Power Amplifiers For The S C X And Kubands An

Eda Perspective 1st Edition Mladen Boani

download

https://ebookbell.com/product/power-amplifiers-for-the-s-c-x-and-
kubands-an-eda-perspective-1st-edition-mladen-boani-5355642

Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.

A Highpower Cavity Amplifier For The New 900mhz Band Arrl Robert
Sutherland

https://ebookbell.com/product/a-highpower-cavity-amplifier-for-the-
new-900mhz-band-arrl-robert-sutherland-49544604

Distributed Power Amplifiers For Rf And Microwave Communications

Narendra Kumar

https://ebookbell.com/product/distributed-power-amplifiers-for-rf-and-
microwave-communications-narendra-kumar-5130172

Envelope Tracking Power Amplifiers For Wireless Communications

Zhancang Wang

https://ebookbell.com/product/envelope-tracking-power-amplifiers-for-
wireless-communications-zhancang-wang-5435408

Highefficiency Load Modulation Power Amplifiers For Wireless

Communications Zhancang Wang

https://ebookbell.com/product/highefficiency-load-modulation-power-
amplifiers-for-wireless-communications-zhancang-wang-11297678
Bandwidth And Efficiency Enhancement In Radio Frequency Power
Amplifiers For Wireless Transmitters Karun Rawat

https://ebookbell.com/product/bandwidth-and-efficiency-enhancement-in-
radio-frequency-power-amplifiers-for-wireless-transmitters-karun-
rawat-56900978

Reconfigurable Rf Power Amplifiers On Silicon For Wireless Handsets

1st Edition Laurent Leyssenne

https://ebookbell.com/product/reconfigurable-rf-power-amplifiers-on-
silicon-for-wireless-handsets-1st-edition-laurent-leyssenne-1898898

Advanced Design Techniques For Rf Power Amplifiers Analog Circuits And

Signal Processing 1st Edition Anna N Rudiakova

https://ebookbell.com/product/advanced-design-techniques-for-rf-power-
amplifiers-analog-circuits-and-signal-processing-1st-edition-anna-n-
rudiakova-2356380

Linearization And Efficiency Enhancement Techniques For Silicon Power

Amplifiers From Rf To Mmw 1st Edition Eric Kerherv

https://ebookbell.com/product/linearization-and-efficiency-
enhancement-techniques-for-silicon-power-amplifiers-from-rf-to-
mmw-1st-edition-eric-kerherv-5138366

Modeling And Design Techniques For Rf Power Amplifiers Arvind Raghavan

https://ebookbell.com/product/modeling-and-design-techniques-for-rf-
power-amplifiers-arvind-raghavan-1384210
Signals and Communication Technology

Mladen Božanić
Saurabh Sinha

Power
Amplifiers for
the S-, C-, X-
and Ku-bands
An EDA Perspective
Signals and Communication Technology
More information about this series at http://www.springer.com/series/4748
Mladen Božanić Saurabh Sinha
•

Power Amplifiers for the S-,

C-, X- and Ku-bands
An EDA Perspective

123
Mladen Božanić Saurabh Sinha
Engineering and Built Environment Engineering and Built Environment
University of Johannesburg University of Johannesburg
Johannesburg Johannesburg
South Africa South Africa

ISSN 1860-4862 ISSN 1860-4870 (electronic)

Signals and Communication Technology
ISBN 978-3-319-28375-3 ISBN 978-3-319-28376-0 (eBook)
DOI 10.1007/978-3-319-28376-0

Library of Congress Control Number: 2015958906

© Springer International Publishing Switzerland 2016

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.
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.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained
herein or for any errors or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by SpringerNature

The registered company is Springer International Publishing AG Switzerland
Preface

In the age where telecommunication has become a standard, almost every portable
device has some kind of transmitter and receiver allowing it to connect to a cellular
network or available Wi-Fi networks. We are also driving cars that are smarter and
equipped with new technologies, such as radars for collision detection. Other types
of radars are used in both civilian and military applications. Nowadays, we even
receive signals from satellites on our phones from Global Positioning Systems.
Radio frequency (RF) identification devices are becoming more and more common
and are being used in many applications, from access control to medical applica-
tions. In other words, the spectrum around us is full of transmitted signals waiting to
be received. Each signal is transmitted by some kind of power amplifier. As a result,
all researchers are likely to face the challenge of designing an RF or microwave
power amplifier at some stage of their careers.
Design of power amplifiers, however, is not an easy task. Even the great number
of power amplifier classes suggests that no single configuration is capable of
delivering acceptable performance for several frequency bands and for several
applications at once. Thus, the aim of this book is twofold. First, the idea is to
provide researchers with enough power amplifier theory to gain sufficient knowl-
edge to choose the best power amplifier stage for the specific application and to
understand the most important defining equations and parameters. Second, the
design equations to achieve this are very complex, and if they are used to design by
hand, they tend to put off researchers and designers. Thus, this book also aims to
provide its readers with some ideas on how to simplify the design process by
introducing their own software-based procedures or, in other words, by developing
their own electronic design automation (EDA). Although MATLAB is used
throughout the book to illustrate the concept of EDA (sometimes also termed
computer-aided design or CAD), the exact programming language is not important.
The accent is on how to identify what is needed as the end result of the power
amplifier design, and how to develop custom EDA to reach this result; essentially,
this book focuses on the methodology of power amplifier design.

v
vi Preface

This methodology is conceptualized so that it “trivializes” the approach to power

amplifier design by removing the “black magic” approach typically used in the
process. This advances research by allowing the readers to shift the focus of
research from power amplifiers onto other less-explored components of the system
or even on specification refinement. It can be used equally well by researchers
focusing on integrated design, or researchers focusing on discrete implementations,
which are typically used for high power or higher frequencies. The researchers
focusing on the thin line between integrated circuits and discrete implementations,
sometimes termed systems-on-package (SOP), are not excluded. Even though
different approaches sometimes need to be followed for design in different fre-
quency bands up to and including the Ku-band, similar principles of EDA apply.
This book is organized in two parts. Part I focuses on the main concepts of
power amplification, and this part can be used like any reference book. It first
presents a review of transmission bands and their implications for transceiver
system design. The feasibility of different passive component implementations in
each frequency range is investigated, and power amplifiers are placed into the
context of the transceiver system. The theory behind transistor operation at large
signal levels is included, and various semiconductor fabrication technologies are
discussed for full system integration or power transistor fabrication, together with
substrates for the implementation of discrete passives and SOP packaging. Other
basic aspects of communication theory, such as S-parameters, Y-parameters, Smith
charts, resonance, loaded quality factor (Q-factor), insertion loss impedance
transformation, and Fourier theory, are reviewed, which allows for better under-
standing of power amplifier concepts. This is followed by an in-depth analysis of
power amplifier stages. Most of the commonly used power amplification classes
(among others A, AB, C, J, D, E, E−1, F, and F−1) are discussed, and the defining
equations are included. Passives and their Q-factors are also covered in Part I. This
includes resistors, capacitors, integrated inductors, solenoids, toroids, RF chokes,
and transformers. Special focus is placed on both discrete and integrated inductor
designs, as they tend to exhibit low Q-factors and are therefore paramount to power
amplifier design. Micro-electro-mechanical systems are discussed as a promising
technology for the design of passives. Lumped and transmission line impedance
matching, which is important if the power amplifier is to be connected to the rest
of the transceiver system with minimum losses, is also discussed from an analytical,
graphical, and EDA perspective, for both real and complex sources and loads.
Part II of the book focuses on developing CAD procedures to aid practical power
amplifier design based on the theory reviewed in Part I. Although this book is not
intended for computer engineers, it is assumed that every RF researcher has some
basic programming skills. The automation and intelligent design ideas for inductor
design are presented in this part, which is followed by automation and intelligent
design ideas of various on- and off-chip power amplifier classes described in Part I.
Previously described design of passive elements and matching are incorporated into
this methodology. Real-life power amplifier design examples using the proposed
methodology are explored by means of examples, and developed algorithms
are illustrated both graphically and in MATLAB. IC layout and fabrication
Preface vii

are considered as one alternative for practical implementations, and discrete

implementations are considered as a second alternative. A practical aspect of
packaging of discrete systems is also explored. Other practical aspects that are not
necessarily covered by the EDA flow are also discussed in this part, and a formal
procedure for practical power amplifier design is presented. Part II of the book
concludes with the ideas for future research.
Acknowledgment

The authors would like to recognize the research-capacity grant of the Department
of Higher Education and Training, South Africa, for sponsoring the work covered
in this book. Furthermore, the authors would like to recognise Dr. Riëtte de Lange,
Postgraduate School, University of Johannesburg, South Africa, for her effective
administration of this grant.

ix
Contents

1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Power Amplifier as Part of a Transceiver System . . . . . . . . . . . 2
1.2 Active and Passive Devices for Power Amplifier Design . . . . . . 3
1.3 Classification of Power Amplifiers . . . . . . . . . . . . . . . . . . . . . 6
1.4 Basic Principles of Operation of Power Amplifiers . . . . . . . . . . 7
1.4.1 Power Amplifier Block Diagram . . . . . . . . . . . . . . . . 7
1.4.2 Output Power and Gain. . . . . . . . . . . . . . . . . . . . . . . 8
1.4.3 Power Consumption . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4.4 Power Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4.5 Output Power Capability . . . . . . . . . . . . . . . . . . . . . . 10
1.4.6 Maximum Operating Frequency of Power
Amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.4.7 Temperature Aspects of Power Amplifiers . . . . . . . . . . 11
1.4.8 Matching for Desired Power . . . . . . . . . . . . . . . . . . . 11
1.4.9 Biasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4.10 Conduction Angle . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4.11 Distortion, Linearization and Increase of Power
Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... 14
1.4.12 Impact of Power Amplifier Turn-on Characteristics. ... 15
1.4.13 Noise in Power Amplifiers. . . . . . . . . . . . . . . . . . ... 16
1.4.14 Measuring Large-Signal Power Amplifier
Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.4.15 Measuring Amplifier Power Gain and Stability. . . . . . . 17
1.5 Justification for Computer-Aided Design . . . . . . . . . . . . . . . . . 19
1.6 Organization of the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

xi
xii Contents

Part I Power Amplifier Theory

2 Review of Telecommunication Aspects for Power Amplifier
Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.1 Wavelength and Transmission Bands . . . . . . . . . . . . . . . . . . . 29
2.2 Review of Modulation Schemes . . . . . . . . . . . . . . . . . . . . . . . 31
2.2.1 Phase Shift-Keying . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.2.2 Frequency Shift-Keying . . . . . . . . . . . . . . . . . . . . . . 32
2.2.3 Phase-Amplitude Modulation . . . . . . . . . . . . . . . . . . . 32
2.2.4 Quadrature Amplitude Modulation . . . . . . . . . . . . . . . 34
2.2.5 On-Off Keying. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.2.6 Orthogonal Frequency-Division Multiplexing . . . . . . . . 35
2.3 Antennas and Propagation . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.4 The Power Transistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.4.1 Semiconductor Technologies for Transistor
Fabrication . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 37
2.4.2 Temperature Aspects of Transistors . . . . . . . . . ..... 39
2.4.3 Transistor Models and Large-Signal Transistor
Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.5 Substrates for Discrete Implementations . . . . . . . . . . . . . . . . . 44
2.6 The Smith Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2.7 Admittance (Y-) and Scattering (S-) Parameters . . . . . . . . . . . . 48
2.7.1 Y-Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.7.2 S-Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2.7.3 Conversion Between Y-Parameters
and S-Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
2.8 Resonant Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
2.8.1 Bandwidth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.8.2 Resonant Frequency . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.8.3 Quality Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
2.8.4 Component Quality Factor. . . . . . . . . . . . . . . . . . . . . 55
2.8.5 Insertion Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.8.6 Coupling of Resonant Circuits . . . . . . . . . . . . . . . . . . 56
2.9 Fourier Analysis of Periodic Signals . . . . . . . . . . . . . . . . . . . . 57
2.10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3 Continuous-Mode Power Amplifiers . . . . . . ....... . . . . . . . . . . 61
3.1 Class-A Power Amplifier. . . . . . . . . . . ....... . . . . . . . . . . 62
3.1.1 Current and Voltage Waveform Analysis. . . . . . . . . . . 64
3.1.2 Power and Efficiency . . . . . . . ....... . . . . . . . . . . 65
3.1.3 Bandwidth. . . . . . . . . . . . . . . ....... . . . . . . . . . . 68
3.1.4 RFC and Coupling Capacitor . . ....... . . . . . . . . . . 69
Contents xiii

3.2 Class-B Power Amplifier. . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

3.2.1 Current and Voltage Waveform Analysis. . . . . . . . . . . 69
3.2.2 Power and Efficiency . . . . . . . . . . . . . . . . . . . . . . . . 70
3.3 Class-AB and Class-C Power Amplifiers . . . . . . . . . . . . . . . . . 73
3.3.1 Current and Voltage Waveform Analysis. . . . . . . . . . . 73
3.3.2 Power and Efficiency . . . . . . . . . . . . . . . . . . . . . . . . 76
3.4 Class-A/AB/B/C Power Amplifier . . . . . . . . . . . . . . . . . . . . . 79
3.5 Push-Pull Class-A/AB/B/C Power Amplifier . . . . . . . . . . . . . . 80
3.5.1 Current and Voltage Waveform Analysis. . . . . . . . . . . 80
3.5.2 Power and Efficiency . . . . . . . . . . . . . . . . . . . . . . . . 82
3.6 Class-J Power Amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
3.6.1 Current and Voltage Waveform Analysis. . . . . . . . . . . 84
3.6.2 Power and Efficiency . . . . . . . . . . . . . . . . . . . . . . . . 85
3.6.3 Calculating XL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3.7 Doherty Power Amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.8 Recent State-of-the-Art Examples . . . . . . . . . . . . . . . . . . . . . . 90
3.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4 Switch-Mode Power Amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . .. 93
4.1 Class-D Power Amplifier. . . . . . . . . . . . . . . . . . . . . . . . . . .. 94
4.1.1 Class-D Complementary Push-Pull
Voltage-Switching Power Amplifier . . . . . . . . . . . . .. 94
4.1.2 Class-D Complementary Push-Pull Current-Switching
Power Amplifier. . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.2 Class-E Power Amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.2.1 Class-E Zero-Voltage-Switching Power Amplifier. . . . . 104
4.2.2 Class-E Zero-Current-Switching Power Amplifier . . . . . 111
4.2.3 Class-E ZVS Power Amplifier with Finite
DC-Feed Inductance . . . . . . . . . . . . . . . . . . . . . . . . . 116
4.3 Class-F Power Amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
4.3.1 Maximally Flat Class-F3 Power Amplifier . . . . . . . . . . 119
4.3.2 Maximally Flat Class-F35 Power Amplifier . . . . . . . . . 124
4.3.3 Maximally Flat Class-F2 and Maximally Flat
Class-F24 Power Amplifier (Class-F−1 Power
Amplifiers) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
4.3.4 Class-F Power Amplifier with Quarter-Wavelength
Transmission Line . . . . . . . . . . . . . . . . . . . . . . . . . . 130
4.4 Other Power Amplifier Classes . . . . . . . . . . . . . . . . . . . . . . . 133
4.5 Recent State-of-the-Art Examples . . . . . . . . . . . . . . . . . . . . . . 133
4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
xiv Contents

5 Passives for Power Amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

5.1 Resistors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
5.2 Capacitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
5.2.1 Discrete Capacitors. . . . . . . . . . . . . . . . . . . . . . . . . . 140
5.2.2 Integrated Capacitors . . . . . . . . . . . . . . . . . . . . . . . . 141
5.3 Inductors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
5.3.1 Discrete Inductors . . . . . . . . . . . . . . . . . . . . . . . . . . 144
5.3.2 Integrated Active Inductors . . . . . . . . . . . . . . . . . . . . 148
5.3.3 Bond Wires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
5.3.4 Spiral Inductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
5.3.5 MEMS Inductors . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
5.3.6 Other On-Chip Inductor Implementations . . . . . . . . . . 161
5.4 RF Chokes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
5.5 Transformers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
5.6 Quarter-Wavelength Transformer . . . . . . . . . . . . . . . . . . . . . . 162
5.6.1 General Transmission Line . . . . . . . . . . . . . . . . . . . . 164
5.6.2 Input Impedance of the Quarter-Wave Transformer. . . . 166
5.6.3 Bandwidth of the Quarter-Wave Transformer . . . . . . . . 166
5.6.4 Impedance of the Quarter-Wave Transformer
Terminated with a Resonant Tank . . . . . . . . . . . . . . . 168
5.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
6 Impedance Matching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
6.1 Importance of Impedance Matching . . . . . . . . . . . . . . . . . . . . 173
6.2 Load-Pull Characterization. . . . . . . . . . . . . . . . . . . . . . . . . . . 174
6.3 Lumped-Element Matching . . . . . . . . . . . . . . . . . . . . . . . . . . 176
6.3.1 Wideband Two-Element Networks (L-Networks) . . . . . 177
6.3.2 Narrowband Three-Element Networks
(T- and Π-Networks) . . . . . . . . . . . . . . . . . . . . . . . . 180
6.4 Lumped Matching Using Smith Charts . . . . . . . . . . . . . . . . . . 183
6.5 Transmission-Line Impedance Matching . . . . . . . . . . . . . . . . . 190
6.5.1 Variations of Transmission-Line Matching
Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
6.5.2 Quarter-Wave Transformer Impedance Matching . . . . . 192
6.6 Impedance Matching Using MATLAB . . . . . . . . . . . . . . . . . . 193
6.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194

Part II Developing EDA for Power Amplifier Design

7 Intelligent Automated Design Ideas for Inductor Synthesis. . . . . . . 199
7.1 Design of Integrated Inductors . . . . . . . . . . ........ . . . . . . 199
7.1.1 Cut-and-Try Approach . . . . . . . . . ........ . . . . . . 200
7.1.2 Synthesis Approach . . . . . . . . . . . ........ . . . . . . 201
Contents xv

7.1.3 MEMS Inductors . . . . . . . . . . . . . . . . . . . . . . . . . . . 214

7.1.4 Integrated Transformers. . . . . . . . . . . . . . . . . . . . . . . 218
7.1.5 Verification of the Inductor Model and the Search
Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
7.2 Bond Wires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
7.2.1 Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
7.2.2 Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
7.2.3 Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
7.2.4 Development of the Inductor Design Routine. . . . . . . . 226
7.2.5 Design Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
7.3 Discrete Inductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
7.4 Design Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
7.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
8 Full Power Amplifier System Design . . . . . . . . . . . . . . . . . . . . . . . 235
8.1 Subroutine for Design of Class-E Zero-Voltage-Switching
Power Amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
8.1.1 Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
8.1.2 Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
8.1.3 Description of the Subroutine. . . . . . . . . . . . . . . . . . . 238
8.1.4 Subroutine Outputs. . . . . . . . . . . . . . . . . . . . . . . . . . 238
8.2 Subroutine for Design of the Class-F Power Amplifiers. . . . . . . 239
8.2.1 Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
8.2.2 Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
8.2.3 Description of the Subroutine. . . . . . . . . . . . . . . . . . . 244
8.2.4 Subroutine Outputs. . . . . . . . . . . . . . . . . . . . . . . . . . 244
8.3 Subroutine for Design of Class-F Power Amplifiers
with Quarter-Wave Transmission Line . . . . . . . . . . . . . . . . . . 244
8.3.1 Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
8.3.2 Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
8.3.3 Description of the Subroutine. . . . . . . . . . . . . . . . . . . 248
8.3.4 Subroutine Outputs. . . . . . . . . . . . . . . . . . . . . . . . . . 249
8.4 Subroutine for Design of Class-A/AB/B/C Power
Amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
8.4.1 Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
8.4.2 Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
8.4.3 Description of the Subroutine. . . . . . . . . . . . . . . . . . . 253
8.4.4 Subroutine Outputs. . . . . . . . . . . . . . . . . . . . . . . . . . 254
8.5 Subroutine for Design of the Class-J Power Amplifiers . . . . . . . 254
8.5.1 Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
8.5.2 Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
8.5.3 Description of the Subroutine. . . . . . . . . . . . . . . . . . . 258
8.5.4 Subroutine Outputs. . . . . . . . . . . . . . . . . . . . . . . . . . 258
xvi Contents

8.6 Subroutine for Impedance Matching . . . . . . . . . . . . . . . . . . . . 260

8.6.1 Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
8.6.2 Subroutine Outputs. . . . . . . . . . . . . . . . . . . . . . . . . . 261
8.6.3 Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
8.6.4 Description of the Subroutine. . . . . . . . . . . . . . . . . . . 261
8.7 Complete System Integration . . . . . . . . . . . . . . . . . . . . . . . . . 263
8.7.1 Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
8.7.2 Routine Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
8.7.3 Description and Flow Diagram of the Power
Amplifier Design Routine . . . . . . . . . . . . . . . . . . . . . 268
8.7.4 MATLAB Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
8.8 Design Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
8.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
9 Practical Considerations of Integrated and Discrete Power
Amplifier Solutions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
9.1 Practical Considerations Common to Integrated, Discrete
and Hybrid (System-on-Package) Solutions . . . . . . . . . . . . . . . 309
9.1.1 Gain of the Power Transistor . . . . . . . . . . . . . . . . . . . 309
9.1.2 Component Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
9.1.3 Feasibility of Component Values . . . . . . . . . . . . . . . . 311
9.1.4 Influence of the Frequency and the Wavelength . . . . . . 312
9.1.5 Coupling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
9.2 Integrated Circuit Considerations . . . . . . . . . . . . . . . . . . . . . . 313
9.2.1 Design Rule Checks and Technology
Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
9.2.2 Extraction of a Spiral Inductor Layout
into a GDSII File . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
9.2.3 Bond-Pad Considerations. . . . . . . . . . . . . . . . . . . . . . 314
9.2.4 Bond-Wire Considerations . . . . . . . . . . . . . . . . . . . . . 314
9.2.5 Package Lead Considerations . . . . . . . . . . . . . . . . . . . 319
9.3 Systems-on-Package Considerations . . . . . . . . . . . . . . . . . . . . 321
9.4 From Theoretical Design Using Custom EDA to Practical
Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
9.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328
10 Future Directions and Final Remarks . . . . . . . . . . . . . . . . . . . . . . 331
10.1 Power Amplifiers Utilizing Transmission Lines . . . . . . . . . . . . 331
10.1.1 Class-E Power Amplifier Utilizing Transmission
Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
10.1.2 Class-F Power Amplifier Utilizing Transmission
Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
Contents xvii

10.2 Near-Terahertz Frequencies . . . . . . . . . . . . . . . . . . . . . . . . . . 335

10.3 Other Design Automation Ideas and Ideas for Expansion
of Devised Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
10.4 Final Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
About the Authors

Mladen Božanić SMIEE, obtained his B.Eng. (with distinction), B.Eng. (Hons)
(with distinction) and Ph.D. degrees in Electronic Engineering from the University
of Pretoria (UP) in 2006, 2008 and 2011 respectively. In 2008, he joined Azoteq, a
fabless IC design company originating in South Africa where he was responsible
for the silicon-level design, simulation characterization design for testability
(DFT) of various analog, RF, digital and mixed-mode circuits. While actively
working in the industry, he also participates in research activities, currently with the
University of Johannesburg (UJ) where he is serving as a Senior Research Fellow.
Since 2011, Dr. Božanić has been fulfilling the role of a Specialist Editor of the
South African Institute of Electrical Engineers (SAIEE). He is a recipient of
SAMES Award and CEFIM Fellowship Award, and an author or co-author of over
10 peer-reviewed journal and conference articles, one book chapter and one book.
Saurabh Sinha SMIEEE, FSAIEE, FSAAE, obtained his B.Eng., M.Eng. and
Ph.D. degrees in Electronic Engineering from the University of Pretoria (UP),
South Africa. He achieved both his B.Eng. and M.Eng. with distinction. As a
published researcher, he has authored or co-authored over 85 publications in
peer-reviewed journals and at international conferences. In addition, he is the
managing editor of the South African Institute of Electrical Engineers (SAIEE)
Africa Research Journal. Prof. Sinha served the UP for over a decade, his last
service being as Director of the Carl and Emily Fuchs Institute for Microelectronics,
Department of Electrical, Electronic and Computer Engineering. On 1 October
2013, Prof. Sinha was appointed Executive Dean of the Faculty of Engineering and
the Built Environment (FEBE) at the University of Johannesburg (UJ). Professor
Saurabh Sinha is the 2014–2015 Vice-President, IEEE Educational Activities and
serves on the IEEE Board of Directors.

xix
Chapter 1
Introduction

In today’s communication age, almost every portable device has some sort of
transmitter—be it a radio for third generation (3G), long-term evolution (LTE) or
Worldwide Interoperability for Microwave Access (WiMAX) networks, Bluetooth
or WiFi [1–4]. We are driving cars that are smarter, equipped with new tech-
nologies, such as radars for collision detection. Other types of radars are used in
both civilian and military applications. We receive signals from satellites in our cars
and on our phones from global positioning systems (GPS), as well as at home
(satellite TV receivers). Radio-frequency identification (RFID) devices are
becoming more and more common and are finding use even in medical applications
[5, 6].
Essentially, the spectrum around us is full of transmitted signals waiting to be
received. Each signal was transmitted by some sort of power amplifier (sometimes
abbreviated as PA). Thus, every circuit designer is likely, sooner or later in his or
her career, to face the challenge of designing a radio-frequency (RF) or microwave
(sometimes abbreviated MW) transmitter, and inherently, a power amplifier for one
of the following bands: S-, C-, X- or Ku-bands [7] operating at ultra-high fre-
quencies (UHF) and super-high frequencies (SHF) from 2 to 18 GHz.
The great number of power amplifier types (termed classes) suggests that no
single configuration is capable of delivering acceptable performance in all fre-
quency bands and for all applications. One of the aims of this book is to provide its
readers with enough power amplifier theory to gain sufficient knowledge to choose
the best power amplifier stage for the specific application and to understand the
most important defining equations and parameters. The power amplifier increases
the power level of the input signal, resulting in a signal with a higher output power
level. Therefore an important focus of power amplifiers is output power as well as
power gain. The design equations and process to achieve this are very complex and
if they are used to design by hand, they tend to almost frighten the designers. Thus
another aim of the book is to provide readers with some ideas on how to simplify
the design process by introducing software-based routines in a programming lan-
guage of their choice, and provide enough examples to make this task easier.
The software-aided methodologies presented this book are conceptionalized so
that they can be used equally well for designing increasingly popular integrated
circuits (ICs), or well-established discrete implementations (typically used for high
© Springer International Publishing Switzerland 2016 1
M. Božanić and S. Sinha, Power Amplifiers for the S-, C-, X- and Ku-bands,
Signals and Communication Technology, DOI 10.1007/978-3-319-28376-0_1
2 1 Introduction

power), or combinations of these two (e.g. integrated power amplifier with

matching or any other components off-chip). Although different approaches need to
be followed for design in different bands (albeit without an exactly defined border
where one approach stops and the next one begins) up to and including the
Ku-band, similar principles of computer-aided design (CAD) and electronic design
automation (EDA) will apply.
This introductory chapter touches on basic principles of signal transmission and
reception, active and passive devices, basic operation of power amplifiers and their
classification, basic design parameters and characterization. It also presents the
justification for the use of a computer-aided approach in power amplifier design.
The purpose of this chapter is to touch on the basic topics of this book in order to
interest the reader in later chapters.

1.1 Power Amplifier as Part of a Transceiver System

A typical transceiver system consists of a transmitter and receiver [8], as shown in

Fig. 1.1. In order for a signal to be transmitted over a channel, the signal is first
processed. This usually means that some sort of digital encoding is performed. The
signal is thereafter modulated onto a carrier frequency using one of the modulation
schemes suitable for a particular band of operation, for example phase-shift keying
(PSK), quadrature PSK (QPSK), direct sequence spread spectrum (DSSS),
quadrature amplitude modulation (QAM), on-off keying (OOK) or orthogonal
frequency division multiplexing (OFDM) [9–13]. The code-division multiple

Transmitter Power Antenna

Modulation
amplification
Input
signal Signal
processing

Channel
fc

Recovered Demodulation
signal Signal
processing Antenna
fc

Signal
detection
Receiver

Fig. 1.1 Power amplifier as part of a simple telecommunication system [8]

1.1 Power Amplifier as Part of a Transceiver System 3

access (CDMA) technique, a type of channel access method where several trans-
mitters can send information simultaneously, is often confused with some modu-
lation schemes.
The first two stages will set the correct signal voltage levels, carrier frequency and
bandwidth. However, this signal is still unsuitable for transmission. The amount of
power needs to be increased in order to drive the antenna. Power amplification is
normally the third stage of the transmitter and last before the antenna, through which
the power amplifier converts the direct current (DC) input power from the supply
rails into a significant amount of RF or microwave power [2].
On the receiver side, a similar process occurs, but in reverse order. First, the
signal is detected from a channel using a low-noise amplifier (LNA) [14]. A carrier
recovery scheme may be employed. Thereafter, the signal is demodulated, and
reverse signal processing and filtering are used to reproduce the original signal.
The power amplifier, marked bold in Fig. 1.1, needs to deliver high efficiency,
high linearity, high power gain and large dynamic range simultaneously [15].
Consequently, it consumes the largest amount of DC power. The increasing
demand for a higher data rate and increasing modulation complexity, comple-
mented by the need to keep the transmitter costs low, calls for innovative art of
transmitter design [16, 17]. The power amplifier therefore remains a bottleneck in
the design of wireless transceivers.
Inclusion of a power amplifier is a particular problem in integrated devices,
especially if integration is done in pure silicon complementary metal-oxide semi-
conductor (CMOS) processes, mainly owing to the amount of power that needs to
be generated on chip and the size of passive components. For this reason, most
commercial wireless devices use an external power amplifier using discrete com-
ponents to drive an antenna. The driving transistor device is usually fabricated in
semiconductor technologies superior to the silicon (Si) CMOS, such as
silicon-germanium (SiGe), gallium-arsenide (GaAs), indium-phosphate (InP),
gallium-nitride (GaN), silicon carbide (SiC) and others [18–24]. However, in dis-
crete power amplifier implementations, other factors can introduce limitations, such
as the design and material of the printed-circuit board (PCB) used, and insertion
loss of lumped devices and discrete designs are not straightforward either.

1.2 Active and Passive Devices for Power Amplifier Design

A basic power amplifier is designed around a minimum of one active device. This
could be metal-oxide semiconductor field-effect transistors (MOSFETs), bipolar
junction transistors (BJTs), heterojunction bipolar transistors (HBTs),
high-electron-mobility transistors (HEMTS), or another type. Vacuum-tube power
amplifiers are still used [9].
4 1 Introduction

MOSFET devices have generally been considered less suitable for the power
amplification task because they require more current to achieve the same amount of
power amplification than their bipolar counterparts (HBTs) [25], but this difference
is becoming smaller as superior MOSFET technologies emerge. Apart from dif-
ferences based on fundamental device properties, transistors (both integrated and
discrete) will yield different performance when fabricated in different technologies.
Several factors, including the transistor transition frequency fT (frequency at which
transistor gain-bandwidth product becomes zero), the breakdown voltage of the
transistor and the driving capability of the transistor, need to be taken into con-
sideration when choosing the best technology or a transistor for power amplifier
implementation [26]. It is worthwhile noting that the performance of active devices
is severely affected by the trend in device scaling, but technologies capable of
reaching even mm-wave frequencies have been reported [19]. There are several
figures of merit that can be used to quantify the suitability of semiconductor
material for power transistor fabrication, and Johnson’s figure of merit (JFOM) and
Baliga’s figure of merit (BFOM) will be mentioned later in this book [27, 28].
For illustration purposes, two power transistors (a layout of an HBT transistor in
IBM 7WL technology and a photograph of a power Darlington pair) are shown in
Fig. 1.2. Active devices will be discussed in more detail in Chap. 2.
Additional to the active devices, a power amplifier contains a number of passive
components, such as inductors and capacitors used for filtering and matching. Other
passive components, including among others transformers that are used for power
combining [9], and transmission lines can also be found in the power amplifier.
At RF, designing with ideal devices seldom generates good results even on the first
design iteration. Real devices and their parasitic effects need to be considered. The

Fig. 1.2 An example of the HBT transistor layout for integrated power amplifier implementations
(a), and a photograph of a Darlington power transistor for discrete power amplifier implemen-
tations (b)
1.2 Active and Passive Devices for Power Amplifier Design 5

greater the frequency, the more difficult it is to find a device with the expected
performance. This principle applies particularly to inductors, which tend to have
inferior performance, both on- and off-chip. Here, the substrate on which the
passive component is fabricated plays a major role, and a quality factor (Q-factor) is
used as a measure of quality. Instead of using lumped passives, passives imple-
mented using transmission lines (i.e. open and short-circuited stubs) can be used at
RF, but they are mostly practical off-chip. At mm-wave frequencies, transmission
lines can be used on-chip; however, mm-wave frequencies are not one of the main
focuses of this book but they will be discussed in Chap. 10 when dealing with
future directions.
Figure 1.3 shows a photograph of a wire-wound inductor, an integrated spiral
octagonal inductor in IBM 7WL technology and a 2:1 wire-wound transformer.

Fig. 1.3 Photographs of different inductors and transformers: a wire-wound inductor (a), an
integrated spiral inductor (b), and a 2:1 transformer (c)
6 1 Introduction

1.3 Classification of Power Amplifiers

Several groupings of power amplifiers are possible and all groupings are used
interchangeably. Power amplifiers are commonly grouped into broadband and
narrowband amplifiers. Sometimes, they are grouped depending on whether they
are intended for linear or constant envelope operation [29]. Finally, the most
common grouping of power amplifiers is grouping into classes according to the
nature of their voltage and current waveforms. The variety of power amplifier
classes reflects the inability of any single circuit to satisfy stringent requirements for
linearity, power gain, output power and efficiency, all described later in this chapter.
A letter or combinations of letters of the alphabet are used to define different
power amplifier classes. This classification is based on the shape of the voltage (vD)
and current (iD) waveforms of the driving transistor. The following classes are
commonly used for different applications:
• Classes A, B, AB and C are classes exhibiting continuous mode of operation
(i.e. the driving transistor is always on) [13, 30].
• Classes D, DE, E, F, FE, G, H, J and S [7, 8, 31] are switch-mode classes (i.e.
the driving transistor functions as a switch).
Inverse classes, where the shape of voltage and current waveforms across the
power transistors are swapped around, are also possible. Common examples are
inverse Class-C (C−1), inverse Class-E (E−1) and inverse Class-F (F−1) amplifiers
[32, 33]. Most of the real-life power amplifiers operate with current and voltage
waveforms that lie between two different classes. If more than one power amplifier
of different classes are combined in parallel to cater for different modes of operation
of the transmitter (usually one main and one peaking), a Doherty power amplifier is
created [34]. Common combinations of Doherty amplifiers are a Class-AB or
Class-B amplifier combined with a Class-C amplifier and a Class-F amplifier
combined with another Class-F, Class-F−1 or Class-C amplifier.
Not all the classes are suitable for design all the way up to the Ku-band. For
example, Class-D amplifiers are the switching-mode power amplifiers generally
used in low-frequency applications (e.g. audio) [9, 35, 36], and the use of this class
of power amplifier at high frequencies is limited by prominent parasitic reactances
that lead to substantial losses. However, they can be considered at higher fre-
quencies when operating in the current mode [35]. Class-G and Class-H amplifiers
are also commonly used for audio applications, with some limited use in digital
telephony and CDMA at low megahertz frequencies, not applicable to the topic of
this book.
Traditionally, power amplification at RF and microwave was done with amplifier
classes A to C, often termed classic amplifiers [7]. These classes (with exception of
Class C) generally have high linearity but suffer from low efficiencies. Class-E,
1.3 Classification of Power Amplifiers 7

Class-F amplifiers and other switchmode classes are considered modern amplifiers,
since they can be used in many high-end applications. They suffer from low lin-
earity, but their efficiencies can reach 100 % in theory.
Because of their importance, all amplifier classes mentioned will be presented in
separate sections in this book.

1.4 Basic Principles of Operation of Power Amplifiers

1.4.1 Power Amplifier Block Diagram

Figure 1.4 shows a block diagram general single-ended power amplifier [9]. In this
model, VDD is the voltage supply, RL is the load, RFC is the RF choke—ideally an
inductor with infinite reactance and zero series resistance. RFC is large enough to
ensure the substantially constant current through the drain. In some designs, RFC
can be replaced by a finite inductor, if the output filter can be designed to resonate
with it. The output filter mentioned is also shown in this figure [37]. It can include
harmonic tuning and wave shaping, impedance matching or any other passive
circuitry. The transistor T1 is shown as an n-channel MOS (NMOS) transistor, but it
can be any power transistor (MOS, HBT, BJT, HEMT or other) used in a particular
power amplifier application.
Note that throughout this book, terms for device terminals associated with MOS
transistors (gate, source, drain) and terms for device terminals of BJTs (such as
base, collector, emitter) are used interchangeably.

Fig. 1.4 General model of a VDD

power amplifier [9]
IDC

RFC

T1 Output filter iO

+ +
Drive and bias
vD vO RL

_ _
8 1 Introduction

1.4.2 Output Power and Gain

The task of a power amplifier is to deliver a given power into the load [8]. This
power is determined by the power supply voltage VDD and the load RL. The
maximum power that can be delivered is
2
VDD
P¼ : ð1:1Þ
2RL

Equation (1.1) is applicable to sinusoidal waveforms. Depending on the shape of

waveforms generated for a specific power amplifier stage, it may be possible to
deliver more power to the same size load. Integrated power amplifiers are generally
designed for low values of RL.
Power gain is defined as a ratio of output power to input power:

Pout
G¼ : ð1:2Þ
Pin

1.4.3 Power Consumption

Total DC power consumption is an important quantity in a power amplifier design,

especially for battery-powered portable devices. The DC input power of a power
amplifier is the current drawn from the voltage supply over a period of time, or

ZT ZT
1 VDD
Pdc ¼ VDD iD dt ¼ iD dt ¼ VDD IDC ; ð1:3Þ
T T
0 0

where IDC is the DC component of the current waveform.

1.4.4 Power Efficiency

Efficiency is a measure of performance of a power amplifier. The performance of a

power amplifier will be better if its efficiency is higher, irrespective of its definition.
Several definitions of efficiency are commonly used with power amplifiers [7, 9,
13]. In order to define the efficiency, the RF output power needs to be defined first.
Assuming sinusoidal voltage and current, the RF output power is given by
1.4 Basic Principles of Operation of Power Amplifiers 9

i1 v1 i21 RL
Pout ¼ veff ieff ¼ ¼ : ð1:4Þ
2 2

Drain (or collector) efficiency η is defined as the ratio of RF output power (Pout)
to DC input power (Pdc), or

Pout
g¼ ; ð1:5Þ
Pdc

where ieff and veff are effective and i1 and v1 are the peak fundamental components of
current and voltage respectively, and the DC input power is given by Eq. (1.3).
Power added efficiency (PAE) takes into account the input power (Pin) by
subtracting it from the output power:
 
Pout  Pin Pout  PGout 1
PAE ¼ ¼ ¼g 1 : ð1:6Þ
Pdc Pdc G

The PAE will give a good indication of the performance of a power amplifier for
high amplifier gains but it can even become negative for low gains. This rela-
tionship is shown in Fig. 1.5. As a consequence, power amplifiers achieve their
greatest efficiency only in saturated operation [10] and during non-saturated or idle
cycles, a large amount of power is wasted. This is why many commercial devices
use Doherty power amplifiers, with different driving stages activated depending on
the intensity of the transmitted signal [38].
Overall efficiency is the ratio of output power to the sum of input power and DC
input power:

Pout
OE ¼ : ð1:7Þ
Pdc þ Pin

Fig. 1.5 Normalized PAE 1

versus power amplifier gain
[8] 0.8

0.6
PAE

0.4

0.2

0
4 8 12 16 20
Gain (dB)
10 1 Introduction

Instead of instantaneous efficiencies defined above, average efficiency may be

more applicable to signals with time-varying amplitudes:

Pout
gAVG ¼ : ð1:8Þ
Pdc

To achieve maximum drain efficiency, no power can be dissipated over the drain
(collector) of the driving transistor [7]. Thus, the power in the drain must be zero, or

ZT
1
PD ¼ vD iD dt ¼ 0: ð1:9Þ
T
0

This implies that the product of the current and voltage needs to be zero at any
moment:

iD vD ¼ 0: ð1:10Þ

Therefore, for 100 % efficiency, at any time, either iD = 0 or vD = 0, or in other

words, the ideal power amplifier, must have non-overlapping current and voltage
waveforms.
Efficiencies in practical power amplifier implementations are generally much
lower than ideal values calculated for each stage because of a number of factors:
low quality factor of passives, saturation voltage in the transistors and transistor
parasitics, tuning errors and temperature variations.

1.4.5 Output Power Capability

Output power capability is defined as ratio of the maximum power delivered to the
load and the product of maximum values of iD and vD:

PoutðmaxÞ
cp ¼ : ð1:11Þ
IDMðmaxÞ VDSðmaxÞ

1.4.6 Maximum Operating Frequency of Power Amplifiers

Another limiting factor in power amplifier design is the maximum operating fre-
quency for a predetermined power and supply voltage. It is dependent on the
transistor output capacitance COUT, and for a Class-E power amplifier it can be
expressed as [39]:
1.4 Basic Principles of Operation of Power Amplifiers 11

1 Pout
fMAX ¼  2
: ð1:12Þ
2p2 COUT VDD

This relation shows that the greater the amount of power that needs to be
delivered, the more limiting the driving transistor will be in reaching higher
frequencies.

1.4.7 Temperature Aspects of Power Amplifiers

As discussed in previous sections, efficiency is the ability of the power amplifier to

convert the electrical energy into output power. Excess power is converted to heat,
which can limit the performance of an amplifier [40]. With increased efficiency, the
amount of heat generated decreases. However, even with high efficiency, high
output power configurations [41] will dissipate significant amounts of heat. The
amount of heat generated and the way in which that heat is dissipated in a power
amplifier system depends on the technology in which the IC is fabricated or type of
a substrate used for discrete implementation, as well as on the type of active device
(transistor) used for power amplification.
Typically, if heating poses a problem, the excess heat can be removed by means
of heat sinks, heat slugs or heat spreaders [42–44] Any of the three mentioned
components is basically a piece of metal used to dissipate heat away from the chip
better than the substrate or the package is able to. Typically, they are made of
aluminum or copper.

1.4.8 Matching for Desired Power

A power amplifier needs to be inserted between the modulator and the antenna with
minimum insertion loss. This calls for careful impedance matching.
Figure 1.6 shows a block diagram of a power amplifier illustrating matching on
the input and output side. At the input side of the power amplifier, care needs to be
taken so that the correct current and voltage waveforms are delivered at the gate or

ZS
Input matching Output matching
VS Power amplifier ZL
network network

Source Antenna

Fig. 1.6 Block diagram of a power amplifier showing input and output matching networks
12 1 Introduction

base of the transistor to achieve a particular class of operation, thus matching needs
to be performed simultaneously with biasing described later.
On the other hand, at the output side load has to be chosen correctly. Searching
for the optimum impedance for maximum power output, PAE and gain for the
power amplifier is usually achieved using load pull.
From Eq. (1.1), it is obvious that the only two parameters influencing the output
power are the voltage supply, VDD or VCC, and the load impedance, RL. The supply
is normally fixed for a given application, so that the only degree of freedom left to
the designer is the impedance of the load. This impedance will often differ from
standard impedances of 50 or 75 Ω, and in IC impedances of less than 10 Ω are not
uncommon. Impedance matching networks are used to convert standard impe-
dances to required load impedances as defined by amplifier design equations or
obtained by load pull. At mm-wave frequencies, where wavelengths are corre-
spondingly small, this matching can be accomplished with transmission lines [45].
At UHF and SHF, the transmission lines are impractically long to be used on a chip
but they can be implemented on a PCB. Matching using discrete or integrated
passive components can be deployed both on- and off-chip, provided that suffi-
ciently high-Q circuit elements of required value can be achieved at the matching
frequency.
Two-component networks (L networks) and three-component networks (T and
Π networks) are commonly used. Eight L-network configurations are possible, as
shown in Fig. 1.7a, b, where X1 and X2 can be any combination of inductors and
capacitors, ZS is the source impedance and ZL is the load impedance. Such an L
network is a broadband (high-pass or low-pass) network. Conversely, the T and Π
networks with passives X1, X2 and X3, shown in Fig. 1.8a, b, are narrowband
networks.

1.4.9 Biasing

Input impedance matching, mentioned in the previous section, is used to ensure that
correct amplitudes of alternating current (AC) signals appear at input. Biasing,

(a) (b)

X1 X2
ZS X2 ZL ZS X1 ZL

Fig. 1.7 Two-component matching networks where passive component is parallel to a load and
b source [45]
1.4 Basic Principles of Operation of Power Amplifiers 13

(a) (b)

X1 X2 X1
ZS X2 ZL ZS X2 X3 ZL

Fig. 1.8 Three-component matching networks: a T network and b Π network [45]

described in this section, provides the appropriate quiescent point for the power
amplifier [45].
The biasing point should remain constant irrespective of transistor parameter
variations or temperature fluctuations. Active and passive biasing networks are
possible. Figure 1.9 shows one-resistor and three-resistor biasing networks com-
monly used with BJT power amplifiers.
Adaptive bias techniques can be used with power amplifiers in order to avoid
too-large or too-small current, as well as to improve linearity and efficiency [13].

Fig. 1.9 Passive biasing (a) VCC

networks for a BJT power
amplifier: a one-resistor
configuration, b three-resistor
configuration [45] RFC
R
RFout
RFC

CB

RFin

VCC
(b)

RFC

R2 RFout
RFC

R1 R3
CB

RFin
14 1 Introduction

1.4.10 Conduction Angle

Conduction angle, 2θ, is another important parameter of power amplifiers. It rep-

resents a portion of the cycle during which an amplifier is conducting current. In
Class A, AB, B and C amplifiers, it directly determines the shape of current and
voltage waveforms and thus it is commonly used to differentiate between Class A,
AB, B and C amplifiers. In switch-mode power amplifiers (e.g. Class E or F), the
conduction angle is limited to describing the biasing level at which the input
waveform is applied to the active device. Thus, a Class-E power amplifier can be
biased at Class-AB, B or C level based on the portion of the cycle during which the
switch is activated.

1.4.11 Distortion, Linearization and Increase of Power

Output

High linearity is one of the main requirements of each power amplifier. Distortion is
manifested either by the harmonics of the carrier frequency (harmonic distortion,
where the nth harmonic is designated as nfc) or by intermodulation products (in-
termodulation distortion IMD, designated by fIMD = nf1 ± mf2) [7]. In practice, the
IMD is tested by a two-tone test or a two-tone test, whereby two or more sinusoidal
waveforms connected in series are applied to the amplifier.
Carrier-to-intermodulation ratio (C/I) should be higher than 30 dBc, where dBc
indicates the number of decibels below the carrier.
A prominent IMD type is third-order intermodulation distortion (IMD3). If a
system with at least a third-order non-linearity can be approximated by a polyno-
mial series

vout ðtÞ ¼ a þ bvin ðtÞ þ cv2in ðtÞ þ dv3in ðtÞ ð1:13Þ

and a two-tone signal

vin ¼ A cos x1 t þ B cos x2 t ð1:14Þ

is used as an input of the amplifier, then the output of the nonlinear amplifier is

vout ¼ a þ bðA cos x1 t þ B cos x2 tÞ

þ cðA2 cos2 x1 t þ B2 cos2 x2 t þ 2AB cos x1 t cos x2 tÞ
ð1:15Þ
þ dðA3 cos3 x1 t þ A2 B cos2 x1 t cos x2 t
þ AB2 cos x1 t cos2 x2 t þ B3 cos3 x2 tÞ:
1.4 Basic Principles of Operation of Power Amplifiers 15

Fig. 1.10 Intermodulation

tones

2 1- 2 1 2 2 2- 1

In this equation, dA2 B cos2 x1 t cos x2 t and dAB2 cos x1 t cos2 x2 t are the
third-order intermodulation terms at frequencies 2ω1 − ω2 and 2ω2 − ω1, illustrated
in Fig. 1.10.
Total harmonic distortion (THD) is the ratio of the sum of the power in all
harmonic components to the power contained in the fundamental frequency,
expressed as [46]
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P1 2
n¼2 Von
THD ¼ ; ð1:16Þ
V1

where Von is the root-mean square (RMS) value of the voltage of the nth harmonic
and V1 is the RMS value of the voltage of the signal at fundamental frequency.
Distortion can usually be improved by various linearization techniques [47–49].
These include feedback and feedforward techniques, various analogue predistortion
(APD) and digital predistortion (DPD) techniques, as well as pulse-width modu-
lation (PWM) and supply modulation.
To improve the output power, the efficiency can be boosted by means of
adaptive bias and the already mentioned Doherty techniques [50]. Power combining
is usually used to increase the total output power of a power amplifier system. This
can be performed on- and off-chip [8, 41, 51] and is typically done with the aid of
transformers.
Chireix outphasing is another technique that is gaining popularity because of
advances of low-power and high-speed digital processing [15]. In this system, an
amplitude-modulated (AM) signal is split into two phase-modulated signals with
constant amplitude, which are amplified separately and combined. This leads to
increased efficiency [52]. Outphasing systems have been implemented successfully
for Class-B, D, F and E amplifiers.

1.4.12 Impact of Power Amplifier Turn-on Characteristics

In many power amplifier applications, the power amplifier is switched off between
transmissions to save power [1]. In the time period during which the amplifier is
transitioning on or off, it will operate outside design specifications and the amount
16 1 Introduction

of power consumed will increase. If the delay time the amplifier takes to turn on is
defined as Td and time during the ON cycle is marked as Ton, then the total power
consumption of the power amplifier during the ON and turn-on part of the cycle is
given by

Td Pturnon
Ptotal ¼ þ Pstatic ; ð1:17Þ
Ton

where Pturn-on is the power it consumes during turn-on, and Pstatic is the power it
consumes during normal operation. It can be deduced from this equation that unless
Ton ≫ Td, power during turn-on can become prominent and should thus not be
ignored.
In addition to increased power consumption, intermodulation and harmonic
distortion will also increase during the turn-on phase.

1.4.13 Noise in Power Amplifiers

Voltage and current waveforms driving the power amplifier are generated in the
modulation block of the transmitter. Three types of noise are applicable to these
waveforms: AM noise, frequency-modulated (FM) noise, and phase noise (PM
noise). AM noise arises from amplitude variations inside the oscillator producing
the carrier frequency. FM and PM noise are due to frequency spreading around the
carrier frequency. Around the carrier, the PM noise is most prominent. Noise is
measured in units of dBc/Hz, or the number of decibels below the carrier per hertz.
Over a bandwidth of one Hz in single sideband, noise power is defined as the
noise-to-carrier power ratio

N
NCP ¼ 10 log ; ð1:18Þ
C

where N is noise power and C is the carrier power. In addition to carrier-induced

noise, power amplifiers can introduce thermal noise, which should be kept below
−130 dBm.

1.4.14 Measuring Large-Signal Power Amplifier

Performance

A practical means of large-signal performance measurement of the power amplifier

is experimentally accomplished by large-signal scattering parameters (S-parameters
1.4 Basic Principles of Operation of Power Amplifiers 17

ZS
vS [S] ZL
Zin
(Z 0)

ΓS Γin Γout ΓL

Fig. 1.11 Two-port model of a power amplifier showing the scattering matrix and reflection
coefficients

for short). Two-port S-parameters are normally used, where S11 and S22 indicate the
quality of input and output matching respectively and S21 and S12 indicate the
forward and reverse gain. All four two-port scattering parameters are normally
termed scattering matrix and denoted [S]. A two-port power amplifier model
showing the scattering matrix and reflection coefficients defined later is shown in
Fig. 1.11.

1.4.15 Measuring Amplifier Power Gain and Stability

Amplifier power gain and stability are usually defined in terms of reflection coef-
ficients and are treated together [45].
Gain of the amplifier between the source and the load is defined as transducer
gain and is the ratio between the power delivered to the load and power supplied
from the source:

PL
GT ¼ : ð1:19Þ
PA

In Fig. 1.11, the reflection coefficient seen looking to the source is defined as:

ZS  Z0
CS ¼ : ð1:20Þ
Z S þ Z0

Similarly, the reflection coefficient seen looking to the load is defined as

ZL  Z0
CL ¼ : ð1:21Þ
Z L þ Z0

The input impedance of an antenna is affected by objects in its vicinity [53],

therefore ΓL will not be zero even in almost perfectly matched systems. The
transducer gain can be expressed as a relation of S-parameters and reflection
coefficients as
18 1 Introduction

ð1  jCL j2 ÞjS21 j2 ð1  jCS j2 Þ

GT ¼ : ð1:22Þ
jð1  S11 CS Þð1  S22 CL Þ  S21 S12 CL CS j2

If the model of the driving transistor is known, it is possible to derive simpler

gain equations that are not derived from S-parameters [54].
Depending on frequency and termination, an amplifier can become unstable and
begin to oscillate. Therefore, any amplifier must also meet stability conditions in the
frequency range of interest. For stability, two more reflection coefficients, also
shown in Fig. 1.11, have to be defined:

S12 S21 CL
Cin ¼ S11 þ ð1:23Þ
1  S22 CL

and

S12 S21 CS
Cout ¼ S22 þ : ð1:24Þ
1  S11 CS

If we define a new quantity Δ as

D ¼ S11 S22  S12 S21 ð1:25Þ

then we can express Γin and Γout as

S11  CL D
Cin ¼ ð1:26Þ
1  S22 CL

and

S11  CL D
Cin ¼ ð1:27Þ
1  S22 CL

Stability implies that the magnitudes of all reflection coefficients are less than
unity. In other words,

jCL j\1; jCS j\1; jCin j\1; jCout j\1: ð1:28Þ

In can be shown, using the theory of stability circles, that the system will be
unconditionally stable if

1  jS11 j2  jS22 j2 þ jDj2

k¼ [1 ð1:29Þ
2jS12 jjS21 j

and
1.4 Basic Principles of Operation of Power Amplifiers 19

jDj\1: ð1:30Þ

Quantity k introduced in Eq. (1.29) is called the stability or Rollett factor.

In a matched power amplifier system, the output capacitance of the driving
transistor (COUT) changes with the drain voltage, and as a result, the amplifier
stability suffers at low voltages [55].
It should be noted that in packaged devices, the package parasitics need to be
included in the stability measurement or calculation.

1.5 Justification for Computer-Aided Design

Computer-aided design is not uncommon in circuit design, and as a matter of fact, is

used in many stages of circuit design.
A typical design flow for an electronic circuit could look as follows: Firstly, a
conceptual design or modeling is performed in a mathematical package. Hand
design of subsystems is then done before the design options are drawn in a sche-
matic editor. Performance of the circuit is then normally simulated in a SPICE1 or
RF SPICE [56] based simulator. Digital circuitry can be designed using a hardware
description language (e.g. VHDL2 [57]) and simulated in digital simulators (such as
Modelsim [58]). Furthermore, a synthesizer can be used to convert digital blocks
designed in hardware description languages into digital gates. Automatic
place-and-route tools can be used to connect synthesized digital cells automatically
with minimum involvement of a design engineer in order to create functional ICs.
Different place-and-route tools can be used for route tracks on PCBs for discrete
implementations. Automated tools can also be used for parasitic extraction, and
electromagnetic (EM) simulators or other specialist software can be used to cater for
transmission lines or other passive devices.
It is clear, however, that even with the increased computing power and new
innovative EDA ideas, an amount of hand design still needs to be performed, and
several hand-design steps (gaps), marked with bold blocks in the flow chart in
Fig. 1.12, will remain. This is particularly applicable to analog design, specifically
to the circuits operating at increased frequencies. A power amplifier is a very
common example of a circuit that requires a lot of hand design work, and therefore
fits into one of the gaps in the flow diagram. In the case of rapid design (where a
number of different devices need to be designed in a short span of time), even a
small amount of hand design work could quickly turn into a very tedious and
time-consuming task.

1
Simulation Program with Integrated Circuit Emphasis.
2
Very High Speed Integrated Circuit Hardware Description Language.
20 1 Introduction

Fig. 1.12 Role of CAD/EDA

in circuit design, where hand Conceptual design/modeling
design stages are termed gaps (mathematical package)
and are marked in bold

Hand design
(subsystems)

Schematic design
(schematic editors)

Simulation (SPICE-based
and digital simulators)

Automated tasks (e.g. syn-

thesis, place and route)

Hand design
(full system)

In this book, we describe how power amplifier design equations can be used as a
starting point to develop a set of software routines that will aid the design process.
Furthermore, we describe and give examples of CAD design of passives, particu-
larly inductors, which have been identified as being traditionally difficult to
implement because of low-quality factors and their indeterministic behavior at high
frequencies. Because of the strong influence of substrates and many degrees of
freedom that need to be considered in inductor design, we also demonstrate intel-
ligent search procedures for inductors that replace iterated procedures commonly
used. Finally, it will be shown how the complex task of matching can be simplified
by introducing matching algorithms. We also try to identify the basic parameters
each designer needs to take into consideration when performing the design. Some
of the parameters (e.g. the carrier frequency and antenna impedance) may be more
obvious and easier to determine than others (e.g. process parameters such as sub-
strate resistivity). There are parameters over which the designer typically has no
control (e.g. thickness of a metal for inductor implementation or a carrier fre-
quency). Other parameters can be treated as design parameters (e.g. output power).
Therefore, throughout this book we help the reader to identify and isolate the
needed information on design parameters.
1.5 Justification for Computer-Aided Design 21

Algorithms presented in this book are coded in MATLAB from Mathworks [59].
This package is a scripting programming language that supports a great number of
mathematical functions that add to the simplicity of the code. The authors are of the
opinion that most of the readers of this book would have at least a basic knowledge
of MATLAB to understand the examples provided.
The authors verified that the MATLAB scripts provided throughout the book as
examples work as expected in at least two versions of MATLAB: version 2007b
and version 2014b. It is thus likely that they will work correctly in any version
released between version 2007b and 2014b, and also in any newer version, but it is
impossible to verify this. The examples in MATLAB, however, are just for illus-
tration purposes and any other programming or scripting language can be used to
accomplish the same task (e.g. python, C#, Delphi). Licenses for certain languages
may be free of charge but may still have good mathematical libraries.

1.6 Organization of the Book

This book is organized in two parts. The first part focuses on the main concepts of
power amplification and this part can be used like any reference book. The second
part focuses on developing CAD routines to aid power amplifier design practically.
This chapter summarized the basic reasoning behind introducing custom EDA
into the design flow. Also, the basic principles of power amplifiers are discussed in
some detail, where some information serves as background information to the
reader and will not be discussed further in this book, but many topics will be
expanded in later chapters.
Chapter 2 will present a review of communication systems as applicable to
power amplifiers. The chapter will include a review of transmission bands and their
implications for transceiver system design. The feasibility of different passive
component implementations in each frequency range will be investigated. Power
amplifiers will be placed into the context of the transceiver system, and different
modulation schemes suitable for a particular band of operation will also be intro-
duced (including PSK, QPSK, DSSS, QAM, OOK and OFDM) The chapter will
also include the theory behind transistor operation as applicable to transceiver
theory. Various semiconductor fabrication technologies will be discussed for full
system integration or power transistor fabrication (SiGe, Si, GaAs, GaN).
Substrates for the implementation of discrete passives and their packaging will also
be discussed. Furthermore, the chapter will focus on the S-parameters and Y-pa-
rameters review, Smith charts and some other aspects of RF and microwave
engineering. The concepts of resonance and resonant tank, loaded quality factor,
insertion loss and impedance transformation will also be introduced.
Chapters 3 and 4 will describe power amplifier stages in great detail. Most of the
commonly used power amplification classes (among others A, AB, C, D, E, E−1, F
and F−1) will be discussed and the defining equations will be included.
Power-combining methods and methods for improving the efficiency of amplifiers
22 1 Introduction

(e.g. Doherty) will be discussed. The two chapters will also include the physics of
the amplifier operation and examples of both integrated and non-integrated
state-of-the-art designs found in the literature. Chapter 3 will focus on continuous
(classic) stages, while switch-mode classes will be covered in Chap. 4.
In Chap. 5, passive components will be discussed. The chapter will cover
resistors, capacitors, integrated inductors, solenoids, toroidal inductors, RF-chokes
and transformers, among others. Q-factors of these devices will be investigated in
detail. Special focus will be placed on both discrete and integrated inductor designs,
as they tend to exhibit low Q-factors and are therefore paramount to power
amplifier design. Micro-electro-mechanical systems (MEMS) will be discussed as a
promising technology for the design of passives.
Chapter 6 will be the last chapter of Part 1 and will deal with impedance
matching, which is important if the power amplifier is to be connected to the rest of
the transceiver system with minimum losses. Impedance matching with lumped
elements and transmission lines will be discussed, together with aspects of
matching both on- and off-chip. Analytical, graphical, and EDA matching solutions
will be presented, both for real and complex sources and loads.
Chapter 7 will be the first chapter in Part 2 and will present inductor design
automation and intelligent design ideas. The chapter will try to cover all inductors
needed to design a stand-alone system; this will include both filtering and matching
inductors. The Q-factor and its dependence on various inductor design parameters
will be described in detail. Together with Chap. 5, this chapter will contain
information beneficial not only for the design of power amplifiers but also for the
design of other devices that require high-quality passives, such as LNAs and
DC-to-DC converters.
Chapter 8 will introduce automation and intelligent design of various on- and
off-chip power amplifier classes in step-by-step manner. The algorithms will be
illustrated by means of flow charts and their development will be demonstrated in
MATLAB, with various examples demonstrating the use of each procedure. With
ideas for inductor design presented in Chap. 7, the automation of the quarter-wave
transformer and impedance matching networks will be considered in Chap. 8.
Finally, the ideas of both chapters will be merged to present the development of a
fully functional power amplifier design program as a proof-of-concept to the
readers.
Chapter 9 will be dedicated to practical power amplifier considerations for
discrete, integrated, and hybrid power amplifier implementations. Packaging will be
discussed in some detail for both system-on-chip (SOC) and system-on-package
(SOP) architectures. Layout of integrated circuits will be reconsidered here with a
few additional useful subroutines for rapid layout design. Finally, a suggestion on
how to execute a practical design of the power amplifier will be shared with the
readers.
Chapter 10 will cover future power amplifier directions. Topics in this chapter
will include mm-wave and transmission line theory, as well an introduction to near
terahertz (THz) transmissions. Other EDA opportunities will also be discussed.
Finally, this chapter will include some concluding remarks.
References 23

References

1. Watkins G, Wang S. The impact of power amplifier turn-on characteristics in cognitive radio
networks. Microwave J. 2014;57(2):86–92.
2. Raab FH, Asbeck P, Kenington PB, Cripps S, Popovic ZB, Pothecary N, Sevic JF, Sokal NO.
RF and microwave power amplifier and transmitter technologies—part 1. High Freq Electron.
2003;2:22–36.
3. Fang J, Moreno J, Quaglia R, Camarchia V, Pirola M, Guerrieri SD, Ramella C, Ghione G.
3.5 GHz WiMAX GaN Doherty power amplifier with second harmonic tuning. Microwave
Opt Technol Lett. 2012;54(11):2601–5.
4. Wagner R. 3G power ampliifiers: moving from GaAs to CMOS. Microwave J. 2012;55(4):44–
8.
5. Khannur PB, Chen X, Yan DL, Shen D, Zhao B, Raja MK, Sindunata R, Yeoh GW, Singh R.
Universal UHF RFID reader IC in 0.18-μm CMOS technology. IEEE J Solid-State Circuits.
2008;43(5):1146–55.
6. Kaynak M, Tekin I, Gurbuz Y. Fully integrated low-power SiGe power amplifier for
biomedical applications. IET Microwaves Antennas Propag. 2011;5(2):214–9.
7. Kazimierczuk MK. RF power amplifiers. 2nd ed. Chiechester: Wiley; 2015.
8. Rogers J, Plett C. Radio frequency integrated circuit design. 2nd ed. Boston: Artech House;
2010.
9. Raab FH, Asbeck P, Cripps S, Kenington PB, Popovic ZB, Pothecary N, Sevic JF, Sokal NO.
Power amplifiers and transmitters for RF and microwave. IEEE Trans Microw Theory Tech.
2002;50(3):814–26.
10. Barton TW, Perreault DJ. Four-way microstrip-based power combining for microwave
outphasing power amplifiers. IEEE Trans Circuits Syst—I: Regul Pap. 2014;61(10):2987–98.
11. Naude N, Bozanic M, Sinha S. Anlogue direct sequence spread spectrum transceiver with
carrier recovery employing complex spreading sequences. In: Malaga P, editor. IEEE
mediterranean electrotechnical conference; 2006. pp. 1227–30.
12. Iguchi S, Saito A, Watanabe K, Sakurai T, Takamiya M. Design method of class-F power
amplifier with output power of 20 dBm and efficient dual supply voltage transmitter. IEEE
Trans Circuits Syst—I: Regul Pap. 2014;61(10):2978–86.
13. Lee ML, Liou CY, Tsai WT, Lou CY, Hsu HL, Mao SS. Fully monolithic BiCMOS
reconfigurable power amplifier for multi-mode and multi-band applications. IEEE Trans
Microw Theory Tech. 2015;63(2):614–24.
14. Szczepkowski G, Farrell R. Study of linearity and power consumption requirements of CMOS
low noise amplifiers in context of LTE systems and beyond. ISRN Electron. 2014;2014:1–11.
15. Calvillo-Cortes DA, van der Heijden MP, Acar M, de Langen M, Wesson R, van Rijs F, de
Vreede LCN. A package-integrated chireix outphasing RF switch-mode high-power amplifier.
IEEE Trans Microw Theory Tech. 2013;61(10):3721–32.
16. Eswaran U, Ramiah H, Kanesan J, Reza AW. Class-E GaAs HBT power amplifier with
passive linearization scheme for mobile wireless communications. Turkish J Electr Eng
Comput Sci. 2014;22:1210–8.
17. Anastasijevic A, Coja D, Neskovic N, Neskovic A, Budimir D. Joint power amplifier and I/Q
modulator impairments modelling and compensation for LTE transmitters using artificial
neural networks. Int J Electron Commun (AEÜ). 2014;69(1):529–38.
18. Poulin D. The III-V vs silicon battle. Microwave J. 2009;52(4):22–38.
19. Avenier G, Diop M, Chevalier P, Troillard G, Loubet N, Bouvier J, Depoyan L. 0.13 m SiGe
BiCMOS technology fully dedicated to mm-wave applications. IEEE J Solid-State Circuits.
2009;44(9):2312–21.
20. Nikandish G, Medi A. A design procedure for high-efficiency and compact-Size 5–10-W
MMIC power amplifiers in GaAs pHEMT technology. IEEE Trans Microw Theory Tech.
2013;61(8):2922–33.
24 1 Introduction

21. Sheppard ST, Doverspike K, Pribble WL, Allen ST, Palmour JW, Kehias LT,
Jenkins TJ. High-power microwave GaN/AlGaN HEMTs on semi-insulating silicon carbide
substrates. IEEE Electron Device Lett. 1999;20(4):161–3.
22. Nguyen C, Micovic M. The State-of-the-art of GaAs and InP power devices and amplifiers.
IEEE Trans Electron Device. 2001;48(3):472–8.
23. Pribble WL, Palmour W, Sheppard ST, Smith RP, Allen ST, Smith TJ, Ring Z, Sumakeris JJ,
Saxler AW, Milligan JW. Applications of SiC MESFETs and GaN HEMTs in power amplifier
design. In: 2002 IEEE MTT-S international microwave symposium digest. Seattle; 2002.
pp. 1819–22.
24. Nellis K, Zampardi P. A comparison of linear handset power amplifier in different bipolar
technologies. IEEE J Solid-State Circuits. 2004;39(10):1746–54.
25. Chiou HK, Liao HY, Chen CC, Wang SM, Chen CC. A 2.6-GHz fully integrated CMOS
power amplifier using power combining transformer. Microwave Opt Technol Lett. 2010;52
(2):299–302.
26. Pierco R, Torfs G, De Keulenaer T, Vandecasteele B, Missinne J, Bauwelinck J. A Ka-band
SiGe BiCMOS power amplifier With 24 dBm output power. Microwave Opt Technol Lett.
2015;57(3):718–22.
27. Johnson EO. Physical limitations on frequency and power parameters of transistors. RCA Rev.
1965;26:163–77.
28. Baliga BJ. Power semiconductor device figure of merit for high-frequency applications.
Electron Device Lett. 1989;10(10):455–7.
29. Lee TH. The design of CMOS radio-frequency integrated circuits. 2nd ed. Cambridge:
Cambridge University Press; 2004.
30. Joo S, Han S, Song M, Kim H, Ham J, Seo M, Kim M, Park C, Yang Y. 1-Watt highly linear
HBT power amplifier driver integrated circuit with low current consumption. Microwave Opt
Technol Lett. 2013;55(7):1680–3.
31. Grebennikov A. High-frequency class-FE tuned power amplifiers. IEEE Trans Circuits Syst-I:
Regul Pap. 2008;55(10):3284–92.
32. Stameroff AN, Ta HH, Pham AV, Leoni RE. Wide-bandwidth power-combining and inverse
class-F GaN power amplifier at X-band. IEEE Trans Microw Theory Tech. 2013;61(3):1291–
300.
33. Raab FH. Class-E, class-C, and class-F power amplifiers based upon a finite number of
harmonics. IEEE Trans Microw Theory Tech. 2001;49(12):1462–8.
34. Eccleston KW, Smith KJI, Gough PT. A compact class-F/class-C Doherty amplifier.
Microwave Opt Technol Lett. 2011;53(7):1606–10.
35. Aflaki P, Negra R, Ghannouchi FM. Enhanced architecture for microwave currentmode
class-D amplifiers applied to the design of an S-band GaN-based power amplifier. IET
Microwave Antennas Prop. 2009;3(6):997–1006.
36. Grebennikov A, Sokal NO. Switchmode RF power amplifiers. 1st ed. Burlington: Elsevier;
2007.
37. Van JH, Kim MS, Jung SC, Park HC, Ahn G, Park CS, Kim BS, Yang Y. A high frequency
and high power quasi-class-E amplifier design using a finite bias feed inductor. Microwave
Opt Technol Lett. 2007;49(5):1114–8.
38. Yang LY, Chen HS, Chen YJE. A 2.4 Ghz fully integrated cascode-cascade CMOS Doherty
power amplifier. IEEE Microwave Wirel Compon Lett. 2008;18(3):197–9.
39. Thian M, Fusco VF. Transmission-line class-E power amplifier with extended maximum
operating frequency. IEEE Trans Circuits Syst—II: Express Briefs. 2011;58(4):195–9.
40. Dongyue J, Wanrong Z, Pei S, Hongyun X, Yang W, Wei Z, Lijian H, Yongping S, Jia L,
Junning G. Multi-finger power SiGe HBTs for thermal stability enhancement over a wide
biasing range. Solid-State Electron. 2008;52(6):937–40.
41. Aoki I, Kee S, Magoon R, Aparicio R, Bohn F, Zachan J, Hatcher G, McClymont D,
Hajimiri A. A fully-integrated quad-band GSM/GPRS CMOS power amplifier. IEEE J
Solid-State Circuits. 2008;43(12):2747–58.
References 25

42. Brama R, Larcher L, Mazzanti A, Svelto F. A 30.5 dBm 48 % PAE CMOS class-E PA with
integrated balun for RF applications. IEEE J Solid-State Circuits. 2008;43(8):1755–62.
43. Gaw C, Arnold T, Martin R, Zupac D. Evaluation of SiGe: C HBT intrinsic reliability using
conventional and step stress methodologies. Microelectron Reliab. 2006;46(8):1272–8.
44. Chen A, Lo HY. Semiconductor packaging: materials interaction and reliability. 1st ed. Boca
Ranton: CRC Press; 2012.
45. Ludwig Bretchko. RF circuit design: theory and applications. 1st ed. Upper Saddle River:
Prentice Hall; 2000.
46. Chen FY, Chen JF, Lin RL. Low-harmonic push-pull class-E power amplifier with a pair of
LC resonant networks. IEEE Trans Circuits Systems-I: Regul Pap. 2007;54(3):579–89.
47. Kim D, Choi D, Moon Y, Baek D, Baek K, Choi Y. The design and realization of a highly
linear power amplifier module for a WiMAX/WiBro (802.16e) base station. Microwave Opt
Technol Lett. 2010;52(9):1952–5.
48. Jiang H, Wilford A. Digital predistortion for power amplifiers using separable functions. IEEE
Trans Signal Process. 2010;58(8):4121–30.
49. Walling S, Allstot DJ. Linearizing CMOS switching power amplifiers using supply regulators.
IEEE Trans Circuits Systems II: Express Briefs. 2010;57(7):497–501.
50. Nam J, Kim B. The Doherty power amplifier with on-chip dynamic bias control circuit for
handset application. IEEE Trans Microw Theory Tech. 2007;55(4):633–42.
51. An KH, Lee O, Kim H, Lee DH, Han J, Yang KS, Kim Y. Power-combining transformer
techniques for fully-integrated CMOS power amplifiers. IEEE J Solid-State Circuits. 2008;43
(5):1064–75.
52. Zhang R, Acar M, van der Heijden MP, Apostolidou M, Leenaerts DMW. Generalized
semi-analytical design methodology of class-E outphasing power amplifier. IEEE Trans
Circuits Syst-I: Regul Pap. 2014;61(10):2951–60.
53. Keerti A, Pham AVH. RF characterization of SiGe HBT power amplifiers under load
mismatch. IEEE Trans Microw Theory Tech. 2007;55(2):207–14.
54. Komijani A, Natarajan A, Hajimiri A. A 24-GHz, + 14.5-dBm fully integrated power amplifier
in 0.18-μm CMOS. IEEE J Solid-State Circuits. 2005;40(9):1901–8.
55. Eroglu A. Stabilization of class E amplifiers with a diode network. AEU-Int J Electron
Commun. 2010;64(3):224–30.
56. Roberts GW, Sedra AS. SPICE. 2nd ed. Oxford: Oxford University Press; 1997.
57. Navabi Z. VHDL: Analysis and Modeling of Digital Systems. 2nd ed. Boston: McGraw-Hill;
1998.
58. Granberg T. Handbook of digital techniques for high-speed design. 1st ed. Upper Saddle
River: Prentice Hall; 2004.
59. Attaway S. Matlab: a practical introduction to programming and problem solving. 3rd ed.
Oxford: Butterworth-Heinemann; 2013.
 Part I
Power Amplifier Theory
Chapter 2
Review of Telecommunication Aspects
for Power Amplifier Design

In this chapter RF and microwave telecommunication theory, together with semi-

conductor and substrate theory, as applicable to power amplifier design, is pre-
sented to complement the basic theory of power amplifiers presented in Chap. 1.
Several topics are addressed.
The chapter starts with a review of the frequency spectrum and various trans-
mission bands and their implications for transceiver system design. The feasibility
of passive component implementations in each frequency range is investigated and
power amplifiers are placed into the context of the transceiver system. A review of
various digital modulation schemes, commonly used with power amplifiers, is also
presented.
Secondly, the theory behind transistor operation under large-signal and high
power is discussed. Different transistor types are discussed and their advantages and
disadvantages compared to other types are listed. In addition, various semicon-
ductor fabrication technologies are compared for complete system integration or
just for active device (transistor) integration. This is complemented by a discussion
of different substrates for discrete implementations, PCBs and packaging. Heating
presents a typical problem when large amounts of power are dissipated, thus
temperature aspects of transistors and semiconductor materials are also discussed.
Last part of this chapter focuses on the scattering (S-) parameters and admittance
(Y-) parameters review, Smith charts and some other aspects of RF and microwave
engineering, such as resonance, loaded quality factor, bandwidth, insertion loss and
impedance transformation. Fourier analysis of periodic signals is also discussed.

2.1 Wavelength and Transmission Bands

Frequency of operation has a major influence on the behavior of passive and active
devices. Below 30 GHz, transceivers constructed by lumped elements can be more
compact than designs based on transmission lines. Above 30 GHz, transceivers and
their elements require accurate modeling and high-precision manufacturing. Above
60 GHz, transmission lines and waveguides are more practical. As described in
Chap. 1, this book presents a design methodology for power amplifiers up to
© Springer International Publishing Switzerland 2016 29
M. Božanić and S. Sinha, Power Amplifiers for the S-, C-, X- and Ku-bands,
Signals and Communication Technology, DOI 10.1007/978-3-319-28376-0_2
30 2 Review of Telecommunication Aspects for Power Amplifier Design

Ku-band, the top end of which is located at 18 GHz [1–3]. Thus, the classification
of the different bands in the frequency spectrum is beneficial for power amplifier
design and is included in this section. Frequency bands are defined by the inter-
national telecommunication union [4].
The frequency is related to wavelength according to relation
v
k¼ ; ð2:1Þ
f

where v is the phase speed of the wave and f is the wave frequency. The phase
speed of the electromagnetic wave is the speed of light, which is about 3 × 108 m/s.
At lower frequencies, the wavelengths of signals are very large, so the size of the
electrical components has little impact on these signals. At 2.4 GHz, the wavelength
is 12.5 cm. This means that any component or a connection should not be greater
than a tenth of the wavelength (12.5 mm) for a system to behave with minimal loss.
This can still be accomplished on a PCB. At 18 GHz, the wavelength is 1.7 cm and
transmission lines can only be avoided on chip without incurring mismatches due to
connections longer than about 1/10 of the wavelength. In all other cases, careful
matching is paramount.
The extremely low frequency (ELF), voice frequency (VF) and
very-low-frequency (VLF) ranges span from 30 to 30 kHz and contain audible
frequencies and are thus not suitable for radio transmission. Low frequencies
(LF) span from 30 to 300 kHz and are used for long-range navigation, submarine
communication and telegraphy. Medium frequencies (MF) or medium waves span
from 300 to 3 MHz and are used for commercial radio. The high-frequency
(HF) range with frequencies from 3 to 30 MHz is used for military tactical radios
and for amateur radio operators because of the long-distance propagation properties
of the waves with 30-m-long waves.
The very-high-frequency (VHF) range with frequencies from 30 to 300 MHz
and the UHF range with frequencies from 300 to 3 GHz are used for television
broadcast, cordless and cellular telephone transmission, as well as for other wireless
applications, such as wireless local area networks (WLANs) and Bluetooth®. This
is also suitable for industrial heating and microwave ovens.
The SHF range includes frequencies from 3 to 30 GHz and the
extremely-high-frequency (EHF) range includes frequencies from 30 to 300 GHz.
These two ranges are mostly used for satellite communication and radar
applications.
UHF, SHF and EHF frequency ranges are further divided into L-band (1–2 GHz),
S-band (2–4 GHz), C-band (4–8 GHz), X-band (8–12.4 GHz), K-band (18–26.5 GHz),
Ku-band (26.5–40 GHz), V-band (40–75 GHz) and W-band (75–110 GHz).
Informally, the spectrum is also divided into RF, microwave and mm-waves, with the
boundaries between the three bands somewhat loosely defined. The frequency
spectrum is illustrated in Table 2.1. This table also shows the feasibility of passives as
applicable to each frequency range.
2.2 Review of Modulation Schemes 31

Table 2.1 The frequency spectrum and feasibility of passives

Frequency Wavelength Range UHF/SHF/EHF Other names/feasibility
range range name band name of passives
30–300 Hz 10,000–1000 km ELF – No transmission
300–3000 Hz 1000–100 km VF
3–30 kHz 100–10 km VLF RF/lumped
30–300 kHz 10–1 km LF passives
300–3000 kHz 1000–100 m MF
3–30 MHz 100–10 m HF
30–300 MHz 10–1 m VHF
300–1000 MHz 100–30 cm UHF
1–2 GHz 30–15 cm L-band
2–3 GHz 15–10 cm S-band
3–4 GHz 10–7.5 cm SHF Microwave/lumped
4–8 GHz 7.5–3.75 cm C-band passives, transmission
lines off-chip
8–12.4 GHz 37.5–24 mm X-band
12.4–18 GHz 24–17 mm Ku-band
18–26.5 GHz 17–1.1 mm K-band
26.5–30 GHz 1.1–1 mm Ka-band
30–40 GHz <1 mm EHF mm-wave/transmission
40–75 GHz V-band lines on and off chip
75–110 GHz W-band
110–300 GHz –

2.2 Review of Modulation Schemes

As discussed in Chap. 1, the power amplifier is driven by a signal that is generated

by a modulation block. Modulation implies that properties of the carrier signal, that
is the signal that can physically be amplified and transmitted, are varied so that the
information of interest is superimposed [5, 6]. Modulation is accomplished by
means of a modulator. On the receiver side, a demodulator is used to recover the
same information.
Modern telecommunication systems are moving from employing analog mod-
ulation towards employing digital modulation. Thus, various digital modulation
schemes are discussed in this review. In a digital modulation scheme, the carrier
signal is modulated by a discrete signal. Digital modulation is therefore a type of
analog-to-digital conversion. Some basic digital modulation techniques, already
defined in Chap. 1, are redefined here for readers’ convenience:
• Phase-shift keying (PSK),
• Frequency-shift keying (FSK),
• Phase-amplitude modulation (PAM),
32 2 Review of Telecommunication Aspects for Power Amplifier Design

• Quadrature amplitude modulation (QAM),

• On-off keying (OOK),
• Orthogonal frequency-division multiplexing (OFDM), and
• Various direct-sequence spread spectrum (DSSS) techniques.

2.2.1 Phase Shift-Keying

PSK is accomplished by modulating the digital information signal onto the carrier
signal by changing its phase. A finite number of phases (M) is used, usually two for
bits 0 and 1 (binary PSK or BPSK), four for bit combinations 00, 01, 10 and 11
(QPSK) or eight for eight 3-bit combinations (octal PSK). A typical waveform of
BPSK is shown in Fig. 2.1, and signal space diagrams (constellations) of BPSK,
QPSK and octal PSK are shown in Fig. 2.2.

2.2.2 Frequency Shift-Keying

FSK is accomplished by modulating the digital information signal onto the carrier
signal by changing its frequency. Usually, two discrete frequencies are used to
represent zeros and ones of a binary digital signal. This concept is illustrated in
Fig. 2.3.

2.2.3 Phase-Amplitude Modulation

In a PAM system, the message is encoded as the amplitude in series of pulses. The
number of amplitude levels required is M = 2k for k-bit blocks of symbols. As in
PSK, usually one-bit, two-bit or three-bit symbol combinations are used, corre-
sponding to M = 2, 4 and 8 respectively. This concept is illustrated in Fig. 2.4.
Signal amplitude

0 1 0 1

Time

Fig. 2.1 The BPSK waveform

2.2 Review of Modulation Schemes 33

(a) (b)
01

0 1 10 00
M=2 M=4

11

(c)
010
011 001

100 000
M=4

101 111
110

Fig. 2.2 Signal constellations of a BPSK, b QPSK and c octal PSK

Signal amplitude

0 1 0 1

Time

Fig. 2.3 Binary FSK waveform

Fig. 2.4 The PAM signal (a) 1

0
constellations for a M = 2,
b M = 4 and c M = 8 M=2

(b)
00 01 11 10
M=4

(c) 000 001 011 010 110 111 101 100

M=8
34 2 Review of Telecommunication Aspects for Power Amplifier Design

2.2.4 Quadrature Amplitude Modulation

QAM is a technique for bandwidth efficiency improvement, where two or more

separately modulated signals are combined on the carriers that are out of phase.
Thus it is a combination of a PSK scheme with another scheme, usually PAM.
When combining M1-level PAM with M2-phase PSK, M = M1M2 combined signal
constellation can be created, as illustrated in Fig. 2.5.

2.2.5 On-Off Keying

OOK is the simplest modulation technique. In this scheme, if the carrier signal is
present, it indicates a digital one, and if the carrier signal is absent, it indicates a
digital zero, as illustrated in Fig. 2.6.

M=4
M=8

M = 16
M = 32

Fig. 2.5 QAM constellations for different values of M

Signal amplitude

0 1 0 1

Time

Fig. 2.6 OOK waveform

2.2 Review of Modulation Schemes 35

2.2.6 Orthogonal Frequency-Division Multiplexing

OFDM is a technique used in cases where it is more feasible to transmit data over a
large number of carriers simultaneously rather than using a single carrier with a
high data rate. This scheme is practical for high data transmission, such as for
high-definition television or LTE networks. OFDM combines carriers with the same
amplitude and modulation scheme, but separated in frequency so that modulation
products arising from one frequency are negligible at the frequencies of the other
carriers.

2.3 Antennas and Propagation

After modulation and power amplification, the amplified signal is passed onto an
antenna in order to be transmitted. Antennas are therefore an important part of every
transmitter and used to radiate the electromagnetic energy into the channel effec-
tively [7]. On the receiving side, antennas are used for receiving the electromagnetic
energy from the channel.
Each antenna has characteristic input impedance, which is usually designed to be
50 Ω. For power amplifier design, the antenna efficiency is also important, which is
the ratio of radiated power to the power fed to the antenna [2]:

PRAD
gA ¼ : ð2:2Þ
PFED

An antenna also has its radiation characteristics, which are mostly determined by
its length and the way in which it is excited. The principle of antenna operation is
based on the Ampere-Maxwell’s law:

@D
D H ¼ Jþ ; ð2:3Þ
@t

where @D @t is the displacement current, J(t) is the time varying current density and H
(t) is the time varying magnetic field around the antenna.
The power density at the distance r from the antenna is

PT
pðrÞ ¼ GT ; ð2:4Þ
4pr 2

where GT is the gain of the antenna in the particular direction and PT is the
transmitted power. From this equation, it is clear that the power density decreases
quadratically with the distance and that high gains are needed to transmit over long
distances. The amount of power received by the antenna on the receiver side with
gain GR is given by the Friis formula
Random documents with unrelated
content Scribd suggests to you:
 Mr. Jenner. Would you please examine it? You need no more than
just to look at it, so you will be able to testify that that is a true and
correct copy of the book you have testified about, published by
Eagle Publishing Co., which contains on its reverse cover side the
letter to which you made reference.
Mr. Surrey. Yes, sir; it is.
The Chairman. What did it cost you to publish that?
Mr. Surrey. It came to $2.50 and some cents. In a limited
quantity—3,000.
The Chairman. Do you now propose to offer all of the exhibits?
Mr. Jenner. Yes; I have three more FBI photos, and then I will
have completed.
The Chairman. Very well.
Mr. Jenner. Mr. Surrey, I show you three more photographs which
are identified first as Commission Exhibit No. 997. Would you read
the material that appears on the reverse side of that first, please?
(The document referred to was marked Commission Exhibit No.
997 for identification.)
Mr. Jenner. Looking now at the face of the photograph,
Commission Exhibit No. 997, does—do the inscriptions on the
reverse side correctly describe that area of General Walker's home
and the Mormon church references?
Mr. Surrey. Yes; they do.
Mr. Jenner. You are familiar with that area?
Mr. Surrey. Yes; I am.
Jenner. And its physical appearance, except for the foliage on the
trees, is as that area looked on the night of April 10, 1963? Is that
correct?
Mr. Surrey. That is correct.
 Mr. Jenner. I hand you Commission Exhibit No. 1016.
(The document referred to was marked Commission Exhibit No.
1016 for identification.)
Mr. Jenner. Read the inscription on the reverse side, please. You
are familiar with that area shown on the photograph?
Mr. Surrey. Yes; I am.
Mr. Jenner. Do the descriptions on the reverse side of the
photograph correctly describe that area?
Mr. Surrey. With the exception that I do not know these cars and
so forth.
Mr. Jenner. I am talking about the area.
Mr. Surrey. The physical area; yes, they do.
Mr. Jenner. And that area looks the same today as it did on the
evening of April 10, or the day of April 10, 1963?
Mr. Surrey. Yes, sir.
Mr. Jenner. I now hand you the last of these, Commission Exhibit
No. 1017, and ask you first to read the inscription and then examine
the photograph.
(The document referred to was marked Commission Exhibit No.
1017 for identification.)
Mr. Surrey. Yes, sir; these are substantially correct.
Mr. Jenner. As of today, as well as as of April 10, 1963?
Mr. Surrey. Yes, sir.
Mr. Jenner. All right.
Now, Mr. Chief Justice, I offer in evidence the various exhibits
which we have identified in the record with the exhibit numbers, and
ask that the exhibits take the exhibit numbers I recited in each
instance as to each exhibit, being Exhibits Nos. 996 through 1000
and 1002 through 1017.
 The Chairman. They may all be admitted under those numbers.
(The documents heretofore marked Commission Exhibits Nos.
996 through 1000 and 1002 through 1017 were received in
evidence.)
Mr. Jenner. That includes, Mr. Chief Justice, the diagrams which
the witness has prepared for us.
The Chairman. Yes.
Mr. Jenner. As I reported to you, Mr. Chief Justice, the file on the
Walker incident reached us about 20 minutes before we opened this
morning. I think I have covered everything. Could I have the
privilege of 5 minutes to take a look?
The Chairman. Yes.
Mr. Jenner. I will do it very quickly.
The Chairman. Yes.
Mr. Jenner. Who is Mr. Coleman? Do you know a man by that
name?
Mr. Surrey. Not personally.
Mr. Jenner. Walker Kirk Coleman.
Mr. Surrey. As I just read on the back of your exhibit, he is the
boy that reported seeing several automobiles at the time of the
assassination.
Mr. Jenner. That is immaterial to this issue.
You have never seen either of the two men you have mentioned
before or since the occasion you saw that automobile with the two
men in it on the evening of April 8, 1963?
Mr. Surrey. Not to my knowledge. I never was very close to
them.
Mr. Jenner. Were you able to—what kind of an automobile was it,
do you know?
 Mr. Surrey. It was a Ford, a new Ford at that time.
Mr. Jenner. Sedan?
Mr. Surrey. Four-door sedan.
Mr. Jenner. And it was new?
Mr. Surrey. Yes.
Mr. Jenner. To your knowledge, have you ever seen that
automobile before or since?
Mr. Surrey. No, sir.
Mr. Jenner. What color was it, if you noticed?
Mr. Surrey. It was either a dark brown or a maroon.
Mr. Jenner. You followed it awhile and then gave up the chase?
Mr. Surrey. That is correct. Actually, they made a turn which—I
am familiar with downtown Dallas—and they made a turn which
would indicate they were doubling back or not going in a straight
direction. And I thought perhaps I had been spotted in my
convertible. So I left them there.
Mr. Jenner. I will close, Mr. Chief Justice, by asking the witness—
was the Mormon church in session? Had there been——
Mr. Surrey. There had been services.
Mr. Jenner. The evening of April 10?
Mr. Surrey. They were still dispersing.
Mr. Jenner. When you arrived at approximately 9:30 in the
evening of April 10, were people still leaving the Mormon church?
Mr. Surrey. Yes; they were.
Mr. Jenner. I have no more questions.
The Chairman. That will be all, Mr. Surrey. You may be excused
now.
 The Commission is adjourned.
(Whereupon, at 12:20 p.m., the President's Commission
recessed.)
 Thursday, June 18, 1964

TESTIMONY OF JAMES J. ROWLEY

AND ROBERT CARSWELL
The President's Commission met at 9 a.m., on June 18, 1964, at
200 Maryland Avenue NE., Washington, D.C.
Present were Chief Justice Earl Warren, Chairman; Senator John
Sherman Cooper, Representative Hale Boggs, Representative Gerald
R. Ford, and Allen W. Dulles, members.
Also present were J. Lee Rankin, general counsel; and Samuel A.
Stern, assistant counsel.
 TESTIMONY OF JAMES J. ROWLEY
(Members present at this point: Chief Justice Earl Warren.)
The Chairman. The Commission will come to order.
Chief, it is our procedure to read a little statement as to the
purpose of the meeting, for the benefit of the witness.
Chief Rowley will be asked to testify with respect to the
protective measures taken by the Secret Service in Dallas, changes
in such measures made as a result of the Dallas experience, and
with regard to the investigation of the assassination and any
information he may have respecting the assassination of the
President.
Would you raise your right hand and be sworn?
You solemnly swear the testimony you are about to give before
the Commission will be the truth, the whole truth, and nothing but
the truth, so help you God?
Mr. Rowley. I do.
The Chairman. Will you be seated, please. Mr. Rankin will conduct
the examination.
Mr. Rankin. Mr. Chief Justice, before starting the examination, I
would like to make a brief statement for your benefit and for the
benefit of the Commission, of the problems that are probably going
to develop in this area with regard to the security of the country, and
a suggestion about how we might handle them as we proceed with
the witness.
 I have suggested to Chief Rowley that as he moves along in his
testimony he might have various matters that he would think should
not be on the record because of the security of the country, and if
he would just suggest that, when he came to that point, and say
specifically that it did involve the security of the country, then we
would proceed to go off the record, if it was satisfactory to the
Commission, and consider those questions off the record. And then
return to the record as soon as we had completed those security
matters.
Would that be satisfactory?
The Chairman. I think that is an appropriate way to proceed.
Mr. Rankin. Chief Rowley, will you state your name and address
for the record, please?
Mr. Rowley. James J. Rowley, 3501 Rittenhouse Street NW.,
Washington, D.C.
Mr. Rankin. Do you have an official position with the
Government?
Mr. Rowley. I have, as Chief of the U.S. Secret Service.
Mr. Rankin. How long have you occupied that position?
Mr. Rowley. Since September 1, 1961.
Mr. Rankin. What is the nature of the duties of that position?
Mr. Rowley. The nature of the duties is the general overall
supervision of the activities of the Secret Service.
Mr. Rankin. And, in a general way, what is the official
responsibility under the statutes of the United States of the Secret
Service?
Mr. Rowley. Well, we are responsible under title 18, section 3056,
to investigate all violations that affect the currency, securities, and
coinage of the United States. That involves Government bonds,
Government checks, and such other functions and duties as are
authorized by law, subject to the direction of the Secretary of the
Treasury.
In addition, we have the responsibility of the protection of the
President, members of his immediate family, the Vice President,
President-elect, Vice President-elect, and the former President for a
reasonable period of time as he leaves office.
Mr. Rankin. Will you please tell us what experience you had with
the Secret Service prior to the time that you became chief.
Mr. Rowley. I was in charge of the White House detail from 1946
to 1961.
Mr. Rankin. Now, will you tell us briefly the training that you had
in regard to Government Service?
Mr. Rowley. I first entered the Government as a member of the
FBI in 1937, and spent a year with the FBI, after which I went back
to New York for a period of 9 months. I entered the Secret Service
on September 12, 1938. I spent time in criminal investigation in the
New York City office, and the Utica office of Secret Service and in
April of 1939, I was assigned to Washington, eventually to the White
House detail.
(At this point, Mr. Dulles entered the hearing room.)
Mr. Rowley. I served as a member of the White House detail, as
an agent on a shift, as an assistant agent in charge, agent in charge
of the shift, and advance man, in preparing for Presidential visits,
both domestically and abroad.
Mr. Rankin. What educational training did you have?
Mr. Rowley. I had 2 years of college toward a B.S., then I was
graduated from law school, and secured a master's degree in law.
Mr. Rankin. Was one of the duties of your position as chief of the
Service to have general supervision over the trip of President
Kennedy and Vice President Johnson to Dallas around November 22,
1963?
 Mr. Rowley. Well, that would be part of my job—the general
supervision of the trip. The actual direct supervision would have
been under the jurisdiction of Mr. Behn, who was in charge of the
White House detail.
Mr. Rankin. Could you describe briefly the nature of Mr. Behn's
responsibilities in that work?
Mr. Rowley. Well, it would have been, as mine was in the period I
was there, that he was responsible for developing all arrangements
with the members of the White House staff, designating the
members of the detail to develop advance work, assigning agents to
the various shifts, directing their training as it applied to the White
House detail, and participating in any event that he thought would
be necessary in connection with his work at the White House.
Mr. Rankin. Did you become familiar with what did happen on
that trip, in your position as chief?
Mr. Rowley. Yes; I was first informed while addressing a
graduating class of our Secret Service school on that day. I was
summoned by Mr. Behn to the White House, at which time he told
me that the President had been shot. He was then at the hospital,
and subsequently we were notified that the President had died; that
the Vice President would take the oath of office in the airplane at
Love Field.
In the meantime, I asked my deputy, who was in his office while
I was at the White House, to arrange with the Immigration Service
to close the border, Texas being in close proximity to the border.
There might have been a conspiracy or something, we didn't want to
take any chances. And then I immediately dispatched an inspector
from my staff to the Capitol to protect the Speaker, and directed the
other activities as we got the information from Dallas.
Mr. Rankin. Did you learn in connection with the trip when the
assassination occurred that certain of the Secret Service agents had
been in the press club and what is called the Cellar, at Fort Worth,
the night before?
 Mr. Rowley. Well, that came to my attention through a broadcast
that Mr. Pearson made, that the agents were inebriated the night
before at the Fort Worth Press Club. I immediately dispatched
Inspector McCann to Fort Worth to investigate the report, and to
interview the agents.
Mr. Rankin. What did you learn?
Mr. Rowley. I learned that there were nine agents involved at the
Press Club. And I might say this—the agents on duty throughout
that day had no opportunity to eat. When they arrived at Fort Worth,
they were informed that there was a buffet to be served at the Fort
Worth Club. This is what I ascertained in personal interviews. Upon
going over there, they learned there was no buffet, and some of
them stayed for a drink. Three, I think, had one Scotch, and others
had two or three beers. They were in and out—from the time they
arrived, I would say roughly around 12:30, until the place closed at
2 o'clock.
Now, after that some of them went to the Cellar. This is a place
that does not serve alcoholic beverages. They went there primarily, I
think, out of curiosity, because this was some kind of a beatnik place
where someone gets up and recites, or plays the guitar.
Mr. Rankin. Did you learn whether or not there were any
violations of the regulations of the Secret Service by these men?
Mr. Rowley. Yes; there was a violation. At that time there was a
section in our manual in effect that said that during——
Mr. Rankin. Will you give us first the number?
Mr. Rowley. Section 10.
Mr. Rankin. Is that chapter 1, page 7?
Mr. Rowley. Chapter 1, page 7; yes, sir.
Mr. Rankin. Now, will you tell the Commission about what the
regulation was?
 Mr. Rowley. "The use of liquor. Employees are strictly enjoined to
refrain from the use of intoxicating liquor during the hours they are
officially employed at their post of duty or when they may
reasonably expect that they may be called upon to perform an
official duty."
The one that applies here—"However, all members of the White
House detail and special agents cooperating with them on
presidential and similar protective assignments are considered to be
subject to call for official duty at any time while in travel status.
Therefore, the use of intoxicating liquor of any kind, including beer
and wine, by members of the White House detail and special agents
cooperating with them or by special agents on similar assignments,
while they are in a travel status, is prohibited."
Mr. Rankin. Can you tell the Commission how many men were
involved in these trips to the Press Club and the Cellar, where these
things were done?
Mr. Rowley. There were 9 men involved at the Press Club, and
there were 10 men involved at the Cellar.
Mr. Rankin. Now, how many men, of those 10 men, were in the
Presidential motorcade on the day of the assassination?
Mr. Rowley. Four—four men were in the followup car.
The Chairman. Who were they?
Mr. Rankin. Do you know their names?
Mr. Rowley. Yes; Landis, Hill, Ready, and Bennett.
Mr. Rankin. Did you make any investigation to determine whether
or not their violation of the Secret Service regulations had anything
to do with the assassination of the President?
Mr. Rowley. Yes. They performed their duties from the time they
departed in the followup car from Love Field until the point of the
tragedy in a most satisfactory manner. There was nothing deficient
in their actions or their alertness. They went through the heaviest
part of downtown Dallas, through the crowds, and performed in an
exemplary manner.
Mr. Rankin. How do you know that?
Mr. Rowley. From the reports that I got from their superiors.
Mr. Rankin. In the work that you did with the White House detail
before you became Chief of the Secret Service, did you know the
various responsibilities of the members of the White House detail?
Mr. Rowley. Yes, sir.
Mr. Rankin. Did you ever participate in such motorcades yourself?
Mr. Rowley. I have; yes, sir.
Mr. Rankin. How much?
Mr. Rowley. Well, I have participated, in rough numbers, over a
period of 22 years—roughly, maybe, a thousand or more.
Mr. Rankin. Will you briefly describe the functions of the Secret
Service agents in connection with the President's car?
The Chairman. Have you finished this other matter?
Mr. Rankin. No; I just wanted to——
The Chairman. All right. Go right ahead.
Mr. Rowley. When the President's car leaves the airport or a
railroad station or any other location, the agents accompany him to
the car and stand to the right and left, in the same order as their
designated positions on the followup car, and screen him. And then
the car moves out, slowly, because the rest of the cars have to have
an opportunity to follow in the motorcade, so that none lingers
behind, or is left behind. And then the agent in the lead car
determines that the motorcade is intact and is moving, then he steps
up his speed, which is a cue to the Presidential driver to step up his
speed, and then they go at a speed consistent with the crowd that is
there, and so forth.
 Now, upon leaving the airport, if there is a huge crowd there,
the men are still on the ground running on the right and left side of
the President, both rear and front of the vehicle. After they get out
of the crowd, then the men in the front beside the Presidential
vehicle drop back and take their positions in the followup car.
This is so that they are not in the way of the men running on the
right and left rear. They move back last and have a clear opportunity
to jump onboard the followup car in the event the speed of the
motorcade is stepped up.
When the motorcade comes to intersections or turns which are
always vulnerable points, in that if you make a right turn, that is the
closest point for someone to come out, the agents on the right side
before reaching that point, will jump off, to be available alongside
the President's car in the event someone darts out with some
malicious plan.
There have also been times when, innocently, ladies and young
people will come out to throw a bouquet of flowers. And then if
there is a crowd that is sparse, they return to their position in the
followup car.
Now, when they come into a big crowd, they take it on foot, and
at a little jog, if necessary.
In some instances, if the crowd continues for a prolonged
distance, the agents work together. In other words, there are rear
steps on the right and left rear of the Presidential car with handrails.
These have two purposes. One, for agents to ride on and to screen
the President from anything from above; the second, in a situation
like this, to keep an additional man available in case of trouble, and
also to alternate with the men to the right rear of the President, who
are jogging along warding off the crowd.
Mr. Rankin. Now, what positions did the four men that you
referred to that were involved in the press club and the Cellar matter
occupy on the day of the assassination?
 Mr. Rowley. Well, Mr. Ready occupied the right front, Mr. Landis
to his rear——
Mr. Rankin. What do you mean by right front?
Mr. Rowley. Right front running board position of the followup
car. It was his responsibility or duty to jump off in crowds and to
take the position at the right rear of the President's car.
Mr. Landis, if necessary, to jump off if the occasion demanded
and take the right front of the President's car.
Mr. Hill was on the left front running board of the followup car,
and his responsibility was at the rear of the President's car. His
position was assigned there because he was in charge of the First
Lady's detail, and she was seated on the left side.
And Mr. McIntyre was to his rear on the left running board. So
his assignment would have been up to the left front of the
President's car. Mr. Bennett was in the rear seat of the followup car.
Mr. Rankin. Now, how can you tell that the fact that they were
out as they were the night before and violated the regulations, had
nothing to do with the assassination?
Mr. Rowley. Well, based on the reports of my investigating agents
and the facts as to how they performed at the time of the tragedy.
Mr. Hill, who was on the left side, responded immediately—as he
looked toward the Presidential car, being on the left side, he scanned
from left to right, and when he saw there was something happening
to the President following a noise, he immediately jumped from his
position to get aboard from his side.
Mr. Ready scanned to the right so he was looking away from the
President, because he was looking around from the right side. As a
consequence, he wasn't aware of what was happening in the front.
The car was also going on a turn at that time.
Mr. Rankin. What about the other two?
 Mr. Rowley. The other two were watching—they reacted normally
—the man on the left side looked to his left rear, and the man,
Landis, looked to his right rear.
Mr. Rankin. Have you done anything to discipline these men for
violation of the regulations of the Secret Service?
Mr. Rowley. Well, I did consider what type of punishment would
be provided.
Then I also considered the fact that these men in no way had—
their conduct had no bearing on the assassination. And, therefore, I
thought that in the light of history, to place a stigma on them by
punishing them at that time, from which inevitably the public would
conclude that they were responsible for the assassination of the
President—I didn't think this was fair, and that they did not deserve
that, with their family and children.
(At this point, Representative Ford entered the hearing room.)
Mr. Dulles. May I ask one question there?
You described the assignment of the four men with respect to
the followup car and the President's car. Do they have different
assignments with regard to watching what is happening around
them, or does that depend on the circumstances in which they are?
Mr. Rowley. Both. When they start off they have a certain area
that they have to watch. Like the man in the right front would
naturally watch slightly to the right and in front of him. The fellow
on the side, behind him, will watch to the right and rear. In other
words, as they are going by a building, he should scan the building.
In the meantime, he picks up where the man in the front has
finished. In other words, the scan of the man in the front will cover
the building to his front and side; the fellow behind will scan
alongside from rear to forward. Their scanning joins. This is the way
they are accustomed to doing it.
Mr. Dulles. Who would cover straight ahead?
Mr. Rowley. The man in the front seat has that responsibility.
 Mr. Rankin. Chief Rowley, how do you construe subparagraph (c)
of your regulation 10 regarding the use of alcoholic liquors?
The Chairman. Will you read it for the record?
Mr. Rankin. Will you kindly read it?
Mr. Rowley. "Violation or slight disregard of the above paragraphs
or the excessive or improper use of intoxicating liquor at any time
will be cause for removal from the service. In interpreting the words
'excessive' and 'improper,' slight evidence tending to indicate unusual
or questionable conduct will be considered proof that the use of
liquor has been improper or excessive. Association with others who
drink to excess will be considered as an indication of using more
than a moderate amount of liquor. The excuse that liquor was used
for medicinal purposes will not be accepted."
Mr. Rankin. How do you construe and apply that?
Mr. Rowley. Well, in this instance, it was wrong.
Mr. Rankin. Now, were these men under this regulation
considered to be on travel status, so that they should not be using
intoxicating liquor?
Mr. Rowley. Yes, sir.
Mr. Rankin. And there is no question about that in your mind?
Mr. Rowley. No, sir.
Mr. Rankin. Has anything been done to reprimand and cause
them to realize that this is a violation of your regulations?
Mr. Rowley. They were interviewed by the inspector at the time.
The seriousness of the matter was impressed upon them. And I
think they recognize the seriousness of their acts.
The men we recruit are men that are college graduates and
mature, and we screen them very carefully, particularly before we
assign them to the White House detail. They know and we know
that they are in a fishbowl 24 hours a day, and that, therefore, their
conduct is always subject to scrutiny, and so forth, and that they are
responsible individuals. Their records have indicated that they have
been performing in a high degree. They have worked endless hours
of overtime. They are dedicated. And if they were not, they would
not be on the detail.
They realize the seriousness of the violation, and I went over it
with my special agent in charge. He understands it. And I am quite
sure that they all understand it at this time.
Mr. Rankin. I would like to have you examine Commission Exhibit
No. 1018, Chief Rowley, and see if that is the regulation of the
Secret Service that you have been referring to.
Mr. Rowley. Yes; that is what I have been reading here, sir.
Mr. Rankin. Mr. Chief Justice, I would like to offer as a part of the
record the regulation, Commission Exhibit No. 1018.
The Chairman. It may he admitted.
(The document was marked for identification as Commission
Exhibit No. 1018, and received in evidence.)
Mr. Rankin. Chief Rowley, have you had any other complaints
similar to this in regard to the conduct of the Secret Service agents
on the Presidential or White House detail?
Mr. Rowley. We had one in the last month. We had charges
leveled at us by an agent of the Secret Service——
Mr. Rankin. Will you tell us about that?
Mr. Rowley. Who is currently under indictment, and who will be
brought to trial on criminal charges on the 29th of June. And, for
that reason, while I have no reluctance to discuss it, I think we
should go off the record, because I don't want to in any way
prejudice the case.
The Chairman. There is no reason to discuss that case here, Chief.
 Is there anything in particular that would affect this situation you
wanted to know about, Mr. Rankin?
Mr. Rankin. Mr. Chief Justice, the only thing would be the
investigation as to whether or not there was comparable conduct. I
didn't know whether the Commission would like to know what that
investigation was and what the results of it were.
The Chairman. Well, I suppose there is no objection to the Chief
telling us what this complaint was, but not insofar as it bears on the
crime that he is charged with.
Mr. Rowley. Well, it ties in with the crime, because he said he
was framed.
Now, he said he was framed because he was prepared to go
before your Commission, sir, to testify about this thing that
happened 3 years ago, and in the charges he said he advised me, as
well as others, and nothing was done. He said he was framed for
this reason.
The Chairman. Had he ever made any complaint to you before?
Mr. Rowley. He had never made any complaint to me. It came as
a complete surprise.
Representative Ford. The complaint to you came subsequent to
the filing of criminal charges against him?
Mr. Rowley. He said he had made the charges at the time the
alleged incidents occurred, Mr. Congressman, that he notified me,
before he left an assignment 3 years ago.
Let me give you the background, so there is no
misunderstanding. We have what we call an orientation program.
The men we recruit from the colleges, and the type of men that we
want, we cannot always get off the civil service roster. Therefore, we
have an understanding with Civil Service that we can take men
under schedule A. Within a period of 2 years, they will have to be
assigned to the White House or dropped from the Service.
 Now, in order to determine their ability and fitness for
assignment, since some people are better criminal investigators than
they are in protection work, we have an orientation program which
includes duty on the White House detail. Mr. Bolden was one of the
men selected to come in the summer of 1961. He was also a
replacement for some regular agent on the detail who was on leave.
It was a 30-day assignment. This afforded us an opportunity to
observe him, determine whether he was equipped and so forth.
And he was on the White House detail for this short period of
time. The time that he describes was a 5-day weekend up in Hyannis
Port.
Mr. Rankin. I don't think that quite answers——
Mr. Rowley. I am giving the background.
Mr. Rankin. I think the question is as to when you got the
complaint.
Mr. Rowley. Well—excuse me. [Continuing.] Before he left his
detail assignment, you see, he alleges that he told me about the
condition that was going on up in Hyannis.
Representative Ford. Before he left on this 30-day assignment?
Mr. Rowley. When he left to return to his office in Chicago.
Mr. Rankin. And what is the fact in that regard?
Mr. Rowley. The fact is he never informed me. He never informed
any of his supervisors or anyone on the detail.
Mr. Rankin. I think the record should show, Mr. Chairman, that we
were never advised that he wanted to testify, nor had we any inquiry
or anything about the matter, until after we learned about it in the
newspapers. And, even then, he didn't ask to testify. And we asked
the FBI to check into it, and he had counsel, and they refused to tell
anything about the matter at that time.
Mr. Dulles. Could I ask a question?
 Did I understand you to say that the Civil Service prescribes that
certain men must be assigned to the White House for a certain
detail?
Mr. Rowley. No, Mr. Dulles; we have an arrangement with the
Civil Service that they will permit us to recruit these men, not from
the register, but under what they call schedule A. They give us an
opportunity, 2 years, to train these men, with the understanding that
within 2 years' time they will have to be assigned to the White
House detail or we will not be able to retain them in the Service.
However, during that 2 years, we urge them to take the civil
service examination, so that they get on the register. And then when
they do—quite frequently this occurs—they are selected from the
register, and once they become permanent, if they are not interested
in the White House detail, then they continue their work as a
criminal investigator in the field.
Mr. Dulles. But if they do not take that special examination, then
—and become a part of the civil service, then they have to be
assigned to the White House, to stay on?
Mr. Rowley. Yes.
Mr. Dulles. I was a little worried when you said certain people
had to be assigned to the White House, that you were under
compulsion to assign certain people to the White House in order to
retain them.
Mr. Rowley. No; anyone who works in the White House, whether
he is an electrician, a painter, or anything, for a period of 2 years, he
automatically becomes eligible for permanent civil service status.
Representative Ford. Is that by law or by regulation?
Mr. Rowley. That I cannot say. I would always interpret it as
under law. I may be wrong on that, Mr. Ford, but this is what
happens. When our men spend 2 years on the detail at the White
House, they come within that classification.
 Mr. Rankin. Chief, can you clarify Commissioner Dulles' inquiry?
The Civil Service does not direct that you put certain people in the
White House?
Mr. Rowley. Oh, no; we do that in order to—I see your point, sir.
We do that in order to give them the permanency that they should
have to continue their employment with the Secret Service.
Mr. Rankin. But that is the choice of the Secret Service rather
than anybody else?
Mr. Rowley. Yes, sir.
Mr. Dulles. I gather the Civil Service prescribed if they did not do
this, they could not be retained. Is that correct?
Mr. Rowley. That is right. In other words——
Mr. Dulles. There is some pressure, I should think.
Mr. Rowley. There is no pressure, because we voluntarily entered
into an agreement with them, sir, for this arrangement, explaining
that we frequently don't get from the register the type of men that
we want, and that, therefore, we want the opportunity to recruit the
men from the universities or colleges. Once they have served on the
White House detail for a period of 2 years, then they would get this
permanent status. However, during the 2 years, they have an
opportunity and they are encouraged to take the civil service
examination, so they get career status. But there is no pressure from
the Civil Service. It is a convenience or agreement that they have
arranged with us.
Mr. Rankin. Maybe I can help, Chief. Schedule A is an exemption
from the regular civil service roster, is it not?
Mr. Rowley. That is correct.
Mr. Rankin. And the register is a list of employees from which you
have to otherwise select Government employees if they are not
exempt by reason of their positions, is that correct?
Mr. Rowley. That is right.
 Representative Ford. In other words, Civil Service Commission
has set up for the White House detail all inclusive——
Mr. Rowley. Not necessarily for the White House detail, Mr. Ford.
For the Secret Service—to allow us to get the type of individuals that
we want for both criminal investigation and protective work. Because
if you say exclusively for the White House detail, the fellow might
not be equipped for the White House detail.
Representative Ford. In other words, every person recruited by
Secret Service for any capacity is recruited in the first instance under
schedule A.
Mr. Rowley. Yes; if he hasn't—if he is not on the register for civil
service. We first go to the Civil Service, when we want to select
somebody, to see if there is anyone on there that meets our
qualifications. And then, if not, then we hire them under schedule A,
which is sort of a blanket exemption.
Representative Ford. But I gather from what you have said, or I
think you are intimating that most of your recruiting actually is from
colleges, and they are under schedule A.
Mr. Rowley. That is right; yes, sir. Most of them from your State,
sir—Michigan State University.
Representative Ford. It is a fine school.
Mr. Rowley. That is where it started, actually. They were the first
ones. Now we also recruit on the west coast, in California, they have
terrific schools out there.
Mr. Rankin. Chief Rowley, I don't think you covered the Bolden
matter as to whether you had an investigation made. Did you?
Mr. Rowley. Yes; I did, sir.
Mr. Rankin. Did you find out anything about the conduct of your
agents?
Mr. Rowley. I found out there was no truth to the charges of
misconduct. There were 11 charges lodged against us.
 One charge, the ninth charge, a part of it was true. The boys did
contribute for food. In other words, up there in Hyannis, when they
are up there for a week, or a weekend, they would be assigned to a
house, which economically was beneficial to them. One shift, and
some of the drivers would be in this house. This house was in a
remote area from the shopping area and so forth. So they agreed
when they arrived there to contribute, to buy food for breakfast, it
being an 8 to 4 shift. Eight to four meant they would have breakfast
there and dinner.
Mr. Rankin. What do you mean by that, Chief? Did they get a
certain house and were able to live together there to reduce their
expenses?
Mr. Rowley. That is correct.
Mr. Rankin. And then they each contributed to that common
expense?
Mr. Rowley. That is correct.
Mr. Rankin. And did someone cook for them?
Mr. Rowley. One of the agents who enjoyed it as a hobby cooked
the meals for them, while the others took care of the dishes.
Mr. Rankin. They did contribute to supporting that?
Mr. Rowley. They contributed to supporting that, sir.
Mr. Rankin. Was there criticism of that action?
Mr. Rowley. There was criticism of the action to this extent: That
when they went shopping they bought two or three cases of beer
which they had available in the icebox when the men came off duty
in the evening.
Mr. Rankin. Now, were they on a travel status or subject to——
Mr. Rowley. Not on travel status under our regulations. They
could be there a week, and they would be working their 8 hours.
They were not working any longer than their 8 hours. It was
comparable to their assignment here in Washington.
Mr. Rankin. So it was really a summer White House position?
Mr. Rowley. Summer White House is what we called it.
Mr. Rankin. And did you investigate the charges to see whether
they were valid?
Mr. Rowley. I investigated. This portion was correct. There was
some substance to that portion.
He also said he was left on post for a period of 2 hours and
wasn't relieved. That an agent had used this time to take care of his
private car. We established there was no agent up there who had a
private car.
Further, we established that he was left on post because
according to our arrangements it was routine that whenever the
President went out for a cruise, the agents on the outer perimeter at
the time would remain on duty, and the agents in the inner
perimeter would accompany the President on the cruise in the
followup boat. Naturally, when they were out on the boat, there was
no one available to start what we call the push, to rotate the men
from one post to another. In other words, in the White House or any
place where we establish posts, every half hour one man starts from
the office and starts making the push. The first man is relieved and
he relieves the next one, so there is no monotony on their jobs.
They each have a different area. They are conversant or acquainted
with each and every phase of the physical area. But because he was
on one post, and not relieved, he complained.
So the next day, to bend over backwards, and show there was
not any prejudice, the agent in charge took him on the cruise, so he
would not feel he was being ignored.
Mr. Rankin. Now, from your investigation, did you find any
violation at Hyannis of the regulations of the Secret Service?
Mr. Rowley. No, sir.
 Mr. Rankin. Have you been informed of any other claims that
Secret Service agents had been violating the regulations while on
duty?
Mr. Rowley. No, sir; I haven't been informed of any others. And it
seems in the last few days or few weeks we have been getting
complaints that we haven't had in many years. And I think, as I
mentioned earlier, because of the fact that we are very careful with
the type of men we screen, their record has been above reproach
over the years. They have conducted themselves in an exemplary
manner. My files are replete with commendations on behalf of the
agents wherever they have traveled and worked with committees
and individuals in connection with Presidential travels, both here and
abroad, which testifies to the impression that they have made.
Mr. Rankin. Have you ever had a Secret Service agent indicted or
a complaint filed against him, a criminal complaint, prior to this
time?
Mr. Rowley. This is the first time I remember anything like this
happening since I have been with the Secret Service.
Representative Ford. Mr. Rankin, I don't recall Chief Rowley
saying precisely what the reprimands were specifically for these
violations of the regulations in this one instance.
You spoke highly of their background, and you spoke very high
in their praise. But I did not hear what reprimand, if any, had
actually been lodged against them.
Mr. Rowley. There was no reprimand. You are talking about the
current thing?
Representative Ford. I am talking about the Dallas trip.
Mr. Rowley. I stated in considering what would be an appropriate
punishment at the time, I felt that these men, by their conduct, had
no bearing on the assassination of the President in Dallas. That to
institute formal punishment or disciplinary action would inevitably
lead the public to conclude that they were responsible for the
assassination of President Kennedy. I did not think in the light of
history that they should be stigmatized with something like that, or
their families or children. And, for that reason, I took the position
that I did.
Representative Ford. So there was no official reprimand or
disciplinary action?
Mr. Rowley. No, sir.
Mr. Rankin. Did you talk to the agents, to indicate and make it
plain to them that this was a violation of the regulations?
Mr. Rowley. I talked to some of the agents, as did my inspector
at the time, who interviewed each and every one of them.
Mr. Rankin. And I think the Commission would be interested in
whether you can be assured, or assure them that the action you
took was sufficient so that this would not happen again.
Mr. Rowley. Well, I am confident that it would not happen again,
Mr. Rankin.
Mr. Rankin. Can you tell us why you think so?
Mr. Rowley. Because they realize the seriousness of their action.
Initially I can understand the situation—they thought they were
going for a dinner, buffet, and they got into the place and it wasn't
there.
I talked personally with the agents there, and they just thought
while they were there they would have a drink. It was one of those
situations.
The important thing was that it was pointed out to them this
was wrong, this was a violation. These men are young men with
futures, they realize the true situation, innocent as they may have
seemed to think it was.
But I am quite confident that we will not have a repetition of
that.
 And in talking to Mr. Behn—I am confident, too, in him—I know
that he will see to it that they are well supervised.
Mr. Rankin. When they are out on a trip of this kind, Chief
Rowley, as I understand your regulations, it is understood by the
regulations and by the Secret Service that they are on duty all the
time—that is, subject to call?
Mr. Rowley. Yes, sir.
Mr. Rankin. And even though it is late in the evening or they had
gone to bed in the early hours of the morning, they could be called
to go on duty and perform their responsibility of taking care of the
President or the Vice President, or whoever they are charged with; is
that right?
Mr. Rowley. That is right.
Mr. Rankin. So that do they understand that when they are out
on that kind of duty, they are subject to call at all times, and
anything they do contrary to regulations is a violation, because they
are subject to the call and must be ready at any moment to perform
their duties.
Mr. Rowley. They certainly do, because there have been
situations, whether or not they have had it with the Kennedy
administration I don't know—but I know there have been situations
where we have moved fast, all hours of the night. I remember one
instance, that has never been disclosed—as Mr. Dulles knows, you
never advertise your successes, you just get the other things—that I
would like to give you as an example off the record, to answer your
question, if I may.
The Chairman. Off the record.
(Discussion off the record.)
The Chairman. Back on the record.
Chief, it seems to me that on an assignment of that kind, to be
alert at all times is one of the necessities of the situation. And I just
wonder if you believe that men who did what these men did, being
out until early morning hours, doing a little—even a small amount of
drinking—would be as alert the next day as men should be when
they are charged with the tremendous responsibility of protecting
the President.
Mr. Rowley. Well, we checked on that, Mr. Chief Justice, and the
agent in charge reported that they were in good physical condition. I
don't condone these late hours; no. This is not a rule. This case is an
exception. However, because of the activities of any travel such as
the Presidents today make from one place to another, to maybe
seven States in a weekend, there is constant going.
I don't condone this at all. But these men are young. They are of
such age that I think that they responded in this instance adequately
and sufficiently as anyone could under the circumstances.
The Chairman. Well, I am thinking of this. As you go along in the
motorcade, you have men who are scanning the buildings along the
way, don't you?
Mr. Rowley. Yes, sir.
The Chairman. And they have submachineguns in one of the cars.
Mr. Rowley. No; for security reasons, I would like to—we don't
have machine-guns now, sir.
The Chairman. I just thought I heard that from the record here,
that they had some kind of guns.
Mr. Rowley. They had a weapon, a new weapon; yes, sir.
Mr. Chairman. Well, whatever it is.
Now, other people, as they went along there, even some people
in the crowds, saw a man with a rifle up in this building from which
the President was shot. Now, don't you think that if a man went to
bed reasonably early, and hadn't been drinking the night before,
would be more alert to see those things as a Secret Service agent,
than if they stayed up until 3, 4, or 5 o'clock in the morning, going
to beatnik joints and doing some drinking along the way?
Mr. Rowley. If I remember that witness' testimony—and that was
one of the first statements that he made—that witness was with his
wife, and he happened to look up there, and I think he said, "There
is a man with a rifle, it is a Secret Service man," and let it go at that.
He didn't inform any of the authorities.
The Chairman. No; nobody did. But I say wouldn't an alert Secret
Service man in this motorcade, who is supposed to observe such
things, be more likely to observe something of that kind if he was
free from any of the results of liquor or lack of sleep than he would
otherwise?
Mr. Rowley. Well, yes; he would be. But then, on the other hand,
Mr. Chief Justice, in some instances the men come in from a trip at
1:30 in the morning, which there have been cases on travels that I
have made, and have to be up at 3:30 or 4 o'clock, and out in time
for a 5 o'clock departure. Then you go all that day until 1 or 2
o'clock the next morning. This is what has happened in the past.
The Chairman. I am not talking about the past. We are talking
about nine men here who were out until rather unusual hours of the
morning.
Mr. Rowley. Yes, sir.
The Chairman. They were to be on duty the next day.
The next day—or if not sooner.
The next day they were supposed to be alert to anything that
might occur along the line of march. Don't you think that they would
have been much more alert, sharper, had they not been doing these
things?
Mr. Rowley. Yes, sir; but I don't believe they could have
prevented the assassination.
 The Chairman. Isn't it a substantial violation of these rules to do a
thing of that kind?
Mr. Rowley. Yes, sir—on the basis of this section here.
The Chairman. Yes.
Now, Chief I noticed, also, in reading some of the reports that
three of these men whom you speak of, were actually on night duty,
protecting the life of the President. And around 4 o'clock in the
morning, when they were protecting him at the Texas Hotel, they
said that they had a coffee break, and they went from the hotel over
to the beatnik joint. Now, is that consistent with your regulations?
Mr. Rowley. In this case, I talked to these three agents. They
were relieved at different times—because their posts are in the
corridor of a stuffy hotel——
The Chairman. Of the what?
Mr. Rowley. The corridor that they were on post outside the
President's suite was a stuffy one, and they went downstairs to get a
breath of fresh air. And they walked—it was a block—and out of
curiosity they went into this place. One fellow looked in and left, he
didn't buy any coffee. Another fellow went in and felt, I suppose,
when he went in that he would buy a cup of coffee. But they were
on what we call reliefs, the same as we relieve them around the
White House. There are only so many posts, but you have a group
of men in one of the rooms of the hotel where they are available,
like an alert squad, and they relieve everyone on post every half
hour. It is a part of the rotation of positions we have.
The Chairman. Do you have any regulations concerning where
they shall remain when they are relieved for this short period of
time?
Mr. Rowley. No, sir.
The Chairman. They can go any place they want?
Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.

More than just a book-buying platform, we strive to be a bridge

connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.

Join us on a journey of knowledge exploration, passion nurturing, and

personal growth every day!

ebookbell.com