2013 Book CurrentFeedbackOperationalAmpl PDF

ACSP · Analog Circuits and Signal Processing
Raj Senani
D.R. Bhaskar
A.K. Singh
V.K. Singh
Current Feedback
Operational
Amplifiers and
Their Applications
ANALOG CIRCUITS AND SIGNAL PROCESSING
Series Editors:
Mohammed Ismail. The Ohio State University
Mohamad Sawan. École Polytechnique de Montréal
For further volumes:

http://www.springer.com/series/7381
Raj Senani • D.R. Bhaskar • A.K. Singh
V.K. Singh
Current Feedback
Operational Amplifiers
and Their Applications
Raj Senani D.R. Bhaskar
Division of Electronics Jamia Millia Islamia
and Communication Engineering Electronics and Communication
Netaji Subhas Institute Engineering
of Technology F/O Engineering and Technology
New Delhi, India New Delhi, India
A.K. Singh V.K. Singh

Electronics and Communication Department of Electronics Engineering
Engineering Institute of Engineering and Technology
HRCT Group of Institutions Lucknow, India
F/O Engineering and Technology
Mota, Ghaziabad, India
ISBN 978-1-4614-5187-7 ISBN 978-1-4614-5188-4 (eBook)

DOI 10.1007/978-1-4614-5188-4
Springer New York Heidelberg Dordrecht London
Library of Congress Control Number: 2012955675
# Springer Science+Business Media New York 2013

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed. Exempted from this legal reservation are brief excerpts
in connection with reviews or scholarly analysis or material supplied specifically for the purpose of being
entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication
of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publisher’s location, in its current version, and permission for use must always be obtained from
Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center.
Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

Preface
In spite of all electronic systems prominently being dominated by digital circuits

and systems, the analog circuits have neither become obsolete nor avoidable.
In fact, despite the dominance of digital circuits, analog circuits and techniques
continue to be indispensable and unavoidable in many areas since all real life
signals are analog in nature. Thus, several types of processing of natural signals
or interface of such signals with digital processing circuits has to be necessarily
carried out by analog circuits. Also, many basic functions such as amplification,
rectification, continuous-time filtering, analog-to-digital conversion and digital-to-
analog conversion etc. need analog circuits and techniques.
Traditionally, the integrated circuit (IC) op-amp has usually been considered to
be the workhorse of all analog circuit designs. However, over the years, it was
found that there are many situations such as realization of voltage controlled current
sources, current controlled current sources, instrumentation amplifiers, non-
inverting integrators and non-inverting differentiators etc., where the traditional
voltage mode op-amp (VOA)-based circuits suffer from two drawbacks namely
employment of more than the minimum required number of passive components
and requirement of perfect matching of several of them (due to which any mismatch
may not only deteriorate the performance of the intended circuits but may also lead
to instability in some cases). Furthermore, VOA-based amplifiers exhibit a gain
bandwidth conflict and their frequency range of operation is limited by the effect of
finite gain bandwidth product (GBP) of the op-amps on one hand and due to the
slew-induced distortion (resulting due to finite slew rate of the op-amps) on the
other hand. Consequently, there has been continuous search for alternative analog
circuit building blocks to overcome these difficulties while still matching the
versatility of the VOAs in realizing almost all kinds of analog functions.
During the past four decades, many alternative new analog circuit building
blocks have been proposed out of which only the Operational Transconductance
Amplifiers, Current Conveyors and Current Feedback Operational Amplifiers
have been made available as of-the-shelf ICs and have therefore attracted the
attention of educators, researchers and circuit designers worldwide who have
explored their various applications. Among these building blocks, the current
v
vi Preface
feedback operational amplifier (CFOA), sometimes also referred as operational

trans-impedance amplifier, has received notable attention in literature because of
its two very significant properties namely, a very high slew rate (theoretically
infinite; practically as high as several thousand volts per μs as against a very modest
0.5 V/μs for the general purpose and most popular μA741 type op-amp) and its
capability of offering gain bandwidth decoupling (thereby implying the feasibility
of maintaining essentially a constant bandwidth and variable gain, for low to
medium values of the gains). Though CFOAs have some limitations as compared
to the traditional VOAs, their advantageous features coupled by their versatility and
flexibility, particularly of a specific type which has its compensation pin accessible
externally, overshadows their demerits in a number of applications.
This monograph is basically concerned with CFOAs and their applications and
includes an extensive discussion about various types of CFOAs, the basic circuits
realizable using them, their merits and demerits and their applications in the
realization of continuous time analog filters, simulation of inductors and other
type of impedances, synthesis of sinusoidal oscillators and miscellaneous linear
and non-linear applications (including a variety of relaxation oscillators and chaotic
circuits). Also covered are numerous examples of the use of CFOAs in realizing a
number of other newly proposed active circuit building blocks. The monograph
closes by giving a brief account of the recent developments in the design of bipolar
and CMOS CFOAs, a discussion about various modified forms of CFOAs proposed
in the recent literature from time to time, outlining the current directions of research
in this area and including a supplementary list of references for further reading.
It is hoped that this monograph, which contains a comprehensive collection of
over 200 CFOA-based analog circuits with their relevant theory and design/
performance details, should turn out to be a useful source of reference for
academicians (both educators and students), practicing engineers and anybody
interested in analog circuit design using CFOAs. Readers may also find a number
of interesting and challenging problems worthy of further investigations, from the
various suggestions given in the respective chapters of this monograph.
Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Prologue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 An Overview of Analog Circuits and Their Applications . . . . . . . 2
1.3 The Ubiquitous Op-Amp: The Drawbacks
and Limitations of Some Op-Amp Circuits . . . . . . . . . . . . . . . . . 3
1.3.1 Op-Amp Circuits Which Employ More Than
the Minimum Number of Resistors and Require
Passive Component-Matching . . . . . . . . . . . . . . . . . . . . . 3
1.3.2 The Gain-Bandwidth Conflict . . . . . . . . . . . . . . . . . . . . . . 7
1.3.3 Slew-Rate Based Limitations . . . . . . . . . . . . . . . . . . . . . . 8
1.4 A Brief Review of the Evolution of Alternative Analog
Circuit Building Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4.1 The Operational Transconductance Amplifiers . . . . . . . . . 9
1.4.2 The Current Conveyors . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.4.3 The Current Feedback Op-Amp (CFOA) . . . . . . . . . . . . . 14
1.4.4 The Operational Trans-resistance Amplifier . . . . . . . . . . . 15
1.4.5 The Four-Terminal-Floating-Nullor . . . . . . . . . . . . . . . . . 17
1.4.6 The Current Differencing Buffered Amplifier . . . . . . . . . . 18
1.4.7 The Current Differencing Transconductance
Amplifier (CDTA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.5 The Necessity and the Scope of the Present Monograph . . . . . . . . 20
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2 CFOAs: Merits, Demerits, Basic Circuits
and Available Varieties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.2 AD844: The CFOA with Externally-Accessible
Compensation Pin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3 The Merits and the Advantageous Features of the CFOAs . . . . . . 28
2.3.1 The Reason and the Origin of the High Slew Rate . . . . . . . 28
vii
viii Contents
2.3.2 De-coupling of Gain and Bandwidth: Realisability

of Variable-Gain, Constant-Bandwidth Amplifiers . . . . . . . 30
2.4 The Demerits and Limitations of CFOAs . . . . . . . . . . . . . . . . . . . 31
2.4.1 Demerits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.4.2 Difficulties with Capacitive Feedback . . . . . . . . . . . . . . . . 32
2.4.3 Effect of Stray Capacitances and Layout Issues . . . . . . . . . 32
2.5 Basic Circuits Using CFOAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.5.1 VCVS Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.5.2 Instrumentation Amplifier Using CFOAs . . . . . . . . . . . . . 34
2.5.3 VCCS, CCVS and CCCS Configurations . . . . . . . . . . . . . 35
2.5.4 Unity Gain Voltage and Current Followers . . . . . . . . . . . . 36
2.5.5 Integrators and Differentiators . . . . . . . . . . . . . . . . . . . . . 36
2.6 Commercially Available Varieties of CFOAs . . . . . . . . . . . . . . . . 42
2.6.1 The Mixed-Translinear-Cells (MTC) as Building
Blocks of CFOAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.6.2 Elantec Dual/Quad EL2260/EL2460 . . . . . . . . . . . . . . . . . 44
2.6.3 Intersil HFA 1130 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
2.6.4 AD8011 from Analog Devices . . . . . . . . . . . . . . . . . . . . . 45
2.6.5 THS 3001 from Texas Instruments Inc. . . . . . . . . . . . . . . . 46
2.7 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3 Simulation of Inductors and Other Types
of Impedances Using CFOAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.2 An Overview of Op-Amp-RC Circuits for Grounded
and Floating Inductor Simulation and Their Limitations . . . . . . . . 49
3.3 Realization of Gyrator and Grounded Impedances
Using CFOAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3.4 Single-CFOA-Based Grounded Impedance Simulators . . . . . . . . . 56
3.4.1 Lossy Grounded Inductors/FDNRs . . . . . . . . . . . . . . . . . . 57
3.4.2 Single-CFOA-Based Grounded Negative
Capacitance and Negative Inductance Simulators . . . . . . . 60
3.5 Floating Inductors and Floating Generalized
impedance Simulators Using CFOAs . . . . . . . . . . . . . . . . . . . . . . 60
3.6 Floating Inductance Circuits Employing Only Two CFOAs . . . . . 65
3.6.1 Lossless/Lossy Floating Inductance Simulator . . . . . . . . . . 65
3.6.2 A Lossy Floating Inductance Simulator . . . . . . . . . . . . . . 67
3.7 Applications of Simulated Impedances
in Active Filter Designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.7.1 Applications in the Design of Second Order Filters . . . . . . 68
3.7.2 Application in the Design of Higher Order Filters . . . . . . . 69
3.8 Realization of Voltage-Controlled Impedances . . . . . . . . . . . . . . . 71
3.8.1 Grounded Voltage Controlled Impedance Simulators . . . . . 72
3.8.2 Floating Voltage Controlled Impedance Simulators . . . . . . 73
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Contents ix
4 Design of Filters Using CFOAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4.2 The Five Generic Filter Types, Their Frequency
Responses and Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
4.3 Voltage-Mode/Current-Mode Biquads Using CFOAs . . . . . . . . . . 83
4.3.1 Dual Function VM Biquads . . . . . . . . . . . . . . . . . . . . . . . 83
4.3.2 Single Input Multiple Output (SIMO)
Type VM Biquads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
4.3.3 Multiple Input Single Output (MISO)
Type VM Biquads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.3.4 MISO-Type Universal Current-Mode (CM) Biquads . . . . . 99
4.3.5 Dual-Mode Universal Biquads Using
Single CFOA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.3.6 Mixed-Mode Universal Biquads . . . . . . . . . . . . . . . . . . . . 103
4.4 Active-R Multifunction VM Biquads . . . . . . . . . . . . . . . . . . . . . . 107
4.5 Inverse Active Filters Using CFOAs . . . . . . . . . . . . . . . . . . . . . . 110
4.6 MOSFET-C Filters Employing CFOAs . . . . . . . . . . . . . . . . . . . . 112
4.6.1 MOSFET-C Fully Differential Integrators . . . . . . . . . . . . . 113
4.6.2 MOSFET-C Fully Differential Biquads . . . . . . . . . . . . . . . 115
4.6.3 MOSFET-C Single-Ended Biquad . . . . . . . . . . . . . . . . . . 116
4.7 Design of Higher Order Filters Using CFOAs . . . . . . . . . . . . . . . 118
4.7.1 Signal Flow Graph Based Synthesis of nth
Order Transfer Function Using CFOAs . . . . . . . . . . . . . . . 119
4.7.2 Doubly Terminated Wave Active Filters
Employing CFOA-Based on LC Ladder
Prototypes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
4.7.3 Higher Order Modular Filter Structures
Using CFOAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5 Synthesis of Sinusoidal Oscillators Using CFOAs . . . . . . . . . . . . . . . 131
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
5.2 The Evolution of Single Element Controlled
Oscillators: A Historical Perspective . . . . . . . . . . . . . . . . . . . . . . 131
5.3 Advantages of Realizing Wien Bridge Oscillator
Using CFOA vis-à-vis VOA . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
5.4 Single-Resistance-Controlled Oscillators (SRCO)
Using a Single CFOA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.4.1 A Novel SRCO Employing Grounded Capacitors . . . . . . . 138
5.5 Two-CFOA-Two-GC SRCOs: The Systematic
State Variable Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
5.6 Other Two-CFOA Sinusoidal Oscillator Topologies . . . . . . . . . . . 143
5.7 Design of Active-R SRCOs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
5.7.1 Active-R Sinusoidal Oscillators Using CFOA-Pole . . . . . . 148
x Contents
5.7.2 Low-Component-Count CFOA-Pole

Based Active-R SRCOs . . . . . . . . . . . . . . . . . . . . . . . . . 149
5.7.3 Other Two-CFOA Based Active-R SRCOs . . . . . . . . . . . 150
5.7.4 CFOA-Pole-Based RC Oscillator . . . . . . . . . . . . . . . . . . 150
5.7.5 A Simple Multiphase Active-R Oscillator
Using CFOA Poles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
5.8 SRCOs Providing Explicit Current Output . . . . . . . . . . . . . . . . . 152
5.9 Fully-Uncoupled SRCOs Using CFOAs . . . . . . . . . . . . . . . . . . . 157
5.10 Voltage-Controlled-Oscillators Using CFOAs
and FET-Based VCRs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
5.11 State-Variable Synthesis of Linear VCOs Using CFOAs . . . . . . . 161
5.12 Synthesis of Single-CFOA-Based VCOs
Incorporating the Voltage Summing Property
of Analog Multipliers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
5.13 MOSFET-C Sinusoidal Oscillator . . . . . . . . . . . . . . . . . . . . . . . 173
5.14 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
6 Miscellaneous Linear and Nonlinear Applications
of CFOAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
6.2 Electronically-Variable-Gain Amplifier . . . . . . . . . . . . . . . . . . . 181
6.3 Cable Driver Using CFOA . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
6.4 Video Distribution Amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
6.5 Schmitt Triggers and Non-sinusoidal Waveform
Generators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
6.6 Precision Rectifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
6.7 Analog Squaring Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
6.8 Analog Divider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
6.9 Pseudo-exponential Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
6.10 Chaotic Oscillators Using CFOAs . . . . . . . . . . . . . . . . . . . . . . . 193
6.11 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
7 Realization of Other Building Blocks Using CFOAs . . . . . . . . . . . . . 201
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
7.2 Applications of the CFOAs in Realizing Other
Building Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
7.2.1 CFOA Realizations of Various Kinds
of Current Conveyors (CC) . . . . . . . . . . . . . . . . . . . . . . 202
7.2.2 CFOA-Realization of the
Four-Terminal-Floating-Nullors (FTFN) . . . . . . . . . . . . . 204
7.2.3 CFOA Realization of Operational
Trans-resistance Amplifier (OTRA) . . . . . . . . . . . . . . . . 205
Contents xi
7.2.4 CFOA Realization of Current Differencing

Buffered Amplifier (CDBA) Based Circuits . . . . . . . . . . 207
7.2.5 CFOA Realization of Circuits Containing
Unity Gain Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
7.2.6 Current Differencing Transconductance
Amplifier (CDTA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
7.2.7 Current Follower Transconductance
Amplifiers (CFTA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
7.2.8 Current Controlled Current Conveyor
Transconductance Amplifier (CCCC-TA) . . . . . . . . . . . . 211
7.2.9 Differential Input Buffered Transconductance
Amplifier (DBTA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
7.2.10 Voltage Differencing Differential Input
Buffered Amplifier (VD-DIBA) . . . . . . . . . . . . . . . . . . . 213
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
8 Advances in the Design of Bipolar/CMOS CFOAs
and Future Directions of Research on CFOAs . . . . . . . . . . . . . . . . . 223
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
8.2 Progress in the Design of Bipolar CFOAs . . . . . . . . . . . . . . . . . . 223
8.2.1 Bipolar CFOA with Improved CMRR . . . . . . . . . . . . . . . 223
8.2.2 Bipolar CFOA with Higher Gain Accuracy,
Lower DC Offset Voltage and Higher CMRR . . . . . . . . . 224
8.2.3 Bipolar CFOA Architectures with New
Types of Input Stages . . . . . . . . . . . . . . . . . . . . . . . . . . 225
8.2.4 Novel CFOA Architecture Using a New
Current Mirror Formulation . . . . . . . . . . . . . . . . . . . . . . 227
8.3 The Evolution of CMOS CFOAs . . . . . . . . . . . . . . . . . . . . . . . . . 227
8.3.1 CMOS CFOA with Rail-to-Rail Swing Capability . . . . . . 229
8.3.2 CMOS CFOA for Low-Voltage Applications . . . . . . . . . 229
8.3.3 Fully-Differential CMOS CFOAs . . . . . . . . . . . . . . . . . . 229
8.3.4 CMOS CFOAs with Increased Slew Rate
and Better Drive Capability . . . . . . . . . . . . . . . . . . . . . . 230
8.3.5 Other CMOS CFOA Architectures . . . . . . . . . . . . . . . . . 231
8.4 Various Modified Forms of CFOAs and Related
Advances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
8.4.1 The Modified CFOA . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
8.4.2 Current-Controlled CFOA . . . . . . . . . . . . . . . . . . . . . . . 232
8.4.3 Current Feedback Conveyor . . . . . . . . . . . . . . . . . . . . . . 233
8.4.4 The Differential Voltage Current Feedback
Amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
8.4.5 Differential Difference Complementary
Current Feedback Amplifier . . . . . . . . . . . . . . . . . . . . . . 235
xii Contents
8.5 Future Directions of Research on CFOAs and

Their Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
8.6 Epilogue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
Erratum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E1
References for Additional Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
About the Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
Acknowledgements
The motivation for writing this book came from the involvement of our research
group in writing two short chapters for the Springer monograph Integrated Circuits
for Analog Signal Processing (edited by Prof E. Tlelo-Cuautle) one of which was
related to Current Feedback Operational Amplifiers (CFOA). During the process of
writing these chapters, it dawned upon the first author that the topic of CFOAs and
their applications deserved a full monograph by itself. Accordingly, a detailed
proposal of the present monograph was submitted to Charles Glaser, Senior Editor
Engineering, Springer US, who, after getting the proposal reviewed, gave us a go-
ahead to prepare the proposed monograph.
The authors are thankful to the facilities provided by the Analog Signal
Processing (ASP) Research Lab., Division of ECE, Netaji Subhas Institute of
Technology (NSIT), New Delhi, where the first author works and where this entire
project was carried out.
The authors gratefully thank their respective family members for their unflinch-
ing encouragement, moral support and understanding shown by them over several
decades, in general and during the entire course of the preparation of this mono-
graph, in particular.
The authors also take this opportunity to thank Charles Glaser, Rebecca
Hytowitz and Susan Westendorf and in particular Shashi Rawat, who provided all
necessary support in the preparation of the manuscript of the monograph. The
authors would also like to thank the other colleagues from their research group
namely, S.S. Gupta, R.K. Sharma and Pragati Kumar for their support and
understanding.
The authors, all of whom are members of the research group at the ASP Research
Lab. at NSIT, have also been involved in teaching a number of ideas contained in
this monograph to their students in various courses related to Analog Integrated
Circuit Design at their respective Institutes. A popular query from our students has
been as to: in which book the material taught to them could be found? Their query
has also been an important deriving force to write this monograph and we thank our
numerous students for this and do hope that this monograph, at least partly, answers
to their persistent query.
xiii
Abbreviations
A/D Analog-to-digital
ABB Active building block
AD Analog devices
ADC Analog-to-digital convertor
AM Analog multiplier
BJT Bipolar junction transistor
BW Bandwidth
CB Complementary bipolar
CC Current conveyor
CCCC-TA Current controlled current conveyor transconductance amplifier
CC-CFOA Current controlled current feedback operational amplifier
CCCS Current-controlled-current-source
CCIII Third generation current conveyor
CCVS Current controlled voltage source
CDBA Current differencing buffered amplifier
CDTA Current differencing transconductance amplifier
CE Characteristic equation
CFC Current feedback conveyor
CFOA Current feedback operational amplifier
CFTA Current follower transconductance amplifier
CMOS Complementary metal oxide semiconductor
CMRR Common mode rejection ration
CO Condition of oscillation
CVC Current voltage conveyor
D/A Digital-to-analog
DBTA Differential-input buffered transconductance amplifier
DDA Differential difference amplifiers
DDCC Differential difference current conveyor
DDCCFA Differential difference complimentary current feedback amplifier
DOCC Dual output current conveyor
DVCC Differential voltage current conveyor
xv
xvi Abbreviations
DVCC+ Differential voltage second generation current conveyor

(positive-type)
DVCFA Differential voltage current feedback amplifier
DVCFOA Differential voltage current feedback operational amplifier
ECO Explicit-current-output
ELIN Externally linear but internally nonlinear
FDCC Fully-differential current conveyor
FDCCII Fully differential second generation current conveyor
FDCFOA Fully differential current feedback operational amplifier
FDNC Frequency-dependent-negative-conductance
FDNR Frequency-dependent-negative-resistance
FET Field effect transistor
FI Floating inductance or floating impedance
FPBW Full power band width
FTFN Four-terminal floating nullor
GBP Gain bandwidth product
GC Grounded capacitor
GIC Generalized impedance converter
GNIC Generalized negative impedance converter
GNII Generalized negative impedance inverter
GPIC Generalized positive impedance converter
GPII Generalized positive impedance inverter
IC Integrated circuit
ICC Inverting current conveyor
MCFOA Modified current feedback operational amplifier
MTC Mixed translinear cell
NE Node equation
NMOS N-type metal oxide semiconductor
OFC Operational floating conveyor
OTA Operational transconductance amplifier
OTRA Operational trans-resistance amplifier
PMOS P-type metal oxide semiconductor
SEC Single element controlled
SR Slew rate
SRC Single resistance controlled
SRCO Single resistance controlled oscillator
TAC Transconductance and capacitance
THD Total harmonic distortion
TI Texas instruments
VCC Voltage-controlled capacitance
VCCS Voltage-controlled-current-source
VCFI Voltage controlled floating impedance
VCL Voltage controlled inductance
VCO Voltage controlled oscillator
Abbreviations xvii
VCR Voltage-controlled-resistor
VCVS Voltage controlled voltage source
VCZ Voltage-controlled impedance
VD-DIBA Voltage differencing differential input buffered amplifier
VDTA Voltage differencing transconductance amplifier
VLF Very low frequency
VOA Voltage-mode op-amp
WBO Wien bridge oscillator
Chapter 1
Introduction
1.1 Prologue
Since all natural signals are analog, the analog circuits and techniques to process
them are unavoidable in spite of almost everything going digital. In particular,
several analog functions/circuits such as amplification, rectification, continuous-
time filtering, analog-to-digital (A/D) and digital-to-analog (D/A) conversion are
impossible to be performed by digital circuits regardless of the advances made in
the digital circuits and techniques. Thus, analog circuits are indispensable in many
applications such as processing of natural signals, digital communication, Disk-
drive electronics, processing of signals obtained from optical and acoustical
transducers and wireless and optical receivers, to name a few. Besides these
applications, there are other areas like simulating artificial neurons, artificial neural
networks and a number of applications in image processing and speech recognition
which are better carried out by analog VLSI or mixed signal VLSIs than digital
circuits. Realistically speaking, all electronic design is essentially analog; in fact,
even high-speed digital design is basically analog in nature. In conclusion, the all-
round proliferation of digital circuits and techniques has not made analog circuits
and techniques obsolete rather, it has thrown more challenges to analog circuit
designers to evolve new methods and circuits to design analog signal processing
circuits compatible with concurrent digital technology.
This monograph focuses on Current feedback operational amplifiers (CFOA)
and their applications.
Although most of the chapters of this book deal with various applications of
CFOAs which take as the basis, the commercially available off-the-shelf IC CFOAs
and hence, it would appear that all such circuits are essentially evolved for discrete
circuit applications, however, with some changes, the basic circuit topologies can
also be carried over to fully integratable circuit designs. For example, using bipolar
The original version of this chapter was revised. An erratum to the chapter can be found at
http://dx.doi.org/10.1007/978-1-4614-5188-4_9
R. Senani et al., Current Feedback Operational Amplifiers and Their Applications, 1

Analog Circuits and Signal Processing, DOI 10.1007/978-1-4614-5188-4_1,
2 1 Introduction
CFOAs and the passive resistors realized by BJT-based translinear current-

controlled resistances, the resulting circuits become suitable for implementation
in bipolar IC technology. Similarly, when a CMOS CFOA is considered along with
the resistors realized by CMOS voltage-controlled-resistors (VCR), the given
CFOA configuration would be possible to be integrated as an IC in CMOS technol-
ogy. It is interesting to note that in either case, the resulting integratable version can
easily possess an additional property of electronic tunability which may usually not
be available in the discrete counterpart. In fully-integrable versions of CFOA-based
circuits, the various parameters of the realized circuits can be electronically
adjusted through external DC bias currents in the former case and through external
DC voltages in the latter case.
1.2 An Overview of Analog Circuits and Their Applications
In the world of analog circuits, it is widely believed that almost any function can be
performed using the classical voltage-mode op-amp (VOA). Thus, on one hand, one
can realize using op-amps, all linear circuits such as the four controlled sources
(VCVS, VCCS, CCVS and CCCS), integrators, differentiators, summing and
differencing amplifiers, variable-gain differential/instrumentation amplifiers,
filters, oscillators etc., on the other hand, op-amps can also be used to realize a
variety of non-linear functional circuits such as comparators, Schmitt trigger,
sample and hold circuits, precision rectifiers, multivibrators, log-antilog amplifiers
and a variety of relaxation oscillators. Though a large variety of op-amps are
available from numerous IC manufacturers, the internally-compensated types,
such as μA741 from Fairchild and (to some extent) LF356 from National
Semiconductors can be regarded to be the most popular ones for general purpose
applications. In view of this, therefore, it is not surprising that till about 1990 or so,
analog electronic circuit design was heavily dominated by VOAs.
Although the ‘current feedback operational amplifier’ (CFOA), sometimes also
called ‘operational trans-impedance amplifier’, had been in existence since around
1985 or so, it actually started receiving attention of the analog circuit designers only
when it was recognized that the circuits built using CFOAs can exhibit a number of
advantages in analog circuit design such as, gain-bandwidth independence, relatively
higher slew rate and consequently higher frequency range of operation and advantage
of requiring a minimum number of external passive components without component-
matching in most of the applications; see [1–10] and the reference cited therein.
While several dozens of books by various publishers have been published on
traditional operational amplifiers and their applications, to the best knowledge of the
authors, no such treatment has so far been given to its close relative—the CFOA. It is
this reason which necessitated the writing of this monograph which is exclusively
devoted to the CFOAs and their applications which are currently available only in
research papers published in various international journals over the past two decades.
This chapter gives a brief overview of analog circuits and their applications,
outlines some difficulties and limitations of certain types of op-amp circuits, surveys
1.3 The Ubiquitous Op‐Amp: The Drawbacks and Limitations. . . 3
the state of the art of some prominent alternative building blocks and outlines the
necessity and scope of the present monograph which deals with the CFOAs and their
applications in modern analog circuit design and signal processing.
1.3 The Ubiquitous Op-Amp: The Drawbacks

and Limitations of Some Op-Amp Circuits
Whereas the fact that the traditional VOA is a time-proven building block need not
be emphasized in view of its wide spread recognition as the work horse of analog
circuit design for several decades now, a comparably less acknowledged fact is that
there are several applications in which the use of VOA does not lead to very
appealing circuits. Some such VOA-based circuits are brought out by the examples
which follow:
1.3.1 Op-Amp Circuits Which Employ More

Than the Minimum Number of Resistors
and Require Passive Component-Matching
There are a number of basic op-amp circuits which not only require more number of
passive components than necessary but also call for the use of a number of matched
resistors or require certain conditions/constraints to be fulfilled for realizing the
intended functions. Some examples are as follows.
1.3.1.1 Voltage-Controlled-Current-Sources (VCCS)

and Current-Controlled-Current-Sources (CCCS)
Consider two well-known VCCS configurations shown in Fig. 1.1a, b.

A straight forward analysis of the circuits of Fig. 1.1 shows that the relation
between the output current and input voltage (assuming ideal op-amp) for the
circuit of Fig. 1.1a is given by

Vin R3 1
I0 ¼ þ V0 (1.1)
R1 R2 R4 R 1
whereas for circuit of the Fig. 1.1b it is:

Vin R2 R2 1
I0 ¼ þ V0 (1.2)
R1 R3 R1 R3 R4
4 1 Introduction
a b
R1 Vo
Vin
R2 R2 R3
−
I0
R1 Vo
+
A ZL I0
+
−
Vin R4 ZL
R4 R3 -
Fig. 1.1 VCCS. (a) Non-inverting VCCS, (b) inverting VCCS
a b
R2
iin − R1 R2
A R1 R1
+
−
+ R1 iin
I0 − R1
+ + I0
−
A ZL
R1 ZL
R1 R1
Fig. 1.2 CCCS. (a) inverting CCCS, (b) Non-inverting CCCS
From the above, it may be seen that to realize a VCCS, the op-amp circuits not
only require more than the minimum number of resistances necessary1 but also
require that all the four resistors should have either a relationship R1 ¼ R2R4/R3 or
else all the four resistors be equal-valued and matched so that the output current
becomes independent of the output voltage and depends only on the input voltage,
as required. Thus, any mismatch in resistor values from the intended ones would
degrade the performance of the circuit.
Figure 1.2a shows the realization of a non-inverting CCCS while the circuit of
Fig. 1.2b realizes an inverting CCCS.
Assuming ideal op-amps, the expressions for the output current in terms of input
current for the two circuits are given by

R2 R2
I0 ¼ iin and I0 ¼ iin (1.3)
R1 R1
respectively. Thus, in these cases also as many as four resistors need to be equal-
valued and matched and a total of five resistors are needed whereas (1.3) indicates
1
The minimum number of resistors necessary to realize VCCS and CCVS is one.
a b
R2
C0 R1
R1
Vin + R2
−
A Vo −
Vin +
A Vo
C0
R3
R4 R3 R4
Fig. 1.3 (a) Non-inverting integrator, (b) non-inverting differentiator
that theoretically two resistors should be the minimum number of resistors neces-
sary for realizing a CCCS.
1.3.1.2 Non-inverting Integrator/Differentiator Using a Single Op-Amp
Figure 1.3a shows a non-inverting integrator popularly known as Deboo’s integrator

[11] realized with a single op-amp whereas Fig. 1.3b shows non-inverting differen-
tiator using exactly the same number of passive components. This circuit was
independently proposed by Horrocks [12] and Ganguli [13] separately. In retrospec-
tion, the circuits of Fig. 1.3a, b are also derivable from each other by inverse
transformation of Rathore [14].
A straight forward analysis of the first circuit reveals that its transfer function is
given by

R3 þ R4
V0 R
¼ 1 (1.4)
Vin R4 R2 R3 R1
sC0 R4 þ
R1 R 2
V0 ðR3 þ R4 Þ
for R1 R3 ¼ R2 R4 ; ¼ (1.5)
Vin sC0 R1 R4
On the other hand, the transfer function of the non-inverting differentiator is

given by

R4 R2 R3 R1
sC0 R4 þ
V0 R1 R 2
¼ (1.6)
Vin R3 þ R4
R1
V0 sC0 R4 R1
for R1 R3 ¼ R2 R4 ; ¼ (1.7)
Vin ðR3 þ R4 Þ
6 1 Introduction
Fig. 1.4 Instrumentation R0 R0

V1 +
amplifier
−
R2
+
R1 Vo
−
R2
−
V2 + R0 R0
In both the cases, for ease of design, one normally takes all identical resistors
(though not necessary) i.e. R1 ¼ R2 ¼ R3 ¼ R4 ¼ R. Thus, the transfer function of
the integrator is given by
V0 2
¼ (1.8)
Vin sC0 R
while that of the differentiator is given by
V0 sC0 R
¼ (1.9)
Vin 2
Thus, in both the cases, the circuits require more than the minimum required
number (only one) of resistors. Furthermore, any mismatch of the resistor values
may lead to the difference term (R2R4 R1R3) as in (1.4) and (1.6) becoming
negative. This mismatch in case of the integrator may lead to instability since the
pole of the transfer function would move into the right half of the s-plane. On the
other hand, in case of the differentiator, the mismatch would degrade the perfor-
mance, since it will not remain an ideal differentiator any more.
1.3.1.3 Instrumentation Amplifier
The conventional instrumentation amplifier is another circuit which uses more

number of resistors than the minimum number required. This circuit is shown in
Fig. 1.4.
The input output equation of this circuit is given by

2R2
V0 ¼ 1 þ ð V1 V2 Þ (1.10)
R1
Note that while for realizing a variable gain, two resistors should be the
minimum number of resistors necessary, this circuit employs as many as seven
a b
kR Vin +
A Vo
R −
Vin −
A Vo
+
(k-1)R
R
Fig. 1.5 K-gain amplifiers. (a) Inverting amplifier, (b) non-inverting amplifier
resistors out of which four must be either perfectly matched (or the constraint
between them required to get V0 / ðV1 V2 Þ must be exactly satisfied).
1.3.2 The Gain-Bandwidth Conflict
A major demerit of the various controlled source implementations (except CCVS)

using the traditional VOAs is the so-called ‘gain-bandwidth-conflict’. This can be
explained as follows.
Consider the well-known realization of the K-gain non-inverting and inverting
amplifiers using op-amps, shown in Fig. 1.5.
Consider the one pole model of the op-amp as
A 0 ωp A 0 ωp ωt
A¼ ffi ffi for ω >> ωp (1.11)
s þ ωp s s
where ωt ¼ A0ωp is the gain bandwidth product of the op-amp.

For the non-inverting amplifier, the non-ideal transfer function is given by
V0 ωt
¼ (1.12)
Vin s þ ωKt
whereas the non-ideal gain function of the inverting amplifier of gain K is found
to be
ωt
V0 ðKþ1Þ
¼ K ωt (1.13)
Vin s þ ðKþ1Þ
It is, thus, seen that in the former case, maximum gain is K (at DC) and its 3-dB
bandwidth is ωKt whereas in the latter case, the maximum gain is K but the 3-dB
8 1 Introduction
ωt
bandwidth is ðKþ1Þ . Thus, in both the cases, the gain and the bandwidth cannot be
set independent of each other i.e. there is a gain-bandwidth conflict.
By a non-ideal analysis, it can be easily confirmed that (with the exception of
the CCVS) this gain bandwidth conflict is also present in the VCCS, CCCS and the
instrumentation amplifier circuit discussed earlier.
1.3.3 Slew-Rate Based Limitations
Another factor, which limits the application of VOA-based circuits in higher

frequency ranges, is the finite slew rate of the op-amp which is defined as the
maximum rate of change of the output voltage with respect to time i.e. Slew Rate

0
ðSRÞ ¼ dVdt max . Internally-compensated type op-amps have the first stage as a
differential transconductance-type amplifier followed by a high gain intermediate
stage, with the frequency compensating capacitor Cc connected across the interme-
diate stage such that it is charged by the current delivered by the input transcon-
ductance stage. When a large differential input is applied to an op-amp
configuration (such as to an op-amp configured as a voltage follower), the input
stage gets saturated and delivers a constant maximum current equal to the dc bias
current Ibias of this stage by which the compensation capacitor is charged. Thus, the
voltage across the compensating capacitor (which is equal to the output voltage of
the op-amp) can change with a maximum rate of change equal to the finite and fixed
dc bias current of the input transconductance stage divided by the value of the
compensating capacitor and hence, the SR ¼ Ibias/Cc and is, therefore, limited.
Thus, at large input voltages or high frequencies or a combination of the two, the
output voltage fails to respond with the same speed as the input (due to finite
maximum SR) and this results in slew-induced distortion. Conversely, to avoid
slew-induced distortion, the input voltages and their frequencies are constrained to
be kept small.
Thus, the finite slew rate affects both the dynamic range of the op-amp circuits as
well as the maximum frequency of the input signal which can be applied without
causing noticeable distortion in the output waveform. It may, however be kept in
mind that the operational frequency range of an op-amp circuit or the maximum
frequency of the input signal which can be applied to an op-amp circuit is also
limited by the finite gain-bandwidth product of the op-amp (which results in finite
close loop 3-dB bandwidth as explained earlier).
The maximum frequency fmax up to which an op-amp can operate without being
slew-rate limited is a function of both the frequency and peak amplitude Vop of the
output. This fmax is given by
SR
f max ¼ (1.14)
2πVop
1.4 A Brief Review of the Evolution of Alternative Analog Circuit Building Blocks 9
As the output voltage peak amplitude increases, the maximum frequency at

which slew-rate-limiting occurs decreases. The frequency at which the op-amp
becomes slew-rate-limited is called full power band width (FPBW) and is same as
in (1.14) above. It is interesting to note that FPBW of a given op-amp amplifier
circuit can be considerably less than the small-signal bandwidth of the same circuit.
1.4 A Brief Review of the Evolution of Alternative

Analog Circuit Building Blocks
The various methods of linear analog circuit design, encompassing the classical as
well as the modern approaches, can be broadly classified in two major categories:
(a) Building block approach: In this approach, first an ideal building block is
postulated and synthesis/realization methods are formulated around such build-
ing blocks which are then realized using BJTs or MOSFETs.
(b) Transistor-level approach: In this approach, BJTs and FETs are used directly as
non-linear elements and synthesis/realization methods are developed to achieve
the required functions such that resulting circuits are externally linear but
internally nonlinear (ELIN).
A vast majority of developments in the analog circuit design belong to the first
category, whereas the so-called translinear, log domain and square root domain
circuits constitute the second category.
In the following, we outline a number of popular and prominent analog circuit
building blocks, which have been extensively investigated as alternatives to the
classical VOAs over the last four decades and have been shown to offer a number of
significant features and advantages over VOAs and VOA-based circuits in various
analog signal processing/signal generation applications.
1.4.1 The Operational Transconductance Amplifiers
Because of inability of the traditional op-amp-RC filters in making it possible to

realize precision fully-integratable filters in monolithic form and because of the
need for having fully-integratable continuous-time filters in both bipolar and
CMOS technologies, the operational transconductance amplifier (OTA)-C or gm-
C circuits had been widely investigated by a number of research groups throughout
the world during eighties and nineties and were found to be useful in numerous
applications. Because of the electronic controllability of their transconductance,
OTAs have been extensively used for designing a variety of linear and non-linear
electronically-controllable signal processing and signal generation circuits.
The symbolic notation of the OTA is depicted in Fig. 1.6 and is characterized by
the equations i1 ¼ 0 ¼ i2 ; i0 ¼ gm ðv1 v2 Þ; gm ¼ f ðIB Þ or f ðVB Þ:
10 1 Introduction
i1
v1 +
io
gm
v2 −
i2
(IB or VB)
Controlling Signal
Fig. 1.6 Symbolic notation of the OTA
The OTA-C circuits (also known as transconductance and capacitance (TAC)

circuits) employ only transconductors and capacitors to build various functional
circuits and thus, generally do not require any external resistors. Furthermore, since
the internal circuits of OTAs also can be designed without using any resistors, the
resulting circuits are completely resistor-less. Since the transconductance of an
OTA is electronically controllable through an external DC bias voltage/current, the
OTA-C circuits are suitable for realizing electronically-controllable functions and
are amenable to integration in both bipolar and CMOS technologies. Thus, the
period 1985–1995 witnessed a phenomenal research activity on the various aspects
of OTAs and OTA-C circuits.
A variety of OTAs is commercially available from a number of manufacturers
out of which 3080 type and 13600/13700 type are most popular ICs which have
been extensively used by several researchers for discrete implementation of OTA-C
circuits. The circuit schematic of these two popular commercial OTAs are given
below in Figs. 1.7 and 1.8, whereas an exemplary circuit schematic of a CMOS
OTA is shown in Fig. 1.9.
Among various applications of the OTAs, that of realizing fully integratable and/
or electronically-tunable filters has received major attention in literature. Together
with the filters, the use of the OTAs, in conjunction with only capacitors as passive
elements, to synthesise sinusoidal oscillators has also been extensively investigated
in literature. Developments in OTA-C oscillators were motivated by the resulting
features of electronic-controllability of the oscillation frequency linearly through
the external DC bias currents of the OTAs. Such OTA-C oscillators were exten-
sively investigated by a number of researchers, for instance, see [15–20]. Compre-
hensive catalogues of all possible OTA-C sinusoidal oscillators realisable with only
three/four OTAs and two capacitors were also made available in [21–25]. The
feasibility of implementing these circuits in CMOS was also demonstrated in a
number of works; see [26–29] and the references cited therein.
It may be mentioned that the continued publication of improved CMOS
implementations of OTAs and OTA-based application circuits in various technical
journals even now (for instance see [30]) shows that all possible ideas related to the
design of OTAs (bipolar, CMOS and Bi-CMOS [30–32]) as well as application
circuits using OTAs, have still not been completely exhausted.
+V
Q5 Q6
Q8 Q9
Q7
Q10 i0
Q1 Q2
Vin+
Q13
Vin−
IB Q11 Q12
Q3 Q4
−V
Fig. 1.7 Simplified Schematic diagram of 3080 type IC OTA
+V
Q5 Q6 Buffer
Diode
input
bias Q17
Q8 Q9
Q7 Q18
D1 Q10 i0
Q1 Q2 Buffer
Vin+ output
Q13
D2
Vin−
Q15 Q11 Q12

IB
Q4 Q16
Q3
−V
Fig. 1.8 Simplified Schematic diagram of 13600/13700 type IC OTA
1.4.2 The Current Conveyors
Historically, the progress in the current-mode circuits/techniques can be visualized

to be considerably stimulated due to two major developments. The first one of these
was the proposition of new building blocks known as Current Conveyors by Smith
and Sedra during 1968–1970 [33, 34] and the second one was the introduction of the
so-called translinear circuits by Gilbert in 1975 [35].
12 1 Introduction
a vDD
vDD
I0
vin− M1 M2 vin+
M3
v vss
bias
vss
b
vDD vDD
vss vss
Fig. 1.9 An exemplary CMOS OTA architecture [28]. (a) The schematic of the CMOS OTA,
(b) Symbolic notation and the circuits of PMOS and NMOS current mirrors
The current-mode techniques, in spite of generating some controversy [36], have

indeed given way to a number of interesting/important analog signal processing/
signal generating circuits as is evident from the vast amount of literature on current-
mode circuits and techniques published during the past four decades. Due to the
advances made in integrated circuit (IC) technology during the last two decades,
circuit designers have quite often exploited the potential of current-mode analog
techniques for evolving elegant and efficient solutions to several circuit design
problems.
The most popular current-mode building block has been, undoubtedly,
the Current Conveyor (CC) introduced by Smith and Sedra as first genera-
tion Current Conveyor or CCI in 1968 [33] and later refined to second generation
Current Conveyor in 1970 by Sedra and Smith [34]. Because of the extensive work
done by researchers for more than four decades, the CCs and other current-mode
circuits notably the current feedback op-amps (CFOA) have begun to emerge as an
important class of building blocks with properties and capabilities that enable them
to rival their voltage-mode counterparts (e.g. the traditional voltage mode op-amp)
in a wide range of applications.
The first generation and second generation Current Conveyors (popularly known
as CCI and CCII respectively), are 3-port active elements for which the symbolic
notations are shown in Fig. 1.10.
Fig. 1.10 CCI+, CCI,

CCII+ and CCII Current X X
Conveyors CCI+ Z CCII+ Z
Y Y
iy=ix; vx=vy; iz=ix iy=0; vx=vy; iz=ix
X X
CCI− Z CCII− Z
Y Y
iy=ix; vx=vy; iz=−ix iy=0; vx=vy; iz=−ix
A CCI () is characterized by the hybrid matrix

2 3 2 32 3
iy 0 1 0 vy
4 vx 5 ¼ 4 1 0 0 54 ix 5 (1.15)
iz 0 1 0 vz
while, a CCII () is characterized by

2 3 2 32 3
iy 0 0 0 vy
4 vx 5 ¼ 4 1 0 0 54 ix 5 (1.16)
iz 0 1 0 vz
In current-mode circuits, the operating variable becomes current rather than the
voltage such that voltage swings could be kept small while permitting large swings
in the signal currents. Motivated by the attractiveness of current-mode approach,
extensive research has been carried out on Current Conveyors and translinear
circuits implementable in both bipolar and CMOS technologies during the past
four decades. This research has led to the developments by which applications of
Current Conveyors (and now CFOAs) have been found in almost all the domains
which were once dominated by the traditional voltage-mode op-amps.
These developments have resulted in a large number of integratable bipolar and
CMOS implementations of a wide variety of current conveyors and other related
building blocks such as current voltage conveyors (CVC), dual output CC (DOCC),
differential voltage CC (DVCC), differential difference CC (DDCC), third genera-
tion CC (CCIII), inverting CC (ICC), fully-differential CC (FDCC), Operational
floating conveyor (OFC), Operational Transresistance Amplifier (OTRA), Current
differencing buffered amplifier (CDBA), Current differencing transconductance
amplifiers (CDTA), Voltage differencing transconductance amplifier (VDTA),
Current follower transconductance amplifier (CFTA), Four terminal floating nullor
(FTFN) etc.
In the following, we give a brief account of the prominent developments on the
evolution of only some of these building blocks which are closely related to CCs.
14 1 Introduction
Fig. 1.11 A simplified form + V

of Fabre-Normand
þ (adapted
Translinear CCII
from [37, 38] # 1985 Taylor
& Francis)
MTC
Rbias Y X Z
− V
In 1985, Fabre [37] and Normand [38] independently proposed a Current

Conveyor implementation based on a mixed translinear cell (MTC), a simplified
form of which is as shown in Fig. 1.11.
The Mixed Translinear cell (MTC) and mirror arrangements force the current
out of Z-terminal to be equal to the current out of X-terminal, while the voltage at
X-terminal will be equal to the voltage at Y-terminal; with no current flowing into
the Y-terminal, thereby exhibiting exactly the properties of a CCII þ .
Of all the building blocks evolved as alternatives to the classical op-amp, no
other building block has received as much attention as the CC even when an IC CC
was not available. In fact, there have been at least four different IC CCs produced at
different times: PA630 from Phototronics Ltd, Canada in 1989; CCII01 from LTP
Electronics in 1991, AD844 disguised as a high slew rate op-amp but containing a
CCII+ inside and more recently, Max 4223 from MAXIM introduced in 2010.
However, the most popular of these has been undoubtedly AD844 which can realize
both CCII+ and CCII (apart from its normal use as a current feedback op-amp);
the other varieties have simply not taken off! It is worth mentioning that the number
of publications on the proposals on hardware realization of CCs runs into several
hundreds and so do the number of papers dealing with the applications of the CCs.
1.4.3 The Current Feedback Op-Amp (CFOA)
A Current Feedback Op-amp is essentially a translinear Current Conveyor (CCII+)

followed by a translinear voltage buffer (see Fig. 1.12a for the symbolic notation
and b for a typical bipolar implementation). One of the most popular CFOA
namely, the AD844 from Analog Devices, is a 4-terminal building block
characterized by the following equations:
iy ¼ 0; vx ¼ vy ; iz ¼ ix and vw ¼ vz (1.17)
Fig. 1.12 Current feedback a y y

operational amplifier. (a)
w CCII 1
w
Notation and internal x z +
constituents, (b) bipolar ix x

implementation (adapted iz iz
z
from [1] # 1990 Analog
Devices, Inc.) b
+Vcc
Q6 Q7
IB Q13
Q5 Q8 Q17
Q14
Q1 Q2
Y X Z W
Q4 Q3
Q15
Q9 Q10 Q18
IB Q16
Q11 Q12
-Vee
Fig. 1.13 Symbolic Vp Ip

notation of the operational +
transresistance amplifier Iz Vz
OTRA
Vn In
−
CFOAs have attracted prominent attention in analog circuit design due to their
two significant properties namely, the gain-bandwidth independence and very high
slew rates together with their commercial availability as off-the self ICs from
almost all leading IC manufactures.
Since this monograph is primarily about CFOAs, further aspects of the CFOAs
would be elaborated in more details in the subsequent chapters.
1.4.4 The Operational Trans-resistance Amplifier
Operational trans-resistance amplifier (OTRA) has attracted considerable attention

of analog designers in the context of recent developments in current-mode analog
integrated circuits. The symbolic notation of the OTRA is given in Fig. 1.13.
An OTRA is characterized by the matrix equation
2 3 2 32 3
Vp 0 0 0 Ip
4 Vn 5 ¼ 4 0 0 0 54 In 5 (1.18)
Vz Rm Rm 0 Iz
16 1 Introduction
Fig. 1.14 An exemplary a

CMOS implementation +VDD
of the operational
transresistance amplifier I0 M7
by Toker et al. (adapted
M3 Vg1
from [39] # 2000 IEEE). M1 M5
(a) The schematic of the
CMOS OTRA. (b) Schematic n p z w
1
of CMOS buffer
M2 M6
M4 Vg2
M8
I0
−Vss
b
+VDD
M5
M3
M6
z w
M4
M1
M7
M2
M8
-Vss
An exemplary CMOS OTRA advanced by Toker et al. [39] is shown in

Fig. 1.14a whereas the voltage follower employed therein is shown in Fig. 1.14b.
OTRAs have so far been used in the realization of MOSFET-C filters, immit-
tance simulators, square wave generators etc. for instance, see [40–43, 57] and the
references cited therein.
From the literature survey, it has been found that the possible applications of
OTRAs are clearly in infant stage and newer applications are still continuously
being explored. Also, the work towards improving the CMOS circuit design of
OTRAs is still continuing. Conceptually, an OTRA2 has two low-impedance inputs
and one low-impedance output. Since both input and output terminals of OTRA are
characterized by low-impedance they offer the advantage of eliminating response
limitations incurred by parasitic capacitances.
2
Although a specific type of OTRA namely, the so called Norton amplifier had been commercially
available since long from several manufacturers such as LM3900 from National Semiconductors,
these commercial realizations do not provide virtual ground at the input terminals and they allow
the input current to flow only in one direction. The former disadvantage limits the functionality of
the Norton amplifier whereas the later calls for the use of external DC bias circuits leading to
complex and clumsy designs even for simple functions.
I1 I01
Y W
V1
X Z
V2 I I02
2
Fig. 1.15 Symbolic notation of the four terminal floating nullor
+VDD
I0
I0
M5 M6
M3 w
M1 z y
M7 M8
x
M2 M4
I0
I0
−Vss
Fig. 1.16 CMOS implementation of the FTFN (adapted from [49] # 2000 IEE)
1.4.5 The Four-Terminal-Floating-Nullor
It was demonstrated in [44] and [45] (see also [46] and the references cited therein)
that the Four terminal floating nullor (FTFN)3 is a very general and flexible building
block compared to the other active elements such as voltage-mode op-amps. This led
to a growing interest in the design of amplifiers, gyrators, inductance simulators,
oscillators and current-mode filters using FTFN as the active element [47]. FTFNs
have been implemented using either a supply current sensing method with an
op-amp and current mirrors [48] or using two CCIIs (as suggested in [44]) or two
current feedback op-amp ICs AD844 from Analog Devices. An FTFN can be
considered to be high gain transconductance amplifier with floating input and output
terminals and can also be called an operational floating amplifier (OFA) [58]. The
ideal nullor notation of the FTFN is shown in Fig. 1.15 whereas, a typical CMOS
implementation of this building block, proposed by Cam et al. [49] which is an
embodiment of the interconnection of two CMOS CCII+, is given in Fig. 1.16 and is
characterized by V1 ¼ V2, I01 ¼ I02, I1 ¼ 0 ¼ I2, Vw and Vz being arbitrary.
Though nullors have been regarded in the circuit theory literature as universal
elements and have found numerous applications as well as several integratable FTFN
architectures have been evolved but a perfect FTFN implementation is still elusive.
3
It may be mentioned that acronym ‘FTFN’ was first coined explicitly in [44] and [45].
18 1 Introduction
vp vw
p w
+ ip iw +
CDBA
+ in iz +
n z
vn vz
Fig. 1.17 Symbolic notation of the current differencing buffered amplifier
+VDD
I0
Vg1
M5 M7 M11
I0 Vg1
M3 Vg2
M1 M8
z
n p 1 w
M2 M9
M4 −Vg2
−Vg1 M12
I0 M10
I0
−Vg1
M6
−Vss
Fig. 1.18 CMOS implementation of CDBA (adapted from [51] # 1998 IEE)
1.4.6 The Current Differencing Buffered Amplifier
The Current Differencing Buffered Amplifier (CDBA) was introduced by Acar and
Ozoguz [50]. The symbolic notation of the CDBA, is given in Fig. 1.17 and an
exemplary realization of CDBA advanced by Ozoguz et al. [51], is reproduced
here in Fig. 1.18., and is characterized by Iz ¼ Ip In, Vp ¼ 0 ¼ Vn and Vw ¼ Vz.
The buffer shown in Fig. 1.18 is same as in Fig. 1.14b.
From the available literature on CDBAs (for instance see, [50–52] and the
references cited therein), it is found that the advantages of CDBAs have not
been fully exploited in available applications so far and work is still continuing in
this area.
Ip
Vp p X+ IX+
CDTA
In
Vn n X− IX−
Z
IZ
VZ
Fig. 1.19 Symbolic notation of the CDTA
VDD
M8 M10 M15 M17 M18 M20
IB1 M16 M19
M2
M1 M5 M11
IX+
n p Z IX−
M13 M14
M4
M3 M6 M12
IB3
M21 M22
IB2
M7 M9 M23 M24
VSS
Fig. 1.20 CMOS realization of CDTA (adapted from [56] # 2006 Elsevier)
1.4.7 The Current Differencing Transconductance

Amplifier (CDTA)
The CDTA was introduced in [53] by Biolek. The CDTA is a five terminal active
element and has been shown to be a useful building block for the realization of a class
of analog signal processing circuits [59, 60]. The symbolic notation of CDTA is
shown in the Fig. 1.19. The CDTA can contain an arbitrary number of X-terminals,
providing currents Ix of both directions. The port relations characterizing CDTA are
given by
Vp ¼ Vn ¼ 0; Iz ¼ Ip In ; and IX ¼ g VZ (1.19)
where g is the transconductance.

A representative CMOS implementation of the CDTA is shown in Fig. 1.20.
20 1 Introduction
A comprehensive treatment of a large variety of analog circuit building blocks,

along with the introduction of several new ones, has been dealt in a recent paper by
Biolek et al. [54].
1.5 The Necessity and the Scope of the Present Monograph
From the brief exposition presented in the earlier sections of this chapter the
following may now be summarized:
• Of the various alternative building blocks discussed, only the OTAs, CCs and
CFOAs are commercially available as off-the-shelf ICs whereas the remaining
building blocks are, as of now, not available. Thus, the circuits built around the
other building blocks have so far been studied through SPICE simulations only.
• CCIIs and CFOAs are closely related; in fact, a CFOA is internally a CCII+
followed by an on-chip voltage buffer and is, therefore, more versatile as it can
realize both CCII+ and CCII.
It is worth pointing out that a number of books are available which deals
exclusively with op-amp-based circuits (too many to be mentioned) and OTA-
based circuits, for instance, see [55]. However, to the best knowledge of the authors,
any book dealing exclusively with CFOAs and their applications has not been
published so far.
The present monograph is, therefore, an attempt to fill this void and is targeted to
educators, students, researchers and practicing engineers. This monograph provides
• A state-of-the art survey of CFOAs, their characteristics, merits and limitations
and various types of commercially-available off-the-shelf integrated circuit
CFOAs
• A repertoire of prominent application circuits using CFOAs (covering both
linear and no-linear applications) at a single place, with critical comments on
the merits and demerits of various configurations (instead of being required to
search a vast amount of literature published in various professional journals over
the last more than 15 years).
• An appraisal of recent advances made in the design of bipolar and CMOS
CFOAs and their variants
• A number of open problems and ideas for research for more advanced research-
oriented readers
• A comprehensive list of references on Current feedback operational amplifiers
and their applications (including those referred in the text as well as those
suggested for further reading).
References 21
References
1. Analog Devices (1990) Linear products data book. Analog Devices Inc., Norwood, MA
2. Fabre A (1992) Gyrator implementation from commercially available transimpedance opera-
tional amplifiers. Electron Lett 28:263–264
3. Svoboda JA, McGory L, Webb S (1991) Applications of commercially available current
conveyor. Int J Electron 70:159–164
4. Toumazou C, Payne A, Lidgey FJ (1993) Current-feedback versus voltage feedback
amplifiers: history, insight and relationships. IEEE Int Symp Circuits Syst 2:1046–1049
5. Franco S (1993) Analytical foundations of current-feedback amplifiers. IEEE Int Symp
Circuits Syst 2:1050–1053
6. Bowers DF (1993) The so-called current-feedback operational amplifier-technological break-
through or engineering curiosity? IEEE Int Symp Circuits Syst 2:1054–1057
7. Toumazou C, Lidgey FJ (1994) Current-feedback op-amps: a blessing in disguise? IEEE Circ
Devices Mag 10:34–37
8. Soliman AM (1996) Applications of the current feedback operational amplifiers. Analog Integr
Circ Sign Process 11:265–302
9. Lidgey FJ, Hayatleh K (1997) Current-feedback operational amplifiers and applications.
Electron Commun Eng J 9:176–182
10. Senani R (1998) Realization of a class of analog signal processing/signal generation circuits:
novel configurations using current feedback op-amps. Frequenz 52:196–206
11. Deboo GJ (1967) A novel integrator results by grounding its capacitor. Electron Des 15:90
12. Horrocks D (1974) A non-inverting differentiator using a single operational amplifier. Int J
Electron 37:433–434
13. Ganguly US (1976) Precise noninverting operator realization with high-resistive input imped-
ance. Proc IEEE 64:1019–1021
14. Rathore TS (1977) Inverse active networks. Electron Lett 13:303–304
15. Abuelma’atti MT, Almaskati RH (1987) Active-C oscillator. Electron Wireless World
93:795–796
16. Linares-Barranco B, Rodriguez-Vazquez A, Huertas JL, Sanchez-Sinencio E, Hoyle JJ (1988)
Generation and design of sinusoidal oscillators using OTAs. Proc IEEE Int Symp Circ Syst
3:2863–2866
17. Abuelma’atti MT, Almaskati RH (1989) Two new integrable active-C OTA-based linear
voltage (current)-controlled oscillations. Int J Electron 66:135–138
18. Senani R (1989) New electronically tunable OTA-C sinusoidal oscillator. Electron Lett
25:286–287
19. Abuelma’atti MT (1989) New minimum component electronically tunable OTA-C sinusoidal
oscillators. Electron Lett 25:1114–1115
20. Senani R, Amit Kumar B (1989) Linearly tunable Wien bridge oscillator realised with
operational transconductance amplifiers. Electron Lett 25:19–21
21. Senani R, Tripathi MP, Bhaskar DR, Amit Kumar B (1990) Systematic generation of OTA-C
sinusoidal oscillators. Electron Lett 26:1457–1459, also see (1991) ibid, 27:100–101
22. Senani R, Amit Kumar B, Tripathi MP, Bhaskar DR (1991) Some simple techniques of
generating OTA-C sinusoidal oscillators. Frequenz 45:177–181
23. Bhaskar DR, Tripathi MP, Senani R (1993) A class of three-OTA-two-capacitor oscillators
with non-interacting controls. Int J Electron 74:459–463
24. Bhaskar DR, Tripathi MP, Senani R (1993) Systematic derivation of all possible canonic OTA-
C sinusoidal oscillators. J Franklin Inst 330:885–900
25. Bhaskar DR, Senani R (1994) New linearly tunable CMOS-compatible OTA-C oscillators
with non-interacting controls. Microelectron J 25:115–123
26. Rodriguez-Vazquez A, Linares-Barranco B, Huertas JL, Sanchez-Sinencio E (1990) On the
design of voltage-controlled sinusoidal oscillators using OTAs. IEEE Trans Circ Syst 37:198–211
22 1 Introduction
27. Linnares-Barranco B, Rodriguez-Vazquez A, Sanchez-Sinencio E, Huertas JL (1989) 10 MHz

CMOS OTA-C voltage-controlled quadrature oscillator. Electron Lett 25:765–767
28. Linnares-Barranco B, Rodriguez-Vazquez A, Sanchez-Sinencio E, Huertas JL (1991) CMOS
OTA-C high-frequency sinusoidal oscillators. IEEE J Solid State Circ 26:160–165
29. Sanchez-Sinencio E, Silva-Martinez J (2000) CMOS transconductance amplifiers,
architectures and active filters: a tutorial. IEE Proc Circ Devices Syst 147:3–12
30. Guo N, Rout R (1998) Realisation of low power wide-band analog systems using a CMOS
transconductor. IEEE Trans Circ Syst II 45:1299–1303
31. Wilson G (1992) Linearized bipolar transconductor. Electron Lett 28:390–391
32. Lee J, Hayatleh K, Lidgey FJ (2002) Linear Bi-CMOS transconductance for Gm-C filter
applications. J Circ Syst Comput 11:1–12
33. Smith KC, Sedra AS (1968) The current conveyor—a new circuit building block. Proc IEEE
56:1368–1369
34. Sedra AS, Smith KC (1970) A second generation current conveyor and its applications. IEEE
Trans Circ Theory 17:132–134
35. Gilbert B (1975) Translinear circuits: a proposed classification. Electron Lett 11:14–16
36. Schmid H (2003) Why ‘Current Mode’ does not guarantee good performance. Analog Integr
37. Fabre A (1985) Translinear current conveyors implementation. Int J Electron 59:619–623
38. Normand G (1985) Translinear current conveyors. Int J Electron 59:771–777
39. Toker A, Ozoguz S, Cicekoglu O, Acar C (2000) Current-mode all-pass filters using current
differencing buffered amplifier and a new high-Q bandpass filter configuration. IEEE Trans
Circ Syst II 47:949–954
40. Chen JJ, Tsao HW, Liu SI (2001) Voltage-mode MOSFET-C filters using operational
transresistance amplifiers (OTRAs) with reduced parasitic capacitance effect. IEE Proc Circ
Devices Syst 148:242–249
41. Cam U, Kacar F, Cicekoglu O, Kuntman H, Kuntman A (2004) Novel two OTRA-based
grounded immittance simulator topologies. Analog Intger Circ Sign Process 39:169–175
42. Gupta A, Senani R, Bhaskar DR, Singh AK (2012) OTRA-based grounded-FDNR and
grounded-inductance simulators and their applications. Circuits Syst Sign Process 31:489–499
43. Hou CL, Chien HC, Lo YK (2005) Square wave generators employing OTRAs. IEE Proc Circ
44. Senani R (1987) A novel application of four-terminal floating nullors. Proc IEEE
75:1544–1546
45. Senani R (1987) Generation of new two-amplifier synthetic floating inductors. Electron Lett
23:1202–1203
46. Huijsing JH (1990) Operational floating amplifier. IEE Proc 137:131–136
47. Kumar P, Senani R (2002) Bibliography on nullors and their applications in circuit analysis,
synthesis and design. Analog Integr Circ Sign Process 33:65–76
48. Higashimura M (1991) Realization of current-mode transfer function using four-terminal
floating nullor. Electron Lett 27:170–171
49. Cam U, Toker A, Kuntman H (2000) CMOS FTFN realization based on translinear cells.
Electron Lett 36:1255–1256
50. Acar C, Ozoguz S (1999) A new versatile building block: current differencing buffered
amplifier suitable for analog signal-processing. Microelectron J 30:157–160
51. Ozoguz S, Toker A, Acar C (1998) Current-mode continuous-time fully integrated universal
filter using CDBAs. Electron Lett 35:97–98
52. Pathak JK, Singh AK, Senani R (2011) Systematic realization of quadrature oscillators using
current differencing buffered amplifiers. IET Circ Devices Syst 5:203–211
53. Biolek D (2003) CDTA-building block for current-mode analog signal processing. Proc
ECCTD Poland III: 397–400
54. Bilolek D, Senani R, Biolkova V, Kolka Z (2008) Active elements for analog signal
processing: classification, review, and new proposals. Radioengineering 17:15–32
References 23
55. Deliyannis T, Sun Y, Fidler JK (1999) Continuous-time active filter design. CRC, Boca Raton, FL
56. Keskin AU, Bilolek D, Honcioglue E, Biolkova V (2006) Current-mode KHN filter employing
current differencing transconductance amplifiers. Int J Electron Commun (AEU) 60:443–446
57. Cakir C, Cam U, Cicekoglu O (2005) Novel all pass filter configuration employing single
OTRA. IEEE Trans Circ Syst II 52:122–125
58. Huijsing JH (1993) Design and applications of operational floating amplifier (OFA): the most
universal operational amplifier. Analog Integr Circ Sign Process 4:115–129
59. Prasad D, Bhaskar DR, Singh AK (2010) New grounded and floating simulated inductance
circuits using current differencing transconductance amplifiers. Radioengineering 19:194–198
60. Prasad D, Bhaskar DR, Singh AK (2008) Realisation of single-resistance-controlled sinusoidal
oscillator: a new application of the CDTA. WSEAS Trans Electron 5:257–259
Chapter 2
CFOAs: Merits, Demerits, Basic Circuits
and Available Varieties
2.1 Introduction
Current feedback op-amps (CFOA) started attracting attention of the analog circuit
designers and researchers when it was realized that one can design amplifiers
exhibiting a characteristic which was the most significant departure from the
characteristics exhibited by well-known VOA-based realizations in that CFOA-
based circuits could realize variable-gain and yet constant bandwidth, as against the
unavoidable gain-band-width-conflict in case of the VOA-based designs (as explained
in Chap. 1). Furthermore, it was recognized that due to much higher slew rates of the
order of several hundred to several thousand V/μs (which can be as large as 9,000 V/μ
s for modern CFOAs), as compared to a very modest 0.5 V/μs for the general purpose
and most popular μA741-type VOA, CFOAs could lead to circuits capable of
operating over much wider frequency ranges than those possible with VOAs.
In this chapter, we focus on the merits and demerits of CFOAs; discuss the
various basic analog circuits realizable with CFOAs and highlight a variety of
commercially available IC CFOAs from the various leading IC manufacturers.
2.2 AD844: The CFOA with Externally-Accessible

Compensation Pin
Although in view of the popularity of the CFOAs they have been manufactured as
integrated circuits by a number of IC manufacturers, there are two varieties which
are in use. There are CFOAs which are pin-compatible to VOAs and do not have
externally accessible compensation pin. On the other hand, AD 844-type CFOA
from Analog Devices [1] has the option that its compensation pin (number 5) is
externally-accessible while still maintaining pin-capability with VOAs.

26 2 CFOAs: Merits, Demerits, Basic Circuits and Available Varieties
+Vcc
Q6 Q7
IB Q13
Q5 Q8 Q17
Q14
Q1 Q2
Y X Z W
Q4 Q3
Q15
Q9 Q10 Q18
IB Q16
Q11 Q12
-Vee
Fig. 2.1 A simplified schematic of the CFOA AD844 (adapted from [1] # 1990 Analog
Devices, Inc.)
The AD844 from Analog Devices is a high speed monolithic (current feedback)
op-amp which has been fabricated using junction- isolated complementary bipolar
(CB) process. It has high bandwidth (around 60 MHz at gain of 1 and around
33 MHz at gain of 10) and provides very fast large signal response with excellent
DC performance. It has very high slew rate, typically, 2,000 V/μs. Although it is
optimized for use in current to voltage conversion applications and as inverting
amplifier, it is also suitable for use in many non-inverting and other applications.
Typical applications recommended by the manufacturers include Flash ADC input
amplifiers, High speed current DAC interfaces, Video buffers and cable drivers and
pulse amplifiers.
The AD844 can be used for replacement of traditional VOAs but due to its
current feedback architecture results in much better AC performance, high linearity
and excellent pulse response. The off-set voltage and input bias currents of the
AD844 are laser- trimmed to minimize DC errors such that drift in the offset voltage
is typically 1 μV/ C and bias current drift is around 9 nA/ C. AD844 is particularly
suitable for video applications and as an input amplifier for flash type analog-to-
digital convertors (ADC). A simplified schematic of the AD844 CFOA [1] is shown
in Fig. 2.1.
It is interesting to point out that due to AD844 being sold, disguised as a large
bandwidth, high slew-rate op-amp, initially it almost got unnoticed that its internal
architecture, is, in fact, a translinear second generation plus type Current Conveyor1
1
The Current Conveyors were introduced as new circuit building blocks by Sedra and Smith in
[2, 3]; the first generation Current Conveyor (CCI) in [2] and the more versatile, the second
generation Current Conveyor (CCII) in [3].
2.2 AD844: The CFOA with Externally-Accessible Compensation Pin 27
Fig. 2.2 A block diagram

of the internal architecture
Y
of CFOA AD844
CCII+ 1 W
X
ix iz
Z
(CCII+) followed by a (translinear) voltage buffer. Its simplified symbolic diagram

showing this identification is shown in Fig. 2.2.
Since the internal architecture of AD844 consists of a CCII + followed by a
voltage buffer, this flexibility was later found to be useful in allowing the AD844 to
be used as a CCII + and CCII (using two CCII+), as pin-by-pin replacement of a
VOA (with Z-pin left open) and lastly, as a 4-terminal building block in its own
right.
In view of its front end being a CCII + and the back end being a voltage
follower, the terminal equations of the CFOA can be written as
iy ¼ 0; vx ¼ vy ; iz ¼ ix and vw ¼ v z (2.1)
In the internal architecture of the CFOA, transistors Q1–Q4 are configured as a

mixed translinear cell (MTC) while the collector currents of transistors Q2 and Q3
are sensed by two modified p-n-p and n-p-n Wilson Current Mirrors consisting of
transistors Q5–Q8 and Q9–Q12 respectively to create a replica of current ix at the
terminal- Z thereby yielding iz ¼ ix. The two constant current sources, each equal
to IB, force equal emitter currents in transistors Q1 and Q4 thereby forcing input
current iy ¼ 0 when a voltage Vy is applied at the input terminal Y. It can be easily
proved that with ix ¼ 0, Vx ¼ Vy and the Z-port current iz will be zero. However,
for the case of ix 6¼ 0, an exact analysis [4] of the circuit using exponential
relations between collector currents and base-emitter voltages for the transistors
Q1–Q4 yields

Vy Vx
Iz ¼ Ix ¼ 2IB Sinh (2.2)
VT
from which an approximate relation between Vx, Vy and rx (for Ix < < 2IB) can be
expressed as follows
VT
Vx ffi Vy þ rx ix where rx ¼ (2.3)
2IB
If terminal-Z is terminated into an external impedance/load ZL, a voltage Vz

is created which passes through the voltage follower made from another MTC
composed of transistors Q13–Q18 for which transistors Q13 and Q16 provide the DC
bias currents. The last stage is characterized by an equation similar to (2.3) which
provides Vw ffi Vz .
2.3 The Merits and the Advantageous Features of the CFOAs
Two major merits and advantageous features of the CFOAs are (1) its very high
(theoretically infinite) slew rate and (2) its capability of realizing amplifiers
exhibiting gain-bandwidth decoupling. In the following, we elaborate these two
characteristics of the CFOAs.
2.3.1 The Reason and the Origin of the High Slew Rate
In this sub-section we explain the origin and the reason for a very high slew rate
of CFOAs as compared to conventional op-amps [5]. Figure 2.3a shows a simplified
schematic of an internally compensated type IC op-amp exhibiting the differential
transconductance stage consisting of transistors Q1-Q2-Q3-Q4, the intermediate
gain stage (normally made from a cascade of CC-CE stages) having an inverting
gain –Av2 and the output stage which is a class AB type push-pull amplifier having
both complementary transistors in emitter follower mode providing a voltage gain
Av3 close to unity.
A straight forward analysis of the first stage reveals that the output current Iout
is given by

Vid
Iout ¼ IB tanh (2.4)
2VT
A graphical representation of the above equation is shown in Fig. 2.3b. From this
characteristic, it is seen that the output current Iout saturates to + IB when Vid is
large and positive while Iout saturates to –IB when Vid is large and negative. Thus,
the maximum current available to charge the compensating capacitor Cc, is IB.
If such an op-amp is configured as a voltage follower by a feedback connection
from Vout to the inverting input terminal of the op-amp and a large step signal is
applied to the non-inverting input terminal at t ¼ 0. This forces the transistor Q1
into saturation and Q2 into cut off due to which Iout ¼ IB and thus, the capacitor Cc
is charged linearly through constant current IB.
In view of the high gain of the intermediate stage, for simplicity, its input node
can be treated to be at virtual ground potential in which case one can write
dVout
Iout ¼ Cc (2.5)
dt
2.3 The Merits and the Advantageous Features of the CFOAs 29
a +V
IB
Cc
+ Q1 Q2
vid
- Iout
-AV2 AV3=1
Vout Vout
Q3 Q4
-V Output stage
Input stage
b
Iout
+IB
Vid
-IB
Fig. 2.3 (a) Simplified model of an internally compensated IC op-amp. (b) The tanh-
characteristics of the input differential transconductance stage
Hence, the slew-rate (SR) is given by

dVout Iout IB
SR ¼ ¼ ¼ (2.6)
dt max Cc Cc
With Cc ¼ 30 pF and IB ¼ 19 μA (as applicable to a μA741 type op-amp biased

with 15 V DC power supplies), the above figure turns out to be around 0 .63 V/μ
s which is close to the data sheet value of 0.5 V/μs. For a sinusoidal output V0 ¼ Vm
sin ωt, it can be shown that the maximum frequency ωmax, for which the limitation
imposed by the finite slew rate will not come into play, is given by
SR
ωmax ¼ (2.7)
Vm
Consider now a simplified schematic of the CFOA shown in Fig. 2.4a. An

analysis of the input stage of the CFOA, which is made from MTC consisting of
a +V
IB Q5 Q6
b
Q1 Q2 Iout
Iout
Vy Vx 1
Vout
Cc
Q4 Q3
(Vy - Vx)
IB Q7 Q8
-V
Fig. 2.4 (a) Simplified model of the CFOA. (b) The transfer characteristic between io
and (Vy Vx)
transistors Q1 Q4 shows that the current output coming out of Z- terminal (which
charges the compensating capacitor) is given by

Vy Vx
Iout ¼ 2IB Sinh (2.8)
VT
A plot of the resulting transfer characteristic is shown in Fig. 2.4b. Thus, in this
case, it is found that for a large differential input voltage (Vy–Vx), the output current
which is the charging current of the compensating capacitor would be theoretically
infinite. Thus, in contrast to VOAs, CFOAs have ideally infinite slew rate. In
practice, slew rates from several hundred V/μs to as high as 9,000 V/μs are
attainable. Consequently, a CFOA implementation of a circuit will not have the
same kind of limitations on the maximum operational frequency range as prevalent
in the corresponding VOA-based circuit. In other words, a CFOA-based circuit
would operate satisfactorily over a frequency range much larger than possible for a
VOA circuit realizing the same function.
2.3.2 De-coupling of Gain and Bandwidth: Realisability

of Variable-Gain, Constant-Bandwidth Amplifiers
It has been explained in the previous chapter that all VOA-based controlled sources
suffer from the drawback of gain- bandwidth-conflict. An important advantage of
employing CFOAs is that this gain bandwidth conflict can be overcome due to the
current feedback prevalent in the same configurations realized with CFOAs
2.4 The Demerits and Limitations of CFOAs 31
Fig. 2.5 The non-inverting

amplifier using a CFOA Vin y
w Vo
z
x
R2
R1
(Interestingly, we will see that even two alternative ways of realizing VCVS from
CFOAs are also free from the gain-bandwidth-conflict).
Consider now the CFOA-based non-inverting amplifier of Fig. 2.5.
From an analysis of this circuit, taking CFOA characterization as iy ¼ 0; vx ¼ vy ;
iz ¼ ix and vw ¼ vz ¼ iz Zp where Zp is the parasitic impedance looking into the
Z-terminal and consists of a resistance Rp (typically, around 3 MΩ) in parallel with a
capacitance Cp (typically in the range 4.5 5.5 pF), the maximum gain of the circuit is
found to be

V0 1 þ R2
R1
¼ (2.9)
Vin 1 þ R2
Rp
whereas the 3 dB bandwidth is given by

1 R2 1
ω3dB ¼ 1þ ffi ; for R2 << Rp (2.10)
Cp R2 Rp C p R2
It is, thus, seen that the bandwidth of the circuit can be fixed by setting the
feedback resistor R2 while the gain can be still varied through the variable resistor
R1 and therefore, the gain and bandwidth have become de-coupled and it has
become possible to realize a constant-bandwidth, variable gain amplifier.
2.4 The Demerits and Limitations of CFOAs
2.4.1 Demerits
Despite its significant advantages over traditional VOAs, as explained in previous

section, CFOAs generally have the following demerits:
• Relatively inferior DC precision.
• Relatively poor DC offset voltage due to the use of both PNP and NPN
transistors.
• Lower CMRR and PSRR than VOAs due to the unsymmetrical complimentary-pair
input stage and unequal and un-correlated input bias currents.
A detailed analysis of the input DC current, input offset voltage and maximum
input voltage range for the input stage of a CFOA is given in [6] while a compre-
hensive analysis of output stage has been dealt in [7].
2.4.2 Difficulties with Capacitive Feedback
It should be kept in mind in devising CFOA-based circuits that a capacitive

feedback between X and W is not recommended as it often leads to instability.
Therefore, an inverting Miller integrator cannot be realized with a CFOA in the
same way as the conventional op-amp-based Miller integrator.
2.4.3 Effect of Stray Capacitances and Layout Issues
Another important practical consideration to be taken care of is to take care of the

stray capacitances on the inverting input node (X-input) and across the feedback
resistor which invariably lead to peaking or ringing in the output response and
sometimes even to oscillations. In view of this, appropriate care has to be taken in
making an appropriate PCB layout and eliminate any stray capacitances. The
performance of a CFOA-based circuit can be improved considerably with a good
layout, good decoupling capacitors and low inductance wiring of the components.
2.5 Basic Circuits Using CFOAs
We now show how a number of basic analog circuits such as the four controlled
sources, the voltage and current followers, the instrumentation amplifier and the
integrators and differentiators can be realized in a number of advantageous ways
using CFOAs sans the disadvantages associated with VOA-based realizations of the
same functions.
2.5.1 VCVS Configurations
Consider now the various other VCVS realizations depicted in Fig. 2.6a–c.
2.5 Basic Circuits Using CFOAs 33
a
R2
Vin x
R1 w Vo
z
y
b c
Vin y Vin x
R1 w
w Vo Vo
z z
x y
R1 R2 R2
Fig. 2.6 Realization of various other VCVS circuits using a CFOA (a) inverting VCVS, (b)
alternative non-inverting VCVS, (c) alternative inverting VCVS
A non-ideal analysis of all the three circuits reveals their non-ideal gains as:
2 R
V0
¼ R1 for the circuit of Fig: 2:6a (2.11)
Vin 1 þ R2 Rp
2 R
V0
¼ R1 for the circuit of Fig: 2:6b (2.12)
Vin 1 þ RR2p

V0 R2
R1
¼ for the circuit of Fig: 2:6c (2.13)
Vin 1 þ RR2p
whereas the 3-dB bandwidth in all cases is given by the same value as in (2.10).
Thus, in all the cases, the bandwidth can be set by the feedback resistor R2 after
which the gain can still be made variable through a single variable resistance R1.
Thus, the gain bandwidth conflict is not present in any of the four circuits. It is,
therefore, possible to design constant-bandwidth variable-gain amplifiers using
CFOAs which unfortunately cannot be done with the same topologies such as
those of Figs. 2.5 and 2.6a realized with a traditional VOA.
However, it must be kept in mind that, in practice, constant bandwidth is
achievable only for low to medium gains (typically, 1–10). Furthermore, the
feedback resistor R2 also cannot be chosen arbitrarily since this critically affects
Fig. 2.7 An instrumentation y

V1
amplifier using CFOAs w V01
(CFOA-version of Wilson’s x z
CCII-based circuit [8])
R2
R1
x
w
V2 y z
the stability of the amplifier. In fact, the CFOA parameters rx (typically, around
50 Ω) and Z-pin parasitics Rp sC1 p (where Rp ¼ 3MΩ; Cp ¼ 4.5 pF) with the
feedback resistance R2 decide the stability of the non-inverting and inverting
amplifiers using CFOAs (if realized with CFOAs configured exactly similar to
their VOA-counterparts). The manufacturer determines the optimum value of the
feedback resistor R2 during the characterization of the IC. Normally, lowering R2
decreases stability whereas increasing R2 decreases the bandwidth.
2.5.2 Instrumentation Amplifier Using CFOAs
We now show that, contrary to the traditional instrumentation amplifier which

requires three VOAs and as many as seven resistors out of which four are required
to be completely matched, the use of CFOAs makes it possible to realize a variable
gain instrumentation amplifier with no more than two CFOAs along with a mini-
mum number of only two resistors. Such a circuit is readily evolved from a known
CCII-based circuit proposed by Wilson [8] and is shown here in Fig. 2.7.
Considering the finite input resistance looking into terminal-X of the CFOA as rx
and taking parasitic output impedance looking into terminal-Z as a resistance RP
in parallel with capacitance CP, the maximum gain of this circuit is found to be:
V01 R2
¼ (2.14)
V1 V2 ðR1 þ 2rx Þ 1 þ R2
Rp
whereas its 3-dB bandwidth is given by the some expression as in (2.10). Thus, it
is seen that the bandwidth of the amplifier can be fixed at a constant value by fixing
R2 while the gain can be made variable by changing R1. Thus, CFOA-based
instrumentation amplifier also does not have the gain-bandwidth-conflict while
employing a minimum possible number of passive components for realizing a
variable gain.
a
vin Y
Z0
W
Z vin X
X
W
Z
i0 Y
Z0
i0
b iin X
iin W
X Z
Y
W v0
Z
Y X
v0
ZW
Y
Z0
Z0
c
iin Z2
X iin X
W X W Y
Z Z
Y W Y W
Z Z
Y X
i0 i0
Z1 Z1 Z2
Fig. 2.8 Various controlled sources (a) Voltage controlled current sources. (b) Current controlled
voltage sources. (c) Current controlled current sources
2.5.3 VCCS, CCVS and CCCS Configurations
In Fig. 2.8 we show the CFOA-based realization for non-inverting and inverting
VCCS, CCVS and CCCS circuits. It may be noted that contrary to VOA-based
circuits for VCCS and CCCS requiring as many as four identical resistors the
corresponding realizations using CFOAs as in Fig. 2.8a–c employ a minimum
possible number of passive components namely only one in case of Fig. 2.8a, b
and two in case of Fig. 2.8c respectively thus, no component matching whatsoever
is needed. Furthermore, it is straight forward to verify that all these circuits possess
the most notable property of CFOA-based circuits i.e. no gain-bandwidth-conflict
in the realization of any controlled sources.
a b
vin Y
vin Y W
Z
W X
Z RF
X
2k
v0 v0
Fig. 2.9 Unity gain voltage followers using CFOA
2.5.4 Unity Gain Voltage and Current Followers
Figure 2.9 shows two different ways of realizing a unity gain voltage follower using
CFOAs. In the first case since between terminals Y and X there is already a voltage
follower inside the chip, the same voltage buffer can be used as a voltage follower.
In the second case, a slightly modified version from [9] is presented which contains
a feedback resistor RF for the self-compensation of the voltage follower.
A non-ideal analysis of the voltage follower of Fig. 2.9b considering the X-port input
resistance rx and Z-port parasitic impedance consisting of a resistance Rp in parallel with
a capacitance Cp, reveals the following non-ideal gain function for this circuit

V0 1 þ RRFp 1 þ sCp Rp ==RF
¼ (2.15)
Vin 1 þ rx þR F 1 þ sCp ðRp ==ðrx þ RF ÞÞ
Rp
If RF> > rx, it is seen that a pole-zero cancellation would take place and the
resulting voltage gain will be close to unity and will be perfectly compensated for.
It is found that for a voltage follower made from AD844-type CFOA, the circuit
works quite well with RF ¼ 2 kΩ [9].
The two possible realizations for unity gain current follower are shown in
Fig. 2.10. As expected, none of the two circuits requires any resistors and both
the circuits offer ideally zero input resistance and ideally infinite output resistance.
2.5.5 Integrators and Differentiators
In this subsection we first explain some integrators and differentiators [10] realiz-
able similar to their VOAs counter parts. Due to the reason spelt out earlier an
inverting integrator with a CFOA is not feasible. Since a capacitive feedback from
iin X
W
Z
Y
iin X
X
W W
Z Z
Y Y
i0
i0
Fig. 2.10 Unity gain current followers using CFOA
RG RF C1
Vin
x
w V0
y z
Fig. 2.11 An inverting integrator using a CFOA [10]
W to X leads to instability. However, a slightly modified version with an additional

resistance incorporated in the feedback path is still possible as shown in Fig. 2.11.
Addition of resistor RF is acceptable since at high frequency the resistor is
dominant and hence feedback impedance would never drop below the resistor
value. The transfer function of this circuit is given by
" #
V0 RF s þ RF1C1
¼ (2.16)
Vin RG s
1 1
; for ω << (2.17)
sC1 RG RF C 1
On the other hand, to realize a non-inverting integrator, one can make Deboo’s
integrator [11] almost in the same manner as is done with a VOA (see Fig. 2.12)
however; this circuit suffers from the drawback of requiring four identical resistors
and also has to fulfill a condition to ensure stable operation.
This circuit is characterized by the following transfer function.
!
V0 1 þ RRGF
ffi (2.18)
V1 sR1 C1
Fig. 2.12 CFOA-version RG RF

of non-inverting Deboo’s
integrator [10, 11]
x
w V0
y z
R1
V1
R2
RA C1
Fig. 2.13 Active- R2

compensated non-inverting
CFOA integrator R0 C0 R1
x
Vin wy
z
y
w V0
x z
R3
whereas the condition required for stable operation is
R2 RF
(2.19)
R1 ==RA RG
To circumvent the above problems, in Fig. 2.13 we show an alternative circuit

for creating non-inverting integrator using two CFOAs [12]. This circuit has an in-
built compensation for the non-ideal effects of the CFOA parasitic impedances.
The circuit of Fig. 2.13 realizes a non-inverting integrator since its transfer
function is given by
V0 1 C0 R0 R2
¼ where T ¼ (2.20)
Vin sT R1
Considering the Z-port parasitic impedance Zp ¼ Rp == SC1 P for both the CFOAs, a
non-ideal analysis reveals
V0 R1
ffi εðsÞ (2.21)
Vin sC0 R0 R2
Fig. 2.14 A differential

integrator using a CFOA V1 Y
W V0
Z
V2 X
R0
C0
Fig. 2.15 Dual input R2

integrator proposed by Lee
and Liu (adapted from [13]
# 1999 IET) R1 C
R3
y w
+
x z +
V2
+ r1 V0
V1
− − r2 −
for Ri < <Rpi, i ¼ 0–3. The error function εðsÞ is given by
1 þ sT2
εðsÞ ¼ with T1 ¼ CP R2 ; T2 ¼ CP R1 R3 =R2 (2.22)
1 þ sT1 þ s2 T1 T2
From the above, the phase error is given by
ϕ ffi ωðT2 T1 Þ ω3 T12 T2 (2.23)
Hence, for negligible phase error, one requires T1 ¼ T2 which gives the required
condition as R3 ¼ R22 =R1 .
From the above, it is seen that with R3 ¼ R22 =R1, the phase error is minimized and
active-compensation is achieved.
In the above cases, the circuits devised using CFOAs are exactly similar to their
VOA counterparts. However, since a CFOA has an in built CCII+, there is an
alternative way of realizing inverting/non-inverting integrators. A general circuit to
realize an integrator in an alternative manner is shown in Fig. 2.14. An analysis of
this circuit shows that the output voltage is given by
1
V0 ¼ ðV1 V2 Þ (2.24)
sC0 R0
Thus, both inverting and non-inverting integrators can be realized from this
circuit as special cases by grounding V1 or V2 respectively. A differentiator is
obtainable from the same circuit by interchanging the resistor and the capacitor.
We now show a circuit which can perform the operation of dual input integrator
using a single CFOA proposed by Lee and Liu [13] (Fig. 2.15).
Fig. 2.16 Dual input x w

differentiator using Vout
y z
CFOAs proposed by Lee C
and Liu (adapted from [13] z
# 1999 IET) V1 x w
R3 y
R5
R1
w z
R2
x y
R4
V2
Analysis of this circuit reveals
V2 RR21 V1 2rr21
Vout ¼ (2.25)
sC R3 1 þ RR21 þ R2 2rr21 RR21
If we choose R2/R1 ¼ r2/2r1 the circuit realizes a dual input integrator with
output voltage given by
1
Vout ¼ ð V2 V 1 Þ (2.26)
sτ

1 1
τ ¼ CR1 1 þ R3 þ (2.27)
R1 R 2
From equations (2.26) and (2.27) it is seen that the time constant of the integrator
can be varied by changing the resistor R3. The circuit operates well within the
frequency range of 450 Hz to 1 MHz with a phase error of 5 .
A dual-input differentiator [13] is shown in Fig. 2.16. The input of this circuit
with R4 ¼ (R2 + R3) and R5 ¼ R2, is given by

R2 R2
Vout ¼ V2 α ð1 αÞ þ sCðV1 V2 Þ R2 þ R1 1 þ (2.28)
R3 R3
If α ¼ R2 =ðR2 þ R3 Þ, (2.28) reduces to

1 1
Vout ¼ sCR2 ðV1 V2 Þ 1 þ R1 þ (2.29)
R2 R3
R R4 R2
nR nR C
+ y
Vm(Vin) Vm-1 Vm-2 V3 R5 V2 R3 V1 R1 w V0
z
− x
r
r
Fig. 2.17 Integrator with time constant multiplication proposed by Lee and Liu (adapted from
[14] # 2001 IET)
Hence, the time constant can be varied by changing R1. Over an operating
frequency range of 1–100 kHz, this circuit works well with a phase error of the
order of 10 .
In Fig. 2.17 we show another integrator circuit which was proposed by Lee and
Liu in [14] and has the facility for time constant multiplication. Analysis of this
circuit, as in [14], shows that its transfer function is given by
V0 1
¼ h p ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi m pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi m i
Vin snRC pffiffiffiffiffiffiffiffiffi
1 ffi n þ 2 þ n2 þ 4n n þ 2 n2 þ 4n þ1
2 n þ4n
m 2
(2.30)
By appropriate selection of m and n, a desired multiplication factor can be

achieved. For instance, if we take m ¼ 3 and n ¼ 10 with V3 as the input, it is
possible to achieve a multiplication factor of 143.
The transfer function of the differentiator circuit of Fig. 2.18 is given by

R5 1 þ RR21 þ R2 1 þ RR54 RR31 þ RR32 þ 1
V0 ¼ sC Vin (2.31)
1 RR21 RR54
If R2 ¼ R, R4 ¼ 2R, R1 ¼ R3 ¼ R5 ¼ 2nR the above equation can be

expressed as

1
V0 ¼ 4sRC n 1 þ þ ð n þ 1Þ ð n þ 1Þ (2.32)
2n
As an example, if we select R2 ¼ 1 kΩ, R4 ¼ 2 kΩ, R1 ¼ R3 ¼ R5 ¼ 10 kΩ

then the multiplication factor turns out to be 166.
In reference [14], it has been demonstrated that the circuit of Fig. 2.17 works
well with in the frequency range 200Hz to 1 MHz with a phase error of less than 12o
whereas for the differentiator of Fig. 2.18, the operating frequency range has been
found to be 100 Hz to 10 kHz with a phase error less than 6 .
Fig. 2.18 Differentiator x

with time constant
w V0
multiplication proposed y
by Lee and Liu (adapted C
z
from [14] # 2001 IET) Vin x
z
w
R1
y
R2
R3
R4
R5
It must be mentioned that yet another differential integrator implemented from

two AD 844-type CFOAs and capable of operating up to several MHz without
encountering any stability problem was presented by Maundy et al. in [15].
2.6 Commercially Available Varieties of CFOAs
Although a wide varieties of CFOAs are available from various IC manufacturers,

optimized with respect to a chosen parameter, it is interesting to note that
the key building blocks used are two types of mixed translinear cells. In the
following, we identify these two basic blocks and then briefly describe the internal
architecture and characteristics/parameters of some exemplary IC CFOAs available
from leading analog IC manufacturers.
2.6.1 The Mixed-Translinear-Cells (MTC) as Building Blocks

of CFOAs
Most of the CFOA architectures have the internal structure of a CCII + followed
by a voltage buffer. Since a CCII + itself has a voltage follower between its Y and
X terminals, it, therefore, follows that a typical CFOA architecture would have
two voltage followers (VF): one between Y and X terminals and the other between
Z and W terminals. Furthermore, there has to be a mechanism of sensing the current
flowing into the low-input impedance terminal X of the input VF, creating a copy
of the same and making it available at the high output impedance Z-terminal where
a compensating capacitor can be connected either internally or externally. Two
standard configurations for realizing VFs are the two mixed translinear Cells
2.6 Commercially Available Varieties of CFOAs 43
a
+V +V
IB
p1 IB p1 I2
I2
Q1 Q2 I
x
Y X Y MTC-I X
Q Q
4 3
I4 I4
IB q1 IB q1
−V −V
b
+V
p2 +V
IB I2
Q2 IB
p2
Q1
−V Ix
Y X Y MTC-II X
+V
Q3
Q4
IB
IB
q2 I4 q2
−V
−V
Fig. 2.19 The two types of mixed-translinear cells (MTC) (a) MTC-I, (b) MCT-II (adapted
from [4] #1997 Taylor & Francis)
(MTC) [4, 16] shown in Fig. 2.19a, b. An analysis of the type-I MTC reveals that
the current Ix and differential input (Vy–Vx) are related by the following equation:

Vy Vx
Ix ¼ 2IB sinh (2.33)
VT
Incidentally, type-II MTC of Fig. 2.19b, although has a different topology, it is

also governed by the same equation [4]. This equation can be re-arranged as:

Vy Vx Ix Ix
¼ sinh1 ffi ; for Ix << 2IB (2.34)
VT 2IB 2IB
+V
R2 R3 R4 R5
V1
Q5
Q3 Q4
Q2
Q7 Q9
Q8
Q6
Y X W
Q10
Q12
Q11 Q13
Q14 Q15 Q16

Q17
V2
R7 R8 R9 R10
−V
Fig. 2.20 Elantec dual/quad CFOA EL2260/EL2460 (adapted from [17] # 1995 Intersil
American Inc.)
VT
Vx ffi V y þ I x r x where rx ¼ (2.35)
2IB
Note that when rx is zero, one gets Vx ¼ Vy (as it should be, in the ideal case).
2.6.2 Elantec Dual/Quad EL2260/EL2460
Figure 2.20 shows a simplified schematic of Elantec dual/quad 130 MHz CFOA EL
2260/EL 2460 [17]. As can be seen, this architecture has both input and output buffers
as type- II MTC and no compensating lead is available externally. This CFOA
provides 130 MHz 3-dB band width (for a gain of +2) with a slew rate of 1,500 V/μs.
2.6.3 Intersil HFA 1130
Intersil HFA1130 (Fig. 2.21) CFOA is an ideal choice for applications requiring
output limiting which allows the designer to set the maximum positive and negative
output levels thereby protecting the later stages from damage or input saturation [18].
2.6 Commercially Available Varieties of CFOAs 45
+V
Qp3 Qp4
IB
Qn2 50K
Qp1
ICLAMP
-V Z
VIN+ +1
+V
Qn1 Qn6 VH
Qn5 200 ohm
Qp2
Qp6
IB
Qn3 Qn4 Qp5
−V
RF
VIN- VOUT
Fig. 2.21 Intersil HFA1130 output-limiting low-distortion CFOA (adapted from [18] # 2005
Intersil American Inc.)
The mechanism of high clamp (VH circuit) can be explained as follows. The unity
gain buffer made from type-II MTC forces VIN to track VIN+ and sets up a slewing
current ¼ (VIN VOUT)/RF. This current through the mirror action of the current
mirrors Qp3–Qp4 and QN3–QN4 creates a replica of this current at the high impedance
node Z. The base voltage of Qp5 is 2VBE (QN6 and QP6) less than VH which permits
the conduction of Qp5 whenever the voltage at the Z node reaches a voltage ¼ Qp5’s
base +2VBE (Qp5 and QN5) in this manner the transistor Qp5 clamps node Z whenever
Z reaches to a voltage level ¼ VH. The resistance 50 K acts as a pull-up resistance to
ensure functionality with the clamp input floating. There is similar circuit (not shown
in this diagram) which provides a symmetrical low clamp control by voltage VL.
HFA1130 has a slew rate of the order of 2,300 V/μs and 3 dB bandwidth of
850 MHz and is capable of providing a high output current of the order of 60 mA
and is recommended for applications in the design of residue amplifier, video
switching and routing, pulse and video amplifiers, Flash A/D Driver, RF/IF signal
processing and Medical imaging systems.
2.6.4 AD8011 from Analog Devices
Figure 2.22 shows a simplified schematic of the two-stage CFOA AD8011 from
Analog Devices [19]. The input stage is a type-I MTC with a complementary
second gain stage created from the pair of transistors Q5 and Q6. The circuit
+V
IB I1
Q4
Q1 IQ
Cc BUFFER W
Y X Cp
Q2
Cc
Q5
IB I2
-V
Fig. 2.22 Simplified schematic of the Analog Devices two-stage CFOA AD8011 (adapted from
[19] # 1995 Analog Devices Inc.)
provides low distortion; high speed and high current drive while running on low
quiescent currents. This CFOA has a 3 dB bandwidth of 57 MHz, slew rate of
3,500 V/μs, output current of 30 mA with quiescent power of 12 mW.
2.6.5 THS 3001 from Texas Instruments Inc.
Figure 2.23 shows the CFOA THS3001 from TI has 420 MHz 3-dB bandwidth for
gain of +1, and has slew rate of 6,500 V/μs with current output drive as high
as100mA. The simplified schematic of this CFOA is shown in Fig. 2.23.
This CFOA is built by using a 30-V dielectrically isolated, complementary
bipolar process with NPN and PNP transistors possessing fT of several GHz. This
configuration implements an exceptionally high performance CFOA having wide
bandwidth, high slew rate, fast settling time (40 ns) and low distortion (THD
80 dBc at 10 MHz).
Lastly, it may be pointed out that a wide varieties of CFOAs optimized for
enhancement of one or more of the several specific performance features such as
higher slew rate, increased output current drive capability, wider bandwidth etc. are
available from leading IC manufacturers. For further information, the readers are
referred to the datasheets of various IC manufacturers. Lastly, it may be pointed out
that CFOAs with slew rate as high as 9,000 V/μs (such as THS3202 from Texas
Instruments Inc.) are available as off-the-shelf items.
2.7 Concluding Remarks 47
VCC
IB
Q7 Q8 Q12
Q13
Q9
Q1 Q2
W
Y X
Q3 Q4 Q10 Q14
Q11
Q5 Q6
IB
VEE
Fig. 2.23 A simplified equivalent of 420-MHz, high-speed CFOA THS 3001 type CFOA
(adapted from [10] # 2009 Texas Instruments Inc.)
2.7 Concluding Remarks
In this chapter, we have outlined the distinct merits of CFOAs over VOAs particularly
the mechanism leading to a high (theoretically infinite) slew rate and the resolution of
the gain-bandwidth conflict resulting in the notable property of the CFOA-based
circuits (particularly VCVS structures) of providing constant-bandwidth with variable
gains. We have also outlined the various de-merits of the CFOAs [5] namely, their
inferior CMRR, unsymmetrical input bias dc currents, high input offset voltage and
lower PSRRs etc.
Various basic analog circuit building blocks using CFOAs were outlined and a
number of examples of commercially available CFOAs from leading IC
manufacturers were highlighted.
In spite of their limitations, CFOAs are quite useful for numerous applications
which can be carried out more efficiently with CFOAs than with VOAs, with one or
more of the following advantages: employment of smaller number of external
passive components, elimination of passive component-matching requirements in
several cases and higher operational frequency range. In fact, the nature of many
high frequency applications of CFOAs is such that the very high slew rate puts the
CFOA in the spotlight [20].
In view of the above, it must be emphasized that the focus of the subsequent
chapters of the present book would be primarily on those applications where the
CFOAs are found to provide significant advantages and/or resulting in novel
circuits—the type of which cannot be realized with conventional VOAs.
References
1. 60 MHz 2000 V/μs Monolithic op-amp AD844 (1990) Analog Devices, Inc. Norwood,
MA 02062-9106, USA
2. Smith KC, Sedra A (1968) The current conveyor—a new circuit building block. Proc IEEE
56:1368–1369
3. Sedra A, Smith KC (1970) A second-generation current conveyor and its applications. IEEE
Trans Circ Theor 17:132–134
4. Abuelma’atti MT, Al-Zaher HA (1997) Nonlinear performance of the mixed translinear loop.
Int J Electron 83:467–471
5. Lidgey FJ, Hayatleh K (1997) Current-feedback operational amplifiers and applications.
Electron Commun Eng J 9:176–182
6. Hayatleh K, Tammam AA, Hart BL (2010) Analysis of the input stage of the CFOA.
Int J Electron Commun (AEU) 64:344–350
7. Hayatleh K, Tammam AA, Hart BL (2010) Open-loop output characteristics of a current
feedback operational amplifier. Int J Electron Commun (AEU) 64:1196–1202
8. Wilson B (1989) Universal conveyor instrumentation amplifier. Electron Lett 25:470–471
9. Payne A, Toumazou C (1992) High frequency self-compensation of current-feedback devices.
IEEE Int Symp Circ Syst 3:1376–1379
10. THS3001 420-MHz High-speed Current-feedback amplifier. Texas Instruments Incorporated
September 2009
11. Deboo GJ (1967) A novel integrator results by grounding its capacitor. Electron Design 15
12. OA-31 Current feedback amplifiers. National Semiconductor Corporation November 1992
13. Lee JL, Liu SI (1999) Dual-input RC integrator and differentiator with tunable time constants
using current feedback amplifiers. Electron Lett 35:1910–1911
14. Lee JL, Liu SI (2001) Integrator and differentiator with time constant multiplication using
current feedback amplifier. Electron Lett 37:331–333
15. Maundy B, Gift SJG, Aronhime PB (2004) A novel differential high-frequency CFA integra-
tor. IEEE Trans Circ Syst-II 51:289–293
16. Fabre A (1994) New formulations to describe translinear mixed cells accurately. IEE Proc Circ
17. EL2260, EL 2460: Dual/Quad 130 MHz current Feedback Amplifiers. Intersil American
Inc. January 1995, Rev B
18. HFA1130 850 MHz, Output limiting, low distortion current feedback operational amplifier.
Intersil American Inc. 2005
19. Drachler W (1995) Two stage current-feedback amplifier. Analog Dialogue 29:1–2
20. Harvey B (1993) Current feedback opamp limitations: a state-of-the-art review. IEEE
Int Symp Circ Syst 2:1066–1069
Chapter 3
Simulation of Inductors and Other Types
of Impedances Using CFOAs
3.1 Introduction
Simulation of inductors by active RC networks has been a very prominent and popular
area of analog circuit research. Due to the well-known difficulties of realizing on-chip
inductors of moderate to high values and high quality factors, simulated inductors have
been the alternative choice for realizing inductor-based circuits in integrated circuit
(IC) form. Simulated inductors are also useful in discrete designs in which case they
can replace bulky passive inductors and offer the advantages of reduced size, reduced
cost and complete elimination of undesirable mutual-couplings when several inductors
are being used in a circuit. The traditional voltage mode op-amp (VOA)-based
simulated inductors had been extensively investigated in the seventies to nineties, for
instance see [1–19] and the references cited therein. The well-known Antoniou’s
Generalized Impedance Convertor (GIC)-based circuit [4] requiring two op-amps
and five passive components is regarded to be the best choice available for simulating
a lossless grounded inductance. Apart from simulated inductors, two other useful
circuit elements known as frequency-dependent-negative-resistance (FDNR) [2]-an
element having input impedance of type Z(s) ¼ 1/Ds2 and frequency-dependent-
negative-conductance (FDNC)-an element having input impedance of type Z(s) ¼
Ms2 also find numerous applications in active filters and sinusoidal oscillator designs.
3.2 An Overview of Op-Amp-RC Circuits for Grounded

and Floating Inductor Simulation and Their Limitations
The objective of this chapter is to present a survey of some prominent CFOA-based

circuits for the simulation of inductors and other types of impedances evolved since
1992 till date.

50 3 Simulation of Inductors and Other Types of Impedances Using CFOAs
b
R + –
R
R
a
R R C0
+ –
R3 C4
+
–
R1 R2
– +
R5
R
R
c R2
d
R2
C0
R1
+ C
– 0
–
–
+
R1
+
e R2 f R2
C0 –
+
– –
R1 C0
R1
g R2
h
+
R3 R1 –
R3 R4
C0 R4
– R2
R1 C0
+
Fig. 3.1 Some well-known op-amp-RC circuits for grounded inductor simulation
However, in view of the large amount of work done on the simulation of

grounded and floating inductors using VOAs, it is useful to take a quick review
of some prominent VOA-based inductance simulators before discussing CFOA-
based impedance simulation circuits and putting them in right perspective.
Some well-known op-amp-RC circuits for grounded inductance simulation from
[1–9] are shown in Fig. 3.1.
3.2 An Overview of Op-Amp-RC Circuits for Grounded and Floating Inductor. . . 51
The various circuits shown in Fig. 3.1 simulate lossless/lossy inductors of

different kinds. The Antoniou’s GIC-based circuit [2, 4] of Fig. 3.1a has input
impedance given by Zin ðsÞ ¼ ðsC4 R1 R3 R5 Þ=R2 : On the other hand, the circuit of
Fig. 3.1b from [2] also simulates a lossless grounded inductance of value L ¼ Co R2.
The circuits shown in Fig. 3.1c by Ford and Girling [7] and that of Fig. 3.1d due to
Rao-Venkateshwaran [6] simulate lossy parallel RL type inductors with their input
admittances given by

1 1 1 1 1
Yin ðsÞ ¼ þ þ ; and Yin ðsÞ ¼ þ (3.1)
R1 R2 sCo R1 R2 R2 sCo R1 R2
respectively. The circuit of Fig. 3.1e by Prescott [8] and that of Fig. 3.1f by Rao-
Venkateshwaran [6] simulate lossy series RL with
Zin ðsÞ ¼ ðR1 þ R2 Þ þ sCo R1 R2 ; and Zin ðsÞ ¼ R2 þ sCo R1 R2 (3.2)
respectively.
The circuit of Fig. 3.1g due to Senani [9] simulates another type of parallel RL
with the input admittance given by
" #
1 1 R3 ð1 þ RR34 Þ
Yin ðsÞ ¼ þ þ þ (3.3)
R1 R2 R1 R2 sC0 R1 R2
while the circuit of Fig. 3.1h (also due to Senani [9]) simulates a grounded series RL
impedance of value

R1 R2 R4
Zin ðsÞ ¼ R1 þ R2 þ þ sC0 R1 R2 ð1 þ Þ (3.4)
R3 R3
Note that in case of the circuits of Fig. 3.1g, h both, the inductance value is
controllable through a single variable resistance R4.
From the circuits given in Fig. 3.1, it may be noted that the op-amp based
lossless simulated inductors require more than the minimum required number of
passive components. If the capacitor needs to be grounded, as desirable for
integrated circuit fabrication [20], then, apart from two op-amps, at least seven
matched-resistors are needed as in the circuit of Fig. 3.1b. On the other hand, the
passive component-matching can be avoided (as in the circuit of Fig. 3.1a) but the
capacitor employed is still floating and the circuit still needs as many as four
resistors. The circuits of Fig. 3.1c–f use a canonical number of passive components
but require two op-amps if a variable inductance is needed. Detailed studies such as
those of [5] have revealed that the grounded inductor circuits of the type shown in
Fig. 3.1c–f, can be used satisfactorily (i.e., within permissible deviation in the
inductance value and its quality factor) typically upto only a small fraction of the
gain bandwidth product (GBP) of the op-amps employed (i.e. only about 10 kHz or
so with an op-amp having GBP of 1 MHz).
+ − + −
R3 C4 R2 R1
R1 R2 R5 C4 R3
− + − +
1 2
b c
Z1
+
1
+
− −
Z5 Z4 Z3 Z2
2
Z6 Z7
+ − Z8
d e
C1 i1
R2 R R
+ −
+ −
1
V1
R2 R R
R1
1 2 i2 C
C1 −
+
2 +
V2
−
R R
Fig. 3.2 Some well-known floating inductance simulation circuits. (a) Lossless FI realization
based on Antoniou’s GIC [11, 13, 14], (b) Lossless FI using only three op-amps-based upon
Riordon gyrator [10, 12], (c) Lossy FI of Dutta Roy [15] and Wise [16]; Takagi and Fujii [43],
(d) lossy FI by Sudo and Teramoto [17], (e) Lossy FI by Senani and Tewari [18]
Some prominent floating inductance/impedance simulation circuits using VOAs

[10–19] are shown in Fig. 3.2.
The circuit of Fig. 3.2a simulates a lossless floating inductance having imped-
ance value
3.2 An Overview of Op-Amp-RC Circuits for Grounded and Floating Inductor. . . 53
Z12 ðsÞ ¼ sC4 R1 R3 R5 =R2 (3.5)
at the cost of employing four op-amps along with two capacitors and requiring
matching of passive components in the two identical GIC networks. The circuit of
Fig. 3.2b also requires one cancellation condition:
1 Z7
¼ (3.6)
Z5 Z6 Z8
and one matching condition
1 Z7
¼ (3.7)
Z1 Z6 Z8
to realize a lossless floating impedance of value:
Z1 Z3 Z5
Z12 ¼ (3.8)
Z2 Z4
but has the advantage of employing one less op-amp than the circuit of Fig. 3.2a.
In inductance simulation mode, this circuits has the advantage of employing only a
single capacitance in contrast to two capacitors in the circuit of Fig. 3.2a.
The circuits of Fig. 3.2c simulate a lossy series RL type inductor with
ZðsÞ ¼ ðR1 þ 2R2 Þ þ 2sC0 R1 R2 (3.9)
and have the disadvantage of employing two capacitors but have the advantage of
requiring only two op-amps as unity gain amplifiers and a small number of only
three resistors with a simple, practically adjustable condition of floatation [15]. The
circuit of Fig. 3.2d simulates a lossy parallel RL inductor with
1 1
Y12 ¼ þ (3.10)
R2 sC1 R2 R1
This circuit although employs two matched capacitors and two matched resistors
but has the advantage of realizing a single resistance tunable inductance by having
variable resistance R1.
Lastly, the circuit of Fig. 3.2e employs only two op-amps, only a single capacitor
but needs four matched resistors to realize a floating lossy inductance of value
ZðsÞ ¼ R þ sCR2 (3.11)
From the above described circuits, we note that for lossless FI simulation, three
to four op-amps along with a non-canonical number of passive elements are needed
and that the circuits usually require component-matching conditions and/or cancel-
lation constraints for realizing the lossless floating inductance. On the other hand,
circuits capable of simulating lossy (series RL/parallel RL) floating inductors

although can be realized with usually two VOAs, these also employ non-canonical
number of passive components (as in the circuits of Fig. 3.2c–e) and need identical
resistor values or cancellation constraints in case of single-capacitor simulations.
Apart from the above mentioned difficulties and demerits, it has been
established [19] that the useful frequency range of the FIs of the type of
Fig. 3.2c is, disappointingly, restricted to a very small fraction of the GBP of
the op-amps employed.
Since the well-known and popular method of designing active filters based upon
passive RLC prototypes require both grounded and floating impedances (FI), the
problem of realizing synthetic floating impedances was widely investigated during
the late 1960s to 1990s. Eventually, it was found that using op-amps, it is impossi-
ble to realize synthetic FIs without requiring any component matching conditions.
In the subsequent sections of this chapter, we would highlight some prominent
CFOA-based circuit configurations for realizing synthetic impedances, in both
grounded and floating forms. It will be shown that using CFOAs, not only new
types of simulators have been possible (the type of which cannot be realized with
VOAs) but also that the resulting circuits offer a number of advantageous features
not possible with VOA-based impedance simulators. We will discuss a number of
CFOA-based impedance simulation circuits which can operate over frequency
ranges several orders higher than those possible for VOA-based impedance
simulators.
3.3 Realization of Gyrator and Grounded Impedances

Using CFOAs
Among the first few applications of the CFOAs, which appeared soon after the
CFOA was noticed as an interesting building block for analog cricuit design, was
the gyrator implementation proposed by Fabre in [21]. This circuit is shown in
Fig. 3.3. With port 2 terminated into a capacitance, the circuit simulates a lossless
grounded inductor of value Leq ¼ C0R1R2 looking into port 1.
In fact, CFOA-based grounded impedance circuits can be synthesized systemat-
ically starting from first principles [22]. To this end, we recall that impedance
converters can be realized from inpedance inverters and vice versa. Also, a positive
impedance inverter can be realised with two voltage-controlled current sources
(VCCS) of opposite polarity connected back-to-back. Employing the non-inverting
and inverting VCCS each realized with a single CFOA and no more than a
single impedance, two impedance converter/inverter cicuits are shown in Fig. 3.4.
An interconnection of the two VCCS as shown in Fig. 3.4a makes an effective use
of the on-chip buffer to create a 2-port which realizes an impedance inverter. With
its port 2 terminated into an impedance Z3, the impedance looking into port 1 is
given by
3.3 Realization of Gyrator and Grounded Impedances Using CFOAs 55
R2
X
WZ
Y
W i2
Z
+
i1 Y X
+ V2 C0
R1
V1
− −
Fig. 3.3 A gyrator and inductance simulator proposed by Fabre (adapted from [21] #1992 IEE)
a b
i1
1 y
v1 y w
Z2 z
x
w x Zin y
z z z
x w Z1 Z3 w
y x
Z1 i2 Z2
v2
Fig. 3.4 Realization of grounded impedance converters/inverters. (a) Realization of GPII/GPIC,

(b) Realization of GNII/GNIC (adapted from [22] # 1998 Walter de Gruyter GmbH & Co. KG,
Germany)
Z1 Z2
Zin1 ðsÞ ¼ (3.12)
Z3
This circuit is, hence, a generalized positive impedance inverter (GPII). The
same circuit with port 2 terminated in Z2, Z1 deleted and the terminal thus created
named as port 3, would function as a generalized positive impedance converter
(GPIC).
The realization of the corresponding generalized negative impedance inverter
(GNII) and generalized negative impedance converter (GNIC) elements is obtain-
able by a back-to-back interconnection of two inverting VCCSs or two non-
inverting VCCSs. One of these two implementations of the GNII/GNIC is shown
in Fig. 3.4b which realizes input impedance given by
Z1 Z2
Zin ðsÞ ¼ (3.13)
Z3
Y R2
W X
Z
X W
Z
Y Iin
Y Vin
R1 C Z
W
X
Zin
R3
Fig. 3.5 Active-compensated lossless grounded inductance by Yuce and Minaei (adapted from
[23] # 2009 Springer)
The circuit configurations of Fig. 3.4a, b can be used to realize a variety of useful
circuit elements such as simulated inductance, FDNR and resistively-variable
capacitance, both in positive as well as in negative forms by appropriate selection
(resistive/capacitive) of the three impedances Z1, Z2 and Z3. It is important to point
out that no such circuits (i.e.using only two active elements and only three passive
elements) are possible with traditional VOAs. Known VOA-based circuits for PII
and NII typically require two VOAs and seven matched resistors [1] where as
VOA-based GIC although does not need component-matching but still requires five
impedances. By contrast, the CFOA-based circuits described above, apart from the
capability of operating at relatively higher frequencies than VOA-based circuits,
offer the following advantages: (1) use of a bare minimum (only three) of passive
components (2) single-resistance tunability of the realized impedances in all the
cases and (3) no component-matching requirements/realization constraints.
An interesting grounded lossless inductance circuit having reduced parasitic effect
was suggested by Yuce and Minaei in [23] and is shown in Fig. 3.5. In this circuit, a
third CFOA is employed in the mode of a current inversion type negative impedance
converter (NIC) to reduce the Z-terminal parasitic resistance of the first CFOA
although it slightly increases the total capacitance at the Z-terminal of the first CFOA.
3.4 Single-CFOA-Based Grounded Impedance Simulators
It is known that as compared to their lossless counterparts, lossy grounded inductors

and FDNRs can be realized more economically typically requiring only a single
op-amp along with three passive components. Similarly, it has been found that
employing CFOAs too, lossy grounded inductors and FDNRs as well as grounded
ideal negative inductance and negative capacitance elements can be realized with only
3.4 Single-CFOA-Based Grounded Impedance Simulators 57
R1
Y
R1
W
Z
X
L=C3R1R3
Zin C2 C3 Taking C2=C3
R3
C1
Y C1
W
Z
X
D=C1C3R3
Zin Taking R2=R3

R2 C3
R3
Fig. 3.6 Series-RL and series-CD simulators proposed by Liu and Hwang (adapted from [24]
# 1994 IEE)
a single CFOA along with three/four passive components, with most of them offering
the significant advantage of single-element-controllability of the realized inductance/
capacitance/FDNR which is not possible in the mentioned VOA-based circuits.
A number of authors have proposed such single-CFOA-based grounded imped-
ance simulators. In the following, we present some prominent and representative
circuits from the references [24–30].
3.4.1 Lossy Grounded Inductors/FDNRs
In [24] Liu and Hwang presented a general single-CFOA circuit for grounded
impedance simulation. Two interesting single CFOA-based circuits capable of
simulating series-RL type lossy inductor and series-CD type lossy FDNR resulting
from their general configuration, as special cases, are shown in Fig. 3.6.
In the first circuit, the inductance value is controllable through a variable
resistance R3 while in the second case, the FDNR value is adjustable through a
variable capacitance C3. Also, in both the circuits, the parasitic output impedance
C
X
W R1+R2
Z
Y
R1
L=C R1R2
R2
Fig. 3.7 Kacar and Kuntman’s lossy grounded inductor (adapted from [28] # 2011
Radioengineering Society, Czech and Slovak Technical Universities)
looking into port Z (consisting of a resistance Rp in parallel with a capacitance Cp)

can be absorbed in R3 and C3 respectively.
Kacar and Kuntman [28], through a general scheme, derived four simulated
inductance circuits; however, only one of them realizes a positive lossy inductance.
This circuit is shown in Fig. 3.7.
Yuce in [25] presented four novel CFOA-based grounded series-RL circuits
which are shown here in Fig. 3.8. By a straight forward analysis, for all the circuits,
the input impedance is found to be of type
Vin
Zin1 ðsÞ ¼ ¼ Req þ sLeq (3.14a)
Iin
where
9
Req ¼ R1 ; Leq ¼ CR1 R2 ðCircuits of Fig:3:8a; bÞ >
=
Req ¼ R1 ; Leq ¼ 2CR1 R2 ðCircuit of Fig:3:8cÞ (3.14b)
>
;
Req ¼ R1 =2; Leq ¼ ðCR1 R2 Þ=2ðCircuit of Fig:3:8dÞ
Thus, an interesting feature of these circuits is that the realized inductance value
can be controlled independently by resistance R2 in all of them.
In [29], Abuelma’atti, through two single-CFOA-based generalized impedance
simulation networks, derived a number of positive and negative impedances as
special cases. A special case from [29], which simulates series RL impedance, is
shown here in Fig. 3.9.
It is interesting to point out that the grounded series-CD impedances can be
readily obtained from the grounded series-RL impedances, by the application of
RC-CR transformation [1]. Two such exemplary grounded series-CD simulators,
obtained from the circuits of Fig. 3.8c, b are shown in Fig. 3.10.
3.4 Single-CFOA-Based Grounded Impedance Simulators 59
a C b
R1 R1
iin
Y
X Vin
W
W Z
Z X
Y
iin R2 C
R2
Vin
c d
R1
iin Y
Y iin ZW
X
Vin ZW C
X Vin R1
C2 R2
R2
Fig. 3.8 Single-CFOA-based canonical lossy grounded inductors proposed by Yuce (adapted
from [25] # 2009 Springer)
C4
Y R1
W 2
Z
X
R1
R2 C4R2
L= (R1-R3); R1> R3
2
R3
Fig. 3.9 An exemplary lossy inductance simulator proposed by Abuelma’atti (adapted from [29]
# 2012 Springer)
a b
1
C1 C1
1 1
Y Y Ceq
W W
X Z X Z
R2 R5 C3 Deq
C3
Fig. 3.10 Series CD simulators derived from Yuce’s inductance simulators. (a) Ceq ¼ C1; Deq
¼ (C1C3R2)/2, (b) Ceq ¼ C1; Deq ¼ C1C3R5
A notable property of both the circuits is that the finite non-zero input impedance
rx looking into terminal X of the CFOA can be easily absorbed in resistor R2 in
the first case and in R5 in the second case, while the parasitic capacitance Cp of the
Z-terminal of the CFOA can be merged with C3 in both the cases. Furthermore,
FDNR value is single-resistance-controllable through R2 in the former and through
Rs in the latter.
3.4.2 Single-CFOA-Based Grounded Negative Capacitance

and Negative Inductance Simulators
There have been a number of works [28–30] where single-CFOA-based grounded

negative capacitance and grounded negative inductance have been simulated. Some
prominent circuits are shown in Fig. 3.11. In all cases, the equivalent capacitance
value is single-resistance-controllable.
A number of negative inductance circuits which have been derived by
Abuelma’atti [29] from two single-CFOA-based generalized schemes are shown
in Fig. 3.12 which offer single-resistance control of the realized negative induc-
tance value by the resistor R4 in the first two circuits and by R2 in the third circuit.
3.5 Floating Inductors and Floating Generalized

impedance Simulators Using CFOAs
Using traditional VOAs, no circuit is known to exist which can realize a lossless or
lossy floating inductance (FI) without requiring any component matching condition.
It was demonstrated in [31–34] for the first time that using a negative type CCII, it is
3.5 Floating Inductors and Floating Generalized impedance Simulators Using CFOAs 61
a b
C0 R1
Y Y
ZW Z
W
X
X
Ceq Ceq
R2
R2
R1 C0
c
X
ZW
Y
C4
Ceq
R1
R2
Fig. 3.11 Some grounded negative capacitance circuits (a), (b) Ceq ¼ (C0R2/2R1) (adapted
from [30] # 2011 Springer; (c) Ceq ¼ [C4/(1 + R1/R2)] (adapted from [29], # 2011 Springer)
a b
R1
R1
Y
Y c
ZW
X ZW
X
C4
Y
ZW
X
C5 C2
R4 R4 R2
R3
Fig. 3.12 Some grounded negative inductance simulation circuits proposed by Abuelma’atti.
(a) Z(s) ¼ sC5R1R4, (b) Z(s) ¼ s 2C2R1R4, (c) Z(s) ¼ s (C4R2R3)/2 (adapted from [29]
# 2012 Springer)
Fig. 3.13 CFOA-based

lossless floating inductor 1
employing a grounded V1 i y
capacitor 1 w y z
x z w
x
C0
R1
R2
x x
2 i2 y z
w
y z
w
V2
possible to simulate FIs without requiring any component-matching or cancellation

constraints. Subsequently, it was shown that four current conveyors along with two
resistors are needed [35] if a lossless FI is to be realized employing a grounded
capacitor as preferred for integrated circuit (IC) implementation [20]. Although if
only two resistors and a capacitor are permitted (not necessarily grounded) then
three CCIIs will suffice to realize a lossless floating inductance [32]. However, in
such circuits, CCIIþ as well CCII are needed. Note that, if realized with CFOAs,
such circuits will, therefore, require four CFOAs (one each for the three CCII+s and
two for the CCII) and even then the resulting circuits would not have a grounded
capacitor.
We now show that using four CFOAs, a lossless floating inductance can be
realized using only two resistors and as desired, a grounded capacitor. Such a
circuit, derivable from the four CC-based circuits of [35], is shown in Fig. 3.13.
A straight forward analysis of this circuit reveals that it is characterized by the
equation
V1 V 2
i1 ¼ i2 ¼ (3.15)
sC0 R1 R2
and thus, simulates a lossless floating inductance of value Leq ¼ C0R1R2 with the
advantages of (1) employment of a minimum possible number of passive elements,
(2) use of a grounded capacitor as preferred for IC implementation [20] and (3)
single-resistance-tunability of the realized inductance value through R1 or R2.
In the following, we present a number of three-CFOA-based FIs free from any
component-matching conditions. Three such circuits are shown in Fig. 3.14 all of
which can be considered to be floating GPII/GPIC elements.
The first circuit [22] is obtained from the grounded impedance converter/inverter
circuit of Fig. 3.4b (with its port 2 terminated into an impedance Z2) by un-
grounding the impedances Z1 and Z3, tying them together and connecting to the
voltage V2, inserting a third CFOA appropriately to make it possible to have the
port 2 current i2 of the overall circuit given by
3.5 Floating Inductors and Floating Generalized impedance Simulators Using CFOAs 63
a Z1
w
2
z
Z3 y x V2
i2
i1 xz Z2 y
w w
1 y xz
V1
b
1 i1
y
zw x
x
Z2 y zw
Z1
Z3
x
w
y z
2
i2
c
R2
i1 W X
V1 Z WZ i2
X Y
Y V2
C1
Y
R3 Z
W
X
Fig. 3.14 Realization of floating impedance converters/inverters. (a) Floating GPII/GPIC

configuration, (b) Configuration for realizing positive/negative generalized floating impedances,
(c) A circuit for lossless FI simulation
Z2
X
WZ
Y 2
W
1 Z i2 V2
V 1 i1
X Y Z1
Z5
Y
Z
W
X
Z4
Y Z
Z3 W
X
Fig. 3.15 A floating GIC using CFOAs (adapted from [38] # 2008 Taylor & Francis)
Z3
i2 ¼ i1 ¼ ðv 1 v2 Þ (3.16)
Z1 Z2
and therefore, realizing a floating impedance of value (Z1Z2/Z3) between ports

1 and 2.
An alternative FI realization from [22], having the same characterization as in
(3.16) but with a new feature of having one of the impedances grounded, is shown
in Fig. 3.14b.
Note that with Z1 ¼ R1, Z2 ¼ R2 and Z3 ¼ 1/sC3 the circuit permits the reali-
zation of a lossless floating inductor employing a grounded capacitor (GC) which is
an attractive feature for IC implementation [20, 35]. Another novel feature of this
configuration is that the same structure can be used to realize floating GNII/GNIC
elements too; see [22] by the simple artifice of changing some interconnections.
It must be mentioned that a slightly different formulation, using exactly the same
number of active and passive components, was presented by Chang and Hwang [36]
which was derived from an earlier voltage mode notch, low pass and band pass filter
proposed by them [37] and is shown here in Fig. 3.14c.
A still more generalized floating GIC configuration proposed by Psychalinos
et al. [38] is shown in Fig. 3.15.
The floating impedance simulated by this circuit is given by
Z2 Z3 Z4
Z12 ¼ (3.17)
Z1 Z5
3.6 Floating Inductance Circuits Employing Only Two CFOAs 65
From the above expression, it is easy to visualize that by selecting various

impedances in the circuit appropriately, a variety of floating impedances can be
simulated by this circuit. The following special cases are of practical interest.
1. Floating FDNR is obtained by choosing any two of the three impedances Z2, Z3
and Z4 as capacitors. Taking Z3 and Z4 as capacitors while taking the remaining
impedances as resistors leads to
R2
Z12 ¼ (3.18)
s2 R 1 C 3 C 4 R5
It may be noted that the value of the simulated FDNR is controllable by a single
grounded resistance R5 and no component-matching is required.
2. Floating inductance is realizable with either of Z1 or Z5 selected as a capacitor.
With Z1 taken as a capacitance C1 with all other impedances being resistors, the
equivalent impedance is given by
C1 R2 R3 R4
Z12 ¼ s (3.19)
R5
3. Floating Capacitance is realizable when any one of Z2, Z3 or Z4 is taken as a

capacitance. For instance, taking Z4 as a capacitor, the floating impedance
realized is given by
R2 R3
Z12 ¼ (3.20)
sC4 R1 R5
3.6 Floating Inductance Circuits Employing Only Two CFOAs
We now show how lossless and lossy FIs can be simulated using only two CFOAs.
3.6.1 Lossless/Lossy Floating Inductance Simulator
In this section we present a circuit [39] which employs only two CFOAs along with
only five passive components (namely two capacitors and three resistors) to realize
a lossy/loss-less FI. This circuit is shown in Fig. 3.16. Assuming ideal characteri-
zation of the CFOAs, a straight forward analysis of the circuit reveals its y- matrix
to be given by

1 1 1 1 1
½Y ¼ þ (3.21)
R1 R 2 sC1 R1 R2 1 1
R2
I1 x w
y I2
z R1
+ C1
z y
+
V1 w x
R1
C1
V2
_ _
Fig. 3.16 A floating inductance configuration using only two CFOAs (adapted from [39] # 2012
Springer)
Thus, with R1 < R2 , the circuit simulates floating parallel-RL admittance with
equivalent resistance Req and equivalent inductance Leq are given by
1 1 1
¼ ; Leq ¼ C1 R1 R2 (3.22)
Req R1 R2
On the other hand, with R1 ¼ R2 ¼ R0 , the circuit simulates a lossless FI with
Leq ¼ C1 R0 2 : (3.23)
With the parasitic impedances of the CFOAs accounted for, i.e. considering the
finite input impedance looking into terminal-X as Rx and the output impedance
looking into terminal-Z consisting of a parasitic resistance Rp in parallel with a
parasitic capacitance Cp, it is found that in view of the symmetry of the circuit, the
non-ideal y-parameters are such that Y11 0 ¼ Y22 0 and Y12 0 ¼ Y21 0. The values of these
admittance parameters are found to be.
1 sC1 sC1 Zp
Y11 0 ¼ Y22 0 ¼ þ
R1 1 þ sC1 Zp 1 þ sC1 Zp ðR2 þ 2Rx Þ
Z
þ p (3.24)
1 þ sC1 Zp ðR1 R2 þ 2R1 Rx Þ
Y12 0 ¼ Y21 0
" #
sC Z sC1 Zp Z
¼ 1 p þ p
R1 1 þ sC1 Zp 1 þ sC1 Zp ðR2 þ 2Rx Þ 1 þ sC1 Zp ðR1 R2 þ 2R1 Rx Þ
(3.25)
3.6 Floating Inductance Circuits Employing Only Two CFOAs 67
y R2 C0
I1 zw
x
y C0R1R2
+ R1 w R1
z I2
V1 x
+
- V
-2
Fig. 3.17 A new series-RL type lossy floating inductance simulator
It may be seen that with Zp ! 1, Rx!0, the y-parameters in (3.24) and (3.25)
reduce to those in (3.21).
From the above non-ideal expressions of the y-parameters of the circuit it may be
easily visualized that the high frequency performance would be affected because of
these parasitic impedances. The equivalent non-ideal inductive and resistive
components resulting from all the four y-parameters of (3.24) and (3.25) would
be frequency-dependent.
With Rx ¼ 50 Ω; Rp ¼ 3 MΩ; Cp ¼ 4:5 pF and the circuit designed with C1 ¼ 1 nF;
R0 ¼ 1 kΩ to realize a lossless FI of 1 mH, MATLAB frequency responses of
jY11 j¼ jY22 j; jY12 j¼ jY21 j have shown [39] that in the circuit of Fig. 3.16, the
y-parameters remain intact (and hence, the circuit is useable) up to a frequency of
around 10 MHz. This frequency range of the circuit has also been confirmed from a
SPICE simulation of the circuit for the same component values using a macro model
of AD844.
It is worth pointing out that in [40] Yuce and Minaei have described two FI
circuits using the so-called modified CFOA (MCFOA). Each circuit therein employs
two MCFOAs, two resistors and a single (grounded) capacitor. However, a MCFOA
is not available commercially as an off-the-shelf integrated circuit. Furthermore,
when an MCFOA is implemented with AD844-type CFOAs, as many as three
CFOAs are needed for each MCFOA. Thus, each of the proposed FI circuits
presented in [40] would require six CFOAs. Thus, the present circuit, although
requires two identical capacitors and three resistors, it has the advantage of
employing only two CFOAs.
3.6.2 A Lossy Floating Inductance Simulator
A possible circuit1 for realizing floating series RL impedance is presented here in

Fig. 3.17. By straight forward analysis, it is found that this circuit is characterized by
1
R. Senani and D.R. Bhaskar, ‘New floating lossy inductors, without component-matching,
Employing only two CFOAs’, May 2012 (unpublished).

i1 1 1 1 v1
¼ (3.26)
i2 ðR1 þ sCR1 R2 Þ 1 1 v2
It may be noted that, like the circuit described in the previous sub-section this
circuit also employs only two CFOAs but by contrast, uses a minimum possible
number of (only three) passive elements and has the novel feature of not requiring
any component-matching conditions whatsoever.
3.7 Applications of Simulated Impedances

in Active Filter Designs
We now show how the circuits presented in this chapter can be used in the design of
second order and higher order filters.
3.7.1 Applications in the Design of Second Order Filters
Those circuits which simulate lossless grounded inductance (for example, the two-
CFOA-based circuits described earlier) and lossless floating inductance (such as the
four/three CFOA-based circuits described earlier) can be used as direct replacements
for grounded and floating inductors respectively in the passive RLC prototype second
order filters. The resulting CFOA-based filters will possess the desirable property of
employing grounded capacitors.
It is interesting to observe that in all the four circuits of Fig. 3.8 [25–27] if the only
grounded element therein is ungrounded and a two port network is thus created, the
resulting circuit will have the short circuit admittance matrix of the form
" #
iA y11 y12 vA
¼ 1
þ 1 (3.27)
i1 ðReq þsLeq Þ ðReq þsLeq Þ v1
where the values of y11 and y12 vary from circuit to circuit but they are of no
consequence because if a capacitor is connected from node 1 to ground, since the
impedance looking into node 1 represents a series RL, a low pass filter would be
realizable from all the circuits by connecting a capacitor across node 1 by applying
input at node having voltage VA, as shown in Fig. 3.18 [25–27].
Thus, all the four circuits shown in Fig. 3.18 realize second order low pass filter
function having transfer function
Vo 1
¼ (3.28)
Vin s2 Leq C1 þ sC1 Req þ 1
where Leq and Req for different circuits are same as given in (3.14b).
3.7 Applications of Simulated Impedances in Active Filter Designs 69
a b
R1 V1
V0 R1
i1 V1 i1
X C1 V0 Y
ZW
W
Y C1 X Z
V0 R2
R2 C0
iA
iA VA=Vin
C0
VA=Vin
c
R1
V1 i1
V0 Y
W d
Z
C1 X C0
Y
i1 V1 R1 W
Z
X
C0
R2 R2
C1 V0
iA iA
VA=Vin VA=Vin
Fig. 3.18 Low pass filters based on series-RL type lossy inductance simulators proposed by Yuce
(adapted from [27] # 2012 Springer)
Furthermore, two of these circuits, namely those of Fig. 3.18a, d, permit taking
the output from the low-output-impedance terminal-W of the CFOA and thus,
should be considered to be the best circuits of this set. Experimental results using
AD 844 CFOA demonstrate [25] that the low pass filter of Fig. 3.18 can be readily
used to realize LPF having f0 ¼ 1.59 MHz.
3.7.2 Application in the Design of Higher Order Filters
1. Grounded-capacitor based designs: A given RLC-prototype higher order pas-

sive filter can be readily converted into a CFOA-based active filter by simulating
grounded and floating inductors by lossless grounded inductor circuit of Fig. 3.3
and any of the four-CFOA based or three-CFOA-based FI simulators of Fig. 3.13
RS L1 Lm
+ +
Vin C1 Cn RL V0
− −
Fig. 3.19 Passive RLC prototype higher order low-pass ladder filter
or 3.14b, c. The resulting CFOA based higher order filters would have the
advantage of employing all grounded capacitors.
2. Designs with reduced number of CFOAs: On the other hand, the circuits which
simulate a lossy inductor (series-RL/parallel-RL impedance) or non-ideal FDNR
(series-CD/parallel-CD impedance) in grounded and floating forms can also be
utilized as direct elements in higher order filter designs by using Senani’s
network transformations [41, 42], thereby leading to economic designs requiring
considerably reduced number of CFOAs. This although has been demonstrated
for current-conveyor- based filters [41], unity-gain voltage- follower-based
filters and op-amp-OTA-based filters in [42] however, we now show how to
do this using the CFOA-based circuits described here. Consider now a passive
RLC prototype shown in Fig. 3.19.
If we now apply Senani’s transformation T-4 from the four network transformations
proposed in [42] (T-2 was also proposed independently by Takagi and Fujii in [43])
on this ladder, each impedance is to be multiplied by a frequency-dependent-
scaling-factor

1þs
FðsÞ ¼ (3.29)
s
which transforms a resistor into series-RC impedance, an inductor into series-RL

impedance and a capacitor into a series-CD impedance. The resulting transformed
ladder turns out to be as shown in Fig. 3.20 which realizes exactly the same transfer
function as the original RLC ladder of Fig. 3.19.
A CFOA-based circuit implementation can now be obtained by simulating the
shunt CD-branches and series RL-branches by appropriate CFOA-based
realizations. An exemplary implementation using the floating series-RL circuit of
Fig. 3.17 and grounded series-CD impedance simulator of Fig. 3.10 is shown in
Fig. 3.21.
It is, thus, clear that the above mentioned circuits of lossy floating series-RL
impedance and lossy grounded CD-impedances can be directly used as elements in
the transformed ladder network of type shown in Fig. 3.20, thereby leading to
higher order CFOA-based filter designs with a reduced number of CFOAs, than
those obtainable by simulating lossless floating inductors/FDNRs.
3.8 Realization of Voltage-Controlled Impedances 71
NC NA1
+ +
Vin NB1 ND V0
− −
Fig. 3.20 Transformed version of the passive RLC ladder of Fig. 3.19 as per [42, 43]
Y
W
X Z
NA1
Y
Z W
X
NC
NAm
+ +
Vin Y NBn
W ND V0
X Z
NB1
− −
Fig. 3.21 An exemplary CFOA-implementation of the higher order low pass filter
3.8 Realization of Voltage-Controlled Impedances
Electronically-controlled impedances such as voltage-controlled-resistances (VCR)

and voltage-controlled impedances (VCZ) find applications in automatic gain con-
trol, amplitude stabilization/control in oscillator circuits, design of analog
multipliers/dividers, voltage-controlled filters and voltage-controlled oscillators etc.
[44–57]. Op-amp-FET structures for realizing linearized positive/negative VCRs,
providing wide dynamic range and low distortion, were first presented in [44, 45]
Fig. 3.22 Grounded a

voltage-controlled impedance
configurations (adapted from y Z2
[22] 1998 # Walter de w x
Gruyter GmbH & Co. KG, z z
Zin x w
Germany)
y
Z3
R0
vc
R0
b
y Z2
w x
z z
Zin x w
y
Z1 R0
R0
vc
subsequent to which, generalized linear VCZ configurations were evolved in [46–51,

53, 54, 57]. Consequently a number of configurations, using a variety of active
elements such as op-amps, operational-mirrored-amplifiers, current-controlled-
conveyors, differential voltage current conveyors, and op-amp-OTA combinations,
have so far been presented in the literature for realizing such elements in grounded
and/or floating forms.
In the following, we show how CFOAs in conjunction with JFETs/MOSFETs,
can be employed to realize novel voltage controlled impedances in grounded and
floating and positive as well in negative forms.
3.8.1 Grounded Voltage Controlled Impedance Simulators
Consider now the VCZ structures shown in Fig. 3.22 [22] which are obtained by
modifying the grounded impedance simulation circuits described earlier. Note that
in both the circuits, one of the impedances has been replaced by a FET along with
two equal valued resistors with their junction connected to the gate and the free
ends connected to voltage Vc and the unused W terminal of a CFOA respectively
thereby making the gate voltage as (Vc + VDS)/2 which results in the cancellation
of square non-linearity of the FET thereby realizing a linear voltage controlled
resistor (VCR). Assuming JFET to be confined to operate in the triode region, the
drain current is given by

IDSS v2DS
iD ¼ 2 vGS Vp vDS (3.30)
Vp 2
Since
1
vGS ¼ ðvc þ v1 Þ; vDS ¼ v1 (3.31)
2
Substituting, (3.31) in (3.30) yields the modified input resistance realized by the
FET circuit as
v1 2V 2
rDS ¼ ¼ p (3.32)
iD IDSS vC 2Vp
Thus, the nonlinear term proportional to v2DS in (3.30) is effectively cancelled,

without requiring any additional active devices and as a consequence, the overall
circuit realizes a linear voltage controlled (VC) input impedance (grounded) of value
Z2
Zin ðsÞ ¼ rDS (3.33)
Z3
This circuit can thus, realize linear VC-capacitance (VCC), VC-resistance

(VCR) and VC inductance (VCL) by appropriate selection (resistive/capacitive)
of impedances Z2 and Z3.
Alternatively, Zin (s) ¼ Z1Z2/rDS is obtained from the circuit of Fig. 3.22a by
replacing Z3 in the same manner, as shown in Fig. 3.22b, which can be then used to
realize linear VCR, VCC and VC-FDNR elements (the last one by selecting Z1 and
Z2 both capacitors).
The same techniques applied to the circuit of Fig. 3.4b would yield structures
providing Zin(s) ¼ Z2 rDS/Z3 or Zin ðsÞ ¼ ðZ1 Z2 =rDS Þ respectively thus,
facilitating realization of linear negative VCR, VCC and VCL in the former case
and linear negative VCC, VCR and VC-FDNR in the latter case.
3.8.2 Floating Voltage Controlled Impedance Simulators
We now show a linear VC-floating impedance (VCFI) configuration which is

obtained from the circuit of Fig. 3.14b by replacing Z3 by a FET and using
the W- terminal of the relevant CFOA for the non-linearity cancellation circuitry.
This circuit is shown in Fig. 3.23 and realizes an FI of value Z12 ¼ Z1Z2/rDS. From
this expression it is readily seen that this circuit can realize linear VCR, VCC and
VC-FDNR elements in floating form. A novel feature of this circuit is that from the
same circuit one can realize a VCR, a VCC and VCFDNR elements in floating
forms with negative values also by changing the connections [a1a2, b1b2] to
[a1b2, a2b1].
i1 i2
V1 V2
Z1
Z2 x x
a1 a2 z z
x w w
w y y
z b1 b2
y
R0
vc
R0
Fig. 3.23 Floating Voltage-controlled impedance configuration (adapted from [22] # 1998
Walter de Gruyter GmbH & Co. KG, Germany)
i1
1 v1 y
w 1
a2 z x
R3
x a1
v3 R1
w 3
z y
b2 x
b1
w 2
z y
v3
y i2
4 w 2
r0 Z2 z v2
x
v3
r0 iD Z4
VC
Fig. 3.24 Generalized, floating, linear VCZ configuration by Senani, Bhaskar, Gupta and Singh
(adapted from [57] # 2008 John Wiley and Sons. Ltd)
While the circuit of Fig. 3.23 is capable of realizing floating VCR, voltage-
controlled-capacitance (VCC) and VC-FDNR elements, this circuit, however,
cannot realize floating VCL and VC-FDNC (frequency-dependent-negative-
conductance characterized by Z(s) ¼ Ms2) elements.
A novel configuration which is capable of realizing linear VCR, VCL and VC-FDNC
elements in positive as well as negative, floating as well as grounded - all possible
forms, from the same topology, under appropriate conditions, is shown in Fig. 3.24.
From a straight forward analysis of the circuit of Fig. 3.24, the equivalent
floating impedance realized by the circuit between terminals (1) and (2) is given by:

R 1 R3
Z12 ¼ rDS (3.34)
Z2 Z4
from which linear, floating, positive, VCR, VCL and VC-FDNC elements can be
realized from the circuit by the following choice (resistive/capacitive) of impe-
dances Z2 and Z4:
1. VCR: Z2 ¼ R2 and Z4 ¼ R4
2. VCL: either Z2 ¼ 1/sC2 or Z4 ¼ 1/sC4
3. VC-FDNC: Z2 ¼ 1/sC2 and Z4 ¼ 1/sC4.
It is interesting to mention that the various negative-valued elements
corresponding to the equivalent impedance given in (3.34) can be obtained by the
simple artifice of connecting a1b2 and a2b1 in the circuit of Fig. 3.24, thereby
leading to floating negative impedance given by

R1 R3
Z12 ¼ rDS (3.35)
Z2 Z4
Furthermore, the grounded forms of all the above-mentioned floating

impedances can be obtained by grounding either port-1 or port-2. However, in the
circuit of Fig. 3.24, with port-2 grounded, CFOA2 becomes redundant (y-terminal
of CFOA-3 and R1 can be connected to ground directly) and as a consequence, the
circuit can be simplified to have only three CFOAs while still being capable of
realizing a grounded impedance.

R1 R3
Zin ¼ rDS (3.36)
Z2 Z4
It is important to keep in mind that in order to reduce the effect of various

parasitic impedances of the CFOAs (i.e. finite input resistance Rx (typically,
50–100 Ω) at port x and parasitic impedance Zp at port z which contains a parasitic
resistance Rp (typically, 3 MΩ) in parallel with a parasitic capacitance Cp (typi-
cally, 4.5 pF)), in all the cases, the external circuit impedances are to be chosen such
that they are larger than Rx but smaller than the magnitude of Zp (over the frequency
range of interest).
Floating VCR: The experimentally observed v-i characteristics using AD844 type
CFOAs biased with 15 V DC in conjunction with BFW11 JFETs (with R1 ¼ R3
¼ r0 ¼ 10 kΩ and R2 ¼ R4 ¼ 1 kΩ) shows (Fig. 3.25) the linear range of the
resulting circuit to be nearly 6 V DC which is about two orders of magnitude
larger than that of a conventional FET-VCR.
Floating negative VCR: The v–i characteristic of the negative VCR realized from
the circuit (for the same component values as in positive floating VCR) is shown in
Fig. 3.26, which shows a linear range of the same order ( 6.0 V DC) as that of the
positive VCR (Fig. 3.25).
Fig. 3.25 v–i characteristics of the floating VCR
Fig. 3.26 v–i characteristics of the floating negative VCR
For the details about the workability of this circuit in realizing VC-L, VC-FDNR
and V-FDNC elements and applications thereof, the reader is referred to [57].
Lastly, it may be mentioned that if R1, R3 are replaced by general impedances
Z1
and Z3, the equivalent floating impedance realized becomes Z1 Z3
Z2 Z4 rDS and in
this mode, the circuit2 can be treated as a floating positive/negative generalized-

impedance-converter.
In view of the above novelty, the circuit of Fig. 3.24 can, therefore, be consid-
ered to be a universal linear voltage controlled floating impedance circuit.
In this chapter we presented a variety of CFOA-based circuits for simulating

inductors and other kinds of impedances such as FDNR and FDNC, in grounded/
floating and positive/negative forms. It was shown that CFOA-based grounded
impedance as well as floating impedance circuits require a minimum possible
number of passive components without requiring any component-matching
conditions. It was also shown that most of the economic grounded simulators
realizable with only a single CFOA and three passive components provided an
attractive advantage of single resistance control of the inductance value which
cannot be attained by any single VOA-three passive component circuits such as
those in [5, 7, 8] known in literature. Applications of some of these circuits in the
design of second order and higher order filters were exemplified.
It was also demonstrated that CFOAs provide novel solutions to the realization
of linear voltage-controlled impedances in grounded/floating, positive/negative all
forms, while maintaining the same advantageous features.
Although a new floating series-RL impedance employing only two CFOAs has
been presented here; there could be a family of such circuits which is still waiting to
be discovered! It is believed that the circuits presented in this chapter provide a good
repertoire of impedance simulation circuits which may be employed in the new
designs of filters and oscillators and may also find interesting other applications.
Finally, it may be pointed out that negative capacitance elements appear to have
an application in the area of high frequency oscillator design where the frequency of
oscillation can be scaled up by having in the expression for frequency a capacitive
difference term in the denominator such that by making this difference term small,
the frequency generated by the circuit can be scaled up. On the other hand, a
negative inductance, together with negative FDNR and negative FDNC elements,
are still elements of academic curiosity and have yet to find practical applications.
This constitutes an interesting area for research.
2
This circuit was first reported in R. Senani, ‘Novel linear voltage controlled floating-impedance
configuration’, ELL/96/53450, dated 25th November 1996 (unpublished) and has been subse-
quently published later in [57].
References
1. Mitra SK (1969) Analysis and synthesis of linear active networks. Wiley, New York.
pp 468–469
2. Bruton LT (1980) RC-active circuits theory and design. Prentice-Hall, Inc., Englewood Cliffs,
NJ, pp 145–180
3. Newcomb RW (1968) Active integrated circuit synthesis. Prentice-Hall, Inc., Englewood
Cliffs, New Jersey, p 151
4. Antoniou A (1969) Realisation of gyrators using operational amplifiers, and their use in RC-
active-network synthesis. Proc IEE 116:1838–1850
5. Ahmed MT, Dutta Roy SC (1976) Evaluating the frequency limitations of operational ampli-
fier RC networks: the dominant-pole technique and its application to inductance simulation.
J Inst Electron Telecom Eng 22:703–708
6. Rao KR, Venkateshwaran S (1970) Synthesis of inductors and gyrators with voltage- con-
trolled voltage sources. Electron Lett 6:29–30
7. Ford RL, Girling FEJ (1966) Active filters and oscillators using simulated inductance. Electron
Lett 2:52
8. Presstcott AJ (1966) Loss—compensated active gyrator using differential—input operational
amplifiers. Electron Lett 2:283–284
9. Senani R (1982) Author’s reply in Comments on new canonic active RC realizations of
grounded and floating inductors. Proc IEEE 70:101–103
10. Von Grunigen DC, Ramseier D, Moschytz GS (1986) Simulation of floating impedances for
low-frequency active filter design. Proc IEEE 74:366–367
11. Reddy MA (1976) Some new operational-amplifier circuits for the realization of lossless
floating inductance. IEEE Trans Circ Syst 23:171–173
12. Riordan RHS (1967) Simulated inductors using differential amplifiers. Electron Lett 3:50–51
13. Senani R (1989) Three op-amp floating immittance simulators: A retrospection. IEEE Trans
Circ Syst 36:1463–1465
14. Rathore TS, Singhi BM (1980) Active RC synthesis of floating immittances. Int J Circ Theor
Appl 8:184–188
15. Dutta Roy SC (1974) A circuit for floating inductance simulation. Proc IEEE 64:521–523
16. Wise DR (1974) Active simulation of floating lossy inductances. Proc IEE 121:85–87
17. Sudo S, Teramoto M (1977) Constitution of floating inductance using operational amplifiers.
IEICE Trans E60-E:185–186
18. Senani R, Tiwari RN (1978) New Canonic active RC realizations of grounded and floating
inductors. Proc IEEE 66:803–804
19. Ahmed MT, Dutta Roy SC (1977) A critical study of some non-ideal floating inductance
simulators. Arch Elektron Uevertragungstch (AEU) 31:182–188
20. Bhushan M, Newcomb R (1967) Grounding of capacitors in integrated circuits. Electron Lett
3:148–149
21. Fabre A (1992) Gyrator implementation from commercially available transimpedance opera-
tional amplifiers. Electron Lett 28:263–264
23. Yuce E, Minaei S (2009) On the realization of simulated inductors with reduced parasitic
impedance effects. Circ Syst Sign Process 28:451–465
24. Liu SI, Hwang YS (1994) Realization of R-L and C-D impedances using current feedback
amplifier and its applications. Electron Lett 30:380–381
25. Yuce E (2009) Novel lossless and lossy grounded inductor simulators consisting of a canonical
number of components. Analog Integr Circ Sign Process 59:77–82
26. Abuelma’atti MT (2011) Comment on “Novel lossless and lossy grounded inductor simulator
consisting of canonical number of components. Analog Integr Circ Sign Process 68:139–141
References 79
27. Yuce E (2012) Reply to comment on “Novel lossless and lossy grounded inductor simulator
consisting of canonical number of components”. Analog Integr Circ Sign Process 72:505–507
28. Kacar F, Kuntman H (2011) CFOA-based lossless and lossy inductance simulators.
Radioengineering 20:627–631
29. Abuelma’atti MT (2012) New grounded immittance function simulators using single current
feedback operational amplifier. Analog Integr Circ Sign Process 71:95–100
30. Lahiri A, Gupta M (2011) Realizations of grounded negative capacitance using CFOAs. Circ
Syst Sign Process 30:143–155
31. Senani R (1979) Novel active RC circuit for floating—inductor simulation. Electron Lett
15:679–680
32. Senani R (1980) New tunable synthetic floating inductors. Electron Lett 16:382–383
33. Senani R (1984) Floating ideal FDNR using only two current conveyors. Electron Lett
20(5):205–206
34. Senani R (1986) On the realization of floating active elements. IEEE Trans Circ Syst
33:323–324
35. Senani R (1982) Novel lossless synthetic floating inductor employing a grounded capacitor.
Electron Lett 18:413–414; also see Erratum, ibid, August 1982 issue
36. Chang CM, Hwang CS (1995) Comment on voltage-mode notch, low pass and band pass filter
using current-feedback amplifiers. Electron Lett 31:246
37. Chang CM, Hwang CS, Tu SH (1994) Voltage-mode notch, lowpass and band pass filter using
current-feedback amplifiers. Electron Lett 30:2022–2023
38. Psychalinos C, Pal K, Vlassis S (2008) A floating generalized impedance converter with
current feedback operational amplifiers. Int J Electron Commun (AEU) 62:81–85
39. Senani R, Bhaskar DR (2012) New lossy/loss-less synthetic floating inductance configuration
realized with only two CFOAs. Analog Integr Circ Sign Process 73:981–987
40. Yuce E, Minaei S (2008) A modified CFOA and its applications to simulated inductors,
capacitance multipliers, and analog filters. IEEE Trans Circ Syst-I 55:266–275
41. Senani R (1985) Novel higher-order active filter design using current conveyors. Electron Lett
21:1055–1057
42. Senani R (1987) Network transformations for incorporating nonideal simulated immittances in
the design of active filters and oscillators. Proc IEE Circ Devices Syst 134:158–166
43. Takagi S, Fujii N (1985) High-frequency active RC simulation of impedance-scaled LC filters
using voltage followers. Proc IEEE Int Symp Circ Syst Kyoto, Jpn:295–298
44. Nay K, Budak A (1983) A voltage-controlled resistance with wide dynamic range and low
distortion. IEEE Trans Circ Syst 30:770–772
45. Nay KW, Budak A (1985) A variable negative resistance. IEEE Trans Circ Syst 32:1193–1194
46. Senani R, Bhaskar DR (1991) Realization of voltage-controlled impedances. IEEE Trans Circ
Syst 38:1081–1086, also see ibid, 1991: 39: 162
47. Senani R, Bhaskar DR (1992) A simple configuration for realizing voltage-controlled
impedances. IEEE Trans Circ Syst 39:52–59
48. Senani R, Bhaskar DR (1994) Versatile voltage-controlled impedance configuration. IEE Proc
Circ Devices Syst 141:414–416
49. Senani R (1995) Universal linear voltage-controlled impedance configuration. IEE Proc Circ
Devices Syst 142:208
50. Ndjountche T (1996) Linear voltage-controlled impedance architecture. Electron Lett
32:1528–1529
51. Leuciuc A, Goras L (1998) New general immittance converter JFET voltage-controlled
impedances and their applications to controlled biquads synthesis. IEEE Trans Circ Syst
45:678–682
52. Senani R (1994) Realisation of linear voltage-controlled resistance in floating form. Electron
Lett 30:1909–1911
53. Senani R (1995) Floating GNIC/GNII configuration realized with only a single OMA. Electron
Lett 31:423–425
54. Ndjountche T, Unbehauen R, Luo FL (1999) Electronically tunable generalized impedance

converter structures. Int J Circ Theor Appl 27:517–522
55. Maundy B, Gift S, Aronhime P (2008) Practical voltage/current controlled grounded resistor
with wide dynamic range extension. IET Circ Devices Syst 2:201–206
56. Senani R, Bhaskar DR (2008) Comment on practical voltage/current controlled grounded
resistor with wide dynamic range extension. IEE Circ Devices Syst 2:465–466, also see ibid,
2: 467
57. Senani R, Bhaskar DR, Gupta SS, Singh VK (2009) A configuration for realizing floating,
linear, voltage-controlled resistance, inductance and FDNC elements. Int J Cir Theor Appl
37:709–719
Chapter 4
Design of Filters Using CFOAs
4.1 Introduction
In the area of analog circuit design, considerable attention has been devoted to the
realization of the so-called universal biquad filters using a variety of active elements
such as the classical op-amps, OTAs, various forms of current conveyors and a host of
other building blocks of relatively more recent origin such as OTRAs, CFOAs, CDBAs,
CDTAs and CFTAs etc. The term universal biquad filter, strictly speaking, is supposed
to mean circuits which are capable of realizing, from the same topology, all the five
basic filtering functions namely, Low pass (LP), band pass (BP), high pass (HP), band
stop (BS also referred to as band reject, band elimination or notch filter) and all pass
(AP). However, quite often, some authors also use the term biquad loosely, to refer to
configurations which can realize three (LP, BP and HP) or even two functions only.
Several researchers and practicing engineers have often wondered and even
questioned the utility of a circuit which simultaneously realizes three responses
(usually LP, BP, and HP) from the same circuit arguing that at a given time, after all,
the circuit would be used for realizing only a single type of filter and thus, there may not
be any great utility of simultaneously realizing all the three responses from the same
circuit. It is, therefore, also argued that a circuit realizing three simultaneous responses
is not necessarily better than the one which realizes only a single type of filter response.
In the above context, we would like to point out that simultaneous realization of
several filter functions from the same topology, particularly if the realized
responses are LP, BP and HP, indeed finds many applications such as in phase
locked loops, FM stereo demodulators, touch-tone telephone tone decoders and
crossover networks used in a three-wave high-fidelity loud speaker; see [1]. More-
over, a universal biquad, if available as a standard integrated circuit, gives the
versatility and flexibility of designing any second order or higher order filter using
such biquad filters as standard building blocks. It is worth mentioning that several

82 4 Design of Filters Using CFOAs
such universal filters using op-amps are, indeed, commercially available as ICs, for
example, UAF-42 from Texas Instruments, AF-151 from National Semiconductors
and MF10 from National Semiconductors, to name a few.
Since the advent of CFOAs in the area of analog circuit design, there have been
numerous investigations and proposals for realizing universal filters using CFOAs
as building blocks. The objective of this chapter is to highlight some prominent
filter circuit configurations employing CFOAs. Besides this, some work has also
been done on realizing MOSFET-C biquads and higher order filters using CFOAs
and hence, some prominent results in these two areas would also be highlighted.
4.2 The Five Generic Filter Types, Their Frequency

Responses and Parameters
Before moving further, it is useful to outline the second order transfer functions of
the five standard filtering functions to establish the notations employed and to
understand the basic terminology and the parameters which shall be often used in
the discussion of various circuits in the subsequent sections of this chapter. These
five basic filter functions are as follows:
Low Pass:
H0 ω20
Tð s Þ ¼ (4.1)
s2 þ Qω00 s þ ω20
High Pass:
H 0 s2
Tð s Þ ¼ (4.2)
Band pass:

ω0
H0 Q0 s
Tð s Þ ¼ ω0 (4.3)
s2 þ Q0 s þ ω20
Band stop:

H0 s2 þ ω20
TðsÞ ¼ 2 ω0 (4.4)
s þ Q0 s þ ω20
All pass:

H0 s2 Qω00 s þ ω20
Tð s Þ ¼ (4.5)
4.3 Voltage-Mode/Current-Mode Biquads Using CFOAs 83
It may be noted that in all cases, H0 represents the maximum gain or the gain
factor whereas ω0 represents the central frequency (sometimes also referred as
resonant frequency) in case of BP and BS responses. In case of BP and BS
responses, Qω00 represents the bandwidth. Lastly, Q0 represents the quality factor
which is normally taken as p1ffiffi in case of LP and HP filters to attain maximally-flat
2
response in the pass band.
4.3 Voltage-Mode/Current-Mode Biquads Using CFOAs
The CFOA-based biquad circuits can be classified in two main categories1:

variable-topology type biquads and fixed topology type biquads. The latter
can be further subdivided into two main categories : voltage mode (VM) and
current mode (CM) each of which can be further divided into the following
categories: single input single output (SISO) type, single input multiple output
(SIMO), multiple input single output (MISO) type. Finally, there are universal
mixed-mode biquads which can realize both VM/CM and in an extended case,
even transimpedance type and transadmittance type biquads. In the following
sections, we present prominent circuits in each type chosen from a vast amount
of literature [2–52] existing on the topic. In the literature, various authors have
proposed a number of topologies employing one to five CFOAs exhibiting
different characteristic features. In general, single CFOA-based biquad circuits
are not capable of realizing ideally infinite input impedance in case of voltage-
mode (VM) filter realizations and ideally zero input impedance in case of
current-mode (CM) filter realizations. On the other hand, there are number
of two-CFOA-based VM topologies which possess interesting properties, a
number of these have been included. A number of three-CFOA-based circuits
have also been proposed out of which we have included only those which
employ both grounded capacitors as preferred for IC implementation. Thus,
we have endeavored to include only the most prominent universal biquad
circuits in the following sections.
4.3.1 Dual Function VM Biquads
Two single input dual output biquads introduced by Soliman in [2] derived from
RLC filters, are shown in Fig. 4.1.
A routine circuit analysis (assuming ideal CFOAs) of the circuit of Fig. 4.1a
yields the following transfer functions:
1
In the category of SIMO-type CM biquad, surprisingly, no configuration based on CFOAs is
known to have been published in the technical literature till the time of writing this chapter.
Fig. 4.1 Dual function a

circuits proposed by Soliman
(a) non-inverting low pass y
R1 w - VBP
and inverting band pass x
Vin
filters, (b) non-inverting low z
x
pass and non-inverting band w + VLP
R2 y z
pass filters (adapted from [2]
C2
# 1998 Taylor & Francis)
C1
b R1
x VBP
w y
Vin y + w + VLP
z x
z
R C1 R2 C2
1 1
s
C1 C2 R1 R2 C2 R1
TðsÞjLP ¼ ; TðsÞjBP ¼ (4.6)
1 1 1 1
s2 þ þ s2 þ þ
C 2 R2 C 1 C 2 R1 R2 C2 R2 C1 C2 R1 R2
Similarly, the transfer functions for the circuit of Fig. 4.1b are given by
1 1
s
C 1 C 2 R1 R2 C1 R1
TðsÞjLP ¼ ; TðsÞjBP ¼ (4.7)
1 1 1 1
s2 þ þ s2 þ þ
C1 R C1 C2 R1 R2 C1 R C1 C2 R1 R2
Both the circuits of Fig. 4.1 employ two CFOAs and two GCs and have the
attractive feature in that the parasitic input resistance rx of port-X and the parasitic
output capacitance Cp of port-Z of both the CFOAs can be absorbed in the external
passive elements. On the other hand, the circuit of Fig. 4.1a uses only two resistors
and does not provide infinite input impedance while the circuit of Fig. 4.1b does
provide an infinite input impedance though it employs three resistors. SPICE simula-
tion results given in [2] demonstrate that filters having ω0 of the order of 1 Mrad/s and
Q0 of the order of 10 using AD 844 macromodel show excellent performance.
4.3.2 Single Input Multiple Output (SIMO) Type VM Biquads
Indeed, the state variable Kerwin-Huelsman-Newcomb (KHN) [53] biquad, popu-

larly known as KHN-biquad, which originally employed classical VOAs, has been
not only the first but also one of the most prominent active filter arrangements due
Fig. 4.2 Senani’s low- V03

component-count KHN- r1
equivalent biquad (adapted V01 w
C1
from [3] # 1998 Walter de z
Gruyter GmbH & Co. KG) x x y C2
w R1
z z
x y y R2
w
V02
Vin r2
to its novel feature of simultaneously providing LP, BP and HP filter responses

from the basic three-VOA-structure. With a fourth amplifier added (configured as a
summer) this configuration also makes it possible to realize BS/notch and AP
functions, subject to fulfillment of appropriate conditions. As a consequence, the
technical literature is flooded with a large number of KHN-equivalent biquads
using various building blocks such as OTAs, CCII, CCIII, and numerous others.
In [3], Senani derived a minimum-component CFOA version of the KHN-biquad
which is shown in Fig. 4.2. The circuit realizes a LP response at V01, a BP response
at V02 and a HP response at V03 with the relevant parameters of the realized filters
given by
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r2
ω0 ¼ ðLP; BP; HPÞ (4.8)
r1 C1 C2 R1 R2
rffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C1 R1 r2
Q0 ¼ ðLP; HPÞ (4.9)
C2 R2 r1

ω0 1
Bandwidth ðBWÞ ¼ ¼ ðBPÞ (4.10)
Q0 R1 C 1
H0 ¼ 1ðfor noninverting LP at V03 Þ (4.11)
and
r2
H0 ¼ ðfor non-inverting BP at V02 and noninverting HP at V01 Þ (4.12)
r1
Note that as compared to VOA-based KHN biquad, this circuit has the advantage
of providing ideally infinite input impedance and employing a reduced number of
resistors (only four as against six in the original circuit).
Yet another three-CFOA biquad, which realizes LP, BP and BS functions was
proposed by Bhaskar [4] and is shown in Fig. 4.3. The circuit is, in fact, a single
resistance controlled oscillator (SRCO) in the form shown in Fig. 4.3, but becomes
R4 R1
x
w y
y z w x z
x z R2 w
y
C2
C1
R3
Fig. 4.3 Bhaskar’s SRCO-cum-Multifunction biquad (adapted from [4] # 2003 Walter de
Gruyter GmbH & Co. KG)
a multifunction biquad filter when resistor R2 is disconnected from the junction of

R1, R4 and the w-output of the middle CFOA and an input Vin is applied at the free
end of the resistor R2. In this case, the circuit realizes a BP at the w-output of
the middle CFOA, inverting LP at w-output of first CFOA and inverting BS at the
w-terminal of the third CFOA.
The characterizing parameters of the various filter responses are given by
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
ω0 ¼ (4.13)
C1 C2 R3 R4
rffiffiffiffiffiffiffiffiffiffi
C1 R3
Q0 ¼ (4.14)
C2 R4
1 R1
BW ¼ ðBP and BSÞ and jH0 j ¼ (4.15)
R3 C 1 R2
In case of BP and BS, BW can be adjusted by R3 after which ω0 can be adjusted

through R4 and finally, gain H0 can be adjusted by R1 or R2.
Another three-CFOA-based biquad proposed by Chang et al. [5] is shown in
Fig. 4.4, which realizes BS at V01, LP at V02 and BP at V03. The relevant filter
parameters of this circuit are given by
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 1 C 1 R2 R3
ω0 ¼ ; Q0 ¼ (4.16)
C1 C2 R2 R3 R1 C2
R1
BW ¼ ðBP and BSÞ (4.17)
C 1 R2 R3
R3
HBS ¼ HLP ¼ 1 and jH0 jBP ¼ (4.18)
R1
Fig. 4.4 Voltage-mode BS, C1

LP and BP filter configuration
(adapted from [5] # 1998
x
IEE) z
R3 V03
y w
Vin
R1 C2
z
y y
V01 z w V02
w x x
R2
While the CFOA-version of the KHN biquad of Fig. 4.2 has ideally infinite input
Impedance it does not provide any independent/orthogonal tunability of the various
parameters of the realizable filters. On the other hand, the circuits of Figs. 4.3 and 4.4
although do provide the required tunability of the parameters but do not offer infinite
input impedance.
SPICE simulation of the biquad of Fig. 4.3 [4] and experimental results of the
circuit of Fig. 4.4 [5] demonstrate that using AD844 type CFOAs, the circuits can
be satisfactorily used to design filters having f0 of the order of 100 kHz.
We now present another state-variable biquad circuit proposed by Singh and
Senani [6] which offers infinite input impedance, possesses the feature of the
tunability of the parameters and in addition, makes it possible to apply passive
compensation for the degradation of the high frequency response of the filter in case
of HP (at Vo1) response. This circuit which realizes BP at Vo2 and LP at Vo3 is
shown in Fig. 4.5. The relevant parameters of this circuit are given by
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
R2 C1 R3 R2
ω0 ¼ ; Q 0 ¼ R5 ; BW ¼ (4.19)
C1 C2 R3 R4 R6 C 2 R2 R4 R6 C 1 R3 R5
R6 R5 R2
HLP ¼ ; HBP ¼ and HHP ¼ (4.20)
R1 R1 R1
It may be observed that after fixing ω0, the quality factor Q0 or the BW can be
adjusted by R5 and finally, H0 can be adjusted by R1.
Consider now the effect of the various parasitics, namely, the finite input
resistance rx at port- X and compensation pin parasitics (Rp | | 1/sCp) at port-Z.
Since resistors R1, R3 and R4 are connected at port-X of CFOAs, rx of the respective
CFOA can be easily accommodated in these resistors. Similarly, the Z-port para-
sitic capacitances of CFOA-II and CFOA-III can be easily accommodated in C1 and
C2 respectively. A non-ideal analysis of the integrator made from CFOA-II gives
the transfer function
R1
x V03
I w y V02
y II w y
Vin z x III w V01
z
x z
R3 R4 C2
C1
R2 z R5
w x
IV
y
R6
Fig. 4.5 The CFOA-based state-variable biquad proposed by Singh and Senani (adapted from [6]
# 2005 IEICE)
1 1
Tð s Þ ¼ h i ; (4.21)
sðC1 þ Cp2 Þ þ 1
ðR3 þ rx2 Þ sðC1 þ C p2 ÞðR3 þ rx2 Þ
Rp2

provided ω C1 þ Cp2 R1p2 .
With C1 ¼ 1 nF, Cp3 ¼ 4.5 pF and Rp3 ¼ 3 MΩ, the above constraints implies
f >> 53 Hz, which does not appear to be very restrictive.
It has been shown in [6] that the effect of Z-port parasitics of CFOA-I and
CFOA-IV constituting the summer, can be accomplished by shunting resistors R1,
R5 and R6 by small external capacitors Cc1, Cc5 and Cc6 as shown in Fig. 4.6.
An analysis of this circuit reveals that the output of CFOA-I is now given by

Vin ðsCc1 R1 þ 1Þ Vo2 ðsCc5 R5 þ 1Þ Vo1 ðsCc6 R6 þ 1Þ

Vo3 ¼
R1 R5 R6
( ) (4.22)

Rp1 Rp2 kR2

sðCp1 þ Cp2 ÞðRp1 Rp2 kR2 Þ þ 1
Thus, if we select Cc1, Cc5 and Cc6 such that

Cc1 R1 ¼ Cc5 R5 ¼ Cc6 R6 ¼ Cp1 þ Cp2 ðRp1 Rp2 kR2 Þ (4.23)
then (4.22) reduces to
R2 R2 R2
Vo3 ¼ Vin Vo2 Vo1 ; assuming ðRp1 Rp2 kR2 Þ ffi R2 (4.24)
R1 R5 R6
Cc1 R1
Cc5
x w
I V03
Vin y x V02
z w R5
IV
z y R6
V01
R2 Cp1 Rp1 Cp2
Rp2
Cc6
Fig. 4.6 A simple passive-compensation of the summer (adapted from [6] # 2005 IEICE)
Fig. 4.7 Frequency response of the HP filter realized from the biquad of Fig. 4.5 with and without
compensation (adapted from [6] # 2005 IEICE)
which is the ideal value of Vo3 and thus, perfect compensation for the Z-port
parasitics of CFOA-I and CFOA-IV would be achieved subject to the satisfaction
of the conditions given in equation (4.23). The SPICE simulations of the uncom-
pensated and compensated designs as contained in [6] are shown in Fig. 4.7 from
where it is seen that the passive compensation is able to extend the operational
frequency range by almost one decade.
A very interesting circuit shown in Fig. 4.8 was advanced by Soliman in [7]
which although uses five CFOAs but has the following novel features (1) realizing
three basic filter functions namely LP, BP and HP at V03,V02 and V01 respectively
(2) providing ideally infinite input impedance and zero output impedance at all the
three outputs (3) providing controllability of ω0 and Q0 (or bandwidth) and (4) using
all grounded passive elements thereby making the circuit attractive for IC
implementation.
Fig. 4.8 Soliman’s non- y V01

Vin V02
inverting HP-BP-LP filter w y
x w y
using all grounded passive z x z w V03
elements (adapted from [7] x z
# 1996 Springer) Ri R R1 R2 C2
C1
y y
w w
x x
z
z
R4 R3
The various filter functions realized by the circuit of Fig. 4.8 are given by
R 2 R
V01 R s V02 C1 R1 Ri s
¼ i ¼ (4.25)
Vi DðsÞ Vi DðsÞ
R
V03 C1 C2 R1 R2 Ri
¼ (4.26)
Vi DðsÞ
R R
where DðsÞ ¼ s2 þ sþ (4.27)
C 1 R1 R4 C1 C2 R1 R2 R3
The parameter ω0 and Q0 of the realized filters are given by

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
R C1 R1
ω0 ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi and Q0 ¼ R4 (4.28)
C1 C2 R1 R2 R3 C 2 R2 R3 R
It may be noted from equation (4.28) that ω0 and Qω00 can be adjusted indepen-
dently; the former, by R2 or R3 and subsequently, the latter by R4.
Two universal VM biquad configurations, each employing five CFOAs and eight
passive elements and possessing the unique feature of providing all the five
standard filters at five different output terminals were introduced by Abuelma’atti
and Alzaher in [8, 9]. Here, we present one of these circuits which has the advantage
of offering ideally infinite input impedance and the tunability of the various filter
parameters. The circuit requires only a single matching condition in case of AP
(at Vo5) response (R1 ¼ R8). This configuration which realizes BS at Vo1, LP at
Vo2, BP at Vo3 and HP at Vo4 is shown in Fig. 4.9.
The various characteristic parameters of this configuration are given by
1 R2
ω0 ¼ ; BW ¼ (4.29)
C4 C6 R3 R5 C 6 R1 R5
Fig. 4.9 Universal filter R1

structure introduced by x
w V05
Abuelma’atti and Al-zaher y z
Springer) R8
x
w V01
Vin y y
z w V03
x z
R2
R5 C6
x y
w V04 w V02
y z x z
R3
R7 C4
R2 R2 R7
HLP ¼ ¼ HBS ; HBP ¼ 1 ¼ HAP and HHP ¼ (4.30)
R1 R1 R5
Hence, this circuit enjoys grounded-resistance-controllability of ω0 by R3

while realizing all the five standard functions from five different low-output-
impedance output terminals. Experimental results of the circuit realized with
AD844 type CFOAs [8] demonstrate that the circuit works well in all the five
modes with f0 of the order of 100 kHz realizable quite well.
4.3.3 Multiple Input Single Output (MISO) Type VM Biquads
A variety of circuits are available in literature which realize MISO-type VM

biquads using one to three CFOAs. In the following, we have included some
prominent configurations from among those available in [10–15].
We first consider an interesting MISO-type VM biquad using a single CFOA
proposed by Horng et al. [10], as shown in Fig. 4.10.
Assuming ideal CFOA, the output voltage V0 of the circuit, in terms of various
input voltages is given by

s V3 þ s
2
V4 þ1
R 1 C2 V1 1
C1 R2 V2 þ C1 C21R1 R2 V1
1
C2 R3
V0 ¼ (4.31)
s þ s R1 C2 þ C1 R2 C2 R3 þ R1 R21C1 C2
2 1 1 1
From (4.31), the various filter responses can be obtained by proper selection of
inputs as follows
Fig. 4.10 VM universal V0

biquadratic filter by Horng C1
et al. (adapted from [10]
# 2002 IEICE)
R2
V1 y
w
V2 x
R3 z
V3
C2
V4
R1
(1) LP: if V4 ¼ V3 ¼ 0, C1 ¼ C2, R2 ¼ R3 and V1 ¼ V2 ¼ Vin (2) BP: if

V1 ¼ V3 ¼ 0 and the input signal is either V2 or V4 (3) HP: if V1 ¼ V2 ¼ V4
¼ 0 and V3 ¼ Vin (4) BS: if V4 ¼ 0, C1 ¼ C2, R2 ¼ R3 and V1 ¼ V2 ¼ V3 ¼ Vin
(5) AP: if V1 ¼ V2 ¼ V3 ¼ V4 ¼ Vin and R2 ¼ R3 with C1 ¼ C2
The filter parameters ω0 and Q0 are given by
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 R3 C1 C2 R1 R2
ω0 ¼ and Q0 ¼ (4.32)
C1 C2 R1 R2 C1 R2 R3 þ C2 R1 R3 C1 R1 R2
This circuit does not provide a low impedance output node, hence would require
a voltage follower (i.e., another CFOA) to be able to connect load impedance
without altering the transfer function. This difficulty is, however, overcome in
some of the available two-CFOAs-based biquads. There are a number of MISO-
type universal filters available in literature employing two CFOAs [11–14].
A MISO-type universal filter proposed by Abuelma’atti and Al-Shahrani in [11]
which, indeed, offers ideally infinite input impedance though only in one case, is
shown in Fig. 4.11.
Assuming ideal CFOAs, the output voltage V0 of the circuit in terms of the input
voltages is given as
1 1
s 2 V3 s V2 þ V1
R1 C 2 R1 R3 C1 C2
V0 ¼ (4.33)
1 1
s2 þ s þ
C2 R2 R1 R3 C1 C2
From (4.33), the various filter responses can be obtained by proper selection of
inputs as follows: (1) LP: if V2 ¼ V3 ¼ 0 and V1 ¼ Vin (2) BP: if V1 ¼ V3 ¼ 0
and V2 ¼ Vin (3) HP: if V1 ¼ V2 ¼ 0 and V3 ¼ Vin (4) BS: if V2 ¼ 0 and V1 ¼
V3 ¼ Vin (5) AP: if V1 ¼ V2 ¼ V3 ¼ Vin and R1 ¼ R2
Fig. 4.11 Low component R3

universal filter proposed
by Abuelma’atti and
x
Al-Shahrani (adapted w y
from [11] # 1996 y w V0
V1 z V2 x
Taylor & Francis) z
R1
C1
C2
R2
V3
Fig. 4.12 Universal filter R4

proposed by Abuelma’atti
and Al-Shahrani (adapted
from [12] # 1997 Hindawi y R5
Publishing Corporation) R1 w x
x w
V1 + y +
z z
V2
+ C2 V0
V3
+ C3
-
-
The various filter parameters are given by

1 C2 1
ω0 ¼ ; Q 0 ¼ R2 and BW ¼ (4.34)
C1 C2 R1 R3 C 1 R1 R3 C 2 R2
Thus, although this circuit does offer independent tunability of ω0 and BW

independently, the former by R1 and/or R3 and the latter by R2, it does not have
ideally infinite input impedance in case of realizing BP, BS and AP filters. Neither
the circuit retains both capacitors grounded in all the cases.
The same authors in [12] presented yet another biquad using exactly the same
number of active and passive components which is shown in Fig. 4.12.
A straight forward analysis of this circuit gives
1 1
s 2 V3 s V2 þ V1
C3 R5 C2 C3 R1 R5
V0 ¼ (4.35)
1 1
s2 þ s þ
C3 R4 C2 C3 R1 R5
Fig. 4.13 Universal filter

proposed by Liu and Wu
(adapted from [13] # 1995 R3
IEEE)
y R1
R2 w x
Vin1 x z w
y +
+ z
C1 C2 V0
R4
Vin2
+
Vin3
+ -
-
The filter parameters ω0, BW and Q0 of this circuit are given by

1 1 C3
ω0 ¼ ; BW ¼ and Q0 ¼ R4 (4.36)
C2 C3 R1 R5 C3 R4 C2 R1 R5
This circuit can realize all the standard filter functions by proper selection of
inputs but does not have infinite input impedance in any of the modes. However, LP
realization employs both grounded capacitors as desirable for IC implementation.
This circuit enjoys independent controllability of ω0 and BW and orthogonal tuning
of ω0 and Q0.
Experimental results based on AD 844 CFOAs show [12] that the circuit can be
used to realize BP and Notch filters with f0 of the order of 1 MHz quite
satisfactorily.
Figure 4.13 shows yet another biquad proposed by Liu and Wu [13] using two
CFOAs and six passive elements which realizes all the five standard filter
responses.
The expression for the output voltage in terms of its input voltages for this circuit
is given by
1 1
s2 Vin3 s Vin2 þ Vin1
C2 R4 C1 C2 R1 R2
V0 ¼ (4.37)
1 1
s2 þ s þ
C2 R3 C1 C2 R1 R2
From (4.37), it is clear that the various filter responses can be obtained by proper
selection of inputs.
The filter parameters ω0, BW and Q0 of this biquad are given by
1 1 C2
ω0 ¼ ; BW ¼ and Q0 ¼ R3 (4.38)
C1 C2 R1 R2 C2 R3 C1 R1 R2
Fig. 4.14 MISO-type biquad R5

deduced from a Mason graph
C2
by Wu et al. (adapted from R4
[14] # 1994 Taylor & V3
Francis) x
w y z
y w V0
V1 z x
C1 R3
V2
Thus, Q0 is independently tunable through resistor R3. However, none of the

filters have ideally infinite input impedance.
Another universal biquad employing two CFOAs, two capacitors and three
resistors, deduced from a Mason graph, was presented by Wu et al. in [14] which
is shown in Fig. 4.14.
The expression for V0 in terms of its input voltages for this circuit is given by
1 R3 þ R4
s 2 V3 s V2 þ V1
C2 R3 C1 C2 R3 R4 R5
V0 ¼ (4.39)
1 R3 þ R4
s2 þ s þ
C2 R4 C1 C2 R3R4 R5
The various filter parameters ω0, BW and Q0 of this biquad are given by
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
R3 þ R4 1 C2 ðR3 þ R4 Þ
ω0 ¼ ; BW ¼ and Q0 ¼ R4 (4.40)
C1 C2 R3 R4 R5 C 2 R4 C1 R3 R4 R5
A high input impedance biquad using two CFOAs proposed by Liu [15] is shown
in Fig. 4.15. Although this circuit does not belong to VM MISO-type category,
however, it provides infinite input impedance to realize the three filter functions.
The various filter responses can be obtained by proper selection of admittances
y1, y2, y3 and y4 as follows:
LP at node Vo2: (1) y1 ¼ 1/R1, Y2 ¼1/R2, y3 ¼ (sC3 + 1/R3) and y4 ¼ sC4 (2)
y1 ¼ 1/R1, Y2 ¼1/R2, y4 ¼ (sC4 + 1/R4) and y3 ¼ sC3
BP at node Vo1: (1) y1 ¼ sC1, y2 ¼ (sC2 +1/R2), y3 ¼ 1/R3 and y4 ¼ 1/R4 (2)
y1 ¼ 1/R1, y2 ¼ (sC2 +1/R2), y3 ¼ sC3 and y4 ¼ sC4
HP at node vo2: (1) y1 ¼ sC1, y2 ¼ sC2, y3 ¼ (sC3 + 1/R3) and y4 ¼1/R4 (2)
y1 ¼ sC1, y2 ¼ sC2, y3 ¼ sC3 and y4 ¼ (sC4 + 1/R4)
The circuit provides orthogonal control of ω0 and Q0 by grounded resistors or
capacitors which makes it suitable for easy conversion into voltage tuned filter by
replacing grounded resistor by a JFET.
Y3
Y4
y z
Vin yz w V02
w x
x V01
Y2
Y1
Fig. 4.15 High input impedance biquad introduced by Liu (adapted from [15] # 1995 IEE)
R3
x R1
w x R2
y w x
V1 z y w V0ut
V2 z y
V3 z
C1
C2 R4 R5
Fig. 4.16 Topaloglu-Sagbas-Anday configuration (adapted from [16] # 2012 Elsevier)
The biquad presented by Topaloglu et al. in [16] (shown in Fig. 4.16) is a three-
input single-output second-order universal filter which realizes all the five basic
filtering functions, by selecting different input signal combinations using three
CFOAs, five resistors and two GCs as preferred for integrated circuit implementa-
tion. The circuit offers the attractive feature of providing ideally-infinite input
impedance at all the three input terminals.
The output voltage Vout of this circuit, in terms of the various input signals, is
given by

R5 2 R5 R5 R5
V3 s þs V2 s þ V1
R2 C2 R2 R4 C2 R1 R2 C1 C2 R1 R2 R3
Vout ¼
1 R 5
s2 þ s þ
R4 C 2 C1 C2 R1 R2 R3
(4.41)
Fig. 4.17 An alternative R4

three-input-three-output
universal biquad
x1 R1
w1
z1 x2 Vo R2
y1 Vo1 w2 2
z2 x3
Vin1 y2 w3
Vin z3 Vo3
2 y3 2
C1 Vin3
C2
R3
The various filter responses can be realized as follows: (1) HP: V1 ¼ 0, V2 ¼

V3 ¼ Vin and R1 ¼ R4 (2) BP (inverting): V1 ¼ V3 ¼ 0, V2 ¼ Vin (3) LP: V2 ¼
V3 ¼ 0, V1 ¼ Vin (4) BS: V1 ¼ V2 ¼ V3 ¼ Vin, R1 ¼ R4 and R2 ¼ R5 (5) AP:
V1 ¼ V2 ¼ V3 ¼ Vin, 2R1 ¼ R4 and R2 ¼ R5
Note that this circuit although employs three CFOAs but offers (1) ideally
infinite input impedance in all the cases (2) employment of both grounded
ω0
capacitors and (3) tunability of the parameters ω0 and Q0 but needs realization
constraints in case of HP, BS, and AP responses. SPICE simulations have
demonstrated [16] the workability of this structure in realizing corner/centre
frequencies between 75 and 80 kHz using AD844-type CFOAs.
An alternative MISO-type VM biquad configuration is, derivable from the KHN-
equivalent biquad of [3] shown in Fig. 4.2 of this chapter by un-grounding the
y-input terminals of CFOA1 and CFOA2 and applying two additional inputs on
the terminals thus created. The resulting circuit, which uses the same number of
CFOAs and GCs as in the circuit of Fig. 4.16 but requires one less resistor, is shown
here in Fig. 4.17.
By straight forward analysis, the output voltage Vo1 of the circuit of Fig. 4.17 is
given by

R3 1 R3 R3
s 2
Vin1 þ s þ Vin2 s Vin3
R4 R1 C 1 C C R RR C2 R2 R4
Vo1 ¼ 1 2 1 2 4
1 R3
s2 þ s þ
R1 C1 C1 C2 R1 R2 R4
(4.42)
Note that with Vin2 ¼ 0 ¼ Vin3, Vin1 ¼ Vin, the circuit becomes exactly same as
the circuit of Fig. 4.2 and realizes a LP at Vo3, BP at Vo2 and HP at Vo1 without
requiring any realization constraints.
R3
x
w 4
z y
R1 Vin4
x R2
1 w x
y z 2 w x
y z R4 3 w Vout
Vin1 C1
y z
Vin2 C2
Vin3 R5
Fig. 4.18 A multiple-input single output type universal biquad proposed by Nikoloudis and
Psychalinos (adapted from [17] # 2010 Springer)
However in the modified form of Fig. 4.17, the additional filter functions BR and
AP are realizable as follows:
BR: Vin2 ¼ Vin3 ¼ Vin1 ¼ Vin, taking output at Vo1 with C1 R1 ¼ C2 R2 and
R3 ¼ R4
AP: Vin2 ¼ Vin3 ¼ Vin1 ¼ Vin, taking output at Vo1 with C1 R1 ¼ 2C2 R2 and
R3 ¼ R4
A careful comparison of this circuit with that of Fig. 4.16 reveals the following:
(1) The circuit of Fig. 4.17 employs one less resistor than the circuit of Fig. 4.16 (2)
the circuit of Fig. 4.16 requires realization conditions in case of HP, BS2 and AP
whereas the circuit of Fig. 4.17 requires realization conditions only in BS and AP.
Furthermore, like the circuit of Fig. 4.16, this circuit also provides ideally infinite
input impedance in all the cases.
Lastly, it may be pointed out that this version has not been described earlier in
[3] or elsewhere.
We now present a MISO-type universal biquad proposed by Nikoloudis and
Psychalinos [17], which is shown in Fig. 4.18.
The expression for the output voltage, in terms of the input voltages for this
circuit, is given by

R5 R5 1 1 R5
s2 vin3 s vin2 vin4 þ vin1
R4 R4 R2 C 2 R3 C 2 R1 R2 R4 C1 C2
vout ¼ (4.43)
R5 R5
s2 þ sþ
R3 R4 C2 R1 R2 R4 C1 C2
2
A re-analysis of the circuit of Fig. 4.16 reveals that this circuit needs two realization constraints
G1 ¼ G4 and G2 ¼ G5, in case of BS filter realization also, which appear to have been missed in
[16] inadvertently.
Fig. 4.19 A MISO-type CM y

universal biquad introduced
C1 ix w
by Sharma and Senani z
(adapted from [21] # 2004 x
Taylor & Francis)
io
C2
i3 R3 i2 R2 i1 R1
The various filter functions obtained are as follows: LP: if vin2 ¼ vin3 ¼ vin4
¼ 0 and vin1 ¼ vin, HP: if vin1 ¼ vin2 ¼ vin4 ¼ 0 and vin3 ¼ vin, BP: if vin1 ¼
vin2 ¼ vin3 ¼ 0 and vin4 ¼ vin (non-inverting) or: vin1 ¼ vin3 ¼ vin4 ¼ 0 and
vin2 ¼ vin (inverting), BS: if vin1 ¼ vin3 ¼ vin, vin2 ¼ vin4 ¼ 0, AP: if vin1 ¼ vin2
¼ vin3 ¼ vin and vin4 ¼ 0. In addition, R2 ¼ R3 and R4 ¼ R5.
It is interesting to note that this circuit, except having two realization constraints
in case of AP, has all the desirable properties namely, (1) infinite input impedance
in all the cases, (2) employment of both grounded capacitors, (3) independent single
resistance-controllability of ω0 and Qω00 and (4) realisability of all the five standard
filter responses. SPICE simulations show [17] that the configuration successfully
realizes filters with f0 in the vicinity of 100 kHz or so.
4.3.4 MISO-Type Universal Current-Mode (CM) Biquads
A MISO-type of universal CM biquad can be realized with only a single CFOA and
a number of such circuits have been advanced by various researchers, for instance,
see [18–22]. In the following, we describe one such circuit which provides a
number of advantageous features as compared to other alternatives. This circuit
from [21] is shown here in Fig. 4.19.
By straight forward analysis, the network function of interest for this circuit is
found to be
h i h i h n o i
i3 s
R 1 C2 þ i2 s
R1 C 2þ R1 R31C2 C1 i1 s2 þ s R31C1 þ C12 R12 þ R13 þ R2 R31C1 C2
io ¼ h n o i
s2 þ s R31C1 þ C12 R12 þ R13 þ R2 R31C1 C2
(4.44)
From (4.44), the various filter functions can be realized as follows:

(1) LP: by setting i1 ¼ 0 and choosing i2 ¼ i3 ¼ iin (2) BP: by setting i1 ¼ i2
¼ 0, and choosing i3 ¼ iin
; (3) HP: by choosing i1 ¼ i2 ¼ i3 ¼ iin, along with
R1 ¼ R2 and R3
R2 ¼ 1 þ CC21 (4) notch: by setting i2 ¼ 0 and choosing i1 ¼ i3 ¼ iin,

along with ¼ 1 þ CC21 þ RR32 and (5) AP: by setting i2 ¼ 0 and choosing i1 ¼ i3
R3
R1

¼ iin, along with RR31 ¼ 2 1 þ CC21 þ RR32
The various parameters of the realized filters are now given by
rffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffi 1
1 R2 C2 R2 C1 R3 C 1
ωO ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ; QO ¼ þ þ (4.45)
R2 R3 C 2 C 1 R3 C1 R3 C2 R2 C 2

R2
HO jLP ¼ ; HO jHP ¼ HO jnotch ¼ HO jAP ¼ 1 (4.46)
R1
and

R1 C2 R1 R1 1

HO BP ¼ þ þ (4.47)
R3 C 1 R3 R2
It is seen that LP and BP responses do not have realization constraints in terms of

component values. As a consequence, from the expression for Q0 it is found that the
upper bound on Q0 for these cases is Q0max ¼ 1/(2√2). In other cases, due to the
realization conditions involving all component values, it may be difficult to achieve
this value of Q0.
The low values of Q0, however, does not hamper the usability of the proposed
circuit in practical applications. One such application is in the realization of
constant-Q graphic equalizers [54], where band-pass and band-reject filters with
Q range varying from 0.1 to 1 are needed for attaining wide-band characteristics. As
another example, simple LP filters realizable with only one active element may also
be used as low-cost anti-aliasing/band-limiting filters [55]. Simulations demon-
strate that like other circuits, f0 of the order of 100 kHz is realizable easily using
AD844 type CFOAs.
An inspection of the expressions for the various parameters of the filters, in
conjunction with the relevant realization conditions, reveals that in all cases,
constant-HO, constant-QO and variable-ωO realizations can be achieved by simul-
taneous variation of all the three resistors i.e. R1 ¼ R2 ¼ R3 ¼ R. If these are
replaced by equal-valued MOSFET-based voltage-controlled-resistances (VCR),
(such as the one proposed by Banu-Tsividis [56]) driven by a common control
voltage Vc, electronic tunability of ωO is achievable.
4.3.5 Dual-Mode Universal Biquads Using a Single CFOA
In this section, we first present two single CFOA-based MISO-type CM/VM

multifunction biquad configurations [19] which are shown in Figs. 4.20 and 4.21
respectively.
Fig. 4.20 MISO-type CM

Multifunction biquad
i2
proposed by Sharma and R2 C2
Senani (adapted from [19]
# 2003 Elsevier)
y
R8 w
z
x
iO
i3 i1
R3 C3 R1
Fig. 4.21 MISO-type VM

Multifunction biquad
introduced by Sharma and
Senani (adapted from [19]
# 2003 Elsevier)
A straight forward analysis of the circuit of Fig. 4.20 reveals that the relation
between the three input currents i1, i2, i3 and the output current io is given by
k2 i2 k3 i3 k1 i1
i0 ¼ (4.48)
DðsÞ
where

1 1 1 1 1 1 1
k2 ¼ þ þ þs þ
C2 C3 R3 R8 C2 C3 R1 R3 R8 C2 R1 R8

1 s
k3 ¼ þ
C2 C3 R2 R8 C3 R8

1 1 1 1 1 1 1
k1 ¼ s2 þ s þ þ þ þ
C 3 R3 R8 C2 R2 C2 C3 R2 R3 R8

1 1 1 1 1 1 1
DðsÞ ¼ s þ s
2
þ þ þ þ
C 3 R3 R8 C2 R2 C2 C3 R2 R3 R8
Table 4.1 Parameters of the filters realizable from the circuits of Figs. 4.20 and 4.21
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
ωo and ωo =Qo are same for all CM and VM filters and are given by ωo ¼ C2 C13 R2 R13 þ R18 and
n o
BW ¼ Qωoo ¼ C13 R13 þ R18 þ C21R2 The values of HO are as follows:
Filter type HO (current-mode) HO (voltage-mode)

Low-pass HoV jLP ¼ R2 HoI jLP
R R R Ro
1þR1 þR8 R1
HoI jLP ¼ 3 3
2
R1 R
R2 1þR8
3

Band-pass 1þR8 C2
R C
HoV jBP ¼ Ro
HoI jBP
HoI jBP ¼
1 3 R2
C2 R R
C3 1þR8 þR8

3 2
High-pass HoI jHP ¼ 1 HoV jHP ¼ Ro

HoI jHP
R1

Notch HoI jNotch ¼ 1 HoV jNotch ¼ R1 HoI jNotch
Ro
Transforming the three current sources i1, i2, i3 along with parallel resistances
R1, R2, R3 into voltage sources v1, v2, v3 with series resistances R1, R2, R3,
connecting a resistor Ro at z and then taking output Vo from the w-terminal of the
CFOA, leads to the multifunction voltage mode biquad shown in Fig. 4.21.
The realization conditions for CM biquad of Fig. 4.20/VM biquad of Fig. 4.21 are
as follows: (1) LP: choosing i1 ¼ 0, i2 ¼ i3 ¼ iin for CM and v1 ¼ 0, v2 ¼ vin, v3

¼ RR32 vin for VM along with 1 þ RR81 ¼ CC23 (2) BP choosing i1 ¼ 0, i3 ¼ i2 ¼ iin

for CM and v1 ¼ 0, v2 ¼ vin, v3 ¼ RR32 vin for VM along with 1 þ RR13 þ RR83 ¼ RR12

(3) HP: taking i1 ¼ i2 ¼ i3 ¼ iin, for CM and v1 ¼ vin, v2 ¼ RR21 v1, v3 ¼ RR31 v1 for

VM along with RR12 2 þ RR83 ¼ 1 þ RR13 þ RR83 and CC23 2 þ RR83 ¼ 1 þ RR81 RR82 (4)

Notch taking i1 ¼ i2 ¼ i3 ¼ iin for CM and v1 ¼ vin, v2 ¼ RR21 v1 , v3 ¼ RR31 v1 for

VM along with 1 þ RR31 þ RR81 ¼ RR32 and CC23 2 þ RR83 ¼ 1 þ RR81 RR82
Table 4.1 shows the values of the various parameters of the realized filters.
It may be seen that the realisability conditions for the various responses in voltage
mode are analogous to those of current mode case (with ij replaced by vj, j ¼ 1–3),
which are also given in Table 4.1. Note that the expression for ωo and ωo =Qo remain
exactly the same in both modes, however in voltage-mode version, the gains Ho in all
the cases become controllable through the resistor Ro. Circuit makes it possible to
realize filters having f0 in the vicinity of 1 MHz [19] using AD 844 type CFOAs.
A limitation of the circuits of Figs. 4.20 and 4.21 is the non-availability of
independent tunability of ωo and ωo =Qo thereby restricting the realizations to low Qo
values. However, this does not hamper the usability of the proposed circuits in
practical applications particularly those outlined in Sect. 4.3.4. Lastly, although no
tunability is available for Ho and BW (or QO), ωO can be tuned electronically if all
the four resistors are replaced by appropriately-valued floating CMOS voltage-

controlled-resistors (VCR) circuits. Results in [19] show that in a specific case,
tunability of f0 over a range of more than a decade (500 Hz–20 kHz) is attainable
with this method.
4.3.6 Mixed-Mode Universal Biquads
The term ‘mixed-mode’ in the context of universal biquads is normally used for
circuits which can realise two or more of the voltage-ratio, current-ratio, trans-
impedance and transadmittance mode filters, from the same configuration under
appropriate conditions.
We now present a configuration which realizes all the five standard filter
responses in both CM and VM [3] and thus, can be called a universal VM/CM
biquad. The starting point of the development is the passive LCR filter of Fig. 4.22a,
which is basically a VM BP filter. Applying the source transformation on this
circuit, the circuit becomes as shown in Fig. 4.22b. If we denote the currents in C0,
L0 and R0 as IC0, IL0 and IR0 respectively, the three current transfer functions can be
obtained as
1
IL0 L0 C0
¼ (4.49)
IIN 1 1
s2 þ sþ
C0 R0 L0 C0
1
s
IR0 R0 C0
¼ (4.50)
IIN 1 1
s þ
2 sþ
C0 R0 L0 C0
and
IC0 s2
¼ (4.51)
IIN 1 1
s2 þ sþ
C0 R0 L0 C0
From (4.49)–(4.51) it is, thus, obvious that the circuit of Fig. 4.22b is an
excellent vehicle for making a CM biquad, by simulating the inductor actively
and by sensing the currents IC0, IL0, and IR0 and making them available at high
output impedance nodes, by using appropriate circuitry. A CFOA-based circuit
which implements these mechanisms is shown in Fig. 4.22c. The various
characterizing equations of this circuit are given by
1

C 0 C 1 R1 R2 VIN s2 VIN
IL0 ¼ IIN ; IC0 ¼ IIN (4.52)
DðsÞ Rs DðsÞ Rs
a b
R0 V0
V0
+
L0 L0
Vin C0 Iin R0 C0
IR0 IC0 ILO
-
c
x
x
x VC0 yz z VL0
C0 yz y
z ICO R1 C1
x y x VR0 ILO
R0 yz R2
Vin IR0 x
y x yz
Iin Rs ICO
Fig. 4.22 Voltage-mode (VM) current-mode (CM) universal biquad filter proposed by Senani
(adapted from [3] # 1998 Water de Gruyter GmbH & Co. KG)
and
s

C0 R0 VIN
IR0 ¼ IIN (4.53)
DðsÞ Rs
where

1 1
DðsÞ ¼ s þ 2
sþ (4.54)
R0 C0 R1 R2 C0 C1
The various filter parameters are given by

1 1
ω0 ¼ for LP, BP, and HP, BW ¼ for BP,
C0 C1 R1 R2 R0 C 0
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C0
Q 0 ¼ R0 for LP; HP and H0 ¼ 1 for LP; BP and HP (4.55)
C 1 R1 R2
From the above equations, it is clear that Q0 is tunable by varying R0 in case of

LP and HP, and for BP, the bandwidth is adjustable by R0 whereas the centre
frequency is tunable with R1 (or R2).
To realize BS filter, we create –iC0 by using an additional CFOA (shown by

dotted lines) as an inverting current follower and then add –iC0 and iL0 which yields
1
s2 þ
C0 C1 R1 R3 VIN
Io1 ¼ IIN (4.56)
DðsÞ Rs
For realizing the AP function, the three current outputs –IC0, IR0 and IL0 need to
be added (i.e., Io2 ¼ IR0 –IC0 + IL0) and requires the node thus created to be treated
as the output terminal. This results in
1 1
s2 s þ
R0 C0 C0 C1 R1 R3 VIN
Io2 ¼ IIN (4.57)
DðsÞ Rs
The structure can be converted into a VM biquad by terminating the various

output currents in to load resistors and then taking the outputs from the z-terminals
of the respective CFOAs. For example, with the z-terminal of CFAOs terminated
into a load RL and voltage output taken from the w-terminal of CFOAs, we obtain
(with IIN ¼ 0) a non-inverting BP function having the transfer function

RL 1
s
VBP Rs C0 R0
¼ (4.58)
VIN DðsÞ
Thus, the same kind of transfer functions are realizable in VM too, except that
each carries a negative sign and, H0 can be adjusted by means of the various
load resistors mentioned above. Experimental results using AD 844 type CFOAs
show the workability of the circuit with f0 around 159 kHz achievable quite
satisfactorily [3].
We now present a novel mixed-mode universal biquad configuration from [23]
which is shown in Fig. 4.23.
This circuit employs two inverting integrators and two specially devised sum-
mers, each realized by a single CFOA. The various transfer functions realizable
from this configuration in its various modes of operation are as follows.
(i) Voltage-mode universal biquad filter: Assuming ideal CFOAs, a routine analysis
of the circuit yields the following five transfer functions

r2 r3

Vo3 H0 ω0 2
r1 r2 R1 C1 R2 C2
¼ ¼ (4.59)
Vin DðsÞ 1 r3
s2 þ sþ
R1 C 1 r2 R1 C1 R2 C2
r2
r6
r1
Vin x R1
w x V02 R2
y y
w x V03 r4
z V01 z w x
y w V04
z y
z
C1
C2 r7
AP BS
r3 r5
Fig. 4.23 CFOA-based mixed mode universal biquad (adapted from [23] # 2005 IEICE)

r3
ω0 s
Vo2 H0 ðQ0 Þs r 1 R1 C 1
¼ ¼ (4.60)
Vin DðsÞ 1 r3
s2 þ sþ
R1 C1 r2 R1 C1 R2 C2

r3 2
s
Vo1 H0 s2 r1
¼ ¼ (4.61)
Vin DðsÞ 1 r3
s2 þ sþ
R1 C1 r2 R1 C1 R2 C2

r3 r3
s þ
2
Vo4 H0 ðs2 þ ω0 2 Þ r r2 R1 C1 R2 C2
¼ ¼ 1 (4.62)
Vin DðsÞ 1 r3
s2 þ sþ
R1 C 1 r2 R1 C1 R2 C2
and

ω0 1 r3
H 0 s2 s þ ω0 2 r3
r1 s 2
s þ
Vo4 Q0 R1 C 1 r2 R1 C1 R2 C2
¼ ¼ (4.63)
Vin DðsÞ 1 r3
s2 þ sþ
R1 C1 r2 R1 C1 R2 C2
with the switch at position ‘AP’ and choosing r2 ¼ r3; r4 ¼ r5 ¼ r6 ¼ r7.

Thus, the circuit realizes a LP filter at Vo3, BP response at Vo2, a HP response at
Vo1, and BR and AP responses at Vo4 under appropriate conditions.
(ii) Current-mode universal biquad filter: With r1 and r7 deleted, the circuit can be
converted into an universal current-mode biquad with ideally zero input impedance
and ideally infinite output impedance. With an input current Iin injected into input
4.4 Active-R Multifunction VM Biquads 107
terminal ‘m’ (x-terminal of the first CFOA) and output current Iout taken out from
the node ‘n’ (z-terminal of the last CFOA) the circuit can realize all the five filter
responses in current mode. The general transfer function for this single-input-
single-output universal current-mode filter is given by

s2 1 1
r3 r6 sþ
Iout C1 R1 r5 C1 C2 R1 R2 r4
¼ (4.64)
Iin 1 r3
s2 þ sþ
C 1 R1 C1 C2 R1 R2 r2
The circuit realizes a LP with r5 and r6 open circuited; a BP with r6 and r4 open
circuited; a HP with r5 and r4 open circuited; a BR with r5 open circuited (along with
r2 ¼ r4 ¼ r6 ¼ r0 (say) thereby yielding H0 ¼ r3/r0) and finally, an AP with r2 ¼ r3
¼ r4 ¼ r5 ¼ r6 yielding H0 ¼ 1.The gains for LP, BP, and HP filters are r3/r4, r3/r5
and r3/r6 respectively.
In the CM biquad, LP and HP responses have only H0 controllable (through r4
and r6 respectively); in BR and AP, H0 is not tunable, however, BW and ω0 can be
independently adjusted (through R1 and R2) respectively and finally, in BP realiza-
tion, BW, ω0 and H0, all are independently controllable (through R1, R2 and r5
respectively).
(iii) Trans-admittance universal biquad filter:
In this mode, we retain the input resistor r1 but take the output Iout from z-terminal
of the last CFOA. The various responses realized and their features are similar to
those of case (ii).
(iv) Trans-impedance universal biquad filter:
In this case, with r1 deleted, the input will be a current Iin, however, the output
voltages will be Vo1, Vo2, Vo3 and Vo4. The realisability conditions, parameters of
filters and their features are similar to those of case (i).
Thus, the configuration of Fig. 4.23 is an universal mixed-mode biquad capable
of realizing all the five standard responses in all the four possible modes.
The hardware implementation of the circuit using AD 844 type CFOAs has been
demonstrated [23] to work well in realizing filters with corner/centre frequencies of
the order of 100 kHz.
4.4 Active-R Multifunction VM Biquads
The active–R biquads utilizing the CFOA-pole have been shown to be superior
alternatives to the active-R circuits designed using the compensation-poles of the
traditional voltage-mode op-amps [24]. In this section, we present two CFOA-pole-
based active-R biquads [24, 25] which overcome the limitations of the op-amp-based
active-R biquads such as strong temperature-dependence of filter center frequency
and the limited dynamic range (due to the finite slew rate of the VOAs).
Fig. 4.24 Non-ideal iy

equivalent circuit of +y
the CFOA including w
the various parasitics 1 +
Cy V0
Ry iz -
1
Vd
ix z
Rx Cp Rp
-x
ix
The non-ideal equivalent circuit of the CFOA is shown in Fig. 4.24, where Rx is
the input resistance at x-port, Rp||1/sCp is the parasitic impedance at the z-port and
Ry||1/sCy represents the parasitic impedance at y-port.
An analysis of the circuit of Fig. 4.24 yields
Vo ðsÞ 1
¼ (4.65)
Vd ðsÞ Rx Cp s þ 1
Rp Cp
Thus, s ¼ 1/CpRp represents the pole of the CFOA.

Figure 4.25 shows an active-R biquad employing two CFOAs and five resistors
proposed by Toumazou, Payne and Pookaiyaudom.
Assuming CFOAs to be characterized by the non-ideal model of Fig. 4.24, a
straightforward analysis of the circuit of Fig. 4.25 shows that this circuit would
give BP response at Vo1 and LP response at V02 with the relevant filter parameters
given by
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 Cz1
ω0 ¼ and Q0 ¼ R3 (4.66)
kR2 R4 Cz1 Cz2 kCz2 R2 R4
Ra
where k ¼
Ra þ Rb
From (4.66) it is clear that Q0 of this biquad can be controlled independently
through R3. The circuit was shown [24] capable of realizing f0 of the order of
1 MHz using AD 846 type dual CFOAs.
Yet another active-R CFOA-based biquad introduced by Singh and Senani in
[25] is shown in Fig. 4.26.
A routine circuit analysis (taking R0 1 ¼ Rp1 which is required to ensure that BP
response contains only the first order term of s in the numerator) analysis shows the
realisability of the LP filter at V01 and BP filter at V02. The various filter parameters
of this circuit are given by
4.4 Active-R Multifunction VM Biquads 109
Fig. 4.25 Active-R VM V01

biquad proposed by Rb
Toumazou et al. (adapted y
from [24] #1995 IEEE) V02
Ra
R3 x
y x
R1 R2
Vin
For BP:

1 1 1
BW ¼ þ (4.67)
Cp23 Rp1 R0 4
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u R0 2

u
u 1þ
t Rp1
ω0 ¼ (4.68)
Cp1 Cp23 Rp1 R0 2

Rp1 Rp1 1
HBP ¼ 0 1 þ 0 (4.69)
R3 R4
where Cp23 ¼ Cp2 + Cp3; R10 4 ¼ R14 þ R1p2 þ R1p3 ; R0 2 ¼ R2 þ Rx2 , R0 3 ¼ R3 þ Rx3
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
!
u
uCp23 1 þ RRp10
For LP: Q0 ¼ t 2
;
Cp1 1 þ Rp10
R4
0
R2 R0 2 1
HLP ¼ 1þ while ω0 remains same: (4.70)
R0 3 Rp1
A comparison of these two biquads reveals that the biquad shown in Fig. 4.26 has
the following novel features : (1) high input impedance is available (2) Y-port
parasitics of all the three CFOAs become ineffective (3) Rxi can be accommodated
in Ri; i ¼ 1–3 and (4) tunability of parameters is available (in case of BP filter, having
set BW by R4, ω0 and HBP can be independently adjusted by R2 and R3, respectively
and in case of LP response, having fixed ω0 by R2, Q0 and HLP can then be adjusted
independently through R4 and R3 respectively). The hardware realizations have
demonstrated that this circuit can also realise filters with f0 of the order of 1 MHz.
V02
w
w z
z x y x
R1 z V01
x y w
R4 y R2
R3
Vin
Fig. 4.26 Active-R biquad proposed by Singh and Senani (adapted from [25] # 2001 IEEE)
4.5 Inverse Active Filters Using CFOAs
In communication, control and instrumentation systems, there are applications

where inverse filters are required to correct the distortion of the signal caused by
the signal processors or transmission system. This correction can be done by using
an inverse filter which is required to have the frequency response as reciprocal of
the frequency response of the system which has caused the distortion. Although
several techniques are known to design such inverse digital filters however, in the
domain of analog signal processing, very few attempts have been made in the past
to synthesize inverse continuous time analog filters, for instance, see [50–52] and
the references cited therein.
In most of the earlier works (see references cited in [50, 51]) quite often four-
terminal floating nullors (FTFNs) have been used as building blocks to realize
inverse active filter. However, so far, FTFNs are not available as off the shelf ICs
whereas CFOAs are. In view of the commercial availability of CFOAs as off-the-
shelf ICs coupled with their popularity, recently, a number of attempts have been
made to realize inverse filters using CFOAs [50–52]. In the following, we show
some exemplary realizations for inverse low pass, inverse band pass, inverse high
pass and inverse band reject filters from [50, 51]. These circuits are shown in
Fig. 4.27.
The transfer functions and the parameters of the filters realized by these circuits
are as follows:
Circuit of Fig. 4.27a:
V0 1
¼ (4.71)
Vin R1 2
s
R0
s 1
s2 þ þ
C1 R2 C1 C2 R2 R3
4.5 Inverse Active Filters Using CFOAs 111
a b
R1
R2 x
R2 w V0
z
y
x x R3
w x Vin y
z w z w V0
w
y y z z
Vin y
C2 x
C1 y
z
R0 w
C1 R1 V0
x
R3 C2
c
Vin y
w x
x
z
R3 z
w V0 d
y
C1 R1
R1 R0
x R2
w x R4
y
z Vin y z w x
w y z w V0
C1 y z
x C2
R0
R2 C2 R3
Fig 4.27 Inverse active filters proposed by Gupta et al. (a) Inverse HP filter (adapted from [50]
#2009 Springer), (b) Inverse low pass filter, (c) Inverse band pass filter, (d) Inverse band reject
filter (adapted from [51] 2009 Taylor & Francis)
R1 1 1
Rin ¼ 1; H0 ¼ ; BW ¼ ; ω0 ¼ (4.72)
R0 C 1 R2 C1 C2 R2 R3
Circuit of Fig. 4.27b:
V0 1
¼ (4.73)
Vin R2 2
1þ
R3 C1 C2 R1 R2
2 2
s2 þ sþ
C 2 R3 C1 C2 R1 R2
R2 2 2
Rin ¼ 1; H0 ¼ 1 þ ; BW ¼ ; ω0 ¼ (4.74)
R3 C2 R3 C1 C2 R1 R2
Circuit of Fig. 4.27c:
V0 1
¼ s (4.75)
Vin R2
R0 C R
2 2
s 1
s þ
2 þ
C2 R2 C1 C2 R1 R3
R2 1 1
Rin ¼ 1; H0 ¼ ; BW ¼ ; ω0 ¼ (4.76)
R0 C 2 R2 C1 C2 R1 R3
Circuit of Fig. 4.27d:
V0 1
¼ (4.77)
Vin R4 2 1
s þ
R0 C1 C2 R1 R2
s 1
s2 þ þ
C2 R3 C1 C2 R1 R2
Condition for realizing the inverse notch response: R4 ¼ R0

1 1
Rin ¼ R3 ; H0 ¼ 1; BW ¼ ; ω0 ¼ (4.78)
C 2 R3 C1 C2 R1 R2
The Inverse filters shown in Fig. 4.27 can be readily implemented to have f0 of
the order of 159 kHz [51]. Recently, a generalized CFOA-based configuration for
realizing inverse filters has also been presented [52] from which, using different
selection of various circuit elements, all the four types of inverse filters can be
realized as special cases.
4.6 MOSFET-C Filters Employing CFOAs
MOSFET-C filters were evolved as fully-integratable continuous-time alternatives

to the clock-frequency-tunable switched-capacitor filters and digital filters to be
implemented in VLSI technology. Traditionally, MOSFET-C filters were derived
from classical op-amp based active-RC filters using dual complementary output
type op-amps, MOS capacitors and MOSFETs as the basic elements [57–60]. Apart
from compatibility with CMOS VLSI techniques, an interesting property of the
MOSFET-C filters is that the various parameters of the realized filters, namely ω0,
Q0 and H0 can be electronically-tuned through external voltages applied at the
gate terminals of the appropriate pairs of MOSFETs replacing a given resistor
in the parent active-RC filter from where a corresponding MOSFET-C filter is
4.6 MOSFET-C Filters Employing CFOAs 113
evolved. Both, balanced-input balanced-output types as well as single-input single-

output type MOSFET-C circuits have been evolved so far. These circuits fall into
the category of ‘externally linear internally non-linear’ (ELIN) class of networks
and the use of the technique so developed was not limited to only filters but was
extendable to other functional circuits such as oscillators, voltage-controlled
amplifiers, automatic gain control circuits and others. Such filters have also been
proposed subsequently using other building blocks such as CCII, CCCS, and
OTRAs for instance, see [61–70] and the references cited therein. In this section,
we discuss a number of MOSFET-C filter configurations based upon the use of
CFOAs as active building blocks.
4.6.1 MOSFET-C Fully Differential Integrators
Two interesting methods of realizing MOS-C lossy and lossless integrators with
in-built mechanism for cancellation of the square nonlinearity of the MOSFETs are
shown in Fig. 4.28a, b. These circuits were proposed by Mahmoud and Soliman
[71]. Assuming MOSFETs to be operating in triode region having equal threshold
voltages (VTH), by a straight forward analysis, using the equation of drain current as

W VDS
ID ¼ μs Cox ðVGS VTH Þ VDS (4.79)
L 2
the transfer function of the circuit of Fig. 4.28a is given by
V0 1
¼ (4.80)
Vi sC1 R1
where R1 ¼ 2K1 ðVG11 VTH Þ is the equivalent resistance of the MOS transistor M1 and

K1 ¼ μn COX WL11
Similarly, the transfer function of the circuit of Fig. 4.28b is given by
1
V0
¼ R 1 C1 (4.81)
Vi s þ R21C1
where R2 ¼ 2K2 ðVG21 VTH Þ is the equivalent resistance of MOS transistor M2

Two alternative circuits were also proposed by the same authors in [72] which
are shown in Fig. 4.29.
a b
Vi Y
Vi Y
Vo W Vo
X ZW X
Z
C1 C1
M2 VG2
VG1 M1 VG1 M1
C1 C1
X X
Z
W Vo Z
Vi Y W Vo
Vi Y
Fig. 4.28 MOS-C integrators proposed by Mahmoud and Soliman (a) lossless integrator, (b)
lossy integrator (adapted from [71] # 1998 Taylor & Francis)
Vo
a b C W
VG1 Z
VG12
C Y X
K1
M1 K2 K1
V1 M3
M2 V2 M1 M2
K1 K1
X Y V1
VG11 VG21
Z M4
VG2 K2
W
VG22
Vo
Fig. 4.29 Alternative MOS-C integrators proposed by Mahmoud and Soliman (a) lossless
integrator, (b) generalized integrator (adapted from [72] # 1999 IEEE)
An analysis of the circuit of the Fig. 4.29a (for matched transistors (M1, M2) and
triode region operation) gives
1
Vo ¼ R
V1 (4.82)
s þ R1
where
1
R¼ (4.83)
KðVG1 VG2 Þ
is the equivalent resistance of matched transistors for ðVGi VTH Þ > max ðV1 ; Vo Þ;
for i ¼ 1, 2
and where K1 ¼ μn Cox W1
L1 is the transconductor parameter of each transistor M1
and M2.
On the other hand, an analysis of the circuit of Fig. 4.29b (for matched transistor
pairs (M1, M2) and (M3, M4)) yields its output voltage as
1 1
V1 þ V2
R1 R2
Vo ¼ (4.84)
1 1
sþ þ
R1 R2
where
1
R1 ¼ (4.85)
K1 ðVG11 VG12 Þ
is the equivalent resistance of matched transistors (M1, M2) for ðVG1i VTH Þ >
maxðV1 ; V0 Þ; for i ¼ 1; 2 and
1
R2 ¼ (4.86)
K2 ðVG21 VG22 Þ
is the equivalent resistance of matched transistors (M3, M4) for ðVG2i VTH Þ >
maxðV2 ; V0 Þ; for i ¼ 1; 2.
4.6.2 MOSFET-C Fully Differential Biquads
A MOSFET-C biquad filter proposed by Mahmoud and Soliman [71] is shown in

Fig. 4.30.
Assuming MOSFETs to be operating in triode region, by a straight forward
analysis of the circuit, the two transfer functions realizable by this circuit are
given by
s 1
VBP VLP R3 R4 C1 C2
¼ R1 C1 and ¼ (4.87)
Vi DðsÞ Vi DðsÞ
where
1 1
DðsÞ ¼ s2 þ sþ (4.88)
R2 C1 R3 R4 C1 C2
VBP
Vi Y Y VLP
W W Y
Z Z Z
X X W Vo
X
C2 C2
VG2
M2
VG1 M1 VG3 M3 VG4 M4
C2 C2
X
X X VLP Vo
Z Z Z W
W VBP W Y
Vi Y Y
Fig. 4.30 MOSFET-C LP/BP filter proposed by Mahmoud and Soliman (adapted from [71]
1 C1
ω0 ¼ ; Q0 ¼ R2 (4.89)
R3 R4 C1 C2 R3 R4 C2
and Ri ¼ 1/Ki(VGi–VTH) for (i ¼ 1, 2, 3, 4) is the equivalent resistance of ith MOS

transistor.
From the above, it is seen that for the realized filters, the parameter ω0 can be
controlled by VG3 and/or VG4 whereas Q0 in case of LP and bandwidth ω0/Q0 in
case of BP can be controlled by external voltage VG2.
4.6.3 MOSFET-C Single-Ended Biquad
A single-ended MOSFET-C biquad was advanced in [72] which uses the lossy
integrator of Fig. 4.29a and lossy summing integrator of Fig. 4.29b and is shown in
Fig. 4.31.
Assuming that all transistors are operating in triode region, the two transfer
functions realized by this circuit are given by
1
s
VBP R 1 C1
¼ (4.90)
Vi 1 1

R2 R1 1
s2 þ s C1 þ
C1 C2 R2 R3
VG22
K1 M1
y
Vi 1 w VHP
xz VG12
K1
M2
VG11 K2 M4
C1
VG13
K3 M5
y VLP
K3 M6 2 w
x z
VG23
K3
M7 C2
K3 M8
VG13
K2 M3
VG21
Fig. 4.31 Another MOSFET-C biquad (adapted from [72] # 1999 IEEE)
1
VLP
¼
C
1 2R1 R3
C
(4.91)
Vi
R2 R1
1 1
1
s þs
2 þ
C1 C1 C2 R2 R3
where
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 C1 R1 2 R2
ωo ¼ ; Q0 ¼ (4.92)
C1 C2 R2 R3 C 2 R3 ð R1 R2 Þ 2
Ri ¼ Ki ðVG1i1VG2i Þ ¼ Ki V1G12i , for i ¼ 1, 2 and 3 is the equivalent resistance of ith MOS

transistor and

W
Ki ¼ μn Cox i for i ¼ 1; 2 and 3 (4.93)
L
For R2 ¼ R3 ¼ R, C1 ¼ C2 ¼ C, the angular frequency and quality factor are

simplified to
1 R1
ωo ¼ and Q0 ¼ (4.94)
RC R1 R
The fully-differential filter of Fig. 4.30 although employs as many as six CFOAs,
four capacitors and only three MOSFETs but has the advantage of providing
orthogonal tunability of ω0 and Q0 and elimination of the common mode noise
due to fully differential nature of the circuit. On the other hand, the configuration of
Fig. 4.31 employs as many as eight MOSFETs, but has only two CFOAs and only
two grounded capacitors. It is a single-ended structure and hence, does not provide
the advantage offered by the fully differential design but there is a considerable
degree of flexibility in tuning the circuit for the desired values of the parameters
ω0 and Q0 through various external control voltages.
Using CMOS CFOAs, the realisability of the filters having f0 ¼ 1 MHz in the
case of the structure of Fig. 4.30 and f0 ¼ 500 kHz in the case of the circuit of
Fig. 4.31, have been successfully achieved [71, 72].
Lastly, it must be mentioned that the MOSFET-C CFOA-based filters have been
essentially evolved for fully integratable electronically tunable filter designs and in
this context, it is, therefore implied that integratable CMOS CFOAs such as those in
[73–75] should be assumed in all the circuits described in this section.
4.7 Design of Higher Order Filters Using CFOAs
Although first order filter sections and universal second order biquadratic filter
functions constitute basic building blocks which by themselves may be adequate for
some filtering applications not having very stringent specifications, on the other
hand, there are many applications in which the rate of rejection/selection in stop
band/pass band offered by a second order filter (i.e., 40 dB/decade) may not be
adequate enough. In such cases, higher order active filters are needed. Higher order
active filters can usually be made from a cascade of first order and/or a number of
second order biquads. Alternatively, biquads can also be employed to synthesize
higher order filters through the so called coupled-biquad topology [76]. Besides
this, there are several other methods of designing higher order filters which are
based upon a direct synthesis of the given nth order transfer function. In this section,
we discuss designing CFOA-based higher order filters.
While there has been a lot of activity on universal VM and CM biquad filter
realization using CFOAs, comparatively only a few researchers have explored the
methods of designing higher order filters using CFOAs. In [77], Acar and Ozoguz
have presented a signal flow graph (SFG) based approach for synthesizing an
arbitrary nth-order transfer function. Rathore and Khot in [78] have given a
systematic method of deriving CFOA-based all-grounded-capacitor filter from
current mode RLC prototype ladders. Said et al. in [79] proposed a new technique
for current mode realization of doubly terminated LC ladder filters in which a
higher order filter is designed by using element transfer method. Besides second
order biquads, third order Butterworth filters have continued to attract the attention
of the researchers from time to time. Nandi et al. [78] presented a CFOA-based
configuration of a third order Butterworth low-pass filter using internal device
4.7 Design of Higher Order Filters Using CFOAs 119
transadmittance and parasitic components along with passive RC elements.

Koukiou and Psychalinos [80] have presented modular filter structures using
CFOAs. More recently, multiple loop feedback filters using CFOAs have been
reported by Katopodis and Psychalinos [81].
In the following, we describe some of the prominent methods for designing VM/
CM higher order filters using CFOAs from amongst those outlined above.
4.7.1 Signal Flow Graph Based Synthesis of nth Order Transfer

Function Using CFOAs
This method involves a signal-flow graph approach, with multiple-loop-feedback

topology, in which any nth-order voltage transfer function using active-RC circuit
employing n grounded capacitors, (n + 2) CFOAs and at most (3n + 2) resistors
(which may be either grounded or virtually grounded) are used.
4.7.2 Doubly Terminated Wave Active Filters Employing

CFOA-Based on LC Ladder Prototypes
This technique [79] relies on current mode realization of doubly terminated LC

ladder filters using CFOA-based linear transformation of port variables. A design
example of 3rd order low pass filter has been considered and compared with the
systematic approach for deriving CFOA-based all-grounded-capacitor filters
from current mode RLC prototype ladder introduced by Rathore and Khot [78].
Figure 4.32 shows the realization of two-port networks for the series and shunt
elements, while Figs. 4.33, 4.34, 4.35, 4.36 show the realization of all possible
combinations for the one port network in the input and output of the doubly
terminated ladder filter using CFOA as an active element.
Figure 4.37 shows the active realisation of a 3rd order low pass filter employing
CFOA-based circuit equivalents for each passive element.
PSPICE simulation results show [79] that this realization has low power dissi-
pation, low total harmonic distortion and smaller active component-count as com-
pared to one introduced in [78] (shown here in Fig. 4.38) The structure of Fig. 4.37
has successfully achieved f0 of the order of 1 MHz.
4.7.3 Higher Order Modular Filter Structures Using CFOAs
CFOAs can be employed to realize higher order wave active filters using lossy
integration-subtraction and simple summation and subtraction building blocks.
a
Y2
Y1 R1
C0 X1 x
R1 w
y
z
C1
C1
X2
b
Y1
Y2
L0 C1
X1 x 2R1
w
y z
C1 2R1
R1 X2
Fig. 4.32 (a) Floating capacitor realization. (b) Floating inductor realization(adapted from [79]
# 2011 Elsevier)
a 2R1
C1
x
w Y2
y z
X1
R1
C0 2R1
Y1
C1
X2
b R1
x
w Y2
y
Y1 z
C1 C1
L0
X1
R1 X2
Fig. 4.33 (a) Grounded capacitor realization. (b) Grounded inductor realization (adapted from
[79] # 2011 Elsevier)
C1 2R
R0 R1 x
w Y1
y z
R2
R1 C1 2R2
E C0 E
X1
Fig. 4.34 Active realisation for input terminal with shunt capacitor and source resistance (adapted
from [79] # 2011 Elsevier)
a b
R2
x
w Y2
y z
C0 R1 C1
RL
X2
Fig. 4.35 Active realisation for output terminal with shunt capacitor and load resistance (adapted
Rs L01
Vin C01 C02 RL V0
Fig. 4.36 Third order Chebyshev lowpass ladder
C1 2R1 R1
x
C1 w V0
x y z
R2 w x 2R1
y z w
y z
R1 2R1 C1
2R1 C1
R2 2R1
E C1
R1
Vi
Fig. 4.37 Active realization of a third order low pass filter using CFOA-based circuits (adapted
Figure 4.39a shows an implementation of lossy integration-subtraction circuit and

Fig. 4.39b gives the realization of subtraction of two input voltages.
Assuming ideal CFOAs, the output voltage Vo for the circuit shown in Fig. 4.39a
is given by
1
V0 ¼ ðVin1 Vin2 Þ where τ ¼ Ra Ca is the time constant (4.95)
1 þ τs
x R6
R7 y w
z
R8
R x
x w y
V0 w y
z
z
R2 R4
R C
x w
y
z
R3
R2
C
R x
x w y
w y
z z
R3
R1 R C
R1
Vi x
w
y
z
R5
Fig. 4.38 Active realization of third order low pass filter using CFOA based circuits (adapted
from [79] # 2008 John Wiley & Sons, Ltd)
Similarly, the output voltage for the circuit of Fig. 4.39b is given by
V0 ¼ ðVin1 Vin2 Þ (4.96)
CFOA-based summation block is shown in Fig. 4.39c for which the output
voltage is given by
V0 ¼ ðVin1 þ Vin2 Þ (4.97)
Using the circuits of Fig. 4.39, the resulting wave equivalent of an inductor in
series-branch is as shown in Fig. 4.40.
The complete set of wave equivalents derived in [80] has been shown in
Tables 4.2 and 4.3 respectively.
For the construction of the complete wave filter (1) equal port resistances are
assumed and (2) cross-cascade connection of the incident and reflected waves has
been considered(because the incident wave at each port is equal to the reflected
wave of the preceding port). For a 3rd order low pass LC ladder filter as shown in
Fig. 4.41, the block diagram representation shown in Fig. 4.42 is derivable.
The validity of this method for a 3rd order low pass filter (cut-off frequency
100 kHz) has been confirmed using CFOA-based wave active equivalents. The
commercially available AD844 devices were used as CFOAs.
a b
Ra
Vin2 x
w V0 Ra
Vin1 y z Vin2 x
w V0
Vin1 y z
Ra Ca
Ra
c
Ra
Vin1
Ra
Vin2 y
w V0
x z
Ra
2Ra
Fig. 4.39 (a) CFOA-based lossy integration-subtraction, (b) CFOA-based subtraction circuit
(adapted from [80] # 2010 Radioengineering), (c) CFOA-based summation block (adapted
from [80] # 2010 Radioengineering)
a Ra
y
y Ra w B2
A1 y w B1 x z
w x z Ra
A2 x z Ra 2Ra
Ra Ra
Ra Ca
A1 T A2
B1 B2
Fig. 4.40 CFOA-based wave equivalent of an inductor in series-branch (a) circuit level represen-
tation, (b) Symbolic notation (adapted from [80] # 2010 Radioengineering)
Table 4.2 Wave equivalents of two-port sub networks in series-branch (adapted from [80] #
2010 Radioengineering)
L
A1 T A2
R R
B1 B2
C
A1 T A2
R R
B1 B2
L C A1 T1 B2
B1 A2
R R T2
L B2
A1 T1
C T2 A2
B1
R R
Due to the fact that the wave equivalents of the passive components can be
realized as manipulated versions of the wave equivalent of an inductor in series-
branch, the resulted filter configurations have modular structures. The design
procedure for obtaining higher-order filters is quite simple as just one step is needed
for the realization of an arbitrary order filter due to availability of basic building
blocks. The drawback of this method is that a more complex circuitry is needed as
compared to filters derived according to the operational or the topological emula-
tion of the corresponding passive proto-type filters.
From the literature survey it has been revealed that not much has been done in
the area of higher order filter design using CFOAs and in fact, only three methods
have so far been advanced in the technical literature which are the SFG based
higher order filter synthesis, doubly-terminated wave active filters and the higher
order modular filters. Thus, there is enough scope for exploring the use of CFOAs in
realizing higher order filters in real life applications.
Table 4.3 Wave equivalents of two-port sub networks in shunt-branch (adapted from [80]
# 2010 Radioengineering)
A1 T -1 A2
C
B1 -1
R R
B2
A1 T -1 A2
L
R R B1 -1 B2
A1 T1 B2
C L
B1 -1
-1 A2
T2
R R
B2
C A1 T1
L
T2 -1 A2
B1 -1
R R
Rs L1 L3
Vin C2 RL Vout
Fig. 4.41 Third order LC ladder prototype filter (adapted from [80] # 2010 Radioengineering)
Vinc
TC2 -1
Vout
Vin TL1 -1
TL2
Fig. 4.42 Block diagram of the corresponding wave active filter (adapted from [80] # 2010
Radioengineering)
In this chapter we discussed a number of circuits and techniques for both VM and CM
universal filters realizable with CFOAs which were selected from a large number of
such circuits [2–49, 85] available in technical literature. While choosing the second
order biquad filter circuits to be included in this chapter, the main focus has been on
only those configurations which provide as many as possible of the following
advantageous features: Ideal input impedance (Rin ¼ 0, for CM-type filters and
Rin ¼ infinite, for VM-type filters), independent/sequential tuning of filter para-
meters ω0, BW, Q0 (and also H0 if possible), use of only two capacitors, ideal output
impedance (R0 ¼ 0, VM-type filters and, R0 ¼ infinite, for CM-type filters) and use
of grounded capacitors as preferred for integrated circuit implementation.
In VM filters, both SIMO-type as well as MISO-type structures were included.
In the category of CM universal biquads however, MISO-type universal biquads
only could be included since any SIMO-type CM universal biquad using CFOAs
has not been reported in the literature so far and this appears to be an interesting
problem for future research. Lastly, two mixed-mode biquads were presented. The
first one can realize both CM and VM responses of all the five standard filters.
On the other hand, the other, a three CFOA-based structure, can realize, in addition,
trans-impedance and trans-admittance type biquads also. Two active-R biquads
realizing LP and BP responses have also been described.
In view of the recent interest in inverse filters, some representative inverse filter
structures were also included.
The authors of this book believe that the area of devising good universal
CM/VM biquads using CFOAs is not exhausted yet and there is ample scope of
devising newer configurations having better features than those available in the
circuits discovered so far [2–49, 85].
In the category of MOSFET-C filters using CFOAs, we have elaborated two
techniques for realizing MOSFET-C biquads using CFOAs. It is obvious that for
fully integrated designs and their implementations CMOS-CFOAs such as those of
[74, 75, 82] would be required.
From the survey of literature, it is found that not much work has been carried out
by on the development of MOSFET-C networks using CFOAs beyond that
contained in references [63, 64, 73, 75, 82, 83] which has been highlighted in this
chapter. One can expect that lot of new configurations and ideas concerning new
CFOA-based MOSFET-C circuits might be waiting to be explored.
From the published works [76–84] it has been revealed that not much has been
done in the area of higher order filter design using CFOAs and in fact, only three
methods have so far been elaborated in the technical literature which are the SFG
based higher order filter synthesis of Sect. 4.7.1, doubly-terminated wave active
filters of section 4.7.2 and the higher order modular filters of Sect. 4.7.3 Thus, there
is enough scope for exploring the use of CFOAs in realizing higher order filters in
real life applications.
References 127
References
1. Ibrahim MA, Minaei S, Kuntman H (2005) A 22.5 MHz current-mode KHN-biquad using
differential voltage current conveyor and grounded passive elements. Int J Electron Commun
(AEU) 59:311–318
2. Soliman AM (1998) Generation of CCII and CFOA filters from passive RLC filters. Int J
4. Bhaskar DR (2003) Realisation of second-order sinusoidal oscillator/filters with non-
interacting Controls using CFAs. Frequenz 57:1–3
5. Chang CM, Hwang CS, Tu SH (1994) Voltage-mode notch, low pass and band pass filter using
current-feedback amplifiers. Electron Lett 30:2022–2023
6. Singh AK, Senani R (2005) CFOA-based state-variable biquad and its high-frequency com-
pensation. IEICE Electron Express 2:232–238
7. Soliman AM (1996) Applications of current feedback operational amplifiers. Analog Integr
8. Abuelma’atti MT, Al-Zaher HA (1998) New universal filter with one input and five outputs
using current-feedback amplifiers. Analog Integr Circ Sign Process 16:239–244
9. Abuelma’atti MT, Al-Zaher HA (1998) New universal filter with one input and five outputs
using current-feedback amplifiers. Proc Natl Sci Counc ROC (A) 22:504–508
10. Horng JW, Chang CK, Chu JM (2002) Voltage-mode universal biquadratic filter using single
current-feedback amplifier. IEICE Trans Fundament E85-A:1970–1973
11. Abuelma’atti MT, Al-Shahrani SM (1996) New universal filter using two current-feedback
amplifiers. Int J Electron 80:753–756
12. Abuelma’atti MT, Al-Zaher HA (1997) New universal filter using two current-feedback
amplifiers. Active Passive Electron Comp 20:111–117
13. Liu SI, Wu DS (1995) New current-feedback amplifier-based universal biquadratic filter. IEEE
Trans Instrum Meas 44:915–917
14. Wu DS, Lee HT, Hwang YS, Wu YP (1994) CFA-based universal filter deduced from a mason
graph. Int J Electron 77:1059–1065
15. Liu SI (1995) High input impedance filters with low components spread using current-
feedback amplifiers. Electron Lett 31:1042–1043
16. Topaloglu S, Sagbas M, Anday F (2012) Three-input single-output second-order filters using
current-feedback amplifiers. Int J Electron Commun (AEU) 66:683–686
17. Nikoloudis S, Psychalinos C (2010) Multiple input single output universal biquad filer with
current feedback operational amplifiers. Circ Syst Sig Process 29:1167–1180
18. Horng JW, Chou PY, Wu JU (2010) Voltage/Current-mode multifunction filters using current-
feedback amplifiers and grounded capacitors. Active Passive Electron Comp 5:785631.
doi:10.1155/2010/785631
19. Sharma RK, Senani R (2003) Multifunction CM/VM biquads realized with a single CFOA and
grounded capacitors. Int J Electron Commun (AEU) 57:301–308
20. Sharma RK, Senani R (2004) On the realization of universal current mode biquads using a
single CFOA. Analog Integr Circ Sign Process 41:65–78
21. Sharma RK, Senani R (2004) Universal current mode biquad using a single CFOA. Int
J Electron 91:175–183
22. Chang CM, Soliman AM, Swamy MNS (2007) Analytical synthesis of low-sensitivity high-
order voltage-mode DDCC and FDCCII-grounded R and C All-pass filter structures. IEEE
Trans Circ Syst-I 54:1430–1443
23. Singh VK, Singh AK, Bhaskar DR, Senani R (2005) Novel mixed-mode universal biquad
configuration. IEICE Electron Express 2:548–553
24. Toumazou C, Payne A, Pookaiyaudom S (1995) The active-R filter technique applied to
current-feedback op-amps. IEEE Int Symp Circ Syst 2:1203–1206
25. Singh AK, Senani R (2001) Active-R design using CFOA-poles: new resonators, filters, and
oscillators. IEEE Trans Circ Syst-II 48:504–511
26. Singh AK, Senani R, Tripathi MP (1999) Low-component-count high frequency resonators
and their applications using op-amp compensation-poles. Frequenz 53:161–169
27. Horng JW, Hou CL, Huang WS, Yang DY (2011) Voltage/current-mode multifunction filters
using one current feedback amplifier and grounded capacitors. Circ Syst 2:60–64
28. Liu SI (1995) Universal filter using two current-feedback amplifiers. Electron Lett 31:629–630
29. Horng JW, Lee MH (1997) High input impedance voltage-mode low pass, band pass and high
pass filter using current-feedback amplifiers. Electron Lett 33:947–948
30. Senani R, Gupta SS (1997) Universal voltage-mode/current-mode biquad filter realised with
current feedback op-amps. Frequenz 51:203–208
31. Soliman AM (1998) A new filter configuration using current feedback op-amp. Microelectron
J 29:409–419
32. Horng JW (2000) New configuration for realizing universal voltage-mode filter using two
current feedback amplifiers. IEEE Trans Instrum Meas 49:1043–1045
33. Abuelma’atti MT, Al-Zaher HA (2000) New grounded-capacitor grounded-resistor controlled
universal filter using current-feedback amplifiers. Proc Natl Sci Counc ROC (A) 24:205–209
34. Horng JW (2001) Voltage-mode multifunction filter using one current feedback amplifier and
one voltage follower. Int J Electron 88:153–157
35. Shah NA, Malik MA (2003) Multifunction filter using current feedback amplifiers. Frequenz
59:264–268
36. Gift SJG, Maundy B (2004) High-performance active band pass filter using current-feedback
amplifiers. Int J Electron 91:563–570
37. Shah NA, Iqbal SZ, Rather MF (2005) Versatile voltage-mode CFA-based universal filter. Int J
Electron Commun (AEU) 59:192–194
38. Mita R, Palumbo G, Pennisi S (2005) Non-idealities of Tow-Thomas biquads using VOA- and
CFOA-based miller integrators. IEEE Trans Circ Syst-II 52:22–27
39. Shah NA, Rather MF, Iqbal SZ (2005) A novel voltage-mode universal filter using a single
CFA. J Active Passive Electron Devices 1:183–188
40. Djebbi M, Assi A, Sawan M (2005) Design of monolithic tunable CMOS band-pass filter using
current feedback operational amplifiers. Analog Integr Circ Sign Process 45:143–154
41. Singh VK, Singh AK, Bhaskar DR, Senani R (2006) New universal biquads employing
CFOAs. IEEE Trans Circ Syst-II 53:1299–1303
42. Sagbas M, Koksal M (2007) Voltage-mode three-input single-output multifunction filters
employing minimum number of components. Frequenz 61:87–93
43. Bhaskar DR, Prasad D (2007) New current mode biquad filter using CFOAs. J Active Passive
Electron Devices 2:292–298
44. Manhas PS, Pal K, Sharma S, Mangotra LK, Jamwal KKS (2007) Realization of high-Q band
pass filter using low voltage current feedback amplifiers. J Active Passive Electron Devices
4:13–20
45. Ferri G, Guerrini N, Piccirilli MC (2003) CFA based fully integrable KHN Biquad. Int Symp
Sig Circ Syst 2:569–572
46. Palumbo G, Pennisi S (1999) Filter circuits synthesis with CFOAs-based differentiators. 16th
IEEE Instrument Measurement Technology Conference (IMTC). pp 546–550
47. Yuce E (2010) Fully integrable mixed-mode universal biquad with specific application of the
CFOA. Int J Electron Commun (AEU) 64:304–309
48. Hou CL, Huang CC, Lan YS, Shaw JJ, Chang CM (1999) Current-mode and voltage-mode
universal biquads using a single current-feedback amplifier. Int J Electron 86:929–932
49. Dostal T (1995) Correspondence: Insensitive voltage-mode and current-mode filters from
commercially available transimpedance op-amps. IEE Proc Circ Devices Syst 142:140–143
50. Gupta SS, Bhaskar DR, Senani R, Singh AK (2009) Inverse active filters employing CFOAs.
Electr Eng 91:23–26
References 129
51. Gupta SS, Bhaskar DR, Senani R (2011) New analogue inverse filters realized with current
feedback op-amps. Int J Electron 98:1103–1113
52. Wang HU, Chang SH, Yang TY, Tsai PY (2011) A novel multifunction CFOA-based inverse
filter. Circ Syst 2:14–17
53. Kerwin WJ, Huelsman LP, Newcomb RW (1967) State-variable synthesis for insensitive
integrated circuit transfer functions. IEEE J Solid State Circ SC-2:87–92
54. Bohn DA (1986) Constant-Q graphic equalizer. J Audio Eng Soc 34:16
55. Baker BC (1999) Anti-aliasing, Analog filter for data acquisition systems. Application notes
no. AN699 of Microchip Tech Inc 1-10
56. Banu M, Tsividis Y (1982) Floating voltage-controlled resistors in CMOS technology. Elec-
tron Lett 18:678–679
57. Banu M, Tsividis Y (1983) Fully integrated active RC filters in MOS technology. IEEE J Solid
State Circ SC-18:644–651
58. Tsividis Y, Banu M, Khoury J (1986) Continuous-time MOSFET-C filters in VLSI. IEEE
Trans Circ Syst 33:125–140
59. Ismail M, Smith SV, Beale RG (1988) A new MOSFET-C universal filter structure for VLSI.
IEEE J Solid State Circ 23:183–194
60. Sakurai S, Ismail M, Michel JY, Sanchez-Sinencio E, Brannen R (1992) A MOSFET-C
variable equalizer circuit with simple on-chip automatic tuning. IEEE J Solid State Circ
27:927–934
61. Liu SI, Tsao HW, Lin TK (1990) MOSFET capacitor filters using unity gain CMOS current
conveyors. Electron Lett 26:1430–1431
62. Liu SI, Tsao HW, Wu J (1991) CCII-based continuous-time filters with reduced gain-
bandwidth sensitivity. IEE Proc Circ Devices Syst 138:210–216
63. Meng XR, Yu ZH (1996) CFA based fully integrated Tow-Thomas biquad. Electron Lett
32:722–723
64. Gunes EO, Anday F (1997) CFA based fully integrated nth-order lowpass filter. Electron Lett
33:571–573
65. Salama KN, Elwan HO, Soliman AM (2001) Parasitic-capacitance-insensitive voltage-mode
MOSFET-C filters using differential current voltage conveyor. Circ Syst Sig Process 20:11–26
66. Schmid HP, Moschytz GS (2000) Active- MOSFET-C single-amplifier biquadratic filters for
video frequencies. IEE Proc Circ Devices Syst 147:35–41
67. Chiu W, Tsay JH, Liu SI, Tsao H, Chen JJ (1995) Single-capacitor MOSFET-C integrator
using OTRA. Electron Lett 31:1796–1797
68. Chen JJ, Tsao HW, Liu SI, Chiu W (1995) Parasitic-capacitance-insensitive current-mode
filters using operational transresistance amplifiers. IEE Proc Circuits Devices Syst
142:186–192
69. Chen JJ, Tsao HW, Liu SI (2001) Voltage-mode MOSFET-C filters using operational
transresistance amplifier (OTRAs) with reduced parasitic capacitance effect. IEE Proc Circ
70. Hwang YS, Wu DS, Chen JJ, Shih CC, Chou WS (2007) Realization of higher-order OTRA-
MOSFET-C active filters. Circ Syst Sig Process 26:281–291
71. Mahmoud SA, Soliman AM (1998) Novel MOS-C balanced-input balanced-output filter using
the current feedback operational amplifier. Int J Electron 84:479–485
72. Mahmoud SA, Soliman AM (1999) New MOS-C biquad filter using the current feedback
operational amplifier. IEEE Trans Circ Syst-I 46:1510–1512
73. Mahmoud SA, Soliman AM (2000) Novel MOS-C oscillators using the current feedback
op-amp. Int J Electron 87:269–280
74. Manetakis K, Toumazou C (1996) Current-feedback op-amp suitable for CMOS VLSI tech-
nology. Electron Lett 32:1090–1092
75. Soliman AM, Madian AH (2009) MOS-C Tow-Thomas filter using voltage op amp, current
feedback op amp and operational transresistance amplifier. J Circ Syst Comput 18:151–179
76. Schaumann R, Ghausi MS, Laker KR (1990) Design of analog filters: active RC and switched
capacitor. Prentice Hall, Englewood Cliffs, NJ
77. Acar C, Ozoguz S (2000) Nth-order voltage transfer function synthesis using a commercially
available active component, CFA: signal-flow graph approach. Frequenz 54:134–137
78. Rathore TS, Khot UP (2008) CFA-based grounded-capacitor operational simulation of ladder
filters. Int J Circ Theor Appl 36:697–716
79. Said LA, Madian AH, Ismail MH, Soliman AM (2011) Active realization of doubly terminated
LC ladder filters using current feedback operational amplifier (CFOA) via linear transforma-
tion. Int J Electron Commun (AEU) 65:753–762
80. Koukiou G, Psychalinos C (2010) Modular filter structures using current feedback operational
amplifiers. Radioengineering 19:662–666
81. Katopodis V, Psychalinos C (2011) Multiple-loop feedback filters using feedback amplifiers.
82. Mahmoud SA, Awad IA (2005) Fully differential CMOS current feedback operational ampli-
fier. Analog Integr Circ Sign Process 43:61–69
83. Soliman AM, Madian AH (2009) MOS-C KHN filter using voltage op amp, CFOA, OTRA and
DCVC. J Circ Syst Comput 18:733–769
84. Nandi R, Sanyal SK, Bandyopadhyay TK (2008) Third order lowpass Butterworth filter
function realization using CFA. Int J Electron 95:313–318
85. Fabre A (1995) Comment and reply: Insensitive voltage-mode and current-mode filters from
transimpedance op amps. IEE Proc Circ Dev Syst 142:140–143
Chapter 5
Synthesis of Sinusoidal Oscillators Using CFOAs
5.1 Introduction
Sinusoidal oscillators find numerous applications in various electronic, instrumenta-

tion, measurement, control and communication systems as test oscillators or signal
generators. Since the classical Wien Bridge oscillator (WBO), RC phase shift oscilla-
tor, Twin-T oscillator etc. do not have the provision for varying the frequency of
oscillation through a single variable passive element (resistance or capacitance;
preferably the former), considerable research effort has been made in the eighties
and nineties on evolving single element controlled sinusoidal oscillators using IC
op-amps. Thus, the area of RC-active oscillators using the conventional voltage-mode
op-amps (VOA) had been a very prominent area of analog research before the advent
of the Current Conveyors, CFOAs and other modern active circuit building blocks. A
large number of VOA-based sinusoidal oscillators were published in the technical
literature during 1976–2001 for instance, see [1–9] and the references cited therein.
During the past two decades, there have been numerous investigations, intuitive
as well systematic, on the generation of a variety of sinusoidal oscillators
employing CFOAs. The aim of this chapter is to give an exposition of some of
the prominent CFOA-based sinusoidal oscillators.
5.2 The Evolution of Single Element Controlled Oscillators:

A Historical Perspective
Since the traditional WBO requires either ganged variable capacitors or ganged
variable resistors for realizing variable frequency sinusoidal oscillations, the prob-
lem of realizing a single element controlled oscillator had been a very popular
132 5 Synthesis of Sinusoidal Oscillators Using CFOAs
problem among researchers in the 1980s. In 1976, Hribsek and Newcomb [1]
presented, for the first time, two single-resistance controlled oscillators each
using two op-amps and two grounded capacitors as preferred for IC implementa-
tion; see [11, 63] and references cited therein.
The first real single-element-controlled oscillator using a single op-amp without
any constraints was introduced by Soliman and Awad [2] in 1978 but in this circuit,
the oscillation frequency could be controlled only through a variable capacitor—not
a very convenient option as variable capacitors have a very limited tuning range. A
single-op-amp based single-resistor-controlled-oscillator (SRCO), capable of being
operated as (1) a variable frequency oscillator, (2) a voltage-controlled-oscillator
(VCO) and (3) a very low frequency (VLF) oscillator, without any constraints, was
proposed by Senani in 1979 in [3].1 This circuit employs only a single op-amp, six
resistors and two capacitors and provides independent single resistance controls of
both condition of oscillation (CO) and frequency of oscillation (FO) through two
separate grounded resistors. The circuit could also be employed to realize a VCO by
replacing the frequency-controlling grounded resistor by a FET used as a voltage-
controlled-resistor and is convenient for incorporating the additional amplitude
stabilizing/control circuitry easily due to the grounded nature of the resistor
governing the CO of the oscillator [4]. Bhattacharyya and Darkani in [5] derived
the complete family of sixteen such single-op-amp-RC canonical SRCOs. Methods
of generating equivalents of such op-amp based oscillators have also been evolved;
for instance, see [12] and [6] and references cited therein.
Interestingly, the problem of devising newer SRCOs employing one or more active
building blocks (ABB) has continued to attract the attention and imagination of
researchers even now and a large number of SRCOs have been evolved using a variety
of other ABBs too during the last two decades. The other ABBs considered have been
second generation Current Conveyors (CCII) (and their many variants), operational
transconductance amplifiers (OTA), Four terminal floating nullors (FTFN), Current
differencing buffered amplifiers (CDBA), Current differencing transconductance
amplifiers (CDTA), Operational Trans-resistance amplifiers (OTRA) etc. In fact, the
search for newer topologies of SRCOs is aimed at ultimately achieving more and more
or all of the following desirable features: employment of grounded capacitors as
preferred for IC implementation, use of a minimum possible number of active and/
or passive components, suitability for VCO realization, achieving quadrature signal
generation, providing explicit voltage mode as well as current mode outputs, achiev-
ing a high frequency-stability, exhibiting higher operational frequency range and
minimization of the effects of parasitic impedances or non-ideal parameters etc.
In this chapter, we present a variety of SRCOs employing CFOAs with a focus
on the works exhibiting a synthesis approach and limiting to only some representa-
tive circuits in various categories from the vast amount of literature accumulated
during the past two decades in this area (for instance, see [10, 13–67] and the
references cited therein).
1
It has come to the attention of the first author only at the time of finalizing this chapter (13–17
September 2012) that a quite similar single op-amp SRCO employing only five resistors and two
capacitors, was proposed by Soliman and Awad in 1978 in [87].
5.3 Advantages of Realizing Wien Bridge Oscillator Using CFOA vis-à-vis VOA 133
5.3 Advantages of Realizing Wien Bridge Oscillator

Using CFOA vis-à-vis VOA
The interest in using CFOAs for realizing sinusoidal oscillators grew after it was
demonstrated by Martinez et al. [13, 14] that the use of CFOA rather than VOA in
the classical Wien Bridge oscillator offers improved performance, as compared to
its VOA-based counterpart, in terms of frequency accuracy, dynamic range, distor-
tion level and frequency span. In the following, we show, as demonstrated in [14]
that in the CFOA-version, the condition of oscillation (CO) and frequency of
oscillation (FO) become decoupled.
Consider now the Wien bridge oscillator (WBO) using a conventional VOA (see
Fig. 5.1a), an ideal analysis gives the closed loop characteristic equation (CE) as
2
ð3 kÞ 1
s þs
2
þ ¼0 (5.1)
RC RC
from where the (CO) and (FO) are given by
CO : k 3 (5.2)
1
FO : ω0 ¼ (5.3)
RC
When VOA is assumed to have a one-pole open loop gain function characterized
by Av ðsÞ ffi ωst for ω ωp where ωp is the pole frequency and ωt is the gain-
bandwidth product of the op-amp, through a re-analysis of the circuit [14], the
following non-ideal FO ð^
ω0 Þ and CO are obtained:
!
2 1 1
ω 0 Þ ¼ ω0 2
ð^ and k3 (5.4)
1 þ 3τkω0 1 τω0 ð1 ωω0 2 Þ
2
a b
R C
R C
+
y
−
w
x
z
R2 R2
Fig. 5.1 Wien bridge C R R1 R R1
C
oscillators: (a) realized
with a VOA, (b) realized
with a CFOA
where

1 R2
τ¼ and k¼ 1þ (5.5)
ωt R1
From the above equations, it is seen that because the closed loop amplifier gain k
appears in the expressions of FO and CO both, therefore, any change in the signal
amplitude calibration (distortion) by changing k, disturbs the oscillation frequency
also and vice versa.
In the case of the CFOA-based WBO circuit of Fig. 5.1b on the other hand, the
non-inverting amplifier gain is given by
k
kðsÞ ¼ (5.6)
1 þ sτ
where τ ¼ RpCp with Rp//(1/sCp) being the parasitic output impedance looking
into terminal-Z of the CFOA. The non-ideal FO and CO are given by
ω0 2
ωÞ 2 ¼
ð^ (5.7)
1 þ 3τω0
and

^2
ω
κ 3 þ τω0 1 2 (5.8)
ω0
From the above equations, difference can be seen in the behavior of the CFOA-
version of the Wien bridge oscillator as compared to its VOA counterpart. It may be
noted that CO and FO in the CFOA-version are de-coupled in the sense that k does
not appear in (5.7) hence, any change in adjusting the CO by changing k, does not
have any effect on FO.
5.4 Single-Resistance-Controlled Oscillators (SRCO)

Using a Single CFOA
As normally happens in any area of research, the initial results are quite often
derived intuitively which lead to systematic formulation of methodologies subse-
quently to enable the generation of all possible circuits belonging to a specific class.
In the area of SRCO realization using CFOAs also, a number of circuits were
5.4 Single-Resistance-Controlled Oscillators (SRCO) Using a Single CFOA 135
a b
y11 y1 y8
y
y8
z w
y x
y9
y6 zw
x
y9 y2
y12
y12 y10
y7 y4 y0
y10 y3
y4 y0
c
y11
y8
y
w
z
x
y9
y10 y4 y0
Fig. 5.2 Generalized single CFOA configurations for systematic generation of canonic SRCOs
[86]. (a) Generalized six-node structure. (b) First converted five-node structure obtained from the
circuit of (a). (c) Second converted five node structure obtained from the circuit of (a)
initially reported in a piece-meal manner by various researchers. It was only after

the research carried over about a decade that systematic approaches started being
formulated by a number of research groups for deriving systematically all possible
CFOA-based SRCOs belonging to specific classes.
One such systematic approach was formulated in [15, 19] which was based upon
the constitution of the most general single-CFOA-based six-node twelve-
admittance structure shown in Fig. 5.2a and two converted five-node structures
obtained therefrom, shown in Fig. 5.2b, c, whose general characteristic equations
were found wherefrom all canonic single-CFOA-based SRCOs i.e. circuits using no
more than three resistors and two capacitors were enumerated. A class of Single-
CFOA-based SRCOs resulting from this approach has been shown in Fig. 5.3.
In the set of eight SRCOs displayed in Fig. 5.3, it may be noted that oscillators 6
and 8 do not permit independent adjustability of CO although FO can be varied
independently; oscillators 1, 2, 3, 5, 7 provide the control of CO through a single
grounded resistor which is attractive from the point of view of ease of incorporating
Circuit number and Oscillator Circuit CO and FO

References where
appeared
1 R3 C 0
= ;
[20] (Fig. 3); y
C1 R4 C 1
w
xz adjustable by R4
[19] (Fig. 5);
[27] (Fig .6(c), (f)) wo = 2
R3 R4 R7 C0 ;
[28] (Fig. A1-B4) R3R7C0C1
controllable by R7
2 R3 C
=1+ 0 ;
[20] (Fig 6); R2 R4 C6
y
C6 w
[19] (Fig 6); xz adjustable by R4
[27] (Fig 6(g)) 1
[28] (Fig A2-B3) R3 R4 C0 wo = ;
R2R3C0C6
controllable by R2
3 C0 R 3
y
C1 = ;
[19] (Fig. 6); w C1 R 4
xz
[13] (Fig. 3);
adjustable by R4
[23] (Fig. 3); R2
[26] (Fig. 1); R3 C0 1
R4 wo =
[27] (Fig .6(a), (e)) R3R2C0C1
[28] (Fig .A1-B3) Controllable by R2
4 R12 C3
C9 = ;
x R1 C9
[40] (circuit 1 of Table- yz
w vo
R12
I); adjustable by R12
[27] (Fig .6(k)) R1 1
C3 wo = ;
R10 2R1R10C3C9
controllable by R10
R0 C
5 =1+ 8 ;
C8 R1 R9 C10
[16] (Fig. 8); y
adjustable by R0
w vo
[27] (Fig .6(n)) xz
R9 1
C10 R0 wo = ;
R1R9C8C10
controllable by R1
6 R8
R0 C
R9
y
w
=1+ 12 ;
[26] (Fig. 6(l)) xz R9 C10
C12 wo = 1 1+1
C10 R0C10C12 R9 R8
R0
controllable by R8
7 R3 C
C11 = 1+ 11 ;
[15] (Fig.7) x R10 C9
C9 w
yz
adjustable by R3
R6 1
wo =
R10 R6R10C9C11
R3
Controlled by R6
8 C9 R4 C9
R8
y = 1+
zw R8 C10
[15] (Fig .8) x
1 ⎛ 1 1 ⎞
R11 wo = ⎜ + ⎟
C10 R4 R4C9C11 ⎜⎝ R8 R11⎟⎠
controllable by R11
Fig. 5.3 The class of Single-CFOA-based canonic SRCOs [86]

R6
a b
R2 Y
Z W Vo
X
Y R3
Z
W Vo
X
C3 C2
R5 C1
C1 C4 C2 R1
R4
R1
Fig. 5.4 Single-CFOA-three-GC SRCOs proposed by Toker et al. (adapted from [27] # 2002
Elsevier)
amplitude stabilization/control circuitry. On the other hand, oscillator 4 provides

FO control through a grounded resistor thereby making it a suitable choice for easy
conversion into a VCO by replacing this frequency-controlling grounded resistor by
a FET or linearized VCR [68].
It may be mentioned that in Fig. 5.3 we have included only canonic SRCOs i.e.
those circuits which use no more than three resistors and two capacitors. For other
interesting SRCOs using more than three resistors and more than two capacitors the
reader is referred to the work reported in [27, 28] from where it is found that
permitting more than two capacitors makes it possible to realize single-CFOA
SRCOs using all grounded capacitors—a feature which is not possible with
canonic-single-CFOA oscillators. Two such three-GC SRCOs devised by Toker
et al. [27] are shown in Fig. 5.4.
The CO and FO for the circuits of Fig. 5.4 are as follows:
For the circuit of Fig. 5.4a

1 1 1 2
CO : C2 þ C1 þ C3 0 (5.9)
R5 R6 R6 R3
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
R3 R5 R1 R6
2 1
1
FO : f0 ¼ (5.10)
2π C2 C3
and for the circuit of Fig. 5.4b

1 1
CO : ðC4 C1 Þ þ C3 ð Þ 0 (5.11)
R2 R4
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
R 2 ðR 4 R 1 Þ
1 1 1
1
FO : f0 ¼ (5.12)
2π C3 C4
P
y
2 w
z
x R1
x
wz1
R3 C2 R2 y
C1
Fig. 5.5 A two-CFOA-GC SRCO proposed by Senani and Singh (adapted from [70])
Thus, in both cases FO can be controlled by a single variable resistor R1 although

an independent control of CO is not available.
It may be seen that out of the two circuits, that of Fig. 5.4a has the advantage that
both the Z-pin parasitic impedances Rp//(1/sCp) can be absorbed in the external
components R4 and C4. To the best knowledge of the authors till date not even a
single SCRO is known which can have only two grounded capacitors while
using only a single CFOA. In the next three sections we show, however, that
given two CFOAs, two-GC SRCOs are possible which can also be synthesized
quite systematically.
5.4.1 A Novel SRCO Employing Grounded Capacitors
A novel SRCO employing two CFOAs and both grounded capacitors (GC) as
preferred for integrated circuit implementation was introduced by Senani and
Singh in 1996 [25]. This circuit is shown in Fig. 5.5 and is characterized by the
following CO and FO:
R 3 ¼ R2 (5.13)
and
1 1
f0 ¼ (5.14)
2π R1 R2 C1 C2
Thus, the CO can be satisfied by adjusting R3 whereas the FO can be indepen-

dently varied by R1.
It must be pointed out that although frequency controlling resistor R1 is not

physically connected to ground; nevertheless, since it is connected to virtual
ground, there is no difficulty in replacing R1 by FET-based voltage-controlled
resistor (VCR) conventional or otherwise such as Senani’s floating VCR (FVCR)
in [69] to obtain voltage controlled oscillations.
In designing this oscillator for providing a specified frequency range, it must be
kept in mind that the various external RC components are to be selected to have
appropriate values so that the parasitic impedances appearing at the x-input and
compensation pins z of the CFOAs have the least effect on the performance.
Alternatively, these internal compensating capacitances can be absorbed into the
main capacitances as they appear in shunt with them, while rx of both CFOAs can
be easily accommodated in the resistors R1 and R3.
The frequency stability is considered to be an important figure of merit for
evaluating/comparing the performance of sinusoidal oscillators. It is defined as
SF ¼ dϕ/du evaluated at u ¼ 1 where u ¼ ω/ωo. Thus, SF can be determined by
finding the open loop transfer function of the circuit of Fig. 5.5 which is given by
s R31C2
TðsÞ ¼ (5.15)
s2 þ s R21C2 þ R1 R21C1 C2
Taking the values of various passive components as: R2 ¼ R3 ¼ R, C1 ¼ C2 ¼ C,

R1 ¼ R/n we get ω0 ¼ √(n)/RC ¼ √(n)/τ, while the open loop transfer function
becomes
ω0ffiffi
s=τ sp n
TðsÞ ¼ ¼ (5.16)
pffiffi0 þ ω2
s2 þ s=τ þ ω20 s2 þ sωn 0
by putting s ¼ jω, we get
j ωω0 p1ffiffin
TðjωÞ ¼ 2
(5.17)
1 ωω0 þ j ωω0 p1ffiffin
pffiffiffi
From the above, the SF has been found to be SF ¼ 2 n which can be made large
by keeping n large. It has been shown in [25] that by breaking the link at P the
resulting open loop circuit can be used as a lowpass/bandpass filter. Also, by
removing external capacitors C1 and C2 and incorporating the Z-pin parasitic
capacitances into design, the circuit can be used as an active-R oscillator with
ω0 still controllable through R1. In active-RC mode, this SRCO works well in
generating oscillation frequencies of the order of 500 kHz while in active-R mode,
it has been possible to extend the generated frequencies till 9.85 MHz.
5.5 Two-CFOA-Two-GC SRCOs: The Systematic

State Variable Synthesis
Subsequent to the publication of the two CFOA-SRCO of [25], it occurred to the

first author of this monograph that if one formulates state equations of this circuit
then with some algebraic manipulations it should be possible to convert these into
node equations which could then be synthesized using CFOAs and RC components.
It was soon realized that a chosen [A] matrix of the state variable characterization of
an SRCO could, therefore, lead to more than one circuit. Furthermore, since for any
specified CO and FO, many different state space representations leading to different
[A] matrices (but all leading to the same characteristic equation (CE)) could be
evolved, this methodology appeared to have the potential of generating a large
number of SRCO (quite likely all possible) circuits. In this section, we give an
account of the state variable methodology from [36–38, 70] and outline some
selected circuits from the complete family of 14 two-CFOA-two-GC SRCOs
generated therefrom.
A canonic second-order (i.e. employing only two capacitors) oscillator can, in
general, be characterized by the following autonomous state equation:

x_ 1 a11 a12 x1
¼ (5.18)
x_ 2 a21 a22 x2
From the above, the characteristic equation (CE)
s2 ða11 þ a22 Þs þ ða11 a22 a12 a21 Þ ¼ 0 (5.19)
gives the condition of oscillation and frequency of oscillation as
CO : ða11 þ a22 Þ ¼ 0 (5.20a)

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
FO : ω0 ¼ ða11 a22 a12 a21 Þ (5.20b)
The methodology of [36–38] involves: (1) a selection of the parameters aij,

i ¼ 1, 2; j ¼ 1, 2, in accordance with the required features (e.g. non-interacting
controls for frequency of oscillation and condition of oscillation through separate
resistors), (2) conversion of the resulting state equation into node equations (NE)
and finally, (3) constitution of a physical circuit from these node equations.
Different circuits are expected to be generated by making different choices of
parameters a11, a12, a21 and a22. For non-interactive controls of condition of
oscillation and frequency of oscillation, let us assume that condition of oscillation
is to be controlled by R1 (independent of R2) and frequency of oscillation is to be
controlled by R2 (independent of R1; with the third resistor R3 featuring in both
condition of oscillation and frequency of oscillation). These conditions can be
5.5 Two-CFOA-Two-GC SRCOs: The Systematic State Variable Synthesis 141
satisfied in a number of ways leading to different [A] matrices. It has been shown in
[37, 38] that a set of 14 different matrices can be conceived.
To illustrate the procedure, consider now the following [A] matrix which satisfies
the above requirements:
" 1 #
0
½A ¼ C1 R 2 (5.21)
C21R3 C2 R3 R1
1 1 1
From the above matrix, the CO and FO are given by
R3 ¼ R1 (5.22)
and
1
fo ¼ pffiffiffiffiffiffiffiffiffiffiffi (5.23)
2π C1 C2 R2 R3
From the above matrix, the following node equations can be written
dx1 x2
C1 ¼ (5.24)
dt R2
dx2 ðx2 x1 Þ x2
C2 ¼ (5.25)
dt R3 R1
The synthesis of the final circuit using node equations (5.24) and (5.25) is shown
in Fig. 5.6a which is self-explanatory.
Following the above explained procedure, a large number of circuits are derived
in [70] out of which a set of 14 SRCOs are demonstrated in [37, 38]. Some
exemplary circuits possessing interesting properties are shown here in Fig. 5.6
(FO is same for all oscillators as given by (5.23)).
The circuits shown in Fig. 5.6 have a number of interesting properties which are
as follows.
Single resistance control (SRC) of frequency of oscillation through a grounded
resistor makes it easier to incorporate FET-based voltage controlled resistors
(VCR) thereby leading to VCO realizations whereas SRC control of condition of
oscillation through a grounded resistor is desirable from the viewpoint of easy
incorporation of amplitude stabilization/ control circuitry. The circuit of Fig. 5.6a is
seen to provide controls of condition of oscillation and frequency of oscillation both
through separate grounded-resistors R1 and R2, respectively and is, therefore,
superior to the remaining SRCOs of Fig. 5.6 SRCOs from this view point.
In case of the circuits of Fig. 5.6a–c, f the z-pin parasitic capacitances can be
easily merged with the main external capacitances and hence, these parasitics do
not affect the circuit behavior adversely. In the circuits of Fig. 5.6d, e, g, the
a
y w
y R3 z
w x
x x2
z + −
R1
x2 + (x2 - x1) x2
x1 dx1 C2
R2 C1 R1
R2 − C1 R3 −
dt dx2
C2
dt
CO: R3 = R1
b c
CO: R1 = R3 CO: C1R1 = C2R3

d
e
CO: C1R1 = C2R3 CO: R3 = R1 (for C1 = C2)

f g
CO: C1R1 = 2C2R2 CO: C1 R1 = 2C2R2
Fig. 5.6 Some exemplary circuits synthesized through the state variable methodology [70].
(a) CO: R3 ¼ R1. (b) CO: R1 ¼ R3. (c) CO: C1 R1 ¼ C2 R3. (d) CO: C1 R1 ¼ C2 R3. (e) CO:
R3 ¼ R1 (for C1 ¼ C2). (f) CO: C1 R1 ¼ 2C2 R2. (g) CO: C1 R1 ¼ 2C2 R2
5.6 Other Two-CFOA Sinusoidal Oscillator Topologies 143
capacitor C1 is connected to z-terminal of CFOA1 and no capacitor is connected at

the z-terminal of CFOA2. However, the parasitic capacitance at the z-terminal of
the CFOA2 is made ineffective by z-terminal being connected to virtual ground
(as in Fig. 5.6d) or is ineffective as the current through z-terminal of CFOA2 is not
coming into picture as in case of Fig. 5.6e. However, in the circuit shown in
Fig. 5.6g, the z-pin parasitic capacitance at the z-terminal of the CFOA2 cannot
be accounted for.
It is found that with C1 ¼ C2 ¼ C and R1 ¼ R3 ¼ R, R2/R ¼ n for the circuits
pffiffiffi
of Fig. 5.6a–d, f–g the frequency stability factor SF found to be SF ¼ 2 n can be
made sufficiently large as ‘n’ can always be kept greater than unity and therefore,
these circuits enjoy excellent frequency stability properties.
The circuit of Fig. 5.6e is notable due to the availability of an explicit current
output.
The family of 14 two-CFOA-GC SRCOs presented in [37, 38, 70] has been found
to work quite well for generating sinusoidal signals up to several hundred kHz.
5.6 Other Two-CFOA Sinusoidal Oscillator Topologies
The SRCOs were shown to have been derived in the previous section through a
systematic synthesis procedure with the objective of possessing the following
features: (a) use of two GCs, (b) use of two CFOAs and (c) independent control
of CO and FO through two separate resistors. All the SRCOs were based upon the
tuning laws of the type
R1 ¼ R3 (5.26)
1
f0 ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (5.27)
2π C1 C2 R2 R3
which give the resulting circuits, the control of CO trough R1 and that of FO by R2.
Although, major attention has been received in the literature on the above kind of
SRCOs, oscillators governed by other type of tuning laws, which thereby provide
CO control through a single variable capacitor or FO control through a single
variable capacitor or provide an expression for FO containing a difference term
are also useful due to the following:
1. Oscillators providing single element control (SEC) of FO through a single
variable capacitor can be used as a transducer oscillator in conjunction with
capacitive transducers.
2. Oscillators providing CO control through a capacitor can be used in some
capacitance measurement schemes, for instance see [71–73].
3. Oscillators having a difference term in the expression of FO may be usefully
employed as very low frequency oscillators [41].
In view of the above, therefore, the catalogue of 14 two CFOA oscillators and
their variants [37, 38, 70] some of which were described in the previous section
does not really exhaust all possible GC-SECOs realizable from two CFOAs.
It, therefore, turns out that given only two CFOAs only two GCs, along with two
to three resistors, a number of other sinusoidal oscillator circuits should be possible
which may have tuning law different than those considered so far and yet satisfy the
single-element-controllability conditions. In Fig. 5.7 , we have presented a number
of such two-CFOAs-GC SECOs. These circuits too are derived by the state variable
methodology by framing new tuning laws, determining the required [A] matrices,
converting the [A] matrices into node equations and finally synthesizing the
resulting node equations by physical circuits using CFOAs and RC elements. The
details of the derivation are given in [48]. The following features of the circuits of
the Fig. 5.7 may now be noted.
• Circuits 1–2 have tuning laws that do not conform to (5.26) and (5.27) and yet
these circuits do possess features (a) and (b).
• Circuit 3 is the only oscillator circuit realizable with a bare minimum of only
four passive components. It may be pointed out that this circuit can be treated to
be the CFOA-version of a similar circuit using CCIIs described earlier in
references [74] and [75] but by contrast, this CFOA-version has the advantage
of providing buffered outlets from the output of either CFOA.
• Circuit 4 although employs three grounded capacitors but still qualifies for
feature (c).
It is worth mentioning that like most CFOA-based circuits, the influence of the
parasitic impedances of the CFOAs can be reduced by selecting the external
resistors to be much larger than the input resistance rx of the X-terminal and smaller
than the parasitic output resistance Rp looking into the compensation terminal-Z of
the CFOA and the external capacitances to be larger than the parasitic output
capacitance Cp of the CFOAs.
Analysis of the frequency stability of the circuits reveals that the frequency
stability factors are quite large for the circuits shown in Fig. 5.7 similar to other
circuits contained in [16, 25, 44].
It is worth noticing that oscillator 1 contains a difference term in the expression
pffiffiffiffiffiffi
for FO of type ω0 ¼ RC 1n
where n is the frequency controlling resistors ratio.
This oscillator qualifies to be used for generating very low frequency oscillations
(i.e. 1 Hz or lower) by choosing n such that (1 n) can be made as small as
possible so that lower values of FO are achievable. On the other hand, oscillators 3
and 4 appear to be suitable for capacitance measurement methods such as those of
[71–73]. In such a case, the unknown capacitance can be connected in place of C1
and then the known variable capacitance C2 is to be varied until the circuit just starts
(or stops) oscillating as described in [71–73].
One more two-CFOAs-two-GC-based SRCO but with an additional frequency
scaling factor in the expression for f0 was introduced by Liu and Tsay in [22] which
is shown in Fig. 5.8.
R3 C2
R3 =
X Vo1 R2 C1
ZW X
Y
ZW R1
Y (1 − )
R2
w0 =
C1
C1C2R1R3
R1 R2 C2
R1
; <1
R2
Circuit 1
C1
X C1 R3
= 1 +
Y
ZW
R1 ZW Vo1
Y C2 R2
X
R3 1
w0 =
C1C2R1(R2+R3)
C2
R2
Circuit 2
C1 y C1 = C2
z V01
w y 1
x w w0 =
z C1C2R1R3
x
R3 C2
R1
Circuit 3
C1 = C3
Y 1
Y Vo1 w0 =
ZW ZW
X R1 C1C2R1R2
X
C1 C2 C3
R2
Circuit 4
Fig. 5.7 Some two-CFOA oscillators with different tuning laws (adapted from [48] # 2006
IEEE)
Fig. 5.8 Another SRCO

grounded-resistor controlled R3
using GCs (adapted from [22]
X Z R2
1 W X
Y 2 ZW
R4 Y
R1
C1 C2
Fig. 5.9 Yet another

grounded resistor controlled X
1 W
sinusoidal oscillator (adapted
from [10] # 1997 IEE) RA Y Z
Y
W 2 C1
R1
Z X
R2
C2
Although not explicitly mentioned in [22], this circuit can be considered to be

derivable from a two-op-amp-GC SRCO published earlier in [11] by realizing the
negative-impedance-converter (NIC) therein by a CFOA without requiring any
resistors and thereby simplifying the circuit as shown in Fig. 5.8. The CO and FO
of this circuit are given by
R1
ðC1 þ C2 Þ ¼ C1 (5.28)
R2
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
1 R3
ω0 ¼ (5.29)
2C1 C2 R1 R2 R4
Thus, the CO can be controlled by R1 while FO can be varied through R4 and/or

R3. It may be noted that the presence of an extra frequency scaling factor in FO (like
that of [11]) is appropriate for the generation of very low frequency oscillations
without having to use large RC components.
Another grounded resistor controlled sinusoidal oscillator using CFOAs was
proposed by Martinez et al. in [10] which is shown in Fig. 5.9.
Fig. 5.10 SRCO proposed

R1 Vo1
by Tangsrirat and
X
Surakampontorn (adapted 1 W Y
from [51] # 2009 Elsevier) Y Z 2 Vo2
W
Z
X
C1 R2 R3 C2
The CO and FO for this circuit are given by

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
RA ¼ R1 and ω0 ¼ (5.30)
C1 C2 R2 RA
In this oscillator also, both CO and FO can be controlled independently through

grounded resistors R1 and R2 respectively.
Martinez and Sanz in [29] gave a method for generation of variable frequency
sinusoidal oscillators based on two integrator loops and presented a sinusoidal
oscillator using two CFOAs. However, this circuit is quite similar to the SRCO
circuit of Senani and Singh [25].
Tangsrirat and Surakampontorn in [51] proposed a single resistance controlled
quadrature oscillator using two CFOAs which is shown in Fig. 5.10.
R2 C1 ¼ R3 C2 (5.31)
and
1
ω0 ¼ (5.32)
R1 R3 C 1 C 2
In this case, the CO can be adjusted by R2 without affecting the oscillation

frequency, while ω0 can be adjusted by R1.
Another quadrature oscillator using two CFOAs has been proposed by Hou and
Wang in [31], which has been obtained by a circuit transformation proposed by
them through which OTA-C circuits can be transformed into CFOA-RC circuits.
A systematic method of realization of low frequency oscillators was proposed by
Elwakil [41], which requires a passive resistor and one active (negative) resistor
which modifies the expression for oscillation-frequency such that it contains a
difference term. By keeping this difference term as small as possible, low frequency
oscillations can be achieved.
Three types of sinusoidal oscillators each using three CCII+, two to four resistors
and two grounded capacitors have been presented by Martinez et al. in [62], whose
practical workability has been verified by AD844 type CFOAs.
Apart from the above, Soliman in [42] has presented a number of sinusoidal
oscillators out of which only one is two-resistor-two capacitor-Two CFOA-based
circuit which is a quadrature oscillator. The other two sinusoidal oscillators use
two CFOA-4R-2C configurations having identical expressions for CO and FO.
However, in these circuits, only CO can be controlled by a single resistance.
5.7 Design of Active-R SRCOs
By taking into account the parasitic capacitance of the Z-terminal of the CFOA into
design, a variety of active-R oscillators have been reported in technical literature.
Some prominent circuits in this category are highlighted in this section.
5.7.1 Active-R Sinusoidal Oscillators Using CFOA-Pole
Such circuits were first presented by Liu et al. in [34], an exemplary circuit out of
If R1, R3 Rx and R2, R4 Rp, the condition of oscillation of this oscillator is
given by
R6 1 1
¼ þ (5.33)
R3 ð R5 þ R6 Þ R4 R2
R7 R5
Y
Y R3 2 W Vo2
1 W X Z
X Z
Vo1
Fig. 5.11 Active-R
R8 R1 R2 R6 R4
sinusoidal oscillator (adapted
from [34] # 1994 Taylor &
Francis)
5.7 Design of Active-R SRCOs 149
Fig. 5.12 Active-R SRCO Vo2

proposed by Singh and Senani
IEEE)) W
Z
Y X
R1
Y
W Vo1
X Z
R2
whereas the frequency of oscillation is given by

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 1 R8 R6
ω0 ¼ þ (5.34)
Cp R2 R4 R1 R3 ðR7 þ R8 Þ R2 R3 ðR5 þ R6 Þ
It is, thus, seen that the oscillation frequency can be independently controlled by
the resistor R1 which does not appear in the condition of oscillation.
5.7.2 Low-Component-Count CFOA-Pole Based Active-R SRCOs
Two such circuits, each employing two CFOAs and only two resistors, were
proposed by Singh and Senani in [43], one of which is shown in Fig. 5.12.
Rp
R0 2 ¼
(5.35)
2Rp 2C
Ry þ 1 þ 1 þ Cpy
and
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u2Rp Rp Rp
u
1 u Ry þ R0 2 R0 1 1 þ 1
f0 ¼ t (5.36)
2πCp Rp 1 þ Cpy
2C
where R0 1 ¼ ðR1 þ Rx1 Þ, R0 2 ¼ ðR2 þ Rx2 Þ, Rp, Cp are the Z-port parasitics, Ry, Cy
are the y-port parasitics and Rx is the x-port parasitic input resistance.
Fig. 5.13 CFOA-pole based

sinusoidal oscillator X
introduced by Liu et al. W Vo
Z
(adapted from [34] # 1994 Y
Taylor & Francis)
R1 R2 R3 C
5.7.3 Other Two-CFOA Based Active-R SRCOs
A number of Two-CFOA active-R SRCO circuits can be easily obtained from those
oscillators presented in Sects. 5.5 and 5.6 where each CFOA has a capacitor
connected from its Z-pin to ground. Thus, from such circuits, active-R VCOs can
be obtained by deleting the external capacitors connected at the Z-terminals of the
CFOAs and employing in their places the Z-pin parasitic capacitances in the design.
5.7.4 CFOA-Pole-Based RC Oscillator
A low component oscillator using one external capacitor and the pole of the CFOA
was presented by Liu et al. in [34] which is shown in Fig. 5.13.
If R1 Rx and R3 Rp, the CO and FO for this circuit are given by
C C Cp
¼ þ (5.37)
R1 R3 R2
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
ω0 ¼ (5.38)
Cp CR2 R3
Although this oscillator has the advantage of using only four passive elements,
but oscillation frequency cannot be independently controlled.
On the other hand, four sinusoidal oscillators each consisting of two capacitors, a
single CFOA with its pole accounted in the design, were proposed by Abuelma’atti
et al. in [21]. A single resistance controlled oscillator out of this set is shown in
Fig. 5.14 for which the CO and FO are given by

1 C4 R3
ðC4 R3 C2 R1 Þ ¼ R1 1 þ (5.39)
Cp Cp Rp
5.7 Design of Active-R SRCOs 151
Fig. 5.14 A SRCO using C4

CFOA-pole proposed by
Abuelma’atti-Farooqi-
R3
Alshahrani (adapted
from [21] # 1996 IEEE) Y W
Z
X
R1 R2
C2
and
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

1 Rp
ω0 ¼ 1þ (5.40)
Cp C4 Rp R3 R2
It is thus seen that in this circuit CO can be controlled by R1 whereas FO is

independently controlled by R2. It has been demonstrated in [21] that using R1 ¼
R3 ¼ 100 Ω, C2 ¼ 4.7 pF, C4 ¼ 22 pF and using AD844 type CFOA biased with
15 V DC power supplies, variable frequency oscillations up to 27.5 MHz, with
peak to peak voltage of 2–9 V, were successfully obtained using this circuit.
Yet another variable frequency oscillator consisting of single CFOA-pole and
single capacitor was proposed by Martinez et al. in [13], which enjoys independent
tunability of oscillation frequency and condition of oscillation. In the same year,
Abuelma’atti and Al-Shahrani in [23] and Abuelma’atti and Farooqi in [26] also
proposed a number of SRCOs using the CFOA-pole.
It is worth pointing out that in [49], Nandi has reported an interesting active-R
oscillator using a building block termed as CFA- (realized from two CFOAs of the
normal kind) along with only three external resistors which is capable of generating
sinusoidal oscillations over a tuning range of 2.8–40 MHz using AD844 type
CFOAs.
5.7.5 A Simple Multiphase Active-R Oscillator

Using CFOA Poles
A multiphase oscillator circuit was proposed by Wu et al. in [32] which is shown in

Fig. 5.15. This circuit has the following merits: (1) uses only parasitic poles of
CFOAs thereby making it suitable for high frequency oscillations and monolithic IC
fabrication due to complete elimination of external capacitors, (2) exhibits large
output voltage swing and (3) has moderately low Total Harmonic Distortion (THD).
Assuming all the CFOAs to be identical, the loop gain for an n-phase oscillator
can be expressed as
Fig. 5.15 Multiphase

oscillator (for n ¼ 3) x
proposed by Wu et al. w V01
(adapted from [32] # 1995 z
y R1
IEE)
R2
x
w V02 R1
y z
R1
R2 x
w V03
z
y
R2
!n
G0
LðsÞ ¼ where ωb ¼ 1=Rb Cp ;
1 þ ωsb (5.41)
Rb ¼ R2 ==Rp and G0 ¼ Rb =Ra ; Ra ¼ ðR1 þ Rx Þ
The frequency and condition of oscillation are given by
π π
ω0 ¼ ωb tan and Rb Ra sec (5.42)
n n
As a special case, the condition of oscillation, for frequency of oscillation of a

three phase sinusoidal oscillator (n ¼ 3), can be expressed as
pffiffiffi R2 R p
ω0 ¼ ωb 3 and Rb 2Ra or R1 Rx (5.43)
2 R2 þ R p
Thus, the circuit of Fig. 5.15 produces the maximum and the minimum oscilla-
tion frequencies when R1 ¼ 0 and R1 ¼ (Rp/2) Rx respectively.
5.8 SRCOs Providing Explicit Current Output
In view of the proliferation of current-mode filters and other signal processing

circuits, the design of oscillators providing an explicit-current-output (ECO) from a
high output impedance node has also become important. Sinusoidal oscillators with
ECO would be useful as signal generators to test various current-mode circuits.
Although, there have been a number of investigations [12, 46, 76–79] on realizing
5.8 SRCOs Providing Explicit Current Output 153
oscillators with ECO using other building blocks, such as first generation current
conveyor [46, 76], differential difference current conveyors [77], differential
difference complementary current feedback amplifier [78], four terminal floating
nullor [12], unity-gain voltage and current followers [79], however, none of these
building blocks are available commercially yet.
On the other hand, because a CFOA of AD844 type does have a current output
terminal and is commercially available, ECO oscillators made from CFOAs are of
practical importance. In this section, we show how the state-variable approach of
synthesis [36] of oscillators can be extended to synthesize systematically current-
mode sinusoidal oscillators with explicit current output using CFOAs as active
building blocks. Of course, current-mode oscillators based on CCII+ can also be
implemented by AD844 however; oscillators using exclusively CCII+ which have
the capability providing explicit current output are known to employ three CCII+
whereas none of the circuits in earlier works [12, 76–79] have been realized with
CFOAs.
The state variable methodology described earlier may be easily tailored to suit
the evolution of the SRCOs with explicit current output. As already described in
Sect. 5.5, the various conditions for non-interacting controls of CO and FO have
some requirements which are repeated here for convenience:
(a) The expression of (a11 + a22) should either not have terms containing R2 or they
should be cancelled out. Thus, in (a11 + a22), there should be two terms left with
opposite signs involving R1 and R3.
(b) Similarly, to have FO independent of R1, the expression (a11a22 a12a21)
should either not have the terms containing R1 or they should be cancelled
out. Thus, FO should be a function of resistors R2 and R3 only (along with C1
and C2).
Let us now construct the required [A] matrix by choosing a11 ¼ C11R1 ; a22 ¼

C21R3 which satisfy the requirement (a). Now, choosing a12 ¼ C11 R11 þ R12 ; a21
¼ C21R3 , we can satisfy the requirement (b). The required [A] matrix, therefore, takes
the following form:
" #
1
C11 1
þ R12
½A ¼ C1 R 1 R1 (5.44)
1
C2 R 3 C21R3
which results in the following CO and FO:
C2
CO : R1 ¼ R3 (5.45)
C1
1
ω0 ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (5.46)
C1 C2 R2 R3
By substitution of (5.44) into (5.18), the following node equations are obtained
dx1 x1 x2 x2
C1 ¼ (5.47)
dt R1 R2
dx2 x1 x2
C2 ¼ (5.48)
dt R3
For meeting the specific objective of having an explicit current-mode output, we

consider implementing (5.47) and (5.48) using CFOAs keeping in mind that the
z-terminal of at least one of the CFOAs has to be left unutilized to make the current
output available from a high output impedance node. The circuit, thus formulated
from (5.47), (5.48) takes the form of circuit (a) of Fig. 5.16 where the mechanism of
constructing the circuit can be understood by following the various current
segments of the (5.47) and (5.48) as marked in the circuit of Fig. 5.16a. Two
other circuits, similarly derived, are shown in Fig. 5.16b, c.
It may be seen that as intended, the explicit current output is available from the
z-terminal of the first CFOA in all the three circuits.
Like most of the CFOA-based oscillators, the parasitics of the CFOA make the
non-ideal expressions of the oscillation frequencies of the circuit of Fig. 5.16 to be
different than their ideal counterparts. Thus, parasitics would limit the operation of
the oscillators at higher frequencies. However, it has been shown in [64] that with
judicious choice of component values, oscillations around 1 MHz range are attain-
able with these circuits. As an example, Fig. 5.17 shows a typical waveform
(1.06 MHz, 2.3 V (p–p)) obtained from the oscillator of Fig. 5.16c using AD 844
CFOAs biased with 15 V DC supplies.
Some other interesting explicit current output oscillators (ECO) using CFOA are
discussed next.
1. Two single CFOA-based ECO oscillators were presented by Senani and Sharma
in [47]. One of the circuits from [47] is shown in Fig. 5.18.
R3 ¼ 6ðR1 þ R2 Þ; provided C1 ¼ C2 ¼ C3 ¼ C (5.49)
1
f0 ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffi (5.50)
2πC 3R1 R2
Although the circuit has the advantage of using a single CFOA, a drawback of
this circuit is that it has three capacitors. On the other hand, f0 can be varied
through a potentiometer by changing ‘n’ (which is the ratio R1/R2) while their
sum (R1 + R2), and hence, the CO remains invariant. However, CO can be
adjusted independently through the resistor R3.
2. Another two-CFOA-based SRCO family with explicit current output has been
proposed recently by Lahiri et al. [58], out of which an exemplary SRCO is
5.8 SRCOs Providing Explicit Current Output 155
Fig. 5.16 Some exemplary a (x1−x2)

SRCOs providing explicit R1
current output (adapted from
[64] # 2010 Jojn Wiley & (x1−x2) x2
Sons Ltd.) R1 R2 −
R3 R2
R3
x
C2 2 w
x y z
dx2 (x1−x2) x2
w 1 C2 −
z dt x1 R1 R2
y
iout C1 dx
C1 1
dt
b
R3
C2
y y
w 1 R1 2 w
z z
x x
iout
R2
C1
c
R2
x
1 w
z
y
C2 iout
y
2 w
z
R3 x
C1
R1
Fig. 5.17 A typical

waveform generated from
the oscillator of Fig. 5.16c
(1.06 MHz, 2.3 Vpp).
Component values:
C1 ¼ C2 ¼ 100 pF,
R1 ¼ 404 Ω,
R2 ¼ R3 ¼ 1 kΩ (adapted
from [64] # 2010 John Wiley
& Sons Ltd.)
Fig. 5.18 CM oscillator

using a single CFOA Y
proposed by Senani and C1 R2 W
Sharma (adapted from [47] Z
X
# 2005 IEICE) Iout
R3 C2
R1
C3
shown in Fig. 5.19. This circuit has the advantage of employing both grounded
capacitors as desirable for IC implementation.
The ongoing search for newer topologies of SRCOs with ECO using CFOAs
may lead to the circuits which might be useful as test oscillators for verifying
various current-mode signal processing circuits2 such as current-mode filters,
current-mode precision rectifiers etc. to which the proposed kind of circuits can
be interfaced without any additional hardware.
2
In spite of the criticism of [88], the current-mode techniques have given way to a number of
important analog signal processing/signal generation circuits over the past three decades.
5.9 Fully-Uncoupled SRCOs Using CFOAs 157
Fig. 5.19 Explicit-current

output second order
sinusoidal oscillator (adapted Y R2
W X
from [58] # 2011 Elsevier) Z
X W
Z Iout
Y
R1
R3 C2 C1
5.9 Fully-Uncoupled SRCOs Using CFOAs
CFOA-based canonic SRCOs which employ a minimum of five passive

components, namely, three resistors and two grounded capacitors (as desirable
from the view point of IC implementation) and possess tuning laws such that
both CO and FO can be controlled /adjusted by two independent resistors, require
at least two CFOAs. A major drawback of such topologies is that as soon as various
non-idealities /parasitics of the CFOAs are accounted for, the theoretically derived
independence of CO and FO vanishes due to the frequency-controlling resistor also
getting involved in the non-ideal expression for the CO. Those oscillators are called
fully-decoupled in which CO and FO are decided by two completely different sets
of components, such that none of the components involved in CO are also involved
in FO and vice versa. Such SRCOs are characterized by tuning laws of the type
CO : ðR1 R2 Þ 0 (5.51)
and
1 1
FO f0 ¼ (5.52)
2π C1 C2 R3 R4
which shows that such circuits would, need at least four resistors along with two
capacitors. Such ‘fully-uncoupled’ SRCOs, however, are not feasible with only
two active elements and call for the employment of at least three active elements as
in [42, 44].
There appear to be only two circuits known in earlier literature employing
CFOAs which belong to the category of fully uncoupled oscillators, namely, the
circuit presented by Soliman in [42] and the one proposed by Bhaskar in [44].
The Solimans’ circuit from [42] is shown in Fig. 5.20, whereas the circuit presented
by Bhaskar [44] is shown in Fig. 5.21.
Fig. 5.20 Fully uncoupled

oscillator (adapted from [42]
# 2000 Springer) Y
1 W Y
X Z 2 W Y
X Z 3 W
X Z
R1 C1 R2 R3
R4 C2
Fig. 5.21 Fully uncoupled

oscillator (adapted from [44] R4 R1
# 2003 Frequenz) x
w y
y z x z
zw
x w
R2 y
C2
C1
R3
It is interesting to note that both the circuits employ exactly the same number of
active and passive components. The ideal CO and FO for the circuit of Fig. 5.20 are
given by
R3 ¼ R4
and
1
2π C1 C2 R1 R2
whereas the CO and FO for the circuit of Fig. 5.21 are given by
R1 ¼ R2
1
2π C1 C2 R3 R4
It has been shown in [59] that the above described fully-uncoupled oscillators
from [42, 44] also fail to retain the independent controllability of FO under the
influence of non-ideal parasitic impedances of CFOAs as all the four resistors
employed in the oscillators appear in the non-ideal expressions of both CO and
FO, thereby completely disturbing the intended property.
In this section, we show two circuits from [59] which retain the independent
controllability of FO even under the influence of CFOA parasitic impedances.
These circuits are shown in Fig. 5.22.
5.9 Fully-Uncoupled SRCOs Using CFOAs 159
Assuming that the CFOAs are characterized by: iy ¼ 0; vx ¼ vy ; iz ¼ ix and vw

¼ vz , both the circuits are governed by a common characteristic equation (CE)
given by:

s 1 1 1
s þ
2
þ ¼0 (5.55)
C1 R0 R1 C1 C2 R2 R3
From this characteristic equation, the CO and FO are found to be
CO : ð R1 R0 Þ 0
1
FO : ω0 ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (5.56)
C1 C2 R2 R3
For an evaluation of the non-ideal performance of these circuits, we consider the

finite input resistance Rxi at the x-port, i ¼ 1–3, parasitic components Ryi in parallel
with 1/sCyi at the y-port and parasitic components Rzi in parallel with 1/sCzi at
the z-port of all the CFOAs i ¼ 1–3. Analysis reveals that in both the cases, the
non-ideal CE of both the circuits continues to remain second order. The non-ideal
CO and FO for both the circuits have been found to be as under
For the circuits of Fig. 5.22a, b
CO:

1 1 1 1 1
ðC2 þ Cz2 Þ þ þ þ
R0 R1 þ Rx1 Ry1 Rz1 Rz3
C1 þ Cz1 þ Cy1 þ Cz3
þ 0 (5.57)
Rz2
FO:
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi0 11=2
1 1 @ 1
f 0
0 ¼ A
2π C1 C2 R2 R3 C þC þC
1 þ y 1 z1 z3 1 þ Cz2
C1 C2
2 31=2
1

6 1 þ Rx2 1 þ Rx3 7
6 7
6 R2 R3
7 (5.58)
6 7
4 R2 R3 1 1 1 1 1 5
þ þ þ þ
Rz2 R0 Ry1 Rz1 Rz3 R1 þ Rx1
From (5.57)–(5.58), it is observed that in both the circuits, the frequency

controlling resistors R2 and R3 do not come into the non-ideal expressions for
CO; therefore, the independent controllability of FO remains intact even under the
influence of the non-ideal parameters/parasitic of the CFOAs employed.
It is worth mentioning that if the circuits are to be converted into voltage
controlled oscillators by replacing the frequency-controlling resistors R2 and/or
R3 by FET-based or CMOS voltage-controlled-resistors (VCR), this does not pose
a
y
1 w y
x z 2 w x
x z R3
3 w
R1 y z
C1 R C2
2
R0
b
y R2
1 w x
z 2 w y
x
y z 3 w
x z
R1
C1 C2
R0
R3
Fig. 5.22 Fully-uncoupled SRCOs proposed by Bhaskar et al. (adapted from [59] # 2012
Springer)
any difficulty since it is well known that grounded/floating VCRs using any of the
above mentioned devices could be realized with exactly the same amount of
hardware, for instance, see [69, 80–82].

Using the definition of frequency stability factor (SF) SF ¼ dϕðuÞ
du where u
u¼1
¼ ωωo is the normalized frequency and ϕðuÞ denotes the phase function of the open
loop transfer function, with C1 ¼ C2 ¼ C, R0 ¼ R1 ¼ R2 ¼ R and R3 ¼ R/n, SF
pffiffiffi
for the above oscillators is found to be SF ¼ 2 n . On the other hand, if both the
resistors R2 and R3 are varied simultaneously i.e., R2 ¼ R3 ¼ R/n, then SF becomes
2n. This figure appears to be the highest (like that of [44]) attained so far as
compared to all SRCOs known so far. Thus, both the circuits of Fig. 5.22 offer
very high frequency stability factors for larger values of n.
Thus, the circuits of Fig. 5.22 possess interesting and practically important
properties not available in any of the earlier known CFOA-based sinusoidal
oscillators.
Lastly, it must be mentioned that the generation of any new three-CFOA-two-
GC-four-resistor fully-uncoupled oscillators which, apart from retaining indepen-
dent controllability of FO, can also retain independent controllability of CO even
under the influence of the non-ideal parameters/parasitics of the CFOAs, appears to
be an interesting but challenging open problem.
5.11 State-Variable Synthesis of Linear VCOs Using CFOAs 161
5.10 Voltage-Controlled-Oscillators Using CFOAs

and FET-Based VCRs
Voltage-controlled oscillators (VCOs) are important building blocks in many

instrumentation, electronic and communication systems.
A well-known method of realizing a sinusoidal VCO is to realize a single-
resistance-controlled-oscillator (SRCO) and then replace the frequency controlling
resistor by a FET-based voltage-controlled-resistor (VCR). This section discusses a
number of CFOA-based VCO configurations which offer different advantageous
features.
A number of CFOA-based SRCOs have been described in the earlier sections of
this chapter in which frequency of oscillation (FO) can be independently controlled
through a single variable resistor without affecting the condition of oscillation
(CO). Thus, in a specific CFOA-based SRCO, this frequency controlling resistor
may be either grounded (having one terminal connected to ground) or floating (none
of the resistor terminals connected to ground).
From a careful examination of the family of 14 SRCOs presented earlier in [38],
it is found that in all, only seven structures are suitable for being converted into
VCOs by replacing frequency controlling resistors by appropriate grounded and
floating VCRs. Out of this set of seven, only five circuits can be realized using no
more than two CFOAs, which are shown here in Fig. 5.23. The CO and FO of these
VCOs are given in Table 5.1. It may be noted that in all the cases, the CO can be
adjusted by R1 without affecting FO whereas FO is controllable independently by
the resistance Rm and hence, by VC.
Taking into consideration the finite X-terminal input resistance Rx and parasitic
impedance at the Z-terminal (consisting of a resistance Rp in parallel with a
capacitance Cp) it is found that the influence of CFOA parasitics on the perfor-
mance of these oscillators can be reduced by choosing external resistances to be
much greater than Rx and much smaller than Rp and selecting external capacitors to
be much larger than Cp.
From the frequency stability analysis it has been found [50] that all the VCOs of
Fig. 5.23 enjoy good frequency stability properties.
Some experimental results for the oscillators of Fig. 5.23a, e from [50] are shown
here in Figs. 5.24, 5.25 and 5.26.
5.11 State-Variable Synthesis of Linear VCOs Using CFOAs
In this section, we describe a systematic approach of synthesizing sinusoidal linear

VCOs (i.e. an oscillator providing linear tuning law of the type f0 / VC between the
oscillation frequency f0 and an external control voltage VC).
a b
Y Y V02
2
R0
Y
1
V01 Z
W
R0
Y V01 R3 2
ZW
Z
W X V02 1 W X
X Z
R3 X
C2
R1
R0 R0
Vc R1
C1 Vc C1 C2
c d
Y
Y 2 W C2
R0 V01 V02
1 W X Z
X
Z Y V01 R3
R3 1 W Y V02
Z ZW
C2 X
2
X
R0 R1 R1 R0
Vc C1 C1
Vc
R0
e
Vc
R0
X V01 R1 R0
1 W X
Y Z
2 W
Z
Y V02
R3
C1 C2
Fig. 5.23 Voltage controlled oscillators proposed by Gupta et al. (adapted from [50] # 2009
Elsevier)
Table 5.1 CO and FO for the oscillators of Fig. 5.23

VCO number Condition of oscillation Frequency of oscillation
(a) R1 ¼ R3 f0 ¼ pffiffiffiffiffiffiffi
1ffi
2π C1 C2 Rm R3
(b), (d) R1 ¼ CC21 R3 f0 ¼ 2π pffiffiffiffiffiffiffi
1ffi
C C R1 2 m R3
(c) R1 ¼ 2 CC21 R3 f0 ¼ 2π pffiffiffiffiffiffiffi

1ffi
C C R1 2 m R3
(e) R1 ¼ R3 C1CþC 2 f0 ¼ 2π pffiffiffiffiffiffiffi

1ffi
C C R
2 1 2 m R3
2Vp2
where Rm ¼ rDS ¼ I
DSS ðVC 2Vp Þ
Fig. 5.24 Variation of

oscillation frequency with
control voltage VC for the
VCO (a) of Fig. 5.23 (adapted
from [50] # 2009 Elsevier).
Fig. 5.25 A typical

VCO (a) of Fig. 5.23,
f0 ¼ 263 kHz, V0 ¼ 1 V
(p–p); VCC ¼ 6 V DC
Elsevier)
It may be noted that the VCOs, presented in Sect. 5.11 are although simple,
however, they do not provide a linear tuning law between the control voltage (say,
VC) and the oscillation frequency f0, since the tuning law for such VCOs is of the form
rffiffiffiffiffiffi
1
f0 / (5.59)
rds
Vp2
where rds ¼ ðin case of VCR realized by JFETÞ (5.60)
IDSS ðVc 2Vp Þ
Fig. 5.26 A typical

VCO (e) of Fig. 5.23,
f0 ¼ 609 kHz, V0 ¼ 1 V
(p–p); VCC ¼ 5 V DC
Elsevier)
1
and rds ¼ ðin case of VCR realized by MOSFETÞ (5.61)
2kðVgs Vth Þ
In (5.60) and (5.61), Vp is the pinch off voltage of the JFET and IDSS is the
saturated drain current (at Vgs ¼ 0) of the FET, and VC is the control voltage, Vth is
the threshold voltage and K ¼ μS COX WL where μs is the surface mobility, Cox is

the capacitance of the gate electrode per unit area and WL is the aspect ratio of the
MOSFET.
However, if an oscillator is evolved with two analog multipliers (AM) appropri-
ately embedded into a circuit configuration, to enable independent control of the
oscillation frequency through an external control voltage VC applied as a common
multiplicative input to both the multipliers, this technique may give rise to a linear
tuning law of the form
f0 / VC (5.62)
Based upon this idea, some VCO configurations have been proposed by various
researchers in the past ([83–85] and the reference cited therein) employing tradi-
tional voltage-mode op-amps (VOA) and AMs. These circuits, however, require
larger number of resistors (5–12) and their usability is limited to low frequency
ranges due to the finite GBP and limited slew rate of VOAs.
In [65] Gupta et al. derived a class of VCOs using state variable technique
through which different circuits could be generated by making different choices of
the parameters of the [A] matrix (i.e. a11, a12, a21 and a22) of the state variable
characterization.
Based upon the already described technique let us construct the [A] matrix of the
oscillator to be synthesized in the following form:
" #
0 1
½A1 ¼ C1 R 2 (5.63)
C2 R 1 R 3
1 1 1 1
C2 R 3
The CE formulated from the above matrix gives the following CO and FO:
CO : R1 ¼ R3 (5.64)
1
C1 C2 R2 R3
The oscillators to be derived

have to incorporate at least two analog multipliers
(characterized byVo ¼ K V1 V2
Vref , where V1 and V2 are two inputs,Vref is the reference
voltage set internally, usually at 10 V in case of AD534 and K can be set up +1 or 1
by grounding appropriate input terminals). In order to provide linear control of
oscillation frequency through an external control voltage VC (to be applied as a
common multiplicative input to both the analog multipliers) the selection of the
matrix parameters
outlined above needs to be modified to include the term β
β ¼ VVrefC . However, it needs to be done in such a way that the final expression of
the CO does not contain the term β and the expression of FO is modified to f0
β
¼ 2π pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C1 C2 R2 R3
ffi so that we can get f0 / β and hence, f0 / VC .
The parameters of the matrix [A] given in (5.63) can now be modified in one of
the following ways:
(i) By including β2 as a factor of a12 or a21
(ii) β as a factor a12 as well as that of a21
(iii) β as a factor in all the parameters of matrix [A]
Using the modification (i) we get the following node equations
dx1 β 2 x2
C1 ¼ (5.66)
dt R2
dx2 x1 x2 x2
C2 ¼ þ (5.67)
dt R3 R1
If we employ two AMs and two CFOAs and try to implement above NEs we can
synthesize VCO-1 as shown in Fig. 5.27. Various current components of (5.66)
and (5.67) have been marked in VCO-1 to make the synthesis clear.
If we apply modification (iii), we get the following node equations:
dx1 β x2
C1 ¼ (5.68)
dt R2
x2
−b2
R1 VC
y
w 2
R2 z x
x R3
1 w
y z
x2 − x1 x2
R3 R1
C1 R1
dx1 dx2
C1 C2 C2
dt dt
VCO-1
β
Fig. 5.27 VCOs derived from matrix [A] CO: R1 R3 , FO: f0 ¼ 2π pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C1 C2 R2 R3
(adapted from [65]
# 2011 World Scientific Publishing Company)
β
Fig. 5.28 Additional VCOs CO: R1 R3, FO: f0 ¼ 2πpffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C1 C2 R2 R3
(adapted from [65] # 2011 World
Scientific Publishing Company)
dx2 β x1 x2 x2
C2 ¼ þ (5.69)
dt R3 R1
The implementation of (5.68) and (5.69) gives us VCO-2 which is shown in

Fig. 5.27. It may be noted that if polarity of β is inverted in both the AMs, the
synthesized circuit still remains VCO with the same CO and FO. Based on the state
variable methodology explained above, along with any or all the modifications
(i)–(iii) suggested above, a number of VCOs have been generated from the suitable
matrices in [65]. Two other circuits from the set of 12 VCOs generated in [65] are
shown here as VCO-3 and VCO-4 in Fig. 5.28.
Fig. 5.29 Experimental

results for VCO-1 of
Fig. 5.27. (a) Variation of
oscillation frequency with
control voltage VC. (b) A
typical waveform generated
for VC ¼ 2 V (adapted from
[65] # 2011 World Scientific
Publishing Company)
The expressions for the frequency stability factors for all the VCOs have been
evaluated in [65] and it has been found that in all the VCOs, SF can be made large.
Variation of oscillation frequency with control voltage VC (for C1 ¼ C2 ¼ 50 pF,
R2 ¼ R3 ¼ 10 kΩ, with DC biasing VCC ¼ 6 V for CFOAs and VCC ¼ 15 V
AMs) and a typical waveform generated by VCO-1 of Fig. 5.27 is shown in Fig. 5.29.
5.12 Synthesis of Single-CFOA-Based VCOs Incorporating

the Voltage Summing Property of Analog Multipliers
In this section, we present a family of CFOA-based VCOs which employ a bare

minimum number of active and passive components namely only one CFOA, only
two multipliers (essential for obtaining linear control of oscillation frequency), two/
three resistors and two capacitors.
The AD534 type AM is a 4-port building block symbolically shown in Fig. 5.30
with three differential inputs (shown as V1, V2 and
VZ1 in the present case) and one
output V0 and is characterized by Vo ¼ K V1 V2
Vref þ VZ where V1 and V2 are two
inputs, Vref is the reference voltage set internally, usually at 10 V in case of AD534
and K can be set up +1 or 1 by grounding appropriate input terminals, VZ is the
voltage applied at the third input terminal of AM which appears at the output
without any multiplying factor.
The VCO circuits presented in this section too are derived using the state-
variable methodology.
If we choose the required [A] matrix in the following form:
" #
1 1
1 1
½ A ¼ C1 R3 R1 C1 R 2 (5.70)
C21R3 0
The CE formulated from the above matrix gives the following CO and FO:
CO : R1 ¼ R3 (5.71)
1
C1 C2 R2 R3
Since the oscillators to be derived have to incorporate at least two analog

multipliers in order to provide linear control of oscillation frequency through an
external control voltage VC which is applied in place of the second input V2 of the
V1 +
− V0
V2 +
−
Z1 Z2
Vz1
Fig. 5.30 Symbolic notation V1V2

of an Analog Multiplier of (Vo = K + VZ1)
Vref
AD 534 type
5.12 Synthesis of Single-CFOA-Based VCOs Incorporating the Voltage Summing. . . 169
bx2 VC
x
w 3 R3
z y
b(x2−x1) y
R1 VC 2 w
y
1 w bx1 x z
x z
R2
C1 • bx1 C2 •
− C2x2
C1x1 R2
bx1
R2
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
β
Fig. 5.31 VCO derived from matrix [A] (CO: R1 ¼ R2, FO: f0 ¼ 2π 1
C1 C2 R1 R2) (adapted from [66])
multipliers (to be applied as a common multiplicative input to both the analog

multipliers), the selection of the
matrix
parameters outlined above needs to be
modified to include the term β β ¼ Vref . It should be done in such a way that the
VC
final expression of the CO does not contain the term β but the expression of FO is
β
modified to f0 ¼ 2π pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C1 C2 R2 R3
ffi so that we can get f0 / β and hence, f0 / VC .
Consider now the following [A] matrix
" #
β β
1
R11
½ A 2 ¼ C1 R2 C 1 R1
(5.73)
C2βR2 0
From matrix [A], we get the following node equations
βðx2 x1 Þ βx1
C1 x_ 1 ¼ þ (5.74)
R1 R2
βx1
C2 x_ 1 ¼ (5.75)
R2
Implementation of the above NEs employing two multipliers results in the

circuit shown in Fig. 5.31. Various current components of (5.74) and (5.75) have
been marked in the circuit to make the synthesis clear.
It may be noted that here we require three CFOAs along with two AMs to
implement NEs of (5.74) and (5.75). We now show that by an alternative represen-
tation of (5.74) and (5.75) and an appropriate incorporation of the z-terminal of AM
it becomes possible to implement the modified NEs with only a single CFOA. Let
us add (5.74) and (5.75) to create a new equation ((5.76) in the following) while we
keep (5.75) as it is (shown as (5.77) in the following):
+
− Vc
2 −
R1 z2z1 +
bx1
Vc + − y
− R2
1 w
b(x2−x1)
−
+ z1z2 x z
R1
R2
C2 •
• C1 C2 x2
C1 x1
VCO-1
Fig. 5.32 VCO derived from matrix [A] (adapted from [66])
βðx2 x1 Þ
C1 x_ 1 þ C2 x_2 ¼ (5.76)
R1
βx1
C2 x_ 2 ¼ (5.77)
R2
It is interesting to note that when the voltage summing property of the AMs is
effectively utilized the implementation of (5.76) and (5.77) then leads to a different
circuit (shown as VCO-1 in Fig. 5.32) which requires only a single CFOA in
contrast to the circuit of Fig. 5.31 needing three CFOAs.
It has been shown in [66] that in addition to (5.73), the following matrices are
suitable for the synthesis of such kind of VCOs.
" #
β 1 β
C1 ðR2 R21 Þ
½A2 ¼ C1 R 1
(5.78)
C2βR2 0
" #
β
C1βR1
½A3 ¼ C1 R1
(5.79)
2β
C2 R2 C2βR2
2 3
β2 β2
6 C R
½A4 ¼ 6 1 1 C1 R1 7
7 (5.80)
4 1 þ β2 β2 5

C2 R2 C2 R2
5.12 Synthesis of Single-CFOA-Based VCOs Incorporating the Voltage Summing. . . 171
2 3
β
6 0 7
½A5 ¼ 6 C1 R2 7 (5.81)
4 β 1 1 1 5

C2 R3 C2 R3 R1
The circuits resulting from the synthesis based upon the above matrices are
shown in Fig. 5.33.
The following may now be noted:
• VCOs 3 and 4 offer the use of both grounded capacitors as desirable for IC
implementation and out of these VCO 4 also has one of the CO controlling
resistor R2 grounded.
• The VCO 5 possesses simultaneously almost all the desirable features namely,
completely non-interacting control of CO through R1 (the CO controlling resis-
tor being grounded), the employment of both grounded capacitors and an
additional degree of freedom via R2 to scale up or down the frequency f0
which is otherwise linearly controllable by β.
The prominent non-idealities of the CFOAs include—a finite non-zero input
resistance Rx at port-X (typically around 50 Ω), y-port parasitic consisting of a
parasitic resistance Ry (typically 2 MΩ) in parallel with a parasitic capacitance Cy
(typically 2 pF) and Z-port parasitic impedance consisting of a parasitic resistance
Rp (typically 3 MΩ) in parallel with a parasitic capacitance Cp (typically, between 4
and 5 pF). In case of an analog multiplier, the finite non-zero output resistance rout,
as per datasheet of AD534, is merely 1 Ω and hence, can be ignored in all the cases.
On the other hand, the input impedance of the AM, being 10 MΩ, is sufficiently
high and hence, its effect can be ignored. The errors caused by the influence of
CFOA parasitics can be kept small by choosing all external resistors to be much
larger than Rx but much smaller than Rp and choosing both external capacitors to be
much larger than Cp.
A non-ideal analysis carried out in [66] shows that the independent control of
CO and FO remains intact for VCO-5 even after consideration of the parasitics.
Hence, VCO-5 is the best circuit from this viewpoint.
All the VCOs have been experimentally studied in [66] using AD844 type
CFOAs and AD534 type AMs biased with 12 V DC power supplies. The compo-
nent values chosen were as under: For VCOs R1 ¼ R2 ¼ 2 kΩ, those for VCO-2
were chosen as R1 ¼ 2 kΩ, R2 ¼ 1 kΩ and for VCO-5 R1 ¼ R2 ¼ R3 ¼ 1 kΩ.
Capacitor values for all the VCOs were taken as C1 ¼ C2 ¼1 nF. As per [66] it has
been possible to generate oscillation frequencies from tens of kHz to several
hundreds of kHz with tolerable errors in the frequency.
In the absence of an automatic amplitude control, it is normally expected that
amplitude of oscillation would also vary when the frequency is varied through the
external control voltage VC. This has indeed been the case for VCOs 1, 2. However,
in case of VCOs 3, 4 and 5, the peak-to-peak output voltage has been found to be
constant, 17 Vp–p in case of VCO 3 and 4 and 10 Vp–p in case of VCO 5 when VC
+
− Vc
Vc +
−
2 − R2
z2z1 + − 1
R1 + z1z2 x
2 w
z
R2 y
Vc +
x −
1 R1
1 w +z z
− z2
1 2
y z z1 2
+ +
− −
Vc
C1 C2 C1 C2
VCO-2 VCO-3
C2
CO: R1= 2R2 CO: R1 = R2
C1
b 1 b 1
FO: f0 = FO: f0 =
2p C1C2R1R2 2p C1C2R1R2
R2 +
− Vc
R2 2
z2z1 + −
x
R1 w
z
y
Vc +z1z2
− y
z
z12 2 − 1 w
− + + −
+ z
Vc x
z2 2
C1 z C2 C1 C2
+1 −−
+ R3 R1
Vc
VCO-4 VCO-5
C2
CO: R1 = R2
C1
CO: R1 = R3
b 1 b 1
FO: f0 = f0 =
2p C1C2R1R2 2p C1C2R2R3
Fig. 5.33 GC-VCOs derived from matrices [A]2 [A]5 (adapted from [66])
was varied from 1–10 V. Thus, VCOs 3, 4 and 5 have been found to be superior than
the other VCOs in this respect.
Some sample results of the various VCOs are shown in Fig. 5.34a, b which show
the variation of oscillation frequency with control voltage VC for VCOs 1 and 5
respectively.
5.13 MOSFET-C Sinusoidal Oscillator 173
a b
VCO-1
80
70
60
frequency, kHz
50
40
30
20
10
0
0 2 4 6 8 10
VC in volts
Fig. 5.34 Experimental results of the VCOs. (a) Variation of frequency with VC for VCO-1, (b)
Variation of frequency with VC for VCO-5 (adapted from [66])
In view of a number of new CMOS CFOA and CMOS multiplier architectures

being evolved in the recent literature, it may be expected that these ideas could possibly
be carried over to the design of completely CMOS-based linear VCOs in future.
5.13 MOSFET-C Sinusoidal Oscillator
Since filters and oscillators are closely related, it is obvious that the techniques
which are employed to synthesize a biquad filter can as well be employed to
synthesize a sinusoidal oscillator. In principle, an active RC band pass filter can
be easily converted into a MOS-C oscillator by replacing resistors by MOS-VCRs.
For this purpose, the classical two-integrator-loop is a natural choice. From the
previous chapter it is known that both lossy and lossless MOSFET-C integrators can
be realized in a number of ways.
Three popular nonlinearity cancellation techniques involving one, two and four
MOSFETs are shown in Fig. 5.35.
Assuming that all the MOS transistors are operating in triode region, the current
I in case of Fig. 5.35a is given by

W
I ¼ 2KðVG VTH Þ for ðVG VTH Þ jV1 j where K ¼ μn COX (5.82)
L
and symbols have their usual meaning.

a c
VG V
I1 M1
I M
V1 −V1 I2 M2
V1
+
b V2 VG
M1 I1
−
V1 V2
M3
+
VG I3
−
M2 M4
V
V1 V2
I2 I4
Fig. 5.35 (a) An NMOS transistor with even nonlinearities cancellation. (b) Two MOS transistors
circuit with full nonlinearities cancellation. (c) Four MOS transistors circuit with full
nonlinearities cancellation (adapted from [67] # 2000 Taylor & Francis)
For the circuit of Fig. 5.35b,
I ¼ ðI1 I2 Þ ¼ KVG ðV1 V2 Þ forðVG VTH Þ maxðV1 ; V2 Þ (5.83)
and lastly, for the circuit of Fig. 5.35c,
I ¼ ðI5 I6 Þ ¼ KVG ðV1 V2 Þ (5.84)
Utilizing the two MOSFETs and four MOSFETs implementation, a two CFOA
grounded capacitor quadrature oscillator was proposed by Mahmoud and Soliman
[67] which is shown in Fig. 5.36.
A straight forward analysis of the oscillator of Fig. 5.36 shows that the condition
of oscillation and frequency of oscillation of the circuit are given by
gm1 ¼ gm3 (5.85)

rffiffiffiffiffiffiffiffiffiffiffiffiffiffi
gm2 gm3
ω0 ¼ (5.86)
C1 C2
where
gmi ¼ Ki VGi ði ¼ 1; 2 and 3Þ (5.87)
Thus, the condition of oscillation can be controlled by the transconductance gm1

and hence, by VG1 while frequency of oscillation can independently be tuned by
gm2 and hence by VG2.
K2 M3
M1 M4
K2
K1 V1 + y
y
− w VG 2 w V2
V x − x z
+ G1 z K2
K1 M5
M2 C2
C1 K2
M6
M8
K3
+
VG 3
−
M7 K3
Fig. 5.36 MOS-C CFOA quadrature oscillator proposed by Mahmoud and Soliman (adapted from
[67] # 2000 Taylor & Francis)
In this chapter, we have elaborated a variety of sinusoidal oscillators which

included canonic SRCOs using a single CFOA, two-CFOA-based oscillators
employing grounded capacitors, SRCOs with explicit current output and fully-
uncoupled SRCOs providing independent control of FO and CO through separate
resistors exhibiting the notable property that independent control of FO remains
intact even under the influence of the various parasitic impedances of a CFOAs.
Also presented were VCOs employing nonlinearity-cancelled FETs through a
general scheme. Subsequently, two different varieties of linear VCOs based upon
the use of analog multipliers in conjunction with CFOAs were described. In the first
one, through a systematic state variable formulation, a class of two-CFOA linear
VCOs were synthesized which provide linear control FO through an external
voltage (i.e. f0 / Vc). This was followed by another family of VCOs, again
synthesized through a state variable methodology, in which the circuit complexity
was reduced by appropriately taking into account the voltage addition feature of the
AD534 type analog multipliers. The resulting VCOs not only provide linear tuning
law but in contrast to the circuits presented in the earlier section, these circuits
can be implemented with only a single CFOA and two analog multipliers. A CFOA-
based MOS-C oscillator was also described.
Lastly, we would like to outline two interesting ideas worthy of further
investigations and research. In Sect. 5.4 a family of eight single-CFOA oscillators
was presented. Unfortunately, till date, there has not been any comparative study of
all the eight circuits to determine as to which one of these is the best of the entire
class and this problem is still open to investigation.
Yet another problem whose solution has not yet been found is whether or not a
Single-CFOA SRCO using only two grounded capacitors exists? In view of the fact
that such a circuit using a single VOA does exist [7], the existence of a similar or
better circuit with a single CFOA may not be ruled out. This constitutes another
interesting problem for research.
References
1. Hribsek M, Newcomb RW (1976) VCO controlled by one variable resistor. IEEE Trans Circ
Syst 23:166–169
2. Soliman AM, Awad SS (1978) A novel sine-wave generator using a single operational
amplifier. Proc IEEE 66:253–254
3. Senani R (1979) New canonic sinusoidal oscillator with independent frequency control
through a single grounded resistor. Proc IEEE 67:691–692
4. Pyara VP, Dutta Roy SC, Jamuar SS (1983) Identification and design of single amplifier single
resistance controlled oscillators. IEEE Trans Circ Syst 30:176–181
5. Bhattacharyya BB, Darkani MT (1984) A unified approach to the realization of canonic
RC-active, single as well as variable, frequency oscillators using operational amplifiers.
J Franklin Inst 317:413–439
6. Singh V (2001) Realisation of operational floating amplifier based 1-op-amp based sinusoidal
oscillators. IEEE Trans Circ Syst 48:377–381
7. Singh V (1980) Novel sinusoidal oscillator employing grounded capacitors. Electron Lett
16:757–758
8. Dutta Roy SC, Pyara VP (1979) Single element controlled oscillators: a network synthetic
approach. Proc IEEE 67:1565–1566
9. Senani R (1988) Analysis, synthesis and design of new types of RC-active sinusoidal
oscillators. Part I and part II. Frequenz 42(223–228):251–256
10. Martinez PA, Sabadell J, Aldea C (1997) Grounded resistor controlled sinusoidal oscillator
using CFOAs. Electron Lett 33:346–348
11. Senani R (1980) Novel sinusoidal oscillator employing grounded capacitors. Electron Lett
16:62–63
12. Senani R (1994) On equivalent forms of single op-amp sinusoidal RC oscillators. IEEE Trans
Circ Syst-I 41:617–624
13. Martinez PA, Celma S, Sabadell J (1996) Designing sinusoidal oscillators with current-
feedback amplifiers. Int J Electron 80:637–646
14. Celma S, Martinez PA, Carlosena A (1994) Current feedback amplifiers based sinusoidal
oscillators. IEEE Trans Circ Syst-I 41:906–908
15. Singh VK, Sharma RK, Singh AK, Bhaskar DR, Senani R (2005) Two new canonic single-
CFOA oscillators with single resistor controls. IEEE Trans Circ Syst-II 52:860–864
Novel configurations using current feedback op-amps. Frequenz 52:196–206
17. Soliman AM (1996) Applications of current feedback operational amplifiers. Analog Integr
18. Celma S, Martinez PA, Carlosena A (1994) Approach to the synthesis of canonic RC-active
oscillators using CCII. IEE Proc Circ Devices Syst 141:493–497
19. Senani R, Singh VK (1996) Comment: synthesis of canonic single-resistance-controlled-
oscillators using a single current-feedback-amplifier. IEE Proc Circ Devices Syst 143:71–72
20. Liu SI, Shih CS, Wu DS (1994) Sinusoidal oscillators with single element control using a
current-feedback amplifier. Int J Electron 77:1007–1013
References 177
21. Abuelma’atti MT, Farooqi AA, Al-Shahrani SM (1996) Novel RC oscillators using the
current-feedback operational amplifier. IEEE Trans Circ Syst-I 43:155–157
22. Liu SI, Tsay JH (1996) Single-resistance-controlled sinusoidal oscillator using current-
feedback amplifiers. Int J Electron 80:661–664
23. Abuelma’atti MT, Al-Shahrani SM (1996) A novel low-component count single-element-
controlled sinusoidal oscillator using the CFOA pole. Int J Electron 80:747–752
24. Soliman AM (2000) Three oscillator families using the current feedback op-amp. Frequenz
54:126–131
25. Senani R, Singh VK (1996) Novel single-resistance-controlled-oscillator configuration using
current feedback amplifiers. IEEE Trans Circ Syst-I 43:698–700
26. Abuelma’atti MT, Farooqi AA (1996) A novel single-element controlled oscillator using the
current-feedback-operational amplifier pole. Frequenz 50:183–184
27. Toker A, Cicekoglu O, Kuntman H (2002) On the oscillator implementations using a single
current feedback op-amp. Comput Electr Eng 28:375–389
28. Gunes EO, Toker A (2002) On the realization of oscillators using state equations. Int J Electron
Commun 56:317–326
29. Martinez PA, Sanz BMM (2005) Generation of two integrator loop variable frequency
sinusoidal oscillator. Int J Electron 92:619–629
30. Abuelma’atti MT, Al-Ghazwani A (2000) New quartz crystal oscillators using the current-
feedback operational amplifier. Active Passive Electron Comp 23:131–136
31. Hou CL, Wang WY (1997) Circuit transformation method from OTA-C circuits into CFA-
based RC circuits. IEE Proc Circ Devices Syst 144:209–212
32. Wu DS, Liu SI, Hwang YS, Wu YP (1995) Multiphase sinusoidal oscillator using the CFOA
pole. IEE Proc Circ Devices Syst 142:37–40
33. Abuelma’atti MT, Al-Shahrani AM (1998) Novel CFOA-based sinusoidal oscillators. Int J
34. Liu SI, Chang CC, Wu DS (1994) Active-R sinusoidal oscillator using the CFA pole. Int J
Electron 77:1035–1042
35. Abuelma’atti MT, Al-Shahrani SM (1997) New CFOA-based grounded-capacitor single-
element-controlled sinusoidal oscillator. Active Passive Electron Comp 20:119–124
36. Senani R, Gupta SS (1997) Synthesis of single-resistance-controlled oscillators using CFOAs:
simple state-variable approach. IEE Proc Circ Devices Syst 144:104–106
37. Gupta SS, Senani R (1998) State variable synthesis of single-resistance-controlled grounded
capacitor oscillator oscillators using only two CFOAs. IEE Proc Circ Devices Syst
145:135–138
38. Gupta SS, Senani R (1998) State variable synthesis of single-resistance-controlled grounded
capacitor oscillator oscillators using only two CFOAs: additional new realizations. IEE Proc
39. Abuelma’atti MT, Al-Zaher HA (1998) New grounded-capacitor sinusoidal oscillators using
the current-feedback-amplifier pole. Active Passive Electron Comp 21:23–32
40. Abuelma’atti MT, Al-Shahrani SM (1997) New CFOA-based sinusoidal oscillators. Int J
Electron 82:27–32
41. Elwakil AS (1998) Systematic realization of low-frequency oscillators using composite
passive–active resistors. IEEE Trans Instrument Measure 47:584–586
42. Soliman AM (2000) Current feedback operational amplifier based oscillators. Analog Integr
43. Singh AK, Senani R (2001) Active-R design using CFOA-poles: new resonators, filters and
oscillators. IEEE Trans Circ Syst-II 48:504–511
44. Bhaskar DR (2003) Realization of second-order sinusoidal oscillator/filters with non-
interacting controls using CFAs. Frequenz 57:12–14
45. Abuelma’atti MT, Al-Shahrani SM (2003) Synthesis of a novel low-component programmable
sinusoidal oscillator. Active Passive Electron Comp 26:31–36
46. Bhaskar DR, Prasad D, Imam SA (2004) Grounded-capacitor SRCOs realized through a
simple general scheme. Frequenz 58:175–177
47. Senani R, Sharma RK (2005) Explicit-current-output sinusoidal oscillators employing only a
single current-feedback op-amp. IEICE Electron Express 2:14–18
48. Bhaskar DR, Senani R (2006) New CFOA-based single-element-controlled sinusoidal
oscillators. IEEE Trans Instrument Measure 55:2014–2021
49. Nandi R (2008) Tunable active-R oscillator using a CFA. IEICE Electron Express
5(8):248–253
50. Gupta SS, Bhaskar DR, Senani R (2009) New voltage controlled oscillators using CFOAs. Int J
Electron Commun 63:209–217
51. Tangsrirat W, Surakampontorn W (2009) Single-resistance-controlled quadrature oscillator
and universal biquad filter using CFOAs. Int J Electron Commun 63:1080–1086
52. Soliman AM (2010) Transformation of oscillators using Op amps, unity gain cells and CFOA.
Analog Integr Circ Sign Process 65:105–114
53. Koren V (2002) RF oscillator uses current-feedback op-amp. EDN:83–84
54. Soliman AM (2011) Transformation of a floating capacitor oscillator to a family of grounded
capacitor oscillators. Int J Electron 98:289–300
55. Abuelma’atti MT (2010) Identification of a class of two CFOA-based sinusoidal RC
oscillators. Analog Integr Circ Sign Process 65:419–428
56. Wangenheim LV (2011) Comment on ‘Identification of a class of two CFOA-based sinusoidal
RC oscillators’. Analog Integr Circ Sign Process 67:117–119
57. Abuelma’atti MT (2012) Reply to comment on “Identification of class of two CFOA-based
sinusoidal RC oscillators”. Analog Integr Circ Sign Process 71:155–157
58. Lahiri A, Jaikla W, Siripruchyanun M (2011) Explicit-current-output second-order sinusoidal
oscillators using two CFOA’s and grounded capacitors. Int J Electron Commun 65:669–672
59. Bhaskar DR, Gupta SS, Senani R, Singh AK (2012) New CFOA-based sinusoidal oscillators
retaining independent control of oscillation frequency even under the influence of parasitic
impedances. Analog Integr Circ Sign Process 73:427–437
60. Bhaskar DR, Senani R, Singh AK (2010) Linear sinusoidal VCOs: new configurations using
current feedback op-amps. Int J Electron 97:263–272
61. Bhaskar DR, Senani R, Singh AK, Gupta SS (2010) Two simple analog multiplier based linear
VCOs using a single current feedback Op-amp. Circ Syst 1:1–4
62. Martinez PA, Sabadell J, Aldea C, Celma S (1999) Variable frequency sinusoidal oscillators
based on CCII+. IEEE Trans Circ Syst-I 46:1386–1390
63. Gupta SS, Senani R (2003) Realizations of current-mode SRCOs using all grounded passive
elements. Frequenz 57:26–37
64. Gupta SS, Sharma RK, Bhaskar DR, Senani R (2010) Sinusoidal oscillators with explicit
current output employing current-feedback op-amps. Int J Circ Theor Appl 38:131–147
65. Gupta SS, Bhaskar DR, Senani R (2011) Synthesis of linear VCOs: the state-variable
approach. J Circ Syst Comput 20:587–606
66. Gupta SS, Bhaskar DR, Senani R (2012) Synthesis of new single CFOA-based VCOs
incorporating the voltage summing property of analog multipliers. ISRN Electron: Article
ID 463680, 8 p
67. Mahmoud SA, Soliman AM (2000) Novel MOS-C oscillators using the current feedback op-
amp. Int J Electron 87:269–280
68. Nay K, Budak A (1983) A voltage-controlled-resistance with wide dynamic range and low
distortion. IEEE Trans Circ Syst 30:770–772
69. Senani R (1994) Realization of linear voltage-controlled resistance in floating form. Electron
Lett 30:1909–1911
70. Gupta SS (2005) Realization of some class of linear/nonlinear analog electronic circuits using
current-mode building blocks. Ph.D. thesis. Faculty of Technology, University of Delhi
71. Natarajan S (1989) Measurement of capacitances and their loss factors. IEEE Trans Instrument
Measure 38:1083–1087
References 179
72. Ahmad W (1986) A new simple technique for capacitance measurement. IEEE Trans Instru-
ment Measure 35:640–642
73. Awad SS (1988) Capacitance measurement based on an operational amplifier circuit: error
determination and reduction. IEEE Trans Instrument Measure 37:379–382
74. Horng JW (2001) A sinusoidal oscillator using current-controlled current conveyors. Int J
75. Fongsamut C, Anuntahirunrat K, Kumwachara K, Surakampontorn W (2006) Current-con-
veyor-based single-element-controlled and current-controlled sinusoidal oscillators. Int J
76. Senani R, Gupta SS (2000) Novel SRCOs using first generation current conveyor. Int J
Electron 87:1187–1192
77. Kilinc S, Jain V, Aggarwal V, Cam U (2006) Catalogue of variable frequency and single-
resistance-controlled oscillators employing a single differential-difference-complementary-
current-conveyor. Frequenz 60:142–146
78. Gupta SS, Senani R (2005) Grounded-capacitor SRCOs using a single differential difference
complementary current feedback amplifier. IEE Proc Circ Devices Syst 152:38–48
79. Gupta SS, Senani R (2006) New single resistance controlled oscillator configurations using
unity-gain cells. Analog Integr Circ Sign Process 46:111–119
80. Moon G, Zaghloul ME, Newcomb RW (1990) An enhancement-mode MOS voltage-
controlled linear resistor with large dynamic range. IEEE Trans Circ Syst 37:1284–1288
81. Elwan HO, Mahmoud SA, Soliman AM (1996) CMOS voltage-controlled floating resistor. Int
J Electron 81:571–576
82. Al-Shahrani SM (2007) CMOS wideband auto-tuning phase shifter cir cuit. Electron Lett
43:804–806
83. Senani R, Bhaskar DR, Tripathi MP (1993) On the realization of linear sinusoidal VCOs. Int J
84. Senani R, Bhaskar DR (1996) New active-R sinusoidal VCOs with linear tuning laws. Int J
Electron 80:57–61
85. Bhaskar DR, Tripathi MP (2000) Realization of novel linear Sinusoidal VCOs. Analog Integr
86. Singh VK (2004) Realization of a class of analog signal processing/signal generation circuits.
Ph.D. thesis. Uttar Pradesh Technical University, Lucknow
87. Soliman AM, Awad SS (1978) A canonical voltage controlled oscillator realized using a single
operational amplifier. Frequenz 32:153–154
88. Schmid H (2003) Why ‘Current Mode’ does not guarantee good performance. Analog Integr
Chapter 6
Miscellaneous Linear and Nonlinear
Applications of CFOAs
6.1 Introduction
In the preceding chapters we have discussed applications which prove the utility of
CFOA as a versatile building block in realizing a variety of linear circuits. It is not
surprising therefore that because of wide spread use of CFOAs they have received
attention as attractive building blocks for realizing a variety of non-linear functions
as well. In this chapter, we would provide an exposition to the application of
CFOAs in realizing miscellaneous linear and non-linear functions and non-
sinusoidal waveform generators which include both relaxation and chaotic
oscillators.
6.2 Electronically-Variable-Gain Amplifier
AD 844 CFOA is an excellent choice as an output amplifier to be used in conjunc-

tion with analog multiplier AD 539 in various connection modes of this multiplier.
A voltage variable gain amplifier realized by a combination of an analog multiplier
AD 539 and the CFOA AD844 [1] is shown in Fig. 6.1.
The output voltage of the circuit is given by V0 ¼ VVinrefVc where Vc is the external
control voltage and Vref ¼ 2 V. The gain Vout/Vin can be electronically controlled
through Vc. In this case Vc is to be taken as a positive voltage which can be varied
from 0 to 3.2 V (max.) while the signal voltage Vin is to be kept nominally 2 V full
scale but can be extended up to 4.2 V.
182 6 Miscellaneous Linear and Nonlinear Applications of CFOAs
+Vs
10 Ω 0.22μF
1 15,16
control input Vc 4 z1, w1
AD 539
3 w2, z2 9,10
Vy1
signal input Vin Ch1 14
Vy2 Ch2 x
6 5 Base common
11 w V0
12 13 y
z
10 Ω 0.22µF
−Vs
Fig. 6.1 Electronically-variable-gain amplifier using AD539 and AD844
300 Ω 300 Ω
+V=+5V
2.2 μF Z0=50 Ω
50 Ω
x
Vin y z w Vout
2.2 μF 50 Ω
50 Ω
−V = −5V
Fig. 6.2 CFOA as a cable driver
6.3 Cable Driver Using CFOA
Most CFOAs find an immediate application to drive low impedance cables. Fig-
ure 6.2 shows an illustrative application that provides a gain of +2 by configuring
CFOA as a non-inverting amplifier. It is easy to see that the arrangement provides
an overall gain of +1 to the signal reaching the load RL. With a CFOA AD844, the
circuit provides 3 dB bandwidth of around 30 MHz.
6.4 Video Distribution Amplifier
Several CFOAs, such as THS 3001, find an excellent application as a video

distribution amplifier as shown in Fig. 6.3.
6.5 Schmitt Triggers and Non-sinusoidal Waveform Generators 183
750 Ω 750 Ω
Z0 = 75 Ω
75 Ω
x
w Transmission line V01
Vi y
z
n lines 75 Ω
75 Ω
Z0 = 75 Ω
75 Ω
Transmission line V0n
75 Ω
Fig. 6.3 Video distribution amplifier
6.5 Schmitt Triggers and Non-sinusoidal Waveform Generators
At the root of any non-sinusoidal signal generator lies either a comparator or a

comparator with hysteresis (using positive feedback) often referred as Schmitt
trigger. The first Schmitt trigger using a CCII+ was presented by Di Cataldo et al. [2].
Consider now the Schmitt trigger of Fig. 6.4 which is, in fact, a CFOA version of
the CCII-based Schmitt trigger of Cataldo et al. [2]. In a CFOA, the output voltage
is ultimately limited to Vsat+ and Vsat with the current flowing into the Z-terminal
being
Vsatþ
Isatþ ¼ (6.1)
R1 þ R2
Vsat
Isat ¼ (6.2)
R1 þ R2
If the two threshold voltages are VTL and VTH, they can be determined as
follows.
If we assume that V0 is in the state Vsat+ then to change this stable state, the
current ix must satisfy the condition ix iz which means
Vin Vy Vsatþ
(6.3)
Rs R 1 þ R2
The higher threshold level VTH is, therefore, given by
R 1 Rs
VTH ¼ Vsatþ (6.4)
R1 þ R2
Fig. 6.4 Schmitt Trigger

circuit using a CFOA Vin X
Rs Vo
(adapted from [2] # 1995 Z
W
John Wiley & Sons Ltd.) Y
R1 R2
Fig. 6.5 Transfer Vo

characteristics of the Schmitt
Trigger of Fig. 6.4 (adapted Vsat+
from [2] # 1995 John Wiley
& Sons Ltd.) (R1-Rs)
Vsat-
(R1+R2)
Vin
(R1-Rs)
Vsat+
(R1+R2)
Vsat-
Similarly, it can be found that the lower threshold level VTL is given by
R1 Rs
VTL ¼ Vsat (6.5)
R1 þ R2
From the above analysis, the transfer characteristic of this Schmitt trigger can be
drawn as shown in Fig. 6.5.
The circuit can be easily converted into a relaxation oscillator by connecting a
capacitor from the input terminal-X to ground. With this addition, the circuit would
generate a square wave output at Vo. Figure 6.6 shows the resulting relaxation
oscillator incorporating the non-ideal model of the CFOA AD844 where the non-
ideal parameter values are typically given by rx ¼ 50 Ω, Ry ¼ 2 MΩ, Rp ¼ 3 MΩ,
Cx ¼ Cy ¼ 2 pF and Cp ¼ 4.5 pF. In reference [3], it has been shown that the
oscillation period of the waveform generated by this circuit is given by

R1
T ¼ 2CT rx ln 2 1 ; where CT ¼ C þ Cx (6.6)
rx
Thus, the time period T is a function of the external capacitor C and the resistors
rx and R1.
Fig. 6.6 Non-ideal model of

the CFOA showing various
parasitic impedances X
(adapted from [3] # 2005 rx
Taylor & Francis) C CX
1 W
1 Vo
ix
Y
Z
Vo
Ry Cy
Rp Cp
Fig. 6.7 An improved

Vin X
CFOA-version of the CCII- Vo
Rs W
based circuit of Schmitt Y Z
Trigger using two CFOAs
John Wiley & Sons Ltd.)
Y
Z
W
X
R1 R2 R3
An improved CFOA-version of the CCII-based Schmitt trigger proposed by

Srinivasulu [4] is shown here in Fig. 6.7.
In this circuit, the two threshold voltage levels are given by

R2 R s
VTH ¼ 1 Vsatþ (6.7)
R1 R 3

R2 Rs
VTL ¼ 1 Vsat (6.8)
R 1 R3
Based upon the above, the transfer characteristics of the circuit can be drawn as
follows (Fig. 6.8).
A square wave/triangular wave generator using the Schmitt trigger of Fig. 6.7 is
shown in Fig. 6.9.
In this circuit, the resistors R and R4 together with the capacitor C constitute an
integrator. A straight forward analysis of this circuit shows that the time period (T)
of the waveforms generated (a square wave at V01 and triangular wave at V02) is
given by
Vo
Vsat+
Vin
(RsR2) 0
- (1- ) Vsat- (RsR2)
(R1R3) (1- )Vsat-
(R1R3)
Vsat-
Fig. 6.8 Transfer characteristic of the Schmitt trigger of Fig. 6.7 (adapted from [4] # 2011 John
Wiley & Sons Ltd.)
X
WZ Rs
Y
Y Vo2
Y R
ZW
W X
XZ Vo1
C
R1 R2 R3 R4
Fig. 6.9 A square/triangular wave generator using Schmitt Trigger of Fig. 6.7 proposed by
Srinivasulu (adapted from [4] # 2011 John Wiley & Sons Ltd.)

R2 Rs
T ¼ RC 1 (6.9)
R1 R 3
Another two-CFOA-based triangular/square wave generator was advanced by

Haque et al. in [5]. The circuit, however, requires two CFOAs, four resistors, one
capacitor and 2n number of diodes to stabilize the Schmitt trigger output levels at
nVD(on).
The frequency of oscillation for triangular/square wave generator by the circuit
of Fig. 6.10 is given by
VCC
R1 Dn
C1 R4 D1
X X
1 2 VR
W VT W
Z Z
Y Y
D1
R2 R3
Dn
VEE
Fig. 6.10 Another two CFOA based triangular/square wave generator (adapted from [5] # 2008
IEEE)

VZNT 1 þ RR23 þ VRR3 ZRT2
VR
R1
f0 ¼ h i (6.10)
4C1 VN 1 þ RR23 VRR R3 2
where ZT is the open loop transimpedance of the CFOA, VN is the peak voltage at
X-input terminal of CFOA2 and VR is the peak voltage of the square waveform.
A novel two CFOA and one grounded capacitor based square/triangular wave
generator was proposed by Minaei and Yuce in [6]. This circuit is shown in
Fig. 6.11.
The operation of this circuit can be explained as follows. Both the CFOAs in this
circuit operate as voltage saturated elements. If we assume Vsquare ¼ Vsat+, the
capacitor charges by a constant current Vsat+/R3 so that a positive ramp appears
at the output of CFOA2 consequently, current flowing through R1 decreases. When
ix becomes iz then output voltage of CFOA1 switches to other stable state Vsat.
Accordingly, we can write
Vsatþ VSatþ VtriðPeakþÞ

¼ (6.11)
R2 R1
From the above equation, the positive peak voltage of the triangular wave
(higher threshold voltage) and the negative peak voltage (lower threshold voltage)
are respectively given by

R1
VtriðPeakþÞ ¼ 1 VSatþ (6.12)
R2
Z Y
W 1
VSquare X
R1
Y
2
W
Z
X VTriangular
R2 R3 C
Fig. 6.11 A low-component-count CFOA-based square/triangular wave generator proposed by

Minaei and Yuce (adapted from [6] # 2012 Springer)

R1
VtriðPeakÞ ¼ 1 VSat (6.13)
R2
Assuming the two saturation voltages to be equal in magnitude, the time period
of the waveforms generated by this circuit is given by

R1
T ¼ 4CR3 1 (6.14)
R2
Finally, we show a triangular/square wave generator made from a single CFOA

as shown in Fig. 6.12a. This circuit was proposed by Abuelma’atti and Al-Shahrani
in [7]. In this circuit, the CFOA behaves as a Schmitt trigger with the input–output
characteristic shown in Fig. 6.12b where the two threshold voltages are given by
R1 r x R1 r x
VTH ¼ Vsatþ and VTL ¼ Vsat (6.15)
R1 þ R 2 R1 þ R2
where Vsat+ and Vsat are two stable states decided by the DC biasing power supply
voltages of the CFOA and rx is the input resistance of the CFOA looking into
terminal-X of the CFOA.
The circuit can be analyzed by starting from any one of the two stable states of
the output voltage V0 (for details, the reader is referred to [7]). The circuit generates
a square wave signal at V0 and a triangular wave signal at Vx. The frequency of the
generated waveforms is given by
6.6 Precision Rectifiers 189
Fig. 6.12 Relaxation a

oscillator proposed by
Abuelma’atti and Al- R2
Shahrani (adapted from [7] y
# 1998 Taylor & Francis).
w Vo
(a) Triangular/square wave z
generator. (b) Transfer x
characteristic of the Schmitt
Trigger composed of CFOA R1 C R3
along with R2 and R1
b Vo
Vsat+
(R1-rx)
Vsat-
(R1+R2)
Vi
(R1-rx)
Vsat+
(R1+R2)
Vsat-

R1 rx 1 R2
f ¼1 2CR3 ffi 1þ ; for R1 >> rx (6.16)
R1 þ R2 2CR3 R1
Out of the various circuits presented, the one in Fig. 6.12 is appealing due to its
lowest-component-count whereas those of Figs. 6.10 and 6.11 have the advantage
of providing low-output impedance outputs for both square and triangular wave
outputs.
6.6 Precision Rectifiers
There have been several attempts of making precision current-mode full wave
rectifiers using current conveyors quite often realized with AD844-type CFOAs.
Here, we present a typical design of a simple full wave precision rectifier circuit
proposed by Khan et al. in [8]. This circuit is shown in Fig. 6.13 and is claimed to
Fig. 6.13 Full wave rectifier VIN y

proposed by Khan et al. 1
z w
(adapted from [8] # 1995 x
IZA
Taylor & Francis) D1
D2
R1
D4
D3
IZB
V0
x z
2 w
y R2
provide a wide dynamic input voltage range over a wide frequency range of
operation. An inspection of the circuit reveals that the current flowing into the
X-terminal of CFOA2 and the one flowing out of Z-terminal CFOA1 are given by:
iz ¼ ix ¼ Vin/(R1 + 2rx).
The diode combinations D1–D2 and D3–D4 are connected in such a manner that
depending upon the polarity of the input voltage Vin the output current of the CFOA
will flow either into the load resistance R2 or will be bypassed to the ground. Thus,
when Vin is positive, ix, iZA and iZB have the directions such that diodes D1–D2
allow iZA to be flowing into the load resistance R2 whereas at the same time, diodes
D3–D4 allow iZB to go the ground. On the other hand, when Vin is negative, ix, iZA
and iZB reverse their directions and as a consequence, now iZB flows into the load
while iZA flows to the ground. In view of this, it is clear that current through the load
R2 will be uni-directional thereby resulting in an output voltage given by
V0 ¼ iZA R2 ¼ iZB R2 ¼ ðR2= ðR1 þ 2rx ÞÞVin (6.17)
It is worth pointing out that by reversing the connections of all the four diodes,
one can obtain a full wave rectified signal with negative sign (i.e. V0). Lastly, it
must be mentioned that as compared to VOA-based precision rectifiers, which
generally require four or more matched resistors, the circuit described here uses a
bare minimum of (only two) resistors and no resistor-matching is needed.
6.7 Analog Squaring Circuit
Figure 6.14 shows a CFOA-version of the CCII-based squaring circuit presented in

[9]. This circuit can be regarded as the simplest squaring circuit using only a single
CFOA, a pair of identical (matched) MOSFETs and a single resistor.
6.8 Analog Divider 191
Fig. 6.14 A voltage squaring VG

circuit proposed by Liu
(adapted from [9] # 1995 M1
Vin
IEE) X
-Vin
M2 Z W +
Y
VG
Vo
R0
-
Assuming MOSFETs M1 and M2 to be operating in saturation mode an analysis

of this circuit shows that the current entering into the terminal-X is given by
IX ¼ KVin2 (6.18)
where K is the transconductance parameter of the MOSFETs.

Since Iy ¼ 0, Vx ¼ Vy, IZ ¼ Ix and Vw ¼ Vz, for an ideal CFOA, the output
voltage of the squaring circuit can be given as
V0 ¼ KR0 Vin
2
(6.19)

The transconductance parameter K of the MOSFETs is given by K ¼ μs C2OX WL ,

where the symbols have their usual meanings.
Thus, the circuit of Fig. 6.14 gives an output voltage V0 which is proportional to
the square of the input voltage Vin.
6.8 Analog Divider
Among various non-linear applications of CFOAs evolved so far, an interesting

application is that of realizing an analog divider using CFOAs and MOSFETs. One
such circuit having all MOSFETs operating in triode region is shown in Fig. 6.15
and was proposed by Liu and Chen [10].
Assuming the input signals Vx and Vy to be small and assuming all MOSFETs to
be identical and (i.e. same K) operating in triode region a straight forward analysis
of the circuit of Fig. 6.15 gives the output voltage as

vy
v0 ¼ ðVGA VGB Þ (6.20)
vx
VGA
VGA M3
X X
M1 W Vo
ZW
Y Y Z
vy
VGB
M4 (VGA+vx)
M2
Fig. 6.15 An analog divider using CFOAs proposed by Liu and Chen (adapted from [10] # 1995 IET)
from where it is seen that the circuit functions as an analog divider with input
signals as vx and vy where the scale factor (VGA VGB) is controllable through
external voltages VGA and VGB.
6.9 Pseudo-exponential Circuits
Pseudo-exponential functions play an important role in allowing wide gain control

changes with a control parameter in a number of communication and signal
processing systems [11]. Although, the exponential v–i characteristics of a BJT or
a MOSFET operating in week inversion mode can be exploited to design exponen-
tial circuits, an alternative way is to employ a Pseudo-exponential function given by
1 þ βx
e2βx (6.21)
1 βx
Two circuits to realize the above function are shown in Fig. 6.16.
Consider first the circuit of Fig. 6.16a. In this circuit, the voltage at node-Y with
identical resistors Ra ¼ Rb is given by
Vin þ Vo
Vy ¼ (6.22)
2
Because of the voltage buffer between terminals Y and X of the CFOA, the same
voltage is transferred to the X-terminals of both the CFOAs. By a straight forward
analysis, it can be proved that

1 þ R2 R11 R13
Vo ¼ Vin (6.23)
1 R2 R11 R13
6.10 Chaotic Oscillators Using CFOAs 193
a
Rb
Vin Y
Ra W
Z
X
R3 Y
Z
W Vo
X
R1 R2
b
Rb
Vin Y
Ra W
Z R2
X
R1 X
W Vo
Y Z
R3
Fig. 6.16 CFOA-based realizations of Pseudo exponential function proposed by Maundy

and Gift (adapted from [11] # 2005 IEEE)
which simplifies to

1þx 1 1
Vo ¼ Vin where; x ¼ R2 (6.24)
1x R1 R3
A similar analysis of the second circuit of Fig. 6.16b shows that this circuit also
realizes the same function as in (6.24). The condition required for both the circuits
to realize Pseudo exponential circuit is given by x < 0, if R1 > R3 or alternatively,
x > 0 if R1 < R3.
6.10 Chaotic Oscillators Using CFOAs
Chua’s circuit [12] has been a very active topic of research in the study of non-
linear dynamical circuits and systems during the past two decades. There has been
considerable interest in devising inductor-less realizations of Chua’s oscillator.
A variety of circuit configurations have been evolved employing traditional
VOAs, current conveyors and CFOAs as building blocks. The advantage of using
CFOAs is the ease with which current state variables can be readily made available
a
iL R iR
+ + +
L NR
C2 VC2 VC1 C1 VR
− − − Chua
diode
b IR
m0
m1
BP VR
−BP
m1 m0
Fig. 6.17 (a) Chua’s oscillator L ¼ 28.53 mH, C1 ¼ 5nF, C2 ¼ 50 nF, R ¼ variable, (b) V–I
characteristics of Chua diode, mo ¼ 0.5, m1 ¼ 0.8, BP ¼ 1
X
R1 Z
W
Y
Y
W
Z
X C
IL
R2
X
Z
Zin = sC0R1R2 W
Y
Fig. 6.18 The simulated inductor with inductor current as output (adapted from [13] # 1998 IEE)
as output. This was demonstrated by Senani and Gupta in [13] and by Elwakil and
Kennedy in [14] in 1998 and 2000 respectively.
It is well known that Chua’s oscillator requires a two segments piece-wise
nonlinear negative resistor called Chua’s diode (see Fig. 6.17). An interesting
circuit for the simulation of the grounded inductor is shown in Fig. 6.18 which
Fig. 6.19 The CFOA-based

Chua diode (adapted from
[13] # 1998 IEE) RN1 RN2
YZ YZ
W W
X X
RN3 RN4
Fig. 6.20 Double scroll

attractor obtained from the
entirely CFOA-based
hardware implementation of
the circuit of Fig. 6.17
has a special feature that the current flowing into the simulating inductor is
available explicitly from the Z-terminal of one of the CFOAs.
The CFOA-based Chua diode is shown in Fig. 6.19 whereas the double scroll
chaotic attractor obtained from the complete implementation obtained by the
placement of the sub-circuits of Figs. 6.19 and 6.18 into the main circuit of
Fig. 6.17 is shown in Fig. 6.20.
The advantage of this implementation is that other than the availability of the
two capacitor voltages as state variables, the third state variable namely, the inductor
current iL, is also available explicitly from the Z-output terminal of one of the CFOAs.
An alternative Chua’s oscillator implementation proposed by Elwakil and
Kennedy [14] using CFOAs is shown in Fig. 6.21 which also provides a current-
mode output.
Other than autonomous chaotic oscillators, several researchers have also
investigated the implementation and application of non-autonomous (derived)
chaotic circuits and mixed-mode chaotic circuits (containing both autonomous
and non-autonomous circuit realizations). A popular mixed-mode chaotic circuit
is shown in Fig. 6.22 [15] in which through a digitally controlled switch, it becomes
L Y
C2 C1
W
Z V0
X
R1
R4
Y
W
Z
X I
R2
R3
Fig. 6.21 CFOA-based Chua’s Oscillator proposed by Elwakil and Kennedy (adapted from [14]
# 2000 IEEE)
Q S1
iR
Q S2 +
RS1 R2 C1 NR
VR
−
L1 RS2
+
R1 VC2
C2
−
L2
Vac
Fig. 6.22 Mixed-mode chaotic circuit proposed by Cam (adapted from [15] # 2004 Elsevier)
possible to realize an autonomous as well as non-autonomous chaotic circuit from

the same general structure. When switch S1 is closed, the circuit becomes a derived
chaotic circuit; on the other hand, when S2, closed, the circuit assumes the form of
an autonomous Chua’s oscillator.
Q S1
iR
Q S2 + +
RS R2 NR
VC1 VR
− C1 −
L1 R3
+
VC3 C3
+ −
R1 R6 Y
C2 VC2
W
− Z
X
Vac R4
R5
Fig. 6.23 Mixed-mode chaotic circuit with Wien bridge configuration proposed by Kilic (adapted
S1
Y
Z RN1
W Y
CF R2 S2 Y
X RF2 X Z Z
Z C1 W W
W R3 X X
Y RF1 RN2
C3
Y RN4
X RN3
Z C2 R6 W
W
R1 X
Y
Vac
R4 R5
Fig. 6.24 Inductorless Wien bridge-based mixed-mode chaotic circuit configuration proposed by
Kilic (adapted from [16] # 2007 Elsevier)
There have been several development of chaotic oscillators based upon classic
Wien bridge oscillator. A mixed-mode chaotic circuit made from the Wien bridge
configuration is shown here in Fig. 6.23 and its complete CFOA implementation
using CFOA-based Wien Bridge oscillator, CFOA-based Chua’s diode and CFOA-
based lossless floating inductance, is shown in Fig. 6.24 [16].
It may be pointed out that a three-CFOA-based floating inductance simulator
was proposed earlier by Senani in [17].
This chapter has presented miscellaneous linear/nonlinear applications of CFOA

which demonstrate that CFOAs are useful building blocks in realizing a variety of
non-linear functions and circuits as well. In most of the cases, the type of circuits
possible with CFOAs cannot be made with conventional op-amps with the same
advantages. In the area of chaotic circuits, there are a lot of other CFOA-based
realizations which have been omitted from the discussion to conserve space; the
interested reader is referred to [18–39] which indicate that CFOAs have been used
quite prominently in realizing various chaotic circuits which confirm their utility in
nonlinear chaotic circuit design. Lastly, it may be pointed out that, to the best
knowledge of the authors, any CFOA-MOSFET based 4-quadrant multipliers and
Square-rooting circuits do not appear to have been attempted so far and thus,
constitute interesting problems for further research.
References
1. Data sheet of 60 MHz 2000V/μS Monolithic Op Amp AD844. Analog Devices, Inc.
1983–2011
2. Di Cataldo G, Palumbo G, Pennisi S (1995) A Schmitt trigger by means of a CCII+. Int J Circ
Theor Appl 23:61–165
3. Abuelma’atti MT, Al-absi MA (2005) A current conveyor-based relaxation oscillator as a
versatile electronic interface for capacitive and resistive sensors. Int J Electron 92:473–477
4. Srinivasulu A (2011) A novel current conveyor-based Schmitt trigger and its application as a
relaxation oscillator. Int J Circ Theor Appl 39:679–686
5. Haque AKMS, Hossain MM, Davis WA, Russel Jr HT, Carter RL (2008) Design of sinusoidal,
triangular, and square wave generator using current feedback operational amplifier (CFOA).
Region 5 Conference IEEE. pp 1–5
6. Minaei S, Yuce E (2012) A simple Schmitt trigger circuit with grounded passive elements and
its application to square/triangular wave generator. Circ Syst Sign Process 31:877–888
7. Abuelma’atti MT, Al-Shahrani SM (1998) New CFOA-based triangular/square wave genera-
tor. Int J Electron 84:583–588
8. Khan AA, El-Ela MA, Al-Turaigi (1995) Current-mode precision rectification. Int J Electron
79:853–859
9. Liu SI (1995) Square-rooting and vector summation circuits using current conveyors. IEE Proc
10. Liu SI, Chen JJ (1995) Realization of analogue divider using current feedback amplifiers. IEE
Proc Circ Devices Syst 142:45–48
11. Maundy B, Gift S (2005) Novel pseudo-exponential circuits. IEEE Trans Circuits Syst-II
52:675–679
12. Chua LO (1992) The genesis of Chua’s circuit. Archiv Elektronik Uebertragungstechnik
46:250–257
13. Senani R, Gupta SS (1998) Implementation of Chua’s chaotic circuit using current feedback
op-amps. Electron Lett 34:829–830
14. Elwakil AS, Kennedy MP (2000) Improved implementation of Chua’s chaotic oscillator using
current feedback Op Amp. IEEE Trans Circ Syst-I 47:76–79
References 199
15. Cam U (2004) A new high performance realization of mixed-mode chaotic circuit using
current-feedback operational amplifiers. Comput Electr Eng 30:281–290
16. Kilic R (2007) Mixed-mode chaotic circuit with Wien-bridge configuration: the results of
experimental verification. Chaos Solitons Fractals 32:1188–1193
18. Elwakil AS, Kennedy MP (1999) A family of Colpitts-like chaotic oscillators. J Franklin Inst
336:687–700
19. Elwakil AS, Kennedy MP (1999) Chaotic oscillators derived from Saito’s double-screw
hysteresis oscillator. IEICE Trans Fundament E82-A:1769–1775
20. Mahmoud SA, Elwakil AS, Soliman AM (1999) CMOS current feedback op amp-based chaos
generators using novel active nonlinear voltage controlled resistors with odd symmetrical
characteristics. Int J Electron 86:1441–1451
21. Elwakil AS, Kennedy MP (1999) Inductor less hyper chaos generator. Microelectron
J 30:739–743
22. Elwakil AS, Kennedy MP (2000) Chua’s circuit decomposition: a systematic design approach
for chaotic oscillators. J Franklin Inst 337:251–265
23. Elwakil AS, Kennedy MP (2000) Novel chaotic oscillator configuration using a diode-inductor
composite. Int J Electron 87:397–406
24. Elwakil AS, Kennedy MP (2001) Construction of classes of circuit-independent chaotic
oscillator using passive-only nonlinear devices. IEEE Trans Circ Syst-I 48:289–307
25. Elwakil AS, Ozoguz S, Kennedy MP (2002) Creation of a complex butterfly attractor using a
novel Lorenz-type system. IEEE Trans Circ Syst-I 49:527–530
26. Bernat P, Balaz I (2002) RC autonomous circuits with chaotic behavior. Radioengineering
11:1–5
27. Ozoguz S, Elwakil AS, Salama KN (2002) n-scroll chaos generator using nonlinear
transconductor. Electron Lett 38:685–686
28. Ozoguz S, Elwakil AS, Kennedy MP (2002) Experimental verification of the butterfly attractor
in a modified Lorenz system. Int J Bifurcation Chaos 12:1627–1632
29. Elwakil AS (2002) Non-autonomous pulse-driven chaotic oscillator based on Chua’s circuit.
Microelectron J 33:479–486
30. Kilic R (2003) On current feedback operational amplifier-based realizations of Chua’s circuit.
Circ Syst Sign Process 22:475–491
31. Yalcin ME, Suykens JAK, Vandewalle J (2004) True random bit generation from a double-
scroll attractor. IEEE Trans Circ Syst-I 51:1395–1404
32. Elwakil AS (2004) Integrator-based circuit-independent chaotic oscillator structure. CHAOS
14:364–369
33. Ozoguz S, Elwakil AS (2004) On the realization of circuit-independent non-autonomous
pulse-excited chaotic oscillator circuits. IEEE Trans Circuits Syst-II 51:552–556
34. Cam U, Kilic R (2005) Inductor less realization of non-autonomous MLC chaotic circuit using
current-feedback operational amplifiers. J Circ Syst Comput 14:99–107
35. Cuautle ET, Hernandez AG, Delgado JG (2006) Implementation of a chaotic oscillator by
designing Chua’s diode with CMOS CFOAs. Analog Integr Circ Sign Process 48:159–162
36. Kilic R, Karauz B (2007) Implementation of a laboratory tool for studying mixed-mode
chaotic circuit. Int J Bifurcation Chaos 17:3633–3638
37. Srisuchinwong B, Liou CH (2007) Improved implementation of Sprott’s chaotic oscillators
based on current-feedback op-amps. ECTI-CON:38–44
38. Petrzela J, Sotner R, Slezak J (2009) Electronically adjustable mixed-mode implementations
of the jerk functions. Contemp Eng Sci 2:441–449
39. Elwakil AS, Ozoguz S (2006) On the generation of higher order chaotic oscillators via passive
coupling of two identical or non-identical sinusoidal oscillators. IEEE Trans Circ Syst-I
53:1521–1532
Chapter 7
Realization of Other Building Blocks
Using CFOAs
7.1 Introduction
Apart from the applications of the CFOAs as (pin-to-pin) replacements of voltage

mode op-amps in which case they have been known to exhibit potential advantages
as compared to the original VOA-based circuits, the CFOAs, as 4-terminal building
blocks too, have yielded very interesting circuits, the type of which cannot be
realized with conventional VOAs. This latter aspect has been amply demonstrated
in the earlier chapters of this monograph.
In this chapter, we demonstrate how CFOAs have found potential applications in
realizing a variety of other active building blocks too thereby quite often resulting
in interesting topologies offering significant advantages.
7.2 Applications of the CFOAs in Realizing

Other Building Blocks
Apart from being employed as four terminal building blocks in their own right,
CFOAs have been employed to realize many other building blocks in the analog
circuits literature, such as CCII+/, unity gain voltage followers (VF) and unity gain
current followers (CF), four terminal floating nullors (FTFN), Current differencing
buffered amplifiers (CDBA), operational transresistance amplifiers (OTRA),
Current differencing transconductance amplifiers (CDTA), third generation Current
conveyors (CCIII), differential voltage second generation Current Conveyors
(DVCC+), Current follower transconductance amplifiers (CFTA), current con-
trolled current conveyor transconductance amplifier (CCCC-TA), differential-
input buffered transconductance amplifier (DBTA), voltage differencing differential
202 7 Realization of Other Building Blocks Using CFOAs
Fig. 7.1 Realization of CCII y +

+ and CCII using CFOAs CFOA
x −
y + −
CFOA
CFOA
x − +
z
z
CCII+ CCII-
iy = 0 iy = 0
vx = vy vx = vy
iz = ix iz = –ix
Fig. 7.2 Realization of Y1 y

a DVCC+ using CFOAs Iy1 zw
(based upon the idea given x
in [1]) y
w
R1 X x z
R2 Ix
z+
Iz+
x
w
Y2 y z
Iy2
iy1 = 0
iy2 = 0
vx = (vy1 – vy2) for R1 = R2
iz = ix
input buffered amplifier (VD-DIBA) etc.; see [1–146]. A few such equivalencies
have been demonstrated earlier in [1, 2]. Also, a theorem facilitating replacement of
CCIIs by CFOAs has been presented in [3].
7.2.1 CFOA Realizations of Various Kinds of Current

Conveyors (CC)
A CCII+ is realizable with only a single CFOA while CCII requires two of them
as shown in Fig. 7.1.
iy ¼ 0 iy ¼ 0
vx ¼ vy vx ¼ vy (7.1)
iz ¼ ix iz ¼ ix
7.2 Applications of the CFOAs in Realizing Other Building Blocks 203
Fig. 7.3 Realization of a x x

CCIII+ using CFOAs (based Ix w
upon the idea given in [1]) y y z
Iy
y
w
x z
R
R z+
Iz+
x
w z
y
iy = −ix
vx = vy
iz = ix
An interesting variant of CCII, known as differential voltage CC (DVCC), can

be realized with three CFOAs but needs two resistors as well (see Fig. 7.2).
iy1 ¼ 0
iy2 ¼ 0
vx ¼ ðvy1 vy2 Þ for R1 ¼ R2
iz ¼ ix (7.2)
Furthermore, yet another variant of CCs, known as a third generation Current

Conveyor (CCIII) can also be realized with three CFOAs and two resistors as
shown in Fig. 7.3.
iy ¼ ix
vx ¼ vy
iz ¼ ix (7.3)
It is worth mentioning that CCII+, CCII, DVCC or CCIII-based voltage-mode

circuits would invariably require a voltage follower after the Z-terminal(s) of those
CCs from which a voltage output is being taken since the Z-terminal being a current
output terminal cannot be connected to the load impedance directly as this will
modify and change the function realized by the circuit. Realizing the CC-based
circuits by CFOAs will easily permit the Z-terminal voltage(s) to be available from
the W-terminal(s) quite easily without requiring any additional external buffer
because of the availability of an internal VF between the Z and W terminals in
the CFOA(s) and thereby providing a remedy to this problem.
The CFOA-based CC implementations have been employed by many
researchers for the verification of their CC-based circuit proposals, for example
see [5–77] and the references cited therein.
Fig. 7.4 Various steps in the a b CCII-

implementation of FTFN
using CFOAs (a) FTFN, (b) y z
X x W
A FTFN using two three- X W
terminal nullors, (c) FTFN
using two CCIIӨ, (d) FTFN Y x
y z Z
using two CFOAs
Y Z CCII-
c d
X y
CCII- z W
x X y
zw
x
W
x
CCII- z Z
Z xz
Y y w
Y y
7.2.2 CFOA-Realization of the Four-Terminal-

Floating-Nullors (FTFN)
It was shown by a number of researchers (such as Nordholt [78], Stevenson [79],

Huijsing [80] and Senani [81] (who coined the term ‘four-terminal floating-nullor’
(FTFN) to represent a fully floating nullor)) that fully floating versions of Op-amps
(termed as operational floating amplifier (OFA)) and FTFNs are more versatile and
flexible building blocks than the traditional op-amps in several applications.
It was suggested in [81] that a composite connection of two CCII can be used
to realize an FTFN (see Fig. 7.4). This follows from the fact that the representation
of FTFN of Fig. 7.4a is equivalent to the combination of two 3-terminal nullors as
shown in Fig. 7.4b where each 3-terminal nullor is equivalent to a CCII, thus,
finally, leading to the implementation of Fig. 7.4c. In fact, two CCII+ or two current
feedback op-amps (CFOA) such as AD844, can be readily used to realize an FTFN
using the same configuration (as in Fig. 7.4d).
A novel application of FTFNs has been in the area of floating impedance
simulation; for instance, see [81, 91, 93].
In this context it may be noted that there are a number of realizations of floating
impedances (FI) which call for two FTFNs and hence, four CFOAs, for instance,
see [93]. In Fig. 7.5 we show a novel FI circuit1 which is realizable with only a
single FTFN and hence, only by two CFOAs. A routine circuit analysis shows that
the floating inductance simulated by both the circuits is given by
1
R. Senani (1987) Generation of new two-amplifier synthetic floating inductors. Electron Lett 23
(22):1202–1203
R6
a R6 b
C0 R5
C0 R5 R3 y
z w
R3 R4
x
R4 i1
FTFN x w i2
i2 + z +
i1 y
R1 V2
+ + V1
R1 R2
V1 V2 R2
− − − −
Fig. 7.5 FTFN based simulated inductor and its implementation (a) FTFN based lossless floating
simulated inductance, (b) CFOA equivalent of a FTFN based floating simulated lossless inductance

C0 R3 R6 ð1 þ RR21 Þ R2 R 1 þ R6 R1 R6
Leq ¼ ; provided that ¼ þ 1þ (7.4)
ð1 þ RR54 Þ R4 R5 R3 R5
From the discussions contained in Chap. 3, it may be seen that although a

lossless FI was shown to be realizable with only two CFOAs as in the circuit of
Fig. 3.15 but the circuit therein needs two matched capacitors. By contrast, the
circuit of Fig. 7.5b has the novelty of employing only a single capacitor.
The CFOA-based FTFN implementation of Fig. 7.4d has been employed by
many researchers for the verification of their FTFN-based propositions, for exam-
ple, see [82–95] and the references cited therein.
7.2.3 CFOA Realization of Operational Trans-resistance

Amplifier (OTRA)
Operational trans-resistance amplifier (OTRA) [119, 120] is characterized by the

terminal equations
vp ¼ 0
vn ¼ 0
v0 ¼ Rm ðip in Þ; Rm ! 1 (7.5)
OTRAs have been employed as alternative building blocks to realize a number

of functions such as all pass filters, inductance simulators, MOS-C biquads, sinu-
soidal oscillators and multivibrators. Although in several publications, CMOS
OTRA architectures have been employed, in many others, the two-CFOA-based
implementation of the OTRA as shown in Fig. 7.6 has been employed.
Ip
p x
y z w
In
n x
V0
y z w
Rm
(5)
Fig. 7.6 Operational trans-resistance amplifier (OTRA) (adapted from [121] # 2004 Springer)
R3
D3
R1
I y
z w
x y
It + vc2 −
z w V0
x
C2
Vtp
Vtn D1 R2
D2
Fig. 7.7 The CFOA-version of the OTRA-based mono-stable multi-vibrator (adapted from [123]
# 2006 IEEE)
Out of the various applications of OTRAs reported till date, a particularly

interesting application was proposed by Lo and Chien [123] who realized a
mono-stable multivibrator featuring both positive and negative triggering modes.
A CFOA-version of this circuit is shown in Fig. 7.7, where the dotted box contains a
two-CFOA realization of the OTRA.
The time period T during which the circuit remains in quasi-stable state and the
recovery time Tr are given by
h i
T ¼ R2 C2 ln ð1 þ K Þ
R1
R2
(7.6)
2 =R1
Tr ¼ R3 C2 ln 2R
1K ; where K ¼ jVVD2
sat j
The novelty of the structure of Fig. 7.7 is that no such circuit has been explicitly
proposed in the existing literature using CFOAs.
Fig. 7.8 Current differencing p −

buffered amplifier (CDBA) CFOA w
(adapted from [96] # 1999 +
Elsevier) vp = 0
+ vn = 0
CFOA iz = in – ip
n −
z vw = vz
CDBA
The CFOA-based OTRA implementation has been employed by many

researchers for the verification of their propositions for example, see [119–131]
and the references citied therein.
7.2.4 CFOA Realization of Current Differencing Buffered

Amplifier (CDBA) Based Circuits
Acar and Ozoguz in 1999 introduced current differencing buffered amplifier

(CDBA) as a new versatile building blocks for analog signal processing [96].
Since then, CDBAs have been employed to realize a variety of linear and non-
linear functions. Although fully integratable circuit implementations of CDBAs
have been proposed by a number of researchers, CFOAs have been found to be
quite handy in realizing them. Since CDBA is characterized by the terminal
equations Vp ¼ Vn ¼ 0, iz ¼ (ip in), and Vw ¼ Vz it was found that it could
be readily implemented by two CFOAs as follows (Fig. 7.8).
vp ¼ 0
vn ¼ 0
iz ¼ in iP
vw ¼ vz (7.7)
As an exemplary application of this equivalence we show in the following

CDBA-based analog multiplier circuit and its CFOA-based implementation.
Assuming MOSFETs M1 and M2 to be matched and operating in triode region,
the analysis of this circuit shows that the output voltage is given by
K12
V0 ¼ : Vx Vy ; jVZ j ¼ ðVþ VT Þ (7.8)
K34 ðVþ VT Þ
where K is the transconductance parameter of MOSFETs M1 and M2 and is given by

W
K12 ¼ μs Cox (7.9)
L 1
a
VB+vx +V
M1
p w V0
Vy M3
n z
M2
VB-vx M4
−V
b
VB+vx
ip y
M1 +V
z w Vw
x y
Vy Vp w V0
z
x
M2 Vn iz M3
in Vz
VB-vx
CDBA M4
−V
Fig. 7.9 CDBA and its implementation (a) A CDBA-based analog multiplier (adapted from [101]
# 2004 Springer). (b) CFOA-based implementation
and K34 is the transconductance parameter of the matched pair of MOSFETs M3 and
M4 (both operating in saturation) and VT is the threshold voltage of the MOSFETs.
It is worth pointing out that although two-CFOA based analog divider using four
MOSFETs has been known in literature [4], any circuit for realizing a 4-quadrant
analog multiplier using an exactly the same number of CFOAs has not been known.
This application demonstrates how by realizing a CDBA through its two-CFOA-
based implementation such a circuit (as in Fig. 7.9b) becomes possible. It must be
emphasized that any such circuit using conventional VOAs and only four
MOSFETs is not known to exist.
The CFOA-based CDBA implementation has been employed by many
researchers for the verification of their propositions, for example, see [96–110]
and the references cited therein.
7.2.5 CFOA Realization of Circuits Containing Unity Gain Cells
A large number of active circuit building blocks of varying complexity have been
introduced by various researchers quite often having three or more external
terminals. Interest in unity gain voltage follower (VF) and unity gain current
Io
- vin = 0
Iin - CFOA
CFOA
CFOA
+
+ i0 = −iin
-
vi
vo CF-
ii
Voltage Follower io
iin -
ii = 0 CFOA vin = 0
v0 = vi + CF+ i0 = iin
Fig. 7.10 Realization of voltage and current followers using CFOA(s)
w Iout
a b z
R1 x y
VF
R1
CF w x
V0 z y R2 Load
C2
R2 w x
Vout zy
R3 C1 C2
R3 C1
Fig. 7.11 Follower based SRCO (a) A VF-CF based SRCO, (b) CFOA implementation
follower (CF) is primarily attributed to the relatively larger bandwidth offered by

them as well as the theoretical novelty that from several building blocks, VFs and
CFs can be realized without requiring any external resistors. For instance, a unity
gain VF is known to be realizable by a single VOA with its inverting terminal
shorted to the output terminal. Likewise, a non-inverting VF and non-inverting/
inverting CFs are also realizable with CFOAs without requiring any external
resistances (see Fig. 7.10). In fact, these CFOA-based implementations of non-
inverting VF and non-inverting/inverting CF have already been used by a number
of researchers to prove the workability of their VFs/CFs based analog signal
processing/signal generating circuits.
As an exemplary application, we demonstrate a VF-based sinusoidal oscillator
and its CFOA implementation here in Fig. 7.11.
The condition of oscillation (CO) and the frequency of oscillation (FO) for both
the circuits are given by
C1 R3
CO : R1 ¼ (7.10)
C1 þ C2
a b
y
ip y w zw
x ix+
Vp xz y
p x+ w
Vp
Ix+ Vn xz R
Ip
in iz
CDTA x
zw
Vz y
In Ix-
Vn n z x- 1
R= ix-
Gm
Iz
Vz
Fig. 7.12 CDTA and its implementation (a) Symbolic notation, (b) CFOA implementation
and FO:
1
2π C1 C2 R2 R3
The novelty of the CFOA-based SRCO of Fig. 7.11 is that from the same circuit,
both VM and CM outputs are available explicitly.
From the above example it, therefore, can be concluded that all the VFs/CFs
based signal processing/generation circuits known so far can be practically realized
in hardware by using AD844-type CFOAs.
The CFOA-based VF/CF implementations of Fig. 7.10 have been employed by
many researchers for the verification of their propositions for example see
[111–118] and the references cited therein
7.2.6 Current Differencing Transconductance Amplifier (CDTA)
The current differencing transconductance amplifier was introduced by Biolek in

[135] as a new building block suitable for current-mode analog signal processing.
The symbolic notation of the CDTA is shown in Fig. 7.12.
A CDTA is characterized by the following matrix equation.
0 1 0 10 1
Iz 0 0 0 1 1 Vz
B Ixþ C B gm 0 C B C
B C B 0 0 0 CB Vxþ C
B Ix C ¼ B gm 0 0 0 0 CB Vx C
C B (7.12)
B C B C
@ Vp A @ 0 0 0 0 0 A@ Ip A
Vn 0 0 0 0 0 In
where gm is the transconductance of the CDTA. An entirely CFOA-based imple-

mentation of the CDTA, based upon the idea given in [135], is shown in Fig. 7.12.
a b
If f x+ y
w
Ix+ x
f
CFTA + x+
gm
-
x-
Ix-
z x-
Iz z
Fig. 7.13 CFTA and its implementation (a) Symbolic notation, (b) CFOA implementation of the
CFTA
CDTAs have received considerable attention in realizing various types of filters,

oscillators, impedance simulators and other applications. For those cases where
CDTAs have been implemented with CFOAs, the reader is referred to [136–140]
and the references cited therein for further details.
7.2.7 Current Follower Transconductance Amplifiers (CFTA)
The current follower transconductance amplifier was introduced by Herencsar et al.

in [141]. This current input, current output building block has been shown to be
particularly useful in realizing analog signal processing functions requiring explicit
current outputs. The symbolic notation of CFTA is shown in Fig. 7.13a and its
characterizing matrix equation is given by (7.13).
0 1 0 10 1
Iz 0 0 0 1 Vz
B Ixþ C B gm 0 0 0 CB Vxþ C
B C B CB C
@ Ix A ¼ @ gm 0 0 0 A@ Vx A (7.13)
Vf 0 0 0 0 If
A discrete version of CFTA can be implemented using one AD844 type CFOA
and one balanced output transconductance amplifier such as MAX 435 and is
shown in Fig. 7.13b.
This CFOA based implementation of the CFTA has been employed in the Reali-
zation of current-mode KHN-equivalent biquad using CFTAs presented in [142].
7.2.8 Current Controlled Current Conveyor Transconductance

Amplifier (CCCC-TA)
The current controlled current conveyor transconductance amplifier (CCCC-TA)

was introduced as a building block for analog signal processing by Siripruchyanun
and Jaikla in [143]. A CFOA-based implementation of this was devised by
Maheshwari et al. in [144] and is shown here in Fig. 7.14.
y
z w
x
Rx y
w + +O
R gm
x -O
+z -
Y X
R
x
w
y -z
Fig. 7.14 A CFOA-based implementation of CCCC-TA (adapted from [144] # 2011 IET)
x
w ix
ip RE z
y
Vx
Vp
in iw
Vn x z x
w z w Vw
y y
iy iz
Vy Vz
Fig. 7.15 A CFOA implementation of DBTA (adapted from [133] # 2009 IEICE)
7.2.9 Differential Input Buffered Transconductance

Amplifier (DBTA)
The differential input buffered transconductance amplifier (DBTA) was introduced by

Herencsar et al. in [132]. DBTA has been found to be an useful building block in
realizing sinusoidal oscillators, quadrature oscillators and universal filters. Quite often,
all these functions can be carried out effectively using only a single DBTA as in [133,
134]. A DBTA is a six port building block characterizing by the following equation:
vp ¼ vy ; vn ¼ vy ; iy ¼ 0; iz ¼ ðip in Þ; vw ¼ vz ; ix ¼ gm vz
(7.14)
where gm ¼ 1=RE
A CFOA implementation of this building block is shown in Fig. 7.15 and has
been used in [133] to verify their proposed quadrature oscillator.
a VD-DIBA b - +
gm
I+
V+ +
w Vw V+ +
V- gm +
- z v Iw V- - gm y
I- - z
x
Iz
Iv Iz Iv
Iw 2.2k
Vz Vv
Vz Vv Vw
Fig. 7.16 CFOA implementation of VD-DIBA (a) Schematic symbol, (b) CFOA implementation
(adapted from [146] # 2011 Hindawi Publishing Corporation)
7.2.10 Voltage Differencing Differential Input Buffered

Amplifier (VD-DIBA)
The voltage differencing differential input buffered amplifier (VD-DIBA) was

introduced by Biolek et al. in [145]. Although some applications of VD-DIBAs
have been reported in the open literature but for the very first time an application
implementing VD-DIBA using OTAs and CFOA was presented in [146].
The schematic symbol of the VD-DIBA is shown in Fig. 7.16a.
A VD-DIBA is characterized by the following matrix equation.
0 1 0 10 1
Iþ 0 0 0 0 0 Vþ
B I C B 0 0 0 0C B C
B C B 0 C B V C
B Iz C ¼ B gm gm 0 0 0 CB Vz C (7.15)
B C B CB C
@ Iv A @ 0 0 0 0 0 A@ V v A
Vw 0 0 1 1 0 Iw
The CFOA-implementation of the VD-DIBA as proposed in [146] is shown in

Fig. 7.16b
In this chapter, we have demonstrated that CFOAs (sometimes only CFOAs and some-
times in conjunction with additional resistors and/or OTA(s)) have potential applications
in realizing a variety of other active building blocks such as CCII+/ CCIII, unity gain
VF and CF, FTFN, CDBA, OTRA, CDTA, DVCC+, CFTA, CCCC-TA, DBTA and
VD-DIBA-all proposed and being employed in recent technical literature.
It was also demonstrated that when CFOAs are used to realize the circuits
employing the considered building blocks, in several cases, quite versatile and
efficient functional circuits result. These applications, therefore, further establish
the flexibility and versatility of CFOAs in analog circuit design.
References
I. General References
1. Yuce E, Minaei S (2007) Realization of various active devices using commercially available
AD844s and external resistors. Electron World 113:46–49
2. Senani R, Bhaskar DR, Gupta SS, Singh VK (2012) Current-feedback op-amp, their
applications, Bipolar/CMOS implementation and their variants, Chapter 2. In: Esteban
Tlelo-Cuautle (ed) Integrated circuits in analog signal processing. Springer, New York,
pp 61–84
3. Aronhime P, Wang K, Qian T (2001) Generalization of a theorem for replacing CCIIs by
CFOAs in current-mode circuits. Analog Integr Circ Sign Process 28:27–33
4. Liu SI, Chen JJ (1995) Realisation of analogue divider using current feedback amplifiers. IEE
Proc Circ Devices Syst 142:45–48
II. Use of CFOAs in Realizing Various Types of Current Conveyors
5. Hou CL, Chen RD, Wu YP, Hu PC (1993) Realization of grounded and floating immittance
function simulators using current conveyors. Int J Electron 74:917–923
6. Svoboda JA (1994) Comparison of RC op-amp and RC current conveyors filters. Int J
7. Chang C, Lee MS (1994) Universal voltage-mode filter with three inputs and one output using
three current conveyors and one voltage follower. Electron Lett 30:2112–2113
8. Vosper JV, Heima M (1996) Comparison of single- and dual-element frequency control in a
CCII-based sinusoidal oscillator. Electron Lett 32:2293–2294
9. Horng JW, Tsai CC, Lee MH (1996) Novel universal voltage-mode biquad filter with three
inputs and one outputs using only two current conveyors. Int J Electron 80:543–546
10. Wang HY, Lee CT (1997) Immittance function simulator using a single current conveyor.
11. Liu SI, Lee JL (1997) Voltage-mode universal filters using two current conveyors. Int J
12. Horng JW, Lee MH, Cheng HC, Chang CW (1997) New CCII-based voltage-mode universal
biquadratic filter. Int J Electron 82:151–155
13. Chang CM (1997) Multifunction biquadratic filters using current conveyors. IEEE Trans Circ
Syst II 44:956–958
14. Lee JY, Tsao HW (1992) True RC integrators based on current conveyors with tunable time
constants using active control and modified loop technique. IEEE Trans Instrum Meas 41
(5):709–714
15. Liu S-I, Kuo J-H, Tsay J-H (1992) New CCII-based current-mode biquadratic filters. Int J
16. Wilson B (1992) Trends in current conveyor and current-mode amplifier design. Int J
17. Brunn E, Olesen OH (1992) Conveyor implementations of generic current mode circuits. Int J
18. Svoboda JA (1994) Transfer function synthesis using current conveyors. Int J Electron
76:611–614
19. Fabre A, Dayoub F, Duruisseau L, Kamoun M (1994) High input impedance insensitive
second-order filters implemented from current conveyors. IEEE Trans Circ Syst I
41:918–921
References 215
20. Martinez PA, Celma S, Gutiérrez I (1995) Wien-type oscillators using CCII+. Analog Integr
21. Hwang YS, Liu SI, Wu DS, Wu YP (1995) Linear transformation all-pole filters based on
current conveyors. Int J Electron 79(4):439–445
22. Soliman AM (1996) New inverting-non-inverting bandpass and lowpass biquad circuit using
current conveyors. Int J Electron 81:577–583
23. Cajka J, Lindovsky D (1997) Universal RC-Active network using CCII+. J Electr Eng
48:98–100
24. Soliman AM (1997) Generation of current conveyors-based all-pass filters from op-amp-
based circuits. IEEE Trans Circ Syst II 44:324–330
25. Al-Walaie SA, Alturaigi MA (1997) Current mode simulation of lossless floating inductance.
26. Vrba K, Cajka J, Zeman V (1997) New RC-active networks using current conveyors.
Radioengineering 6:18–21
27. Vrba K, Cajka J (1997) High-order one port elements for lowpass filter realization. J Electr
Eng 48:31–34
28. Cajka J, Dostal T, Vrba K (1997) Realization of Nth-order voltage transfer function using
current conveyors CCII. Radioengineering 6:22–25
29. Cicekoglu O (1998) New current conveyor based active-gyrator implementation.
Microelectron J 29:525–528
30. Cicekoglu MO (1998) Active simulation of grounded inductors with CCII + s and grounded
passive elements. Int J Electron 4:455–462
31. Ozoguz S, Acar C (1998) On the realization of floating immittance function simulators using
current conveyors. Int J Electron 85:463–475
32. Soliman AM (1999) Synthesis of grounded capacitor and grounded resistor oscillators.
J Franklin Inst 336:735–746
33. Cicekoglu O, Ozcan S, Kuntman H (1999) Insensitive multifunction filter implemented with
current conveyors and only grounded passive elements. Frequenz 53:158–160
34. Chang CM, Tu SH (1999) Universal voltage-mode filter with four inputs and one output using
two CCII + s. Int J Electron 86:305–309
35. Abuelma’atti MT, Tasadduq NA (1999) New negative immittance function simulators using
current conveyors. Microelectron J 30:911–915
36. Soliman AM, Elwakil AS (1999) Wien oscillators using current conveyors. Comput Electr
Eng 25:45–55
37. Abuelma’atti MT (2000) Comment on: Active simulation of grounded inductors with CCII + s
and grounded passive elements. Int J Electron 87:177–181
38. Cicekoglu O (2000) Reply to comment on: Active simulation of grounded inductors with
CCII + s and grounded passive elements. Int J Electron 87:183–184
39. Abuelma’atti MT, Tasadduq NA (2000) Current-mode lowpass/bandpass and highpass filter
using CCII + s. Frequenz 54:162–164
40. Abuelma’atti MT (2000) New sinusoidal oscillators with fully uncoupled control of oscilla-
tion frequency and condition using three CCII+s. Analog Integr Circ Sign Process
24:253–261
41. Ozoguz S, Acar C, Toker A, Gunes EO (2001) Derivation of low-sensitivity current-mode
CCII-based filters. IEE Proc Circ Devices Syst 148:115–120
42. Horng JW (2001) A sinusoidal oscillator using current-controlled current conveyors. Int J
43. Cicekoglu O, Toker A, Kuntman H (2001) Universal immittance function simulators using
current conveyors. Comput Electr Eng 27:227–238
44. Biolek D, Cajka J, Vrba K, Zeman V (2002) Nth-order allpass filters using current conveyors.
J Electr Eng 53:50–53
45. Hwang YS, Hung PT, Chen W, Liu SI (2002) Systematic generation of current-mode linear
transformation filters based on multiple outputs CCIIs. Analog Integr Circ Sign Process
32:123–134
46. Aksoy M, Ozcan S, Cicekoglu O, Kuntman H (2002) High output impedance current-mode
third-order Butterworth filter topologies employing unity gain voltage buffers and equal-
valued passive components. Int J Electron 90:589–598
47. Shah NA, Malik MA (2005) High impedance voltage- and current-mode multifunction filters.
Int J Electron Commun (AEU) 59:262–266
48. Kumar P, Pal K (2005) Variable Q all-pass, notch and band-pass filters using single CCII.
Frequenz 59:235–239
49. Horng JW (2004) High input impedance voltage-mode universal biquadratic filters with three
inputs using plus-type CCIIs. Int J Electron 91:465–475
50. Gift SJG (2004) New simulated inductor using operational conveyors. Int J Electron
91:477–483
51. Abuelma’atti MT, Bentrcia A, Al-Shahrani SM (2004) A novel mixed-mode current
conveyor-based filter. Int J Electron 91:191–197
52. Horng JW (2004) Voltage-mode universal biquadratic filters using CCIIs. IEICE Trans
Fundam E-87-A:406–409
53. Horng JW, Hou CL, Chang CM, Chung WY, Tang HW, Wen YH (2005) Quadrature
oscillators using CCIIs. Int J Electron 92:21–31
54. Fongsamut C, Fujii N, Surakampontorn W (2005) Two new RC oscillators using CCIIs. Proc
ISCIT 2:1138–1141
55. Khan AA, Bimal S, Dey KK, Roy SS (2005) Novel RC sinusoidal oscillator using second-
generation current conveyors. IEEE Trans Instrum Meas 54:2402–2406
56. Horng JW (2005) Current conveyors based allpass filters and quadrature oscillators
employing grounded capacitors and resistors. Comput Electr Eng 31:81–92
57. Abuelma’atti MT, Shahrani SMA, Al-Absi MK (2005) Simulation of a mutually coupled
circuit using plus-type CCIIs. Int J Electron 92:49–54
58. Keskin AU (2005) Single CFA-based NICs with impedance scaling properties. J Circ Syst
Comput 14:195–203
59. Horng JW, Hou CL, Chang CM, Chung WY, Wei HY (2005) Voltage-mode universal
biquadratic filters with one input and five outputs using MOCCIIs. Comput Electr Eng
31:190–202
60. Horng JW, Hou CL, Chang CM, Chung WY (2006) Voltage-mode universal biquadratic
filters with one input and five outputs. Analog Integr Circ Sign Process 47:73–83
61. Yuce E, Cicekoglu O (2006) The effects of non-idealities and current limitations on the
simulated inductances employing current conveyors. Analog Integr Circ Sign Process
46:103–110
62. Pandey N, Paul SK, Bhattacharyya A, Jain SB (2006) A new mixed mode biquad using
reduced number of active and passive elements. IEICE Electron Express 3:115–121
63. Metin B, Cicekoglu O (2006) A novel floating lossy inductance realization topology with
NICs using current conveyors. IEEE Trans Circ Syst II 53:483–486
64. Maundy B, Gift S, Aronhime P (2007) Realization of a GIC using hybrid current conveyors/
operational amplifier circuits. 50th Midwest Symp Circ Syst (MWSCAS 2007), pp 163–166,
DOI:10.1109/MWSCAS.2007.4488562
65. Maundy B, Gift S, Aronhime P (2007) A novel hybrid active inductor. IEEE Trans Circ Syst
II 54:663–667
66. Kumar P, Pal K, Rana S (2008) High input impedance universal biquadratic filters using
current conveyors. J Active Passive Electron Devices 3:17–27
67. Kumar P, Pal K (2008) Universal biquadratic filter using single current conveyor. J Active
Passive Electron Devices 3:7–16
68. Pandey N, Paul SK, Jain SB (2008) Voltage mode universal filter using two plus type CCIIs.
J Active Passive Electron Devices 3:165–173
References 217
69. Yuce E (2008) Negative impedance converter with reduced nonideal gain and parasitic
impedance effect. IEEE Trans Circ Syst I 55:276–283
70. Yuce E (2008) Grounded inductor simulators with improved low-frequency performances.
IEEE Trans Instrum Meas 57:1079–1084
71. Maundy B, Gift S, Aronhime P (2008) Practical voltage/current-controlled grounded resistor
with dynamic range extension. IET Circ Devices Syst 2:201–206
72. Ferri G, Guerrini N, Silverii E, Tatone A (2008) Vibration damping using CCII-based
inductance simulators. IEEE Trans Instrum Meas 57(5):907–914
73. Pal K, Nigam MJ (2008) Novel active impedances using current conveyors. J Active Passive
Electron Devices 3:29–34
74. Yuce E, Minaei S (2008) Electronically tunable simulated transformer and its application to
Stagger-tuned filter. IEEE Trans Instrum Meas 57:2083–2088
75. Senani R, Bhaskar DR (2008) Comment: Practical voltage/current-controlled grounded
resistor with dynamic range extension. IET Circ Devices Syst 2:465–466
76. Skotis GD, Psychalinos C (2010) Multiphase sinusoidal oscillator using second generation
current conveyors. Int J Electron Commun (AEU) 64:1178–1181
77. Maheshwari S (2010) Current-mode third-order quadrature oscillator. IET Circ Devices Syst
4:188–195
III. Realization of Four Terminal Floating Nullors (FTFN)

Using CFOAs
78. Nordholt EH (1982) Extending op-amp capabilities by using a current-source power supply.
IEEE Trans Circ Syst 29:411–412
79. Stevenson JK (1984) Two-way circuits with inverse transmission properties. Electron Lett
20:965–967
80. Huijsing JH (1990) Operational amplifier. IEE Proc Circ Devices Syst 137:131–136
81. Senani R (1987) A novel application of four-terminal floating nullor. IEEE Proc
75:1544–1546
82. Hou CL, Yean R, Chang CK (1996) Single-element controlled oscillators using single FTFN.
83. Liu SI (1997) Single-resistance-controlled sinusoidal oscillators using two FTFNs. Electron
Lett 33:14
84. Abuelma’atti MT, Al-Zaher HA (1998) Current-mode sinusoidal oscillator using two FTFNs.
Proc Natl Sci Counc Repub China A 22:758–764
85. Wang HY, Lee CT (1998) Realization of R-L and C-D immittances using single FTFN.
86. Bhaskar DR (1999) Single resistance controlled sinusoidal oscillator using single FTFN.
Electron Lett 35:190
87. Abuelma’atti MT, Al-Zaher HA (1999) Current-mode quadrature sinusoidal oscillators using
two FTFNs. Frequenz 53:27–30
88. Abuelma’atti MT, Al-Zaher HA (1999) Current-mode sinusoidal oscillator using single
FTFN. IEEE Trans Circ Syst II 46:69–74
89. Gunes EO, Anday F (1999) Realization of voltage/current-mode filters using four-terminal
floating nullors. Microelectron J 30:211–216
90. Cam U, Toker A, Cicekoglu O, Kuntman H (2000) Current-mode high output impedance
sinusoidal oscillator configuration employing single FTFN. Analog Integr Circ Sign Process
24:231–238
91. Cam U, Cicekoglu O, Kuntman H (2000) Universal series and parallel immittance simulators
using four terminal floating nullors. Analog Integr Circ Sign Process 25:59–66
92. Lee CT, Wang HY (2001) Minimum realization for FTFN based SRCO. Electron Lett
37:1207–1208
93. Cam U, Cicekoglu O, Kuntman H (2001) Novel lossless floating immittance simulator
employing only two FTFNs. Analog Integr Circ Sign Process 29:233–235
94. Wang HY, Chung H, Huang WC (2002) Realization of an nth-order parallel immittance
function employing only (n 1) FTFNs. Int J Electron 89:645–650
95. Bhaskar DR (2002) Grounded-capacitor SRCO using only one PFTFN. Electron Lett
38(20):1156–1157
IV. CFOA-Based of Current Differencing Buffered

Amplifier (CDBA)
amplifier for analog signal processing filters. Microelectron J 30:157–160
97. Acar C, Ozoguz S (2000) nth-order current transfer function synthesis using current
differencing buffered amplifier: signal-flow graph approach. Microelectron J 31:49–53
98. Ozcan S, Toker A, Acar C, Kuntman H, Cicekoglu O (2000) Single resistance-controlled
sinusoidal oscillators employing current differencing buffered amplifier. Microelectron J
31:169–174
99. Ozcan S, Kuntman H, Cicekoglu O (2002) Cascadable current mode multipurpose filters
employing current differencing buffered amplifier (CDBA). Int J Electron Commun (AEU)
56:67–72
100. Horng JW (2002) Current differencing buffered amplifiers based single resistance controlled
quadrature oscillator employing grounded capacitors. IEICE Trans Fundam E85-
A:1416–1419
101. Keskin AU (2004) A four quadrant analog multiplier employing single CDBA. Analog Integr
102. Keskin AU (2005) Voltage-mode notch filters using single CDBA. Frequenz 59:1–4
103. Tangsrirat W, Surakampontorn W (2005) Realization of multiple-output biquadratic filters
using current differencing buffered amplifiers. Int J Electron 92:313–325
104. Keskin AU (2006) Multi-function biquad using single CDBA. Electr Eng 88:353–356
105. Keskin AU, Aydin C, Hancioglu E, Acar C (2006) Quadrature oscillator using current
differencing buffered amplifiers (CDBA). Frequenz 60:21–23
106. Koksal M, Sagbas M (2007) A versatile signal flow graph realization of a general transfer
function by using CDBA. Int J Electron Commun (AEU) 61:35–42
107. Tangsrirat W, Pisitchalermpong S (2007) CDBA-based quadrature sinusoidal oscillator.
Frequenz 61:102–104
108. Tangsrirat W, Pukkalanun T, Surakampontorn W (2008) CDBA-based universal biquad filter
and quadrature oscillator. Active Passive Electron Comp: Article ID 247171
109. Tangsrirat W, Prasertsom D, Piyatat T, Surakampontorn W (2008) Single-resistance-controlled
quadrature oscillator using current differencing buffered amplifiers. Int J Electron 95:1119–1126
110. Pathak JK, Singh AK, Senani R (2010) Systematic realization of quadrature oscillators using
current differencing buffered amplifiers. IET Circ Devices Syst 5:203–211
V. Unity Gain VF and CF Based Circuits Realized with CFOAs
111. Celma S, Sabadell J, Martinez P (1995) Universal filter using unity-gain cells. Electron Lett
31:1817–1818
References 219
112. Senani R, Gupta SS (1997) Universal voltage-mode/current-mode biquad filter realized with
current feedback op-amps. Frequenz 51:203–208
113. Abuelma’atti MT, Daghreer HA (1997) New single-resistor controlled sinusoidal oscillator
circuit using unity-gain current followers. Active Passive Electron Comp 20:105–109
114. Weng RM, Lai JR, Lee MH (2000) New universal biquad filters using only two unity gain
cells. Int J Electron 87(1):57–61
115. Kuntman H, Cicekoglu O, Ozcan S (2002) Realization of current-mode third order
Butterworth filters employing equal valued passive elements and unity gain buffers. Analog
Integr Circ Sign Process 30:253–256
116. Gupta SS, Senani R (2004) New single resistance controlled oscillators employing a reduced
number of unity-gain cells. IEICE Electron Express 1:507–512
117. Keskin AU, Toker A (2004) A NIC with impedance scaling properties using unity gain cells.
Analog Integr Circ Sign Process 41:85–87
118. Nandi R, Kar M (2009) Third order lowpass Butterworth filters using unity gain current
amplifiers. IEICE Electron Express 6:1450–1455
VI. Use of CFOAs in Realizing Operational Trans-resistance

Amplifiers (OTRA)
119. Chen JJ, Tsao HW, Chen CC (1992) Operational transresistance amplifier using CMOS
technology. Electron Lett 28:2087–2088
120. Salama KN, Elwan HO, Soliman AM (2001) Parasitic-capacitance-insensitive voltage-mode
MOSFET-C filters using differential current voltage conveyor. Circ Syst Sign Process
20:11–26
121. Cam U, Kacar F, Cicekoglu O, Kuntman H, Kuntman A (2004) Novel two OTRA-based
grounded immittance simulator topologies. Analog Integr Circ Sign Process 39:169–175
122. Hou CL, Chien HC, Lo YK (2005) Square wave generators employing OTRAs. IEE Proc
123. Lo YK, Chien HC (2006) Current-mode monostable multivibrators using OTRAs. IEEE
Trans Circ Syst II 53:1274–1278
124. Kilinc S, Salama KN, Cam U (2006) Realization of fully controllable negative inductance
with single operational transresistance amplifier. Circ Syst Sign Process 5(1):47–57
125. Chen JJ, Tsao HW, Liu SI, Chiu W (1995) Parasitic-capacitance-insensitive current-mode
filters using operational transresistance amplifiers. IEE Proc Circ Devices Syst 142:186–192
126. Lo YK, Chien HC (2007) Switch-controllable OTRA-based square/triangular waveform
generator. IEEE Trans Circ Syst II 54:1110–1114
127. Lo YK, Chien HC (2007) Single OTRA-based current-mode monostable multivibrator with
two triggering modes and a reduced recovery time. IET Circ Devices Syst 1:257–261
128. Lo YK, Chien HC, Chiu HJ (2008) Switch-controllable OTRA-based bistable multivibrator.
IET Circ Devices Syst 2:373–382
129. Lo YK, Chien HC, Chiu HJ (2010) Current-input OTRA Schmitt trigger with dual hysteresis
modes. Int J Circ Theory Appl 38:739–746
130. Sanchez-Lopez C, Martinez-Romero E, Tlelo-Cuautle E (2011) Symbolic analysis of
OTRAs-based circuits. J Appl Res Technol 9:69–80
131. Gupta A, Senani R, Bhaskar DR, Singh AK (2011) OTRA-based grounded-FDNR and
grounded-inductance simulators and their applications. Circ Syst Sign Process 31:489–499
VII. Use of CFOA in Realizing Differential Input and Buffered

Trans-Conductance Amplifier (DBTA)
132. Herencsar N, Vrba K, Koton J, Lattenberg I (2009) The conception of differential-input

buffered and transconductance amplifier (DBTA) and its application. IEICE Electron Express
6(6):329–334
133. Herencsar N, Koton J, Vrba K, Lahiri A (2009) New voltage-mode quadrature oscillator
employing single DBTA and only grounded passive elements. IEICE Electron Express
6:1708–1714
134. Herencsar N, Koton J, Vrba K, Lattenberg I (2010) New voltage-mode universal filter and
sinusoidal oscillator using only single DBTA. Int J Electron 97:365–379
VIII. Current Differencing Transconductance Amplifier (CDTA)

Using CFOAs
135. Biolek D (2003) CDTA-building block for current-mode analog signal processing. Proc
ECCTD’03, Krakow, Poland III, pp 397–400
136. Bekri AT, Anday F (2005) nth-order low-pass filter employing current differencing transcon-
ductance amplifiers. Proc 2005 European Conf Circ Theor Appl 2:II/193-II/196
137. Tangsrirat W (2007) Current differencing transconductance amplifier-based current-mode
four-phase quadrature oscillator. Indian J Eng Mater Sci 14:289–294
138. Prasad D, Bhaskar DR, Singh AK (2008) Realisation of single-resistance-controlled sinusoi-
dal oscillator: a new application of the CDTA. WSEAS Trans Electron 5:257–259
139. Silapan P, Siripruchyanum M (2011) Fully and electronically controllable current-mode
Schmitt triggers employing only single MO-CCCDTA and their applications. Analog Integr
140. Lahiri A (2010) Resistor-less mixed-mode quadrature sinusoidal oscillator. Int J Comput
Electr Eng 2:63–66
IX. Current Follower Transconductance Amplifier (CFTA)

Realized with CFOAs
141. Herencsar N, Koton J, Vrba K, Misurec J (2009) A novel current-mode SIMO type universal
filter using CFTAs. Contemp Eng Sci 2:59–66
142. Herencsar N, Koton J, Vrba K (2010) Realization of current-mode KHN-equivalent biquad
using current follower transconductance amplifiers (CFTAs). IEICE Trans Fundam
E93:1816–1819
References 221
X. CFOA Realizations of Current-Controlled Current Conveyor

Transconductance Amplifier (CCCC-TA)
143. Siripruchyanun M, Jaikla W (2007) Current controlled current conveyor transconductance

amplifier (CCCCTA): a building block for analog signal processing. Electr Eng 19:443–453
144. Maheshwari S, Singh SV, Chauhan DS (2011) Electronically tunable low-voltage mixed-
mode universal biquad filter. IET Circ Devices Syst 5(3):149–158
XI. Voltage-Differencing Differential-Input Buffered Amplifier

(VD-DIBA) Realized with CFOAs
145. Biolek D, Senani R, Biolkova V, Kolka Z (2008) Active elements for analog signal
processing: classification, review, and new proposals. Radioengineering 17:15–32
146. Prasad D, Bhaskar DR, Pushkar KL (2011) Realization of new electronically controllable
grounded and floating simulated inductance circuits using voltage differencing differential
input buffered amplifiers. Active passive Electron Compon: Article ID 101432
Chapter 8
Advances in the Design of Bipolar/CMOS
CFOAs and Future Directions of Research
on CFOAs
8.1 Introduction
Motivated by the widespread and potential applications of the CFOAs, as evidenced

by the publication of several hundred research papers, most of which have been
cited in this monograph, various researchers have worked towards evolving new
bipolar or CMOS architectures for CFOAs possessing one or more of the several
desirable features such as reduced input impedance at X-input terminal, better
accuracy of unity current gain between Z and X terminals and unity voltage gain
between Y and X terminals, higher slew rates, increased CMRR, enhanced gain-
bandwidth-products, lower DC offset voltage, better current drive capability and
reduced operating voltages etc. In this chapter, we will outline major developments
which have taken place on the improvement in the design of Bipolar/CMOS/
BiCMOS CFOAs and will also make some comments on future directions of
research on CFOAs and their applications.
8.2 Progress in the Design of Bipolar CFOAs
Although there have been hundreds of publications on improving the design of

Current Conveyors, surprisingly, in spite of the wide spread applications of the
CFOAs, as exemplified in this monograph and in the references cited in the various
chapters, there has been comparatively a much lesser effort [1–24] on improving
the design of bipolar and CMOS CFOAs.
224 8 Advances in the Design of Bipolar/CMOS CFOAs and Future Directions. . .
+VCC
I3 Q3' I4
I2
CORE CCII+
Y Q1' Q8'
Q5 Q7
I1 +Vbias Q13
Q9' Q3a
Q7'
Q9
Q10' Q6' Q1 Q3b Q11

X Z Q12 W
Q11' Q5'
Q2 Q4b
Q10
Q12' Q2' -Vbias
I1 Q4a Q14
Q6 Q8
I3
I2
Q4' I4
-VEE
Fig. 8.1 CFOA architecture proposed by Tammam et al. employing bootstrapping and a folded-
cascode in the input stage (adapted from [1] # 2003 Taylor & Francis)
8.2.1 Bipolar CFOA with Improved CMRR
Conventional CFOAs generally exhibit a poor CMRR. In [1], Tammam et al. have
carried out an analysis of CMRR of a conventional CFOA and have identified the
mechanism primarily responsible for the CMRR. They have then presented a
modified CFOA input stage circuit design by introducing a combination of
a bootstrapping technique and folded-cascode transistors resulting in a new
CFOA architecture which is shown to result in significant improvement in the
CMRR and gain accuracy. The circuit proposed in [1] is shown in Fig. 8.1.
This architecture has been shown to achieve a CMRR increased by some 62 dB
as compared to the conventional CFOA along with an improvement in the AC gain
error and input dynamic range although at the expense of reduction of slew rate and
slight increase in the supply currents by about 15%.
8.2.2 Bipolar CFOA with Higher Gain Accuracy,

Lower DC Offset Voltage and Higher CMRR
Hayatleh et al. in [2] presented a bipolar CFOA architecture based upon a new type
of input stage incorporating both forward and reverse bootstrapping technique
resulting in the architecture shown in Fig. 8.2. This novel topology was shown to
provide higher gain accuracy; lower DC offset voltage and higher CMRR. It was
8.2 Progress in the Design of Bipolar CFOAs 225
Q7 Q19 Q26 Q34 Q35 +VCC

Q8 Q42
Q17
Q9 Q13 Q15 Q24 Q36
Q27
Q47
Q11 Q22 Q28 Q44
Q46
Q1 Q29 Q40
RQ Q3 Z O/P
Y X
Q41
Q2 Q4 Q30 CP
Q48
Q12 Q24 Q31 Q45
Q49 Q39
Q32
Q10 Q14 Q18
Q25 Q37
Q16 Q38
Q6 Q20 Q33 Q43
Q5
−VEE
Fig. 8.2 CFOA using forward and reverse bootstrapping proposed by Hayatleh-Tammam et al.
(adapted from [2] # 2007 Taylor & Francis)
shown that in comparison with the conventional CFOA, the CMRR increases by
about 46 dB and the input offset voltage reduces by a factor of two. While the
majority of other characteristics are also better, the price to be paid is a reduced
output voltage swing because of vertical transistors stacking.
8.2.3 Bipolar CFOA Architectures with New Types

of Input Stages
Hayatleh et al. in [3] considered six new input stages with the intention of improv-
ing the performance of Bipolar CFOAs over a CFOA made from the conventional
input stage, taking three major characteristics to improve upon, namely, CMRR,
offset voltage and slew rate.
Figure 8.3 shows the schematic of a traditional CFOA architecture in which the
half circuit of the input stage has been shown in dotted box as half circuit-A. Apart
from this basic stage, six new half circuit stages have been presented therein which
are shown here in Fig. 8.4. Accordingly, six different bipolar CFOA formulations
were made and their performance in terms of CMRR, AC gain accuracy, and
frequency response for unity closed loop gain, transient response and input imped-
ance were studied and compared.
Out of the six new CFOA architectures, the one based upon input stage F has
been found to be superior as compared to the rest in respect of slew rate (950 V/μs),
input offset voltage (0.75 mV), input resistance (62.9 Ω), bandwidth (61.6 MHz)
and AC gain error (1.1 mV). This circuit has been shown in Fig. 8.5.
1:1 CURRENT MIRROR
VCC
IQ rs
A
Q3
Z
+ − X1 V0
Q1 CZ
VI Q2
RZ
Q4
rs IQ
VEE
1:1 CURRENT MIRROR
Fig. 8.3 The schematic of a traditional CFOA architecture (adapted from [3] # 2007 Springer)
a b
VCC CM1 CM1

VCC
I I I I
Q5
Q5
Q6
Q6
Q3
Q7
Q3
VIN+ Q1 Q7
Q1 VIN-
VEE VIN+ VIN-
c d
VCC CM1
VCC CM1
I I I
Q5 Q9 Q5
Q6
Q8
Q1 Q3
Q7
VIN+ Q1 Q3
VIN-
VIN+ VIN-
e f
VCC CM1
VCC CM1
I
I
Q8 Q10
Q5
Q9 Q5
Q8
Q1 Q3 Q1 Q3
VIN+ VIN- VIN+ VIN-
Fig. 8.4 The various half circuits of the modified input stages used in CFOA designs (adapted
from [3] # 2007 Springer)
8.3 The Evolution of CMOS CFOAs 227
+V
Q12
Q8 Q7 Q11 Q19
Q42 Q23
Q41 Q40 Q9 Q13
Q43 Q24 Q21
Q44
Q70 Q46
Q45
Q1 Q3 Q17
RQ Z
Q18 W
Y X
CP
Q2 Q4
Q55
Q80 Q56 Q22
Q54 Q25
Q53
Q51 Q10 Q16

Q50
Q52 Q26
Q6 Q5 Q14 Q20
Q15
-V
Fig. 8.5 Improved CFOA architecture using half circuit F (adapted from [3] # 2007 Springer)
8.2.4 Novel CFOA Architecture Using a New Current

Mirror Formulation
Recently, a novel CFOA utilizing a new current cell for biasing the CFOA has been
presented in [4] which is shown in Fig. 8.6. Although a drawback of this circuit is its
moderately higher power supply operation (V ¼ 3 V), this CFOA exhibits
performance characteristics, superior to those obtained with an established input
architecture, in terms of higher CMRR (91 dB), smaller offset voltage (<26 mV)
with an acceptable high slew rate and gain accuracy.
8.3 The Evolution of CMOS CFOAs
In this section, we will highlight some significant contributions made in the design
of CMOS CFOAs.
One of the early attempts in this direction was made by Bruun [5] who presented
a CMOS CFOA based upon a bipolar counterpart and demonstrated that this CMOS
CFOA has performance characteristics comparable to that of bipolar CFOA. This
structure is shown in Fig. 8.7.
Q9 VCC
Q1 Q7 Q37
Q33 Q42
Q5 Q11
Q3
Q15 Q13 Q31 Q35
Q17
Q21 Q19
Q43
Q23 Q27
Q25 Q29 X Q39

RQ Q40
Y W
Q26 Q30
CP
Q24 Q28
Q22 Q20 Q44
Q18
b Q16 Q14 Q32 Q38

Q12
Q4
Q6 Q36
Q34 Q41
Q2 Q8
Q10
VEE
Fig. 8.6 New CFOA utilizing new current cell proposed by Tammam et al. (adapted from [4] #
2012 World Scientific Publishing Company)
VDD
M7 M8 M9 M10
M1 M2 M13 M14
IB
Y X Z W
CT
M3 M4 M15 M16
M5 M6 M11 M12
VSS
Fig. 8.7 CMOS Current feedback op-amp due to Bruun (adapted from [5] # 1994 Springer)
VDD
M10 M9 M11 M29 M28

M14 M16 M30
M18 M32
M12
M1 M2 M20 M21
X Z W
Y M3 M4 M22 M23
M15
M17
M13 M33
M6 M5 M25 M24
M31
M7 M8 IB
VB M26 M27
M19
VSS
Fig. 8.8 The rail-to-rail CMOS CFOA proposed by Mahmoud et al. (adapted from [6] # 2000
Springer)
8.3.1 CMOS CFOA with Rail-to-Rail Swing Capability
Mahmoud et al. in [6], presented a CMOS structure with rail-to-rail swing capabil-
ity. This circuit is shown in Fig. 8.8. The circuit operates as class AB, is capable of
operating with 1.5 V DC power supplies and has a standby current of 200 μA.
This circuit appears to be suitable for low-voltage, low-power applications.
8.3.2 CMOS CFOA for Low-Voltage Applications
Two different variants of CMOS CFOA architectures were advanced by Maundy

et al. in [8]. It was shown that the two variants exhibited overall gain bandwidth
products in excess of 59 and 102 MHz respectively for a gain of 10 and with a
3.3 V DC power supply. One of these circuits is shown in Fig. 8.9.
8.3.3 Fully-Differential CMOS CFOAs
Mahmoud and Awad in [9] introduced a CMOS fully-differential CFOA

(FDCFOA) using 0.35 μm technology which is shown in Fig. 8.10. This implemen-
tation was evolved with a view to be used for the synthesis of fully differential
integrators and filters. This FDCFOA architecture is based upon a novel class AB
fully differential buffer circuit. The circuit is capable of being operated with a
supply voltage of 1.5 V and has a total standby current of 400 μA.
Vdd
M3 M4 M6 M10 M12
M8
Z M14
CC1 CT
V0
M7 CC2
M5 M13
X
M1 M2 M9 M11
Y Vss
Fig. 8.9 CMOS CFOA proposed by Maundy et al. (adapted from [8] # 2002 Springer)
VDD
IB/2 M31 IB/2 IB/2 M30 IB/2

Vbiasp
M17 M5 M6 M11 M20 M34

M36 M9 M32
M38
M22 M15 M13
M1 M2 M3 M4 M24 M25
M26 M27Z2 X2 Y2 Y1 X1
Vo-
M18 Z1 Vo+
Vbiasn M35
M23 M16 M12 M21 M29
Vbiasn
M39 M28 M7 M8
Vbiasn
M10 M33
M37 M14
M19
IB IB IB IB
VSB
VSS
Fig. 8.10 Fully differential current feedback operational amplifier proposed by Mahmoud and
Awad (adapted from [9] # 2005 Springer)
Another CMOS fully-differential CFOA was presented in [15] by Madian et al.

providing a wide range controllable bandwidth of 57–500 MHz without changing
the feedback resistance. The circuit had a standby current of 320 μA using a dual
power supply voltages of 0.75 V.
8.3.4 CMOS CFOAs with Increased Slew Rate

and Better Drive Capability
Two internally-compensated CMOS CFOAs were introduced by Cataldo et al. in

[11]. The circuits were made from class AB stages thereby obtaining increased slew
rate and better drive capability. Experimental results on a prototype implemented in
VDD
M9 M10 M12 M19 M20 M21
M11
M4 X M8
M2 M16 M6 W
Y M1 Z M5
M3 VB2 M7
CP
RP
IB IB IB IB
M15 M17 M24

VB1
M14
M13 M18 M22 M23
VSS
Fig. 8.11 CMOS CFOA due to Cataldo et al. (adapted from [11] # IEEE 2007)
0.35 μm process demonstrated a slew rate of 35 V/μs and a constant bandwidth of

the order of 2 MHz in an inverting amplifier configuration with a 10 kΩ feedback
resistor. One of these circuits is shown in Fig. 8.11.
8.3.5 Other CMOS CFOA Architectures
A CMOS CFOA was proposed by Ismail and Soliman in [7] which was
implemented in 0.5 μm CMOS technology. This circuit exhibited frequency
range of operation up to around 180 MHz using a DC bias power supply 2.5 V
and offered input impedance of the order of 2 Ω at the x-input terminal.
The CMOS CFOA presented in [10] is based upon connecting two high perfor-
mance CCIIs and is aimed at achieving good input/output swing and drive capabil-
ity. The circuit, however, has the drawback of having poor slew rate performance
which, however, was shown to be substantially improved by adopting a class AB
differential stage in the CCIIs.
The CMOS CFOA of [12] exhibited a bandwidth of 34.6 MHz, CMRR of 41 dB,
input impedance of X-terminal as 1.65 kΩ, that of Y-terminal as 227.16 MΩ , the
output impedance of 125.51 MΩ at Z-terminal and of W-terminal 1.65 kΩ.
Mahmoud et al. in [14] presented a low-voltage CMOS CFOA which allows rail-
to-rail input/output operation with high drive capability using a supply voltage of
0.75 V and a total standby current of 304 μA. The circuit had a bandwidth
of 120 MHz and a current drive capability of 1 mA.
In [23], Maundy et al. proposed a new topology for designing a CMOS CFOA by
employing a CMOS CCII+ and a CMOS op-amp in an unconventional manner. The
workability of the configuration was verified making an IC manufactured in 0.18 μm
digital CMOS process.
A new CMOS CFOA based on the design and use in a repeated pattern of a
current transfer cell was presented in [24] which resulted in the reduction of input
referred offset voltage, a CMRR increased by approximately 53 dB and an
improvement in AC gain accuracy as compared to a conventional CMOS CFOA.
8.4 Various Modified Forms of CFOAs and Related Advances
In view of the popularity of the CFOAs in various analog signal processing and
signal generation applications, from time to time, several modified forms of CFOAs
have also been proposed by various researchers. Some of these are: current con-
trolled CFOA (CC-CFOA) proposed by Siripruchyanun et al. [16], modified CFOA
(MCFOA) proposed by Yuce and Minaei [17], differential voltage CFOA
(DVCFOA) proposed by Gunes and Toker [18], differential difference complimen-
tary current feedback amplifier (DDCCFA) proposed by Gupta and Senani [19].
8.4.1 The Modified CFOA
The so-called MCFOA [17], when closely examined, turns out to be exactly same as
the ‘Composite Current Conveyor’ proposed by Smith and Sedra in one of their very
early publications on current conveyors [25]. In fact, its CMOS realization also can
be identified to be a composite connection of a CCII+ and CCII. Furthermore, its
implementation in terms of normal kind CFOAs has three CFOAs out of which the
first one is used as a CCII+ while the remaining two together are configured as
CCII. Thus, any single MCFOA-based circuit is actually a circuit involving a CCII
+ and a CCII and hence, in the opinion of the present authors, the MCFOA as a
building block is not distinctly different than the composite current conveyor of [25].
8.4.2 Current-Controlled CFOA
BiCMOS CC-CFOA architecture was proposed by Siripruchyanun et al. in [16]

The BiCMOS CC-CFOA of [16] consists of two blocks out of which the first is a
CC-CCII+ and the other is a modified voltage follower. The CC-CCII stage consists
of mixed translinear loop consisting of transistors Q6–Q9 which is biased by a
current IB through the current mirrors consisting of M1–M3 and M10–M11. The input
resistance looking into the terminal-X is given by rx ¼ VT/2IB. A replica of the
current ix is generated and conveyed to the Z terminal using the CMOS transistors
M4–M5 and M12, M13. On the other hand, the voltage buffer is realized using
bipolar transistors Q10–Q22 and M14–M21. Since the structure uses only NPN
8.4 Various Modified Forms of CFOAs and Related Advances 233
VDD
M19 M20
M10 M11 M12 M13 M18 M21
Q1
Q14
Q8 Q9
IB Y X Z IA
Q10 Q11
Q6 Q7 Q21
Q12 Q13 W Q19
M16
M2 M15 Q14
M1 M3 M4 M5 M14 M17 Q18
Q15 Q16 Q17 20 Q22

VSS
Fig. 8.12 A BiCMOS CFOA architecture proposed by Siripruchyanun et al. (adapted from [16]
# 2008 WSEAS)
Y Z
W Y
X CC Z
R
X
Fig. 8.13 Current feedback conveyor proposed by Gift and Maundy (adapted from [20] # 2008
John Wiley & Sons)
transistors, good high frequency behavior is expected. SPICE Simulations, using

PMOS and NMOS transistors with parameters of 0.35 μm TSMC CMOS technology,
have shown that rx is controllable from 53 Ω to 12.62 kΩ by varying IB from 1 μA to
300 μA. The circuit when operated from 1.5 V biasing shows a power consumption
of 4.16 mW with 3 dB bandwidth of 76 MHz for iz/ix, 342 MHz for vx/vy and 260 MHz
for vw/vz.
8.4.3 Current Feedback Conveyor
A novel circuit element utilizing a CCII+ (realizable from a CFOA) with its input
circuit in the feedback loop of a CFOA was advanced by Gift and Maundy [20] who
preferred to call it a Current feedback conveyor (CFC). A CFC is shown in
Fig. 8.13.
It was shown to be particularly attractive for realizing Current amplifiers and can
be implemented from variety of CFOAs such as AD844, OPA 603 and OPA 623
and a variety of CCs such as AD844 and OPA660.
8.4.4 The Differential Voltage Current Feedback Amplifier
The differential voltage current feedback amplifier (DVCFA) introduced by Gunes

and Toker in [18] provides a differential Y input (consisting of two terminals Y1
M21 M22 +VDD
M18
M5 M6 M16 M17
M10 M12 RC2
RC1 CC2
IA
Y2 Y3 Y1 X Z
CC1 w
M1 M2 M3 M4 M14 M15
M7 M8 M9 M11 M13 M19
−VSS M20
Fig. 8.14 CMOS implementation of DVCFA by Gunes and Toker (adapted from [18] # 2002
Elsevier)
Y1
C1 W Vo
R3 Y2
y
w
z Vo
x
R3 X Z
R1 C2 R2
R2
C1 R1 C2
Fig. 8.15 A CFOA-based SRCO and its DVCFA-based counterpart employing both GCs as
proposed by Gunnes and Toker (adapted from [18] # 2002 Elsevier)
and Y2), has the characterization Iyi ¼ 0, i ¼ 1, 2; Vx ¼ (Vy1 Vy2), Iz ¼ Ix and

Vw ¼ Vz and is shown in Fig. 8.14.
DVCFAs have been shown to be particularly useful building blocks for
synthesizing SRCOs employing grounded capacitors (GC). In this context, it may
be noticed that while it has been amply demonstrated by a number of researchers
that single resistance controlled oscillators (SRCO) can be realized using only a
single CFOA however, none of the circuits known so far is able to employ both GCs
as desirable for integrated circuit implementation. A DVCFA is particularly useful
in removing this difficulty and it makes GC-SRCOs realizable from a single
DVCFA. A family of eight such GC-oscillators has been derived by Gunes and
Toker in [18]. An exemplary realization therefrom is shown here in Fig. 8.15.
Both the circuits are characterized by exactly the same CO and FO which are
given by
8.4 Various Modified Forms of CFOAs and Related Advances 235
+VDD
M32 M33 M24
M5 M6 M16 M17 M18 M27 M28 M29
M10 M12 RC2 M21 RC3
RC1 CC2 CC3

IA M25
Y2 Y3 Y1 X Z1 w1 Z2 w2
C
M1 M2 M3 M4 C1 M14 M15 M22 M23 M26
M7 M8 M9 M11 M13 M19 M20 M30 M31
−VSS
Fig. 8.16 An exemplary CMOS implementation of the DDCCFA [26]
C1 R1 ¼ C2 R2 ðadjustable by R2 Þ (8.1)
1
ω0 ¼ ðadjustable by R3 Þ (8.2)
C 1 C 2 R1 R3
8.4.5 Differential Difference Complementary Current

Feedback Amplifier
Another extension of the CFOA named Differential Difference Complementary

Current Feedback Amplifier (DDCCFA) obtained by appropriate modification
of the DVCFA of Gunes and Toker [18], was proposed by Gupta and Senani in
[19, 26]. It is an active eight-port building block defined by the following
characterizing equations: iyk ¼ 0; k ¼ 1–3, vx ¼ vy1 vy2 + vy3, iz1 ¼ ix, iz2 ¼
ix, vw1 ¼ vz1 and vw2 ¼ vz2. An exemplary implementation of DDCCFA is
shown in Fig. 8.16.
It was shown in [19] and [26] that a single DDCCFA is sufficient to generate
SRCO circuits possessing the following properties simultaneously: (a) use of a
single active building block, (b) employment of two GCs along with a minimum
number (only three) of resistors, (c) non-interacting controls of CO and FO, (d) a
simple condition of oscillation (i.e. not more than one condition) and (e) availability
of current-mode and voltage-mode outputs both explicitly. An exemplary SRCO
using a DDCCFA is shown in Fig. 8.17.
This circuit is characterized by the following CO and FO
C1 R3
CO : ¼ (8.3)
C2 R1
1
FO : f0 ¼ (8.4)
C1 C2 R2 R3
Fig. 8.17 An exemplary

SRCO using a DDCCFA [26]
C2 R3 C1
y1 z2
y2 w2 V02
y3 V01
w1
x
z1
R1 R2
I0
from where it is seen that CO is adjustable by R1 while FO is independently variable

through R2. Furthermore, both VM and CM outputs are available and circuit
employs both GCs as preferred for IC implementation.
It has also been demonstrated in [19] and [26] that employing a single DDCCFA,
a large number of previously known building blocks can be derived as special cases
as shown in the following:
(a) First generation current conveyor minus type (CCI) can be realized by
grounding terminals y2, y3, interconnecting terminals y1, z1 and leaving
terminals w1, w2 unused.
(b) Second Generation Current Conveyor (CCII+) can be realized by grounding y2,
y3, z2 terminals while leaving terminals w1, w2 unused.
(c) Second Generation Current Conveyor (CCII) can be realized by grounding
terminals y2, y3, z1 and leaving terminals w1, w2 unused.
(d) Dual Output Current Conveyor (DOCC) can be realized by grounding y2, y3
terminals and leaving w1, w2 terminals unused.
(e) Differential Voltage Current Conveyor (DVCC+) [27] can be realized by
grounding terminals y3, z2 while w1, w2 terminals are unused.
(f) Differential Voltage Current Conveyor (DVCC) [27] can be realized by
grounding terminals y3, z2 and leaving terminals w1, w2 unused.
(g) Differential Voltage Complementary Current Conveyor (DVCCC) [27, 28] can
be realized by grounding terminal y3 and terminals w1, w2 are left unused.
(h) Current Feedback Operational Amplifier (CFOA) can be realized by grounding
y2, y3, z2 terminals while leaving terminal w2 unused.
(i) Differential Difference Complementary Current Conveyor (DDCCC) [27, 29]
can be realized by leaving terminals w1 and w2 unused.
(j) Differential Difference Complementary Current Conveyor (DDCCC) [27, 29]
can be realized by grounding y3, z2 terminals and leaving terminal w2 unused.
(k) Inverting Second Generation Current Conveyor (IVCCII+) [29] can be realized
by connecting terminals y1, y3, z2 to ground and terminals w1, w2 are left unused.
8.6 Epilogue 237
(l) Inverting Second Generation Current Conveyor (IVCCII) [29] can be realized
by connecting terminals y1, y3, z1 to ground and terminals w1, w2 are left
unused.
(m) Third Generation Current Conveyor (CCIII+) [30] can be realized by ground-
ing terminals y2, y3 , interconnecting terminals y1, z2 and leaving terminals w1,
w2 unused.
With more than one DDCCFA, other building blocks such as first generation
CCI+, third generation CCIII, current differencing buffered amplifier (CDBA)
[31], four terminal floating nullor (FTFN) [32, 33], operational trans-resistance
amplifier (OTRA) [34] and fully differential second generation current conveyors
(FDCCII) [35] can also be realized. Thus, the DDCCFA can be considered to be a
generalized building block.
8.5 Future Directions of Research on CFOAs

While a considerable number of applications of CFOAs have been elaborated in this

monograph, a number of applications such as those of realizing precision rectifiers,
relaxation oscillators and analog multipliers/dividers etc. are, in the opinion of the
authors, far from being completely investigated and a number of new configurations
might be still waiting to be discovered. Thus, there is ample scope for evolving new
configurations for the above mentioned as well as many newer applications of the
CFOAs, not attempted so far.
Another area in which research work is still continuing is the development of
improved bipolar/CMOS/BiCMOS circuit architectures of the CFOA itself. In this
respect, one can also add a number of recent innovations like several varieties of so-
called modified CFOAs (MCFOA), DVCFOA, and DDCCFA which were shown to
be useful and versatile building blocks capable of providing many advantages in
analog circuit design over the traditional type of CFOAs.
Among the continued efforts on improving the design of Bipolar and CMOS
CFOAs, the systematic synthesis of CFOAs advanced by Torres-Papaqui and Tlelo-
Cuautle [22] through manipulation of voltage followers and current followers looks
very promising.
8.6 Epilogue
In spite of covering various aspects of the design and applications of CFOAs in this
monograph, it can be concluded that the area still holds a lot of promise for the
discovery of further new applications and the design of improved CFOAs. We hope
that this exposition should be able to provide to the general readers, circuit
designers and researchers, the needed stimulus to carry out further work on newer
possibilities in improving the design of the CFOA and in searching newer
applications of the CFOAs. For further studies, an additional list of references has
been provided at the end.
It is interesting to note that the existing literature on CFOAs and their
applications spread over dozens of international journals resulting in the publication
of several hundred research papers on CFOAs have curiously focused their atten-
tion quite dominantly only on one specific type of CFOA namely, the AD844,
which is uniquely different from the variety of other CFOAs from various
manufacturers in that, this happens to be the only CFOA which has its Z-pin
accessible outside the chip. As a consequence of this, researchers and academicians
have found this CFOA to be particularly flexible and versatile because it (1) can be
used as an op-amp, (2) can be used to realize current conveyor based circuits, (3)
can be used to realize circuits based upon many other building blocks of more
recent origin and at the same time, (4) can also be used as a 4-terminal building
block in its own right.
In view of the wide spread use and applications of the CFOA with an external
accessible Z-pin, it is, therefore, extremely surprising that none of the
manufacturers have turned their attention to produce any more CFOAs of this
kind and have limited themselves without exception, to only three terminal
CFOAs which can be used only as a replacement of the traditional VOAs with
the only advantage of offering superior performance in the same topologies (as
compared to their VOA-based counterparts).
In view of the forgoing, the authors strongly feel that to tap the full potential of
CFOAs with external accessible Z-terminal, the leading ICs manufacturers should
produce improved versions of bipolar/CMOS/BiCMOS CFOAs and should neces-
sarily provide one or more (if two, then complimentary) current-mode outputs and
as many buffered voltage-mode outputs which will definitely make such a building
block much more capable, flexible and versatile for various analog signal
processing and signal generation applications in both linear and non-linear modes.
References
1. Tammam AA, Hayatleh K, Lidgey FJ (2003) High CMRR current-feedback operational

amplifier. Int J Electron 90:87–97
2. Hayatleh K, Tammam AA, Hart BL, Lidgey FJ (2007) A novel current-feedback op-amp
exploiting bootstrapping techniques. Int J Electron 94:1157–1170
3. Hayatleh K, Tammam AA, Hart BL (2007) Novel input stages for current feedback operational
amplifiers. Analog Integr Circ Sign Process 50:163–183
4. Tammam AA, Ben-Esmael M, Abazab MR (2012) Current feedback op-amp utilizes new
current cell to enhance the CMRR. J Circ Syst Comput 21:1250038-1–13
5. Bruun E (1994) A dual current feedback Op amp in CMOS technology. Analog Integr Circ
Sign Process 5:213–217
References 239
6. Mahmoud SA, Elwan HO, Soliman AM (2000) Low voltage rail to rail CMOS current
feedback operational amplifier and its applications for analog VLSI. Analog Integr Circ Sign
Process 25:47–57
7. Ismail AM, Soliman AM (2000) Novel CMOS current feedback op-amp realization suitable
for high frequency applications. IEEE Trans Circ Syst-I 47:918–921
8. Maundy BJ, Finvers IG, Aronhime P (2002) Alternative realizations of CMOS current
feedback amplifiers for low voltage applications. Analog Integr Circ Sign Process 32:157–168
9. Mahmoud SA, Awad IA (2005) Fully differential CMOS current feedback operational ampli-
fier. Analog Integr Circ Sign Process 43:61–69
10. Mita R, Palumbo G, Pennisi S (2005) Low-voltage high-drive CMOS current feedback op-
amp. IEEE Trans Circ Syst-II 52:317–321
11. Cataldo GD, Grasso AF, Pennisi S (2007) Two CMOS current feedback operational amplifiers.
IEEE Trans Circ Syst-II 54:944–948
12. Perez AP, Cuautle ET, Mendez AD, Lopez CS (2007) Design of a CMOS compatible CFOA
and its application in analog filtering. IEEE Latin Am Trans 5:72–76
13. Maundy B, Gift S, Magierowski S (2007) Constant bandwidth current feedback amplifier from
two operational amplifiers. Int J Electron 94:605–615
14. Mahmoud SA, Madian AH, Soliman AM (2007) Low-voltage CMOS current feedback
operational amplifier and its application. ETRI J 29:212–218
15. Madian AH, Mahmoud SA, Soliman AM (2007) Low voltage CMOS fully differential current
feedback amplifier with controllable 3-dB bandwidth. Analog Integr Circ Sign Process
52:139–146
16. Siripruchyanun M, Chanapromma C, Silapan P, Jaikla W (2008) BiCMOS current-controlled
current feedback amplifier (CC-CFA) and its applications. WSEAS Trans Electron 5:203–219
17. Yuce E, Minaei S (2008) A modified CFOA and its applications to simulated inductors,
capacitance multipliers, and analog filters. IEEE Trans Circ Syst-I 55(2):266–275
18. Gunes EO, Toker A (2002) On the realization of oscillators using state equations. Int J Electron
Commun (AEU) 56:317–326
19. Gupta SS, Senani R (2005) Grounded-capacitor SRCOs using a single differential difference
complimentary current feedback amplifier. IEE Proc Circ Devices Syst 152:38–48
20. Gift SJG, Maundy B (2008) A novel circuit element and its application in signal amplification.
Int J Circ Theor Appl 36:219–231
21. Madian AH, Mahmoud SA, Soliman AM (2008) Configurable analog block based on CFOA
and its application. WSEAS Trans Electron 5:220–225
22. Papaqui LT, Cuautle ET (2004) Synthesis of CCs and CFOAs by manipulation of VFs and
CFs. Proc 2004 I.E. international behavioral modeling simulation conference. pp 92–96
23. Maundy BJ, Sarkar AR, Gift SJ (2006) A new design topology for low-voltage CMOS current
feedback amplifiers. IEEE Trans Circ Syst-II 53:34–38
24. Tammam AA, Hayatleh K, Hart B, Lidgey FJ (2003) Current feedback operational amplifier
with high CMRR. Electron Lett 39:2
25. Smith KC, Sedra A (1970) Realization of the Chua family of new nonlinear network elements
using the current conveyor. IEEE Trans Circ Theor 17:137–139
26. Gupta SS (2005) Realization of some classes of linear /nonlinear analog electronic circuits
using current-mode building blocks. PhD thesis (Supervisor Raj Senani), Faculty of Technol-
ogy, University of Delhi, Ch 4. pp 128–188
27. Elwan HO, Soliman AM (1997) Novel CMOS differential voltage current conveyor and its
applications. IEE Proc Circ Devices Syst 144:195–200
28. Gupta SS, Senani R (2000) Grounded-capacitor current-mode SRCO: novel application of
DVCCC. Electron Lett 36:195–196
29. Awad IA, Soliman AM (1999) Inverting second generation current conveyors: the missing
building blocks, CMOS realizations and applications. Int J Electron 86:413–432
30. Fabre A (1995) Third-generation current conveyor: a new helpful active element. Electron Lett
31:338–339
amplifier suitable for analog signal-processing filters. Microelectron J 30:157–160
32. Senani R (1987) Generation of new two-amplifier synthetic floating inductors. Electron Lett
23:1202–1203
33. Senani R (1987) A novel application of four terminal floating nullors. Proc IEEE
75:1544–1546
34. Chen JJ, Tsao HW, Chen CC (1992) Operational transresistance amplifier using CMOS
technology. Electron Lett 28:2087–2088
35. El-Adawy AA, Soliman AM, Elwan HO (2000) A novel fully differential current conveyor and
applications for analog VLSI. IEEE Trans Circ Syst-II 47:306–313
Erratum
Current Feedback Operational Amplifiers
Raj Senani, D.R. Bhaskar, A.K. Singh and V.K. Singh
Errata to:
R. Senani et al., Current Feedback Operational Amplifiers
and Their Applications, Analog Circuits and Signal
Processing, DOI 10.1007/978-1-4614-5188-4
In chapter 1 titled “Introduction”, Figures 1.2 and 1.18 are updated.
In chapter 2 titled “CFOAs: Merits, Demerits, Basic Circuits and Available

Varieties”, text has been updated in pages 28, 29, 30, 34, 43, 44, 45 and 47.
In chapter 3 titled “Simulation of Inductors and Other Types of Impedances Using

CFOAs”, the text has been updated in pages 52, 53, 59, 61, 62, 63, and 64.
In chapter 4 titled “Design of Filters Using CFOAs”, text corrections are made
in pages 85, 87, 88, 89, 95, 100, 103, 104, 106, 109, 111, 112, 114, 116, 117, 123
and 125.
–––––––––––––––––
The online version of the original book can be found at
http://dx.doi.org/10.1007/978-1-4614-5188-4
The updated online version of these chapters can be found at

R. Senani et al., Current Feedback Operational Amplifiers and Their Applications, E1

Analog Circuits and Signal Processing, DOI 10.1007/978-1-4614-5188-4_9
© Springer Science+Business Media New York 2017
E2 Erratum
In chapter 5 titled “Synthesis of Sinusoidal Oscillators Using CFOAs”, text has

been updated in pages 136, 143, 150, 151, and 175.
In chapter 6 titled “Miscellaneous Linear and Nonlinear Applications of CFOAs”,

text has been updated in pages 181, 185, 186, 189, 198.
In chapter 7 titled “Realization of Other Building Blocks Using CFOAs”, text has
been updated in pages 205, 208, 209, 210, 211 and 213.
In chapter 8 titled “Advances in the Design of Bipolar/CMOS CFOAs and Future

Directions of Research on CFOAs”, text has been updated in pages 230 and 234.
References for Additional Reading
1. Smith SO (1993) The good, the bad and the ugly: current feedback-technical contributions and
limitations. IEEE Int Sympos ISCAS 93:1058–1061
2. Bruun E (1993) Feedback analysis of trans-impedance operational amplifier circuits. IEEE
Trans Circ Syst-I 40(4):275–278
3. Hart BL (1993) Current and voltage feedback op.-amps: a unified approach. Int J Electron
75(4):715–718
4. Higashimura M (1995) Filters and immittances using current-feedback amplifiers.
Proceedings of the 20th International Conference on Microelectronics. 2: NIS SERBIA:
737742
5. Mahattanakul J, Toumazou C (1996) A theoretical study of the stability of high frequency
current feedback Op-amp integrators. IEEE Trans Circ Syst-I 43(1):2–12
6. Fabre A, Amrani H, Saaid O (1996) Current-mode band-pass filters with Q-magnification.
IEEE Trans Circ Syst-II 43(12): 839–842
7. Serrano L, Carlosena A (1997) Active RC impedances revisited. Int J Circ Theor Appl
25:289–305
8. Dostal T, Prokop R, Sarman R (1997) Functional blocks and biquadratic ARC filters using
trans-impedance amplifiers. Radioengineering 6(1):9–15
9. Martinez PA, Aldea C, Sabadell J, Celma S (1998) Approach to the realization of state
variable based oscillators. IEEE Int Conf Electron Circuits Syst 3:139–142
10. Palumbo G (1999) Bipolar current feedback amplifier: compensation guidelines. Analog
Integr Circuits Signal Process 19:107–114
11. Yang Z (1999) Circuit transformation method from OTA-C circuits into CFA-based RC
circuits. IEE Proc Circuits Devices Syst 146(2):99–100
12. Schmidt H (2000) Approximating the universal active element. IEEE Trans Circ Syst-II
47(11):1160–1169
13. Weng RM, Lai JR, Lee MH (2000) Realization of nth-order series impedance function using
only (n-1) current-feedback amplifiers. Int J Electron 87(1):63–69
14. Gift SJG (2001) Hybrid current conveyor-operational amplifier circuit. Int J Electron
88(12):1225–1235
15. Palumbo G, Pennisi S (2001) Current-feedback amplifiers versus voltage operational
amplifiers. IEEE Trans Circ Syst-I 48(5):617–623
16. Bayard J (2001) CFOA based inverting amplifier bandwidth enhancement. IEEE Trans Circ
Syst-II 48(12):1148–1150
17. Takagi S (2001) Analog circuit designs in the last decade and their trends toward the 21st
century. IEICE Trans Fundam E84-A(1):68–79
18. Gilbert B (2001) Analog at milepost 2000: a personal perspective. Proc IEEE 89(3):289–304
Analog Circuits and Signal Processing, DOI 10.1007/978-1-4614-5188-4,
242 References for Additional Reading
19. Bayard J, Ayachi M (2002) OTA-or CFOA-based LC sinusoidal oscillators-analysis of the

magnitude stabilization phenomenon. IEEE Trans Circ Syst-I 49(8):1231–1236
20. Venkateswaran P, Nagaria RK, Sanyal SK, Nandi R (2003) Dual-input single-tunable integrators
and differentiators using current-feedback amplifier. Int J Electron 90(2): 109–115
21. Wu R, Lidgey FJ, Hayatleh K (2004) Using the ‘T’ feedback network with the current
feedback operational amplifier. Int J Electron 91(11):685–695
22. Mita R, Palumbo G, Pennisi S (2004) Effect of CFOA non-idealities in Miller integrator cells.
IEEE Trans Circ Syst-II 51(5):249–253
23. Natarajan S (2004) A variable frequency oscillator using modern current feedback amplifiers
(CFAs) for high frequency applications. Proceedings of the 36th southeastern symposium
on system theory. pp 417–421
24. Gilbert B (2004) Current mode, voltage mode, or free mode? a few sage suggestions. Analog
Integr Circuits and Signal Process 38:83–101
25. Natarajan S (2005) Inductance simulation using modern current feedback amplifiers (CFAs).
Proceedings of the 37th southeastern symposium on system theory. pp 60–64
26. Natarajan S (2005) A narrow-band VHF-tuned amplifier using modern current feedback
amplifiers (CFAs). Proceedings of the 37th southeastern symposium on system theory.
pp 55–59
27. Gift JGS, Maundy Brent (2005) Improving the bandwidth gain-independence and accuracy of
the current feedback amplifier. IEEE Trans Circ Syst-II 52(3):136–139
28. Lim H, Park J (2006) Frequency-domain analysis of effects of the location of a feedback
resistor in a current feedback amplifier. IEEE Trans Circ Syst-II 53(8):687–691
29. Gupta SS, Sharma RK, Bhaskar DR, Senani R (2006) Synthesis of sinusoidal oscillators with
explicit-current-output using current feedback Op-amps. WSEAS Trans Electron 3(7):
385–388
30. Alzaher HA (2006) A novel band pass filter based on current feedback amplifier. Analog
Integr Circuits Signal Process 46:145–148
31. Pal K (2006) All pass/notch filters using operational amplifier and current conveyors. JAPED
1:289–294
32. Vochyan J (2006) Synthesis of new biquad filters using two CFOAs. Radioengineering
15(4):76–79
33. Nagaria RK (2008) On the new design of CFA based voltage controlled integrator/
differentiator suitable for analog signal processing. WSEAS Trans Electron 6(5):232–237
34. Bayard J (2008) Universal filter using a ‘pseudo’ gyrator. Int J Electron 95(2):77–83
35. Raikos G, Psycalinos C (2009) Low-voltage current feedback operational amplifiers. Circuits
Syst Signal Process 28:377–388
36. Yuce E (2010) Fully-integrable mixed-mode universal biquad with specific application of the
CFOA. Int J Electron Commun (AEU) 64:304–309
37. Pennisi S, Scotti G, Trifiletti A (2011) Avoiding the gain-bandwidth trade off in feedback
38. Ananda Mohan PV (2011) Comments on avoiding the gain-bandwidth trade off in feedback
39. Pennisi S, Scotti G, Trifiletti A (2011) Reply to ‘Comments on Avoiding the gain-bandwidth
trade off in feedback amplifiers. IEEE Trans Circ Syst-I 58(9):2117
40. Rybin YK (2012) The nonlinear distortions in the oscillatory system of generator on CFOA.
Hindawi Publishing Corporation: active and passive electronic components. Article ID
908716: doi:10.1155/2012/908716
About the Authors
Raj Senani received B.Sc. from Lucknow University, B.Sc. Engg. from Harcourt
Butler Technological Institute, Kanpur, M.E. (Honors) from Motilal Nehru
National Institute of Technology (MNNIT), Allahabad and Ph.D. in Electrical
Engg. from the University of Allahabad.
Dr. Senani held the positions of Lecturer (1975–1986) and Reader (1986–1988)
at the EE Department of MNNIT, Allahabad. He joined the ECE. Department of
the Delhi Institute of Technology (now named as Netaji Subhas Institute of
Technology) in 1988 and became a full Professor in 1990. Since then, he has
served as Head, ECE Department, Head Applied Sciences, Head, Manufacturing
Processes and Automation Engineering, Dean Research, Dean Academic, Dean
Administration, Dean Post Graduate Studies and Director of the Institute, a
number of times.
Professor Senani’s teaching and research interests are in the areas of Bipolar and
CMOS Analog Integrated Circuits, Electronic Instrumentation and Chaotic
Nonlinear Circuits. He has authored/co-authored 135 research papers in various
international journals and 4 book chapters for monographs published by Springer.
He is currently serving as Editor-in-Chief for IETE Journal of Education and as an
244 About the Authors
Associate Editor for the Journal on Circuits, Systems and Signal Processing,
Birkhauser Boston (USA) since 2003, besides being on the editorial boards of
several other journals and acting as an editorial reviewer for 30 international
journals.
Professor Senani is a Senior Member of IEEE and was elected a Fellow of the
National Academy of Sciences, India, in 2008. He is the recipient of Second
Laureate of the 25th Khwarizmi International Award for the year 2012. Professor
Senani’s biography has been included in several editions of Marquis’ Who’s Who
series (published from NJ, USA) and a number of other international biographical
directories.
D.R. Bhaskar received B.Sc. degree from Agra University, B.Tech. degree from
Indian Institute of Technology (IIT), Kanpur, M.Tech. from IIT, Delhi and Ph.D.
from University of Delhi. Dr. Bhaskar held the positions of Assistant Engineer in
DESU (1981–1984), Lecturer (1984–1990) and Senior Lecturer (1990–1995) at the
EE Department of Delhi College of Engineering and Reader in ECE Department of
Jamia Millia Islamia (1995–2002). He became a full Professor in January 2002 and
has served as the Head of the Department of ECE during 2002–2005.
Professor Bhaskar’s teaching and research interests are in the areas of Analog
Integrated Circuits and Signal Processing, Communication Systems and Electronic
Instrumentation. He has authored/co-authored 64 research papers in various Inter-
national journals and 3 book chapters for monographs published by Springer. He
has acted/has been acting as a Reviewer for several international journals. Professor
Bhaskar is a Senior Member of IEEE. His biography is included in a number of
international biographical directories.
About the Authors 245
A.K. Singh received M.Tech. in Electronics and Communication Engineering from

IASED and Ph.D., in the area of Analog Integrated Circuits and Signal processing,
from Netaji Subhas Institute of Technology (NSIT), University of Delhi. Dr. Singh
held the positions of Lecturer and Senior Lecturer (June 2000–August 2001) at the
ECE Department, AKG Engineering College, Ghaziabad. He joined ECE Depart-
ment of Inderprastha Engineering College, Ghaziabad, India as a Senior Lecturer in
August 2001 where he became Assistant Professor in April, 2002 and Associate
Professor in 2006.
At present, he is a full Professor at the ECE Department of HRCT Group of
Institutions, Faculty of Engineering and Technology, Ghaziabad, India. His teach-
ing and research interests are in the areas of Bipolar and MOS Analog Integrated
Circuits and Signal Processing. Dr. Singh has authored/co-authored 36 research
papers in various International journals and 2 book chapters for monographs
published by Springer.
V.K. Singh obtained B.E. and M.E. degrees in Electrical Engineering from MNR
Engineering College, Allahabad and Ph.D. in Electronics and Communication
Engineering from Uttar Pradesh Technical University, India. He worked as a
246 About the Authors
Research Assistant (1979–1980) at EE Department of MNR Engineering College

Allahabad, as Teaching Assistant (1980–1981) and Assistant Professor at EE
Department of GB Pant University; as a Lecturer (1986–1992) and Assistant
Professor at Institute of Engineering and Technology (IET) Lucknow (1992–2004)
where he became a full Professor in 2004. He has served as Head of the ECE
Department at IET Lucknow between 1986–1988 and 2007–2010 where currently
he is functioning as Dean of Research and Development since 2007. His teaching
and research interests are in the areas of Analog Integrated Circuits and Signal
Processing and he has published 16 research papers in various international Journals.
Index
A CMRR, 227, 231, 232

Active R biquad, 107–110, 126 Compensating capacitor, 28, 30, 42
Active-R oscillators, 139, 148, 151–152 Compensation, 25–28, 36, 38, 39
Active-R single resistance controlled Controlled source implementations
oscillators (SRCO), 148–152 (CCVS), 35
Amplifiers, 2, 7, 9–11, 17, 20, 25, 26, 28, Controlled sources, 2, 7, 30, 35
30–31, 33, 34, 45, 53, 72, 113, 211, 233 Current-controlled current conveyor (CCCC),
Analog circuit, 1–3 201, 211–213
design, 213 Current-controlled current feedback
Analog dividers, 191–192, 208 operational amplifiers (CFOA),
Analog filters, 110 232–233
Analog multipliers (AM), 164, 165, 168–173, Current-controlled-current-sources (CCCS), 35
175, 181, 207, 208, 237 Current conveyors, 11–14, 26, 62,
Analog signal processing, 1, 9, 12, 19 72, 81, 131, 132, 153, 189,
193, 201, 223, 232
Current difference buffered amplifiers
B (CDBA), 18, 201, 207–208
Biquads, 82–107, 109, 115–116, 118, 126, 205 Current differencing transconductance
Bootstrapping, 224, 225 amplifier (CDTA), 13, 19–20,
Building blocks, 236–238 201, 210, 211, 213
Current feedback, 26, 30
amplifiers, 153, 214, 233–234
C conveyor, 233
Cable deriver, 182 Current feedback operational amplifiers
Canonic single resistance controlled oscillators (CFOA), 1–3, 12–15, 20
(SRCO), 135–137, 157, 175 architecture, 224–229
Capacitive feedback, 32, 36 pole, 148–152
CCCS. See Current-controlled-current-sources Current followers, 32, 36, 37
(CCCS) Current follower transconductance amplifier
CCVS. See Controlled source implementations (CFTA), 201, 211, 213
(CCVS) Current mirror, 232
Chaotic oscillators, 193–197 Current mode (CM)
Chua diode, 194, 195 biquad, 83–107
Chua’s oscillator, 193–197 circuits, 11, 13, 152
Circuit building blocks, 9–20 oscillators, 156
248 Index
D H
DC precision, 31 Higher order filter, 55, 62, 63, 68–71
Deboo’s integrator, 38
Decoupling of gain and bandwidth, 30–31
Differential difference complementary current I
feedback amplifier (DDCCFA), IC implementation, 83, 94
235–237 Impedance converters, 55, 63
Differential input buffered transconductance Impedance inverters, 54, 55, 63
amplifier (DBTA), 201, 212, 213 Inductance simulation, 50, 52, 53, 61
Differential voltage Current feedback amplifier Instrumentation amplifier, 2, 6–8
(DVCFA), 233, 234 Integrated circuits, 15
Differentiators, 2 Integrators, 2
Double scroll attractor, 195 Inverse active filters, 110–112
Inverse transformation, 5
E
Electronically-gain variable amplifier, 181, 182 K
Explicit current output, 152–157, 175 Kerwin Huelsman Newcomb (KHN) biquad,
84, 85, 87, 97
F
Floating generalized impedance convertor L
(GIC), 64 LC Ladder prototype, 118–122, 125
Floating generalized impedance simulator, Linear voltage-controlled oscillators (VCOs),
60–65 161, 163–167, 173, 175
Floating impedances, 53, 54, 63–65, 74, 76, Lossless floating inductance simulator, 52, 62,
77, 204 65–68
Four terminal floating nullor (FTFN), 13, 17, Lossy floating inductance simulator, 65–68
201, 204–205, 213 Lossy grounded inductor, 56–60
Frequency dependent negative conductance, Low component count, 188
49, 67, 70
Frequency dependent negative
resistance, 49, 56–60, 65, M
70, 73, 74, 76, 77 Mixed mode, 195–197
Frequency stability, 132, 139, 143, 144, 160, biquad, 83, 107, 126
161, 167 Mixed translinear cell (MTC), 14
Full power bandwidth (FPBW), 9 Modified current feedback operational
Fully differential biquad, 115–116 amplifiers (CFOA), 232, 237
Fully-differential current feedback Modular filter structure, 119, 121–125
operational amplifiers (CFOA), Monostable, 206
229, 230 MOSFET-C filters, 16, 112–118, 126
Fully differential integrator, 113–115 MOSFET-C sinusoidal oscillator, 173–175
Fully uncoupled oscillator, 157, 158, 160 Multifunction biquad, 86, 100, 101
Multiphase oscillator, 151, 152
Multivibrators, 2, 205, 206
G
Gain bandwidth conflict, 7–8
Generalized positive impedance converters, 55 N
Generalized positive impedance inverters, 55, Negative capacitance, 56, 60, 61, 77
62, 63 Negative impedance inverter, 55
Grounded capacitors, 62, 64, 67–70, 132, Negative inductance, 56, 60, 61, 77
137–139, 144, 148, 156, 157, 171, Network transformation, 70
174–176 Nonlinearity cancellation, 73, 173
Gyrator, 52, 54–56 Norton amplifier, 16
Index 249
O State variable synthesis, 140–143, 161,

Offset voltage, 26, 31, 32, 47, 224–225 163–167
Op-amp circuits, 2–9 Stray capacitances, 32
Operational amplifiers, 2, 15, 20, 230
Operational floating amplifier (OFA), 204
Operational transconductance T
amplifier, 9–11 Third generation current conveyors, 201, 203
Operational transresistance amplifiers Threshold voltage, 183, 185, 187–189
(OTRA), 13, 15–16, 201, Transconductance, 201, 207, 208, 210–212
205–207, 213 Transfer characteristics, 30, 184–186, 189
Transimpedance, 187
Translinear circuits, 11
P Triangular wave generators, 185–188
Passive compensation, 89 Triggering recovery time, 206
Passive LCR filter, 103 Tunability, 87, 90, 93, 97, 100, 102, 103,
Positive impedance inverter, 55 109, 118
Precision rectifiers, 2, 156, 189–190, 237 Tuning laws, 143–145, 157
Pseudo exponential, 192–193
PSRR, 32
U
Unity gain cells, 208–210
Q Universal biquad, 81, 83, 92, 95, 97–107, 126
Quadrature oscillators, 147, 148, 174, 175
Quasi stable state, 206
V
VCCS. See Voltage-controlled-current-sources
R (VCCS)
Rail-to-Rail, 229 VCVS, 31–34, 47
Rectifier, 189–190 Very low frequency oscillator, 132, 143
Relaxation oscillator, 184, 185, 189 Voltage buffer, 14, 15, 20
Voltage-controlled-current-sources
(VCCS), 35
S Voltage controlled impedances, 71–77
Saturation, 188, 191 Voltage controlled oscillators (VCO), 71, 132,
Schmitt trigger, 2, 183–189 161–164, 166–173, 175
Second order filter, 68–69 Voltage controlled resistance (VCR), 137, 139,
Series CD, 57, 58, 60, 70 141, 159–161, 163–167
Series RL, 51, 53, 54, 57, 58, Voltage differencing differential input
67–70, 77 buffered amplifier (VD-DIBA),
Signal flow graph (SFG), 118, 119, 124, 126 201, 213
Simulated impedances, 68–71 Voltage follower, 8, 16
Simulated inductance, 56, 204, 205 Voltage mode
Single element controlled oscillator, biquad, 102
131–132, 143 circuits, 203
Single resistance controlled oscillators
(SRCO), 131, 134–138
Sinusoidal oscillators, 10, 131–176 W
Slew rate, 2, 8–9, 14, 15 Wave active filters, 119–122, 124
Squarer, 9 Wave equivalents, 122, 124, 125
Square rooter, 9, 198 Waveform generators, 181, 183–189
Square wave generators, 16, 186–189 Wien bridge oscillator (WBO), 131, 133–134
Squaring circuit, 190–191 Wilson current Mirror, 27
State variable methodology, 140, 142, 153, 175

2013 Book CurrentFeedbackOperationalAmpl PDF

Uploaded by

Copyright:

Available Formats

2013 Book CurrentFeedbackOperationalAmpl PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2013 Book CurrentFeedbackOperationalAmpl PDF

Uploaded by

Copyright:

Available Formats

ACSP · Analog Circuits and Signal Processing

For further volumes:

A.K. Singh V.K. Singh

ISBN 978-1-4614-5187-7 ISBN 978-1-4614-5188-4 (eBook)

# Springer Science+Business Media New York 2013

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

In spite of all electronic systems prominently being dominated by digital circuits

feedback operational amplifier (CFOA), sometimes also referred as operational

2.3.2 De-coupling of Gain and Bandwidth: Realisability

4 Design of Filters Using CFOAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

5.7.2 Low-Component-Count CFOA-Pole

7.2.4 CFOA Realization of Current Differencing

8.5 Future Directions of Research on CFOAs and

References for Additional Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

About the Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

DVCC+ Differential voltage second generation current conveyor

R. Senani et al., Current Feedback Operational Amplifiers and Their Applications, 1

CFOAs and the passive resistors realized by BJT-based translinear current-

1.2 An Overview of Analog Circuits and Their Applications

1.3 The Ubiquitous Op-Amp: The Drawbacks

1.3.1 Op-Amp Circuits Which Employ More

1.3.1.1 Voltage-Controlled-Current-Sources (VCCS)

Consider two well-known VCCS configurations shown in Fig. 1.1a, b.

whereas for circuit of the Fig. 1.1b it is:

Fig. 1.1 VCCS. (a) Non-inverting VCCS, (b) inverting VCCS

Fig. 1.2 CCCS. (a) inverting CCCS, (b) Non-inverting CCCS

Fig. 1.3 (a) Non-inverting integrator, (b) non-inverting differentiator

1.3.1.2 Non-inverting Integrator/Differentiator Using a Single Op-Amp

Figure 1.3a shows a non-inverting integrator popularly known as Deboo’s integrator

On the other hand, the transfer function of the non-inverting differentiator is

Fig. 1.4 Instrumentation R0 R0

while that of the differentiator is given by

1.3.1.3 Instrumentation Amplifier

The conventional instrumentation amplifier is another circuit which uses more

1.3.2 The Gain-Bandwidth Conflict

A major demerit of the various controlled source implementations (except CCVS)

where ωt ¼ A0ωp is the gain bandwidth product of the op-amp.

1.3.3 Slew-Rate Based Limitations

Another factor, which limits the application of VOA-based circuits in higher

As the output voltage peak amplitude increases, the maximum frequency at

1.4 A Brief Review of the Evolution of Alternative

1.4.1 The Operational Transconductance Amplifiers

Because of inability of the traditional op-amp-RC filters in making it possible to

Fig. 1.6 Symbolic notation of the OTA

The OTA-C circuits (also known as transconductance and capacitance (TAC)

Fig. 1.7 Simplified Schematic diagram of 3080 type IC OTA

Q15 Q11 Q12

Fig. 1.8 Simplified Schematic diagram of 13600/13700 type IC OTA

1.4.2 The Current Conveyors

Historically, the progress in the current-mode circuits/techniques can be visualized

The current-mode techniques, in spite of generating some controversy [36], have

Fig. 1.10 CCI+, CCI,

iy=ix; vx=vy; iz=ix iy=0; vx=vy; iz=ix

iy=ix; vx=vy; iz=−ix iy=0; vx=vy; iz=−ix

A CCI () is characterized by the hybrid matrix

while, a CCII () is characterized by

Fig. 1.11 A simplified form + V

In 1985, Fabre [37] and Normand [38] independently proposed a Current