- = -------------------------------------iD k n W ( v GS V T )
(4.10)
* This name is carried over from the old days of the vacuum tube triode whose characteristics are as shown in Figure 4.11. It
is also referred to as the quadratic region. Likewise, the saturation region is sometimes referred to as the pentode region. The drain current i D is not exactly independent of the draintosource voltage v DS . It increases with increasing v DS due to the socalled channel width modulation caused by reduction of the effective channel length, and since i D is inversely proportional to the channel length, i D increases with v DS and thus (4.8) is an approximation to the exact drain current i D
1 W 2 given by i D = -- kn ----- ( v GS V T ) ( 1 + v DS ) where is a small quantity, typically 0.01 . 2 L
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
49
Assume that k n , L , and W do not change significantly in this interval. Solution: The MATLAB script for Example 4.2 is given below.
kn=18*10^(6); L=5*10^(6); W=60*10^(6); VT=1; rD1 = 10.^(3).*L./(kn.*W.*(1.5VT)); rD2 = 10.^(3).*L./(kn.*W.*(2VT));...% Kilohm values rD3 = 10.^(3).*L./(kn.*W.*(2.5VT)); rD4 = 10.^(3).*L./(kn.*W.*(3VT));... rD5 = 10.^(3).*L./(kn.*W.*(3.5VT)); fprintf(' \n');... disp('vGS (Volts) rD (KOhms)');... % Display vGS and rD values disp('-------------------------');... fprintf('%6.1f %10.2f \n', 1.5,rD1,2.0,rD2,2.5,rD3,3.0,rD4,3.5,rD5) vDS=(0:0.01:100)*10^(-2); iD1=rD1*vDS; iD2=rD2*vDS; iD3=rD3*vDS; iD4=rD4*vDS; iD5=rD5*vDS; plot(vDS,iD1, vDS,iD2, vDS,iD3, vDS,iD4, vDS,iD5); xlabel('vDS in V');... ylabel('iD in mA'); title('Linear region for typical MOSFET'); grid
When this program is executed, MATLAB displays the following: vGS (Volts) rD (KOhms) ------------------------1.5 9.26 2.0 4.63 2.5 3.09 3.0 2.31 3.5 1.85 The plot is shown in Figure 4.12. Relation (4.8) which is repeated below for convenience, represents an n channel MOSFET and we observe that, in saturation, the drain current i D is independent of the drain voltage v DS
1 W 2 i D = -- k n ---- ( v GS V T ) 2 L for v DS v GS V T
(4.11)
Therefore, we conclude that in the saturation mode the n channel MOSFET behaves as an ideal current source whose value is as in (4.11).
410
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
V GS = 3.5 V
iD in mA
In analogy with the transconductance in bipolar junction transistors, the MOSFET transconductance is defined as
i D g m = ----------v GS
(4.12)
v DS = cons tan t
In other words, the transconductance is a measure of the sensitivity of drain current to changes in gatetosource bias. As indicated in the previous chapter, subscripts in upper case represent the sum of the quiescent and small signal parameters, and subscripts in lower case represent just the small signal parameters. Thus, v GS = V GS + v gs and (4.11) can be expressed as
2 2 1 W 1 W i D = -- k n ---- ( V GS + v gs V T ) = -- k n ---- [ ( V GS V T ) + v gs ] 2 L 2 L 2 W W 1 W 2 1 = -- k n ---- ( V GS V T ) + k n ---- ( V GS V T )v gs + -- k n ---- v gs L L 2 L 2
For small signals, transconductance is defined in terms of the second term of the above expression and thus
W i d = k n ---- ( V GS V T )v gs L
Letting
and substituting the last expression above into (4.12) we obtain Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
411
(4.13) (4.14)
or
id W g m = ------- = k n ---- ( V GS V T ) L v gs
(4.15)
v GS = cons tan t
(4.16)
Therefore the output conductance is found by differentiation of the last expression above as
i D W g o = ----------- = k n ---- ( v GS V T v DS ) L v DS
(4.17)
and one that operates in both the depletion and enhancement modes are shown in Figure 4.13(b).
412
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v GS ( V )
Depletion mode
Enhancement mode
i D ( mA ) 30 25 20 15 I DSS 10 5
3 0 2 1 1 2 ( b ) Depletion Enhancement mode
v GS ( V )
As noted in Figure 4.13(b), I DSS is the value of the drain current in saturation with v GS = 0 . Figure 4.14 shows typical i D vs vDS characteristics for a depletiontype n channel MOSFET. Example 4.3 It is known that for a depletiontype n channel MOSFET, k n = 0.3 mA V , W = 100 m ,
L = 10 m , and V T = 3 V .
2
a. At what value should the voltage v DS be set so that this device will be operating in the saturation region when v G S = 2 V ? Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
413
vG S = 5
b. What would the drain current i D be if the voltage v DS is set to its minimum value to keep the device in the saturation region? Solution: a. The dotted curve of the i D vs vDS characteristics of Figure 4.14 separates the saturation region from the triode region. Therefore, the voltage v DS should be such that
v DS v GS V T 2 ( 3 ) 5 V
b. Relation (4.11) indicates that at saturation the current iD is independent of the voltage v DS . Therefore,
1 W 1 100 2 2 i D = -- k n ---- ( v GS V T ) = -- 0.3 -------- [ 2 ( 3 ) ] = 37.5 mA 2 L 2 10
414
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(4.18)
and if the device operates in the saturation mode, the drain current i D is found from
1 W 2 i D = -- k p ---- ( v GS V T ) 2 L for v DS v GS V T *
(4.19)
The constant k p in relations (4.18) and (4.19) is analogous to k n for the n channel MOSFET and typical values of k p are around 10 A V . Example 4.4 I t i s k n o w n t h a t f o r a n e n h a n c e m e n t t y p e p channel M O S F E T , k p = 8 mA V ,
W = 100 m , L = 10 m , and V T = 1.5 V . If v G = 0 , and v S = 5 V , compute the drain cur2 2
rent i D if: a. v D = 4 V b. v D = 1.5 V c. v D = 0 V d. v D = 5 V Solution: As stated above, to create a channel v GS must be equal or less than V T , and v DS must be negative and using the conditions specified in (4.18) and (4.19) we will determine whether the device is in the triode or saturation mode. For convenience, we list the following conditions:
v GS = v G v S v GS < V T v DS = v D v S
* As with the nchannel MOSFET, the drain current i D is not exactly independent of the draintosource voltage v DS . It increases with increasing v DS due to the socalled channel width modulation caused by reduction of the effective channel length, and since i D is inversely proportional to the channel length, i D increases with v DS and thus (4.18) is an approxi1 W 2 mation to the exact drain current i D given by i D = -- k n ----- ( v GS V T ) ( 1 + v DS ) where vGS , v DS , V T , and are all
negative.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
415
a.
and we conclude that the device operates in the triode mode. Then, with (4.17) we obtain
W 1 2 10 2 i D = k p ---- ( v GS V T )v DS -- v DS = 8 ----- [ ( 5 ( 1.5 ) ) ( 1 ) ] 0.5 ( 1 ) = 0.24 mA L 2 1
b.
and we conclude that the device operates at the point where the triode mode ends and the saturation region begins. Then, with (4.19) we obtain
1 W 2 2 i D = -- k p ---- ( v GS V T ) = 0.5 8 10 [ 5 ( 1.5 ) ] = 0.49 mA 2 L
c.
and we conclude that the device operates well in the saturation region.Then, with (4.19) we obtain
1 W 2 2 i D = -- k p ---- ( v GS V T ) = 0.5 8 10 [ 5 ( 1.5 ) ] = 0.49 mA 2 L
d.
v DS = v D v S = 5 5 = 10 V v GS V T = 5 ( 1.5 ) = 3.5 V
416
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
and we conclude that the device operates deeply in the saturation region.Then, with (4.19) we obtain
1 W 2 2 i D = -- k p ---- ( v GS V T ) = 0.5 8 10 [ 5 ( 1.5 ) ] = 0.49 mA 2 L
iD
RD
D B S
V DD
+
vD
v GS
Figure 4.15 shows that the substrate is connected to the source; this is a common practice with MOSFET devices. From Figure 4.15,
v D = V DD R D i D iD = ID + id v D = V DD R D ( I D + i d )
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
417
30 25 20 15 10 5 v GS ( V )
Depletion mode
Enhancement mode
i D ( mA ) 30 25 20 15 I DSS 10 5
3 0 2 1 1 2 ( b ) Depletion Enhancement mode
v GS ( V )
and
vd = RD id
id g m = ------v gs i d = g m v gs
418
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(4.20)
and the minus () sign indicates 180 phase reversal. Relation (4.20) reveals that high gains can be achieved with increased drain resistance R D . However, external resistors are not used in IC MOSFETS because they require large areas and their tolerance may cause some undesirable effects. Instead, constant current sources are used, and the most commonly used current sources are the socalled current mirrors. A typical current mirror is shown in Figure 4.17.
RD D B S G G
V DD
i1
D B S
i2
v GS
A current mirror is essentially an adjustable current regulator. In the circuit of Figure 4.17, both MOSFETs share the same voltage v GS and the ratio of the currents i 1 and i 2 is directly related to the ratio of the geometry of the transistors, that is,
i1 ( W L )1 --- = ------------------i2 ( W L )2
and if the two MOSFETs have the same geometry, then i 1 = i 2 , and hence the name current mirror. A similar current mirror circuit can be implemented with bipolar transistor. MOSFET devices can be used as amplifiers provided that they operate in the saturation region. However, the CMOS devices, discussed on the next section, are the most common amplifiers in the FET technology.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
419
i2 i1 v in
v out
B S
A typical commonsource CMOS amplifier can provide gains of 20 to 100 , and the output is 180 outofphase with the input. It exhibits both high input and high output resistances.
420
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v out
V DC v in
V DD
v in
i2 S
D
D B S
i1
G G
v out
B S
V SS
421
The main advantage of the MESFET is the higher mobility also referred to as surface mobility of the carriers in the channel as compared to the silicontype MOSFETs. The higher mobility results in a higher current, transconductance and transit frequency of the device. However, the presence of the Schottky metal gate limits the forward bias voltage on the gate to the turnon voltage of the Schottky diode. This turnon voltage is typically 0.7 V for GaAs Schottky diodes. The threshold voltage therefore must be lower than this turnon voltage. As a result it is more difficult to fabricate circuits containing a large number of enhancementmode MESFET. The higher transit frequency of the MESFET makes it particularly of interest for microwave circuits. While the advantage of the MESFET provides a superior microwave amplifier or circuit, the limitation by the diode turnon is easily tolerated. Typically depletionmode devices are used since they provide a larger current and larger transconductance and the circuits contain only a few transistors, so that threshold control is not a limiting factor. The buried channel also yields a better noise performance as trapping and release of carriers into and from surface states and defects is eliminated. The use of GaAs rather than silicon MESFETs provides two more significant advantages: first of all the room temperature mobility is more than 5 times larger, while the saturation velocity is about twice that in silicon, and second it is possible to fabricate semiinsulating (SI) GaAs substrates which eliminates the problem of absorbing microwave power in the substrate due to free carrier absorption.
iE
i B2
Base 2
422
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
The Diac
In operation, a positive voltage is applied to B 2 and B 1 is placed at ground potential. The
B 2 E B 1 junctions then act like a voltage divider which reversebiases the emitter junction.
An external voltage having a potential higher than this reverse bias will forwardbias the emitter and inject holes into the silicon bar which move toward B 1 . The emitter B 1 resistance then decreases, and this, in turn, causes the emitter voltage to decrease as the emitter current increases, and a negative resistance characteristic is obtained as shown in Figure 4.22.
Figure 4.22. UJT emitter characteristics curve with important parameters (Courtesy of General Electric)
On the emitter characteristics curve of Figure 4.22, the points of interest are the peak point, and the valley point. The region to the left of the peak point is called the cutoff region, and in this region the emitter is reverse biased and only a small leakage current flows. The region between the peak point and the valley point is referred to as the negative resistance region. The region to the right of the valley point is the saturation region and as we can see, the resistance in this region is positive. Device 2N2646 is a popular UJT and can be used for the design of pulse and sawtooth generators, analogtodigital converters, relay time delay circuits, and frequency dividers. Other UJT devices, referred to as programmable UJTs, can have their parameters set by external components such as resistors and capacitors. Device 2N6027 is known as a programmable UJT.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
423
i 12 N P
2
i i 21
IS VS IS VS
i 12
Structure 2 Symbol Characteristics
Normally, the diac does not conduct (except for a small leakage current) until the breakover voltage V S is reached, typically 20 to 40 volts. At that point the diac goes into avalanche conduction also at that point the device exhibits a negative resistance characteristic, and the voltage drop across the diac snaps back, typically about 5 volts, creating a breakover current I S in the order of 50 to 200 A sufficient to trigger a triac or SCR. The negativeresistance characteristic of the diac makes it useful for very simple relaxation oscillators* and pulse generators, but its major application is in conjunction with a triac, to be discussed in Section 4.8, to produce ac phasecontrol circuits useful for motorspeed control, light dimming, and other AC powercontrol functions.
* Relaxation oscillators. are circuits that generate nonsinusoidal waveforms such as pulse and sawtooth generators. An earlier gas filled tube device called a Thyratron provided a similar electronic switching capability, where a small control voltage could switch a large current.
424
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
P N
Gate
P N
Gate
N P N
Gate Cathode
Gate Cathode
P N
Cathode
Cathode
Figure 4.24. Parts of an SCR, the twotransistor equivalent circuit, and its symbol
SCRs are mainly used where high currents and voltages are involved, and are often used to control alternating currents, where the change of sign of the current causes the device to automatically switch off. Like a diode, an SCR conducts only in one direction. A similar 5layer device, called a triac, to be discussed on the next section, conducts current in both directions. Modern SCRs can switch large amounts of power (up to megawatts). In the realm of very high power applications, they are still the primary choice. However, in low and medium power (from few tens of watts to few tens of kilowatts) they have almost been replaced by other devices with superior switching characteristics like MOSFETs. While an SCR is that is it not a fully controllable switch in the sense that triggering current direction need to be reversed to switch it off, a newer device, the gate turnoff SCR (GTO) can be turned on and off with a signal applied to the gate. The turnon signal is a small positive voltage and the turnoff is a negative small signal. GTOs are used for the output stages of mediumvoltage, high horsepower, variable frequency drives. In highfrequency applications, SCRs are poor candidates due to large switching times arising out of bipolar conduction. MOSFETs, on the other hand, has much faster switching capability because of its unipolar conduction (only majority carriers carry the current). Figure 4.25 shows the voltampere characteristics of a typical SCR.
+I i
PIV
+V
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
425
Conduction angle
Firing angle
As shown in Figure 4.26, the number of degrees from the beginning of the cycle until the SCR is gated to the on condition is referred to as the firing angle, and the number of degrees that the SCR remains conducting is known as the conduction angle. For accurate SCR gating, the firing circuit must be synchronized with the AC line voltage being applied anodetocathode across the device. Without synchronization, the SCR firing would be random in nature and the system response will be erratic. Also, when the firing angle is greater than zero, the voltage applied to the load is no longer sinusoidal. This presents no problem in the case of motor loads, but for radio and television an interference is created and usually the manufacturer of the SCR equipment will include an electromagnetic interference (EMI) filter to rectify the problem. In closedloop systems, such as motor control, an Error Detector Circuit computes the required firing angle based on the system setpoint and the actual system output. The firing circuit is able to sense the start of the cycle, and, based on an input from the Error Detector, delay the firing pulse until the proper time in the cycle to provide the desired output voltage. An analogy of a firing circuit would be an automobile distributor which advances or retards the spark plug firing based on the action of the vacuum advance mechanism. In analog control systems the error detector circuit is usually an integrated circuit operational
426
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
vC
R1
v in
R2
The SCR in the circuit of Figure 4.27 is switched to the on (conducting) state by applying a positive triggering pulse at v in and with the SCR on, there is there is a very small voltage drop from anode to cathode. The capacitor then charges to voltage v C and this voltage is approximately equal to V S . The SCR will be turned off by closing the switch S , and when this occurs, we observe that the positive side of the capacitor is connected to the ground. Since the capacitor voltage cannot change instantly, the anode of the SCR becomes negative with respect to the cathode. The Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
427
vC
R1 R2
+
R5 R6
v in1
v in2
If a train of positive pulses is applied simultaneously at v in1 and v in2 they have no effect on the rectifier that is conducting, but they turn the nonconducting SCR on. Again, with the aid of the capacitor this action turns the conducting SCR off. Thus a train of positive pulses applied simultaneously to the gates of the SCRs causes them to switch on and off alternately.
v out
vC S
v out
T
428
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
R3 R1
vC
R2
iC
C
R4 VS
v out
v in
In the circuit of Figure 4.30 the SCR performs the function of the switch in the ideal generator of Figure 4.29. Resistor R 4 is a large resistor in the order of 1 M and this with the voltage source VS approximate the current source in the ideal sawtooth generator; since a true current source is not used, the output voltage shown in Figure 4.31 is not exactly linear.
VS Vf Vx
v out
V = Vf V x
t I C max Tc Td iC
The voltage v in is the control voltage for the rectifier, and R 3 is a resistance that limits the rectifier current to a safe value when the capacitor is discharging.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
429
(4.21)
When t = 0 , v out has its minimum value V x as shown in Figure 4.31; hence letting t = 0 in (4.21) we obtain
v out = V S + A = V x
A = Vx VS
( t R4 C )
v out = V S ( V S V x )e
(4.22)
* For an introduction to transient analysis, refer to Circuit Analysis I with MATLAB Applications, ISBN 9780 970951120.
430
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(4.23) (4.24)
(4.25)
The exponential factor in (4.25) can now be expanded in a power series to obtain the following more useful expression:
2 2 1 1 t t t t - - v out = V' S + V x V' S 1 ---------- + -- ---------- = V x + V' S ---------- -- V' S ---------- + R 4C R 4C 2 R 4C 2 R 4C
(4.26)
Since we want the output voltage v out to be a linear function of time, our sawtooth generator must be designed so that the second term on the right side of (4.26) should be sufficiently small. Then,
t v out = V x + V' S ---------R 4C
(4.27)
The rightband side of (4.26) is a converging alternating series; hence the error in truncating the series at any point is smaller than the first term in the part out off. Thus, the magnitude of the error at any instant is smaller than
1 t 2 v out = -- V' S ---------- R 4C 2
(4.28)
(4.29)
It now follows from (4.29) and Figure 4.30 that the amplitude of the sawtooth wave is
Tc V = V f V x = V' S ---------R C
4
(4.30)
and the error at the end of the charging interval, which is the amount by which the actual amplitude is less than the amplitude given by (4.30), is smaller than
Tc 2 1 v out = -- V' S ---------- R 4C 2
(4.31)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
431
(4.32) (4.33)
Relation (4.33) reveals that the fractional error depends only on the relative sizes of the sawtooth amplitude and the effective DC supply voltage; in order to have good linearity it is necessary to have an effective supply voltage much larger than the required amplitude of the sawtooth. Good linearity is often obtained by the direct expedient of using a large supply voltage or by generating a sawtooth wave of small amplitude. In more sophisticated designs the effect of a very large supply voltage, often thousands of volts, is simulated by the use of additional circuitry in which amplifiers are frequently used. One method of simulating a large supply voltage is illustrated in Example 4.5 at the end of this section. When the capacitor is charged from an ideal current source, as in Figure 4.28, the effective supply voltage is the opencircuit voltage of the source; this voltage is infinite, giving zero error. In designing a sawtooth generator, V S and V' S must be chosen so that the fractional error given by (4.33) is sufficiently small. The time constant R 4 C can then be chosen to produce the desired slope to the sawtooth. With the control voltage v in in Figure 4.29 held constant, the firing voltage V f and the sawtooth amplitude V are also constant, and the slope of the sawtooth determines the length of the charging period. From Equation (4.30),
V T c = R 4 C ------V' S
(4.34)
and when a train of triggering pulses is applied at v in , T c is determined by the interval between the pulses, and the slope of the sawtooth determines the amplitude V . With a fixed slope, V is a measure of the time interval T c . At the end of the charging interval the SCR switches on, and the capacitor discharges exponentially through the SCR at a rate determined by the resistance R 3 in Figure 4.29. During the discharging interval the voltage across the capacitor changes by the amount
Qf Q x V = Vf V x = -----------------C
(4.35)
The duration of the discharging interval T d can be determined from this relation. To simplify the
432
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(4.36)
Also, since the capacitor current I C max is approximately equal to the current I max through the SCR, we can express (4.36) as
T d I max V -----------------2C
For good linearity we must have V V' S , and since the voltage drop across R 4 during the charging interval is approximately V' S , and V' S R 4 is very nearly the charging current I C for the capacitor, (4.37) can be expressed as
Td 2I C ----- = ----------Tc I max
(4.40)
where I C is the charging current for the capacitor, and I max is the peak discharging current through the SCR. To provide a relatively short discharge time, the peak SCR current must be much greater than the capacitor charging current. Any load that is connected across the output terminals of the sawtooth generator in Figure 4.30 will affect the operation of the circuit, and any leakage resistance that develops across the capacitor has a similar effect. Figure 4.32(a) shows a sawtooth generator with a load resistor R L connected across the output terminals.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
433
x iC R4 + C vC +
VS
RL y
v out
(a)
i R3
SCR Circuit
+
iC C
vC
R TH
V TH
(b) Figure 4.32. A sawtooth generator with resistive load and its Thevenin equivalent
To simplify the circuit of Figure 4.32(a), we apply Thevenins theorem at points x y and we obtain the circuit of Figure 4.32(b) where
RL V TH = ------------------- VS R 4 + RL
(4.41)
and
R 4 RL R TH = ------------------R 4 + RL
(4.42)
It is then evident that the addition of the load resistor R L affects both the slope and the linearity of the sawtooth waveform. The presence of R L affects the effective DC supply voltage, and hence on the linearity, is especially severe. Therefore, it becomes necessary to make R L greater than 1 M . A load resistor of about 100 K would reduce the effective supply voltage and will increase the fractional error by a factor of more than 10 . It is quite possible in such cases for the effective voltage V TH to be less than the firing voltage V f , for the SCR; then the rectifier never fires, and no sawtooth wave is generated. Accordingly, the output from the sawtooth generator must be connected to a very high impedance device such as a MOSFET. The output can also be delivered to a bipolar junction transistor, but a very special circuit configuration must be used to provide the required high input resistance.
434
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v out
(a)
20 V
1 mA VS R1
R4
R2
R3
v out
(b)
In order to examine the circuits of Figure 4.33 in more detail, it is helpful first to simplify and rearrange the circuit. During the charging interval the SCR conducts no current, and it can be omitted from the circuit. If V S is much larger than the small emittertobase voltage drop, the transisElectronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
435
vC
vC
v out rC
20 V
rC v out i E r C
i E = 1 mA
VS R1 E
i E
B (a)
(b)
Figure 4.34. Piecewiselinear and equivalent circuits for the sawtooth generator of Figure 4.33
In most circuits the collector resistance is in parallel with a much smaller load resistance and it can be neglected; this is not the case in the circuit of Figure 4.34(a), however, and the collector resistance has an important effect. As the next step in the simplification of the circuit, it is noted that because of the short circuit between base and ground, the emitter circuit affects the collector circuit only through the action of the controlled source i E . Thus when i E is known, the emitter circuit can be ignored. And finally, the current source and its shunt resistance can be converted to an equivalent voltage source with a series resistance. The resulting equivalent circuit is shown in Figure 4.34(b). From relation (4.23), the effective supply voltage is
V' S = V S V x
(4.43) (4.44)
The equivalent circuit of Figure 4.34(b) shows that by charging the capacitor through the transistor an effective supply voltage of
V' S = 20 + i E r C V x
(4.45)
is obtained. The extinction voltage V x , is a negligible fraction of a volt, and the emitter current
i E is 1 ma. Thus taking typical values of = 0.98 and r C = 1 M , relation (4.45) yields V' S = 20 + 980 = 1000 V
(4.46)
436
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(4.47) (4.48)
and thus
V = 20 V
Thus, in principle the amplitude of the sawtooth output can be as large as the DC supply voltage; however, to avoid possible non linearities in the transistor characteristics at low voltages and to allow for possible variations in the supply voltage, a value of 15 volts might be chosen for V . Accordingly, the values of resistors R 2 and R 3 should be chosen to make the gateterminal potential about 5 volts when the SCR is switched off. To limit the SCR current to the maximum safe value of 100 ma, the value of resistor R 4 should be
V15 R 4 = ---------- = ------ = 150 I max 0.1
(4.49)
(4.50)
From (4.40)
(4.51)
(4.52) (4.53)
T c = 0.00098 sec V T c = R 4 C ------V' S T c V' S 0.00098 1000 - C = ----- ------- = ------------------ ----------- = 0.0655 F 6 R4 V 15 10
(4.54)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
437
Chapter 4 Field Effect Transistors and PNPN Devices 4.8 The Triac
The triac is a device capable of switching on for either polarity of an applied voltage. As can be seen in Figure 4.35, the triac has a single gate lead and is therefore the AC equivalent of the SCR. A triac is essentially two SCRs connected back to back with a common gate and common terminals where each terminal is the anode of one SCR and the cathode of another.
1
i 21
i 12
2
Figure 4.36 shows an SCR circuit and a triac circuit for comparison. The diode D in the SCR circuit ensures a positive trigger voltage. Figure 4.37 shows a comparison of the waveforms at the input, gate, and output of these two devices.
Triac R1 RL R2
v out
AC Power Supply
v out
438
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Output voltage
The triac exhibits voltagecurrent characteristics similar to those of the SCR. Applications for triacs include AC motorspeed control, AC light dimmers, and general AC powercontrol applications.
P N P N
Cathode
P N P N P N
Cathode Cathode
Figure 4.38. Parts of a Shockley diode, the twotransistor equivalent circuit, and its symbol
A Shockley diode can be to turned on by applying sufficient voltage between anode and cathode. This voltage will cause one of the transistors to turn on, which then turns the other transistor on,
* The Shockley diode should not be confused with the Schottky diode, the twolayer metalsemiconductor device known for its high switching speed. We discussed the Schottky diode in Chapter 2.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
439
440
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
P N
Gate 2 Gate 1
P N P N P
Gate 2
P N
Cathode
G2 Cathode Cathode
Figure 4.39. Parts of an SCS, the twotransistor equivalent circuit, and its symbol
As shown in Figure 4.39, the basic construction of the SCS is the same as for the SCR, with the addition of a second gate lead. Thus the SCS has an anode, a cathode, an anode gate, and a cathode gate. The SCS has two advantages over the SCR and the Shockley diode. First, because both gate regions are accessible, they can be biased so as to completely cancel the rate effect we described with the fourlayer diode. Second, since we can now control both end junctions, we can actively turn the SCS off without the need to reduce the applied voltage or current.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
441
Figure 4.40. A simplified arrangement of a latch as a possible replacement for the transistor
442
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v DS = cons tan t
In other words, the transconductance is a measure of the sensitivity of drain current to changes in gatetosource bias. A MOSFET is said to operate in the depletion mode when a negative voltage is applied at the gate. The JFET also operates in this manner. A MOSFET is said to operate in the enhancement mode if a positive voltage is applied at the gate. A MOSFET can be constructed that will operate in either mode depending upon what type of bias is applied, thus allowing a greater range of input signals. In a MOSFET, the voltage at which the channel is closed is known as the pinchoff voltage V P , and the minimum voltage required to form a conducting channel between the drain and source is referred to as the threshold voltage and it is denoted as V T . Typical values for V T are 2 to 4 volts for high voltage devices with thicker gate oxides, and 1 to 2 volts for lower voltage devices with thinner gate oxides. In a MOSFET, the three regions of operation are the cutoff, the quadratic or triode, and the saturation or pentode. In the cutoff mode, v GS < V T and no drain current flows. In the quadratic region the drain current is determined from the expression
W 1 2 i D = k n ---- ( v GS V T )v DS -- v DS L 2 for v DS v GS V T
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
443
where k n is a constant depending on electron mobility, and oxide thickness, permittivity, and capacitance for nchannel MOSFETs. For pchannel MOSFETs this parameter is denoted as k p . The ratio W L denotes the ratio of the channel width W to channel length L . Typical values of k n are around 20 A V , for k p , around 10 A V , and for W L about 10 . For small signals, transconductance is defined as
id W g m = ------- = k n ---- ( V GS V T ) L v gs
2 2
v GS = cons tan t
and the minus () sign indicates 180 phase reversal. In high density integrated circuits, current mirrors are normally used in lieu of resistors. Complementary MOS or CMOS technology combines one NMOS device and one PMOS device into a single device referred to as CMOS. These devices are used extensively in both analog and digital circuits, and integrated circuits. Also, CMOS devices are the most common amplifiers in the FET technology. In CMOS devices only one of its components is on at any given time, that is, either the NMOS device is on and the PMOS device is off, or vice versa. Thus, CMOS chips require less power than chips using just one type of a MOSFET, and unlike bipolar transistors, a CMOS has almost no static power dissipation. A CMOS device can be configured as a commonsource amplifier, commongate amplifier, and commondrain or source follower. The MetalSemiconductorFieldEffectTransistor (MESFET) consists of a conducting channel positioned between a source and drain contact region. The carrier flow from source to drain is controlled by a Schottky metal gate. The control of the channel is obtained by varying the depletion layer width underneath the metal contact which modulates the thick-
444
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Summary
ness of the conducting channel and thereby the current MESFETs use GaAs (gallium arsenide) technology. Only n channel MESFETS are available because holes have a slower mobility. The unijunction transistor (UJT) is a threeterminal, singlejunction device which exhibits negative resistance and switching characteristics totally unlike those of conventional bipolar transistors. UJTs can be used for the design of pulse and sawtooth generators, analogtodigital converters, relay time delay circuits, and frequency dividers. Other UJT devices, referred to as programmable UJTs, can have their parameters set by external components such as resistors and capacitors. The diac is a twoterminal, transistorlike component which exhibits bistable switching for either polarity of a suitably high applied voltage. The diac closely resembles a PNP transistor without an external base terminal. Its major application is in conjunction with a triac to produce AC phasecontrol circuits useful for motorspeed control, light dimming, and other ac powercontrol functions. The silicon controlled rectifier (SCR) is a fourlayer semiconducting device, with each layer consisting of an alternately N or Ptype material, for example NPNP. The main terminals, labelled anode and cathode, are across the full four layers, and the control terminal, called the gate, is attached to one of the middle layers. Another name for the SCR is thyristor. It is similar to a diode with an additional terminal that is used to turn it on. Once turned on, the SCR will remain on as long as current flows through it. If the current falls to zero, the SCR behaves like an open switch. The triac is a device capable of switching on for either polarity of an applied voltage; It is the AC equivalent of the SCR. The Shockley diode, (not to be confused with the Schottky diode), or fourlayer diode, or PNPN diode, is a fourlayer sandwich of PNPN semiconductor material very similar to the SCR but without a gate. The silicon controlled switch (SCS) is very similar to the SCR but all four regions are accessible to the external circuit.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
445
V DD
v out
v+ in
S
3. Draw an equivalent circuit that represents the n channel MOSFET in the saturation mode. 4. An n channel MOSFET with k n = 50 10 , W = 10 m , L = 1 m , and V T = 1 V , is biased with v GS = 3 V and v DS = 5 V . a. Compute the drain current i D b. Compute the transconductance g m c. Compute the output conductance g o if V DS = 2 V d. Compute the output conductance g o if V DS = 0 V 5. An enhancementtype
p channel
6 6
MOSFET
with
k p = 10 10
W = 10 m ,
voltage of v D that will keep the device in saturation? 6. In the circuit below, the SCR is used to control the power delivered to the 50 load by the sinusoidal source. As shown, the gate supply V GG is adjustable. a. Over what range may the conduction angle of the SCR be continuously varied? b. Over what range may the load DC current be continuously varied if the frequency is 60 Hz ?
446
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
R load 50 RG V GG
100 sin t
7. The circuit below is in steady state with the switch open and the SCR is in the conducting state. The voltage drop across the SCR is 1 V while it is conducting. The switch is then closed at t = 0 , and it is assumed that the SCR switches instantly to the nonconducting state and remains nonconducting thereafter. a. What are the initial and final values of the voltage across the capacitor with the polarities shown? b. If the anode potential of the SCR must remain negative for 10 s to ensure that the SCR switches to the nonconducting state, what is the minimum value of the capacitor C that should be used?
V DC R3 1 K R1 10 K R2 100 sin t 20 K C S 50 V R4 50 K
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
447
Chapter 4 Field Effect Transistors and PNPN Devices 4.14 Solutions to EndofChapter Exercises
1. By definition
g m = di DS dv GS = di DS dv in (1)
and since it follows that and with (1) Thus, the voltage gain is 2. From (4.10),
3. In the saturation mode the n channel MOSFET behaves as an ideal current source whose value is as in (4.11) and it can be represented by the equivalent circuit shown below.
G
iD
D
v GS
S
4. a. Since v DS > v GS V T 5 > 3 1 , the device is in saturation and the drain current i D is found from (4.11). Thus,
6 10 1 W 1 2 2 i D = -- k n ---- ( v GS V T ) = -- 50 10 ----- ( 3 1 ) = 1 mA 2 L 2 1
448
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
This result is not surprising. With reference to Figure 4.14 we observe that for v DS v GS V T the slope is zero. d. If V DS = 0 V , the device is in the triode region since v DS < v GS V T 0 < 3 1 . Then,
6 10 1 W g o = k n ---- ( v GS V T v DS ) = 50 10 ----- ( 3 1 0 ) = 1 m 1 L
6.
1
100 sin t V GG
0.5
-0.5
firing angle
0 1 2 3
-1
a. The SCR will fire for the interval 0 t 2 or not at all depending on the value of V GG . b. The load DC current I DC is the average value for the interval 0 2 and its value is determined by the integral
I DC
1 100 sin t 1 ----- ---------------------- d( t ) = -- ( cos t ) = 50 2
If = 0 , I DC = 2 = 0.637 A If = 2 , I DC = 1 = 0.318 A If SCR never fires, I DC = 0 Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
449
a. Just before the switch closes, the voltage drop across the SCR is 1 V and the voltage drop across the capacitor is 49 V with the polarity as shown. The capacitor voltage cannot change instantaneously so immediately after the switch is closed, the capacitor voltage will be 49 V as shown in the figure below, and the SCR will be Off.
R3 50 V 1 K R4 C
V DC
50 K
The capacitor voltage will eventually charge to 50 V with respect to the ground and thus it will undergo a change from 49 V (initial value) to +50 V (final value). b. We want the SCR anode to remain negative for 10 s , that is, until the voltage across the capacitor reaches the value of 0 V . Using the general formula for the capacitor voltage, we must have or
v C ( t ) = V ( V V initial )e 0 = 50 [ 50 ( 49 ) ]e e
t ( RC ) t ( RC )
t ( RC )
= 50 99
t ( R 3 C ) = ln ( 50 99 )
6 3 t R3 ( 10 10 ) 10 C = ------------------------- = ------------------------------------------- = 14.64 nF 0.683 ln ( 50 99 )
450
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Chapter 5
Operational Amplifiers
his chapter begins with an introduction to operational amplifiers (op amps), characteristics, and applications. We will discuss the ideal op amp, analysis of circuits in the inverting and noninverting configurations, and gain and bandwidth on circuit performance. We will also introduce some circuits consisting of op amps and nonlinear devices, and analog computers.
2 3
As shown above the op amp has two inputs but only one output. For this reason it is referred to as differential input, single ended output amplifier. Figure 5.2 shows the internal construction of the popular 741 op amp. This figure also shows terminals V + and V . These are the voltage sources required to power up the op amp. Typically, V + is +15 volts and V is 15 volts. These terminals are not shown in op amp circuits since they just provide power, and do not reveal any other useful information for the op amps circuit analysis.
51
resistor R as shown in Figure 5.3. For the circuit of Figure 5.3, the voltage gain G v is
v out Rf G v = --------- = ------v in R in
(5.1)
52
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
+ v in
R
Rf
+ v out
Note 1: In the inverting mode, the resistor R connected between the noninverting (+) input and ground serves only as a current limiting device, and thus it does not influence the op amps gain. So its presence or absence in an op amp circuit is immaterial. Note 2: The input voltage v in and the output voltage v out as indicated in the circuit of Figure 5.3, should not be interpreted as open circuits; these designations imply that an input voltage of any waveform may be applied at the input terminals and the corresponding output voltage appears at the output terminals. As shown in the relation of (5.1), the gain for this op amp configuration is the ratio R f R in where R f is the feedback resistor which allows portion of the output to be fed back to the input. The minus () sign in the gain ratio R f R in implies that the output signal has opposite polarity from that of the input signal; hence the name inverting amplifier. Therefore, when the input signal is positive (+) the output will be negative () and vice versa. For example, if the input is +1 volt DC and the op amp gain is 100, the output will be 100 volts DC. For AC (sinusoidal) signals, the output will be 180 outofphase with the input. Thus, if the input is 1 volt AC and the op amp gain is 5, the output will be 5 volts AC or 5 volts AC with 180 outofphase with the input. Example 5.1 Compute the voltage gain G v and then the output voltage v out for the inverting op amp circuit shown in Figure 5.4, given that v in = 1 mV . Plot v in and v out as mV versus time on the same set of axes.
R in R f 120 K
+ v in 20 K
+ + v out
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
53
v in = 1 mV
mV
v out = 6 mV
-6
Figure 5.5. Input and output waveforms for the circuit of Example 5.1
Example 5.2 Compute the voltage gain G v and then the output voltage v out for the inverting op amp circuit shown in Figure 5.6, given that v in = sin t mV . Plot v in and v out as mV versus time on the same set of axes.
R in 20 K R f 120 K
+ v in
+ + v out
54
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
mV
Figure 5.7. Input and output waveforms for the circuit of Example 5.2
+
R
+ v in
+ v out
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
55
(5.2)
As indicated by the relation of (5.2), the gain for the noninverting op amp configuration is 1 + R f R in and therefore, in the noninverting mode the op amp output signal has the same polarity as the input signal; hence, the name noninverting amplifier. Thus, when the input signal is positive (+) the output will be also positive and if the input is negative, the output will be also negative. For example, if the input is +1 mV DC and the op amp gain is 75 , the output will be +75 mV DC . For AC signals the output will be inphase with the input. For example, if the input is 0.5 V AC and the op amp gain is G v = 1 + 19 K 1 K = 20 , the output will be
10 V AC and inphase with the input.
Example 5.3 Compute the voltage gain G v and then the output voltage v out for the noninverting op amp circuit shown in Figure 5.9, given that v in = 1 mV . Plot v in and v out as mV versus time on the same set of axes.
R in 20 K Rf 120 K
+ + v out
+ v in
and thus
v out = G v v in = 7 1 mV = 7 mV
56
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
mV
v out = 7 mV v in = 1 mV
1 0 t
Figure 5.10. Input and output waveforms for the circuit of Example 5.3
Example 5.4 Compute the voltage gain G v and then the output voltage v out for the noninverting op amp circuit shown in Figure 5.11, given that v in = sin t mV . Plot v in and v out as mV versus time on the same set of axes.
R in 20 K Rf 120 K
+ + v out
+ v in
Solution: This is the same circuit as in the previous example except that the input is a sinusoid. Therefore, the voltage gain G v is the same as before, that is,
v out Rf G v = --------- = 1 + ------- = 1 + 120 K = 1 + 6 = 7 -------------------v in R in 20 K
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
57
mV
4 2 0 0 -2 -4 -6 -8 0
v out = 7 sin t
v in = sin4t 2
10
12
14
Figure 5.12. Input and output waveforms for the circuit of Example 5.4
v+ out
and thus
v out = v in
For this reason, the op amp circuit of Figure 5.13 it is called unity gain amplifier. For example, if the input voltage is 5 mV DC the output will also be 5 mV DC , and if the input voltage is 2 mV AC , the output will also be 2 mV AC . The unity gain op amp is used to provide a very high resistance between a voltage source and the load connected to it. An application is presented in Example 5.8.
58
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Active Filters
is shown in Figure 5.14(a) and its amplitude frequency response in Figure 5.14(b).
1 R2 R1 Low pass filter frequency response Ideal R in
C2
32
vin
C1
vout
(a)
v out v in
C
Radian frequency (log scale)
(b) Figure 5.14. An active lowpass filter and its amplitude frequency response
In Figure 5.14(b), the straight vertical and horizontal lines represent the ideal (unrealizable) and the smooth curve represents the practical (realizable) lowpass filter characteristics. The vertical scale represents the magnitude of the ratio of outputtoinput voltage v out v in , that is, the gain
G v . The cutoff frequency c is the frequency at which the maximum value of v out v in which is unity, falls to 0.707 Gv , and as mentioned before, this is the halfpower or the 3 dB point.
A highpass filter transmits (passes) all frequencies above a critical (cutoff) frequency c , and attenuates (blocks) all frequencies below the cutoff frequency. An op amp highpass filter is shown in Figure 5.15(a) and its frequency response in Figure 5.15(b).
High pass filter frequency response 1 32 Half power point Ideal Realizable
vin
R1
C3
vout
v out v in
C1
C2
R2
(a)
(b) Figure 5.15. An active highpass filter and its amplitude frequency response
In Figure 5.15(b), the straight vertical and horizontal lines represent the ideal (unrealizable) and Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
59
filter and its frequency response are shown below. An op amp bandpass filter is shown in Figure 5.16(a) and its frequency response in Figure 5.16(b). A bandelimination or bandstop or bandrejection filter attenuates (rejects) the band (range) of frequencies between the critical (cutoff) frequencies denoted as 1 and 2 , where the maximum value of G v which is unity, falls to 0.707 G v , while it transmits (passes) all frequencies outside this band. An op amp bandstop filter is shown in Figure 5.17(a) and its frequency response in Figure 5.17(b).
C1
C2
R2
R3
vin
R1
vout
(a)
11 0 32
8
v out v in
Realizable
Ideal
0 1 0
10 2 1
3
1 0
(b) Figure 5.16. An active bandpass filter and its amplitude frequency response
510
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
R3
C1
C2
vin
vout
R1 R2
C3 (a)
1 32 Ideal Half power points Realizable
v out v in
1 2
Radian frequency (log scale)
(b) Figure 5.17. An active bandelimination filter and its amplitude frequency response
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
511
i i
R
Rf
v in
v out
(5.3)
Proof: No current flows through the () input terminal of the op amp; therefore the current i which flows through resistor R in flows also through resistor R f . Also, since the (+) input terminal is grounded and there is no voltage drop between the () and (+) terminals, the () input is said to be at virtual ground. From the circuit of Figure 5.18,
v out = R f i
v in i = ------R in
f v out = ------- v in R in
The input and output parts of the circuit are shown in Figure 5.19 with the virtual ground being the same as the circuit ground.
512
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
R in
i i
Rf
+
v out
Figure 5.19. Input and output parts for the inverting op amp
These two circuits are normally drawn with the output as a dependent source as shown in Figure 5.20. This is the equivalent circuit of the inverting op amp and the dependent source is a Voltage Controlled Voltage Source (VCVS).* +
v in
R in
v + ------- v in out R
in
Rf
Example 5.6 The op amp circuit shown in Figure 5.21 is called noninverting op amp. Prove that the voltage gain G v is as given in (5.4) below, and draw its equivalent circuit showing the output as a dependent source.
R in Rf
v in
v out
(5.4)
Proof: Let the voltages at the () and (+) terminals be denoted as v 1 and v 2 respectively as shown in Figure 5.22.
* For a definition of dependent sources please refer to Circuit Analysis I with MATLAB Applications, ISBN 9780 970951120.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
513
Rf
i2 v1 v in v 2
i1
v out
By application of KCL at v 1
i1 + i2 = 0
or
(5.5)
Rearranging, we obtain
(5.6)
Figure 5.23 shows the equivalent circuit of Figure 5.22. The dependent source of this equivalent circuit is also a VCVS. + +
v in
+ 1 + ------- v in
R in
Rf
v out
514
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(5.7)
v out = v in
With R in open and R f shorted, the noninverting amplifier of the previous example reduces to the circuit of Figure 5.24.
v in
v out
The voltage difference between the (+) and () terminals is zero; then v out = v in . We will obtain the same result if we consider the noninverting op amp gain
Rf v out G v = --------- = 1 + ------v in R in
Then, letting R f 0 , the gain reduces to G v = 1 and for this reason this circuit is called unity gain amplifier or voltage follower. It is also called buffer amplifier because it can be used to buffer (isolate) one circuit from another when one loads the other as we will see on the next example. Example 5.8 For the circuit of Figure 5.25: a. With the load R load disconnected, compute the open circuit voltage v ab b. With the load connected, compute the voltage v load across the load R load Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
515
a b
v in 12 V
R load 5 K
5 K
Solution:
a. With the load R load disconnected the circuit is as shown in Figure 5.26.
7 K
a b
v in 12 V
5 K
R load 5 K
Figure 5.26. Circuit for Example 5.8 with the load disconnected
b. With the load R load reconnected the circuit is as shown in Figure 5.27. Then,
5 K || 5 K v load = ------------------------------------------------------- 12 = 3.16 V 7 K + 5 K || 5 K
Here, we observe that the load R load loads down the load voltage from 5 V to 3.16 V and this voltage may not be sufficient for proper operation of the load.
7 K
a b
v in 12 V
R load 5 K
5 K
Figure 5.27. Circuit for Example 5.8 with the load reconnected
c. With the insertion of the buffer amplifier between points a and b and the load, the circuit now is as shown in Figure 5.28.
516
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
7 K
a
5V
R load 5 K
+
v load = v ab = 5 V
v in 12 V
5 K
Figure 5.28. Circuit for Example 5.8 with the insertion of a buffer op amp
From the circuit of Figure 5.28, we observe that the voltage across the load is 5 V as desired. Example 5.9 The op amp circuit shown in Figure 5.29 is called summing circuit or summer because the output is the summation of the weighted inputs. Prove that for this circuit,
v out = R f v in 1 v in 2 --------- + --------- R R
in 1 in 2
(5.8)
R in 1 R in 2
Rf
+
v in 1
+ +
R
v out
v in 2
Proof:
We recall that the voltage across the () and (+) terminals is zero. We also observe that the (+) input is grounded, and thus the voltage at the () terminal is at virtual ground. Then, by application of KCL at the () terminal, we obtain
v in 1 v in 2 v out --------- + --------- + --------- = 0 R in 1 R in 2 Rf
(5.9)
and solving for v out we obtain (5.8). Alternately, we can apply the principle of superposition to derive this relation. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
517
v in 1 1 mV
R in 2
10 K
+
R 20 K in 3 30 K
v out
v in 2 4 mV v in 3 10 mV
Solution:
Let v out 1 be the output due to v in 1 acting alone, v out 2 be the output due to v in 2 acting alone, and v out 3 be the output due to v in 3 acting alone. Then by superposition,
v out = v out 1 + v out 2 + v out 3
(5.10)
First, with v in 1 acting alone and v in 2 and v in 3 shorted, the circuit becomes as shown in Figure 5.31.
R in 1 Rf 1 M
v in 1
10 K 1 mV R in 2 20 K R in 3
+
30 K
v out 1
and thus
(5.11)
Next, with v in 2 acting alone and v in 1 and v in 3 shorted, the circuit becomes as shown in Figure 5.32.
518
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Rf
1 M
R in 3 20 K
+
30 K
v out 2
and the voltage at the plus (+) input is computed from the voltage divider circuit shown in Figure 5.33.
R in 2 20 K
+
R in 3 30 K
To v ( + )
v in 2 4 mV
Figure 5.33. Voltage divider circuit for the computation of v ( + ) with v in 2 acting alone
Then,
R in 3 30 K v ( + ) = --------------------------- v in 2 = ---------------- 4 mV = 2.4 mV R in 2 + R in 3 50 K
and thus
v out 2 = 101 2.4 mV = 242.4 mV
(5.12)
Finally, with v in 3 acting alone and v in 1 and v in 2 shorted, the circuit becomes as shown in Figure 5.34. The circuit of Figure 5.34 is also a noninverting op amp whose voltage gain G v is
G v = 1 + 1 M 10 K = 101
and the voltage at the plus (+) input is computed from the voltage divider circuit shown in Figure 5.35.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
519
R in 2
20 K
+
R in 3
30 K
v out 3
+
30 K
To v ( + )
v in 3
10 mV
Figure 5.35. Voltage divider circuit for the computation of v ( + ) with v in 3 acting alone
Example 5.11 For the circuit shown in Figure 5.36, derive an expression for the voltage gain G v in terms of the external resistors R 1 , R 2 , R 3 , and R f .
R1 Rf
v in
R2 R3
+ v out
520
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v1 v2
R3
v in
Rf
R2
+ v out
At node v 1 :
v 1 v in v 1 v out ----------------- + -------------------- = 0 Rf R1 v out 1 1 ----- + ---- v = v in + -------------1 R Rf Rf R1 1 R1 + Rf ----------------- v = R f v in + R 1 v out ---------------------------------- R1 R 1 R1 Rf f
or
R f v in + R 1 v out v 1 = -----------------------------------R1 + Rf
(5.13)
At node v 2 : or
(5.14)
or
in R f v in + R 1 v out = ------------------ ( R 1 + R f )
R3 v
R2 + R3
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
521
(5.16)
(5.17)
Therefore, in an op amp circuit the input resistance provides a measure of the current i s which the amplifier draws from the voltage source v s . Of course, we want i s to be as small as possible; accordingly, we must make the input resistance R in as high as possible. Example 5.12 Compute the input resistance R in of the inverting op amp amplifier shown in Figure 5.38 in terms of R 1 and R f .
R1 Rf
vs
is
v out
Solution: By definition, R in = v s i s and since no current flows into the minus () terminal of the op amp and this terminal is at virtual ground, it follows that
is = vs R1
522
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(5.18)
It is therefore, desirable to make R 1 as high as possible. However, if we make R 1 very high such as
10 M , for a large gain, say 100 , the value of the feedback resistor R f should be 1 G . Obvi-
ously, this is an impractical value. Fortunately, a large gain can be achieved with the circuit of Exercise 8 at the end of this chapter. Example 5.13 Compute the input resistance R in of the op amp circuit shown in Figure 5.39.
Rf 100 K
v in
v out
Solution:
In the circuit of Figure 5.39, v in is the voltage at the minus () terminal; not the source voltage
v S . Therefore, there is no current i S drawn by the op amp. In this case, we apply a test (hypothet-
ical) current i X as shown in Figure 5.40, and we treat v in as the source voltage.
v in
Rf 100 K
iX
+ v out
Figure 5.40. Circuit for Example 5.13 with a test current source
By definition, the output resistance R out is the ratio of the open circuit voltage to the short circuit current, that is, Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
523
(5.19)
The output resistance R out is not the same as the load resistance. The output resistance provides a measure of the change in output voltage when a load which is connected at the output terminals draws current from the circuit. It is desirable to have an op amp with very low output resistance as illustrated by the following example. Example 5.14 The output voltage of an op amp decreases by 10% when a 5 K load is connected at the output terminals. Compute the output resistance R out . Solution: Consider the output portion of the op amp shown in Figure 5.41.
R out
v out
(5.20)
With a load R load connected at the output terminals, the load voltage v load is
R load v load = ------------------------------ v out R out + R load
(5.21)
and from (5.20) and (5.21) Therefore, and solving for R out we obtain
R load v load = ------------------------------ G v v in R out + R load v load 5 K ----------- = 0.9 = ------------------------------v OC R out + 5 K R out = 555
(5.22)
From (5.22) we observe that as R out 0 , v load = G v v in and with (5.20), v load = v OC .
524
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
10 10
10
10
1 1
f ( Hz )
10 10
2
10
10
10
10
Figure 5.42. Typical op amp openloop frequency response curve (log scales)
The frequencyresponse curve of Figure 5.42 shows that the bandwidth is only 10 Hz with this configuration. The unity gain frequency is the frequency at which the gain is unity. In Figure 5.46 the unity gain frequency is 1 MHz . We observe that the frequency response curve shows that the Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
525
(5.23)
Denoting the openloop gain as A ol , the 3 dB bandwidth as BW 3 dB , and the unity gain frequency as f ug , we can express (5.23) as
A ol BW 3 dB = f ug
(5.24)
The use of negative feedback increases the bandwidth of an operational amplifier circuit but decreases the gain so that the gain times bandwidth product is always equal to the unity gain frequency of the op amp. The frequencyresponse curve shown in Figure 5.43 is for a circuit in which negative feedback has been used to decrease the circuit gain to 100 (from 100,000 for the operational amplifier without feedback). We observe that the halfpower point of this curve is slightly above 10 KHz .
Gain 3 dB point ( open loop )
10
5
10 10
10
10
10
100
10
10
10
10
f ( Hz )
526
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v1 v in = v 2 v 1 v2
R in
R out
Ao l ( v2 v1 )
v out = A ol v in
(5.25)
We observe that when v 2 > v 1 , the output voltage v out is positive and when v 1 > v 2 , v out is negative. Accordingly, we call the lower terminal v 2 the noninverting input and the upper terminal
v 1 the inverting input. However, if v 2 = v 1 , v out = 0 and we call this condition commonmode
rejection. In other words, the op amp rejects any signals at its inputs that are exactly the same. Example 5.15 Figure 5.45 shows a circuit that can be used as a highpass filter. We want to find the range of the openloop gain A o l for which the system is stable. Assume that the input impedance is infinite and the output impedance is zero.
R
v in
+
C1 C2 R1 Rf
v out
Solution:
For stability, the coefficients on the denominator of the transfer function must all be positive. To Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
527
v in
1 s C1
1 s C2
v out
(5.26)
At Node V 2 ,
2 s C 2 ( V 2 V 1 ) + -----
R1
(5.27)
Also, or
V out = A o l ( V 2 V 3 ) = A o l ( V 2 0 ) = A o l V 2 V 2 = V out A o l
(5.31) (5.32)
Solving (5.31) and (5.32) for V 1 , equating right sides, and rearranging, we obtain the transfer function
A o l [ R 1 R f C 1 C 2 ]s G ( s ) = ---------- = ----------------------------------------------------------------------------------------------------------------------------------2 V in [ R 1 R f C 1 C 2 ]s + [ R 1 C 2 ( 1 A ol ) + R f C 1 + R f C 2 ]s + 1
V out
2
(5.33)
For stability, the coefficient of s in the denominator of (5.33) must be positive, that is,
528
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Transresistance Amplifier
R1 C2 ( 1 Ao l ) + Rf C1 + Rf C2 > 0
(5.34) (5.35)
or
or
v in
v out
The circuit of Figure 5.47 is the same as that of Figure 5.39, the circuit for Example 5.13, where we found that R in = 0 . Figure 5.48 shows the circuit model of the transresistance amplifier which was introduced in Exercise 4 in Chapter 1. As in Example 5.13, a test current i x at the inverting input produces an output voltage v out whose value is v out = v in R f i x , and since v in = 0 , the transresistance R m of the circuit of Figure 5.47 is
v out v out R m = --------- = --------- = R f i i
in x
(5.36)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
529
i out v out
R m = ---------
v out i in
i out = 0
v in
R in R m i in
and the minus () sign indicates that them output voltage v out and the test current i x are 180 outofphase with each other.
(5.37)
V out ( s )
Example 5.16 Derive the closedloop transfer function for the circuit of Figure 5.50. Solution: To derive the transfer function, we first convert the given circuit to its s domain equivalent, and for convenience we denote the series devices as Z 1 ( s ) and the parallel devices as Y f ( s ) . The circuit then is as shown in Figure 5.51.
530
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
+
v in
R1
C1
+
v out
Rf
+
V in ( s )
R1
1 s C1 1 s Cf
V out ( s )
(5.41)
iC dt
(5.42)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
531
iC = i
v out
and assuming the initial condition that at t = 0 , the voltage across the capacitor is V 0 , we can express (5.42) as
1 v C = --C
0 i C d t + V 0
(5.43)
Since the inverting input is at virtual ground, the output voltage v out is the negative of the capacitor voltage v C , that is, v out = v C , and thus
1 v out = --C
0 i C d t V 0
R1
0 vin dt V0
Example 5.17 The input voltage to the amplifier in Figure 5.53(a) is as shown in Figure 5.53(b). Find and sketch the output voltage assuming that the initial condition is zero, that is, V 0 = 0 .
C 1 F 2 v in ( V )
R1 v in 1 M
t (s)
532
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
0 vin dt V0 0 2 dt = 2 t 0 = 6
3 3
and with R 1 C = 10 10
0 vin dt
This result shows that the output voltage v out decreases linearly from zero to 6 V in the time interval 0 t 3 s and remains constant at 6 V for t > 3 s thereafter as shown in Figure 5.54.
v out ( V )
3 t (s)
6 Figure 5.54. Output waveform for the integrator circuit in Figure 5.53
The output voltage waveform in Figure 5.54 indicates that after the capacitor charges to 6 V , it behaves like an open circuit and effectively the negative feedback is an open circuit. Now, let us suppose that the input to the op amp integrator circuit of Figure 5.57(a) is the unit step function u 0 ( t ) * as shown in Figure 5.55(a). Then, ideally the output would be a negative ramp towards minus infinity as shown in Figure 5.55(b).
v in ( V ) u0 ( t ) v out ( V )
t (s)
(a)
t (s)
(b)
Figure 5.55. The output voltage of the circuit in Figure 5.53 when the input is the unit step function
* For a detailed discussion on the unit step function refer to Signals and Systems with MATLAB Computing and Simulink Modeling, Fourth Edition, ISBN 97819734404119.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
533
5 M
0.2 F
v in ( V )
v in
1 M
v out
(a) (b)
t (s)
Solution: This is the same circuit and input voltage waveform as in Example 5.17 except that a 5 M feedback resistor has been added and the capacitor value was changed to 0.2 F to simplify the computations. Since it is stated that the initial condition is zero, the capacitor charges in accordance with the relation
vC = V ( 1 e
( 1 R f C )t
(5.47)
where
and at t = 3 s
v out
t=3 s
= 10 ( 1 e
3 1
534
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
= v out
t=3 s
( 1 RC ) t
= 10e
10 1
= 10 4.54 10
= 45.4 V
9.5
As we can see from Figure 5.57, the addition of the feedback resistor makes the circuit of Figure 5.56 somewhat less than an ideal integrator.
iC
Rf
+
v in
iC
v out
We observe that the right side of the capacitor is virtually grounded and therefore the current through the capacitor is
dv in dv C i C = C -------- = C --------dt dt
Also,
(5.48)
and we observe that the output voltage v out is the derivative of the input voltage v in . Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
535
+
v in
vC ( t )
+
v out
= R f C = 10 C 10
3 9
R f = --- = --------- = 1 M 10
b. Differentiation in the time domain corresponds to multiplication by s in the complex frequency domain, minus the initial value of f ( t ) at t = 0 .* Thus,
dv C -------- sV C ( s ) v C ( 0 ) dt
* For all Laplace transform properties, refer to Circuit Analysis II with MATLAB Applications, ISBN 97809709511 51.
536
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
or
V out ( s ) ------------------ = sR f C V in ( s )
c. With s = j , the transfer function can be expressed in magnitude and phase form as
V out ( j ) --------------------- = jR f C V in ( j ) V out ---------- = R f C V in = 90 V out ---------V in = 2 10 10 10
f = 1 KHz 3 6 9
= 2 10 10
= 2
and the phase angle is 90 at all frequencies. d. As f , the capacitor behaves as a short circuit and so with the addition of a resistor R 1 in series with the capacitor, the closed loop voltage gain G v is
Gv = Rf R1
R1
i1
R2
i i
v in 2
i2
RN
Rf
v in N
iN
v out
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
537
v in 1 i 1 = --------R1
v in 2 i 2 = --------R2
v in N i N = ---------RN
v out = R f i = R f ( i 1 + i 2 + + i N )
f f f v out = R f i = R f ( i 1 + i 2 + + i N ) = ----- v in 1 + ----- v in 2 + + ------ v in N R1 R2 RN
(5.49)
(5.50)
(5.51)
and this indicates that the circuit of Figure 5.60 can be used to find the negative sum of any number of input voltages. The circuit of Figure 5.60 can also be used to find the average value of all input voltages. The ratio R f R is selected such that the sum of the input voltages is divided by the number of input voltages applied at the inverting input of the op amp. The circuit of Figure 5.61 shows the basic noninverting summing and noninverting averaging op amp circuit. In Figure 5.61 the voltage sources v in 1, v in 2, , v in N and their series resistances
R 1, R 2, , R N
and their parallel resistances R 1, R 2, , R N .* The circuit of Figure 5.61 can now be represented as in Figure 5.62.
For voltage source with series resistance to current source with parallel resistance transformation please refer to Circuit Analysis I with MATLAB Applications, ISBN 9780970951120
538
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v in 1
R1 R2 RN
+
v in 2
v out
v in N
+
R eq
v out
In the circuit of Figure 5.62, the voltage V 2 at the noninverting input is V 2 = R eq I eq and thus the output voltage is
R R v out = ----f + 1 V 2 = ----f + 1 ( R eq I eq ) R R
(5.52)
v in 1
R1 R2
v in 2
+
R3
v out
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
539
v in 1
+
R2 R3
v out 1
(5.53)
Next, with the input voltage v in 1 grounded and v in 2 acting alone, the circuit of Figure 5.63 reduces to that of Figure 5.65 and as indicated, we denote the voltage at the noninverting input as v 2 .
R1 Rf
v2
R3
+
v in 2
R2
v out 2
(5.54)
540
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(5.55)
or
(5.56)
To be useful, a differential input amplifier must have a high common mode rejection ratio (CMRR) defined as
Ad Differential gain CMRR = ----------------------------------------------------- = ---------- * Common mode gain A cm
(5.57)
where the differential gain A d is the gain with the input signals applied differentially, and common mode gain is the ratio of the output commonmode voltage V cm out to the input commonmode voltage V cm in , that is, A cm = V cm out V cm in . Ideally, A cm is zero but in reality is finite and much smaller than unity. It is highly desirable that the differential input op amp of Figure 5.63 produces an output voltage v out = 0 when v in 2 = v in 1 , and as we now know, this is referred to as commonmode rejection. We also want a nonzero output when v in 2 v in 1 . If the commonmode rejection condition is achieved, that is, when v in 2 = v in 1 and v out = 0 , relation (5.57) above reduces to
Rf Rf R1 + 1 ----- = ------------------------R1 R2 R3 + 1 Rf ----- + 1 R = R 2 + 1 R ---- R1 1 R f
3
R f + R 1 = R f + ----- R f R
3 2 R 1 = ----- R f R 3
R2
Rf R3 ----- = ----R1 R2
(5.58)
* The common mode rejection is normally expressed in dB, that is, CMR ( dB ) = 20 log ( A v Acm ) = 20 log CMRR .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
541
R1
(5.59)
Next, we will derive the input resistance for the differential input op amp circuit of Figure 5.63. For convenience in (5.58) we let R 2 = R 1 . Then R 3 = R f and with these simplifications the circuit of Figure 5.63 is as shown in Figure 5.66.
Rf
v in 2 v in 1 i
R1
+
R1 Rf
v out
Figure 5.66. Differential input op amp for derivation of the input resistance
Application of KVL around the input circuit starting at the minus () input terminal and going counterclockwise, and observing that there is a virtual short between the inverting and non inverting inputs, we obtain
R 1 i + R 1 i ( v in 2 v in 1 ) = 0 v in 2 v in 1 = 2R 1 i
v in 2 v in 1 R in = ------------------------i R in = 2R 1
Relation (5.59) reveals that for a large differential gain, we must make the feedback resistor R f as large as possible and the resistance R 1 as small as possible. But with small R 1 the input impedance will also become small as we can see from (5.62).
542
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Instrumentation Amplifiers +
v1
A1
vo 1
R2
R3
R4
v1 v2
R1
+
R2 R3
A3
v out
v2
A2
R4
vo 2
Figure 5.67. High input resistance differential input op amp for use with instrumentation circuits
In the circuit of Figure 5.67 we have made the assumption that there is no current flowing at the inputs of amplifiers A 1 and A 2 and thus the same current flows through the resistive R 2 R 1 R 2 network. Thus,
vo 1 v1 v1 v2 v2 v1 I = ------------------ = ---------------- = ---------------R1 R2 R2
(5.63)
We also have made the assumption that there is no voltage difference between the amplifiers A 1 and A 2 input terminals. The output voltage v o 1 of amplifier A 1 is
v o 1 = R 2 I + R 1 I + R 2 I + v o 2 = ( R 1 + 2R 2 )I + v o 2
and with the second term on the right side of (5.63) the above relation can be expressed as
v1 v2 2R 2 v o 1 = ( R 1 + 2R 2 ) ---------------- + v o 2 = 1 + -------- ( v 1 v 2 ) + v o 2 R1 R
1
2R 2 v o 1 v o 2 = 1 + -------- ( v 1 v 2 ) R
1
(5.64) (5.65)
2R 2 v o 2 v o 1 = 1 + -------- ( v 2 v 1 ) R
1
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
543
R3
(5.66)
(5.67)
To make the overall gain of the circuit of Figure 5.67 variable while maintaining CMRR capability, op amp manufacturers recommend that the resistor R 1 be replaced with a fixed value resistor in series with a variable resistor. The fixed resistor will ensure that the maximum gain is limited, while the variable resistor (a potentiometer) can be adjusted for different gains. The interested reader may refer to Exercise 16 at the end of this chapter.
According to the 741 op amp specifications, the maximum input offset voltage is 6 mv and assuming that the closed loop gain is 100, the output voltage with respect to ground can be
100 6 10
3
= 0.6 V and this value can be either positive or negative even though the input
544
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Figure 5.69. Typical op amp with externally connected capacitors for frequency compensation
100 90 80 Open-loop gain (dB) 70 60 50 40 30 20 10 0 3 10 10
4
No compensation
With compensation
10 10 Frequency (Hz)
10
10
Figure 5.70. Frequency responses with and without external frequency compensation
545
(5.68)
Of course, relation (5.68) is the slope of the output voltage under maximum rate of change conditions. Typical slew rates range from 0.1 V s to 100 V s , and most internally compensated op amps have slew rates in the order of 1 V s . Figure 5.71 shows a step function of amplitude 10 V applied to the input of a unity gain op amp, and the waveform at the output of this op amp.
v in ( V ) 10 v out ( V )
Slew rate = Slope
+v in
t
+
R
v out
10
Figure 5.71. The resultant slew rate when a step function is applied to a unity gain op amp
The linearly rising slew rate shown in Figure 5.71 will not be produced if the input voltage is smaller than that specified by the manufacturer. In this case, the slew rate will be a rising exponential such as the rising voltage across a capacitor. In most op amps the slew rate is set by the charging rate of the frequency compensating capacitor and the output voltage is
v out = V f ( 1 e
ug t
(5.69)
where V f is the final value of the output voltage as shown in Figure 5.71, ug = 2f ug , and f ug is the unity gain frequency as defined in (5.24), i.e., A ol BW 3 dB = f ug .
Vf v out
546
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
about 0.7 V . Likewise, when the output voltage is positive, its value is limited to the value ( VZ + VF ) .
2
D1
D2 VZ + VF
1
v out
R1
v in
Rf
v out
+
Slope = R f R 1
v in ( V Z2 + V F )
Figure 5.73. A positive and negative voltage limiter circuit and its transfer characteristics
Example 5.20 In the circuit of Figure 5.73, V Z = V Z = 6.3 V , V F = 0.7 V , R 1 = 5 K , and R f = 100 K .
1 2
Describe the output waveforms when a. v in = 0.3 sin 10 t b. v in = 0.6 cos 100 t c. v in = 3 cos ( 1000 t + 6 ) Solution: Since this is an inverting amplifier, its gain is R f R 1 = 100 5 = 20 * and since we are interested in peak values, the frequencies and phase angles are immaterial for this example. With the given values, the output peaks on positive halfcycles are limited to
V Z + V F = 6.3 + 0.7 = 7 V
1
* The gain is always expressed as a positive quantity. The minus sign simply implies inversion.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
547
This value is lower than 7 V and neither Zener diode conducts. Therefore, the output voltage is an unclipped sinusoid. b. With v in = 0.6 V peak,
v out ( peak ) = ( R f R 1 )v in = 20 ( 0.6 ) = 12 V +
and this would be the output peak voltage if the Zener diodes were not present. Since they are, the output peak voltage is clipped to v out ( peak ) = 7 V . + c. With v in = 3 V peak,
v out ( peak ) = ( R f R 1 )v in = 20 ( 3 ) = + 60 V
This is indeed a very large voltage for the output of an op amp and even without the Zener diodes, the op amp would saturate. But with the Zener diodes present, v out ( peak ) = 7 V . + Figure 5.74 shows a positive voltage limiter and its transfer characteristics where both the Zener diode and the junction diode limit the positive halfcycle of the output voltage. As shown by the transfer characteristics, v out cannot rise above the voltage level V Z + V F because the Zener
1
diode enters the Zener (avalanche) region and the output is clipped. However, the negative half cycles are not clipped unless the op amp is driven into negative saturation.
D1 D2 v out VZ + VF
1
v in
R1
Rf
v out
Slope = R f R 1
v in
Figure 5.74. A positive voltage limiter circuit and its transfer characteristics
Figure 5.75 shows a negative voltage limiter and its transfer characteristics.
548
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
+
v in
R1
Rf
v out
v in
Slope = R f R 1
( VZ2 + VF )
Figure 5.75. A negative voltage limiter circuit and its transfer characteristics
Figure 5.76 shows a limiter where only a single Zener diode is used. This circuit is often referred to as a halfwave rectifier with limited positive output.
D v out
R1 Rf
+
v in
VZ v out
Slope = R f R 1
v in VF
Figure 5.76. A halfwave rectifier with limited positive output and its transfer characteristics
In the circuit of Figure 5.76, the output is limited by the Zener diode during the positive half cycles of the output voltage, and during the negative halfcycles of the output voltage is limited by the Zener diode forward voltage drop V F . Figure 5.77 shows another limiter where only a single Zener diode is used.
D v out VF v out
+
v in
R1
Rf
VZ
Slope = R f R 1
v in
Figure 5.77. A halfwave rectifier with limited negative output and its transfer characteristics
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
549
5.21 Comparators
A comparator is a circuit that senses changes in a varying signal and produces an output when a threshold value is reached. As a comparator, an op amp is used without feedback, that is, the op amp is used in the open loop configuration. Figure 5.78 shows a differential input amplifier without feedback used as a comparator.
+V CC v out +V CC 6 v ref > v in
+
v in
7 4 V CC
+
v ref
v out
v ref < v in
v ref
V CC
As shown in Figure 5.78, v out = +V CC if v ref > v in , and v out = V CC if v ref < v in . The switching time from V CC to +V CC is limited by the slew rate of the op amp. Comparators are used extensively in analogtodigital conversion as we will see in a subsequent section. Op amp applications are limitless. It is beyond the scope of this text to describe all. It will suffice to say that other applications include zerocrossing detectors also known as sinewave to square wave converters, sample and hold circuits, squarewave generators, triangularwave generators, sawtoothwave generators, TwinT oscillators, Wien bridge oscillators, variable frequency signal generators, Schmitt trigger, and multivibrators. We will discuss the Wien bridge oscillator*, the digitalto analog converter, and the analogtodigital converter in the next sections, and the Schmitt trigger and multivibrators in Chapter 7.
550
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
R1
C1
C2
R2
10 K
D1
D2 V out
50 K
Figure 5.79 shows that the Wien bridge oscillator uses two RC networks connected to the non inverting input of the op amp to form a frequency selective feedback circuit and this causes oscillations to occur. It also amplifies the signal with two negative feedback resistors. The input signal to the noninverting input is in phase with the output V of the op amp at the particular frequency
1 f 0 = -------------2RC
(5.70)
R = R1 = R2 C = C1 = C2
The Wien bridge oscillator requires precision resistors and capacitors for reliable operation. The feedback signal at the noninverting input of the op amp leads the output V of the op amp at frequencies below f 0 and lags V at frequencies above f 0 . The amount of negative feedback to the inverting input of the op amp and amplitude can be adjusted with the 50 K potentiometer. The diodes prevent excessive feedback amplitude. Example 5.21 For the oscillator circuit in Figure 5.79 R 1 = 100 K . What values of R 2 , C 1 , and C 2 are required to obtain a frequency of approximately 1 KHz ? Solution: The value of resistor R 2 must be 100 K . The values of capacitors C 1 , and C 2 must also be equal. From (5.70),
1 1 f 0 = 1 KHz = -------------- = -----------------------------5 2RC 2 10 C
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
551
100 K
10 K
D1
D2 V out
50 K
5V
5V
5V
5V
5V
In Figure 5.81 V D = 0 , V C = 1 , V B = 0 , and V A = 1 , that is, switches A and C are HIGH (5 volts) and switches B and D are LOW (0 volts). The first 16 binary numbers representing all possible combinations of the four switches with voltage settings V A (least significant position) through V D (most significant position), and their decimal equivalents are shown in Table 5.1.
* Refer also to Digital Circuit Analysis and Design with Simulink Modeling and Introduction to CPLDs and FPGAs, ISBN 9781934404058.
552
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
DigitaltoAnalog Converters
TABLE 5.1 Voltage levels for the circuit of Figure 5.81 and binary and decimal equivalents
Voltage Level
VD
LOW LOW LOW LOW LOW LOW LOW LOW HIGH HIGH HIGH HIGH HIGH HIGH HIGH HIGH
VC
LOW LOW LOW LOW HIGH HIGH HIGH HIGH LOW LOW LOW LOW HIGH HIGH HIGH HIGH
VB
LOW LOW HIGH HIGH LOW LOW HIGH HIGH LOW LOW HIGH HIGH LOW LOW HIGH HIGH
VA
LOW HIGH LOW HIGH LOW HIGH LOW HIGH LOW HIGH LOW HIGH LOW HIGH LOW HIGH
A 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1
Binary Equivalent B C D 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
Decimal Equivalent 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
A digitaltoanalog (D/A or DAC) converter is used to convert a binary output from a digital system to an equivalent analog voltage. If there are 16 combinations of the voltages V D through V A , the analog device should have 16 possible values. For example, since the binary number 1010 (decimal 10) is twice the value of the binary number 0101 (decimal 5), an analog equivalent voltage of 1010 must be double the analog voltage representing 0101. Figure 5.82 shows a DAC with binaryweighted resistors.
R----R --8 VD VC R --4 VB R --2 VA V analog R
2
VN
We can prove that the equivalent analog voltage V analog shown in Figure 5.82 is obtained from the relation
V A + 2V B + 4VC + 8V D + V analog = -----------------------------------------------------------------------------1+2+4+8+
(5.71)
The proof is left as an exercise at the end of this chapter. The DAC with binaryweighted resistors shown in Figure 5.82 has the disadvantage that it Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
553
We can prove that the equivalent analog voltage V analog shown in Figure 5.83 is obtained from the relation
V A + 2V B + 4VC + 8V D + V analog = -----------------------------------------------------------------------------n 2
(5.72)
where n is the number of digital inputs. The proof is left as an exercise at the end of this chapter. Figure 5.84 shows a fourbit R2R ladder network and an opamp connected to form a DAC. The op amp shown is an inverting amplifier and in this case the reference voltage V ref should be negative so that the amplifier output will be positive. Alternately, a noninverting op amp could be used with a positive value of V ref .
Rf 2R V analog
2R 2R
R 2R
R 2R
Example 5.22 Figure 5.85 shows a fourbit DAC where all four switches are set at the ground level. Find the analog voltage value at the output of the unity gain amplifier for each of the sets of the switch positions shown in Table 5.2. Fillin the rightmost column with your answers.
554
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
DigitaltoAnalog Converters
20 K
10 K 20 K D ( lsb )
10 K 20 K C
10 K 20 K B 20 K A ( msb ) 0 1 0
V analog
8 VDC
Figure 5.85. DAC circuit for Example 5.22 TABLE 5.2 Switch positions for Example 5.22 (a) (b) (c) (d) A2 1 1 1 0
0
B2 1 0 0 1
C2 1 0 1 0
D2 1 1 0 0
Solution: This is a 4bit DAC and thus we have 2 = 16 distinct binary values from 0000 to 1111 corresponding to decimals 0 through 15 respectively. From (5.72): a. b. c. d.
V A + 2V B + 4VC + 8V D V analog = ----------------------------------------------------------------- = 1 8 + 2 8 +4 4 8 + 8 8 = 7.5 V ------------------------------------------------------------------n 2 2 V A + 2V B + 4VC + 8V D V analog = ----------------------------------------------------------------- = 1 8 + 2 0 +4 4 0 + 8 8 = 4.5 V ------------------------------------------------------------------n 2 2 V A + 2V B + 4VC + 8V D V analog = ----------------------------------------------------------------- = 1 8 + 2 0 +4 4 8 + 8 0 = 2.5 V ------------------------------------------------------------------n 2 2 V A + 2V B + 4V C + 8V D V analog = ----------------------------------------------------------------- = 1 0 + 2 8 +4 4 0 + 8 0 = 1.0 V ------------------------------------------------------------------n 2 2
4
Based on these results, we can now fillin the rightmost column with the values we obtained, and Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
555
Figure 5.86. Output vs. inputs for an R2R network with levels 0 to 4 volts
A typical DAC must include an op amp to match the resistive network to a lowresistance load and to provide gain also. Placing an impedancematching device (the op amp in this case) at the output of the resistive network is called buffering the output of the network. Figure 5.87 shows an R2R type DAC with buffered output and gain.
R1 Rf V analog 2R VD
2R 2R VA
R 2R VB
2R VC
Figure 5.87. Op amp placed between the resistive network and output for buffering and gain
556
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
AnalogtoDigital Converters
converter. We will not discuss the latter; the interested reader may refer to http://www.allaboutcircuits.com/vol_4/chpt_13/9.html. We begin our discussion with the flash converter because of its simplicity.
A8 A7
Overflow
10.5 V
Comparator
9V 7.5 V
A6 A5 A4 A3 A2 A1 A0 B2
8 to 3 Line Encoder
Vx Vy
Ai
6V
Inputs
Output
B1 B0
4.5 V 3V 1.5 V
0 V
Analog Input
The truth table of the 8to3 line encoder is shown in Table 5.3. Obviously, a practical ADC would require a circuit with many more comparators and a suitable line encoder.
* For a detailed discussion of line encoders please refer to Digital Circuit Analysis and Design with Simulink Modeling and Introduction to CPLDs and FPGAs, ISBN 9781934404058.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
557
A2
The updown counter* has a digital output that increases with each clock pulse when the count up line is HIGH and the countdown line is LOW. Conversely, it counts down with each clock pulse when its countup line is LOW and its countdown line is HIGH. Op amp A 2 is a compar* For a detailed discussion on updown counters, refer to Digital Circuit Analysis and Design with Simulink Modeling and Introduction to CPLDs and FPGAs, ISBN 9781934404058.
558
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
AnalogtoDigital Converters
ator and its output goes HIGH when its inverting input becomes slightly more negative than its noninverting input. Conversely, its output goes LOW when its inverting input I becomes slightly more positive than its noninverting input. When its output is HIGH (positively saturated), the countup line is also HIGH. Likewise, when the output of A 2 is LOW (negatively saturated), the countup line is LOW also. Thus, depending on whether the output of A 2 is HIGH or LOW, the updown counter counts digitally up or down, respectively. When the updown counter is counting up, an upward staircase voltage appears at point X. When the counter is counting downward, a downward staircase is present at point X. Op amp A 1 serves as a voltage follower and it buffers the DAC resistive network.The output voltage of the resistive network is applied to the non inverting input of A 1 and subsequently to inverting input of A 2 and it is compared with the analog input voltage V analog which is the voltage to be digitized. If V analog exceeds the voltage of the resistive network, the output of A 2 goes HIGH, and the updown counter counts up, bringing the resistive networks output voltage up, in steps, to the analog input V analog . But if V analog is less than the resistive networks output voltage the output of A 2 goes LOW, and the updown counter counts down, bringing the voltages at the input of the comparator in line with each other. Because this system has feedback which keeps the voltage output of the resistive network approximately equal to the analog input voltage V analog , this ADC is also known as the feedbacktype ADC. In this way the output of the updown counter is always a digital equivalent of the analog input V analog . In a typical ADC of this type, the output of the counter is fed to a circuit to provide a digital readout. The line amplifiers are required just in case the maximum value of V analog exceeds the level of output voltages of the updown counter. The junction diodes prevent excessive differential inputs to the comparator, and the Zener diodes are selected to clip the comparators output to levels compatible with the updown counter.
559
v in
(b)
The output voltage v out is fed to a logic circuit which initiates counting in a binary counter. Thus, the circuitry of the this ADC translates a voltage level into a time measurement which can be counted with the binary counter. The singleslope converter of Figure 5.90(a) is not practical because it is subject to errors caused by variation of the timing components R and C, temperature, and aging. These errors will change the conversion results and make measurement repeatability difficult to attain. To overcome these sensitivities, the dualslope integrating architecture is used. Figure 5.91 shows a typical dualslope ADC.
S1 v max v in v ref R S2 C v int v out (a) Figure 5.91. An ADC with dualslope architecture
T1 T2
(b)
In the circuit of Figure 5.91(a) Switch S 1 is used to discharge the capacitor before the start of the conversion cycle and thus v int = 0 . The conversion cycle begins with opening Switch S 1 and placing Switch S 2 to the v in position. A constant current whose value is I N = v in R charges the capacitor for a reference time T 1 as shown in Figure 5.91(b). The output voltage v out is fed to a logic circuit which initiates counting in a binary counter. The binary counter counts the pulses from a fixed clock frequency f C and the counting is proportional to time T 1 . At the end of time
T 1 the binary counter is reset to zero and Switch S 2 is placed to the v ref position where v ref has
a negative value. The current into the integrator reverses direction and its value is
560
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
as shown in Figure 5.91(b) and when this voltage reaches zero volts, the comparator initiates a signal to stop the binary counter. Let n 1 be the counter reading at the end of T 1 and n 2 be the counter reading at the end of T 2 . Since v max T 1 = v in ( RC ) , v max T 2 = v ref ( RC ) , and n 1 T 1 = n 2 T 2 , it follows that
v in n 2 = n 1 -------- v ref
(5.73)
and this expression indicates that the content of the counter at the end of the conversion procedure is directly proportional to the voltage v in to be digitized. The major advantage of the dual slope architecture ADC over the singleslope architecture ADC is that the entire cycle of the conversion is insensitive to errors since any error introduced by the components R and C during time T 1 will be cancelled out during time T 2 .
(a)
1 2 3 4 5 6 7 8 9 10
t ( s )
(b)
From Figure 5.92(b) we see that at t = 1 s , v = 1.0 mv , at t = 2 s , v = 2.5 mv , and so on. But with the quantization we chose, we do not know intermediate values, for instance, we do not know what the amplitude of the voltage at t = 1.6 s . Of course, we could quantize the time axis in smaller intervals but no matter how small the intervals are, there will always be a small number that will fall between the time intervals. Subdividing the time axis into a finite number of intervals is known as sampling. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
561
562
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
10
= 1024
(5.74)
where a 1 , a 2 , b 1 , b 2 , c 1 , and c 2 are constant coefficients. Solution: We observe that the arithmetic operations involved in (5.74) are addition and multiplication by constant coefficients. The two additions can be performed by summing op amps. Multiplication by a coefficient greater than unity can be performed with an op amp with feedback, and if the coefficient is less than unity, we can use a voltage divider. It is convenient to express the given equations as
c1 b1 x = ---- ---- y a1 a1 c2 a2 y = ---- ---- x b2 b2
(5.75)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
563
obtained by amplifier A 2 .
V S1 R adj1 R 1 R2 R f1 A1
c1 a1
x
R adj3
( a 2 b 2 )x
R3 V S2 R adj2
R f2 A2
c2 b2
R4
y
R adj4
( b 1 a 1 )y
Figure 5.93. Analog computer for the solution of two simultaneous equations with two unknowns
In Figure 5.93 the scaling factors chosen must ensure that the voltages representing the unknowns do not exceed the output capability of the op amps. This procedure can be extended to the solution of simultaneous equations with more than two unknowns. Before we consider the next example for a circuit to solve a simple differential equation, we need to discuss a practical integrator circuit that provides some means of setting a desired initial value at the beginning of the integration cycle. It is also necessary to provide means to stop the integrator at any time, and for the integrator output to remain constant at the value it has reached at that time. An integrator circuit that provides these means is shown in Figure 5.94. Initially, the integrator sets the initial condition with the switches as shown in Figure 5.94(a), and denoting the initial condition as V 0 . The voltage across the capacitor cannot change instantaneously, and thus
564
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(5.76)
VS
VS
S1
S1
v in
v out
v in
R in S2 ( b ) Compute Mode
v out
VS
R1
R2
S1
v in
R in S2 ( c ) Hold Mode
v out
When switched to the compute mode, the circuit integrates the input voltage and the value of the output voltage is
1 v out = v out ( t = 0 ) ----------R C
in
0 vin dt
(5.77)
The integrator is then switched to the hold mode and remains constant at the value reached at the end of the compute mode. We observe that the switches in the hold mode are positioned as in the set initial condition mode. Example 5.26 Using one op amp, resistors, capacitors, and switches design an analog computer that will solve the simple differential equation
dv C 1 -------- = --- i C C dt
(5.78)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
565
iC 1 --- i C C
Initially, a voltage representing the value of dv C dt is applied to the integrator and the output of the integrator will be a voltage representing i C . Then, the output is multiplied by 1 C and fed back to the input.
566
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
1. Very high input impedance (resistance) 2. Very low output impedance (resistance) 3. Capable of producing a very large gain that can be set to any value by connection of external resistors of appropriate values 4. Frequency response from DC to frequencies in the MHz range 5. Very good stability 6. Operation to be performed, i.e., addition, integration etc. is done externally with proper selection of passive devices such as resistors, capacitors, diodes, and so on.
The gain of an inverting op amp is the ratio R f R in where R f is the feedback resistor which allows portion of the output to be fed back to the minus () input. The minus () sign implies
that the output signal has opposite polarity from that of the input signal.
The gain of an noninverting op amp is 1 + R f R in where R f is the feedback resistor which allows portion of the output to be fed back to the minus () input which is grounded through the R in resistor. The output signal has the same polarity from that of the input signal. In a unity gain op amp the output is the same as the input. A unity gain op amp is used to pro-
vide a very high resistance between a voltage source and the load connected to it.
Op amps are also used as active filters. The input resistance R in is defined as the ratio of the applied voltage v S to the current i S drawn by the circuit, that is, R in = v S i S . The output resistance R out is the ratio of the open circuit voltage to the short circuit current, that is, R out = v OC i SC . Operational amplifiers can operate either a closedloop or an openloop configuration. The operation, closedloop or openloop, is determined by whether or not feedback is used. Most operational amplifiers operate in the closedloop configurations, that is, with feedback. The gain of any amplifier varies with frequency. The specification sheets for operational amplifiers state the openloop at DC or 0 Hz . At higher frequencies, the gain is much lower and decreases as frequency increases at the rate of 20 dB decade . The unity gain frequency, is the frequency at which the gain is unity. The gainbandwidth product is an important op amp parameter and is related to the unity gain
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
567
decreases the gain so that the gain times bandwidth product is always equal to the unity gain frequency of the op amp.
Commonmode rejection refers to the condition that if the inputs to the op amp are exactly the same, i.e., if v in2 = v in1 , then v out = 0 . In other words, the op amp rejects any signals at
A differential input op amp allows input signals to be applied simultaneously to both input ter-
minals and produce an output of the difference between the input signals. A differential input amplifier must have a high common mode rejection ratio (CMRR) defined as
Ad Differential gain CMRR = ----------------------------------------------------- = --------Common mode gain A cm
where the differential gain A v is the gain with the input signals applied differentially, and common mode gain is the ratio of the output commonmode voltage V cm out to the input commonmode voltage V cm in , that is, A cm = V cm out V cm in . Ideally, A cm is zero but in reality is finite and much smaller than unity.
High input resistance differential input amplifiers are suitable for use in differential measure-
ment applications and the associated circuits are referred to as instrumentation amplifiers.
General purpose op amps are normally internally frequency compensated so that they will be
stable with all values of resistive feedback. Other types of op amps are without internal frequency compensation and require external connection of frequency compensating components to the op amp. Typically, the compensating components alter the open loop gain characteristics so that the rolloff is about 20 dB decade over a wide range of frequencies.
The slew rate (SR) is defined as the maximum rate of change of an output voltage produced in
568
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Summary
response to a large input step function and it is normally expressed in volts per microsecond, that is,
dv out SlewRate = SR = -----------dt max Op amps can also be used in circuits with nonlinear devices. We have introduced the positive and negative voltage limiters, halfwave rectifier with limited positive or negative output. A comparator is a circuit that senses changes in a varying signal and produces an output when a
threshold value is reached. As a comparator, an op amp is used without feedback, that is, the op amp is used in the open loop configuration.
A Wien bridge oscillator produces a sinusoidal output. This oscillator uses two RC networks connected to the noninverting input of the op amp to form a frequency selective feedback cir-
cuit and this causes oscillations to occur. It also amplifies the signal with two negative feedback resistors.
A digitaltoanalog converter (D/A converter or DAC) is used to convert a binary output from a digital system to an equivalent analog voltage. If there are 16 combinations of the voltages V D
precision resistors.
A DAC formed with an R2R ladder network, requires more resistors, but only two sets of pre-
A DAC must include an op amp to match the resistive network to a lowresistance load and to provide gain also. Placing an impedancematching device (the op amp in this case) at the out-
put of the resistive network is called buffering the output of the network.
An analogtodigital converter (A/D converter or ADC) converts an analog signal to its binary
equivalent. Popular types of ADCs are the flash converter, the successive approximation, and the dualslope.
Quantization is the process where an analog signal is subdivided into small discrete time and
amplitude intervals. Subdividing the time axis into a finite number of intervals is known as sampling. The difference between an exact analog value and the closest discrete value after sampling is known as quantization error. alent, the analog signal, before conversion to a digital, must be sampled at least twice the frequency of the highest frequency component of the signal. This minimum frequency is known as the Nyquist rate. The Nyquist rate is not sufficient; because of electronic equipment limitations and component imperfections, this theoretical value cannot be achieved. A general rule is to sample at five to ten times the highest frequency of the analog signal.
The sampling theorem states that to faithfully reproduce an analog signal from it digital equiv-
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
569
Resolution is the smallest change in the parameter being measured that causes a detectable
change in the output of the instrument. Typically, an eight, ten, or at most twelve bits of digital output provides adequate resolution. Present day analog computers are build with op amps. In an analog computer the numbers representing the variables are voltages.
570
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v in 1
3 K
+
27 K
10 K
+ v ou t1
90 K
10 mV
v out 2
4 K
+ +
3 K 6 K 5 K
i 5K
60 mV
3. For the circuit below, R in 1 , R in 2 , and R in 3 represent the internal resistances of the input voltages v in 1 , v in 2 , and v in 3 respectively. Derive an expression for v out in terms of the input voltage sources, their internal resistances, and the feedback resistance R f .
R in 1 R in 2 R in 3 Rf
v in 1
+ v out
+ v in 2
v in 3
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
571
10 K
+ +
40 mV 20 K 40 K
v out
5. The opamp circuit (a) below can be represented by its equivalent circuit (b). For the circuit (c), compute the value of R load so that it will receive maximum power.
R1 Rf
v in
v in v out
+
R1
+
(a)
(b)
----- v in v out R1
Rf
2 K
20 K
v in
5 K
+
(c)
+
15 K R load
v out
+
v 5K
20 K
v out
572
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
7. For the circuit below compute the gain G v = v out v in .
R2 40 K
+ v in
R1 200 K R5
40 K R4 50 K 50 K
R3
+ v out
R2
+ v in
R1
R4
+ + v out
9. The specifications for the popular 741 op amp state that the openloop gain is 200, 000 and the unity gain frequency is 1 MHz . Plot the frequency response for openloop gain versus frequency curve and the closedloop frequency response for a gain of 10 . 10. For the op amp shown below the openloop gain is 100, 000 .
v in 1 v in 2
v out
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
573
v in 2
+
+
Rf 100 K 10 F
12. For the op amp circuit below, find: a. the transresistance R m b. the output voltage v out
i in
10 K
+ v in 1 mA
+ v out
+ v in
Cf
+ v out
a. Derive the closedloop transfer function b. Derive an expression for the DC gain c. Derive an expression for the 3 dB frequency d. If R 1 = 1 K , compute the values of R f and C f such that the circuit will have a DC gain of 40 dB and 1 KHz 3 dB frequency e. Compute the frequency at which the gain is 0 dB f. The phase angle at the frequency where the gain is 0 dB
574
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
14. For the op amp integrator circuit below the time constant is = 100 s .
Cf
R1
v in
10 K
v out
a. Find the value of the capacitor C f b. Find the magnitude and phase angle of the gain at f = 1 KHz c. Find the frequency at which the magnitude of the gain is unity 15. For the circuit below, it is required that the current i f should not exceed 1 mA , and the output voltage should be such that v out = ( 4v in 1 + 3v in 2 ) 12 V . Select appropriate values for resistors R f , R 1 , and R 2 to meet these specifications.
v in 1
R1 R2
if
Rf
v in 2
v out
16. As stated earlier in Section 5.16, to make the overall gain of the circuit of Figure 5.67 variable, we replace resistor R 1 with a fixed value resistor R fix in series with a variable resistor
R var as shown below.
R1
R fix
R var
Choose appropriate values for the fixed value resistor R fix and the variable resistor R var so that the overall voltage gain can vary from 10 to 100 . 17. Suggest a nulling circuit other than that in Exercise 16 for: a. an inverting mode op amp b. a noninverting mode op amp
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
575
v in 1
v in 2
+
R2 R3
v out
19. For the circuit below, derive the transfer function G v ( s ) = V out ( s ) V in ( s ) and the magnitude and phase of G v ( j ) = V out ( j ) V in ( j ) at unity gain.
R1 Rf
v in
+
C
R2
v out
20. The circuit below is referred to as a differential input integrator. The resistors and capacitors are of equal value. Derive an expression for V out ( s ) in terms of the input voltages and the circuit constants.
R V in 1 ( s )
V in 2 ( s )
R C
C V out ( s )
21. The figure below shows a digitaltoanalog converter with binaryweighted resistors.
R ----R --8 VD VC R --4 VB R --2 VA V analog R
2
VN
Prove that
576
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
22. The figure below shows a digitaltoanalog converter with known as R2R ladder network.
2R 2R VA VB R 2R VC R 2R VD R 2R V analog
Prove that
VA + 2V B + 4V C + 8V D + Vanalog = -----------------------------------------------------------------------------n 2
where n is the number of digital inputs. 23. Design an analog computer circuit to perform the following integration.
v out =
Assume ideal op amp action and that the integrating capacitor has a value of 1 F . 24. Design an op amp circuit with input v in and output v out = 5v in 3dv in dt 25. For the op amp circuit shown below, derive an expression for the output voltage v out in terms of the input voltage v in and the circuit constants.
C R2 L
v in
R1
v out
26. For the op amp circuit shown below, derive an expression for the output voltage v out in terms of the input voltage v in .
1 M 0.5 F
1 M
v in
v out
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
577
v in 1
3 K
+
27 K
90 K
10 mV
v in 2
v out 2
v out1 = ( 27 3 ) 10 = 90 mV
and thus
v in2 = v out1 = 90 mV
Then
----v out2 = 1 + 90 ( 90 ) = 0.9 V 10
4 K
+ +
3 K
6 K
v1
v load
R load 5 K
i 5K
60 mV
3 K || 6 K = 2 K
Then
3 v load 6 20 10 20 mV i 5K = ------------ = ---------------- = ---------------------- = 4 10 A = 4 A 3 5 K R load 5 10
578
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Rf
v in 1
+ v out
+ v in 2
v in 3
By superposition
v out = v out 1 + v out 2 + v out 3
where
v out 1
v in2 = 0 v in3 = 0
Rf = --------- v in 1 R in1
We observe that the minus () input is a virtual ground and thus there is no current flow in R in1 and R in2 . Also,
v out 2
v in1 = 0 v in3 = 0
Rf = ---------- v in 2 R in 2
and
v out 3
v in1 = 0 v in2 = 0
Rf = ---------- ( v in 3 ) R in 3
Then,
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
579
10 K
v1 v2
40 mV
20 K 40 K
v out
Multiplication by 50 10 yields
6 3 2 80 10 50 10 ---------------------------------------------------------- v out = 4 10 50 10 3 50 10 3 3
v out = 40 mV
Vout = -0.0400 5. We attach the 5 K , 15 K , and R load resistors to the equivalent circuit as shown below. By Thevenins theorem,
15 K v TH = v OC = v ab = ------------------------------------- ( 10v in ) 5 K + 15 K
580
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v in
2 K
+ 10v in
a b
R load
or
v TH = 7.5v in
Because the circuit contains a dependent source, we must compute the Thevenin resistance using the relation R TH = v TH i SC where i SC is found from the circuit below.
5 K
10v in
a b i SC
15 K
We observe that the short circuit shorts out the 15 K and thus
10v in 3 i SC = ---------------- = 2 10 v in 5 K
Then
7.5v in R TH = ------------------------------ = 3.75 K 3 2 10 v in
v TH
3.75 K R load
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
581
+ +
Rf 1 + ------- v R in
in
v out
Attaching the external resistors to the equivalent circuit above we obtain the circuit below.
12 K 84 K 4 K 72 mV
a
5 K
v 5K 6v 5K
To find the Thevenin equivalent at points a and b we disconnect the 5 K resistor. When this is done there is no current in the 4 K resistor and the circuit simplifies to the one shown below.
12 K 72 mV
a+
84 K
v ab b
6v 5K
By KVL or Also
( 12 K + 84 K )i + 6v 5K = 72 mV 72 mV 6v 5K i = --------------------------------------------( 12 K + 84 K )
72 mV 6v 5K v TH = v ab = v 5K = 72 mV ( 12 K )i = 72 mV 12 K -------------------------------------- 96 K 3 v TH = 72 mV 9 mV + -- v 5K 4
or
582
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
and thus
v TH = v ab = v 5K = 252 mV
The Thevenin resistance is found from R TH = v OC i SC where i SC is computed with the terminals a and b shorted making v 5K = 0 and the circuit is as shown in Figure (a) below. We also perform voltagesource to currentsource transformation and we obtain the circuit shown in Figure (b) below.
12 K 4 K 12 K 4 K
+
72 mV
a i SC (a)
84 K
6 A
84 K
a i SC b
(b) 12 K || 84 K = 10.5 K
Now,
Therefore,
and the Thevenin equivalent circuit with the 5 K resistor is shown below.
R TH 12 K
a
+ v 5K
v TH 252 mV
5 K
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
583
+ v in
R1
v1
40 K R4 50 K 50 K
R3
v2
+ + v out
200 K R5
Node 1 at v 1 : or
v 1 v in v 1 v out v 1 0 v1 ------------------- + -------------------- + ---------------- + ---------------- = 0 200 K 40 K 50 K 50 K v out v in 1 1 1 1 ------------------- + ---------------- + ---------------- + ---------------- v = ------------------- + ---------------- 1 200 K 40 K 50 K 50 K 200 K 40 K
or
584
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
+ v in
R1
v1
+ + v out
Node 1 at v 1 :
0 v in 0 v 2 --------------- + ------------- = 0 R2 R1 R2 v 2 = ----- v in (1) R1 v 2 0 v 2 v 2 v out ------------- + ----- + -------------------- = 0 R3 R2 R4 1 1 1 ----- + ----- + ----- v = v out --------R R3 R4 2 R4 2 1 v 2 = ------------------------------------------------ v out (2) R4 R2 + R4 R3 + 1
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
585
10
10
10 1
f ( Hz )
1 10 100 10
3
10
10
10
10.
v in 1 v in 2
v out
From (5.25), Page 527, v out = A o l ( v in 2 v in 1 ) , and we are given that A o l = 100, 000 . Then, a.
v out 3 5 v in 1 = v in 2 --------- = 3 10 ------- = 2.95 mV 5 Ao l 10 v out 3 5 v in 2 = v in 1 + --------- = 2 10 + ------- = 1.95 mV 5 Aol 10 v out = A o l ( v in 2 v in 1 ) = 10 ( 30 20 ) 10
5 6
b.
c.
= 1 V
586
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v in 2
+
+
Rf 100 K 5 10 s
The pole of G ( s ) is located at s = ( 1 A ol ) = A ol 1 and it is positive for A o l > 1 . Therefore, the circuit is unstable since the pole is located on the right halfplane. 12.
i in
10 K
+ v in 1 mA
+ v out
This circuit is the same as that of Figure 5.47, Page 5-29, where we found that R m = R f . Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
587
Z1 ( s )
+ V in ( s )
R1
1 s Cf
+
V out ( s )
a.
b.
The DC gain is defined at the point where the frequency is zero, that is, s = 0 . Then, (1) above reduces to
Rf V out Gain DC = ---------- = ----R1 V in
c.
and we recognize this as the transfer function of a lowpass filter whose cutoff 3 dB fre-
588
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
1 f 3 dB = -----------------2R f C f
log scale
20 dB decade
Pass band
Stop band
d. 40 dB represents a ratio of 100 . Therefore, R f R 1 = 100 and with R 1 = 1 K , we find that R f = 100 K . Also, since
1 f 3 dB = -----------------2R f C f
e. The sketch above shows that the gain drops at the rate of 20 dB decade and thus the drop of the gain from 40 dB to 0 dB represents two decades. And since at 40 dB , f 40 dB = 1 KHz ,
f 0 dB = 100 f 40 dB = 100 1000 = 100 KHz
The MATLAB script below plots the phase angle = atan ( RC ) as a function of frequency with RC = 1 . The script expresses the frequency normalized to the 3 dB frequency where f 3 dB = 1 KHz . Then, the normalized phase angle can be expressed as
n = atan ( f f 3 dB ) . f=(0.1:0.01:10); f3dB=1; theta=atan(f./f3dB).*180./pi; semilogx(f,theta); grid
* Please refer to Chapter 7 in Circuit Analysis I with MATLAB Applications, ISBN 9780970951120
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
589
Degrees
10
The plot above shows that the f 3 dB = 1 KHz frequency occurs at = 45 and the asymptotic (straight) line indicates that it drops at the rate of 45 ( decade ) . Therefore, the phase at 100 KHz is 90 and because this is an inverting amplifier, we must add 180 so the total phase shift is 270 or +90 . 14. The sdomain equivalent circuit is shown below.
IC ( s ) 1 s C f
I(s)
V in ( s ) R 1 = 10 K
V out ( s )
s , and thus
10 = 0.01 F
b.
590
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
f = 1 KHz
= 90
c. From (1) above we observe that the gain will be unity when = 10 r s or
10 f = ------- 1.6 KHz 2
4
15.
v in 1
R1 R2
if i
Rf
v in 2
v out
Since the magnitude of v out must 12 V or less and the feedback current i f must be limited to
1 mA , there must be
out R f = ------------ = --------- = 12 K -
12
if
10
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
591
16.
R1 R fix R var
The overall voltage gain can be found from relation (5.67), Page 544, that is,
R4 2 R2 v out A d = ----------------------------- = ----- -------- + 1 R3 R1 ( v in 2 v in 1 )
so we need to replace R 1 with R fix + R var . For convenience, we can make R 4 = R 3 and the above relation reduces to
2 R2 A d = -------------------------- + 1 R fix + R var
Let us choose a 100 K potentiometer for R var . For maximum gain, 100 for our case, the potentiometer must be set for zero resistance, in other words, R var = 0 . Then,
2 R2 A d max = --------- + 1 = 100 (1) R fix
For minimum gain, 10 for our case, the potentiometer must be set for full resistance, in other words, R var = 100 K . Then,
2R2 A d min = ----------------------- + 1 = 10 (2) 5 R fix + 10
592
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
b.
10 K Pot
+V R1 R2 V v in Rf v out
In both circuits of (a) and (b) above, the potentiometer can be adjusted to provide the value and polarity of the DC input voltage required to null the output to zero volts. 18.
v in 1
R1
Rf
v in 2
+
R2 R3
v out
R in = 2R 1
but this relation holds only if R 2 = R 1 . Since for the circuit above we want R in = 50 K , we must make R 1 + R 2 = 50 K , and if R 2 = R 1 , then R 1 = R 2 = 25 K . The circuit above will perform as a differential input op amp if
f 3 ----- = -----
R1
R2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
593
+
V in ( s )
V out ( s )
V1 = V2
2 V 1 = ------------------------- V in ( s ) 1 sC + R 2
Next,
V out ( s ) = V 1 R f I ( s )
V in ( s ) R2 1 sC V out ( s ) = ------------------------- V in ( s ) R f --------------- ------------------------- R 1 1 sC + R 2 1 sC + R 2 Rf R 1 R 2 R f sC V out ( s ) R2 1 sC G v ( s ) = ------------------ = ------------------------- ----- ------------------------- = -----------------------------------V in ( s ) 1 sC + R 2 R 1 1 sC + R 2 R 1 ( 1 sC + R 2 )
594
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Therefore,
V out ( j ) --------------------- = 1 V in ( j )
and because the magnitude is independent of frequency, this circuit is referred to as first order allpass filter. The phase angle is
= 180 2 tan ( C R 2 )
1
where the angle of 180 is due to the fact that the feedback resistor is connected to the inverting input. Because of this phase shift, this circuit is also referred to as a phase shifter. 20.
R V in 1 ( s )
V in 2 ( s )
R C
C V out ( s )
We will apply the superposition principle. With V in 1 ( s ) acting alone and V in 2 ( s ) = 0 , the circuit reduces to that shown below.
R
V in 1 ( s )
C
V out 1 ( s )
+
R
The circuit above is essentially a Miller integrator like that shown in Figure 5.52, Page 532, and in Exercise 14 we found that Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
595
(1)
Next, with V in 2 ( s ) acting alone and V in 1 ( s ) = 0 , the given circuit reduces to that shown below.
R V in 2 ( s )
R C
VC
C V out 2 ( s )
21.
R -----
2
VN
R --8 VD VC
R --4 VB
R --2 VA
V analog R
596
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
or
22.
1
2R V1 2R VA VB R V2 2R VC
2
R V3
3
R 2R VD 2R
V analog
By Cramers rule*
* For a review of matrices and determinants please refer to Numerical Analysis Using MATLAB and Excel, ISBN 9781 934404034.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
597
or
1V 3 = ----- ( 3V A + 6V B + 12V C + 8V D ) 32
and
where 16 = 2 = 2 . 23.
v out =
This integration can be performed by a summing integrator such as that shown below where the factor 1 RC was chosen to form the coefficients of v in 1 , v in 2 , and v in 3 in the given expression.
Reset
v in 1 v in 2 v in 3
1 M 500 K 100 K
1 F
v out
598
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
i v in
R1
Rf
v out
i v in
R1
i v out
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
599
i Rf
v in
1 M
v out
Alternately,
5100
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Chapter 6
Integrated Circuits
his chapter begins with an introduction to electronic logic gates and their function in terms of Boolean expressions and truth tables. Positive and negative logic are defined, and the transistortransistor logic (TTL), emittercoupled logic (ECL), CMOS, and BiCMOS logic families are discussed. Earlier logic families are presented in the exercises section.
Inverter
AND gate
OR gate
Four other logic gates, known as NAND, NOR, Exclusive OR (XOR), and Exclusive NOR (XNOR), are derivatives of the basic AND and OR gates and will be discussed later in this chapter.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
61
Figure 6.3. High (H) and Low (L) assignments in a voltage waveform
With the H and L assignments as shown in Figure 6.3, the logic circuitry designer has the option of assigning a logical 1 to H and logical 0 to L, or logical 0 to H and logical 1 to L. The former convention is known as positive logic, and the latter as negative logic. Thus, Figure 6.4(a) represents positive logic and Figure 6.4(b) represents negative logic.
log ical 1 L positive log ic (a) log ical 0 H L log ical 0 log ical 1 negative log ic (b) H
We will discuss the three common IC logic families in subsequent sections. For our present discussion it will suffice to list the High and Low voltage levels for the TransistorTransistor Logic ( TTL or T 2 L ), EmitterCoupled Logic ( ECL ), and Complementary Metal Oxide Semiconductor ( CMOS ) logic. Typical values are shown in Table 6.1.
TABLE 6.1 Typical values of High and Low voltage levels for three IC families
IC Family TTL ECL CMOS High voltage (Volts DC) 5.0 0.9 5.0 to 15.0 Low voltage (Volts DC) 0.2 1.75 0
62
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
The Inverter
TABLE 6.2 The truth table for the inverter
Input L H Output H L
With positive logic, the truth table for the inverter is written as shown in Table 6.3.
TABLE 6.3 The truth table for the inverter when positive logic is assumed
Input 0 1 Output 1 0
With negative logic, the truth table for the inverter is written as shown in Table 6.4.
TABLE 6.4 The truth table for the inverter when negative logic is assumed
Input 1 0 Output 0 1
Figure 6.6 shows the TTL SN7404 Hex Inverter IC where SN identifies the packaging and Hex implies that there are 6 inverters within the IC.
Vcc A6 Y6 A5
12 11
Y5 A4
10 9
Y4
8
1A 1Y 2A 2Y 3A 3Y GND
1 2 3 4 5 6 7
14 13 12 11 10 9 8
Vcc 6A 6Y 5A 5Y 4A 4Y
14
13
A1
Y1 A2
Y2 A3
Y3 GND
One important parameter is the input clamp voltage denoted as V IK and this refers to the maximum negative voltage that may be applied at the input terminals without damaging the IC. A typical value for this parameter is 1.2 V . To insure that this value is not exceeded, diode D 1 is included to clamp the input voltage to less than 1.2 V with respect to the ground. We recall from Chapter 2 that a practical diode, when forwardbiased, may be represented as an ideal diode in series with a 0.7 V source and a small resistance as shown in Figure 6.7. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
63
vD
0.7 V
The data sheets provided by the manufacturer list several parameters and we will discuss the most important later in this chapter. Figure 6.8 shows the internal details of the TTL IC 7404 inverter.
+Vcc 4 1 .6 130 T4 T1 Input A D2 Output Y T2
D1
T3 1
Note: Unless otherwise noted, henceforth Low will mean a nominal 0 V input or output signal and High will mean a nominal 5 V input or output signal. The inverter circuit of Figure 6.8 functions as follows: We assume that the input voltage is the output of a previous stage and it is Low at V in = 0.2 V with respect to the ground as shown in Figure 6.9. With this input, transistor T 1 is ON and the voltage V 1 at the collector of transistor T 1 is V 1 = V CE sat + V in = 0.2 + 0.2 = 0.4 V and under this condition, transistors T 2 and T 3 are both OFF. This is because if transistors T 2 and T 3 were
64
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
The Inverter
both ON, the voltage V 1 would have to be V 1 = V BE T2 + V BE T3 = 0.7 + 0.7 = 1.4 V . Accordingly, we accept the fact that with the input Low, transistors T 2 and T 3 are both OFF. However, the current I that flows through the 1.6 K resistor and then into the base of transistor T 4 , is sufficient to turn it ON. Then, V out = V CC V CE T4 V D2 = 5.0 0.2 0.7 = 4.1 V and thus the output is High and the inversion operation has been performed.
+Vcc 4 1 .6 130
ON V in = 0.2 V
Input A T1
I
OFF V1
T2
T4
ON
V CE sat 0.2 V
D2 Output Y
D1
OFF
T3 1
V out = 4.1 V
-
Next, with the input High the circuit is as shown in Figure 6.10 and the circuit functions as follows: We assume that the input voltage is the output of a previous stage and it is High at V in = 5.0 V with respect to the ground. With this input, the baseemitter junction of transistor T 1 is reverse biased, current flows from V CC through the 4 K resistor and thus the basecollector junction is forwardbiased. Under these conditions, transistor T 1 is said to be operating at the inverse active mode. The current that flows through the base collector junction of transistor T 1 is sufficient to turn transistor T 2 ON and this causes transistor T 3 to turn ON also. The voltage V 2 at the collector of transistor T 2 is V 2 = V CE T2 + V BE T3 = 0.2 + 0.7 = 0.9 V . However, this voltage is not sufficient to turn transistor T 4 ON because for transistor T 4 to be ON there should be V 2 = V BE T4 + V D2 + V out = 0.7 + 0.7 + 0.2 = 1.6 V . Therefore, we conclude that transistor T 4 is OFF, V out = 0.2 V , and thus the output is Low and the inversion operation has been performed.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
65
V2
V in = 5.0 V
Input A T1
T4
OFF
ON
T2
Acts as diode
0.2 V
D2 Output Y
D1
ON
T3 1
V out = 0.2 V
0.7 V
-
It should be noted that it is the diode D 2 that prevents transistor T 4 from being ON; if D 2 were not being there, T 4 would be also ON since there would be V BE T4 = 0.9 0.2 = 0.7 V which is just sufficient to turn T 4 ON.
With positive logic, the truth table for a 3input AND gate is written as shown in Table 6.6. Table 6.6 shows that the output of an AND gate is logical 1 (true) only when all inputs are logical 1 . Figure 6.12 shows the TTL SN7408 Quad 2input AND gate where Quad implies that there are 4 AND gates within the IC.
66
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
TABLE 6.6 Truth table of 3input AND gate when positive logic is used
A 0 0 0 0 1 1 1 1 Inputs B 0 0 1 1 0 0 1 1 C 0 1 0 1 0 1 0 1 Output D 0 0 0 0 0 0 0 1
1A 1A 1B 1Y 2A 2B 2Y GND
1 2 3 4 5 6 7 14 13 12 11 10 9 8
Vcc 4B 4A 4Y 3B 3A 3Y
1B 2A 2B 3A 3B 4A 4B
1Y
2Y
3Y
4Y
Figure 6.13 shows the internal details of a twoinput TTL AND gate. We will defer the analysis of this circuit and the reason for deferring the discussion after we introduce the NAND gate.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
67
Figure 6.13. Circuit for the SN7408 Quad 2input AND gate (Courtesy Texas Instruments)
Figure 6.14. Symbol for 3input OR gate TABLE 6.7 Truth table for 3input OR gate with positive logic
A 0 0 0 0 1 1 1 1 Inputs B 0 0 1 1 0 0 1 1 C 0 1 0 1 0 1 0 1 Output D 0 1 1 1 1 1 1 1
Table 6.7 shows that the output of an OR gate is logical 1 (true) whenever one or more of its inputs are logical 1 .
68
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Vcc 4B 4A 4Y 3B 3A 3Y
1B 2A 2B 3A 3B 4A 4B
1Y
2Y
3Y
4Y
Figure 6.16 shows the internal details of the TTL SN7432 Quad 2input OR gate. We will not describe the functioning of the SN7432 Quad 2input OR gate at this time. We will defer the circuit operation until we first describe the NOR gate operation in a subsequent section.
Figure 6.16. Circuit for the SN7432 Quad 2input OR gate (Courtesy Texas Instruments)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
69
Figure 6.17. Symbol for 3input NAND gate TABLE 6.8 Truth table for 3input NAND gate with positive logic
A 0 0 0 0 1 1 1 1 Inputs B 0 0 1 1 0 0 1 1 C 0 1 0 1 0 1 0 1 Output D 1 1 1 1 1 1 1 0
Table 6.8 shows that the output of an NAND gate is logical 0 (false) only when all inputs are logical 1 . Figure 6.18 shows the TTL SN7400 Quad 2input NAND gate where Quad implies that there are 4 NAND gates within the IC.
1A 1A 1B 1Y 2A 2B 2Y GND
1 2 3 4 5 6 7 14 13 12 11 10 9 8
Vcc 4B 4A 4Y 3B 3A 3Y
1B 2A 2B 3A 3B 4A 4B
1Y
2Y
3Y
4Y
Figure 6.19 shows the internal details of the IC SN7400 NAND gate where transistor T 1 is equivalent to two identical NPN transistors with their bases and collectors tied together; therefore, they are fabricated as a single device with 2 emitters but only one collector and one base as shown in Figure 6.19.
610
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Figure 6.19. Circuit for the TTL SN7400 Quad 2input NAND gate
We observe that, with the exception of two inputs for the 2input NAND gate, this circuit is practically the same as that of the inverter circuit of Figure 6.8 where transistor T 4 is stacked on top of transistor T 3 and the operation is complementary, that is, when transistor T 3 is ON transistor T 4 is OFF and vice versa. This arrangement is referred to as totempole configuration. The 2 input TTL NAND gate circuit of Figure 6.19 functions as follows: Let us assume that either Input A or B or both are Low at V in = 0.2 V with respect to the ground as shown in Figure 6.20. Under any of these three conditions, transistor T 1 is ON and the voltage V 1 at the collector of transistor T 1 is V 1 = V CE sat + V in = 0.2 + 0.2 = 0.4 V and thus transistors
T 2 and T 3 are both OFF. This is because if transistors T 2 and T 3 were both ON, the voltage V 1
would have to be V 1 = V BE T2 + V BE T3 = 0.7 + 0.7 = 1.4 V . Accordingly, we accept the fact that under any of these three conditions, transistors T 2 and T 3 are both OFF. However, The current I that flows through the 1.6 K resistor and then into the base of transistor T 4 is sufficient to turn it ON. Then, V out = V CC V CE T4 V D2 = 5.0 0.2 0.7 = 4.1 V and thus the output is High and this satisfies the 2input NAND gate operation. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
611
ON
ON T 1 V in = 0.2 V
Input A Input B
I OFF
V1 V CE sat
0.2 V
T2
D3 D2 D1 1 T3
OFF
Output Y
V out = 4.1 V
Figure 6.20. Circuit of the SN7400 2input NAND gate when either Input A of B or both are Low
Next, with both inputs High the circuit is as shown in Figure 6.21 and the circuit functions as follows: With V in = 5.0 V at both inputs, the baseemitter junction of transistor T 1 is reversebiased, current flows from V CC through the 4 K resistor and thus the basecollector junction is forwardbiased. In other words, transistor T 1 is operating at the inverse active mode. The current that flows through the base collector junction of transistor T 1 is sufficient to turn transistor T 2 ON and this causes transistor T 3 to turn ON also. The voltage V 2 at the collector of transistor T 2 is V 2 = V CE T2 + V BE T3 = 0.2 + 0.7 = 0.9 V . However, this voltage is not sufficient to turn transistor T 4 ON because for transistor T 4 to be ON the voltage V 2 should have the value V 2 = V BE T4 + V D2 + V out = 0.7 + 0.7 + 0.2 = 1.6 V . T h e r e f o r e , t r a n s i s t o r T 4 i s O F F ,
V OUT = 0.2 V , and this satisfies the 2input NAND gate operation.
It should be noted that it is the diode D 3 that prevents transistor T 4 from being ON; if D 3 were not being there, T 4 would be also ON since there would be V BE T4 = 0.9 0.2 = 0.7 V which is
612
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
T1
V2
ON
T2
V in = 5.0 V
Input A Input B
0.2 V
Acts as diode
D3 D2 D1 1 T3
ON
Output Y
OFF
Figure 6.21. Circuit of the SN7400 2input NAND gate when both inputs are High
Now, referring back to the AND gate circuit of Figure 6.13, we observe that this circuit essentially consists of a NAND gate followed by an inverter as shown in Figure 6.22.
A B AB AB Y
We mentioned earlier that one of the parameters specified by the manufacturer is the input clamp voltage denoted as V IK and this refers to the maximum negative voltage that may be applied at the input terminals without damaging the IC. A typical value for this parameter is 1.2 V . To insure that this value is not exceeded, diodes D 1 and D 2 are included in the circuit of the 2input NAND gate to clamp the input voltage to less than 1.2 V with respect to the ground.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
613
Figure 6.23. Symbol for 3input NOR gate TABLE 6.9 Truth table for 3input NOR gate with positive logic
A 0 0 0 0 1 1 1 1 Inputs B 0 0 1 1 0 0 1 1 C 0 1 0 1 0 1 0 1 Output D 1 0 0 0 0 0 0 0
Table 6.9 shows that the output of an NOR gate is logical 1 (true) only when all inputs are logical 0 (false). Figure 6.24 shows the TTL SN7402 Quad 2input NOR gate where Quad implies that there are 4 NOR gates within the IC.
1A 1Y 1A 1B 2Y 2A 2B GND
1 2 3 4 5 6 7 14 13 12 11 10 9 8
Vcc 4Y 4B 4A 3Y 3B 3A
1B 2A 2B 3A 3B 4A 4B
1Y
2Y
3Y
4Y
Figure 6.25 shows the internal details of the TTL IC SN7402 NOR gate.
614
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
T2 T4 D1 D2 1 T6
Input B
Output Y
Figure 6.25. Circuit for the TTL SN7402 Quad 2input NOR gate
The TTL 2input NOR gate circuit of Figure 6.25 functions as follows: When both inputs are Low at 0.2 V with respect to the ground, transistors T 1 and T 2 are both ON. The voltages at the base of transistors T 3 and T 4 are the same as the collector voltages of transistors s T 1 and T 2 , that is, 0.2 V , and thus transistors T 3 and T 4 are OFF. However, transistor T 5 is ON and therefore, V out = V CC V CE T5 V D5 = 5.0 0.2 0.7 = 4.1 V . If Input A is Low and Input B is High, transistor T 1 will be ON and transistor T 3 will be OFF. But transistor T 2 will behave as a junction diode turning transistor T 4 ON, and a voltage drop across the 1 K resistor will be developed and it will be sufficient to turn transistor T 6 ON, and thus V out = 0.2 V . If Input A is High and Input B is Low or both inputs are High, we find that
V out = 0.2 V also.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
615
Figure 6.26. Symbol for the XOR gate with positive logic TABLE 6.10 Truth table for 2input XOR gate
Inputs A 0 0 1 1 B 0 1 0 1 Output C 0 1 1 0
Figure 6.27 shows the TTL SN7486 Quad XOR gate where Quad implies that there are 4 XOR gates within the IC.
1A 1A 1B 1Y 2A 2B 2Y GND
1 2 3 4 5 6 7 14 13 12 11 10 9 8
Vcc 4B 4A 4Y 3B 3A 3Y
1B 2A 2B 3A 3B 4A 4B
1Y
2Y
3Y
4Y
Figure 6.28 shows the internal details of the TTL IC SN7486 XOR gate. The exclusiveNOR (XNOR) logic gate has two inputs and one output. The symbol for the XNOR gate is shown in Figure 6.29. and the truth table with positive logic is shown in Table 6.11. Table 6.11 shows that the output of a XNOR gate is logical 1 (true) only when the inputs are the same, that is, both logical 0 or both logical 1 . For this reason, the XNOR gate is also known as equivalence gate. There is no IC XNOR gate in the TTL family but one can be formed with an XOR gate (SN7486) followed by an inverter (SN7404). We can also implement the XNOR function using the TTL SN7486 IC with negative logic.
616
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
FanIn, FanOut, TTL Unit Load, Sourcing Current, and Sinking Current
Typical of all outputs VCC Equivalent of each input VCC 4 K 130
Input
Output
Figure 6.29. Symbol for 2input XNOR gate TABLE 6.11 Truth table for 2input XNOR gate with positive logic
Inputs A 0 0 1 1 B 0 1 0 1 Output C 1 0 0 1
6.9 FanIn, FanOut, TTL Unit Load, Sourcing Current, and Sinking Current
The fanin of a gate is the number of its inputs. Thus, a 3input NAND gate has a fanin of 3. Fanout is a term that defines the maximum number of digital inputs that the output of a single logic gate can feed. Generally, TTL gates can feed up to 10 other digital gates or devices. Thus, a typical TTL gate has a fanout of 10. In some digital systems, it is necessary for a single TTL logic gate to drive more than 10 other gates or devices. When this is the case, a device called a buffer can be used between the TTL gate and the multiple devices it must drive. A buffer of this type has a fanout of 25 to 30. An inverter (NOT gate) can serve this function in most digital circuits if complementation is also required. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
617
(6.1)
Thus, for a fanout of 10 which is typical for TTL gates, when the output is in High state, there is a current of 10 40 A = 0.4 mA , and when the output in Low state there is a current of 0 1.6 mA = 16 mA . We often hear the expressions passive pullup and active pullup. Figure 6.30 is an example of passive pullup where the resistor pullsup the output voltage towards V CC . and the word passive is used to indicate that the pullup device used is passive, i.e., a resistor.
+Vcc
R OFF T
The arrangement in Figure 6.31(a) is often referred to as active pullup since when transistor T 1 is ON, it pulls the output up towards V CC . The word active is used here to indicate that transistor T 1 , like all other transistors, is an active device. Sourcing and sinking currents refer to the current flow in TTL circuits. A driver gate* is said to be sourcing current when its output is a logic 1 as shown in Figure 6.31(a). A driver gate is sinking current when its output is a logic 0 as shown in Figure 6.31(b). The passive pullup arrangement is often used with open collector TTL devices such as the one shown in Figure 6.32. The advantage of the opencollector configuration over the totempole configuration is that the outputs of two or more opencollector TTL gates can be tied together to realize the AND function as shown in Figure 6.33.
* A driver gate is a gate whose output serves as an input to another gate referred to as a unit load.
618
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
FanIn, FanOut, TTL Unit Load, Sourcing Current, and Sinking Current
+Vcc
+Vcc
T1
T1
I sin k = 16 mA
( 10 unit loads )
T2
T2
V OH = 2.4 V
V OL = 0.4 V
(a)
+Vcc 4 T1 T2
Input A Input B X (open collector point)
1.6
D1
D2 T3 1
Output Y
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
619
To see why the outputs of two or more totem pole TTL gates should not be tied together for wired AND operation, let us consider the arrangement shown in Figure 6.34. With the collectors of transistors T 2 and T 4 as shown, the current I can be as high as 55 mA which most likely will damage transistors T 1 and T 4 , because per manufacturers specifications these transistors can sink only 16 mA .
+Vcc 130 T1
ON
+Vcc 130 T3
OFF
T2
OFF
T4
ON
620
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Data Sheets
TABLE 6.12 Typical parameters for TTL logic circuits provided by IC manufacturers
Parameter 1 2 3 4 5 Supply voltage V CC T A = 25C Input clamp voltage V CC = Min , T A = 25C , I IN = 12 mA Conditions Min Typical Max
4.75
5.00
5.25 1.5
Units V V V
V IK
High Level Input Voltage V IH Low Level Input Voltage V IL High Level Output Voltage V OH Low Level Output Voltage V OL High Level Input Current I IH Low Level Input Current I IL
V CC = Min V CC = Min V CC = Min , V IL = 0.8 V , I OH = 0.4 mA V CC = Min , V IH = 2.0 V , I OL = 16 mA V CC = Max , V IN = 2.4 V V CC = Max , V IN = 5.5 V V CC = Max , V IN = 0.4 V
2.0 0.8 2.4 3.4 0.2 0.4 40 1 1.6 18 2.4 0.8 7 55 4.4 1.6 15
V V
6 7
A mA mA mA mA mA
ns
8 9
10 Low Level Supply Current I CCL 11 High Level Supply Current I CCH
12 Propagation Delay V CC = 5.0 V , T A = 25C , R L = 400 Time High to Low C L = 15 pF Level t PHL 13 Propagation Delay V CC = 5.0 V , T A = 25C , R L = 400 Time Low to High C L = 15 pF Level t PLH
11
22
ns
3. The high level input voltage V IH , also referred to as logical 1 input voltage is the minimum input voltage level that the IC device will recognize as a valid logical 1 input as shown in Figure 6.35. 4. The low level input voltage V IL , also referred to as logical 0 input voltage is the maximum input voltage level that the IC device will recognize as a valid logical 0 input as shown in Figure 6.35. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
621
5.0 V
V IL
Figure 6.35. Valid logical 1 and valid logical 0 input regions for a typical IC gate
5. The high level output voltage V OH , also referred to as logical 1 output voltage is the minimum output voltage level that the IC device will recognize as a valid logical 1 output. 6. The low level output voltage V OL , also referred to as logical 0 output voltage is the maximum output voltage level that the IC device will recognize as a valid logical 0 output when a current of 16 mA , representing a fanout of 10 , is sinked by transistor T 2 as indicated in Figure 6.36.
+Vcc
130
T1
Output = 0.4 V
T2
I sin k = 16 mA
622
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Data Sheets
7.a. For a 2input NAND gate, when the input voltage is 2.4 V , the maximum value of the sum of the currents I IH furnished by the input lines should not be greater than 40 A for ten unit loads. 7.b. For a 2input NAND gate, when the input voltage is 5.5 V , the maximum value of the sum of the currents I IH furnished by the input lines should not be greater than 1 mA for ten unit loads. 8. When the input voltage is low at 0.4 V , the maximum value of the current I IL should not exceed 1.6 mA where the minus sign indicates the direction of the current flow as shown in Figure 6.37 where for a logical 0 input voltage the ten gates at the right side source (supply) 1.6 mA each, and the single gate on the left sinks (accepts) 1.6 10 = 16 mA .
VCC
16 mA
1.6 mA 1.6 mA
1.6 mA
10gates
Figure 6.37. An illustration of the input current I IL for a 2input NAND gate
9. The output shortcircuit current I OS is the output current when the output is shorted to ground as shown in Figure 6.38. The data sheet specifies an output short circuit current of 55 mA maximum. When transistor T 1 is ON, transistor T 2 is OFF, and the output voltage is grounded, the resistor will act as a current limiter to limit the current in the range 18 mA to 55 mA . Thus, the resistor provides short circuit protection. 10. The Low Level Supply Current I CCL is the current that the voltage supply V CC must furnish when only one of the four TTL 2input NAND gates within the IC SN7400 has its output at logical 0 . If two outputs are at logical 0 , V CC must furnish twice the amount of current specified in the data sheet, that is, 2 4.4 = 8.8 mA . Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
623
I short circuit
T2
11. The High Level Supply Current I CCH is the current that the voltage supply V CC must furnish when only one of the four 2input NAND gates within the IC SN7400 has its output at logical 1 . If two outputs are at logical 1 , V CC must furnish twice the amount of current specified in the data sheet, that is, 2 1.6 = 3.2 mA . 12. The propagation delay time t PHL is the time required for a logical 1 Input A or logical 1 Input B to appear as logical 0 at Output Y. A typical value is 7 ns . 13. The propagation delay time t PLH is the time required for a logical 0 Input A or logical 0 Input B to appear as logical 1 at Output Y. A typical value is 11 ns .
624
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
IC1
T1 VB1=v s
T2 VB2= v s
IE
RE
VEE
Example 6.1 Find the differential voltage gain for the circuit in Figure 6.40.
+Vcc RC1 1 K IC1 RC2 1 K IC2 VOUT1 VOUT2 T1 VB1 T2 VB2 Differential Output
RE = 8 K IE
VEE = 13.4 V
Solution: The basetoemitter resistance is small compared to the external resistor R E and thus
V EE V BE 13.4 0.6 I E ------------------------ = ----------------------- = 1.6 mA RE 8 K
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
625
From Chapter 3, relation (3.78), at room temperature g m = 40I C where I C is in milliamps. Thus,
g m = 40 0.8 10
3
= 0.032
di C g m = ---------dv BE
iC = IC
The differential circuit of Figure 6.39 forms the basis for the emittercoupled logic (ECL) electronic gates. The basic ECL circuit is shown in Figure 6.41.
RC1 250
VEE = 5.2 V
In contrast to TTL where the transistors are either cut off or saturated, depending on the state of the circuit, in ECL the transistors always operate in the active region so they can change state
626
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
and
Therefore, with V in = 1.75 V (logical 0 ), we obtain V out 1 = 0 V and V out 2 = 1.03 V . Next, let V in = 0.9 V , i.e., logical 1 . Because V E = 1.92 V , transistor T 1 conducts hard, and transistor T 2 is OFF, and V out 2 is essentially connected to the ground. Therefore all current flows through transistor T 1 and the collector current of this transistor is
V E ( V EE ) 1.92 ( 5.2 ) I C1 I E = ----------------------------- = ----------------------------------- = 4.1 mA 800 RE
and
Therefore, with V in = 0.9 V (logical 1 ), we obtain V out 1 = 1.03 V and V out 2 = 0 V . We observe that: 1. V out 1 and V out 2 are complements of each other. 2. The output voltage levels are not the same as input voltage levels, that is, the input levels are 1.75 V and 0.9 V and the outputs are 0 V and 1.03 V . However, the output levels can be made comparable to the input levels by attaching two emitter followers to the outputs as shown in Figure 6.42. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
627
RC1 250
Emitter follower
T2
V'OUT2
VOUT1
VREF = 1.32 V
100 2 V
With the emitterfollowers connected at the outputs V' out 1 and V' out 2 of the basic differential circuit and recalling that in the emitterfollower configuration the output voltage is essentially the same as the input, the outputs at the emitterfollowers are
V out 1 = V' out 1 V BE = 1.03 0.75 = 1.78 V ( log ical 0 )
and
Figure 6.43 shows a typical 2input ECL gate and. as before, the resistor values are approximate. The symbol for a 2input ECL gate is shown in Figure 6.44. Two types of ECL gates are the ECL 10K and ECL 100K, the latter having a propagation delay time of 0.75 ns and power consumption of 40 mW . It should also be noted that ECL is not a new technology; Motorola introduced the MECL series in the 1980s.
628
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
RC1 250
RC2 IC1 250 IC2 Emitter follower Emitter follower A+B A+B
T1
T2 VBE2 = 0 .6 V VE RE = 800 IE
T3 VREF = 1.32 V
VEE = 5.2 V
An nchannel MOSFET or simply NMOS device, behaves like an open switch, as shown in Figure 6.46, when V GS < V T . where V T = 1.5 V and it is referred to as the threshold voltage.
D G V GS + _ S S S N V DD D 10
10
or
Open switch
V GS < V T
Figure 6.46. Condition under which an NMOS device behaves as an open switch
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
629
or
Closed switch
V GS > V T
Figure 6.47. Condition under which an NMOS device behaves as a closed switch
A pchannel MOSFET or simply PMOS device, behaves like an open switch, as shown in Figure 6.48, when V GS > V T . where V T = 1.5 V is the threshold voltage.
D G V GS + _ S S S V DD D 10
10
or
Open switch
V GS > V T
Figure 6.48. Condition under which a PMOS device behaves as an open switch
An PMOS device behaves like a closed switch, as shown in Figure 6.49, when V GS < V T .
D G V GS + _ S S S V DD D 10
3
or
Closed switch
V GS < V T
Figure 6.49. Condition under which a PMOS device behaves as a closed switch
For convenience, in our subsequent discussion we will the state of an NMOS or PMOS device ON or OFF (closed or open switch) by a low resistance ( 1 K ) or a very high resistance ( 10 10 ) respectively. Also, since in integrated circuits resistors are being replaced by NMOS or PMOS devices because these devices require much less space than physical resistors, we will represent these with resistor symbols with a typical value of 100 K . We will denote these as N L or P L where the subscript L stands for load.
630
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Thus, the inversion operation has been performed. With V in = 5 V (logical 1 ), the NMOS device N is ON, R X = 10 3 , N L = R D = 10 5 , and the output is
RX 10 V out = ------------------- V DD = -------------------- 5 0.05 V 3 5 RX + RD 10 + 10
V DD = 5 V RD D + V in _ G N S + V out _ + V in _ G S V DD = 5 V V DD = 5 V RD 10
3 5 3
or
G S D N
D NL
+ V out _
RX
10
+ V out _
or
10
10
and again the inversion operation is performed. The truth table for positive logic is shown in Table 6.13.
TABLE 6.13 Truth table for NMOS inverter V in 0 1 V out 1 0
A PMOS inverter has a similar arrangement where the NMOS devices are replaced by PMOS devices.
631
V DD = 5 V
RD
10 +
or
10 3
10
V out
10
or
10
10
Figure 6.51. 2input NMOS NAND gate TABLE 6.14 Truth table for 2input NMOS NAND gate A B V
out
0 0 1 1
0 1 0 1
1 1 1 0
632
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
V DD = 5 V 10 + 10
3 5
or
10
or
10
V out _
10
10
Figure 6.52. 2input NMOS NOR gate TABLE 6.15 Truth table for 2input NMOS NOR gate A B V
out
0 0 1 1
0 1 0 1
1 0 0 0
V DD = 5 V
V DD = 5 V
10
10 + V out 5 V _
10
or
10
3
V out 0 V _
V in = 5 V
In Figure 6.53 with V in = 0 V (logical 0 ), device P is ON, device N is OFF, and the output is Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
633
and the inversion operation has been performed. With V in = 5 V (logical 1 ), device P is OFF, device N is ON, and the output is
RX 10 V out = ------------------- V DD = ---------------------- 5 0 V 3 10 RX + RD 10 + 10
3
and again the inversion operation is performed. The truth table for positive logic is shown in Table 6.16.
TABLE 6.16 Truth table for CMOS inverter V in 0 1 V out 1 0
P
D D
P
D +
or
10 3
or
3
10
10 + V out _
10
+ VA _ + VB _
G S D
or
10
N V out N S
10 10
10 3
or
10 _
634
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
or
10 3
10
10
or
10
10
10
3
10 V out 10
10
or
or
10
V out
10
Figure 6.55. 2input CMOS NOR gate TABLE 6.18 Truth table for 2input CMOS NOR gate VA 0 0 1 1 VB 0 1 0 1 V out 1 0 0 0
635
The symbol of a buffer gate is shown in Figure 6.57 and its truth table is shown in Table 6.19.
A B
Figure 6.57. Symbol for a buffer gate TABLE 6.19 Truth table for buffer amplifier with positive logic A 0 1 B 0 1
The TTL ICs SN7407 and SN7417 are a hex buffers meaning that each contains 6 buffers featuring highvoltage, opencollector outputs for interfacing with highlevel circuits (such as CMOS) or for driving highcurrent loads (such as lamps or relays), and also are characterized for use as buffers for driving TTL inputs. Figure 6.58 shows the circuits for these ICs.
Figure 6.58. Circuit for the IC SN7407/SN7417 buffer (Courtesy of Texas Instruments)
The circuits for the SN7407/SN7417 are completely compatible with most TTL families. The SN7407 has minimum breakdown voltages of 30 V, and the SN7417 has minimum breakdown voltages of 15 V. The maximum sink current is 40 mA for both the SN7407 and SN7417. Inputs are diode clamped to minimize transmissionline effects, which simplifies design. Typical power dissipation is 145 mW , and average propagation delay time is 14 ns .
636
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Figure 6.59. Symbols for the TTL SN74125 and SN74126 devices
The output of the SN74125 device is disabled when G is High, and the output of the SN74126 is disabled when G is Low. The truth table for the SN74125 device is shown in Table 6.20.
TABLE 6.20 Truth table for the SN74125 3state buffer A B G 0 0 1 1 0 1 0 1 0 1 High Z High Z
Figure 6.60 shows the symbol for the TTL SN74366 hex tristate inverting buffer, and Table 6.21 is the truth table for that device.
G A B
Figure 6.60. Symbol for the SN74366 hex tristate inverting buffer TABLE 6.21 Truth table for the TTL SN74125 3state buffer A B G 0 0 1 1 0 1 0 1 1 0 High Z High Z
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
637
V in
D G
V out
As shown in Figure 6.61, the CMOS transmission gate circuit consists of one NMOS and one PMOS connected in parallel and controlled by inverted gate voltages. With this arrangement, both NMOS and PMOS can be switched ON of OFF at the same time rather than alternately. If they are ON, the resistance between V IN and V OUT is very low and the signal is transmitted without degradation. If they are OFF, there is no path from the input to the output. Also, the CMOS transmission gate can be used for both analog and digital signals and the only requirement is that the signal does not exceed the power supply voltages. Figure 6.62 shows the symbol for the CMOS transmission gate.
C
VIN
VOUT
A data bus is a group transmission paths such as groups of wires used as a common path to interconnect several devices. A data bus, or simply a bus can be unidirectional, that is, one in which data can flow in one direction only, or bidirectional where data can flow in either direction where each direction can be controlled by tristate devices as shown in Figure 6.63 where the control line C A B directs the data flow from left to right and C B A from right to left.
638
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A2
B2
An
Bn
CA B
CB A
Figure 6.64. Bipolar NPN transistor with Schottky diode and its representation as Schottky transistor
To achieve lowpower consumption, the resistances are considerably higher in the upgraded versions of the TTL family and the ICs are identified with the LS designation which stands for Low power Schottky device. Thus, the Lowpower Schottky 2input NAND gate is identified as SN74LS00 whose internal construction is shown in Figure 6.65.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
639
T4
T5
Figure 6.65. Internal construction of the SN74LS00 2input NAND gate (Courtesy Texas Instruments)
A comparison of the SN7400 2input NAND gate shown in Figure 6.19 with SN74LS00 2input NAND gate shown in Figure 6.65, reveals that in the latter all resistances are higher, the conventional diodes have been replaced by Schottky diodes, and all transistors except T 5 , have been replaced by Schottky transistors. This is because transistor T 5 never goes into saturation as it is explained below. Let us consider the terminal voltages of the Schottky transistor shown in Figure 6.66.
V BC = 0.5 V B V BE = 0.8 V C
V CE = 0.3 V
E
(6.2)
From Figure 6.65, by application of (6.2) and noting that V CB = V BC , we find that
640
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
and therefore we conclude that transistor T 5 never enters the saturation region. A more recent advancement in TTL technology is the Advanced Lowpower Schottky denoted as SN74ALSXX. Thus, a 2input NAND gate in this technology is indicated as SN74ALS00. The propagation delay time is about 4 ns compared to about 11 ns for the standard TTL, and the power consumption is about 1 mW compared to about 10 mW for the standard TTL. Another recent technology is the BiCMOS. It combines bipolar and CMOS to give the best balance between available output current and power consumption. As we know, bipolar transistors offer high speed, high gain, and low output resistance, whereas CMOS technology offers high input resistance, and this combination results in lowpower logic gates. Basically, a BiCMOS is a twostage amplifier that uses a CMOS device in the first stage and a bipolar transistor in the second stage. This device can be used with both analog and digital signals. We will not discuss the BiCMOS technology in detail. We will only show the BiCMOS inverter and the BiCMOS 2input NAND gate. The BiCMOS inverter circuit is shown in Figure 6.67. This inverter functions as follows:
V DD = 5 V
S G D + V in _ G S
P
T1
D N T2
+ V out _
When V in = 0 V , the NMOS device and transistor T 2 are OFF while the PMOS device and transistor T 1 are ON. Then, V out = V DD V CET1 5 0.2 = 4.8 V . When V in = 5 V , the PMOS device and transistor T 1 are OFF while the NMOS device and transistor T 2 are ON. Then, V out = V CET2 0.2 V . Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
641
S P D
S P D T1
R D + V _ A + V _ B G N S D N S R T2 _ V out +
As we mentioned in Chapter 3, gallium arsenide (GaAs) is the fastest technology with propagation delay times about 20 ps and even though prices for these devices have fallen considerably during the last few years, the cost is still very high in comparison with other IC families. Undoubtedly, faster and with lower power consumption devices will emerge in the future. The interested reader is urged to check occasionally the Internet for new products developed by Texas Instruments, Motorola, National Semiconductor, Fairchild, and other IC manufacturers for new products.
642
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
643
644
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Summary
In contrast to TTL where the transistors are either cut off or saturated, depending on the state of the circuit, in ECL the transistors always operate in the active region so they can change state very rapidly. The basic ECL circuit is essentially a differential amplifier. The ECL gate provides two outputs, one which implements the OR function, and the other implements the NOR function. Two types of ECL gates are the ECL 10K and ECL 100K, the latter having a propagation delay time of 0.75 ns and power consumption of 40 mW . It should also be noted that ECL is not a new technology; Motorola introduced the MECL series in the 1980s. The earlier NMOS and PMOS logic gates have now been superseded by CMOS logic gates. CMOS logic uses both NMOS and PMOS devices to form logic functions. CMOS technology is the dominant semiconductor technology for the manufacturing of microprocessors, memories and application specific integrated circuits. The main advantage of CMOS over NMOS and bipolar technology is the much smaller power dissipation. A buffer or line driver is normally used to change (amplify) the voltage level at the output of a logic circuit. Some other ICs are tristated buffers or gates. The output of these devices can assume three states, logic 0, logic 1, and HighZ (high impedance). The CMOS transmission gate is another tristate device where the control input serves as an open or closed switch. A data bus is a group transmission paths such as groups of wires used as a common path to interconnect several devices. A data bus, or simply a bus can be unidirectional, that is, one in which data can flow in one direction only, or bidirectional where data can flow in either direction where each direction can be controlled by tristate devices. A more recent advancement in TTL technology is the Advanced Lowpower Schottky denoted as ALS. Thus, a 2input NAND gate in this technology is indicated as SN74ALS00. The propagation delay time is about 4 ns compared to about 11 ns for the standard TTL, and the power consumption is about 1 mW compared to about 10 mW for the standard TTL. The BiCMOS technology combines bipolar and CMOS to give the best balance between available output current and power consumption. Bipolar transistors offer high speed, high gain, and low output resistance, whereas CMOS technology offers high input resistance, and this combination results in lowpower logic gates. Basically, a BiCMOS is a twostage amplifier that uses a CMOS device in the first stage and a bipolar transistor in the second stage. This device can be used with both analog and digital signals. Gallium arsenide (GaAs) still remains the fastest technology with propagation delay times about 20 ps and even though prices for these devices have fallen considerably during the last few years, the cost is still very high in comparison with other IC families.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
645
a. Write the truth table for the input combinations of 0 V and 5 V with respect to the ground. b. State the conditions under which this circuit can be classified as an AND gate. c. State the conditions under which this circuit can be classified as an OR gate. d. State some of the deficiencies of this circuit when used as a logic gate. 2. The circuit below is known as Diode Logic (DL) gate.
V1 V2 15 K 5V
V out
Write the truth table for the input combinations of 0 V and 3 V with respect to the ground. Which type of logic gate does this circuit represent? 3. The circuit below is another form of a Diode Logic (DL) gate.
V1 V2 1 K V out
Write the truth table for the input combinations of 0 V and 5 V with respect to the ground. Which type of logic gate does this circuit represent?
646
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
4. The circuit below is a 3input ResistorTransistor Logic (RTL) gate.
750 T1 750 T2 750 T3 1 K V out 5V
V1
V2
V3
Write the truth table for the input combinations of 0 V and 5 V with respect to the ground. Which type of logic gate does this circuit represent? 5. For the circuit below find V out 1 and V out 2 when each of V 1 through V 4 assumes the values of
0 V and 5 V with respect to the ground.
750
V3
T3
V4 V1 V2 750 T1 T2 1 K V out 1 5V
750
T4
1 K
V out 2
750
5V
6. For the circuit of Exercise 5, derive an expression for V out 1 High level when the driving gate has a fanout of 2. Hint: Start with an equivalent circuit. 7. The circuit below is a 3input DiodeTransistor Logic (DTL) gate.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
647
R1 D1 V1 V2 V3 D2 D3
2 K D4 D5 R3
R2
2 K V out
20 K
Write the truth table for the input combinations of 0 V and 5 V with respect to the ground. Which type of logic gate does this circuit represent? 8. It is known that the inverter circuit shown below has the value h FE = 30 minimum. Calculate the levels of the output voltage v out when the levels of the input voltage v in are 0 and 12 volts. You may first assume that v out ( sat ) = 0 V to obtain approximate values and then assume that v out ( sat ) = 0.2 V for exact values.
V CC = 12 V RC R1 2.2 K
v out
R2
v in 15 K
100 K
V BB = 12 V
648
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
V1 = 0 V
= V out
V1 = 0 V
+ V out
V2 = 0 V
V2
0V 5V 0V 5V
V out
0V 1.67 V 1.67 V 3.34 V
b. This circuit can be classified as an AND gate if V out 3.34 V c. This circuit can be classified as an OR gate if V out 1.67 V d. This circuit is not practical for use as a logic gate because if more than two inputs exist, more different output levels result. Moreover, there is no compatibility between the input and output levels. With the above arrangement for instance, with V 1 = V 2 = 5 V the output is V out = 3.34 V and if this output becomes the input of another gate shown below as Gate 3, the outputs drop to 2.72 V as shown in parentheses. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
649
2.
V1 V2 15 K 5V
V out
Both diodes are forwardbiased for any combination of the inputs since the value of V CC is greater than 3 V and can be represented as in the circuit below.
0.7 V V1 V2
0.7 V
15 K 5V V CC
V out
By inspection, when V 1 = V 2 = 0 V , or V 1 = 0 V and V 2 = 3 V , or V 1 = 3 V and V 2 = 0 V , the output is V out = 0.7 V . When V 1 = V 2 = 3 V , the output is V out = 3.7 V and thus the truth table is as shown below.
V1
0V 0V 3V 3V
V2
0V 3V 0V 3V
V out
0.7 V 0.7 V 0.7 V 3.7 V
The truth table indicates that this circuit could be used as an AND gate.
650
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
the output is V out = 5.0 0.7 = 4.3 V and thus the truth table is as shown below.
V1
0V 0V 5V 5V
V2
0V 5V 0V 5V
V out
0V 4.3 V 4.3 V 4.3 V
The truth table indicates that this circuit could be used as an OR gate. 4.
750 T1 750 T2 750 T3 1 K V out 5V
V1
V2
V3
By inspection, when V 1 = V 2 = V 3 = 0 V , none of the three transistors conduct and thus V out = 5 V . When one or more of the input voltages is 5 V one or more of the transistors will saturate and the output will be V out = 0.2 V 0 V . The truth table is as shown below and thus this circuit behaves as a 3input NOR gate.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
651
Inputs
Output
V1
0V 0V 0V 0V 5V 5V 5V 5V
V2
0V 0V 5V 5V 0V 0V 5V 5V
V3
0V 5V 0V 5V 0V 5V 0V 5V
V out
5V 0V 0V 0V 0V 0V 0V 0V
5.
V3 750 T3
V4 V1 V2 750 T1 T2 1 K V out 1 5V
750
T4
1 K
V out 2
750
5V
This circuit is a 2input NOR gate with fanout of 1. We recall that fanout is the number of gate inputs that can be driven by a single gate output. Thus, in the circuit above the output of the first 2input NOR gate drives (is connected to) one of the inputs of the second 2input NOR gate. By inspection, when V 1 = 0 V and V 2 = 5 V , or V 1 = 5 V and V 2 = 0 V , or V 1 = V 2 = 5 V , one of both transistors T 1 and T 2 conduct and thus V out 1 = 0.2 V 0 V . When V 3 = 0 V and V out 1 = 0 V , V out 2 = 5 V . Next, let us examine the case where V 1 = V 2 = 5 V . Under this condition V out1 < 5 V because current flows through the 750 at the base of transistor T 4 which now conducts, and and V BET4 = 0.7 V . We can now find that current from the equivalent circuit below.
652
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
1 K 5V I
0.7 V
The actual value of V out 1 is 5 V minus the voltage drop across the 1 K resistor. Thus,
V out 1 = 5 10 2.46 10
3 3
= 2.54 V
We see then that by connecting one input of a gate to the output of another gate causes the high level to decrease from its open circuit value of 5 V to the value of 2.54 V . 6. For a fanout of 2, the equivalent circuit is as shown in Figure (a) below and its Thevenin equivalent is as shown in Figure (b).
1 K 5V
750 0.7 V
1 K 5V I (b)
750 2 0.7 V
and
1000 V out 1 = 5 --------------------------------- 4.3 = 1.87 V 1000 + 750 2
The equivalent circuit of Figure (a) can be extended to the situation where there are N gates driven by one output of a gate, that is a fanout of N, by replacing the resistance 750 2 by 750 N . Then,
1000 V out 1 = 5 --------------------------------- 4.3 1000 + 750 N
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
653
R1 D1 V1 V2 V3 D2 D3
2 K D4 D5 R3
R2
2 K V out
20 K
With one or more of the inputs V 1 , V 2 , or V 3 at 0 V , the given circuit becomes as shown below where the 0.2 V is assumed to be the output of a previous logic gate with logical 0 .
V CC = 5 V
R1 0.7 V 0.2 V VA
2 K D4 D5 R3 VB
R2
2 K V out = 5 V
20 K
With V A = 0.2 V + 0.7 V = 0.9 V , the transistor is OFF. For the transistor to be ON, the voltage V A with respect to the ground would have to be
V A = V D4 + V D5 + V BE = 0.7 + 0.7 + 0.7 = 2.1 V
above ground. Therefore, V out = 5 V . Next, we will consider the case where all three inputs are high, that is, V 1 = V 2 = V 3 = 5 V . In this case, the three input diodes D 1 , D 2 , and D 3 are reversebiased and thus do not con-
654
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
R1 2 K
VB
R2
2 K
To determine the output voltage V out we need to find out whether the transistor is ON or OFF. Let us assume that it is OFF. Then, by KVL
R 1 I 0.7 V 0.7 V + R 3 I = V CC = 5 V 5 ( 0.7 + 0.7 )V B = V BE = R 3 I = 20 K ------------------------------------- = 3.27 V 2 K + 20 K
This voltage is more than sufficient to turn the transistor on and thus V out = 0.2 V . The two diodes D 4 and D 5 in series are both necessary for the following reason: If only one diode were used, when at least one of the input voltages is Low, the voltages V A and V B would be as shown below.
VA 0.9 V VD R3 VB
From the circuit above, V BE = V B = V B V B = 0.9 0.7 = 0.2 V and thus the transistor will be OFF. But this would not be a good design since any small noise voltage such as 0.5 V when added to 0.2 V would result to 0.2 + 0.5 = 0.7 V and this would turn the transistor ON.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
655
V1
0V 0V 0V 0V 5V 5V 5V 5V
V2
0V 0V 5V 5V 0V 0V 5V 5V
V3
0V 5V 0V 5V 0V 5V 0V 5V
V out
5V 5V 5V 5V 5V 5V 5V 0V
Note: Like the RTL, DTL is no longer used as a logic family. It is one of the earlier families employed years ago with power consumption of about 10 mW and propagation delay time of approximately 30 ns . 8. With v in = 0 V the circuit is as shown below.
V CC = 12 V 2.2 K
VB 15 K 100 K V BB = 12 V
v out = 12 V
and this voltage will certainly keep the transistor at cutoff. Therefore, for v in = 0 V , v out = 12 V and the inversion operation has been performed.
656
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
VB
v out ( sat ) = 0 V
v in = 12 V
15 K I2
V BB = 12 V
Let the minimum base current for saturation be denoted as I B ( min ) . Then, I B ( min ) = I C h FE where
12 V I C = ---------------- = 5.45 mA 2.2 K
and thus
IC --------------------I B ( min ) = ------- = 5.45 mA = 0.182 mA h FE 30
The actual base current can be found from the circuit above. We find that
12 V I 1 = ---------------- = 0.80 mA 15 K 12 V I 2 = ------------------- = 0.12 mA 100 K
and
I B = I 1 I 2 = 0.80 0.12 = 0.68 mA
Since this value is considerably larger than I B ( min ) , we accept the fact that the transistor is in saturation. Next, we assume that v out ( sat ) = 0.2 V to obtain more exact values. The circuit then is as shown below, and since the transistor is in saturation, there must be
v BE ( sat ) = 0.7 V
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
657
IC 5.1 mA I B ( min ) = ------- = ------------------ = 0.17 mA 30 h FE I 1 = 12 V 0.7 V = 0.75 mA ------------------------------15 K ) I 2 = 0.7 V ( 12 V - = 0.13 mA --------------------------------------100 K
and
I B = I 1 I 2 = 0.75 0.13 = 0.62 mA
We observe that the more exact computations do not differ significantly from those obtained by assuming that the transistor in saturation is essentially a short circuit. Therefore, when v in = 12 V , v out ( sat ) = 0.2 V and again the inversion operation is performed.
658
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Chapter 7
Pulse Circuits and Waveform Generators
his chapter is an introduction to several pulsed circuits and waveform generators also known as relaxation oscillators. These circuits are part of a family of circuits that include differentiators, integrators, clipping and clamping circuits, pulsetiming and delay circuits, logic circuits, and switching circuits. In this chapter we discuss the three types of multivibrators, the 555 Timer, and the Schmitt trigger. Sinusoidal oscillators are discussed in Chapter 10.
Output 2
Let us now consider Output 1 and assume that the time ON T 1 and time OFF T 2 are equal and the period is T as shown in Figure 7.3.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
71
Figure 7.3. Waveform showing period T and times ON and OFF for the multivibrator circuit of Figure 7.1
Comparator 1 R Q S Q Output
1 3 V CC
+
Comparator 2
We observe that this IC includes a resistive voltage divider consisting of three identical resistors and this divider sets the voltage at the plus (+) input of the lower comparator at 1 3 V CC and at plus (+) input of the upper comparator at 2 3 V CC . The outputs of the comparators determine
72
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
T1 T
T2
Figure 7.6 shows an astable multivibrator employing a 555 Timer. We will see that with this circuit the duty cycle will always be greater than 0.5 as shown in Figure 7.7. For the circuit of Figure 7.6 let us first assume that the capacitor is uncharged and the SR flipflop is Set. In this case the output is High and the transistor is does not conduct. The capacitor then will charge towards V CC through the resistors R A and R B . When the capacitor reaches the value v C = 1 3 V CC the output of Comparator 2 goes Low and the SR flipflop remains Set.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
73
RA R RB
2 3 V CC
+
R
Comparator 1 R Q S Q
T1 T 2
vC C
1 3 V CC
Output
+
Comparator 2 R
Output waveform
Figure 7.7. Input and output waveforms for the astable multivibrator of Figure 7.6
When the capacitor reaches the value v C = 2 3 V CC , the output of Comparator 1 goes High and resets the SR flipflop and thus the output Q goes Low, Q goes High, the transistor becomes saturated, its collector voltage becomes almost zero, and since it appears at the common node of resistors R A and R B , the capacitor begins to discharge through resistor R B and the collector of the transistor. The capacitor voltage v C decreases exponentially with time constant t d = R B C
74
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(7.1)
where V is the final value and V in is the initial value of the voltage across the capacitor. For the charging interval T 1 the circuit is as shown in figure 7.8.
V CC RA
RB
vC ( t )
C
= V CC ( 2 3 V CC )e
t ( R A + R B )C
(7.2)
= 13
T 1 ( R A + R B )C
= 12
T 1 ( R A + R B )C = ln ( 1 2 ) = ln 1 ln 2 = 0 ln 2
or
T 1 = ln 2 ( R A + R B )C = 0.69 ( R A + R B )C
(7.3)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
75
vC ( t )
C
= ( 2 3 V CC )e
t R B C
(7.4)
= 13
T2 RB C
= 12
T 2 R B C = ln ( 1 2 ) = ln 1 ln 2 = 0 ln 2
or
T 2 = ln 2 ( R A + R B )C = 0.69R B C
(7.5)
(7.6)
The right side of (7.6) cannot be negative; accordingly, the circuit of Figure 7.9 cannot achieve the condition T 2 > T 1 . Moreover, from (7.3) and (7.5) we observe that T 1 and T 2 cannot be equal. Therefore, with the circuit of Figure 7.9 T 1 will always be greater than T 2 and the duty cycle will always be greater than 0.5 of 50 percent. But we can achieve a duty cycle close to 50 percent if we choose resistor R A to be much smaller than resistor R B . Example 7.1 For the astable multivibrator of Figure 7.10 the capacitor has the value of 10 nF . Determine appropriate values for the resistors R A and R B so that the circuit will produce a pulse repetition frequency of 200 KHz with a duty cycle of 60%.
76
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
RA R RB
2 3 V CC
+
R
Comparator 1 R Q
T1 T 2
vC C
1 3 V CC
Output S Q
+
Comparator 2 R
Solution: The period T = T 1 + T 2 is a function of the capacitor and resistor values as shown in relation (7.6), that is,
T = T 1 + T 2 = 0.69 ( R A + 2R B )C
(7.7)
The duty cycle is T 1 T and from relations (7.3) and (7.6) we obtain
0.69 ( R A + R B )C RA + RB RA + RB T Duty cycle = 0.6 = ----1 = ------------------------------------------ = ---------------------- = -----------------------0.69 ( R A + 2R B )C R A + 2R B 14.43 K T R A + R B = 0.6 14.43 = 8.66 K
(7.8)
Subtraction of (7.8) from (7.7) yields Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
77
(7.9)
Figure 7.11 shows an alternate form of an astable multivibrator using the 555 Timer.
V CC
RA RB R
2 3 V CC
+ vC C
R
Comparator 1 R Q S Q
T1 T 2
1 3 V CC
Output
+
Comparator 2 R
With the circuit of Figure 7.11 we can obtain any desired value for the duty cycle as we will see from the relations below. For the charging interval T 1 the circuit is as shown in figure 7.12.
V CC RA
vC ( t )
C
78
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
= V CC ( 2 3 V CC )e
t RA C
(7.10)
= 13
t RA C
= 12
T 1 R A C = ln ( 1 2 ) = ln 1 ln 2 = 0 ln 2 T 1 = ln 2 R A C = 0.69R A C
(7.11)
vC ( t )
C RB
(7.12)
R A 2R B [ 1 3 R B ( R A + R B ) ]V CC = --------------------------------------------------------------------- = ---------------------- 2R A R B [ 2 3 R B ( R A + R B ) ]V CC e
T2 RB C
(7.13)
= 12
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
79
or
RA RB 2R A R B T 2 = ------------------- C ln ---------------------- R A 2R B RA + RB
(7.14)
However, there is one restriction as we can see from relation (7.13), that is, the term R B ( R A + R B ) must be less than 1 3 . Therefore, the period T is the summation of (7.11) and (7.14) from which we obtain the relation
2R A R B RA RB T = T 1 + T 2 = 0.69R A C + ------------------- C ln ---------------------- R A 2R B RA + RB 2R A R B RA RB T = 0.69R A + ------------------- C ln ---------------------- R A 2R B RA + RB
(7.15)
subject to the condition that R B ( R A + R B ) < 1 3 , and this condition can be simplified as shown below.
RB ------------------- < 1 3 RA + RB 3R B < R A + R B 2R B < R A RB < RA 2
(7.16)
Example 7.2 Using the astable multivibrator of Figure 7.11 find an appropriate ratio for resistors R A and R B so that the output will be a square waveform, i.e., a waveform with duty cycle at 50 % . Solution: To achieve a duty cycle at 50 % , we must make T 1 = T 2 and this condition will be satisfied if we equate relations (7.11) and (7.14) subject to the constraint of (7.16). Then,
2R A R B RA RB ln 2R A C = ------------------- C ln ---------------------- R A 2R B RA + RB
710
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(7.17)
This is a nonlinear equation and we could try to solve it with the MATLAB symbolic expression solve(eqn) as
syms x; solve('(x+1)*log(2)=log((2*x1)/(x2))')
but this returns the value of 1 which obviously is meaningless for this example. Therefore, let us plot relation (7.17) with the following MATLAB script to find the xaxis zero crossing.
x=2.1:0.01:2.5; y=(x+1)*log(2)log((2.*x1)./(x2)); plot(x,y);grid
Figure 7.14 indicates that xaxis zero crossing occurs at approximately x = R A R B = 2.363 and thus to achieve a duty cycle at 50 % , we should choose that the ratio R A R B should be close to this value.
0.4 0.2 0 Relation (7.17) -0.2 -0.4 -0.6 -0.8 -1 -1.2 2.1 2.15 2.2 2.25 2.3 2.35 x=RA/RB 2.4 2.45 2.5
Figure 7.14. Plot for finding the ratio RA R B to achieve 50 % duty cycle for the circuit of Figure 7.14
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
711
from which
An unstable multivibrator can also be constructed with an op amp as shown in Figure 7.15.
V ( + ) sat
v out
t C Rf
v() v(+)
R1 (a) R2
V ( ) sat
v out
v(+)
(b)
v()
(c)
t (d)
From the circuit of Figure 7.15(a), application of KCL at the noninverting input of the op amp yields
v ( + ) v out v ( + ) ------------------------ + -------- = 0 R2 R1 v out ( R 1 + R 2 )v ( + ) -------------------------------- = --------R1 R2 R2 R1 v ( +) --------- = ---------------------- = a v out ( R1 + R2 )
(7.18)
712
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
When the potential difference between the inverting and noninverting inputs of the op amp approaches the zero value the op amp comes out of saturation and subsequently the positive feedback from the output to the inverting input of the op amp drives the op amp to positive saturation. The voltage across the capacitor cannot change instantaneously and rises exponentially as shown in Figure 7.15(d) and when it reaches the value aV ( + ) sat the output switches back to the negative saturation state. For the astable multivibrator of the op amp circuit of Figure 7.15(a) we can derive the period of the pulse repetition frequency as follows: From relation (7.1)
v C ( t ) = V ( V V in )e
t Rf C
Taking the natural log of both sides and solving for the time t we obtain
( V V in ) t = R f C ln -----------------------------(V v (t))
(7.19)
Substitution of the voltages shown in Figures 7.15(a) and 7.15(b) yields the appropriate expressions for the timing periods as indicated below.
V ( + ) sat a V ( ) sat V ( + ) sat a V ( ) sat t 1 = R f C ln --------------------------------------- = R f C ln --------------------------------------V(+) aV(+) V(+) ( 1 a )
sat sat sat
(7.20) (7.21)
Relations (7.20) and (7.21) indicate that if the positive and negative values of the saturation voltage have the same amplitude, then t 1 = t 2 and the period of the pulse repetition becomes
T = t 1 + t 2 = 2R f C ln 1 + a ----------1a
(7.22) (7.23)
2R 1 T = t 1 + t 2 = 2R f C ln 1 + -------- 2R
2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
713
Rf
v() v(+)
R1 (a) R2
V ( + ) sat
v out
V ( ) sat
(b)
40 6 2 = 5 10 ln --------------- = 2.35 ms 25 6
714
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Figure 7.17. Block diagram and typical input and output waveforms for a monostable multivibrator
We observe that a positive going pulse at the input signal (transition from A to B) which is often referred to as the leading edge of the pulse (the transition from C to D is referred to as the trailing edge of the pulse) makes the 0 output to change state momentarily from state 0 to state 1 for a time period designated as T which is determined by an external resistor R X and an external capacitor C X as shown in Figure 7.18. Device SN74121 is a typical monostable multivibrator circuit and it is shown in Figure 7.18.
CX Input 5 B A1 4 10 9 RX 14 Q 7 6 V CC Output
74121
A2 3
As shown in Figure 7.18, the 74121 has three trigger inputs: A 1 , A 2 , and B . Depending on the desired circuit operation, any or all of these three pins may be connected to the input trigger signal. In Figure 7.5 A 1 and A 2 are both active low and the B input is active high. Under those conditions, the monostable multivibrator will produce the output pulse shown. As stated above, time period that the output will stay High is determined by an external resistor R X and an external capacitor C X . Typically, the capacitor value is greater than 1 F and the resistor value is greater than 10 K and the pulse width period T at the output is determined from the relation
T = 0.33 R X C X
For instance, if R X = 20 K and C X = 5 F , the pulse width will be T = 100 ms . With the 74121 device, the output pulse width can be varied from 40 ns to 28 s . Also the 74121 is retriggable meaning that the input may be a train of periodic pulses.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
715
Output
Figure 7.19. The use of a monostable multivibrator to block unwanted input pulses
A monostable multivibrator can also be constructed with an op amp as shown in Figure 7.20.
D1 C1 Rf
v(-) v(+)
C2 D2 R1 R3 R2
v out
v in
In the circuit of Figure 7.20, while the input pulse is zero, i.e., v in = 0 , the op amp remains in positive saturation, that is, v out = V ( + ) sat , and this is the stable state of the circuit, hence the name monostable. At this state, diode D 1 is forward biased and the voltage across the capacitor C 1 is the same as that of diode D 1 , about 0.7 V and thus the voltage across the inverting input of the op amp is v ( ) = 0.7 V . By the voltage division expression, the voltage at the noninverting input of the op amp is
R2 v ( + ) = ----------------- V ( + ) sat R1 + R2
(7.24)
Usually, resistors R 1 and R 2 are chosen such that R 2 R 1 = 10 and thus v ( + ) v ( ) . Accordingly, the output of the op amp is held at positive saturation. Next, let us assume that a negative going pulse whose amplitude exceeds the voltage v ( + ) . This voltage at the noninverting of the op amp will go momentarily negative and will drive the output of the op amp into negative saturation V ( )sat . So, with the output at negative saturation, the voltage v ( + ) at the noninverting input will also become negative. But this condition is only tem-
716
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(7.25)
R1 t = R f C 1 ln 1 + ----- R
2
(7.26)
(7.27)
As with astable multivibrators, we can form a monostable multivibrator using a 555 timer circuit as shown in Figure 7.21. As shown in Figure 7.21, in the stable state the SR flipflop will be in the Reset state keeping the transistor in saturation and thus the capacitor voltage will be close to zero volts. The voltage at the input terminal is sufficiently high to keep the output of Comparator 2 Low. The output Q will also be Low. Now, if a negative going pulse is applied at the input, the output of Comparator 2 will go High, the flipflop will be Set and its output Q will go High and output Q will go Low and the transistor will be at cutoff. At this time, the capacitor will begin charging through resistor R1 and will rise exponentially towards the supply voltage VCC . Once the capacitor voltage exceeds the threshold voltage of Comparator 1, the output of this comparator will go High, the flipflop will be Reset, the transistor will go in saturation, and the capacitor will discharge. Thus, the multivibrator will be returned to it stable state at zero volts level.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
717
R1
R2 Comparator 1
C R2
R Q S Q Comparator 2 RB
R2
We can find the pulse width T for the time that the monostable multivibrator changes momentarily from its stable state by observing that the capacitor voltage, when charging, reaches the threshold voltage at the minus () input of Comparator 1. Because the three resistors have the same value, this threshold voltage is 2 3 V CC . From (7.1),
v C ( t ) = V ( V V in )e
t R1 C
and assuming that the negative going pulse is applied at the input at t = 0 when V in = 0 , at
t = T the above relation becomes ( 2 3 )V CC = V CC V CC e e
T R1 C T R1 C
= 13
T R 1 C = ln ( 1 3 ) = ln 1 ln 3 T 1.1R 1 C
(7.28)
718
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
We will illustrate the operation of the circuit of Figure 7.22 with the following example. The triggering signal which is used to induce a transition from one state to another will be discussed later in this section. Example 7.4 Compute the stablestate currents and voltages for the flipflop circuit of Figure 7.23. Assume that the transistors have a minimum h FE value of 50.
* The logic diagrams and characteristic tables for the SetReset, Data, JK, and Toggle flipflops are described in Digital Circuit Analysis and Design with Simulink Modeling and Introduction to CPLDs and FPGAs, ISBN 9781934404058. This circuit is also known as EclessJordan configuration
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
719
Solution: This circuit consists of two crosscoupled inverter circuits such as that of Exercise 8 in Chapter 6. The analysis is facilitated by breaking the given circuit into two parts, the first part indicating the connections between the base of transistor T 1 and the collector of transistor T 2 as shown in Figure 7.24(a), and the second part indicating the connection between the collector of T 1 and the base of T 2 as shown in Figure 7.24(b).
+ 12 V
2.2 K I2 V B1 100 K 12 V (a) 15 K T2 ON I1 I C2 2.2 K V C2 0 V V C1 T1 OFF
+ 12 V
I3 15 K V B2 0 V 100 K (b) I B2 I4 ON T2
OFF T1
12 V
Figure 7.24. Circuits for the computation of the stable states for Example 7.4
Let us first assume that transistor T 1 is OFF and transistor T 2 is ON. Since the saturation voltages are small (about 0.2 V ), we will initially neglect them and we let V C2 ( sat ) 0 V and V B2 ( sat ) 0 V as shown in Figures 7.20(a) and 7.20(b) respectively. By the voltage division expression
15 K V B1 = ------------------------------------------- ( 12 V ) = 1.57 V 100 K + 15 K
and this voltage will certainly keep transistor T 1 at cutoff. To verify that with transistor T 1
720
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Also, Then,
We observe that the value of I 2 is much less than the value of I 1 and in a practical flipflop design this may not be the case. Usually, the minimum base current I B2 is read from the collector characteristics curve for a particular transistor and since no curves were given, we will use the given minimum value of h FE .for the minimum value of I B2 for saturation. Thus,
I C2 5.35 mA I B ( min ) = ------- = --------------------- = 0.11 mA 50 h FE
Let us now compute the value of I B2 from Figure 7.24(b). We find that
12 I 3 = ------------------------------------------ = 0.70 mA 2.2 K + 15 K 12 I 4 = ------------------- = 0.12 mA 100 K I B2 = I 3 I 4 = 0.70 0.12 = 0.58 mA
and Then,
Since ( I B2 = 0.58 mA ) > ( I B ( min ) = 0.11 mA ) , that is, the value of I B2 exceeds the minimum base current value of I B ( min ) required for saturation, we have shown that transistor T 2 is deeply into saturation. Finally, the collector voltage V C1 if found from Figure 7.20(b) as
V C1 = 12 2.2 10 I 3 = 12 2.2 10 0.70 10
3 3 3
= 10.5 V
and this value is close to the collector supply voltage of 12 volts. With the assumption that the saturation are zero volts, the voltages and currents for a stable state for the flipflop of this example are summarized in Table 7.1. The second stable state is the one in which transistor T 1 is ON and transistor T 2 is OFF. The analysis is the same as above where the voltages and currents are interchanged between transistors T 1 and T 2 , and V C2 = 10.5 V . Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
721
I C1 V C1 I C2 V C2 I B1 V B1 I B2 V B2
1.56 V
0.58 mA 0V
R1
RE
The procedure for calculating the stable states is similar to that for a fixedbias flipflop and it is illustrated in Example 7.5.
722
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Solution: This circuit consists of two crosscoupled inverter circuits such as that of Exercise 8 in Chapter 6. The analysis is facilitated by breaking the given circuit into two parts, the first part indicating the connections between the base of transistor T 1 and the collector of transistor T 2 as shown in Figure 7.27, and the second part indicating the connection between the collector of T 1 and the base of T 2 as shown in Figure 7.28. Let us first assume that transistor T 1 is OFF and transistor T 2 is ON. We must first compute the voltage V EN1 shown in Figure 7.27.
V CC 12 V
RC R1 OFF T1 V BE1 V EN1 V BN1 R2 10 K I B2 + I C2 30 K ON I B2 I C2 T2 RE 0.5 K N V CE2 V CN2 V EN2 4 K
Figure 7.27. The circuit of Example 7.5 showing the connections between the base of T 1 and collector of T 2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
723
(7.29)
Figure 7.28. The circuit of Example 7.5 showing the connections between the base of T 2 and collector of T 1
where the minus () sign indicates that the neutral point (ground) is at a higher potential than the emitter. To find the saturation currents I B2 and I C2 we will apply Thevenins theorem by first replacing the collector circuit of transistor T 2 by its equivalent, then by replacing the base circuit of T 2 by its equivalent, combining these two, and inserting the transistor T 2 between them. With transistor T 2 and emitter resistor R E disconnected, the Thevenin equivalent is as shown in Figure 7.29.
V CC
RC R1 OFF T1 R2 10 K 30 K x RC 4 K 12 V R2 R1 30 K x 10 K R TH1 3.64 K V TH1 10.9 V
12 V
4 K
x y
V CC
Figure 7.29. Collector circuit for the computation of Thevenin equivalent, Example 7.5
724
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(7.30) (7.31)
Next, we replace the base circuit of T 2 by its equivalent as shown in Figure 7.30.
V CC
RC 4 K R1 30 K T1 OFF RC R1 R TH2 12 V
R2 10 K
v w
N
4 K 30 K R2 V CC 10 12 V K
v w
v w
Figure 7.30. Base circuit for the computation of Thevenin equivalent, Example 7.5
(7.32) (7.33)
The Thevenin equivalent voltages and resistances of V TH1 , R TH1 , V TH2 , and R TH2 , combined with transistor T 2 and resistor R E yield the circuit of Figure 7.31 where have assumed that V EC ( sat ) = 0.2 V and V EB ( sat ) = 0.7 V . We find the values of I B2 and I C2 by application of KVL around the loops indicated in Figure 7.31.
0.5 ( I B2 + I C2 ) + 0.7 + 7.73I B2 = 2.73 0.5 ( I B2 + I C2 ) + 0.2 + 3.64I C2 = 10.9
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
725
Figure 7.31. The composite Thevenin equivalent circuit of Example 7.5 when transistor T 2 is in saturation
Using MATLAB with the matrix left division operator, we obtain write and execute the following script which produces the values of the currents I B2 and I C2 .
A=[8.23 0.5; 0.5 4.14]; B=[2.03 10.7]'; x=A\B;... fprintf(' \n'); fprintf('IB2=%2.3f mA \t',x(1)); fprintf('IC2=%2.3f mA \t',x(2)); fprintf(' \n')
(7.34) (7.35)
(7.36)
The voltages can now be found from the circuits of Figures 7.27 and 7.28. From Figure 7.26 we observe that V EN2 = V EN1 and from relation (7.18)
V EN1 = V EN2 = ( I B2 + I C2 ) R E = ( 0.09 + 2.57 ) 0.5 = 1.33 V
From Figure 7.27, From Figure 7.28, From Figure 7.27, and
V CN2 = V CE2 + V EN2 = 0.2 1.33 = 1.53 V V BN2 = V BE2 + V EN2 = 0.7 1.33 = 2.03 V R2 10 V BN1 = ----------------- V CN2 = ----------------- ( 1.53 V ) = 0.38 V 30 + 10 R1 + R2 V BE1 = V BN1 V EN1 = 0.38 ( 1.33 ) = +0.95 V
726
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
2.03 V
12 V
(a)
Therefore,
(7.42)
The stable state values of the circuit of Example 7.5 are summarized in Table 7.2
TABLE 7.2 Voltages and currents for a stable state for the flipflop of Example 7.4
Parameter Value 0 mA
10.83 V
2.57 mA
1.53 V
0 mA
1.56 V
0.09 mA
2.03 V 1.33 V
We observe that the output changes from V CN2 = 1.53 V to V CN1 = 10.83 V or vice versa.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
727
Chapter 7 Pulse Circuits and Waveform Generators 7.5.3 Triggering Signals for FlipFlops
A flipflop will remain in one of its stable states indefinitely until a triggering signal is applied to make a transition. In some applications it is desired to have a change of state occur immediately after the application of an abrupt triggering signal. The interval time during which conduction transfers from one transistor to the other is known as the transition time. It is always desirable to reduce the transition time and this can be accomplished by inserting small capacitors, referred to as speedup capacitors, in parallel with the coupling resistors R 1 as shown in Figure7.33. Let us assume that in the flipflop circuit of Figure 7.33 transistor T 1 if OFF, transistor T 2 is ON, and to initiate a transition, a negative pulse is applied at the base of T 2 denoted as point X 2 . The collector voltage of T 2 , denoted as point Y 2 will rise rapidly and it is desirable that this rapid rise will be transmitted with minimum delay to the base of T 1 denoted as point X 1 . Transistor T 1 has an input capacitance C i and in the absence of the parallel capacitor C 1 , the circuit configuration consisting of resistors R 1 and R 2 , and input capacitor C i constitutes an uncompensated attenuator.*
+ V CC
RC V C1 T1 Y1 C1 R1 X1 R2 R2 C1 R1 Y2 X2 T2 RC V C2
The smallest allowable interval between triggers is referred to as the resolving time of the flipflop, and its reciprocal is the maximum frequency at which the flipflop will respond. In practice, the maximum frequency f max of operation is given by the relation
R1 + R2 1 f max = ----- = --------------------2 2C 1 R 1 R 2
(7.43)
The triggering signal which is usually employed to initiate a transition from one state to the other is either a pulse of short time duration or a step voltage. This pulse or step may be used in a man-
728
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Trigger input
T1
V BB
RC Trigger input
V CC
C1 ON T2 R2
RS
C T 1
R1 OFF (b)
+ V BB
In symmetrical triggering, each successive triggering signal initiates a transition, regardless of the state in which the flipflop happens to be. The circuits of Figure 7.35 show how a flipflop may be triggered in a symmetrical manner.
+V CC RC D3 RC C T D1 T1 OFF (a) ON (b) D2 T2 R2 D3 R2 OFF T1 ON T2
C T
D1
D2
V BB
Figure 7.35. Methods of symmetrical triggering through diodes at the collectors or the bases of the transistors
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
729
Chapter 7 Pulse Circuits and Waveform Generators 7.5.4 Present Technology Bistable Multivibrators
The bistable multivibrator (flipflop) circuits weve discussed thus far are the original circuits and were in use in the 1960s. We have included them in this text because they are the circuits from which present technology bistable multivibrators such as those with CMOS technology and op amps have evolved. We will briefly discuss a bistable multivibrator with an op amp in this subsection. The circuit of Figure 7.36 shows how an op amp can be configured to behave as a bistable multivibrator.
v in
C R1 R3 (a)
v in v out v out
R2
(b)
The stable states for the bistable multivibrator of Figure 7.36(a) are the conditions where the output is at positive or negative saturation. It assumes either positive or negative saturation by the positive feedback formed by resistors R 2 and R 3 . A positive or negative going pulse as shown in Figure 7.36(b) causes the circuit to switch states.
v out
R1 V+ upper
v out
R2 V ref
vs
V+ lower
vs
(a)
(b)
(c)
Figure 7.37. Schmitt trigger circuit and waveforms for increasing and decreasing input signals
The Schmitt trigger circuit and transfer characteristics are similar to the comparator. It provides an voltage when its input signal v s reaches some predetermined value set at the noninverting
730
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
and the reference voltage V ref . These threshold voltages can be found by application of KCL at the noninverting input of the op amp. Thus,
V+ V ref V+ V out --------------------- + ----------------------- = 0 R2 R1 1 1 ----- + ----- V = V out + V ref ---------- -------- R1 R2 + R1 R2 R 1 V ref + R 2 V out R1 + R2 ----------------- V = ---------------------------------------+ R1 R2 R1 R2 R 1 V ref + R 2 V out V+ = ---------------------------------------R1 + R2
(7.44)
In (7.44), when V out is the maximum positive output voltage, V+ = V+ upper , and when V out is the maximum negative output voltage, V+ = V+ lower . Normally, the peaktopeak output voltage of the Schmitt trigger of Figure 7.37 is often limited by the use of backtoback zeners across the output terminal and ground. The zener voltages are chosen so that the output swing from positive to negative or vice versa, is compatible with commercially available IC digital devices. Example 7.6 For the Schmitt trigger circuit of Figure 7.38(a), the input signal v s is as shown in Figure 7.38(b). Find and sketch V+ upper and V+ lower . Solution: It is given that the Schmitt trigger saturates positively at +12 V . Then, with (7.44)
4 R 1 V ref + R 2 V out 10 2 + 250 12 V+ upper = ---------------------------------------- = -------------------------------------------- = 2.24 V 10250 R1 + R2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
731
vs ( V )
V out = 12 V 4 2 0 4 7
t ( ms )
Figure 7.38. Schmitt trigger circuit and input signal waveform for Example 7.6
12 V out ( V )
4 2 0
vS
2.24 V 1.66 V 2.24 V 4 7
t ( ms )
12
That is, when the input signal v S drops below this value, the output abruptly swings back to positive saturation, i.e., +12 V , and when the input signal v S rises again to 2.24 V , the output drops again to 12 V as shown in Figure 7.39.
732
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
733
+
R
Comparator 1
T1 T2
vC C
1 3 V CC
R Q S Q
Output
+
R
Comparator 2
v (V) 5 t ( s )
2. For the astable multivibrator below V CC = 5 V and C = 1 nF . Find the values for the resistors R A and R B so that the circuit will produce the waveform shown.
734
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
V CC RA RB
R+
2 3 V CC
Comparator 1
T1 T2
vC C
1 3 V CC
R Q S Q
Output
+
R
Comparator 2
v (V) 5 t ( s )
3. Repeat Exercise 2 to design a circuit that will produce the following waveform.
v (V) 5 t ( s )
4. Design a monostable multivibrator using a 555 timer, a capacitor with value C = 1 nF and appropriate resistor values to produce an output pulse of 20 s duration.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
735
15 K
15 K
12 V
The transistors are identical and the curves supplied by the manufacturer specify that V CE2 ( sat ) = 0.15 V and V BE2 ( sat ) = 0.7 V . Assume that the transistors have a minimum h FE value of 50. Using these voltages, recalculate the stablestate currents and voltages. 6. For the Schmitt trigger circuit of Figure (a) below, the input signal v S is as shown in Figure (b).
vs
V+ R1 40 K R2 10 K 0 5 (b)
2 3 2 2 t (r )
V out = 10 V
vs ( V )
5
V ref = 1 V (a)
736
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
= 0.69 ( R A + 2R B )C = 0.69 10 ( R A + 2R B )
6
The duty cycle is T 1 T and from relations (7.3) and (7.6) we obtain
0.69 ( R A + R B )C RA + RB RA + RB T1 Duty cycle = 0.75 = ----- = ------------------------------------------ = ---------------------- = --------------------0.69 ( R A + 2R B )C R A + 2R B 5.77 K T R A + R B = 0.75 5.77 = 4.33 K (2)
2.
v (V) 5 T1 0 1 T T2 2 3 4 5 t ( s )
The period is T = T 1 + T 2 = 1 + 3 = 4 s . The duty cycle is T 1 T = 1 4 = 0.25 = 25 % . To achieve a duty cycle at 25 % , we must make T 2 = 3T 1 and this condition will be satisfied if we multiply relation (7.11) by 3 and equate it to relation (7.14) subject to the constraint of (7.16). Then, Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
737
Therefore, let us plot the above relation with the following MATLAB script to find the xaxis zero crossing.
x=2.005:0.0001:2.006; y=3.*log(2).*(x+1)log((2.*x1)./(x2)); plot(x,y);grid
0.05
T2=3T1
-0.05
-0.1
-0.15
-0.2 2.005
2.0052
2.0058
2.006
The plot above indicates that the xaxis zero crossing occurs at approximately
x = R A R B = 2.0058
and thus to achieve a duty cycle at 25 % , we should choose that the ratio R A R B should be close to this value but not exactly equal or less than 2. The value of the capacitor can be
738
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
3.
v (V) T 5 T1 0 1 2 3
T2 4 t ( s )
Here, we must make T 1 = 3T 2 and this condition will be satisfied if we multiply relation (7.14) by 3 and equate it to relation (7.11) subject to the constraint of (7.16). Then,
2R A R B RA RB ln 2 R A C = 3 ------------------- C ln ---------------------- R A 2R B RA + RB RB 2R A R B ln 2 ------------------- ln ---------------------- = ------ R A 2R B RA + RB 3 2R A R B ln 2 R A + R B ------- ------------------- = ln ---------------------- R A 2R B RB 3 2R A R B 1 ln 2 R A ------- ------ + 1 = ln ----------------------------- RA RB 2 RB 3
Therefore, let us plot above relation with the following MATLAB script to find the xaxis zero crossing.
x=2.1:0.01:6; y=log(2).*(x+1)3.*log((2.*x1)./(x2)); plot(x,y);grid
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
739
2.5
3.5
4 4.5 x=RA/RB
5.5
The plot above indicates that xaxis zero crossing occurs at approximately
x = R A R B = 4.25
and thus to achieve a duty cycle at 75 % , we should choose that the ratio R A R B should be close to this value. The value of the capacitor can be found from either expression (7.11) or (7.14) but (7.11) is simpler. We require that T 1 = 3 s and assuming R A = 5.1 K , we obtain
T 1 = 3 10
6
Example 7.2 and Exercises 2 and 3 reveal that as the duty cycle decreases the ratio R A R B approaches the value of 2, and as the duty cycle increases, this ratio approaches the value of 11. 4. We will use the monostable multivibrator of Figure 7.21 which is repeated below for convenience. From relation (7.28),
T 1.1R 1 C
With a capacitor of value C = 1 nF to produce an output pulse of T = 20 s duration, resistor R 1 must have the value of
20 10 T R 1 = ----------- = ----------------------- = 18.2 K 9 1.1C 1.1 10
6
740
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
R1
R2 Comparator 1
C R2
R Q S Q Comparator 2 RB
R2
5.
+ 12 V
2.2 K V C1 T1 100 K 100 K T2 2.2 K V C2
15 K
15 K
12 V
As in Example 7.4, we break the given circuit into two parts, the first part indicating the connections between the base of transistor T 1 and the collector of transistor T 2 as shown in Figure (a), and the second part indicating the connection between the collector of T 1 and the base of T 2 as shown in Figure (b).
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
741
I1 V C1 T1
T1
V B1
100 K 12 V
I2
I C2 V C2 0.15 V T2
ON
I3
OFF
ON
V B2 0.7 V
100 K
I B2 I4
T2
(a)
(b)
12 V
Let us first assume that transistor T 1 is OFF and transistor T 2 is ON. By the voltage division expression and the superposition principle,
100 K 15 K V B1 = ------------------------------------------- ( 12 V ) + ------------------------------------------- 0.15 = 1.43 V 100 K + 15 K 100 K + 15 K
and this voltage will certainly keep transistor T 1 at cutoff. To verify that with transistor T 1 beyond cutoff transistor T 2 is in saturation, we calculate I C2 by first finding I 1 and I 2 as follows:
I 1 = 12 V 0.15 V = 5.39 mA --------------------------------2.2 K
Then, and
0.15 ( 12 ) I 2 = ------------------------------------------- = 0.11 mA 15 K + 100 K I C2 = I 1 I 2 = 5.39 0.11 = 5.28 mA I C2 5.28 mA I B ( min ) = ------- = --------------------- = 0.11 mA 50 h FE
Let us now compute the value of I B2 from Figure (b). We find that
12 0.7 I 3 = ------------------------------------------ = 0.66 mA 2.2 K + 15 K ) I 4 = 0.7 ( 12 - = 0.13 mA -------------------------100 K
and Then,
742
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Since ( I B2 = 0.53 mA ) > ( I B ( min ) = 0.11 mA ) , that is, the value of I B2 exceeds the minimum base current value of I B ( min ) required for saturation, we have shown that transistor T 2 is deeply into saturation. Finally, the collector voltage V C1 if found from Figure (b) as
V C1 = 12 2.2 10 I 3 = 12 2.2 10 0.66 10
3 3 3
= 10.5 V
and this value is close to the collector supply voltage of 12 volts. Therefore, with the values given the more accurate voltages and currents for a stable state for the flipflop of this example are summarized in the table below.
Parameter Value 0 mA 10.5 V 5.28 mA 0.15 V 0 mA
I C1 V C1 I C2 V C2 I B1 V B1 I B2 V B2
1.43 V
0.53 mA 0.7 V
A comparison of the values of Table 7.1 and those of the table above reveals that the error in assuming that a transistor in saturation behaves as an ideal short circuit is very small and this error can be neglected. The second stable state is the one in which transistor T 1 is ON and transistor T 2 is OFF. The analysis is the same as above where the voltages and currents are interchanged between transistors T 1 and T 2 , and V C2 = 10.5 V .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
743
vs
V+ R1
V out = 10 V
10 5
vS
R2
40 K 10 K
0 5 10
1.2 V 2.8 V
V ref = 1 V
(a)
(b)
That is, when the input signal v s exceeds this value, the output abruptly swings to negative saturation, i.e., 10 V as shown.
4 4 R 1 V ref + R 2 V out --------------------------------------------------------------------) V+ lower = ---------------------------------------- = 4 10 ( 1 ) + 10 ( 10 - = 2.8 V 4 R1 + R2 5 10
That is, when the input signal v S drops below this value, the output abruptly swings back to positive saturation, i.e., +10 V , and this cycle repeats.
744
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Chapter 8
Frequency Characteristics of SingleStage and Cascaded Amplifiers
his chapter presents certain basic concepts and procedures that are applicable to frequency dependent singlestage and cascaded amplifiers. The intent is to provide the basic principles of the frequency dependence that is an essential prerequisite for the effective design and utilization of most electronic circuits.
(8.1)
where V is the DC component of the composite signal, is the fundamental frequency or first harmonic, 2 , is the second harmonic, 3 is the third harmonic, and so on. Unfortunately, not all signals consist of harmonically related components. In the case of speech or music, for instance, the sinusoidal composition of the sound is continuously changing. In this case, the composite signal can be approximated by the superposition of a number of sinusoidal components expressed as
v ( t ) = V + V 1 cos ( 1 t + 1 ) + V 2 cos ( 2 t + 2 ) + V 3 cos ( 3 t + 3 ) +
(8.2)
where V is the DC component of the composite signal, and 1 , 2 , 3 , and so on, are not harmonically related. Figure 8.1, referred to as the frequency spectrum of the signal reveals information about the frequencies and amplitude, but it contains no information about the phase angles of the components.
* For a thorough discussion on Fourier series, the Fourier transform, and the Laplace transformation please refer to Signals and Systems with MATLAB Computing and Simulink Modeling, ISBN 9781934404119.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
81
nt
The frequency spectrum of Figure 8.1 represents the square waveform shown in Figure 8.2.*
T A
Figure 8.2. Waveform that produces the frequency spectrum in Figure 8.1
Example 8.1 The frequency components of the frequency spectrum of Figure 8.1 are as shown in relation (8.3) below.
4A 1 1 4A v ( t ) = ------ sin t + -- sin 3t + -- sin 5t + = ----- 5 3
n = odd
1 -- sin nt n
(8.3)
What percentage of the average power is absorbed by a. the fundamental frequency b. the fundamental and the third harmonic c. the fundamental, the third, and the fifth harmonic Solution: The instantaneous power is p = v 2 R and for simplicity let us assume that R = 1 . The average power P over a period T is
1 P = -T
T 0
1 2 2 v dt = -- v t T
= v
0
(8.4)
We recall that the RMS value of a sinusoid is the amplitude of the sinusoid divided by 2 . Thus,
* Please refer to Chapter 7, Signals and Systems with MATLAB Computing and Simulink Modeling, ISBN 9781934404119.
8 2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
8A = --------- 1 + 1 + ----- + -- 12 9 25
b. Percentage of power absorbed by the fundamental frequency and the third harmonic components:
8---- 1 + 1 100 = 90% -2 9
c. Percentage of power absorbed by the fundamental frequency, the third, and the fifth harmonic components:
8 1 ---- 1 + 1 + ----- 100 = 93% -2 9 25
This example indicates that most of the power is absorbed by the fundamental frequency and the first two nonzero harmonics. Of course, the power absorbed would be considerably higher if the signal would include a DC component. Figure 8.3 shows a typical curve of amplification versus frequency.
As shown in Figure 8.3, the amplification decreases at high frequencies as a result of the parasitic capacitances in the circuit, and decreases at low frequencies because of the presence of coupling capacitors. Also, for a good designed amplifier, the amplification near the lower and near the higher band of frequencies should not be less than 70 percent of the amplification in the middle of the band. It turns out that if an amplifier retards the phase of each sinusoidal component of the signal in direct proportion to the frequency of that component, then the waveform of the output signal is an exact copy of the input waveform, but it is delayed in time. This can be illustrated by expressing relation (8.2) with the introduction of a constant time delay T . Then,
v ( t T ) = V + V 1 cos ( 1 ( t T ) + 1 ) + V 2 cos ( 2 ( t T ) + 2 ) + V 3 cos ( 3 ( t T ) + 3 ) +
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
83
(8.5)
Relation (8.5) reveals that the introduction of a constant delay adds a phase lag k to each component, directly proportional to the frequency of that component. Reversing the above process we can see that if an amplifier introduces a phase lag that increases linearly with frequency, then = K , and the amplifier produces a time delay T = K , and if the amplification is the same for all frequencies, the signal is not distorted; it is merely delayed in time. In conclusion, the analysis and design of linear amplifiers is concerned with the characteristics of amplification and phase shift versus frequency. For the remaining of this chapter and Chapter 9, we will develop procedures for constructing these characteristics. Subsequently, an easy method to begin with the design of amplifiers at low, medium, and high frequencies is to first define the transfer function G ( s ) gain and the frequencies as the poles of the transfer function and draw the Bode plot using MATLAB.* The results of the following example will be useful in our analysis and development of equivalent circuits. Example 8.2 Prove that the circuits of Figure 8.4(a) and 8.4(b) are equivalent.
gm vg v in
R1 (a)
vg
v1
R3 R2
v2 v out
R4
v in
R1
vg
v1
R3 R2
v2 v out gm vg R4
gm vg
(b)
Figure 8.4. Equivalent circuits for Example 8.2 * Frequency response and Bode plots are discussed in detail in Circuit Analysis II with MATLAB Applications, ISBN 9780970951151.
8 4
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(8.6)
(8.7)
We observe that equations (8.6) and (8.7) are the same and thus the circuits of Figures 8.4(a) and 8.4(b) are equivalent.
rn v in I 1
IB R1 R3
RA
IB
C3 (b) C R2 I2
v in
RB
R3 (a)
C3
rn v in
I1 R1 IB
rn
R2
3 --------------(1 + )
v in I 1
I2
IB
R3 (c)
C3
IB
IB R1 ( 1 + )R 3 (d)
R2
IB
I2
The behavior of the amplifier of Figure 8.5(a) at low frequencies is affected by the action of capacitor C 3 . An incremental model for the amplifier that is valid at low and medium frequencies is Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
85
(8.8) (8.9)
and
R1 I b = ------------------------------- I 1 R 1 + r n + Z eq
(8.10)
and the half power ( 3 dB ) frequency is defined at G ( j ) = 1 2 which occurs when = 1 RC . Accordingly, in (8.10) let us define the frequency
* The reduction theorem is discussed in Appendix E. Please refer to Chapter 11, Signals and Systems with MATLAB Computing and Simulink Modeling, ISBN 9781934404119.
8 6
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(8.11)
Factoring R 1 + r n out of the denominator in (8.10) and multiplying both numerator and denominator by 1 + j 3 , we obtain
R1 I1 1 + j 3 I b = ---------------- --------------------------------------------R1 + rn ( 1 + )R 3 j 1 + ----------------------- + ----R1 + rn 3
(8.12)
(8.13)
In the mediumfrequency range the term 3 is very large, and the second factor on the right side in (8.13) approaches unity; hence the first factor is the current gain at medium frequencies. Relation (8.13) can be expressed in a more convenient form by defining two new terms as
R 1 A m = ---------------R1 + rn
(8.14) (8.15)
and
( 1 + )R 3 k 3 = ----------------------R1 + rn
Substitution of (8.14) and (8.15) in (8.13) and factoring 1 + k3 out the denominator yields
Am 1 + j 3 A C = ------------- -------------------------------------------1 + k 3 1 + j ( 1 + k 3 ) 3
(8.16)
Relation (8.16) can be used as a guide in the design of transistor amplifiers for lowfrequency responses once the quiescent operating point has been established. The bypass capacitor is then chosen to make the break in the amplitude characteristic at = ( 1 + k 3 ) 3 occur at a suitable point below the band of frequencies occupied by the signal. Changing the value of C 3 does not change the shape of the ramp in the amplitude characteristic; it only shifts the ramp parallel to the frequency axis. Example 8.3 For the transistor amplifier in Figure 8.5, it is known that R 1 = 3 K , r n = 12 K , R 3 = 1 K ,
= 80 , and C 3 = 2.5 F . Construct the logarithmic amplitude characteristic for this amplifier in the low and mediumfrequency range.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
87
and in decibels
or
and the break frequency for the factor ( 1 + k 3 ) 3 in the denominator of relation (8.16) is
( 1 + k 3 ) 3 = ( 1 + 5.4 ) 400 = 2560 rad/sec
or
( 1 + k 3 )f 3 = ( 1 + 5.4 ) 64 = 410 Hz
The amount by which the amplification at low frequencies is less than its value at medium frequencies is
20 log ( 1 + k 3 ) = 20 0.8 = 16 dB
Therefore, from DC to the break frequency at 64 Hz , the gain is 25 16 = 9 dB . The amplitude characteristic is shown in Figure 8.6 where the frequency is shown in logarithmic scale and the gain is in dB scale.
30 20 10 0 Gain ( dB )
( 1 + k 3 )f 3 f3
64 Hz 410 Hz
10
100
1000
f ( Hz )
8 8
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
10
10
Figure 8.7. Gain versus frequency plot for Example 8.3 with MATLAB
ib v be r' b
v' be = v r be
C1
+
C2
ic ro ie v ce
gm v
Figure 8.8. The hybrid model for the transistor at high frequencies
An incremental model for the transistor amplifier in Figure 8.9(a) that is valid at medium and high frequencies is shown in Fig. 8.9(b).
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
89
r' b r be
RA
v b'
Ce
I Cc I2
ro
v in
RB
gm v
(b)
R2
v out
R3 (a)
C3
v
R1 I1 (c)
I Cc Ce I2 R eq
v ro
R2
r be
( 1 + A )C c C eq Ce
I2
gm v
v out
I1
R1
r be
ro gm v
R2
v out
(d)
The transistor circuit of Figure 8.9(a) is represented by the hybrid model of Figure 8.9(b), where the resistance R 1 represents resistances R A , R B , and the resistance associated with the source I 1 . The behavior of this amplifier is affected at high frequencies by the parasitic capacitances C e and C c . At high frequencies, these capacitances tend to shortcircuit node b' to ground and this results in a significant reduction of the voltage v and the current g m v . In the circuit of Figure 8.9(b), resistance r' b is typically in the order of 50 to 100 ohms while resistance
R 1 is in the order of several kilohms and thus we can eliminate r' b by replacing the portion of the circuit on the left of r be with a Norton equivalent circuit and Figure 8.9(c) shows the circuit
(8.17)
and for the useful frequency range this current is much smaller than the current g m v . Therefore, with the assumption that I g m v , we find that
ro R2 v out = ---------------- ( g m v ) ro + R2
(8.18) (8.19)
and letting
gm ro R2 A = --------------------( ro + R2 )
810
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(8.20)
Typically, the gain A in (8.20) is in the range of 10 to 1000. Also, r o R 2 , and thus A g m R 2 . Therefore, we can express (8.17) as
I = jC c ( v + Av ) = jC c ( 1 + A )v jC c Av
(8.21)
This relation enable us to replace the current I by a capacitor of the value ( 1 + A )C c as shown in Figure 8.9(d), and as we know from Appendix C, this amplifying property of the transistor which increases the apparent value of the collector capacitance is known as the Miller effect. From the circuit of Figure 8.9(d),
ro I 2 = ---------------- g m v ro + R2 C eq = C e + ( 1 + A )C c r be R 1 R eq = -----------------r be + R 1 1 jC eq R eq v = ------------------------------------- I 1 R eq = --------------------------------- I 1 1 + jR eq C eq R eq + 1 jC eq
(8.25)
(8.26) (8.27)
Next, we let
ro A m = ---------------- g m R eq ro + R2
(8.28)
(8.29)
At medium frequencies is small, that is, 1 , and the second factor on the right side in Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
811
( log scale )
Figure 8.10. Typical logarithmic amplitude characteristics for the transistor amplifier at high frequencies
(8.30)
(8.31)
In Chapter 3 we defined the cutoff frequency (relation 3.88) as = 1 r be C e ; hence we can express (8.31) as
1 + r be R 1 1 = -----------------------------1 + AC c C e
(8.32)
or
1 + r be R 1 f 1 = f -----------------------------1 + AC c C e
(8.33)
Thus, relation (8.33) provides us the halfpower point for the current gain of a transistor at high frequencies. Obviously, the halfpower point frequency f 1 can be greater or smaller than the cutoff frequency f depending on the values of resistor R 1 and the gain A = g m R 2 . Example 8.4 The model for a transistor amplifier at high frequencies is shown in Figure 8.11.
812
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
r' b
b' r be
Ce
Cc
I2
ro gm v
R2
v out
For this circuit it is known that R 1 = 5 K , R 2 = 1 K , the quiescent collector current is I C = 5 mA , = 80 , r' b = 50 , C c = 3 pF , r o = 50 K , and the current gainbandwidth frequency is f T = 400 MHz . Determine: a. the Miller effect ( 1 + A )C c at high frequencies b. the medium frequency current amplification A m c. the halfpower frequency f 1 Solution: Since R 1 r' b and r o R 2 , the given model circuit can be simplified to that shown in Figure 8.12.
I2 R2
Cc R1 I1
r be
Ce
gm v
Figure 8.12. Simplified model of the transistor amplifier for Example 8.6
= 0.2
This indicates that large current amplification is provided at frequencies up to 5 MHz . From (3.93),
gm 0.2 C e = ----- = ----------------------------- = 79.6 pF 8 T 2 4 10
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
813
Thus,
( 1 + A )C c = 201 3 10
12
= 603 10
12
= 603 pF
and this indicates that the dominant effect at high frequencies is the Miller effect. b. From Figure 8.12,
3 r be R 1 ------------------------------R eq = ------------------ = 400 5 10- = 370 3 r be + R 1 5.4 10
and since r o R 2 ,
A m g m R eq = 0.2 370 = 74
This value is close to the value of of the transistor and thus it is near the maximum value that can be obtained by this transistor. c. The halfpower frequency f 1 is found from (8.33). Then,
1 + r be R 1 6 1 + 400 5000 f 1 = f ------------------------------ = 5 10 ------------------------------------------------------------------------------ = 0.635 MHz 12 12 1 + AC c C e 1 + 200 3 10 ( 80 10 )
814
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
0 10
0
10
10
10
10
10
10
10
10
f ( log scale )
Figure 8.13. Combined low and highfrequency amplitude versus frequency characteristics
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
815
R2 C4
RA
R2 C4
v in
RB R3 C3 RB R3 C3
R4
v out
(a)
v1
G eq 1
I1
I2
v2 CE
Cc
I3
v3
CE
Cc
g m v 1 G eq 2
(b)
gm v2
G eq 3
v1
I1 G eq 1 R1
I2
v2
I3
v3
CE
Cc
C1
g m v 1 G eq 2
(c)
CE
C2 g v m 2
G e q3
Figure 8.14. Circuit of a cascaded RCcoupled transistor amplifier and its high frequency models
In Figure 8.14(a), capacitor C 4 is referred to as the interstage coupling capacitor and its function is to block the DC component through the RC coupling network. Transistors can also be direct coupled under certain conditions. A highfrequency model for the cascaded transistor stages is shown in Figure 8.14(b) where we have assumed that the base spreading resistance r' be is negligible. The conductances G eq 1 , G eq 2 , and G eq 3 , are the equivalent conductances connected between the three nodes v 1 , v 2 , v 3 , and ground. Thus,
G eq 1 = ------ + ----- + -----G eq 2 = ------ + ----- + ------ + -----
1 RA
1 RB
1 r be
(8.34)
1 R2
1 RA
1 RB
1 r be
(8.35)
and
816
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(8.36)
The model for the second stage of the amplifier has the same form as the model for a singlestage amplifier shown in Figure 8.9(c); hence the effect of the collector capacitance in this stage can be accounted for by a Miller capacitance as shown in Figure 8.9(d). When this is done, the second stage has the simplified representation shown in Figure 8.14(c). The Miller capacitance has the value (8.37) C 2 = ( 1 + A 2 )C C where A2 = v3 v2 (8.38) Similarly, A1 = v2 v1 (8.39) The model for the first stage of the amplifier differs from that of the second stage in that its load has a capacitive component as well as a resistive component; hence the Miller effect is slightly different in the first stage. The first stage is as shown in Figure 8.14(c) where C 1 is the shunt capacitance that is present in the load of the first stage transistor, and the resistor R 1 is negligible in comparison to the reactance of capacitor C 1 . Typically, R 1 is one kilohm or less. The value of R 1 depends on the midband value of the gain A 1 as defined in (8.39), the collector capacitance C c , and the halfpower frequency for the output circuit of the first stage defined as
G eq 2 2 = ------------------------------CE + C2 + CC
(8.40)
In terms of relations (8.39) and (8.40), the values of R 1 and C 1 can be found from
CE + C2 + CC 1 R 1 = ------------------- = ------------------------------gm CC 2 A1 CC
(8.41) (8.42)
and
C1 = A1 CC
(8.43)
If the resistance R 1 in the circuit of Figure 8.24(c) were negligible in comparison with the reactance of C 1 , then the simplified representation for the first stage would reduce to the same form as that for the second stage, and the process could be repeated for each stage in a cascade of any number of stages. However, relation (8.43) shows that at = 2 , R 1 = 1 2 C 1 where 2 is the halfpower frequency for the output circuit of the first stage. Thus, it is convenient to neglect R 1 at frequencies up to 2 . If R 1 is not neglected, and if the amplifier of Figure 8.24(c) is preceded by Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
817
(8.44) (8.45)
I2 = gm v1 I 1 = [ G e q1 + j ( C E + C c + C 1 ) ]v 1
(8.49)
Example 8.5 For the cascaded amplifier circuit of Figure 8.15, the transistors are identical with R A = 15 K , R B = 3.3 K , R 2 = 2.6 K , R 3 = 0.68 K , and R 4 = 0.5 K . The quiescent collector current in each stage is I C = 4 mA , and at this operating point = 100 , r' b = 50 , C C = 4 pF , f T = 100 MHz , and the output resistance r o is very large and it can be neglected. Compute the overall current gain A c at medium and high frequencies. Sketch and dimension the asymptotes for the highfrequency amplitude characteristics. Solution: From Chapter 3, relation (3.78), From relation (3.83),
g m = 40I C = 40 4 10
3
= 0.16
818
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
C4
RA
v out
v in
Figure 8.15. Cascaded amplifier for Example 8.5 gm 0.16 C E = ------ = -------------------- = 255 pF 8 T 2 10
From (8.27)
ro A m = ---------------- g m R eq ro + R2
and since r o R 2 , the current gain of the second stage amplifier at medium frequencies is
gm 0.16 A m2 = g m R eq = ---------- = ----------------------- = 67 3 G e q3 2.4 10
From (8.35),
1 1 1- 1- 11 1 1 1 G eq 2 = ------ + ----- + ------ + ----- = ---------------- + ------------------ + ------------ + ------------------ = 2.4 m R A R B r be R 2 15 K 3.3 K 50 2.6 K
From (8.46),
gm I3 1 A c 2 = --- = ---------- -----------------------------------------------------I2 G eq 2 1 + j ( C E + C 2 ) G eq 2 1 1 0.16 = ----------------------- ------------------------------------------------------------------------------------------3 = 67 -------------------------------------- 12 7 3 1 + j2.2 10 2.4 10 1 + j ( 255 + 272 ) 10 2.4 10
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
819
From (8.29)
ro A m = ---------------- g m R eq ro + R2
and since r o R 2 , the current gain of the first stage amplifier at medium frequencies is
gm 0.16 A m 1 = g m R eq = ---------- = ----------------------- = 67 3 G eq 2 2.4 10
gm I2 1 A c 1 = --- = ---------- ------------------------------------------------------------------1 + j ( C E + C c + C 1 ) G eq 1 I1 G eq 1 1 1 0.16 = ------------------ ----------------------------------------------------------------------------------------------- = 80 ----------------------------------------- 12 3 7 3 1 + j2.66 10 2 10 1 + j ( 255 + 4 + 272 ) 10 2 10
820
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
80
40 f = 720 KHz f = 600 KHz 0 10
0
20 dB decade 40 dB decade
10
10
f ( log scale )
Figure 8.16. Asymptotes for the highfrequency characteristics of the amplifier for Example 8.5
Figure 8.17 shows a cascaded transistor amplifier circuit and its model that is valid at low and medium frequencies. With r o assumed to be very large, we used the same procedure as in the simplification of Figure 8.5(a) to derive the simple equivalent circuit of Figure 8.5(d). The characteristics of the amplifier are affected at low frequencies by the coupling and bypass capacitors.
V CC R2 C4 R2 C4 RB R4
RA
RA
v out
v in
RB R3 C3 R3 C3
(a) I1 C4 Ib1 R1 I2 C3 R2 C4 Ib 2
3 ---------------(1 + )
I3 R2
C4 I4 R4
v in
rn
R3
rn
R1 ( 1 + ) R3 (b)
v out
Ib 1
Ib 2
Figure 8.17. RCcoupled amplifier circuit and its model at low frequencies
The effect of the coupling capacitor at the output of the amplifier is determined from right side part of Figure 8.17(b) with the use of the current division expression as follows:
2 I 4 = ------------------------------------------ I 3 R 2 + R 4 + 1 j C 4
or Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
821
(8.50)
1 4 = ---------------------------C4 ( R2 + R4 )
R2 j 4 I4 --- = ----------------- ------------------------R 2 + R 4 1 + j 4 I3
(8.51) (8.52)
To simplify the procedure of deriving an expression for the gain I 3 I 2 we make use of the fact that the value required for the bypass capacitor C 3 is much larger than the value required for the coupling capacitor C 4 . Thus, we choose the capacitors so that the lowfrequency halfpower point is determined by C 3 , and with this choice C 4 acts as a short circuit while C 3 is making the transition from a short circuit to an open circuit, and with this transition, the interstage network has the same form as the amplifier shown in Figure 8.5(d), and the current gain for each stage has the same form as (8.16). Thus, with C 4 acting as a short circuit we find that
Am 1 + j 3 A C = ------------- -------------------------------------------- * 1 + k 3 1 + j ( 1 + k3 ) 3
where and
where, with C 4 acting as a short circuit, R' 1 is the parallel combination of R 1 and R 2 in Figure 8.17(b). Thus,
R1 R2 R' 1 = ----------------R1 + R2
(8.57)
At very low frequencies C 3 acts as an open circuit and the interstage coupling capacitor C 4 contributes the factor
j 5 ------------------------1 + j 5
(8.58)
* The minus () sign stems from the fact that the current I 3 in Figure 8.17(b) has a direction opposite to the current I 2 in Figure 8.5(d).
822
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(8.59)
and R eq is the equivalent resistance when C 4 acts as a short circuit and C 3 as an open circuit. Thus, R eq = R 1 || R 2 + r n + ( 1 + )R 3 (8.60) Usually, ( 1 + )R 3 is much larger than the other resistors in that group, and thus
1 5 ----------------------------( R 1 + R 2 )C 4
(8.61)
and with the factor contributed by the interstage coupling capacitor C 4 , the complete expression for the current gain is
Am 1 + j 3 j 5 A C = ------------- -------------------------------------------- ------------------------1 + k 3 1 + j ( 1 + k 3 ) 3 1 + j 5
(8.62)
Normally, the design process begins with the selection of the proper resistor values to establish a suitable quiescent point. Then, C 3 is chosen to make the break at ( 1 + k 3 ) 3 occur at a desired point below the band of the high frequencies of the signal. Finally, C 4 is chosen to make the break at 5 occur at some frequency below 3 where C 3 acts as an open circuit. Example 8.6 A cascaded transistor amplifier circuit and its model that is valid at low and medium frequencies are shown in Figure 8.18. The transistors are identical with R A = 56 K , R B = 12 K , R 2 = 6.8 K , R 3 = 2.2 K , and R 4 = 1 K . The quiescent collector current in each stage is I C = 1 mA , and the transistor parameters are = 200 , r' b = 200 , C C = 20 pF , f T = 8 MHz , and the output resistance r o is very large in comparison with R 2 and thus it can be neglected. The coupling and bypass capacitors are to be chosen to yield suitable performance at low frequencies. a. Find the values of the resistances in the simplified lowfrequency model in Figure 8.28(b). b. The break frequency associated with the coupling capacitor C 4 at the output of the amplifier is to occur at 16 Hz . Compute the value of C 4 required in the output circuit. c. The highest break frequency associated with the interstage coupling network is to be ( 1 + k 3 )f 3 = 160 Hz . Find 3 , and, treating C 4 in the interstage network as a short circuit, find the required value for the bypass capacitor C 3 .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
823
R2
C4
RA
R2
C4 R4
v out
v in
RB R3 C3 (a) I1 C4 Ib 1 I2 R2 C4 Ib 2 RB R3
C3
3 ---------------(1 + )
I3 R2
C4 I4 R4
v in
rn
R1 R3 C3
rn
R1 ( 1 + ) R3 (b)
v out
I b1
I b2
Figure 8.18. Cascaded amplifier circuit and its lowfrequency equivalent for Example 8.6.
d. The break frequency associated with the coupling capacitor C 4 in the interstage network, denoted as 5 in relation (8.87), is to be at least as small as 3 . Treating C 3 as an open circuit, find the minimum value for C 4 in the interstage network. Solution: a. The values of R 2 = 6.8 K , R 3 = 2.2 K , and R 4 = 1 K are given. Also,
RA RB 56 12 R 1 = ------------------- = ----------------- = 9.9 K 56 + 12 RA + RA ( 1 + )R 3 = ( 1 + 200 )2.2 = 442 K
From Chapter 3, relation (3.78), From relation (3.83), and from (3.79), b. From relation (8.51),
C 4 = ---------------------------- = ------------------------------------------------------------ = 1.28F 3 ( R 2 + R 4 ) 4 ( 6.8 + 1 ) 10 2 16
g m = 40I C = 40 1 10
= 40 m
824
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
or
3 = 2f 3 = 2 3.33 = 21 rad/s
The computations for all examples and the endofchapter exercises, can be greatly simplified by writing a MATLAB script such as that shown below and plot the overall current gain.
w=1:10:10^5; RA=56000; RB=12000; R2=6800; R3=2200; R4=1000;... beta=200; rbprime=200; Ic=10^(3); f4=16; w4=2*pi*f4;... R1=RA*RB/(RA+RB); R1prime=R1*R2/(R1+R2); gm=40*Ic; rbe=beta/gm;... rn=rbprime+rbe; k3=(1+beta)*R3/(R1prime+rn); f3=160/(1+k3);... w3=2*pi*f3; C3=1/(w3*R3); w5=w3; C4b=1/((R2+R4)*w4);... C4d=1/((R1+R2)*w5); Am=beta*R1prime/(rn+R1prime);... Acnum=(Am.*(1+j.*w./w3).*(j.*w./w5));... Acden=(1+k3).*(1+j.*w./((1+k3).*w3)).*(1+j.*w./w5);... Ac=Acnum./Acden; AcdB=20.*log10(abs(Ac));... fprintf(' \n'); fprintf('RA = %2.2f K \t',RA/1000); fprintf(' \n');... fprintf('RB = %2.2f K \t',RB/1000); fprintf(' \n');... fprintf('R2 = %2.2f K \t',R2/1000); fprintf(' \n');... fprintf('R3 = %2.2f K \t',R3/1000); fprintf(' \n'); fprintf('R4 = %2.2f K \t',R4/1000); fprintf(' \n');... fprintf('R1 = %2.2f K \t',R1/1000); fprintf(' \n');... fprintf('R1prime = %2.2f K \t',R1prime/1000); fprintf(' \n');... fprintf('rbprime = %2.2f K \t',rbprime/1000); fprintf(' \n');... fprintf('rbe = %2.2f K \t',rbe/1000); fprintf(' \n');... fprintf('rn = %2.2f K \t',rn/1000); fprintf(' \n');... fprintf('Ic = %2.2f mA \t',Ic*1000); fprintf(' \n');...
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
825
RA = 56.00 K RB = 12.00 K R2 = 6.80 K R3 = 2.20 K R4 = 1.00 K R1 = 9.88 K R1prime = 4.03 K rbprime = 0.20 K rbe = 5.00 K rn = 5.20 K Ic = 1.00 mA beta = 200 f4 = 16.00 Hz w4 = 100.53 rps C4b = 1.28 microF gm = 40.00 mmho k3 = 47.92 f3 = 3.27 Hz w3 = 20.55 rps C3 = 22.12 microF w5 = 20.55 rps C4d = 2.92 microF Am = 87.30 The plot for the current gain (in dB scale) versus frequency (in log scale) is shown in Figure 8.19.
826
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
10
10
10
Figure 8.19. Amplitude of current gain versus frequency for Example 8.8
(8.63)
where A m L depends on the circuit parameters, and L is the lowfrequency halfpower point. For instance, for the transistor amplifier circuit of Figure 8.5 at low frequencies, L = 3 is as defined in relation (8.11), and A m L is as defined in relation (8.14).
A ( dB ) 40 20 0 20 dB/decade
20 dB/decade
90 45 0
fL
1 2
fH
f ( Hz )
fL
1 10
1
10 10 10 10
2 3
10 fH
10
1 10 10 10 3 10 4 10 5 10 6 10 7
45 90
f ( Hz )
(a)
(b)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
827
(8.64)
where A mH depends on the circuit parameters, and H is the highfrequency halfpower point. For instance, for the transistor amplifier circuit of Figure 8.9 at high frequencies, H = 1 is as defined in relation (8.28), and A mH is as defined in relation (8.27). We can obtain the combined low and highfrequency characteristics for the voltage gain A v by multiplying relation (8.89) by (8.90); thus,
j L A v = A m L A m H -------------------------------------------------------------( 1 + j L ) ( 1 + j H )
(8.65)
If two identical stages such as that shown in Figure 8.18, are connected in cascade, the overall voltage gain A v is obtained from the relation
( j L ) 2 A v = A m -----------------------------------------------------------------2 2 ( 1 + j L ) ( 1 + j H )
2
(8.66)
The overall amplitude characteristics for two identical stages in cascade is shown in Figure 8.21.
A ( dB ) 40 20 0 f ( Hz ) 40 dB/decade
40 dB/decade
fL
fH
Figure 8.21. Typical overall amplitude characteristics for two identical stages in cascade
By comparing the amplitude characteristics of Figures 8.20 and 8.21, we observe that the bandwidth between the halfpower points decreases as more stages are connected in cascade. It can be shown* that the halfpower bandwidth of the cascade of n identical stages is given by
Bn = B1 2
1n
(8.67)
where B 1 = H and H is the halfpower frequency for on stage amplifier at medium and high frequencies. The quantity 2 1 n 1 is referred to as the bandwidth reduction factor for n identical stages in cascade. For two stages the bandwidth reduction factor is 0.64 , for three stages is 0.51 , and for four stages it is 0.43 .
* The proof is left as an exercise for the reader at the end of this chapter.
828
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(8.68)
and the overall amplitude and phase characteristics are obtained by addition of the constituent characteristics. The asymptotes for a typical amplitude characteristic are shown in Figure 8.22 where the notations 20 and 40 stand for 20 dB/decade and 40 dB/decade respectively.
A ( dB ) 20 0 40
20 40
f L1 f L2
f H2 f H1
f ( Hz )
Figure 8.22. Typical overall amplitude characteristics for two nonidentical stages in cascade
(8.69) (8.70)
Then,
j A v ( j ) = K ---------------------------------------------------( j + j L ) ( j + j H )
The roots of the denominator of an expression for A v ( j ) may be purely imaginary such as
j L = 100 or complex conjugates such as j H = 3 + j 2 . However, we are interested on the
values that lie on the j axis of the complex frequency plane. Therefore, it is convenient to substitute the symbol s for j and after the roots of a quadratic equation in s are found, we can let s = j . Replacing j with s and defining s L = L , and s H = H , relation (8.70) is written as
s A v ( s ) = K -----------------------------------( s s L ) ( s sH )
(8.71)
and we can obtain the steadystate voltage gain for any frequency by letting s = j . For all practical electric and electronic circuits, voltage and current gains can be expressed as the ratio of two polynomials in the variable s , and when the highest power of the denominator is greater than the highest degree of the numerator, the expression is referred to as a rational function. The values of s that make a rational function zero are called zeros of the function; hence s = 0 is a zero of the function of (8.71). The values of s that make rational functions infinite are called poles of the function; hence s L and s H are poles of the function of (8.71). As indicated by the form of (8.71), if all the poles and zeros of a rational function are known, the function is completely specified except for a constant multiplier. For example, if the zeros of a certain voltage gain are 1 and 3 , and if the poles are 2 and 4 , then the voltage gain is Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
829
(8.72)
(8.73)
Thus if the poles and zeros of a rational function are known, the logarithmic amplitude and phase characteristics can be constructed, except for the effect of the constant multiplier. If the constant multiplier is not unity, it adds a constant number of decibels to the amplitude characteristic; if it is not positive, it adds an angle of 180 to the phase shift. It is also interesting to note that the poles and zeros of A v ( j ) correspond to the break frequencies expressed in radians per second. A zero corresponds to a break of 20 dB/decade upward in the asymptotic characteristic, and a pole corresponds to a break of 20 dB/decade downward. The relation between the poles and zeros of the voltage or current gain and the frequency characteristics is displayed in a useful way by the polezero diagram. For instance, Figure 8.23 shows the polezero diagram for the voltage gain of a singlestage amplifier obtained from relation (8.71).
Imaginary axis
s = j
Real axis
sH
sL
As shown in Figure 8.23, the polezero diagram is a set of rectangular coordinates known as the complex plane. The zero of A v ( s ) which occurs when s = 0 , is denoted by a circle at the origin of the complex plane. The poles of A v ( s ) which occur when s = s L = L and s = s H = H are represented by crosses on the negative real axis. For steadystate sinusoidal operation, the variable s = j corresponds to some point on the imaginary axis shown as a filled circle. The design of networks to provide a desired frequency characteristic begins with the choice of a polezero pattern that yields an approximation to the desired performance. The design is then completed by finding a network configuration and the parameter values that yield the desired polzero pattern. Example 8.7 The polezero pattern for the voltage gain of a certain amplifier is shown in Figure 8.24. The amplification at high frequencies is 40 dB . Construct the asymptotes for the logarithmic amplitude characteristic.
830
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
812
200
128 30
Real axis
Figure 8.24. Polezero pattern for Example 8.9 ( j + 30 ) ( j + 200 ) A v ( j ) = K ---------------------------------------------------( j + 128 ) ( j + 812 )
For very large values of the frequency , the above relation reduces to
A v ( j )
= K
is shown in Figure 8.25 where the radian frequencies 30, 128, 200, and 812 are shown in Hz as 4.8, 20, 32, and 129 respectively.
A ( dB ) 40
15.2
0 f ( Hz )
4.8
20 32
129
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
831
Frequency Characteristics of SingleStage and Cascaded Amplifiers 8.7 Amplification and Power Gain in Three or More Cascaded Amplifiers
Transistor amplifiers are conveniently characterized as current amplifiers. A typical transistor radio many employ a cascade of five amplifier stages. The first four are designed to yield the greatest possible current amplification, and the fifth stage is a power amplifier designed to produce the greatest possible power output. Let us consider the equivalent circuit of a voltage amplifier shown in Figure 8.26.
i1 v1
R1 R2
i2 v 1 v2
RL
(8.74)
(8.75)
(8.76)
(8.77)
In communications circuits it is desirable to make RL = R1 , and in this case the second term on the right side of (8.77) is zero. Denoting the voltage gain as A v we obtain
v2 A v = 20 log ---v1
(8.78)
Example 8.8 For the threestage amplifier shown in Figure 8.27, find:
832
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
100v 1
1 M
50v 2
50 K
2v 3
10
a. the voltage amplification and power gain of each stage in dB b. the overall voltage amplification and overall power gain of each stage in dB Solution: a. First stage:
v2 100v 1 A v 1 = 20 log ---- = 20 log ------------- = 40 dB v1 v1 G p 1 = A v 1 + 10 log 1 M = A v 1 + 0 = 40 dB ------------1 M
Second stage:
Third stage:
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
833
where V is the DC component of the composite signal, is the fundamental frequency or first harmonic, 2 , is the second harmonic, 3 is the third harmonic, and so on.
Signals that are not can be approximated by the superposition of a number of sinusoidal com-
ponents expressed as
where V is the DC component of the composite signal, and 1 , 2 , 3 , and so on, are not harmonically related.
The frequency spectrum of a signal reveals information about the frequencies and amplitude,
Amplification decreases at high frequencies as a result of the parasitic capacitances in the cir-
cuit, and decreases at low frequencies because of the presence of coupling capacitors.
For a good designed amplifier, the amplification near the lower and near the higher band of
frequencies should not be less than 70 percent of the amplification in the middle of the band.
If an amplifier retards the phase of each sinusoidal component of the signal in direct propor-
tion to the frequency of that component, then the waveform of the output signal is an exact copy of the input waveform, but it is delayed in time.
The behavior of transistor amplifier at low frequencies is affected by the action of the bypass
Transistor amplifier circuits designed for low frequencies are discussed in Section 8.2. Transistor amplifier circuits designed for high frequencies are discussed in Section 8.3. The amplifying property of the transistor which increases the apparent value of the collector
The frequency dependence of various amplifier circuits is examined by treating the lowfrequency characteristics separately from the highfrequency characteristics. However, the combined low and highfrequency characteristics can be displayed on a single set of coordinates. To achieve higher gains, we use multistage amplifier circuits. The overall characteristics of
multistage amplifiers can be easily approximated and displayed with the use of asymptotes.
834
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Summary
The bandwidth between the halfpower points decreases as more stages are connected in cascade. The halfpower bandwidth of the cascade of n identical stages is given by Bn = B1 2
1n
where B 1 = H and H is the halfpower frequency for on stage amplifier at medium and high frequencies. The quantity 2 1 n 1 is referred to as the bandwidth reduction factor for n identical stages in cascade. For two stages the bandwidth reduction factor is 0.64 , for three stages is 0.51 , and for four stages it is 0.43 .
If two nonidentical stages are connected in cascade, the overall voltage gain is given by ( j L1 ) ( j L2 ) A v = A m 1 A m 2 ------------------------------------------------------------------------------------------------------------------------------------------( 1 + j L1 ) ( 1 + ( j L2 ) ) ( 1 + j H1 ) ( 1 + j H2 ) For all practical electric and electronic circuits, voltage and current gains can be expressed as the ratio of two polynomials in the variable s , and when the highest power of the denominator
is greater than the highest degree of the numerator, the expression is referred to as a rational function. The values of s that make a rational function zero are called zeros of the function; the values of s that make rational functions infinite are called poles of the function. If all the poles and zeros of a rational function are known, the function is completely specified except for a constant multiplier. For example, if the zeros of a certain voltage gain are 1 and 3 , and if the poles are 2 and 4 , then the voltage gain is expressed as
(s + 1)(s + 3) A v ( s ) = K -------------------------------(s + 2)(s + 4)
amplification and power gain. The last stage is used to produce the greatest possible power output, and the previous stages are used for voltage amplification.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
835
RA
R2
v out v in
RB R3 C3
2. A transistor is used as an amplifier for video signals. The highfrequency model has the form shown below.
I2 R1 I1 Cc
r be
Ce
ro
R2
gm v
The quiescent collector current is I C = mA and the collector load resistor is R 2 = 1 K . The transistor parameters are = 50 , r' b = 50 , C c = 3 pF , r o = 30 K , and the current gainbandwidth frequency is f T = 200 MHz . The amplifier circuit is to be designed so that the highfrequency halfpower point for the current gain is 3 MHz . a. Determine the values of g m , r be , C e , and the Miller effect b. Determine the value of R 1 c. With the value of R 1 found in part (b) what is the midband current amplification? 3. For the cascaded amplifier circuit below, the transistors are identical with R A = 56 K , R B = 12 K , R 2 = 6.8 K , R 3 = 2.2 K , and R 4 = 1 K . The quiescent collector current in each stage is I C = 1 mA , and the transistor parameters are = 200 , r' b = 200 ,
836
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Exercises
C C = 20 pF , f T = 8 MHz , and the output resistance r o is very large in comparison with R 2
v in
RB R3 C3 R3 C3
a. Find the parameter values of the simplified highfrequency model circuit shown below.
v1
I1 R1 G e q1 I2
v2 CE
C2
I3
v3
CE
Cc C1
gm v1
G e q2
gm v2
G e q3
b. Neglecting the effect of the resistance R 1 in the model circuit above, find the overall current gain A c = I 3 I 1 . c. Sketch and dimension the asymptotes for the highfrequency amplitude characteristics. 4. Prove that the halfpower bandwidth of the cascade of n identical stages is given by
Bn = B1 2
1n
where B 1 = H and H is the halfpower frequency for one stage amplifier at medium and high frequencies. Hint: Begin with the assumption that the amplitude of the voltage gain of a onestage amplifier is given by
Av = A = Am ( 1 + ( H ) )
2
5. The polezero patterns for the voltage gain of two transistor amplifiers are shown below.
Imaginary axis Imaginary axis
10 1 (a)
Real axis
20 1 (b)
2 1
Real axis
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
837
v1
v 1
v2
v 2
(a)
v3
v 3
RL
50 K
10
v1
(b)
50v 1
v1
(c)
1v 1
838
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
RA
v out v in
RB R3 C3
I2 R2
IB
where
( 1 + )R 3 ( 1 + 80 ) 1 k 3 = ----------------------- = ---------------------------- = 7.5 10.3 + 0.5 R1 + rn 1000 f 3 = -------------------- 118 Hz ( 1 + 7.5 ) 1 3 = ----------R3 C3 1 f 3 = -----------------2R 3 C 3 1 1 C 3 = ---------------- = ----------------------------------- = 1.35 F 3 2R 3 f 3 2 10 118
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
839
and in decibels
The amount by which the amplification at low frequencies is less than its value at medium frequencies is
20 log ( 1 + k 3 ) = 20 0.93 = 18.6 dB
Therefore, from DC to the break frequency at 118 Hz , the gain is 37.6 18.6 = 19 dB . The amplitude characteristics are shown below where the frequency is shown in logarithmic scale and the gain is in dB scale.
40 30 20 10 0 1 118 Hz Gain ( dB )
( 1 + k 3 )f 3
f3
1 KHz
10
100
1000
f ( Hz )
2.
I2 R1 I1 Cc
r be
R eq
Ce
ro
R2
gm v
Figure 8.28. Simplified model of the transistor amplifier for Example 8.6
= 0.04
840
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
This indicates that large current amplification is provided at frequencies up to 4 MHz . From (3.93),
gm 0.04 C e = ------ = ----------------------------- = 31.8 pF 8 T 2 2 10
Thus,
( 1 + A )C c = 41 3 10
12
= 123 10
12
= 123 pF
and this indicates that the dominant effect at high frequencies is the Miller effect. b. From (8.19),
6 gm ro R2 0.04 30 10 A = --------------------- = ------------------------------------ = 38.7 3 ( ro + R2 ) 31 10
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
841
v in
RB R3 C3 R3 C3
a. The parameter values of the simplified highfrequency model circuit shown below are as follows:
v1
I1 R1 G e q1 I2
v2 CE
C2
I3
v3
CE
Cc C1
gm v1
G e q2
gm v2
G e q3
= 0.04
From (8.27)
ro A m = ---------------- g m R eq ro + R2
and since r o R 2 , the current gain of the second stage amplifier at medium frequencies is
gm 0.04 A m 2 = g m R eq = ---------- = -------------------------- = 34.8 3 G eq 3 1.15 10
842
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
From (8.35),
1 1 1- 1- 11 1 1 1 G eq 2 = ------ + ----- + ----- + ----- = ---------------- + ---------------- + ------------- + ------------------ = 0.45 m R A R B r be R 2 56 K 12 K 5 K 6.8 K
From (8.46),
gm I3 1 A c 2 = --- = ---------- -----------------------------------------------------I2 G eq 2 1 + j ( C E + C 2 ) G eq 2 1 1 0.04 = -------------------------- ---------------------------------------------------------------------------------------------3 = 89 ----------------------------------------- 12 6 3 1 + j3.36 10 0.45 10 1 + j ( 796 + 716 ) 10 0.45 10
From (8.27)
ro A m = ---------------- g m R eq ro + R2
and since r o R 2 , the current gain of the first stage amplifier at medium frequencies is
gm 0.04 A m 1 = g m R eq = ---------- = -------------------------- = 89 3 G eq 2 0.45 10
From (8.49), Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
843
The sketch below shows the asymptotes for the highfrequency amplitude characteristics with the gain in dB where A c ( dB ) = 20 log 11864 82 , the frequency in log scale, and the slope of each highfrequency asymptote is 20 dB per decade.
A ( dB )
80
40
f = 47.4 KHz
20 dB decade 40 dB decade
f = 18.3 KHz 0 10
0
10
10
f ( log scale )
4.
for n identical stages of this type connected in cascade, the overall amplification will be
Am A n = -----------------------------------------2 n ( 1 + ( H ) )
n
844
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
= 2
1n
= 2
from which
= B1 2
1n
This value of is the halfpower bandwidth of the cascade of n identical stages. Therefore,
Bn = B1 2
1n
1
Imaginary axis
5.
Imaginary axis
10 1 (a)
Real axis
20 1 (b)
2 1
Real axis
and when = 0 (DC condition), A V = 0 . Therefore, this amplifier will not transmit DC signals. For the amplifier whose polezero pattern is shown in (b),
( j + 1 ) A V = -------------------------------------------------------( j + 2 1 ) ( j + 20 1 )
Therefore, this amplifier will transmit DC signals. b. The MATLAB script and resulting plot are shown below.
w=1:10:100000; N=j.*w+100; D=(j.*w).^2+2200.*(j.*w)+400000;... Av=4.*10.^4.*N./D; AvdB=20.*log10(abs(Av)); semilogx(w,AvdB); grid;... xlabel('Frequency in rad/s (log scale)'); ylabel('Gain (dB scale)');... title ('Amplitude characteristic for Exercise 5')
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
845
10
10
10
6.
v1 v 1 v2 v3
RL
v 2
(a)
v 3
50 K
10
v1
(b)
50v 1
v1
(c)
1v 1
a. If the amplifier is connected directly to the loudspeaker, the total series resistance will be
R eq = 500 + 10 = 510
846
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
b. Let us assume 1 v input to each of the amplifiers in Figures (b) and (c). 1. Current in amplifier of Figure (b) will be
50 1 I b = -------------------------- = 5 mA 10000 + 10
Therefore, the amplifier of Figure (c) will deliver 10 w to the loudspeaker with the smallest input voltage. c. 1. Voltage amplification in amplifier of Figure (b) will be
50v 1 A b = 20 log ---------- = 34 dB v1
Therefore, the amplifier of Figure (b) will produce the greatest voltage amplification. d.
500 10 K 10 K 10 10 RL
v2
v3 50v 2 50v 3
v4
v1
1 M
1v 4
For last stage we must have P L = I 2 R L = 10 w and since R L = 10 , I L = 1 A , and L V L = 10 V . Therefore, v 4 = 10 + 10 = 20 V , and Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
847
The first and second stages were so chosen as to provide the greatest voltage amplification.
848
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Chapter 9
Tuned Amplifiers
his chapter begins with an introduction to tuned amplifiers. We will examine the properties of various tuned amplifiers that find applications to telecommunications systems employing narrowband modulated signals. New tools for the analysis and design of tuned amplifiers are developed.
* For a thorough discussion on passive and active filters, please refer to Signals and Systems with MATLAB Applications, ISBN 9781934404119.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
91
Tuned Amplifiers
the phase, and they give rise to amplitudemodulated (AM), frequencymodulated (FM), and phasemodulated (PM) signals. To understand the requirements that must be met by tuned amplifiers it is necessary to examine the nature of these waves and to understand the way in which they are used in telecommunication systems. The properties of AM waves are developed in the paragraphs that follow; although they are not discussed specifically, FM* and PM waves have similar properties. If the information, or modulating, wave is a voltage v m ( t ) , then an AM carrier can be expressed as
v s ( t ) = [ V c + v m ( t ) ] cos C t
(9.1)
where C is the carrier frequency, and V C is the amplitude of the unmodulated carrier. The waveform of such an AM wave is shown in Figure 9.1.
vm ( t )
Modulating signal Carrier Envelope
vc ( t )
Figure 9.1. Signal and carrier wave that is amplitudemodulated by the signal.
In an AM system the magnitude of v m ( t ) is always less than V C ; hence the envelope of the modulated wave never drops to zero. From relation (9.1) and Figure 9.1 we observe that the waveform of the envelope of the modulated signal is the same as the waveform of the modulating signal v m ( t ) . Thus if the information in the modulated carrier is to be preserved, the waveform of the envelope must be preserved. The information can be recovered by recovering the waveform of the envelope; the peak rectifier circuit that we discussed in Chapter 2 is used for this purpose. The peak rectifier is also known as diode detector, and in this application it is frequently called an envelope detector. To examine the properties of the AM wave further, it is convenient to assume that the modulating signal is a sinusoid expressed as
v m ( t ) = V m cos ( m t + m )
(9.2)
* FM and PM systems are best described by Bessel functions. For an introduction to these functions, please refer to Numerical Analysis Using MATLAB and Excel, ISBN 9781934404034.
9 2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
m = Vm Vc v s ( t ) = V c [ 1 + m cos ( m t + m ) ] cos c t
and the quantity m is known as the modulation index in an AM system, and it is always less than unity. In (9.2) and (9.5) we have assumed that the modulating signal v m ( t ) is sinusoidal and as such, it predictable for all times; hence, it conveys no useful information. In practice, v m ( t ) is nonsinusoidal but if it is periodic in nature as most physical phenomena are, it can be expanded in a Fourier series,* and the analysis of an AM system can be performed by the superposition of the effects of each sinusoidal component known as harmonics. Relation (9.5) can be simplified by using the trigonometric identity for the product of two cosines. Thus,
m m v s ( t ) = V c cos c t + --- V c cos [ ( c + m )t + m ]t + --- V c cos [ ( c m )t m ] 2 2
(9.6)
Relation (9.6) reveals that a sinusoidally modulated AM wave is the superposition of three sinusoidal components, a carrier wave at the carrier frequency, and two side waves at frequencies on either side of the carrier frequency. The frequency spectrum of a typical sinusoidally modulated AM wave is shown in Figure 9.2.
vs ( t ) vs ( t )
Ideal bandpass filter
(a)
(b)
Figure 9.2(a) shows the relative amplitudes and frequencies of the carrier and side waves, except that usually the side waves are much closer to the carrier wave than it is possible to show in the diagram. In the more general case there are many pairs of side waves, and these are referred to as the upper and lower sidebands of the AM wave. In most AM waves the carrier frequency is much larger than the highest frequency in the modulating signal. Thus, a typical AM radio station broadcasting speech and music with frequencies up to 5 KHz has a carrier frequency of 1 MHz
* For a detailed discussion and applications in Fourier series, please refer to Signals and Systems with MATLAB Computing and Simulink Modeling, ISBN 9781934404119.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
93
Tuned Amplifiers
and a total bandwidth of 10 KHz centered at 1 MHz . The advantage of AM signals is that they are narrowband signals; that is, their total bandwidths are small compared to their center frequencies. A narrowband allows the transmission of many simultaneous telephone conversations on a single pair of conductors in a coaxial cable and for the simultaneous use of space by all radio transmitters. Figure 9.2(b) shows the frequency spectra of three AM signals indicated as 1 , 2 , and 3 . Each signal uses a different carrier frequency, and normally the difference between these frequencies is great enough so that the spectra of the signals are completely separated and occupy different frequency bands. These signals can be separated by frequencyselective amplification; the idealized gain characteristic for such an amplifier is indicated in Figure 9.2(b). Amplifiers of this kind are called bandpass amplifiers. Tuning a radio receiver is the operation by which the passband of the amplifier is shifted from one place to another in the frequency spectrum. The study of tuned amplifiers in the sections that follow is concerned with the response of the amplifiers to modulated waves and with design techniques for obtaining optimum performance. Optimum performance is concerned with the rejection of adjacent channels as well as with faithful amplification of the desired channel. Relation (9.6) shows that a symmetry exists between the upper and lower sidebands of an AM wave. The sidebands have equal amplitudes, and they have phase angles with respect to the carrier that are equal in magnitude but opposite in sign. If these components are to add up to give the original AM wave, this symmetry must be preserved. The significance and importance of this symmetry is illustrated further by the phasor diagrams in Figure 9.3, in which each sinusoidal component of the wave is represented by a rotating vector. If the carrier vector is taken as the reference and is assumed to be stationary, then the uppersideband vector rotates counterclockwise at m rad sec , and the lowersideband vector rotates clockwise at the same speed.
m VC m
(a)
m VC
(b)
(c)
In Figure 9.3(a) the sum of the two sideband vectors (represented by the dotted line) is at every instant colinear with the carrier vector and it gives a sinusoidal variation in the amplitude of the total voltage. However, that this result depends on the symmetry between the sidebands. To illustrate what may happen when the symmetry is destroyed, Figure 9.3(b) shows the case in which the sidebands have been shifted 90 outofphase in phase with respect to the carrier. In this case, the sum of the sideband vectors (represented by the dotted line) is at every instant normal to the carrier vector, and the amplitude of the total voltage becomes almost constant as indi-
9 4
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(9.7)
(9.8)
The frequency spectrum for this signal is similar to the spectra shown in Figure 9.2 except that in this case each sideband contains a sinusoidal component corresponding to each sinusoidal component of the modulating signal; each pair of sideband components exhibits the symmetry discussed above. If the modulating signal is allowed a constant time delay T, then (9.7) becomes
v m ( t ) = V 1 cos [ 1 ( t T ) 1 ] + V 2 cos [ 2 ( t T ) 2 ] + V 3 cos [ 3 ( t T ) 3 ] +
(9.9)
or
v m ( t ) = V 1 cos ( 1 t 1 T 1 ) + V 2 cos cos ( 2 t 2 T 2 ) + V 3 cos ( 3 t 3 T 3 ) + (9.10)
Since T is a constant, it follows from (9.10) that imposing a constant time delay on v m ( t ) is equivalent to imposing a lagging phase shift on each sinusoIdal component, proportional to the freElectronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
95
Tuned Amplifiers
quency of that component. Thus if we let 1 T = 1 , 2 T = 2 and so on, the corresponding AM wave of (9.8) with the delayed envelope can be expressed as
1 1 v s ( t ) = V C cos C t + -- V 1 cos [ ( C + 1 )t 1 1 ] + -- V 1 cos [ ( C 1 )t + 1 + 1 ] 2 2 1 1 + -- V 2 cos [ ( C + 2 )t 2 2 ] + -- V 2 cos [ ( C 2 )t + 2 + 2 ] + 2 2
(9.11)
where 1, 2, are directly proportional to the separation of the side frequency from the carrier frequency. Thus, if an AM signal is transmitted through an amplifier having a linear phase characteristic with a negative slope, it can be shown by going through this process in reverse that the result is a pure delay in the envelope. It can also be shown that a constant phase shift added to the linear phase characteristic produces a delay in the carrier. Alternatively, a constant phase shift rotates all vectors in the diagrams of Figure 9.3 by the same amount and leaves the relations among them unchanged. The results obtained in the preceding discussion are useful guides in judging the merits of various tuned amplifiers and in designing these amplifiers for optimum performance. A good amplifier for use with modulated signals should have uniform gain in the frequency band occupied by the signal, and it should have a linear phase characteristic in that frequency band. Example 9.1 A carrier
v C ( t ) = 10 cos 2 10 t
6
Sketch a diagram of the frequency spectrum of the amplitudemodulated wave similar to that of Figure 9.2. Show the amplitude and frequency for each component of the AM wave. Solution: From relation (9.6)
3 3 6 6 3 6 3 v S ( t ) = 10 cos 2 10 t + -- cos [ ( 2 10 + 10 )t + m ]t + -- cos [ ( 2 10 10 )t ] 2 2 5 5 6 3 6 3 + -- cos [ ( 2 10 + 2 10 )t + m ]t + -- cos [ ( 2 10 2 10 )t ] 2 2
9 6
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Example 9.2 A radio receiver* with a parallel GLC circuit whose inductance is L = 0.5 mH is tuned to a radio station transmitting at 810 KHz resonant frequency. The resonant frequency is given by
0 = 1 LC and the quality factor at parallel resonance is given by Q 0P = 0 C G
a. What is the value of the capacitor of this circuit at this resonant frequency? b. What is the value of conductance G if Q 0P = 75 ? c. If a nearby radio station transmits at 740 KHz and both signals picked up by the antenna have the same current amplitude I ( A ), what is the ratio of the voltage at 810 KHz to the voltage at 740 KHz ? Solution: a. or Then, b. or c.
0 = 1 ( LC )
2
I III V 810 KHz = --------------------- = ----- = --- = -------------------------6 Y 810 KHz Y0 G 5.24 10
(9.12)
* For a block diagram of a typical AM radio receiver and description, please refer to Circuit Analysis II with MATLAB Applications, ISBN 9780970951151, Chapter 2, Figure 2.16.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
97
Tuned Amplifiers
Also, where
Y 740 KHz = I V 740 KHz = --------------------Y 740 KHz G + (C 1 (L) )
3 12 2 2
(9.13)
or
Y 740 KHz = ( 5.24 10 ) + 2 740 10 77.2 10
6 2 2 1 -----------------------------------------------------------------3 3 2 740 10 0.5 10
(9.14)
6 6 V 810 KHz 71.2 10 I 5.24 10 ---------------------- = -------------------------------- = -------------------------- = 13.6 6 6 V 740 KHz I 71.2 10 5.24 10
(9.15)
that is, the voltage developed across the parallel circuit when it is tuned at f = 810 KHz is 13.6 times larger than the voltage developed at f = 740 KHz .
v
R1
gm v
R2 I2
In the circuit of Figure 9.5, we assume that the parasitic capacitances are negligible, and with this assumption this circuit is good at frequencies up to a few hundred KHz. The resistance R 1 is in the order of 1 kilohm or less and represents the internal resistance of the current source, the input resistance of the transistor, the biasing transistors, and the losses in the inductor. The admittance Y ( s ) on the left side of the circuit is
1 1 Y ( s ) = ----- + ----- + sC R 1 sL
(9.16) (9.17)
and thus
I 1 ( s ) = Y ( s )V ( s )
9 8
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(9.18)
or
gm s - 2 A c ( s ) = ------ -----------------------------------------------------C s + ( 1 RC )s + 1 LC
(9.19)
(9.20) (9.21)
and
1 2 = ------RC
(9.22)
The quadratic denominator in (9.18) can be expressed as the product of two linear factors, i.e.,
gm s A c ( s ) = ------ ---------------------------------C ( s s1 ) ( s s2 )
s1 = + 0 s2 = 0
2 2
Depending on the values of the circuit parameters, the poles of (9.18) may be either real or complex. If they are real, relation (9.19) has the same form as relation (8.71) in Chapter 8; hence the polezero pattern and the frequency characteristics of the tuned amplifier with real poles are the same as those of the RCcoupled amplifier which cannot provide a truly narrowband characteristic. Therefore, in our subsequent discussion we will be concerned with circuit parameters which will yield complex poles. We are also interested in the current gain at the resonant frequency denoted as A c0 . From the parallel network of Figure 9.5, we observe that at parallel resonance inductive susceptance* and
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
99
Tuned Amplifiers
capacitive susceptance cancel each other and the admittance of the parallel R 1 LC circuit reduces to R 1 and thus Ac 0 = I2 I1 = gm R1 (9.26) As in the case of the RC amplifier in Chapter 8, the halfpower frequencies provide a measure of the bandwidth of the circuit, and the halfpower frequency is the frequency at which the amplification is A = A c ( max ) 2 . Also, there are two halfpower frequencies, one above and the other below the resonance frequency and thus the bandwidth between the two halfpower frequencies is given by
1BW = 2 = ------RC
(9.27)
Example 9.3 The equivalent circuit of singletuned amplifier is shown in Figure 9.6 and has a transconductance g m = 40 m 1 and R 1 = 500 .
I1 R1
R2
gm v
I2
Figure 9.6. Transistor equivalent for the tuned amplifier of Example 9.2
a. Find the values of L , C , and the amplification at resonance so that the halfpower bandwidth will be BW = 10 KHz and the passband will be centered at 1 MHz . b. The quality factor Q 0 * at parallel resonance. Solution: a. For the specified bandwidth it is necessary that
4 1 BW = --------- = 2 10 R1 C
1 C = ----------------------------------- = 32 nF 4 2 10 500
* For a detailed discussion on quality factor Q, please refer to Circuit Analysis II with MATLAB Applications, ISBN 9780970951151
910
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
and with C = 32 10 9 ,
In Example 9.2, the very low value of R 1 = 500 makes a large value of C necessary to obtain the narrow bandwidth BW = 10 KHz , and a relatively high quality factor Q 0 . Moreover, the very low value of L causes the inductive reactance at 1 MHz to have a very low value of
0 L = 2 10 0.8 10
6 6
= 5
Fortunately, we can transform values required by the specifications into practical values of the circuit parameters by the use of a transformer. Figure 9.7(a) shows a practical tuned transistor amplifier in which a tapped inductor is used as a stepdown autotransformer.
V CC I1 RA RC I1 a
v'
C L RB
vs RP
v out
RE CE
L1 (a)
C v s --a2
a2 L
v'
(b)
R'
gm v
R2 I2
Figure 9.7. Impedance transformation in a tuned transistor amplifier and its model
Resistors R A and R B provide the necessary biasing but they also include coils in series with these resistors (not shown) and these coils are known as radio frequency chokes (RFCs). An RFC blocks radio frequencies but passes audio frequencies when both frequencies are applied to the same circuit. The inductors used in the autotransformer have ferrite cores, the coefficient of coupling* among the turns is close to unity, and the transformation ratio is
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
911
Tuned Amplifiers
V1 ----- = V L N1 ----1 = ----- = a < 1 L N
(9.28)
where L is the total inductance, N is the number of turns in the coil, L 1 is the inductance portion of the coil below the tap, and N 1 is the number of turns in the coil below the tap. When the portion of the circuit on the left of the transformer coil in Figure 9.7(a) is transferred to the right side, the small signal model assumes the form shown in Figure 9.7(b), and the capacitance C and the total inductance L are transformed by the factor a 2 . In Figure 9.7(a), the primary side of the stepdown autotransformer is the total inductance and this represents the primary side of the autotransformer, and the secondary side is the portion of the coil below the tap. As we know, a practical inductor contains some resistance in series with it. Let the resistance in series with the coil in the primary be denoted as R P and the resistance in series with the coil in the secondary be denoted as R S . The resistance R P includes the internal resistance of the source supplying the input signal, and when it is referred to the secondary, it assumes the value a 2 R P . The resistance R' shown in Figure 9.7(b) represents the parallel combination of a 2 R P with R S , that is,
a2 RP RS R' = ----------------------a2 RP + RS
(9.29)
The input signal current is also transformed to I 1 a as shown in Figure 9.7(b) and since a is less than unity, we observe that the current amplification is increased by the transformer action. From Figure 9.7(b)
C 1 1 Y' ( s ) = ---- + ---------- + ---- s 2 R' a Ls a 2 I1 ( s ) ---------- = Y' ( s )V' ( s ) a
Also,
* For a detailed discussion on transformers, please refer to Circuit Analysis II with MATLAB Applications, ISBN 9780970951151
912
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(9.33)
Therefore, the transformer action leaves the resonant frequency 0 unchanged, but the bandwidth becomes
2 a2 RP + RS a BW = -------- = ----------------------RP RS C R'C
(9.34)
The LC combination in Figure 9.7(a) forms a lossless coupling network and if R P and R S are fixed, the coupling network can be adjusted for maximum amplification, and this condition occurs when maximum power is delivered to R S . Therefore, maximum amplification is obtained when the value of a is chosen so that a2 RP = RS (9.35) At resonance, the inductive susceptance and capacitive susceptance cancel out and thus the admittance is just R' . Therefore, relation (9.33) reduces to
g m R' A co = ----------a
(9.36)
(9.37)
Example 9.4 The equivalent circuit of singletuned amplifier shown in Figure 9.8 has the transformation ratio
a = 1 20 , transconductance g m = 40 m , and R' = 500 .
I1 a C ---a2 R2
1
vs
a2 L
v'
R'
I2
a. Find the values of L , C , and the amplification at resonance so that the halfpower bandwidth will be BW = 10 KHz and the passband will be centered at 1 MHz . b. The quality factor Q 0 at parallel resonance.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
913
Tuned Amplifiers
Solution: a. For the specified bandwidth it is necessary that
4 a2BW = -------- = 2 10 R'C
500 = 100
914
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Re
Figure 9.9. Polezero pattern for two synchronously tuned amplifiers in cascade
In the study of tuned amplifiers, two important relations are the narrowband approximations. They are very useful in the analysis and design of practically all narrowband amplifiers and filters. For the typical narrowband amplifier they are very good approximations. To understand these approximations, we refer back to relations (9.24) and (9.25), that is,
s1 = + 0 s2 = 0
2 2 2 2
(9.38) (9.39)
= 0 s 1 = + j s 2 = j
From relation (9.42) we observe that 0 is the hypotenuse of a right triangle, and in the complex frequency plane it represents the distance of the poles s 1 and s 2 from the origin as shown in Figure 9.10(a).
s1
Im Im
( s s1 )
Re
s = j
Re
s2
(a)
( s s2 )
(b)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
915
Tuned Amplifiers
A typical AM signal has a center frequency that is 100 times the bandwidth of the signal to be transmitted. An amplifier suitable for use with this signal will have a resonant frequency of 0 = 200 and as we can see from Figure 9.10(b), the poles s 1 and s 2 are extremely close to the imaginary axis. Now, let us consider the sinusoidal operation where s = j and as j moves along the imaginary axis near the pole s 1 , the vector s s 1 gets smaller and experiences large percentage variations. However, since the vector s s 2 is relatively further away, the percentage variations are much smaller and the variations in s and s s 2 tend to cancel. Therefore, for sinusoidal operation in the passband we obtain the narrowband approximations
s = j j 0
(9.45) (9.46)
and
s s 2 2j 0
When these approximations are substituted into relation 9.23, we obtain the simplified expression
gm gm j 0 1 A ( s ) = ------ ---------------------------- = ------ ---------------C ( s s 1 )2j 0 2C ( s s 1 )
(9.47)
Relation (9.47) shows that in narrowband operation the ratio s ( s s 2 ) can be expressed as the constant 1 2 and that the narrowband approximations reduce the quadratic factor in the denominator to a linear factor. This reduction is very important when more complex circuits are to be analyzed. Also, for sinusoidal operation s = j and relation (9.47) can be expressed as
gm 1 A ( j ) = ------ --------------2C j s 1
(9.48)
and since 0 = 200 , we see from Figure 9.10(a) that 0 , and thus
s 1 = + j + j 0
(9.49)
(9.50)
(9.51)
916
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(9.52)
(9.53)
(9.54)
This relation can also be expressed in terms of the quality factor at resonance, i.e., BW = 0 Q 0 . Then,
1 A ( j ) = A 0 --------------------------------------1 + j2Q 0 0
(9.55)
In terms of the bandwidth (relation9.54), the magnitude of the amplification for one stage is
1 A 1 = A = A 01 ----------------------------------------------2 1 + ( 2 BW 1 )
(9.56)
where the subscript unity denotes a single stage. For n stages connected in cascade, the magnitude of the amplification is
1 n n A 1 = A = A 01 ----------------------------------------------n 2 1 + ( 2 BW 1 )
(9.57)
The halfpower frequency for n identical stages connected in cascade occurs when the square of the denominator in (9.57) equals 2. Thus, at the halfpower frequency
2 2 1 + ----------- = 2 BW 1
(9.58)
from which
BW 1 1 n = ----------- 2 1 2
(9.59)
There are two halfpower frequencies, one above and one below the resonant frequency. The bandwidth between these two halfpower frequencies is
BW n = 2 = BW 1 2
1n
(9.60)
The square root of this expression is the bandwidth reduction factor for n identical stages connected in cascade; for two stages it is 0.64 , for three stages it is 0.51 , and for four stages it is 0.43 .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
917
Tuned Amplifiers
Example 9.5 Two synchronously tuned stages connected in cascade are to designed so that the resonant frequency will be f 0 = 25 MHz , and the capacitors will have a value of C = 80 pF for each stage. Determine the values of L and R and the halfpower bandwidth for each stage so that the overall halfpower bandwidth will be 100 KHz . Solution: The halfpower bandwidth of each stage can be determined from relation (9.60), the values of R from the relation BW 1 = 1 RC and the values of L from 2 = 1 LC . For convenience, we will 0 let the following MATLAB script do the calculations for us.
BWn=2*pi*10^5; C=25*10^(12); L=1/((2*pi*25*10^6)^2*C); n=2;... BW1= BWn/sqrt(2^(1/n)1); BWf=BW1/(2*pi); R=1/(BW1*C); fprintf(' \n');... fprintf('C = %5.2e F \t',C);... fprintf('L = %5.2e H \t',L);... fprintf('BW1 = %5.2e rad/sec \t', BW1); fprintf(' \n');... fprintf('BWf = %5.2e Hz \t', BWf);... fprintf('R = %5.2e Ohms \t', R); fprintf(' \n')
Upon execution of this script, MATLAB displays the following results: C = 2.50e-011 F BWf = 1.55e+005 Hz L = 1.62e-006 H BW1 = 9.76e+005 rad/sec R = 4.10e+004 Ohms
The halfpower bandwidth of each stage is 155 KHz and, as expected, is larger than the overall halfpower bandwidth of 100 KHz . Synchronous tuning for two stages connected in cascade does not provide the best approximation to the ideal bandpass. As stated above, the bandwidth reduction for synchronous tuning of two stages connected in cascade is about 0.64 and this is confirmed by the ratio 100 KHz 155 KHz . Figure 9.11 shows the frequency response of a synchronous tuning for two stages in cascade in comparison to a single stage.
918
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Ac
(log scale)
Figure 9.11. Frequency responses for single stage and two synchronously tuned amplifiers.
Im
s1 s3
(2) Re
Ac
(dB scale)
s4 s2 0
(a) (b)
(log scale)
Figure 9.12. Polezero pattern and amplitude characteristic for two staggertuned amplifiers
Poles s 1 and s 2 in Figure 9.12(a) are the poles one of the stages, and poles s 3 and s 4 are the poles of the other stage; all four poles lie on a line parallel to the imaginary axis. With this arrangement, the amplitude characteristic of the staggertuned amplifier will be shown in Figure 9.12(b). We observe that the passband of such staggertuned amplifier is wider and flatter than the passband of the synchronously tuned amplifier. The top flatness of the characteristic depends on the spacing between the poles in Figure 9.12(a), and it is under the control of the designer. Consider, for example the relations (9.20) and (9.21) which are repeated below for convenience, and the polezero pattern of Figure 9.10(a).
2 1 0 = ------LC
(9.61) (9.62)
1 2 = ------RC
From relation (9.62) we observe that if R and C remain constant, the spacing from the imaginary axis remains constant also. From relation (9.61) we observe that a change in L will result in Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
919
Tuned Amplifiers
a change in 0 but since remains constant, the poles must move on a path parallel to the imaginary axis. Hence, the staggering of the poles along the vertical line can be adjusted by adjusting the inductances in the two stages for slightly different values. We found that the current gain of one stage under sinusoidal conditions and after the narrowband approximations have been applied, is given by relation (9.48), that is,
gm 1 A ( j ) = ------ --------------2C j s 1
(9.63)
(9.64)
where s 1 = + j 1 , s 3 = + j 3 , is as shown in Figure 9.10(a), C 1 and C 2 are the capacitances of the first and second stages respectively, and the narrowband approximations have reduced a fourthdegree polynomial in the denominator to a seconddegree polynomial. The narrowband approximations and relation (9.64) reveal that the frequency characteristics of the amplifier in the passband are determined by the two poles s 1 and s 3 , and the poles s 2 and s 4 together with the zeros at the origin contribute the constant 1 4 in the scale factor of that relation. Figure 9.13(a) shows an enlarged view of the polezero pattern in the vicinity of the poles s 1 and s 3 .
Im
s1
1 s j c 3
s3
(a)
c
(b)
The frequency c , which is the center frequency of the passband, corresponds to a point on the imaginary axis equidistant from s s 1 and s 3 , and the vectors 1 and 3 correspond to the linear factors j s 1 and j s 3 . The amplification, given by (9.64), can be expressed as
920
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(9.65)
As the variable s = j moves along the imaginary axis, the area of the triangle s 1 s s 3 remains constant and its value is (9.66) a = The area can also be expressed in trigonometric form as
1 a = -- 1 3 sin 2
(9.67) (9.68)
(9.69)
The poles for the two stages have the same real parts, = 1 2R 1 C 1 = 1 2R 2 C 2 and thus relation (9.69) can be expressed as
2 A = g m R 1 R 2 ----- sin 2
(9.70)
where g 2 R 1 R 2 is the resonant amplification of two stages with synchronous tuning. Therefore, in m relation (9.70), sin is the only factor that varies with changes in . If is less than , a circle with center at the middle of the line joining poles s 1 and s 3 that is parallel to the imaginary axis and with radius , this circle will intersect the imaginary axis at two points, as shown by the dotted semicircle in Figure 9.13(a), and the amplitude characteristic has two peaks as shown in Figure 9.13(b). The amplification at the peaks occurs when sin = 1 and thus
2 A p = g m R 1 R 2 ----2
(9.71) (9.72)
In general,
A = A p sin
If is greater than , the semicircle does not intersect the imaginary axis and the amplitude characteristic has only one peak; it is located at = c and it is less than A p . The frequencies at which the double peaks occur are denoted as (9.73) p = c and since the radius of the semicircle is , it follows that Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
921
Tuned Amplifiers
=
2 2 2
(9.74)
The triplepoint frequencies t are the frequencies outside the double peaks at which the amplification has the same value as at the center frequency are obtained by
t = c 2
(9.75)
Let c be the value of the angle when = c . Then, the amplification at the center frequency, denoted as A c is
A c = A p sin c
(9.76)
(9.77)
(9.78)
When the amplifier is adjusted so that = , the semicircle is tangent to the imaginary axis, and the amplitude characteristic is maximally flat. In this case, the amplifier is also said to be flat staggered, and this condition marks the transition from the overstaggered (double peaks) to the understaggered adjustment. We are mostly interested on the flatstaggered and the overstaggered cases. In all of the amplifiers studied up to this point, the halfpower bandwidth has been related in a simple way to the circuit parameters. In the case of the overstaggered pair of tuned stages, however, a simpler and in some respects more useful measure of the bandwidth is the socalled triple point bandwidth shown by the dotted lines in Figure 9.13(b). It is defined as
W = 2 2 *
(9.79)
2
(9.80)
It is convenient to express relation (9.80) in terms of the design parameter r = which is related to the peaktovalley ratio in relation (9.77). Rearranging relation (9.80) we obtain
W 2 2 ------ = 2 -- 1 = 2 ( r 2 1 ) 2
(9.81)
* We reserve the notation BW for the 3 dB bandwidth and we will use W for the triplepoint bandwidth.
922
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
s1
b
Radius = 2 BW --------2
BW
s3
(a)
(b)
Figure 9.14. Determination of the halfpower frequencies of flatstaggered and overstaggered amplifiers
In Figure 9.14(a) the line segment joining the poles s 1 and s 3 subtends the angle 2 = 90 , the radius of the circle is 2 , and the center of the circle lies a distance to the right of the imaginary axis. Figure 9.14(b) is an enlarged version of the small right triangle oab and by the Pythagorean theorem
BW = 2 2 ( ) 2 -------- 2
2
(9.82) (9.83)
or
2 2 BW = -- + 2 1 ------------ 2
(9.84)
Using relation (9.84), flatstaggered and overstaggered amplifiers can be designed for a specified halfpower bandwidth. Also, a comparison of (9.84) with (9.81) reveals that the halfpower bandwidth and the triplepoint bandwidth are nearly equal except for cases of very slight overstaggering. The relations derived above provide a simple, direct design procedure for overstaggered stages.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
923
Tuned Amplifiers
The design requirements specify the center frequency c , the triplepoint bandwidth W , and the peaktovalley ratio A p A p .These specifications are sufficient to define the poles of the amplitude gain A , and hence the circuit parameters. Specifically, the ratio A p A p defines
r = through relations (9.77) and (9.78). Then, relation (9.81) gives the required value of , and the required value of is obtained from the relation = r . Thus, the two of the three parameters R , L , and C are determined, and the third can be chosen as desired to satisfy the
other requirement as it was illustrated in the previous examples. Example 9.6 A transistorized twostage staggertuned amplifier has a polezero pattern like the one shown in Figure 9.13. The pertinent dimensions are p = 10 6 rad/s , = 1500 rad/s , and = 2500 rad/s . Both transistors have a transconductance of g m = 40 m 1 , and the total shunt resistance in each stage is R 1 = R 2 = 0.75 K . Find the peak amplification A p , the peaktovalley ratio
A p A c , the triplepoint bandwidth W , the halfpower bandwidth BW , and the triplepoint fre-
Therefore, the amplifiers are overstaggered. The triplepoint bandwidth W , is given by relation (9.79), i.e., W = 2 2 where from relation (9.74)
924
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
t 2 = 0.997 10 rad/s
The procedure for finding all pertinent data, except the values of L and C , was well illustrated in Example 9.6. The following example illustrates the procedure for finding of L and C to meet all design specifications. Example 9.7 The pole locations for a twostage staggertuned amplifier and the values of R 1 and R 2 are as given in Example 9.6. Determine the values of L and C required for each stage. Solution: From relation (9.27) and thus
1 BW = 2 = ------RC 1 1 2 = ----------- = ----------R1 C1 R2 C2
and Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
925
Tuned Amplifiers
1 1 L 1 = ------------- = ---------------------------------------------------------------- = 2.28 H 2 12 6 2 0.997 10 0.44 10 t 1 C1 1 1 L 2 = ------------- = ---------------------------------------------------------------- = 2.27 H 2 2 12 6 1.003 10 0.44 10 t 2 C2
Figure 9.15 shows the normalized amplitude characteristics for two tuned stages under four different adjustments.
A -----Ac
2 3
Figure 9.15. Amplitude characteristics for two tuned stages with different degrees of staggering
For the curves in Figure 9.15, is held constant while is increased in steps. Curve 1 is for synchronous tuning, Curve 2 is for understaggering with c = 70 , Curve 3 is for flat staggering, and Curve 4 is for overstaggering with A p A c = 1.1 . The halfpower bandwidths are indicated by the solid dots on these curves. For the flatstaggered case = , r = 1 , and sin c = 1 . Thus, from relations (9.71) and (9.72) we obtain
1 2 A c = A p = -- g m R 1 R 2 2
(9.85)
Letting B fs denote the bandwidth of two stages with flat staggering and BW the 3 dB bandwidth of one of the stages, from (9.84) we obtain
B fs = 2 2BW
(9.86)
(9.87)
926
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
1 = 0.64 ( g m R 1 R 2 ) ( BW )
(9.88)
The foregoing discussion of staggertuned amplifiers considers only the case in which the poles are staggered along a line parallel to the imaginary axis. If the poles do not lie on the same vertical line, skewed frequency characteristics result, and, in accordance with the discussion in Section 9.1, this fact may cause distortion of the envelopes of AM signals.
1 3 5
In Figure 9.16, The center frequency c of the passband corresponds to the center of the circle on which the poles lie, and no matter how many stages are cascaded, the halfpower frequencies are the frequencies at which the circle intersects the imaginary axis. The more stages cascaded in this way, the flatter the amplitude characteristic in the passband and the steeper the characteristic at the edges of the band. The distribution of poles producing these characteristics is known as the Butterworth configuration.
* For a detailed discussion on this topic, please refer to Signals and Systems with MATLAB Computing and Simulink Modeling, ISBN 9781934404119.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
927
passive devices, i.e., resistors, inductors, and capacitors, and thus it provides no gain. For small signals, active filters with op amps are very popular. Tuned amplifiers can also be designed with bipolar junction transistors and MOSFETs.
The analysis and design of tuned amplifiers is greatly facilitated with the small signal models. For a single stage tuned amplifier the current gain at the resonant frequency is given by Ac0 = I2 I1 = gm R1
A radio frequency choke (RFC) blocks radio frequencies but passes audio frequencies when both frequencies are applied to the same circuit.
Tuned amplifiers are connected in cascade to achieve higher selectivity. When two identical stages are cascaded, the two stages are said to be synchronously tuned. The amplitude characteristic for two synchronously tuned amplifiers is the characteristic for
In the study of tuned amplifiers, two important relations are the narrowband approximations.
They are very useful in the analysis and design of practically all narrowband amplifiers and filters.
For synchronously tuned amplifiers the bandwidth between the halfpower points is given by BW n = 2 = BW 1 2
1n
The passband of such staggertuned amplifier is wider and flatter than the passband of the
synchronously tuned amplifier and thus provides a better approximation to the ideal bandpass.
2 A = g m R 1 R 2 ----- sin 2
928
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Summary
For staggertuned amplifiers, the frequencies at which the double peaks occur given by p = c For staggertuned amplifiers, the frequencies outside the double peaks at which the amplifica-
tion has the same value as at the center frequency are obtained by
t = c 2
For staggertuned amplifiers adjusted so that = , the amplifier is also said to be flatstag-
gered, and this condition marks the transition from the overstaggered (double peaks) to the understaggered adjustment. We are mostly interested on the flatstaggered and the overstaggered cases.
For an overstaggered pair of tuned stages the triplepoint bandwidth is defined as W = 2 2
where the distribution of poles produce characteristics is known as the Butterworth configuration.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
929
v
R1
gm v
R2 I2
a. Determine the values of L and C so that the amplifier will have a resonant frequency of 100 KHz and a halfpower bandwidth of 5 KHz . b. Determine the current gain A c 0 = I 2 I 1 , i.e., the current gain at the resonant frequency. 2. For the incremental model of a typical MOSFET tuned circuit below, g m = 5 m 1
V1 R 2
gm V1
R1
C V2
By a procedure similar to that in Section 9.2, we can show that the resonant frequency, the halfpower frequency, and the voltage gain at resonance are given by the relations
2 1 0 = ------LC
1 BW = 2 = --------R1 C
Av0 = gm R1
a. Find suitable values for circuit parameters so that the halfpower bandwidth will be the 10 KHz , and the passband will be centered at 1 MHz . b. Find the voltage amplification at resonance. c. Find the quality factor at resonance. 3. Two synchronously tuned stages are connected in cascade and the transconductance of both transistors used is g m = 40 m 1 where R 1 = 1 K and R 2 = 1.5 K are to designed so that the resonant frequency will be f c = 25 MHz . Find the resonant amplification. 4. A twostage flatstaggered tuned amplifier is to designed with two transistors each with transconductance g m = 40 m 1 where R 1 = 1 K and R 2 = 1.5 K are to designed so that the resonant frequency will be f c = 25 MHz . Find the resonant amplification.
930
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v
R1
gm v
R2 I2
and with R 1 = 1 K ,
1 1 C = ------------------- = -------------------------------------- = 31.8 nF 4 3 BW R 1 3.14 10 10
2 The value of L is found from 2 = 1 LC , or f 0 = 1 2LC . Thus, 0
1 L = ------------------------------------------------------------ = 80 H 5 2 9 ( 2 10 ) 31.8 10
2.
V1 R 2
gm V1
R1
C V2
For high gain we can choose a large value for R 1 . Or, we can choose a large value for C to overcome the parasitic capacitances. Let us assume that the parasitic capacitances are of more concern to us and choose a 200 pF capacitor. Then, the bandwidth requirement yields
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
931
Tuned Amplifiers
10 1 R 1 = ------------- = ----------------------------------- = 80 K 4 BWC 2 10 200
12
80 10 = 100
3. By relation (9.26) the resonant amplification of the first stage is A 1 = g m R 1 , and that of the second stage is A 2 = g m R 2 . Therefore, the overall gain is
A = g m R 1 R 2 = ( 40 10 ) 1 1.5 10 = 60
2 3 2 6
4. By relation (9.85)
1 2 1 3 2 6 A c = A p = -- g m R 1 R 2 = -- ( 40 10 ) 1 1.5 10 = 30 2 2
As expected this gain is half of that of Exercise 3 where the circuit was designed with synchronously tuned stages.
932
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Chapter 10
Sinusoidal Oscillators
his chapter begins with an introduction to sinusoidal oscillators. Subsequently, we will discuss the RC phaseshift and LC oscillator families, and we will describe the Armstrong, Hartley, Colpitts, crystal, and the crystalcontrolled Pierce oscillators.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
101
Sinusoidal Oscillators
An oscillator must provide amplification where the amplification of signal power occurs from the input to the output of the oscillator and a portion of the output is fed back to the input to sustain a constant input. We will briefly discuss positive and negative feedback with reference to the block diagram of a feedback amplifier shown in Figure 10.1.
v in
v out
v out = Av
(10.1)
(10.2)
where is known as the feedback factor. If v f subtracts from v in , then v = v in v f and the feedback is known as degenerative or negative, and the amplifier is provided with negative feedback. If v f is added to v in , then v = v in + v f and the feedback is known as regenerative or positive, and the amplifier is provided with positive feedback From (10.1), (10.2), and v = v in v f , we obtain the general expression for a feedback loop as
v out A A f = --------- = -------------1 A v in
(10.3)
where the quantity A is known as the loop gain, and 1 A is known as the amount of feedback. In a practical amplifier, the open loop gain A is much greater than unity and thus the gain with feedback A f is for all practical purposes equal to 1 and this means that A f is effectively dependent on the feedback network, and since this network can be constructed with accurate devices such as precision resistors and capacitors, the gain A f can be designed precisely. With negative feedback, relation (10.3) is expressed as
v out A A f = --------- = ---------------v in 1 + A
(10.4) (10.5)
102
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Sinusoidal Oscillators
Practically, all amplifiers employ negative feedback, but as we will see shortly, sinusoidal oscillators use positive feedback. Example 10.1 A certain amplifier has an openloop gain A of 70 at an angle of zero degrees, distortion of 10% , and normal input voltage of 1 V . Determine the effects of negative feedback if the feedback factor is 0.1 . Solution: For negative feedback A = 0.1 70 = 7 and in accordance with relation (10.4) the gain with negative feedback is
70 A A f = ---------------- = ----------- = 8.75 1+7 1 + A
The amount of distortion D without feedback is 10% of the normal input voltage v in of 1 V , and with feedback is
0.10 0.1 1 D f = ---------------- = ----------- = 0.0125 = 1.25% 1+7 1 + A
Thus, the introduction of negative feedback has reduced the amplifier gain by 18 dB or we can say that 18 dB of feedback has been applied to the input of the amplifier.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
103
Sinusoidal Oscillators
Example 10.1 above and Exercise 1 at the end of this chapter illustrate the effects of degenerative (negative) feedback. In our subsequent discussion we will be concerned with regenerative (positive) feedback. Let us again consider relation (10.5) which is repeated below for convenience.
v out A A f = --------- = ------------1 A v in
(10.6)
In an oscillator circuit the positive feedback must be large enough to compensate for circuit losses so that oscillations will be sustained. Moreover, a practical oscillator must oscillate at a predetermined frequency and thus the oscillator must include a frequencydetermining device which essentially is a bandpass filter allowing only the desired frequency to pass. Let us now assume that at that predetermined frequency the loop gain A in relation (10.6) above is very close to unity and in this case the closed loop A f is very large. Accordingly, it is possible to obtain a finite output with a very small input and this is the principle behind the operation of an oscillator. To find the predetermined frequency 0 to produce sustained oscillations, we express relation (10.6) in the sdomain, that is,
V out ( s ) A(s) A f ( s ) = ------------------ = ---------------------------1 ( s ) A ( s ) V in ( s )
(10.7)
(10.8)
To sustain oscillations at = 0 the denominator of (10.8) must meet the Barkhausen criterion which states that to provide sinusoidal oscillations the magnitude and phase angle of the term ( j )A ( j ) must be such that ( j 0 )A ( j 0 ) 1 (10.9) and (10.10) ( j 0 ) + A ( j 0 ) = 2n n = 1, 2, 3,
10.3 RC Oscillator
An amplifier and an RC network can form an oscillator as shown in Figure 10.2. This oscillator is also known as phaseshift oscillator. In the oscillator circuit of Figure 10.2 the transistor, being in the commonemitter configuration, provides 180 phase shift and the three RC sections can be selected that each section will provide 60 shift. Therefore, the total phase shift in that oscillator will be 3 60 + 180 = 360 and the condition of relation (10.10) will be satisfied. The condition of relation (10.9) can also be satisfied by making the product A equal to unity. For instance, if A = 10 we can make = 1 10 and the condition of relation (10.9) will also be satisfied.
104
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
LC Oscillators
C R
RL
The output of the oscillator contains only a single sinusoidal frequency. When the oscillator is powered on, the loop gain A is greater than unity and the amplitude of the oscillations will increase. Eventually, a level is reached where the gain of the amplifier decreases, and the value of the loop gain decreases to unity and constant amplitude oscillations are sustained. The frequency of oscillations is determined by the values of resistance and capacitance in the three sections. Variable resistors and capacitors are usually employed to provide tuning in the feedback network for variations in phase shift. For an RC phaseshift oscillator, the amplifier is biased for Class A operation to minimize distortion of the signal. For the RC oscillator of Figure 10.2, the frequency of oscillation is given by
1 0 = --------------------------------------2 C 4RR L + GR
(10.11)
where G is the transistor power gain and the values of the resistors and the transistor gain are such that the following relation is satisfied.
29R 4R L = 23 + --------- + --------RL R
(10.12)
In general, for an RC phaseshift oscillator the frequency of oscillation (resonant frequency) can be approximated from the relation
1 0 = --------------RC n
(10.13)
10.4 LC Oscillators
The LC type of oscillators use resonant circuits. As we know, a resonant circuit stores energy alternately in the inductor and capacitor. However, every circuit contains some resistance and this resistance causes reduction in the amplitude of the oscillations. To sustain oscillations with constant amplitude it is necessary to use regenerative feedback. Figure 10.3 shows a block diagram of a typical LC oscillator. In an LC oscillator the sinusoidal signal is generated by the action of an inductor and a capacitor. The feedback signal is coupled from the LC tank of the oscillator circuit Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
105
Sinusoidal Oscillators
by using a coil (tickler* or a coil pair) as shown in Figure 10.4(a) and 10.4(b) or by using a capacitor pair in the tank circuit and tap the feedback signal between them as shown in Figure 10.4(c).
L Feedback Amplifier C
v out
( a ) Tickler coil
( b ) Coil pair
( c ) Capacitor pair
V CC
V CC
* A tickler coil is an inductor that is inductively coupled to the inductor of the LC tank circuit.
106
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
L1 M L2 C
V CC
In a Hartley oscillator the feedback is provided through a coil pair as shown in Figure 10.4(b). For the oscillator circuit of Figure 10.6 the frequency of oscillation is
1 0 = ---------------------------------------------------------------------------------------------------------2 C ( L 1 + L 2 + 2M ) ( L 1 L 2 M ) ( h ob h ib )
(10.14)
where M is the mutual inductance* and h ob and h ib are the hparameters representing the output admittance with opencircuit input and input impedance with short circuit output respectively, as discussed in Chapter 3.
* For a detailed discussion on mutual inductance, please refer to Circuit Analysis II with MATLAB Applications, ISBN 9780970951151.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
107
Sinusoidal Oscillators
C1 L C2
V CC
The Colpitts oscillator provides better frequency stability than the Armstrong and Hartley oscillators. Moreover, the Colpitts oscillator is easier to tune and thus can be used for a wide range of frequencies. For the oscillator circuit of Figure 10.7 the frequency of oscillation is
0 = C1 + C2 1 h ob - ----------------- + ----------- -----LC 1 C 2 C 1 C 2 h ib
(10.15)
where h ob and h ib are the hparameters representing the output admittance with opencircuit input and input impedance with short circuit output respectively, as discussed in Chapter 3.
108
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Crystal Oscillators
The symbol and the equivalent circuit of a quartz crystal are shown in Figure 10.8 where capacitor C 1 represents the electrostatic capacitance between the electrodes of the crystal and in general C1 C2 .
Electrode C2 Crystal Electrode C1 L R
(10.16)
We now recall that in a series resonant circuit the quality factor Q at resonance is
0S L Q 0S = ----------R
(10.17)
and since the Q of a crystal oscillator is very high, the value of the resistance R in (10.17) must be very low and thus it can be omitted in relation (10.16) which can now be expressed as
( sL + 1 sC 2 ) Z ( s ) = 1 sC 1 || ( sL + 1 sC 2 ) = ( 1 sC 1 ) -------------------------------------------------1 sC 1 + sL + 1 sC 2
or
s + 1 LC 2 Z ( s ) = ( 1 sC 1 ) --------------------------------------------------------------2 s + [ ( C 1 + C 2 ) ( LC 1 C 2 ) ]
(10.18)
The denominator of (10.18) is a quadratic and it implies the presence of two resonant frequencies which can be found by inspection of the equivalent circuit of Figure 10.10. The resonance of the series branch occurs when the imaginary part of the impedance is equal to zero. Thus, letting s = j we obtain Z ( j ) = jL + 1 jC 2 = 0 and denoting this frequency as 0S we obtain
1 0S = -------------LC 2
(10.19)
We also can prove* that the resonance of the parallel combination occurs when
0P = C1 + C2 ----------------LC 1 C 2
(10.20)
* The proof is left as an exercise for the reader at the end of this chapter.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
109
Sinusoidal Oscillators
As stated above, C 1 C 2 and under this condition relation (10.20) reduces to that of relation (10.19). Example 10.2 A crystal oscillator has a nominal frequency of oscillation at 5 MHz with Q = 85, 000 and with reference to the equivalent circuit of Figure 10.8, it is known that L = 50 mH , C 1 = 2 pF , and C 2 = 0.02 pF . Find: a. The series resonance 0S b. The parallel resonance 0P c. The value of the resistance R Solution: a. From (10.19)
1 1 --------------------- = -------------------------------------------------------------------- = 5.033 MHz 3 12 2 LC 2 2 50 10 0.02 10
b. From (10.20)
12 1 1 C1 + C 2.02 10 f 0P = ----- -----------------2 = ----- ------------------------------------------------------------------------------------ = 5.058 MHz 2 50 10 3 2 10 12 0.02 10 12 2 LC 1 C 2
c. From (10.17)
0S L 2f 0S L Q 0S = ----------- = ---------------R R
or
6 2f 0S L 2 5.033 10 R = ---------------- = ---------------------------------------- = 372 3 Q 0S 85 10
1010
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
C1 RC CE RE R1 V CC RB CB
C2 Output
Figure 10.9. Pierce oscillator with PNP transistor in the commonbase configuration
In the oscillator circuit of Figure 10.9, feedback is provided from the collector to the emitter of the transistor through capacitor C 1 and resistors R 1 , R B , and R C are used to establish the proper bias conditions. Besides the crystal, the frequency of oscillation is also determined by the settings of the variable capacitors C E and C B .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
1011
angular, and sawtooth waveforms. An oscillator is essentially an amplifier circuit that provides its own input signal. In this chapter we discussed sinusoidal oscillators. Nonsinusoidal oscillators are known as relaxation oscillators.
Many lowfrequency oscillators use resistors and capacitors to form their frequencydetermining networks and are referred to as RC oscillators. These are used in the audiofrequency range. Other types of employ inductors and capacitors for its frequencydetermining network.
These are known as the LC oscillators and use tank circuits, are commonly used for the higher radio frequencies.
oscillator provides excellent frequency stability and is used from the middle of the audio range through the radio frequency range.
An oscillator must provide amplification where the amplification of signal power occurs from
the input to the output of the oscillator and a portion of the output is fed back to the input to sustain a constant input.
In an oscillator circuit the positive feedback must be large enough to compensate for circuit
losses so that oscillations will be sustained. Moreover, a practical oscillator must oscillate at a predetermined frequency and thus the oscillator must include a frequency determining device which essentially is a bandpass filter allowing only the desired frequency to pass.
The output of the oscillator contains only a single sinusoidal frequency. In an RC oscillator the frequency of oscillations is determined by the values of resistance and
capacitance in the three sections. Variable resistors and capacitors are usually employed to provide tuning in the feedback network for variations in phase shift. For an RC phaseshift oscillator, the amplifier is biased for Class A operation to minimize distortion of the signal. approximated from the relation
itor. The feedback signal is coupled from the LC tank of the oscillator circuit by using a tickler coil or a coil pair.
In an Armstrong oscillator, the feedback is provided through a tickler coil. In a Hartley oscillator the feedback is provided through a coil pair.
1012
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Summary
In a Colpitts oscillator the feedback is provided through a capacitor pair. The Colpitts oscillator provides better frequency stability than the Armstrong and Hartley
oscillators. Moreover, the Colpitts oscillator is easier to tune and thus can be used for a wide range of frequencies.
In crystal oscillators the frequency determining element is a quartz crystal. Because of the
inherent characteristics of the quartz crystal the crystal oscillator may be held to extreme accuracy of frequency stability. The frequency depends almost entirely on the thickness where the thinner the thickness, the higher the frequency of oscillation. However, the power obtainable is limited by the heat the crystal will withstand without fracturing.
Crystal oscillators are used in applications where frequency accuracy and stability ar of utmost importance such as broadcast transmitters and radar. The frequency stability of crystalcon-
trolled oscillators depends on the quality factor Q. Besides the quartz crystal oscillators, other precision oscillators are constructed with cesium or rubidium.
The Pierce oscillator is a modified Colpitts oscillator that uses a crystal as a parallelresonant circuit, and for this reason is often referred to as crystalcontrolled Pierce oscillator.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
1013
v in vs
R2
v out
a. Derive an expression for the feedback factor b. Find an appropriate ratio of R f R 1 so that the closed loop gain will be A f = 10 c. Express the amount of feedback in dB d. Find the output voltage v out e. Find the feedback voltage v f f. Find the input voltage v in at the noninverting input. g. Determine the decrease in the closedloop gain A f if the openloop gain A decreases by 20% . 2. A sinusoidal oscillator consists of an amplifier with gain A = 10 and a bandpass filter whose center frequency is f 0 = 20 KHz . Determine a. The frequency 0 and the gain of the filter to produce sustained oscillations b. The gain of the filter at the frequency 0 3. The figure below shows a crystal oscillator and its equivalent circuit.
Electrode C2 Crystal Electrode C1 L R
Prove that
0P =
C1 + C2 ----------------LC 1 C 2
1014
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v in vs
R2
v out
a. The feedback factor is the portion of the output voltage v out which appears at the inverting input, and by the voltage division expression
R1 v ( ) = ----------------- v out R + R1 f R1 = ----------------Rf + R1
and thus
(1)
from which
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
1015
Sinusoidal Oscillators
d.
v out = A f v s = 10 1 = 10 V
g. Let the reduced open loop gain be denoted as A' and the corresponding closed loop gain as A' f . Then,
A' 0.8 10, 000 A' f = ----------------- = --------------------------------------------------------------- = 9.975 1 + A' 1 + 0.0999 0.8 10, 000
where the minus () sign implies decrease is percentage. 2. a. The frequency 0 at which sustained oscillations will be produced is the center frequency of the bandpass filter and thus
0 = 2f 0 = 2 20 10 = 1.26 10 r s
3 5
3.
Electrode C2 Crystal Electrode C1 L R
This is a parallel circuit and thus the resonant frequency occurs when the imaginary part of the admittance Y is equal to zero. Therefore,
1016
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
1017
Sinusoidal Oscillators
NOTES:
1018
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Appendix A
Introduction to MATLAB
his appendix serves as an introduction to the basic MATLAB commands and functions, procedures for naming and saving the user generated files, comment lines, access to MATLABs Editor / Debugger, finding the roots of a polynomial, and making plots. Several examples are provided with detailed explanations.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A1
A2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Roots of Polynomials
One of the most powerful features of MATLAB is the ability to do computations involving complex numbers. We can use either i , or j to denote the imaginary part of a complex number, such as 3-4i or 3-4j. For example, the statement z=34j displays z = 3.00004.0000i In the above example, a multiplication (*) sign between 4 and j was not necessary because the complex number consists of numerical constants. However, if the imaginary part is a function, or variable such as cos ( x ) , we must use the multiplication sign, that is, we must type cos(x)*j or j*cos(x) for the imaginary part of the complex number.
Solution: The roots are found with the following two statements where we have denoted the polynomial as p1, and the roots as roots_ p1.
p1=[1 10 35 50 24] % Specify and display the coefficients of p1(x) -50
p1 = 1
-10
35
24
roots_ p1=roots(p1)
roots_p1 = 4.0000 3.0000 2.0000 1.0000 Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A3
Solution: There is no cube term; therefore, we must enter zero as its coefficient. The roots are found with the statements below, where we have defined the polynomial as p2, and the roots of this polynomial as roots_ p2. The result indicates that this polynomial has three real roots, and two complex roots. Of course, complex roots always occur in complex conjugate* pairs.
p2=[1 7 0 16 25 52]
p2 = 1 -7 0 16 25 52
roots_ p2=roots(p2)
A4
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
r3 = 1 2 3 4
poly_r3=poly(r3) % Find the polynomial coefficients
poly_r3 = 1 -10
35
-50
24
We observe that these are the coefficients of the polynomial p 1 ( x ) of Example A.1. Example A.4 It is known that the roots of a polynomial are 1, 2, 3, 4 + j5, and 4 j5 . Find the coefficients of this polynomial. Solution: We form a row vector, say r4 , with the given roots, and we find the polynomial coefficients with the poly(r) function as shown below.
r4=[ 1 2 3 4+5j 45j ]
-4.0000+ 5.0000i
poly_r4 = 1 14
100
340
499
246
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A5
(A.1)
at x = 3 . Solution:
p5=[1 3 0 5 4 3 2]; % These are the coefficients of the given polynomial % The semicolon (;) after the right bracket suppresses the % display of the row vector that contains the coefficients of p5. % val_minus3=polyval(p5, 3) % Evaluate p5 at x=3; no semicolon is used here % because we want the answer to be displayed
val_minus3 = 1280 Other MATLAB functions used with polynomials are the following:
conv(a,b) multiplies two polynomials a and b [q,r]=deconv(c,d) divides polynomial c by polynomial d and displays the quotient q and remainder r. polyder(p) produces the coefficients of the derivative of a polynomial p.
and
p 2 = 2x 8x + 4x + 10x + 12
6 4 2
A6
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
p1p2 = 2 -6 Therefore,
-8
34
18
-24
11
-74
6x
5 10
-88
9
78
8
166
7
174
6
108
p 1 p 2 = 2x
and
p 4 = 2x 8x + 4x + 10x + 12
6 5 2
q = 0.5000 r = 0 Therefore,
q = 0.5 r = 4x 3x + 3x + 2x + 3
5 4 2
-3
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A7
der_p5 = 12 Therefore,
-32
10
(A.2)
where some of the terms in the numerator and/or denominator may be zero. We can find the roots of the numerator and denominator with the roots(p) function as before. As noted in the comment line of Example A.7, we can write MATLAB statements in one line, if we separate them by commas or semicolons. Commas will display the results whereas semicolons will suppress the display. Example A.9 Let
5 4 2 p num x 3x + 5x + 7x + 9 R ( x ) = ----------- = -------------------------------------------------------6 4 2 p den x 4x + 2x + 5x + 6
Express the numerator and denominator in factored form, using the roots(p) function. Solution:
num=[1 3 0 5 7 9]; den=[1 0 4 0 2 5 6]; roots_num=roots(num), roots_den=roots(den) % Do not display num and den coefficients % Display num and den roots
-1.1633
A8
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Rational Polynomials
roots_den = 1.6760 + 0.4922i -0.2108 + 0.9870i 1.6760 - 0.4922i -0.2108 - 0.9870i -1.9304 -1.0000
As expected, the complex roots occur in complex conjugate pairs. For the numerator, we have the factored form
p num = ( x 2.4186 j1.0712 ) ( x 2.4186 + j1.0712 ) ( x + 1.1633 ) ( x + 0.3370 j0.9961 ) ( x + 0.3370 + j0.9961 )
We can also express the numerator and denominator of this rational function as a combination of linear and quadratic factors. We recall that, in a quadratic equation of the form x 2 + bx + c = 0 whose roots are x 1 and x 2 , the negative sum of the roots is equal to the coefficient b of the x term, that is, ( x 1 + x 2 ) = b , while the product of the roots is equal to the constant term c , that is, x 1 x 2 = c . Accordingly, we form the coefficient b by addition of the complex conjugate roots and this is done by inspection; then we multiply the complex conjugate roots to obtain the constant term c using MATLAB as follows:
(2.4186 + 1.0712i)*(2.4186 1.0712i)
ans = 6.9971
(0.3370+ 0.9961i)*(0.33700.9961i)
ans = 1.1058
(1.6760+ 0.4922i)*(1.67600.4922i)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A9
ans = x^5-29999/10000*x^4-1323/3125000*x^3+7813277909/ 1562500000*x^2+1750276323053/250000000000*x+4500454743147/ 500000000000 and if we simplify this, we find that is the same as the numerator of the given rational expression in polynomial form. We can use the same procedure to verify the denominator.
R1
R2 C V L
A10
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Plot the magnitude of the impedance, that is, |Z| versus radian frequency . Solution: We cannot type (omega) in the MATLAB Command prompt, so we will use the English letter w instead. If a statement, or a row vector is too long to fit in one line, it can be continued to the next line by typing three or more periods, then pressing <enter> to start a new line, and continue to enter data. This is illustrated below for the data of w and z. Also, as mentioned before, we use the semicolon (;) to suppress the display of numbers that we do not care to see on the screen. The data are entered as follows:
w=[300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900.... 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000]; % z=[39.339 52.789 71.104 97.665 140.437 222.182 436.056.... 1014.938 469.830 266.032 187.052 145.751 120.353 103.111.... 90.603 81.088 73.588 67.513 62.481 58.240 54.611 51.468.... 48.717 46.286 44.122 42.182 40.432 38.845];
Of course, if we want to see the values of w or z or both, we simply type w or z, and we press Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A11
800 600
400 200
500
1000
1500
2000
2500
3000
This plot is referred to as the magnitude frequency response of the circuit. To return to the command window, we press any key, or from the Window pulldown menu, we select MATLAB Command Window. To see the graph again, we click on the Window pulldown menu, and we choose Figure 1. We can make the above, or any plot, more presentable with the following commands:
grid on: This command adds grid lines to the plot. The grid off command removes the grid. The
command grid toggles them, that is, changes from off to on or vice versa. The default* is off.
box off: This command removes the box (the solid lines which enclose the plot), and box on restores the box. The command box toggles them. The default is on. title(string): This command adds a line of the text string (label) at the top of the plot. xlabel(string) and ylabel(string) are used to label the x and yaxis respectively.
The magnitude frequency response is usually represented with the xaxis in a logarithmic scale. We can use the semilogx(x,y) command which is similar to the plot(x,y) command, except that the xaxis is represented as a log scale, and the yaxis as a linear scale. Likewise, the semilogy(x,y) command is similar to the plot(x,y) command, except that the yaxis is represented as a
* A default is a particular value for a variable that is assigned automatically by an operating system and remains in effect unless canceled or overridden by the operator.
A12
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
After execution of these commands, the plot is as shown in Figure A.3. If the yaxis represents power, voltage or current, the xaxis of the frequency response is more often shown in a logarithmic scale, and the yaxis in dB (decibels).
Magnitude of Impedance vs. Radian Frequency 1200 1000 800 |Z| in Ohms 600 400 200 0 2 10
10 w in rads/sec
10
To display the voltage v in a dB scale on the yaxis, we add the relation dB=20*log10(v), and we replace the semilogx(w,z) command with semilogx(w,dB). The command gtext(string)* switches to the current Figure Window, and displays a crosshair that can be moved around with the mouse. For instance, we can use the command gtext(Impedance |Z| versus Frequency), and this will place a crosshair in the Figure window. Then, using
* With the latest MATLAB Versions 6 and 7 (Student Editions 13 and 14), we can add text, lines and arrows directly into the graph using the tools provided on the Figure Window. For advanced MATLAB graphics, please refer to The MathWorks Using MATLAB Graphics documentation.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A13
This function specifies the number of data points but not the increments between data points. An alternate function is
x=first: increment: last
and this specifies the increments between points but not the number of data points. The script for the 3phase plot is as follows:
x=linspace(0, 2*pi, 60); % pi is a builtin function in MATLAB; % we could have used x=0:0.02*pi:2*pi or x = (0: 0.02: 2)*pi instead; y=sin(x); u=sin(x+2*pi/3); v=sin(x+4*pi/3); plot(x,y,x,u,x,v); % The xaxis must be specified for each function grid on, box on, % turn grid and axes box on text(0.75, 0.65, 'sin(x)'); text(2.85, 0.65, 'sin(x+2*pi/3)'); text(4.95, 0.65, 'sin(x+4*pi/3)')
These three waveforms are shown on the same plot of Figure A.4.
1
sin(x) 0.5
sin(x+2*pi/3)
sin(x+4*pi/3)
-0.5
-1
A14
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
.
o x + * s d
: .
< >
p h
For example, plot(x,y,'m*:') plots a magenta dotted line with a star at each data point, and plot(x,y,'rs') plots a red square at each data point, but does not draw any line because no line was selected. If we want to connect the data points with a solid line, we must type plot(x,y,'rs'). For additional information we can type help plot in MATLABs command screen. The plots we have discussed thus far are twodimensional, that is, they are drawn on two axes. MATLAB has also a threedimensional (threeaxes) capability and this is discussed next. The plot3(x,y,z) command plots a line in 3space through the points whose coordinates are the elements of x, y and z, where x, y and z are three vectors of the same length. The general format is plot3(x1,y1,z1,s1,x2,y2,z2,s2,x3,y3,z3,s3,...) where xn, yn and zn are vectors or matrices, and sn are strings specifying color, marker symbol, or line style. These strings are the same as those of the twodimensional plots.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A15
(A.3)
In a twodimensional plot, we can set the limits of the x and yaxes with the axis([xmin xmax ymin ymax]) command. Likewise, in a threedimensional plot we can set the limits of all three axes with the axis([xmin xmax ymin ymax zmin zmax]) command. It must be placed after the plot(x,y) or plot3(x,y,z) commands, or on the same line without first executing the plot command. This must be done for each plot. The threedimensional text(x,y,z,string) command will place string beginning at the coordinate (x,y,z) on the plot. For threedimensional plots, grid on and box off are the default states.
* This statement uses the so called dot multiplication, dot division, and dot exponentiation where the multiplication, division,
and exponential operators are preceded by a dot. These important operations will be explained in Section A.9.
A16
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(A.4)
Plot the volume of the cone as r and h vary on the intervals 0 r 4 and 0 h 6 meters. Solution: The volume of the cone is a function of both the radius r and the height h, that is,
V = f ( r, h )
The threedimensional plot is created with the following MATLAB script where, as in the previous example, in the second line we have used the dot multiplication, dot division, and dot exponentiation. This will be explained in Section A.9.
[R,H]=meshgrid(0: 4, 0: 6); % Creates R and H matrices from vectors r and h;... V=(pi .* R .^ 2 .* H) ./ 3; mesh(R, H, V);... xlabel('xaxis, radius r (meters)'); ylabel('yaxis, altitude h (meters)');... zlabel('zaxis, volume (cubic meters)'); title('Volume of Right Circular Cone'); box on
The threedimensional plot of Figure A.6 shows how the volume of the cone increases as the radius and height are increased. The plots of Figure A.5 and A.6 are rudimentary; MATLAB can generate very sophisticated threedimensional plots. The MATLAB Users Manual and the Using MATLAB Graphics Manual contain numerous examples.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A17
150
100
50
A.8 Subplots
MATLAB can display up to four windows of different plots on the Figure window using the command subplot(m,n,p). This command divides the window into an m n matrix of plotting areas and chooses the pth area to be active. No spaces or commas are required between the three integers m, n and p. The possible combinations are shown in Figure A.7. We will illustrate the use of the subplot(m,n,p) command following the discussion on multiplication, division and exponentiation that follows.
111 Full Screen 211 212 221 222 212 221 223 211 223 224 222 224 221 223 122 Default
121
121
A18
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
or as b=[1 3 6 11]' As shown below, MATLAB produces the same display with either format.
b=[1; 3; 6; 11]
b = -1 3 6 11 b=[1 3 6 11]' b = -1 3 6 11 We will now define Matrix Multiplication and ElementbyElement multiplication. 1. Matrix Multiplication (multiplication of row by column vectors) Let
A = [ a1 a2 a3 an ] % Observe the single quotation character ()
and
B = [ b 1 b 2 b 3 b n ]'
be two vectors. We observe that A is defined as a row vector whereas B is defined as a column vector, as indicated by the transpose operator (). Here, multiplication of the row vector A by the column vector B , is performed with the matrix multiplication operator (*). Then,
A*B = [ a 1 b 1 + a 2 b 2 + a 3 b 3 + + a n b n ] = sin gle value
(A.5)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A19
and
B = [ 2 6 3 8 7 ]'
the matrix multiplication A*B produces the single value 68, that is,
A B = 1 ( 2 ) + 2 6 + 3 ( 3 ) + 4 8 + 5 7 = 68
ans = 68 Now, let us suppose that both A and B are row vectors, and we attempt to perform a rowby row multiplication with the following MATLAB statements.
A=[1 2 3 4 5]; B=[2 6 3 8 7]; A*B % No transpose operator () here
When these statements are executed, MATLAB displays the following message: ??? Error using ==> * Inner matrix dimensions must agree. Here, because we have used the matrix multiplication operator (*) in A*B, MATLAB expects vector B to be a column vector, not a row vector. It recognizes that B is a row vector, and warns us that we cannot perform this multiplication using the matrix multiplication operator (*). Accordingly, we must perform this type of multiplication with a different operator. This operator is defined below. 2. ElementbyElement Multiplication (multiplication of a row vector by another row vector) Let
C = [ c1 c2 c3 cn ]
and
D = [ d1 d2 d3 dn ]
be two row vectors. Here, multiplication of the row vector C by the row vector D is performed with the dot multiplication operator (.*). There is no space between the dot and the multiplication symbol. Thus,
C. D = [ c 1 d 1 c2 d2 c3 d3 cn dn ]
(A.6)
A20
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
and
D = [ 2 6 3 8 7 ]
Dot multiplication of these two row vectors produce the following result.
C. D = 1 ( 2 ) 2 6 3 ( 3 ) 4 8 5 7 = 2 12 9 32 35
ans = -2
12
-9
32
35
Similarly, the division (/) and exponentiation (^) operators, are used for matrix division and exponentiation, whereas dot division (./) and dot exponentiation (.^) are used for element byelement division and exponentiation, as illustrated in Examples A.11 and A.12 above. We must remember that no space is allowed between the dot (.) and the multiplication, division, and exponentiation operators. Note: A dot (.) is never required with the plus (+) and minus () operators. Example A.13 Write the MATLAB script that produces a simple plot for the waveform defined as
y = f ( t ) = 3e
4 t
2
cos 5t 2e
3 t
t sin 2t + ---------t+1
(A.7)
in the 0 t 5 seconds interval. Solution: The MATLAB script for this example is as follows:
t=0: 0.01: 5; % Define taxis in 0.01 increments y=3 .* exp(4 .* t) .* cos(5 .* t)2 .* exp(3 .* t) .* sin(2 .* t) + t .^2 ./ (t+1); plot(t,y); grid; xlabel('t'); ylabel('y=f(t)'); title('Plot for Example A.13')
The plot for this example is shown in Figure A.8. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A21
2 1 0
-1
0.5
1.5
2.5 t
3.5
4.5
Had we, in this example, defined the time interval starting with a negative value equal to or less than 1 , say as 3 t 3 , MATLAB would have displayed the following message: Warning: Divide by zero. This is because the last term (the rational fraction) of the given expression, is divided by zero when t = 1 . To avoid division by zero, we use the special MATLAB function eps, which is a number approximately equal to 2.2 10
16
The command axis([xmin xmax ymin ymax]) scales the current plot to the values specified by the arguments xmin, xmax, ymin and ymax. There are no commas between these four arguments. This command must be placed after the plot command and must be repeated for each plot. The following example illustrates the use of the dot multiplication, division, and exponentiation, the eps number, the axis([xmin xmax ymin ymax]) command, and also MATLABs capability of displaying up to four windows of different plots. Example A.14 Plot the functions
y = sin 2x, z = cos 2x, w = sin 2x cos 2x, v = sin 2x cos 2x
in the interval 0 x 2 using 100 data points. Use the subplot command to display these functions on four windows on the same graph.
A22
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
w=y.* z; v=y./ (z+eps);% add eps to avoid division by zero subplot(221);% upper left of four subplots
plot(x,y); axis([0 2*pi 0 1]);
title('y=(sinx)^2');
title('z=(cosx)^2');
title('w=(sinx)^2*(cosx)^2');
subplot(224); plot(x,v); axis([0 2*pi 0 400]); % lower right of four subplots
0.5
0.5
2
2
4
2
2
2
4
2
w=(sinx) *(cosx)
The next example illustrates MATLABs capabilities with imaginary numbers. We will introduce the real(z) and imag(z) functions that display the real and imaginary parts of the complex quantity z = x + iy, the abs(z), and the angle(z) functions that compute the absolute value (magnitude) and phase angle of the complex quantity z = x + iy = r . We will also use the polar(theta,r) function that produces a plot in polar coordinates, where r is the magnitude, theta Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A23
Z ab
With the given values of resistance, inductance, and capacitance, the impedance Z ab as a function of the radian frequency can be computed from the following expression:
10 j ( 10 ) Z ab = Z = 10 + ------------------------------------------------------5 10 + j ( 0.1 10 )
4 6
(A.8)
a. Plot Re { Z } (the real part of the impedance Z) versus frequency . b. Plot Im { Z } (the imaginary part of the impedance Z) versus frequency . c. Plot the impedance Z versus frequency in polar coordinates. Solution: The MATLAB script below computes the real and imaginary parts of Z ab which, for simplicity, are denoted as z , and plots these as two separate graphs (parts a & b). It also produces a polar plot (part c).
w=0: 1: 2000; % Define interval with one radian interval;... z=(10+(10 .^ 4 j .* 10 .^ 6 ./ (w+eps)) ./ (10 + j .* (0.1 .* w 10.^5./ (w+eps))));... % % The first five statements (next two lines) compute and plot Re{z} real_part=real(z); plot(w,real_part);... xlabel('radian frequency w'); ylabel('Real part of Z'); grid
A24
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
200
400
600
1400
1600
1800
2000
Figure A.11. Plot for the real part of the impedance in Example A.15 % The next five statements (next two lines) compute and plot Im{z} imag_part=imag(z); plot(w,imag_part);... xlabel('radian frequency w'); ylabel('Imaginary part of Z'); grid
600 400 Imaginary part of Z 200
0 -200 -400
-600
200
400
600
1400
1600
1800
2000
Figure A.12. Plot for the imaginary part of the impedance in Example A.15 % The last six statements (next five lines) below produce the polar plot of z mag=abs(z); % Computes |Z|;... rndz=round(abs(z)); % Rounds |Z| to read polar plot easier;... theta=angle(z); % Computes the phase angle of impedance Z;... polar(theta,rndz); % Angle is the first argument ylabel('Polar Plot of Z'); grid
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A25
90 120
1500 60 1000
30
180
210
330
240 270
300
Example A.15 clearly illustrates how powerful, fast, accurate, and flexible MATLAB is.
is a function file and must be saved as myfunction.m For the next example, we will use the following MATLAB functions:
fzero(f,x) attempts to find a zero of a function of one variable, where f is a string containing the name of a realvalued function of a single real variable. MATLAB searches for a value near a point where the function f changes sign, and returns that value, or returns NaN if the search fails.
A26
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Example A.16 Find the zeros, the minimum, and the maximum values of the function
1 1 f ( x ) = --------------------------------------- --------------------------------------- 10 2 2 ( x 0.1 ) + 0.01 ( x 1.2 ) + 0.04
(A.9)
in the interval 1.5 x 1.5 Solution: We first plot this function to observe the approximate zeros, maxima, and minima using the following script.
x=1.5: 0.01: 1.5; y=1./ ((x0.1).^ 2 + 0.01) 1./ ((x1.2).^ 2 + 0.04) 10; plot(x,y); grid
-1
-0.5
0.5
1.5
Figure A.14. Plot for Example A.16 using the plot command
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A27
To save this file, from the File drop menu on the Command Window, we choose New, and when the Editor Window appears, we type the script above and we save it as funczero01. MATLAB appends the extension .m to it. Now, we can use the fplot(fcn,lims) command to plot f ( x ) as follows:
fplot('funczero01', [1.5 1.5]); grid
This plot is shown in Figure A.15. As expected, this plot is identical to the plot of Figure A.14 which was obtained with the plot(x,y) command as shown in Figure A.14.
100 80 60 40 20 0 -20 -40 -1.5
-1
-0.5
0.5
1.5
Figure A.15. Plot for Example A.16 using the fplot command
We will use the fzero(f,x) function to compute the roots of f ( x ) in Equation (A.9) more precisely. The MATLAB script below will accomplish this.
x1= fzero('funczero01', 0.2); x2= fzero('funczero01', 0.3); fprintf('The roots (zeros) of this function are r1= %3.4f', x1); fprintf(' and r2= %3.4f \n', x2)
A28
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
is executed, MATLAB displays a very long expression which when copied at the command prompt and executed, produces the following: ans = 0.6585 + 0.3437i ans = 0.6585 - 0.3437i ans = 1.2012 The real value 1.2012 above is the value of x at which the function y has its minimum value as we observe also in the plot of Figure A.15. To find the value of y corresponding to this value of x, we substitute it into f ( x ) , that is,
x=1.2012; ymin=1 / ((x0.1) ^ 2 + 0.01) 1 / ((x1.2) ^ 2 + 0.04) 10
ymin = -34.1812 We can find the maximum value from f ( x ) whose plot is produced with the script x=1.5:0.01:1.5; ymax=1./((x0.1).^2+0.01)+1./((x1.2).^2+0.04)+10; plot(x,ymax); grid and the plot is shown in Figure A.16. Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A29
-1
-0.5
0.5
1.5
Next we compute the first derivative of f ( x ) and we solve for x to find the value where the maximum of ymax occurs. This is accomplished with the MATLAB script below.
syms x ymax zmax; ymax=(1/((x0.1)^2+0.01)1/((x1.2)^2+0.04)10); zmax=diff(ymax)
zmax = 1/((x-1/10)^2+1/100)^2*(2*x-1/5)-1/((x-6/5)^2+1/25)^2*(2*x-12/5)
solve(zmax)
is executed, MATLAB displays a very long expression which when copied at the command prompt and executed, produces the following: ans = 0.6585 + 0.3437i ans = 0.6585 - 0.3437i ans = 1.2012 ans = 0.0999 From the values above we choose x = 0.0999 which is consistent with the plots of Figures A.15 and A.16. Accordingly, we execute the following script to obtain the value of ymin .
A30
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Display Formats
x=0.0999; % Using this value find the corresponding value of ymax ymax=1 / ((x0.1) ^ 2 + 0.01) 1 / ((x1.2) ^ 2 + 0.04) 10
ymax = 89.2000
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
A31
The disp(X) command displays the array X without printing the array name. If X is a string, the text is displayed. The fprintf(format,array) command displays and prints both text and arrays. It uses specifiers to indicate where and in which format the values would be displayed and printed. Thus, if %f is used, the values will be displayed and printed in fixed decimal format, and if %e is used, the values will be displayed and printed in scientific notation format. With this command only the real part of each parameter is processed. This appendix is just an introduction to MATLAB.* This outstanding software package consists of many applications known as Toolboxes. The MATLAB Student Version contains just a few of these Toolboxes. Others can be bought directly from The MathWorks, Inc., as addons.
* For more MATLAB applications, please refer to Numerical Analysis Using MATLAB and Spreadsheets, ISBN 0970951116.
A32
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Appendix B
Introduction to Simulink
his appendix is a brief introduction to Simulink. This author feels that we can best introduce Simulink with a few examples. Some familiarity with MATLAB is essential in understanding Simulink, and for this purpose, Appendix A is included as an introduction to MATLAB.
+
vC ( t )
+
i(t) vs ( t ) = u0 ( t )
43 F
(B.1)
(B.2)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
B1
Introduction to Simulink
d vC dv C RC -------- + LC ---------- + v C = u 0 ( t ) 2 dt dt
2
(B.3)
(B.4)
t>0
(B.5)
To appreciate Simulinks capabilities, for comparison, three different methods of obtaining the solution are presented, and the solution using Simulink follows. First Method Assumed Solution Equation (B.5) is a secondorder, nonhomogeneous differential equation with constant coefficients, and thus the complete solution will consist of the sum of the forced response and the natural response. It is obvious that the solution of this equation cannot be a constant since the derivatives of a constant are zero and thus the equation is not satisfied. Also, the solution cannot contain sinusoidal functions (sine and cosine) since the derivatives of these are also sinusoids. However, decaying exponentials of the form ke where k and a are constants, are possible candidates since their derivatives have the same form but alternate in sign. It can be shown* that if k 1 e 1 and k 2 e 2 where k 1 and k 2 are constants and s 1 and s 2 are the roots of the characteristic equation of the homogeneous part of the given differential equation, the natural response is the sum of the terms k 1 e be
s1 t s t s t at
and k 2 e
s2 t
+ k2 e
+ v cf ( t )
(B.6)
* Please refer to Circuit Analysis II with MATLAB Applications, ISBN 9780970951151, Appendix B for a thorough discussion.
B2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(B.7)
Solution of (B.7) yields of s 1 = 1 and s 2 = 3 and with these values (B.6) is written as
vc ( t ) = k1 e + k2 e
t 3 t
+ v cf ( t )
(B.8)
t>0
(B.9)
Since the right side of (B.9) is a constant, the forced response will also be a constant and we denote it as v Cf = k 3 . By substitution into (B.9) we obtain
0 + 0 + 3k 3 = 3
or
v Cf = k 3 = 1
(B.10)
+1
(B.11)
The constants k 1 and k 2 will be evaluated from the initial conditions. First, using v C ( 0 ) = 0.5 V and evaluating (B.11) at t = 0 , we obtain
v C ( 0 ) = k 1 e + k 2 e + 1 = 0.5 k 1 + k 2 = 0.5
0 0
(B.12)
Also, and
(B.13)
t=0
(B.14)
By equating the right sides of (B.13) and (B.14) we obtain Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
B3
Introduction to Simulink
k 1 3k 2 = 0
(B.15)
Simultaneous solution of (B.12) and (B.15), gives k 1 = 0.75 and k 2 = 0.25 . By substitution into (B.8), we obtain the total solution as
v C ( t ) = ( 0.75 e + 0.25e
t 3 t
+ 1 )u 0 ( t )
(B.16)
y1 = 3/4*exp(-t)-3/4*exp(-3*t)
y2=diff(y0,2) % The second derivative of y(t)
y2 = -3/4*exp(-t)+9/4*exp(-3*t)
y=y2+4*y1+3*y0 % Summation of y and its derivatives
y = 3 Thus, the solution has been verified by MATLAB. Using the expression for v C ( t ) in (B.16), we find the expression for the current as
dv C 4 3 t 3t t 3t - - i = i L = i C = C --------- = -- -- e 3 e = e e A - dt 3 4 4
(B.17)
Second Method Using the Laplace Transformation The transformed circuit is shown in Figure B.2.
R 1 L
0.25s C 3 4s
Vs ( s ) = 1 s
VC ( s )
I(s) 0.5 s
+ V (0) C
B4
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(B.18)
= 1
s=0
= 0.75
s = 1
= 0.25
s = 3
* For derivation of the voltage division and current division expressions, please refer to Circuit Analysis I with MATLAB Applications, ISBN 9780970951120. Partial fraction expansion is discussed in Chapter 3, Signals and Systems with MATLAB Computing and Simulink Modeling, ISBN 9781934404119 For an introduction to Laplace Transform and Inverse Laplace Transform, please refer Chapters 2 and 3, Signals and Systems with MATLAB Computing and Simulink Modeling, ISBN 9781934404119 ** Usually, in StateSpace and State Variables Analysis, u ( t ) denotes any input. For distinction, we will denote the Unit Step Function as u 0 ( t ) . Please refer to Chapter 5, Signals and Systems with MATLAB Computing and Simulink Modeling, ISBN 9781934404119
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
B5
Introduction to Simulink
By substitution of given values and rearranging, we obtain
di 1 ------- L = ( 1 )i L v C + 1 - 4 dt
or
di L ------ = 4i L 4v C + 4 dt
(B.19)
(B.20)
and
(B.21)
dv C i L = C -------dt dv C 4 x 1 = i L = C -------- = Cx 2 = -- x 2 3 dt 3 x 2 = -- x 1 4
(B.22)
Therefore, from (B.19), (B.20), and (B.22), we obtain the state equations
x 1 = 4x 1 4x 2 + 4 3 x 2 = -- x 1 4
(B.23)
* The notation x (x dot) is often used to denote the first derivative of the function x , that is, x = dx dt .
The detailed solution of (B.23) is given in Chapter 5, Example 5.10, Page 523, Signals and Systems with MATLAB Computing and Simulink Modeling, ISBN 9781934404119
B6
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
1 0.75 e + 0.25e
t 3t
Then,
x1 = iL = e e
(B.24)
3t
and
x 2 = v C = 1 0.75e + 0.25e
t
(B.25)
Modeling the Differential Equation of Example B.1 with Simulink To run Simulink, we must first invoke MATLAB. Make sure that Simulink is installed in your system. In the MATLAB Command prompt, we type:
simulink
Alternately, we can click on the Simulink icon shown in Figure B.3. It appears on the top bar on MATLABs Command prompt.
Upon execution of the Simulink command, the Commonly Used Blocks appear as shown in Figure B.4. In Figure B.4, the left side is referred to as the Tree Pane and displays all Simulink libraries installed. The right side is referred to as the Contents Pane and displays the blocks that reside in the library currently selected in the Tree Pane. Let us express the differential equation of Example B.1 as
2 dv C d vC ---------- = 4 -------- 3v C + 3u 0 ( t ) 2 dt dt
(B.26)
A block diagram representing relation (B.26) above is shown in Figure B.5. We will use Simulink to draw a similar block diagram.*
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
B7
Introduction to Simulink
u0 ( t )
dt
4
dv C -------dt
dt
vC
To model the differential equation (B.26) using Simulink, we perform the following steps: 1. On the Simulink Library Browser, we click on the leftmost icon shown as a blank page on the top title bar. A new model window named untitled will appear as shown in Figure B.6.
B8
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
The window of Figure B.6 is the model window where we enter our blocks to form a block diagram. We save this as model file name Equation_1_26. This is done from the File drop menu of Figure B.6 where we choose Save as and name the file as Equation_1_26. Simulink will add the extension .mdl. The new model window will now be shown as Equation_1_26, and all saved files will have this appearance. See Figure B.7.
2. With the Equation_1_26 model window and the Simulink Library Browser both visible, we click on the Sources appearing on the left side list, and on the right side we scroll down until we see the unit step function shown as Step. See Figure B.8. We select it, and we drag it into the Equation_1_26 model window which now appears as shown in Figure B.8. We save file Equation_1_26 using the File drop menu on the Equation_1_26 model window (right side of Figure B.8). 3. With reference to block diagram of Figure B.5, we observe that we need to connect an amplifier with Gain 3 to the unit step function block. The gain block in Simulink is under Commonly Used Blocks (first item under Simulink on the Simulink Library Browser). See Figure B.8. If the Equation_1_26 model window is no longer visible, it can be recalled by clicking on the white page icon on the top bar of the Simulink Library Browser. 4. We choose the gain block and we drag it to the right of the unit step function. The triangle on the right side of the unit step function block and the > symbols on the left and right sides of the gain block are connection points. We point the mouse close to the connection point of the unit step function until is shows as a cross hair, and draw a straight line to connect the two Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
B9
Introduction to Simulink
blocks.* We doubleclick on the gain block and on the Function Block Parameters, we change the gain from 1 to 3. See Figure B.9.
Figure B.8. Dragging the unit step function into File Equation_1_26
Figure B.9. File Equation_1_26 with added Step and Gain blocks * An easy method to interconnect two Simulink blocks by clicking on the source block to select it, then hold down the Ctrl key and leftclick on the destination block.
B10
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
6. From the Commonly Used Blocks of the Simulink Library Browser, we choose the Integrator block, we drag it into the Equation_1_26 model window, and we connect it to the output of the Add block. We repeat this step and to add a second Integrator block. We click on the text Integrator under the first integrator block, and we change it to Integrator 1. Then, we change the text Integrator 1 under the second Integrator to Integrator 2 as shown in Figure B.11.
7. To complete the block diagram, we add the Scope block which is found in the Commonly Used Blocks on the Simulink Library Browser, we click on the Gain block, and we copy and paste it twice. We flip the pasted Gain blocks by using the Flip Block command from the Format drop menu, and we label these as Gain 2 and Gain 3. Finally, we doubleclick on these gain blocks and on the Function Block Parameters window, we change the gains from to 4 and 3 as shown in Figure B.12.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
B11
Introduction to Simulink
8. The initial conditions i L ( 0 ) = C ( dv C dt )
t=0
ble clicking the Integrator blocks and entering the values 0 for the first integrator, and 0.5 for the second integrator. We also need to specify the simulation time. This is done by specifying the simulation time to be 10 seconds on the Configuration Parameters from the Simulation drop menu. We can start the simulation on Start from the Simulation drop menu or by clicking on the icon.
9. To see the output waveform, we double click on the Scope block, and then clicking on the Autoscale icon, we obtain the waveform shown in Figure B.13.
Figure B.13. The waveform for the function v C ( t ) for Example B.1
Another easier method to obtain and display the output v C ( t ) for Example B.1, is to use State Space block from Continuous in the Simulink Library Browser, as shown in Figure B.14.
Figure B.14. Obtaining the function v C ( t ) for Example B.1 with the StateSpace block.
B12
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
y = Cx + du y = [0 1] x1 x2 + [ 0 ]u
We doubleclick on the StateSpace block, and in the Functions Block Parameters window we enter the constants shown in Figure B.15.
Figure B.15. The Function block parameters for the StateSpace block.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
B13
Introduction to Simulink
The initials conditions [ x1 x2 ]' are specified in MATLABs Command prompt as
x1=0; x2=0.5;
form, we double click on the Scope block, and then clicking on the Autoscale obtain the waveform shown in Figure B.16.
Figure B.16. The waveform for the function v C ( t ) for Example B.1 with the StateSpace block.
The statespace block is the best choice when we need to display the output waveform of three or more variables as illustrated by the following example. Example B.2 A fourthorder network is described by the differential equation
4 d y + a d y + a d y + a dy + a y ( t ) = u ( t ) --------3 -------2 ------1 ----0 3 2 4 dt dt dt dt 3 2
(B.27)
where y ( t ) is the output representing the voltage or current of the network, and u ( t ) is any input, and the initial conditions are y ( 0 ) = y' ( 0 ) = y'' ( 0 ) = y''' ( 0 ) = 0 . a. We will express (B.27) as a set of state equations
B14
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(B.28)
subject to the initial conditions y ( 0 ) = y' ( 0 ) = y'' ( 0 ) = y''' ( 0 ) = 0 , has the solution
y ( t ) = 0.125 [ ( 3 t ) 3t cos t ]
2
(B.29)
In our set of state equations, we will select appropriate values for the coefficients a 3, a 2, a 1, and a 0 so that the new set of the state equations will represent the differential equation of (B.28), and using Simulink, we will display the waveform of the output y ( t ) . 1. The differential equation of (B.28) is of fourthorder; therefore, we must define four state variables that will be used with the four firstorder state equations. We denote the state variables as x 1, x 2, x 3 , and x 4 , and we relate them to the terms of the given differential equation as
x1 = y ( t ) dy x 2 = ----dt d y x 3 = -------2 dt
2
d y x 4 = -------3 dt
(B.30)
We observe that
x1 = x2 x2 = x3 x3 = x4 d y --------- = x 4 = a 0 x 1 a 1 x 2 a 2 x 3 a 3 x 4 + u ( t ) 4 dt
4
(B.31)
(B.32)
(B.33) (B.34)
y = Cx + du
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
B15
Introduction to Simulink
x1 x2 x3 x4 0 0 A= 0 a0 1 0 0 a1 0 1 0 a2 0 0 , 1 a3 x1 x= x2 x3 x4 y ( t ) = x1 , 0 b= 0, 0 1
x=
and u = u ( t )
(B.35)
(B.36)
2. By inspection, the differential equation of (B.27) will be reduced to the differential equation of (B.28) if we let
a3 = 0 a2 = 2 a1 = 0 a0 = 1 u ( t ) = sin t
and thus the differential equation of (B.28) can be expressed in statespace form as
x1 x2 x3 x4 0 0 = 0 a0 1 0 0 0 0 1 0 2 0 0 1 0 0 + 0 sin t 0 x3 1 x4 x2 x1
(B.37)
where
x1 x2 x3 x4 0 0 A= 0 a0 1 0 0 0 0 1 0 2 0 0 , 1 0 x1 x= x2 x3 x4 y ( t ) = x1 , 0 b= 0, 0 1
x=
and u = sin t
(B.38)
B16
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(B.39)
We invoke MATLAB, we start Simulink by clicking on the Simulink icon, on the Simulink Library Browser we click on the Create a new model (blank page icon on the left of the top bar), and we save this model as Example_1_2. On the Simulink Library Browser we select Sources, we drag the Signal Generator block on the Example_1_2 model window, we click and drag the StateSpace block from the Continuous on Simulink Library Browser, and we click and drag the Scope block from the Commonly Used Blocks on the Simulink Library Browser. We also add the Display block found under Sinks on the Simulink Library Browser. We connect these four blocks and the complete block diagram is as shown in Figure B.17.
We now doubleclick on the Signal Generator block and we enter the following in the Function Block Parameters: Wave form: sine Time (t): Use simulation time Amplitude: 1 Frequency: 2 Units: Hertz Next, we doubleclick on the statespace block and we enter the following parameter values in the Function Block Parameters:
A: [0 1 0 0; 0 0 1 0; 0 0 0 1; a0 a1 a2 a3] B: [0 0 0 1] C: [1 0 0 0] D: [0]
Initial conditions: x0 Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
B17
Introduction to Simulink
Absolute tolerance: auto Now, we switch to the MATLAB Command prompt and we type the following: >> a0=1; a1=0; a2=2; a3=0; x0=[0 0 0 0]; We change the Simulation Stop time to 25 , and we start the simulation by clicking on the icon. To see the output waveform, we double click on the Scope block, then clicking on the Autoscale icon, we obtain the waveform shown in Figure B.18.
The Display block in Figure B.17 shows the value at the end of the simulation stop time. Examples B.1 and B.2 have clearly illustrated that the StateSpace is indeed a powerful block. We could have obtained the solution of Example B.2 using four Integrator blocks by this approach would have been more time consuming. Example B.3 Using Algebraic Constraint blocks found in the Math Operations library, Display blocks found in the Sinks library, and Gain blocks found in the Commonly Used Blocks library, we will create a model that will produce the simultaneous solution of three equations with three unknowns. The model will display the values for the unknowns z 1 , z 2 , and z 3 in the system of the equations
a1 z1 + a2 z2 + a3 z3 + k1 = 0 a4 z1 + a5 z2 + a6 z3 + k2 = 0 a7 z1 + a8 z2 + a9 z3 + k3 = 0
(B.40)
B18
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
After clicking on the simulation icon, we observe the values of the unknowns as z 1 = 2 ,
z 2 = 3 , and z 3 = 5 .These values are shown in the Display blocks of Figure B.19.
The Algebraic Constraint block constrains the input signal f ( z ) to zero and outputs an algebraic state z . The block outputs the value necessary to produce a zero at the input. The output must affect the input through some feedback path. This enables us to specify algebraic equations for index 1 differential/algebraic systems (DAEs). By default, the Initial guess parameter is zero. We can improve the efficiency of the algebraic loop solver by providing an Initial guess for the algebraic state z that is close to the solution value.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
B19
Introduction to Simulink
An outstanding feature in Simulink is the representation of a large model consisting of many blocks and lines, to be shown as a single Subsystem block.* For instance, we can group all blocks and lines in the model of Figure B.19 except the display blocks, we choose Create Subsystem from the Edit menu, and this model will be shown as in Figure B.20 where in MATLABs Command prompt we have entered:
a1=5; a2=1; a3=4; a4=11; a5=6; a6=9; a7=8; a8=4; a9=15;... k1=14; k2=6; k3=9;
The Display blocks in Figure B.20 show the values of z 1 , z 2 , and z 3 for the values specified in MATLABs Command prompt.
* The Subsystem block is described in detail in Chapter 2, Section 2.1, Page 22, Introduction to Simulink with Engineering Applications, ISBN 9781934404096. The contents of the Subsystem block are not lost. We can doubleclick on the Subsystem block to see its contents. The Subsystem block replaces the inputs and outputs of the model with Inport and Outport blocks. These blocks are described in Section 2.1, Chapter 2, Page 22, Introduction to Simulink with Engineering Applications, 9781934404096.
B20
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Appendix C
ProportionalIntegralDerivative (PID) Controller
his appendix is a brief introduction to the ProportionalIntegralDerivative Controller, briefly known as PID. The components and the functions of a typical PID are described, and two examples are presented using the Simulink PID Controller Block and the Simulink PID Controller with Approximate Derivative Block.
The Proportional value determines the reaction to the current error, the Integral determines the reaction based on the sum of recent errors, and the Derivative determines the reaction to the rate at which the error has been changing. The weighted sum of these three actions is used to adjust the process via a control element such as the position of a control valve or the power supply of a heating element. By adjusting the three constants in the PID controller algorithm the PID can provide control action designed for specific process requirements. The response of the controller can be described
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
C 1
Proportional
v in
Integral
v out
Derivative
C2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Figure C.4. Output waveforms for the model in Figure C.2 without and with PID control
The Simulink PID controller with approximate derivative calculation contains the functional blocks shown in Figure C.5.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
C3
Figure C.7. Output waveforms for the model in Figure C.6 without and with PID control
The reader should try to minimize the oscillations shown in Figure C.7 by assigning other values for the Derivative divisor. The examples presented in this appendix indicate that the Proportional Gain Kp , Integral Gain Ki , and Derivative Gain Kd are dependent on each another. Generally, Kp and Ki will decrease the rise time and increase the overshoot, and Kd will cause a small change in rise time and decrease the overshoot.
C4
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Appendix D
Compensated Attenuators
his appendix begins with a simple resistive attenuator which may be used to reduce the amplitude of a signal waveform. We will examine the conditions under which it is possible to ensure no distortion even if shunt capacitances are taken into consideration. Also, we will investigate the types of responses which are obtained with a step voltage input if the circuit is improperly adjusted.
v in
R1
R2 (a)
C2
v out
av in
C2 (b)
v out
The presence of the stray capacitance C 2 in Figure D.1(a) represents an unavoidable condition since the output of the resistive attenuator in most cases is followed by the input capacitance of a stage of amplification. Using Thevenins theorem, we can replace the circuit of (a) with that of (b) where R represents the parallel combination of R 1 and R 2 and a in av in is the attenuation factor. We could make both R 1 and R 2 very large so that the input impedance of the attenuator would be large enough to prevent loading down the input signal but his may produce a large rise time which would, in most cases, be unacceptable. This is explained below. The rise time, denoted as t r , is defined as the time it takes the voltage to rise from 10% to 90% of its final value. It is an indication of how fast a circuit can respond to a discontinuity in voltage. In an RC network the time required for the output voltage v out to reach 10% of its final value is
0.1RC , and the time required for the output voltage v out to reach 90% of its final value is 2.3RC . The rise time t r is the difference between these two values and in terms of the RC time
constant it is given by
2.20.35 t r = 2.2 = 2.2RC = ---------- = --------2f c fc
(D.1)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
D1
Compensated Attenuators
where f c is the 3-dB frequency given by f c = 1 2RC . It was stated above that we could make the input impedance of the attenuator large enough to prevent loading of the input signal. If, for example, we form an attenuator with R 1 = R 2 = 1 M and the input capacitance of a stage amplification is C 2 = 15 pF . With these values the rise time for the circuit of Figure D.1(b) will be
t r = 2.2 = 2.2RC = 2.2 0.5 10 15 10
6 12
= 16.5 sec
This is a very large rise time for a practical application and thus it is unacceptable.
C1
v in v in
R1 R2 (a) C2
x
R2 (b) C2
y v out
v out
The circuit of Figure D.2(a) is shown as in Figure D.2(b) to form the four arms of a bridge. The bridge will be balanced if R1 C1 = R2 C2 (D.2) and under this condition there will be no current flow in the branch connecting point x to point y , so when computing the output voltage v out this branch may be omitted and the output will be equal to av in and independent of frequency. Because relation (D.2) must hold precisely, it is practical to make capacitor C 1 variable and the final adjustment for perfect compensation may be made experimentally with the testing of a square-wave. Figure D.3 shows the appearances of the output signal when the input is a step voltage of amplitude V if the compensation is not precise. In Figure D.3(a) the circuit is overcompensated and in Figure D.3(b) is undercompensated.
D2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Compensated Attenuator
Step input, amplitude V Output v out Perfect compensation Step input, amplitude V Perfect compensation Output v out
v out ( 0+ )
v out ( ) = aV
v out ( 0+ )
v out ( ) = aV
(a)
(b)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
D3
Compensated Attenuators
NOTES:
D4
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Appendix E
Substitution, Reduction, and Millers Theorems
his appendix discusses three additional theorems that are especially useful in the simplification of circuits containing dependent sources. In our previous studies* we discussed the superposition principle and Thevenins and Nortons theorems.
x
3A 1
x
3
x
XC = 8 3 V ' = 10 53.1 V
x y
(a)
v xy = 6 V i xy = 3 A
6V
y (c)
3V
y (d)
3V
(b)
y (e)
For the simple resistor circuit of Figure E.1(a) we find by seriesparallel resistance combinations that v xy = 6 V and i xy = 3 A . According to the substitution theorem, the 2 resistor across terminals x and y can be replaced with a source with a 6 V source as shown in Figure E.1(b) and the rest of the network will be unaffected. The current in the branch will be 3 A as before. Other substitutions are possible also. For instance, the substitute branch may consist of a resistance of 1 and a voltage source of 3 V as shown in Figure E.1(c), or it may consist of a resistance of 3 and a voltage source of 3 V of opposite polarity as shown in Figure E.1(d).
* For a detailed discussion of Thevenins and Nortons theorems and the superposition principle please refer to Circuit Analysis I with MATLAB Applications, ISBN 9780970951120.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
E1
(E.1)
With the value X C = 8 3 , the value of the AC voltage source V' necessary to satisfy the substitution theorem must be such V xy = 6 V and I xy = 3 A and by substitution of these values into relation (E.1) we obtain
6 = j ( 8 3 ) 3 + V'
V' = 6 + j8 = 10 53.1
Proof of the substitution theorem is based on the branch equations of the network. Operation of the whole network of B branches is defined by 2B simultaneous equations. Consider, for example, the bridge network shown in Figure E.2.
Zs
a
Z1 Z2
c
Vs Z3 Z5
d
Z4
The network of Figure E.2 has six branches and by application of Ohms law,
V ab = Z ab I ab + Vs V cb = Z cb I cb V ac = Z ac I ac V cd = Z cd I cd V ad = Z ad I ad V db = Z db I db
(E.2)
If we assume that the source voltage V S and six impedances are known constants, it becomes apparent that we have 6 equations with 12 unknowns, and thus we need 6 more equations to solve for all voltages and currents in each branch. We can obtain 3 additional equations by application of KCL, and the last three from KVL. These are shown below.
I ab + I ac + I ad = 0 V ab + V ac + V cb = 0 I ca + I cd + I cb = 0 V ac + V cd + V da = 0 I da + I dc + I db = 0 V cb + V bd + V dc = 0
(E.3)
The 12 equations of (E.2) and (E.3) can now be solved simultaneously for the 12 unknowns.
E 2
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
(E.4)
that is, the voltage across this branch is V ab and the current is I ab . Other branch voltages in the network are V bc , V cd , and so on, and currents are I bc , I cd , and so on, and all these voltages and currents simultaneously satisfy the 2B equations of the network, including equation (E.4). Next, let us assume a substitute branch that has some different value of impedance Z' ab , and some different voltage source V' S , but with Z' ab and V' S so related that the same values of V ab and I ab that satisfied equation (E.4) will also satisfy the new branch equation
V ab = Z' ab I ab + V' S
(E.5)
When this substitution is made, operation of the altered network will be defined by 2B equations, as before; (2B 1) of the equations will remain the same but equation (E.5) will have replaced equation (E.4). However, since equation (E.5) is satisfied by the same values of voltage and current that satisfied equation (E.4), it follows that the whole group of simultaneous equations will be satisfied by the same voltages and currents as before. Voltages and currents of all branches of the network will therefore be the same as before substitution, which was to be proved. The substitution theorem is especially useful in the simplification of circuits containing dependent sources. Consider for example the currentcontrolled voltage source and the voltagecontrolled current source shown in Figure E.3.
Rest of the network (a)
i
Ri
Gv
Figure E.3. Application of the substitution theorem in current and voltagecontrolled dependent sources.
The network of Figure E.3(a) can be any network having the terminal current i , and Ri is the voltage of a currentcontrolled voltage source. Figure E.3(b) shows the conditions under which the dual form can be applied. In this case the voltagecontrolled current source Gv , which is connected across the voltage v . As another example, let us consider the equivalent circuit of some amplifier shown in Figure E.4(a).
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
E3
v in
R3
v in
R1
R2
i (a)
R4
v out
v in
R1
( 1 + ) R2 (b)
R4
v out
In Figure E.4(a), the currents and voltages are not changed if the dependent source R 2 i is replaced by a resistance R 2 , and the simplified equivalent circuit is as shown in Figure E.4(b). A special case of the substitution theorem is the source absorption theorem* and has two dual forms, the voltage source absorption and the current source absorption theorems. The voltage source absorption theorem is illustrated in Figure E.5.
i
Zi
(a)
(b)
As shown in Figure E.5, the voltage source absorption states that if in a branch of the network in which the current through it is i , the voltage source Zi can be replaced by an impedance Z . Example E.1 Use the voltage source absorption theorem to simplify the branch AB shown in Figure E.6.
A
aZi
Z B
* This theorem is analogous to the absorption theorem in Boolean algebra which states that A = ( A + B ) = AB
E 4
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
aZi
Z (a) B A
( a + 1 )Z Z (b)
aZ
(c)
Figure E.7. Steps for simplification of the branch for Example E.1
v = v AB (a)
v = v AB (b)
As shown in Figure E.8, the current source absorption states that if in a branch of the network in which the voltage across it is v , the current source Yv can be replaced by an admittance Y . Example E.2 Use the current source absorption theorem to simplify the parallel combination shown in Figure E.9.
a Y v AB
A B
Solution: Application of the current source absorption theorem yields the simplified branch shown in Figure E.10(c) below.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
E5
aY
Y
B A
( a + 1 )Y B
(a)
(b)
(c)
Figure E.10. Steps for simplification of the parallel combination of Example E.2
v1
A v1
A i1
(a)
(b)
Figure E.11. Network configurations that can be simplified by the reduction theorem
The reduction theorem applies to the network configurations shown in Figure E.11. The network N 1 in Figure E.11(a) is any linear twoterminal network having the terminal voltage v 1 , and N 2 is any other linear twoterminal network. The constant A of the voltagecontrolled voltage source Av 1 is any real number, positive or negative. The two networks and the voltage source are connected in series so that the voltages v 1 and Av 1 are additive with respect to the loop that they form. The theorem states that all currents in N 1 and N 2 remain unchanged if the source Av 1 is replaced with a short circuit and if: 1. Each resistance, inductance, reciprocal capacitance, and voltage source in Network N 1 is multiplied by 1 + A or 2. Each resistance, inductance, reciprocal capacitance, and voltage source in Network N 2 is divided by 1 + A . Proof of the reduction theorem is based on the form of the loop equations. Obviously, if all resistances, inductances, reciprocal capacitances, and voltage sources are multiplied by the same arbi-
E 6
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Part 2 of the theorem is proved. When all currents and voltages in the network are sinusoidal and of the same frequency, the constant of the controlled source may be expressed as a complex number; the reduction theorem holds in this case also. The dual form of the reduction theorem is applicable to the network configuration shown in Figure E.11(b). It states that all voltages in the network of Figure E.11(b) remain unchanged if the current source Ai 1 is replaced with an open circuit and if: 1. All conductances, capacitances, reciprocal inductances, and current sources in N 1 are multiplied by 1 + A or 2. All conductances, capacitances, reciprocal inductances, and current sources in N 2 are divided by 1 + A Proof of the dual form of the theorem follows from nodal equations and is dual to the proof outlined above. In applying the reduction theorem it is essential that the two networks N 1 and N 2 be properly identified before application of this theorem. The two necessary requirements are: a. The terminal voltage (or current) of the network identified as Network N 1 must be the controlling quantity for the source to be eliminated; and b. No current may enter or leave either network through any terminal other than the two terminals by which the networks and controlled source are joined in series (or parallel). It should be noted that connections joining separate parts of a network carry no current and can therefore be ignored in respect to item b. A useful application of the reduction theorem is provided by the cascode amplifier shown in Figure E.12.
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
E7
Figure E.12 shows a partial circuit of a cascode amplifier where a commonbase bipolar transistor sits on top of a commonemitter bipolar transistor. Of course, a cascode amplifier may use MOSFETs of CMOS devices. To illustrate an application of the reduction theorem, let us consider the the incremental model of some cascode amplifier shown in Figure E.13.
v g1 v g1 v in
R1 R2 R3
v3 v g2
v g2
R3
i (a)
v out
v in v in
R3
v3
v 3
( 1 + ) v3
R3 R4 Network 2
R1
( 1 + )R 2 Network 1
v out
(b)
R3 ( 1 + ) v3
( 1 + ) v in v in
2
( 1 + )R 3
R1
( 1 + ) R2
R4
v out
(c) Figure E.13. Incremental model of a cascode amplifier and simplification by application of the reduction theorem
E 8
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
v in C 1
RB
v in i in rn v out is
Rs R1
ib
( 1 + )i b
is
Rs
i b (b)
ro
R3
R2 i out
v out
i out
Network 1
Network 2
(a) v in ki in ----s k
R
1 ----k
ki b rn --k ro v out
R3 R2 i out
v in i in is
Rs R1
ki s
(c)
Figure E.14. Simplification of the emitter follower by application of the reduction theorem
As we know from Chapter 3, the principal properties of the emitter follower are that it provides a relatively low output resistance, it is capable of providing a relatively high input impedance, and it provides current and power amplification but no voltage amplification. Figure E.14(b) shows an incremental model for the circuit that is valid in the range of frequencies where the coupling capacitors act as short circuits and where the performance of the transistor is independent of frequency. The hybrid representation for the transistor is used with = 0 . The resistance R 1 represents the parallel combination of resistors R A and R B , and the circuit is separated into two networks in preparation for the reduction theorem. Application of Part 1 of the currentsource form of the theorem yields the reduced circuit shown in Figure E.14(c). For simplicity, the factor 1 + is denoted as k in Figure E.14(c); in practice k is essentially equal . Dividing the resistances R S , R 1 , and r n , by k is equivalent to multiplying
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
E9
i2
i1
Y1
i2
Y2
v1
v2 = K v1
v1
v2 = K v1
In the networks of Figure E.15, the constant K denotes the voltage gain from Node A to Node B, that is, K = v2 v1 (E.6) or (E.7) v 2 = Kv 1 At Node A of Figure E.15(a), we observe that
i 1 = Y ( v A v B ) = Y ( v 1 v 2 ) = Y ( v 1 Kv 1 ) = Yv 1 ( 1 K )
(E.8)
(E.9)
(E.10)
E10
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Millers Theorem
At Node B of Figure E.15(a), we observe that
i 2 = i 1 = Y ( v 1 v 2 ) = Y ( v 2 v 1 ) = Y ( v 2 v 2 K ) = Yv 2 ( 1 1 K )
(E.11)
(E.12)
(E.13)
We should remember that for the equivalent circuit of Figure E.15(b) the value of the gain K , once established, it cannot be changed. Thus, while Millers theorem can be applied to find the input impedance, it should not be used to determine the output impedance of an amplifier since when determining the output impedance the input signal voltage v S is normally set to zero volts. Millers dual theorem states that under certain conditions, to be established below, the networks of Figures E.16(a) and E.16(b) are equivalent.
i1 i 2 = i 1
B Z A
i1
Z1
i 2 = i 1
B Z2
v1
v2
v1
v2
In the networks of Figure E.16, the constant denotes the current gain from Node A to Node B, that is, (E.14) = i2 i1 or i 2 = i 1 (E.15) At Node A of Figure E.16(a), we observe that
v 1 = Z ( i 1 + i 2 ) = Z ( i 1 + i 1 ) = Zi 1 ( 1 + )
(E.16)
(E.17)
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
E11
(E.18)
(E.19)
(E.20)
(E.21)
Example E.3 For the circuit of Figure E.17, it is given that R 1 = 100R 2 . Use Millers theorem to replace this circuit with an equivalent circuit where resistor R 1 is being replaced with another resistor in parallel with resistor R S . Make any reasonable assumptions.
v1
R1 Rs R2
v out
is
gm v1
Solution: Since R 1 R 2 it is reasonable to assume that practically all the current of the current source flows through resistor R 2 and thus the output voltage is
v out = g m v 1 R 2 = Kv 1
from which
K = gm R2
Figure E.18 shows the equivalent circuit after application of Millers theorem where using relation (E.10) we obtain
R' 1 = R 1 ( 1 K ) = R 1 ( 1 + g m R 2 )
E12
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
Millers Theorem
v1
Rs R' 1 R2
is
v out
gm v1
Figure E.18. Miller equivalent circuit for the circuit of Figure E.17
Being familiar with Millers theorem helps us in better understanding the socalled Miller effect. This effect is very pronounced in amplifier circuits where the input capacitance caused by negative feedback from output to the input causes a significant increase in capacitance. As an illustration, let us consider the amplifier of the circuit in Figure E.19.
i vs
Rs
v in
i C f
K
v out = K v in
(E.22)
Relation (E.22) indicates that the input voltage v in sees an increase in capacitance by a factor of
Cf ( 1 + K ) .
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
E13
E14
Electronic Devices and Amplifier Circuits with MATLAB Computing, Second Edition Copyright Orchard Publications
R1
C. The following publications by The MathWorks, are highly recommended for further study. They are available from The MathWorks, 3 Apple Hill Drive, Natick, MA, 01760, www.mathworks.com. 1. Getting Started with MATLAB 2. Using MATLAB 3. Using MATLAB Graphics 4. Using Simulink 5. Sim Power Systems for Use with Simulink 6. FixedPoint Toolbox 7. Simulink FixedPoint 8. RealTime Workshop 9. Signal Processing Toolbox 10. Getting Started with Signal Processing Blockset 11. Signal Processing Blockset 12. Control System Toolbox 13. Stateflow
R2
Electronic Devices and Amplifier Circuits MATLAB Computing , Second Edition Copyright Orchard Publications
Index
% (percent) symbol in MATLAB A-2 555 Timer 7-2 A A/D conversion 5-56 abs(z) MATLAB function A-23 accuracy defined 5-62 active filter 5-8 mode 3-3 pull-up 6-18 region 3-78 ADC 5-56 Advanced Low-Power Schottky devices 6-41 alternating waveform 1-2 amount of feedback 10-2 amplifier defined 1-3 amplitude-modulated signal 9-2 analog computers 5-63 analog-to-digital converter 5-56 AND gate 6-1 angle(z) MATLAB function A-25 anode 2-6 Armstrong oscillator 10-6 astable multivibrator 7-1 attenuation factor 7-13 attenuator 1-4 avalanche 2-8 average value defined 1-1 averaging op amp 5-37 axis MATLAB command A-16 B backward diode 2-47 band-elimination filter 5-10 bandpass amplifier 9-4 filter 5-10, 9-1 band-rejection filter 5-10 band-stop filter 5-10 bandwidth defined 1-5 Barkhausen criterion 10-4 barrier 2-5 base spreading resistance 3-45 basic differentiator circuit 5-35 beta cutoff frequency 3-58 BiCMOS devices 6-41 bidirectional bus 6-38 bistable multivibrator 7-19 bode(sys,w) MATLAB function 1-13 bodemag(sys,w) MATLAB function 1-13 bound electrons 2-2 box MATLAB command A-13 buffer amplifier 5-15, 6-17, 6-35 Butterworth configuration 9-28 C cascaded tuned amplifiers 9-14 cascode amplifier E-7 cathode 2-6 cesium oscillators 10-8 Class A Amplifier 3-31 AB Amplifier 3-35 B Amplifier 3-34 C Amplifier 3-37 D amplifier 3-38 E amplifier 3-38 clc MATAB command A-2 clear MATLAB command A-2 closed loop gain 5-27 transfer function 5-30 CMOS defined 4-19 Inverter 6-33 Logic Gates 6-32 NAND Gate 6-34 NOR Gate 6-35 transmission gate 6-37 CMRR 5-41 Colpitts oscillator 10-7 column vector in MATLAB A-19 command window in MATLAB A-1 commas in MATLAB A-8 comment line in MATLAB A-2 common base current gain 3-5 collector current gain 3-5 emitter gain 3-4 common mode gain 5-41 rejection ratio 5-41 comparator 5-25, 5-50 compensated attenuator D-2 Complementary Metal Oxide Semiconductor 4-19, 6-2 complementary MOS 4-19 complex conjugate A-4 numbers in MATLAB A-3 conduction angle in SCR 4-26 conv(a,b) MATLAB function A-6 coupling capacitor 8-21 covalent bonding 2-1 critical frequency 5-8 crossover distortion 3-35 crystal controlled Pierce oscillator 10-10 oscillator 10-8 current amplifier 1-18 mirror 4-19 source absorption E-4 cutoff frequency 1-5, 5-8 region in bipolar transistors 3-78 MOSFETs 4-9 UJTs 4-23 D D/A converter 5-53 DAC with with binary-weighted resistors 5-53 buffered output and gain 5-56 Darlington connection 3-60 data bus 6-38 points in MATLAB A-14 sheets from manufacturers 6-20 dB defined 1-4 decade defined 1-7 decibel defined 1-4 deconv(c,d) MATLAB function A-6 default settings in MATLAB A-13, A-16 degenerative feedback 5-25, 10-2 depletion region 2-4 mode 4-7 diac 4-23 differential circuit 6-24 gain 5-41 input amplifier 5-1 integrator 5-76 op amp 5-39 differentiator 5-35 digital-to-analog converter 5-53 diode detector 2-37, 9-2 ideal characteristics of 2-6 display formats in MATLAB A-32 doping 3 dot multiplication operator in MATLAB A-22 drain 4-1 driver gate 6-18 dual-slope ADC 5-60 E Early voltage 3-11 Ebers-Moll transistor model 3-81 ECL circuit 6-24, 6-26 editor window in MATLAB A-1 editor/debugger in MATLAB A-1, A-2 effect of temperature 3-10 effective value defined 1-2 IN-1
electromagnetic interference 4-26 element-by-element division in MATLAB A-21 exponentiation in MATLAB A-21 multiplication in MATLAB A-20 EMI 4-26 emitter coupled logic 6-2, 6-24, 6-26 diffusion resistance 3-45 enhancement-mode MOSFET 4-7 envelope detector 2-37, 9-2 eps in MATLAB A-24, A-29 equivalence gate 6-16 equivalent circuits of op amps 5-13 transistors 3-7 error detector circuit 4-26 exclusive NOR gate 6-16 OR gate 6-15 exit MATLAB command A-2 expand(s) MATLAB function 1-14 external frequency compensation 5-45 F false condition in logic circuits 6-1 fan-in defined 6-17 fan-out defined 6-17 FET 4-1 field-effect transistor 4-1 figure window in MATLAB A-13 firing angle in SCR 4-26 first harmonic 8-1 order all-pass filter 5-95 fixed bias in transistors 3-21 flip-flops 7-19 flash converter 5-56 flat-staggered amplifier 9-22 flip-flop 7-19 fmax(f,x1,x2) MATLAB function A-27 fmin(f,x1,x2) MATLAB function A-27 format command A-32 forward-biased diode 2-5 Foster-Seely discriminator 2-38 Fourier series 8-1, 9-3 four-layer diode 4-38 fplot(fcn,lims) MATLAB function A-27 free electron defined 2-2 movement 2-2 frequency fundamental 8-1 response 1-6, 5-26 spectrum 8-1, 9-4 full-wave rectifier 2-23 function file in MATLAB A-26 fundamental frequency 8-1 fzero(f,x) MATLAB function A-26 IN-2
G gallium arsenide 3-2 gate 4-1 gate turn-off SCR 4-25 grid MATLAB command A-12 gtext(string) MATLAB command A-13 GTO SCR 4-25 H half power defined 1-5, 5-9 frequency 8-6, 9-10 points 1-5 half-wave rectifier defined 2-19 with limited negative output 5-50 positive output 5-49 hard limiter 2-33 harmonics 9-3 Hartley oscillator 10-7 h-equivalent circuit 3-62 hex buffers 6-36 high frequency amplifier 8-9 models for transistors 3-55 level defined 6-2 pass filter 1-23, 5-9 High-Z output of tri-state devices 6-37 hole defined 2-2 movement 2-2 h-parameter equivalent circuit 3-61 hybrid incremental network model 3-50 I ideal characteristics for diode 2-6 IGFET 4-6, 4-7 imag(z) MATLAB function A-23 improper rational function 1-11 impurity atoms 2-3 incremental conductance 2-17 linear models 3-49 resistance 2-17 increments between points in MATLAB A-14 input clamp voltage 6-3, 6-13 resistance 5-22 instrumentation amplifiers 5-42 insulated-gate FET 4-6 integrated circuits 1-1, 6-1 integrator 5-31 internal frequency compensation 5-45 inverse active mode 6-5 inverter 6-2
inverting mode 5-2 summing amplifier 5-37 J JFET 4-1 junction recombination 2-4 L laser diode 2-50 LC Oscillators 10-5 leading edge defined 7-15 light-emitting diode 2-49 lims = MATLAB function A-27 line driver 6-35 linear factor A-9 linearity 5-62 linspace(fv, lv, nv) MATLAB MATLAB command A-14 load line 2-10, 3-40 load resistance 5-24 log(x) MATLAB function A-13 log10(x) MATLAB function A-13 log2(x) MATLAB function A-13 loglog(x,y) MATLAB command A-13 lossless coupling network 9-13 low frequency amplifier 8-5 level defined 6-2 pass filter 1-6, 5-8 lower sidebands 9-4 M magnitude response 1-6 majority carriers 2-4 marginally stable 1-12 matrix multiplication in MATLAB A-20 maximum average forward current 2-8 surge current 2-8 MESFET 4-21 mesh(x,y,z) in MATLAB A-17 meshgrid(x,y) in MATLAB A-17 metal oxide semiconductor defined 4-1 FET 4-6, 4-21 m-file in MATLAB A-1, A-26 microelectronics 1-1 Miller effect 8-11, E-13 Miller integrator 5-31 Millers theorem E-10 minority carriers 2-4 modulation defined 9-1 index in an AM system 9-3 monostable multivibrator 7-14 MOSFET 4-1, 4-6 mutual inductance 10-7
N NaN in MATLAB A-26 NAND gate 6-10 narrowband approximations 9-15 signals 9-4 natural log in MATLAB A-13 negative feedback 10-2 logic 6-2 resistance 2-46 voltage limiter 5-47 NMOS Inverter 6-31 Logic Gates 6-28 NAND Gate 6-31 NOR Gate 6-32 non-inverting averaging 5-38 op amp mode 5-5, 5-13 summing amplifier 5-38 NOR gate 6-13 norator 5-26 NOT gate 6-1 NPN transistor 3-1 N-type semiconductor 2-3 nullor 5-26 Nyquist rate 5-62 O octave defined 1-7 offset null 5-44 op amp 5-1 open collector configuration 6-18 loop defined 5-25 gain 5-27 operational amplifier 5-1 optical coupler 2-51 optoelectronic devices 2-49 optoisolator 2-51 OR gate 6-1 output conductance 4-12 resistance 3-11, 5-23 overcompensated attenuator B-2 overstaggered amplifier 9-22 P
pentode region 4-9 period defined 1-2 periodic signals 8-1 time function 1-2 phase modulated signal 9-2 response 1-6 shift oscillator 10-4 shifter 5-95 photocell 2-50 photodiode 2-47 phototransistor 2-48 piecewise-linear analysis 3-44 Pierce oscillator 10-10 pinch-off 4-3, 4-7, 4-8 PID C-1 PIV 4-27 plot(x,y) MATLAB command A-11 plot(x,y,'rs') MATLAB command A-16 plot(x,y,s) MATLAB command A-15 plot3(x,y,z) MATLAB command A-16 PMOS device 6-30 PNP transistor 3-1 PNPN diode 4-38 polar plot in MATLAB A-24 polar(theta,r) MATLAB function A-23 poles of a function 1-11, 8-29 poly(r) MATLAB function A-5 polyder(p) MATLAB function A-6 polyval(p,x) MATLAB function A-6 positive feedback 10-2 logic 6-2 voltage limiter 5-47 programmable UJT 4-23 P-type semiconductor 2-3 pulsating DC voltage 2-25 Q quadratic factors A-9 region 4-9 quality factor 9-10 quantization defined 5-61 error 5-62 quartz crystal 10-8 quiescent voltages and currents 3-19 quit MATLAB command A-2 R
relaxation oscillators 10-1 resistive network 5-57 resolution defined 5-62 reverse-biased 2-5 rise time defined B-1 RMS value defined 1-2 roots(p) MATLAB function A-3 round(n) MATLAB function A-24 row vector in MATLAB A-3, A-19 rubidium oscillators 10-8 S sampling defined 5-61 theorem 5-62 saturation defined 4-9 current 2-56 region 3-78 Schmitt trigger 7-30 Schottky diode 2-45, 6-39 SCR 4-24 script file in MATLAB A-26 second harmonic 8-1 self-bias transistor 3-25 flip-flop 7-22 semicolons in MATLAB A-8 semilogx(x,y) MATLAB command A-12 semilogy(x,y) MATLAB command A-12 series-fed tuned-collector 10-7 Shockley diode 4-38 signal defined 1-1 silicon controlled rectifier 4-24 single ended output amplifier 5-1 sideband 9-5 slope converter 5-59 tuned transistor amplifier 9-8 sinking current 6-18 sinusoidal oscillators 10-1 slew rate 5-45 slope converter 5-56 SN7400 Quad 2-input NAND gate 6-10 SN7402 Quad 2-input NOR gate 6-14 SN7404 Hex Inverter 6-3 SN7407 hex buffer 6-36 SN7408 Quad 2-input AND gate 6-6 SN74121 Device 7-15 SN74125 device 6-37 SN7417 hex buffer 6-36 SN7432 Quad 2-input OR gate 6-9 SN7486 Quad XOR gate 6-16 SN7486 XOR gate 6-16 solar cell 2-51 source absorption theorem E-4 in JFET 4-1 sourcing current 6-18 stability in systems 1-12, 5-27 stable system 1-12 stagger-tuned amplifiers 9-18 IN-3
passive filter 5-8 pull-up 6-18 peak forward current 2-8 inverse voltage 4-25 point in UJT 4-23 rectifier 2-30, 9-2 reverse voltage 2-8 peak-to-valley ratio 2-47, 9-22
R-2R ladder network 5-54 rational function 8-29 polynomials A-8 RC Oscillator 10-4 real(z) MATLAB function A-23 recombination 2-2 reduction theorem E-6 regenerative feedback 10-2
string in MATLAB A-17 subplot(m,n,p) MATLAB command A-19 substitution theorem C-1 successive approximation ADC 5-58 summing amplifier circuit 5-17 symmetrical triggering 7-29 synchronously-tuned amplifiers 9-15 T tank circuit 10-1 T-equivalent circuit 3-61 text(x,y,string) MATLAB command A-14 thermal voltage 2-7 third harmonic 8-1 threshold voltage 4-7, 6-30 thyristor 4-24 tickler coil 10-6 time-window converter 5-56 title(string) MATLAB command A-13 totem-pole configuration 6-11 trailing edge 7-15 transconductance 1-19, 1-24, 3-54, 4-2, 4-4, 4-11 transistor construction 3-1 future of 4-41 networks 3-61 Transistor-Transistor Logic (TTL) 6-2 transition time 7-28 transresistance 1-19, 1-23, 5-29 triac 4-37 triode region 4-9 triple-point bandwidth 9-22 frequencies 9-22 tri-stated buffers 6-37 true condition in logic circuits 6-1 tuned amplifier 9-1 tunnel diode 2-45 U uncompensated attenuator 7-28, B-1 undercompensated attenuator B-2 understaggered amplifier 9-22 unijunction transistor 4-22 unipolar transistor 4-2 unit load 6-18 step function 5-33 unity gain amplifier 5-8, 5-15 gain frequency 5-26 unstable system 1-13 unsymmetrical triggering 7-29 upper sidebands 9-4 IN-4
V varactor 2-48 varicap 2-48 VCVS 5-13 virtual ground in op amps 5-12 voltage amplifier 1-16 controlled voltage source 5-13 doubler 2-36 follower 5-15 gain 4-17 quadrupler 2-37 source absorption E-4 tripler 2-36 W Wien bridge oscillator 5-50 X xlabel(string) MATLAB command A-12 XNOR gate 6-16 XOR gate 6-15 Y ylabel(string) MATLAB command A-12 Z Zener diodes 2-39 region 2-8 zeros of a function 1-12, 8-29