Maurizio 

Di Paolo Emilio

Microelectronic
Circuit Design
for Energy
Harvesting
Systems
Microelectronic Circuit Design for Energy
Harvesting Systems
Maurizio Di Paolo Emilio

Microelectronic Circuit
Design for Energy
Harvesting Systems

123
Maurizio Di Paolo Emilio
Data Acquisition System
Pescara, Italy

ISBN 978-3-319-47586-8 ISBN 978-3-319-47587-5 (eBook)

DOI 10.1007/978-3-319-47587-5

Library of Congress Control Number: 2016959987

© Springer International Publishing AG 2017

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 Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To Julia, Elisa and Federico
Imagination is more important than
knowledge. [A. Einstein]

When wireless is perfectly applied, the whole

earth will be converted into a huge brain,
capable of response in every one of its parts.
[N. Tesla]
Foreword

Energy is everywhere, sun, wind, temperature and other sources rarely known like
earth vibrations: all of these are sources of “free” energy. There are not fuel costs.
The only investment required is facility and maintenance. Modern technology has
a lot of “waste energy” that needs to be captured. Producing “green energy” and
harvesting waste energy are the solution to create a better and sustainable world!
Thanks to Tesla, it is possible to gain energy at zero cost, but we don’t know
how to obtain it. There is only one solution to resolve this problem: research and
development. Energy is everywhere, as Tesla said, from the most simple resources
to capture (sun, water, wind) to more complex forms to harvest (vibrations, heat,
sound, movements). Energy harvesting is the solution for small and large energy
requirements: from smartphones to automotive. Thinking about the automotive
market, electric cars were invented more than 20 years ago, but only now there
are companies like Tesla Motors that combine technology and design. They gain
billion dollars in a few days. Energy efficiency is a critical point for electric cars
and a big contribution was given by the development of battery technology and
their evolution. In addition, features like “start and stop” or “regenerative brake”
or even energy studies about “regenerative shock absorbers” were introduced only
in the last few years. The lesson that we learned is to optimise the energy recovery
from a mix of sources and not from a single source. This is the way to obtain the best
performance. How often does a smartphone vibrate and how many times are they
illuminated by sunlight or artificial light or even are they exposed to body heat: : :all
sources of energy! Microelectronic Circuit Design for Energy-Harvesting Systems
is the ideal guide to explore and examine in detail this technology. It is only a matter
of time when smartphones without battery will be produced. Understanding energy
harvesting and its power management means discovering the future of power supply.

ix
x Foreword

The author is the right person to discuss these topics. He is an electronic engineer
and also a physicist, so who is better than him to accompany us in this study?

Founder and Editor Emanuele Bonanni

Elettronica Open Source, EOS Book
and Firmware Magazines
Rome, Italy
September 2016
Foreword

My father was a bit of a mathematician, physicist and all-round curious sort. I

grew up listening to him about scientific matters that a 6- or 7-year-old may have
not immediately understood but does get interested in, enough to seek them out
when older. One such concept my father spoke of and that stuck in my mind was
perpetuum mobile or perpetual motion, which allows a machine to run unabated
forever without running out of power. For a very long time I believed in this
machine; it took a great deal of learning and understanding of the physical world
to realise it is not possible eventually; everything runs out of energy for a variety
of reasons. However, one type of energy does convert into another, allowing for
something to act, move or operate. Maurizio’s latest book discusses the concept
of energy harvesting, where power may be collected from a viable source, which
could be as simple as a person walking or just waving a hand. Of course, in addition
to motion or vibration, other unlimited sources of energy could be light, sound or
heat abundantly available from the environment around us. Imagine what a single
car could power if only a fraction of its unwanted outputs are to be harnessed, i.e.
collected for use by another device, most likely electronic! The impact of energy
harvesting could be huge economically and, more importantly, for the environment,
which we must start thinking about how to best protect and preserve. Maurizio
is addressing all these issues in his new book, where he also focuses on different
types of technologies available for energy harvesting. He describes the basic and
advanced concepts of energy harvesting in terms of physics and engineering and
proposes design techniques suited to power supplies for low-power systems. In his
book, Maurizio introduces different principles of transducer operation and materials
and related devices, as well as power management, storage and design of energy-
harvesting systems and their future architectures. I am excited about this future one
that will be so greatly enabled by technology. I am certain that 1 day an electronic
device or system will be able to swap seamlessly between energy sources for its

xi
xii Foreword

power, so it can run forever a little like perpetuum mobile. Who says this is not
doable? I can hardly wait to see what happens next.

Editor, Electronics World

London, UK Svetlana Josifovska
September 2016
Preface

Today’s technology world is certainly evolving and is also bringing most innovative
devices that have ever come to the market. One of those is called energy harvesting.
Energy harvesting is the process of taking energy from external sources and
converting it to electrical energy to supply any mobile device. Michael Faraday’s
law of induction found that moving a magnet though a loop of wire would create
an electrical energy in terms of current. This principle can be one point of starting
for energy harvesting. In the early twentieth century, a great scientist tried to pose
the question to which we seek an answer yet. We all live in a solar system with
an enormous amount of energy sent to the earth; adequate protection system (see
the atmosphere) allows us to live and prevent the destruction of the planet under
the powerful yield of energy from the sun. In addition to this, the earth is a living
organism, or an accumulator and generator for various types of known and unknown
energies. The question is legitimate: is it possible to be able to “catch” this enormous
energy and make it available to users? Scientifically speaking, yes, it is possible;
the thing is possible. A classic example is the photovoltaic cells that convert solar
energy into an electrical signal (photovoltaic effect). This question was asked by
Nikolas Tesla. Currently, most of these electronic devices are powered by batteries.
However, batteries have several disadvantages: they need to be replaced or recharge
periodically and mostly they are not handy with their size and weight compared to
a highly electronic technology. One possibility to overcome these power limitations
is to extract energy from the environment to recharge a battery or even to directly
power the electronic device. The environmental energy is naturally occurring in
large and micro-scale; the technologies have been widely disseminated efficiently:
solar energy is an example that although the overall efficiency remains remarkably
low (around 30 %), its usefulness is much appreciated, or almost. Fossil fuels are
limited, expensive and, above all, not environmentally friendly as they induce a
strong impact on land-based pollution. The photovoltaic system is a classical green
system that converts solar energy to electric current to supply electronic devices. It
needs improvements in terms of efficiency and new materials with the goal to be a
system totally dependent in periods of minimum intensity of the sun. In this context,
however, it is fundamental that the battery management system have a great capacity

xiii
xiv Preface

and excellent long-term efficiency. How much energy is available around us? What
kind of energy sources do we have? What is utility? They are some issues to which
we will answer in this book through an engineering discussion with the basic and
advanced concept about physics and electronic circuit. In addition to large-scale
energy such as solar, there are variants of energy, which could be defined on a
small scale to implement in low-power systems such as wearable and smartphone
devices. Walking can also be used to produce energy by using an electromagnetic
mechanism. The electronics and microelectronics are spreading steadily, and many
companies provide day after day IC systems of energy harvesting for different types
of energies such as electromagnetic. The purpose is always to harvest the energy
dispersed in the environment for reusing it in other forms (electric current) to power
other electronic devices or the same device in a way that we could define recycling
energy loop. Collecting all these energies (heat, light, sound, vibration, movement)
could have a significant impact concerning the economic and environmental factors,
reducing costs and developing new sensor technologies. The main part of every
electronic system is the battery, as in a computer or a smartphone, and thinking
to recharge it by external source of energy in a harvesting automatic mode could
be very impressive with zero-impact work process: a smartphone supplied by
environmental energy without battery. Eliminating the battery is a long-term goal
that in some systems such as photovoltaics is definitely an essential element in
the design. The physical aspects that come into play in an energy-harvesting
system can be described in terms of ability to store the energy, materials science,
microelectronics for power management and systems engineering. All electronic
systems such as computer and smartphones waste energy: why not charge your
phone by using its electromagnetic waves that we know to be of greater intensity
during calls and receiving data? Still, why not detect the energy that the universe
sends to us, such as cosmic rays for the realisation of a low-power system to supply
wearable systems. But it is interesting to note that there are other sources that have
emerged from the action of man, as a consequence of industrial and technological
development. These modern energy sources (or artificial) are directly related to
energy harvesting; vibration or temperature gradients are produced by machines
and engines. Even in the electromagnetic spectrum, we can collect the energy not
only from the natural solar radiation but also by all the artificial radio sources
which is acquiring a great importance with the development of web-based devices
concerning IoT and IIoT. The technology behind energy harvesting is possible,
thanks to a careful analysis and design of power management factors that have
reduced the consumption of electronic systems. Although manufacturers struggle
to reduce battery consumption, running out of power after just a few hours of use
and having to be connected to a power supply to recharge are common problems
that need a solution. The goal of the book is to focus on energy harvesting which
is released into the environment in various types: electromagnetic, vibrational and
heat. The most used sources are vibration, movement, all the mechanical energies
and sound that can be captured and converted into electricity using piezoelectric
materials. The heat can be captured and converted into electricity using thermal
and pyroelectric materials. This book describes basic and advanced concepts of
Preface xv

energy harvesting in terms of physics and engineering and then proposes the design
techniques to obtain power supplies for low-power systems. The first six chapters
describe a special technology of energy harvesting including the different principles
of transducer and related materials, power management, storage and design of
system. In addition, design techniques with conditioning circuital solutions to
efficiently manage a low-power system will be analysed. The final chapter describes
various types of energy-harvesting applications and related market with a focus on
future architectures.

Pescara, Italy Maurizio Di Paolo Emilio

August 2016
Acknowledgements

I would like to express my gratitude to all those who gave me the possibility
to complete this book. In particular, I want to thank Svetlana Josivofska and
Emanuele Bonanni for their foreword and Charles B. Glaser, editorial director, for
the publication of present book. To my family, thank you for the patience and for
encouraging and inspiring me to follow my dreams. I am especially grateful to my
wife, Julia, and my children, Elisa and Federico.

xvii
Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Sensors and Transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2.1 Temperature Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2.2 Magnetic Field Sensors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.3 Potentiometers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.4 Light Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Communications Cabling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3.1 Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4.1 Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4.2 Settling Time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4.3 DC Input Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2 The Fundamentals of Energy Harvesting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.1 What’s Energy? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 Why Energy Harvesting? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Free Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.4 Power Management Unit. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.5 Storage Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3 Input Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.1 Mechanical Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.2 Thermal Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.3 Electromagnetic Energy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.4 Space Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.5 Solar Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.5.1 Photovoltaic Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

xix
xx Contents

4 Electromagnetic Transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.2 Electromagnetic Waves and Antenna. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.3 System Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
5 Piezoelectric Transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.2 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.3 Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
5.4 System Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
6 Thermoelectric Transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
6.2 Seebeck and Peltier Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
6.3 Potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
6.4 Charges in a Semiconductor with a Temperature Gradient. . . . . . . . . 57
6.5 Thermoelectric Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
6.6 Thomson Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
6.7 Thermoelectric Generator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
6.8 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
6.9 Figure of Merit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
7 Electrostatic Transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
7.2 Physical Phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
7.3 Switching System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
7.4 Continuous Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
7.5 Design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
8 Powering Microsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
8.1 Power Conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
8.2 Rectifier Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
8.2.1 Bridge Rectifier Circuit. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
8.2.2 Zener Diode as Voltage Regulator . . . . . . . . . . . . . . . . . . . . . . . . . 81
8.2.3 Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
8.3 Piezoelectric Biasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
8.4 Voltage Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
8.5 MPPT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
8.6 Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
8.7 DC-DC Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
8.7.1 Linear Regulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
8.7.2 Switching Regulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
8.7.3 Buck Converter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Contents xxi

8.7.4 Boost Converter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

8.7.5 Buck-Boost Converter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
8.7.6 Armstrong Oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
8.8 Load Matching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
8.9 AC-DC Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
8.10 Electrical Storage Buffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
8.10.1 Supercapacitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
9 Low-Power Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
9.2 Review of Microelectronics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
9.2.1 Basic of Semiconductor’s Physics . . . . . . . . . . . . . . . . . . . . . . . . . 106
9.2.2 PN Junction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
9.2.3 Diode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
9.2.4 Bipolar Transistor: Emitter Follower . . . . . . . . . . . . . . . . . . . . . . 111
9.2.5 MOS Transistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
9.2.6 Differential Amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
9.2.7 Feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
9.2.8 Effects of Feedback. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
9.2.9 Digital CMOS Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
9.2.10 CMOS Inverter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
9.2.11 Current Mirror . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
9.2.12 Ideal Current Mirror . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
9.2.13 Current Mirror BJT/MOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
9.3 Low-Power MOSFET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
9.3.1 General Characteristics of a MOSFET . . . . . . . . . . . . . . . . . . . . 129
9.3.2 Mosfet Power Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
9.3.3 Stage of Amplification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
9.3.4 Common Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
9.4 Analog Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
9.5 Operational Amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
9.6 Power Supply and Rejection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
9.7 Low Noise Pre-amplifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
10 Low-Power Solutions for Biomedical/Mobile Devices . . . . . . . . . . . . . . . . . . 143
10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
10.2 Design of Wearable Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
10.3 RF Solutions for Mobile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
10.3.1 Ferrite Rod Antenna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
10.3.2 Circular Spiral Inductor Antenna . . . . . . . . . . . . . . . . . . . . . . . . . . 148
10.3.3 Folded Dipole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
10.3.4 Microstrip Antenna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
10.4 Power Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
xxii Contents

10.5 Ultra-Low Power 2.4 GHz RF Energy Harvesting

and Storage System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
11 Applications of Energy Harvesting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
11.2 Building Automation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
11.3 Environmental Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
11.4 Structural Health Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
11.5 Automotive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
11.6 Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
11.7 Solar Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
11.8 Wind Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
11.9 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Chapter 1
Introduction

1.1 Fundamentals

Data acquisition systems (DAQ) are the main instruments used in laboratory
research from scientists and engineers; in particular, for test and measurement,
automation, and so on. Typically, DAQ systems are general-purpose data acquisition
instruments that are well suited for measuring voltage or current signals. However,
many sensors and transducers output signals must be conditioned before that a board
can acquire and transform in digital the signal. The basic elements of DAQ are
shown in Fig. 1.1 and are:
• Sensors and Transducers
• Field Wiring
• Signal Conditioning
• Data Acquisition Hardware
• Data Acquisition Software
• PC (with operating system).
Transducers can be used to detect a wide range of different physical phenomena
such as movement, electrical signals, radiant energy, and thermal, magnetic, or
mechanical energy. They are used to convert one kind of energy into another kind.
The type of input or output of the transducer used depends on the type of signal
detected or process controlled; in other ways, we can define a transducer as a device
that converts one physical phenomena into another one. Devices with input function
are called Sensors because they detected a physical event that changes according to
some events as, for example, heat or force. Instead, device with output function are
called actuators and are used in control system to monitor and compare the value of
external devices. Sensors and transducers belong to category of transducers.

© Springer International Publishing AG 2017 1

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_1
2 1 Introduction

Fig. 1.1 Functional diagram of a PC-based data acquisition system

Table 1.1 Common transducers

Quantity being measured Input device (Sensor) Output device (Actuator)
Light level Photodiode photo-transistor solar cell Lamps–LED–Fiber optics
Temperature Thermistor–Thermocouple Heater–Fan
Force/Pressure Pressure switch Electromagnetic vibration
Position Potentiometer–Encoder Motor
Speed Tacho-generator AC/DC motors
Sound Carbon microphone Buzzer–Loudspeaker

There are many different types of transducers; each transducer has input and
output characteristics and the choice depends on the goal of your system; for
example, from type of signal must be detected and the control system used to
manage it (see Table 1.1).
Sensors produce in output a proportional voltage or current signal in according
to the variation of physical phenomena that are measuring. There are two types
of sensors: active and passive. Active sensors require external power supply to
work; instead, a passive sensors generate a signal in output without external power
supply. Signal conditioning consists in manage an analog signal in order that it
meets the requirements of the next electronics system for additional processing.
Generally, in various applications of control system there is a sensing stage (for
example, a sensor), conditioning stage, and a processing stage. The conditioning
stage can be built, for example, using operational amplifier to amplify the signal
and, moreover, can include the filtering, converting, range matching, isolation, and
1.2 Sensors and Transducers 3

any other processes required to make sensor output suitable for processing stage.
The processing stage manages the signal conditioned in other stages such as analog-
to-digital converter, microcontroller, and so on [1].

1.2 Sensors and Transducers

Transducers and sensors are used to convert a physical phenomena into an electrical
signal (voltage or current) that will then be converted into a digital signal used [2]
for the next stage such as a computer, digital system, or memory board.

1.2.1 Temperature Sensors

Several techniques for detection of temperature are currently used. The most
common of these are RTDs, thermocouples, thermistors, and sensor ICs. The choice
of one for your application can depend on some factors such as required temperature
range, linearity, accuracy, cost, and features. Resistance temperature detectors or
RTD are more commonly known; they are built using several different materials
for the sensing element, for example, the Platinum. Platinum is used for different
reasons: high temperature rating, very stable, and very repeatable. Other materials
used for RTD sensors are nickel and copper.
Thermocouple is composed of two different metals that have a common contact
point where it is produced a voltage (some mV) proportional to the variation of
the temperature. Thermistors are generally composed of semiconductor materials.
There are thermistors with positive and negative temperature coefficient. The ther-
mistors with negative temperature coefficient are used to monitor low temperature
of the order of 10 K [2–4]. The temperature coefficient is defined from the following
Eq. (1.1)

1 dR
˛.t/ D (1.1)
R.T/ dT

In general, a linear curve is used working only over a small temperature range.
To accurate temperature measurements, it is necessary to use the Steinhart–Hart
equation (see (1.2)):

1
D a C b  ln.R/ C c  ln3 .R/ (1.2)
T
where a; b, and c are parameters. The solution of Eq. (1.2) can be written as (1.3):

y 1 y 1
R D e.x 2 / 3 .xC 2 / 3 (1.3)
4 1 Introduction

where
s
 3
b y2
xD C (1.4)
3c 4

and
1
a
yD T
(1.5)
c
Typical values of the resistance of 3000  at room temperature (25 C) are the
following:
• a D 1:40  103
• b D 2:37  104
• c D 9:90  108

1.2.2 Magnetic Field Sensors

Magnetic sensors convert magnetic energy into electrical signals for processing
by electronic system. Magnetic sensors are designed to respond to a wide range
of magnetic field; they are mainly used in different applications, in particular, in
automotive systems for the sensing of position, distance, and speed. For example,
the position of the car seats and seat belts for air-bag control or wheel speed
detection for the anti-lock braking system, (ABS). Magnetic sensors work according
to the Hall Effect (see Fig. 1.2): the production of potential difference (Hall Voltage)
across a conductor where a perpendicular magnetic field is applied [5–9].
The output voltage, called the Hall voltage, (VH ) of the basic Hall Element is
directly proportional to the magnetic field .B/ passing through the semiconductor
material:
 
I
VH D RH  B (1.6)
t

where RH is the Hall Effect coefficient, I is the current flow through the sensor in
Ampere, and t is the thickness of the sensor in mm. Most commercial Hall Effect
devices are manufactured with built-in DC amplifiers, voltage regulators to improve
the sensors sensitivity, and the range of output voltage that it is quite small, only few
microvolts [2, 10–15].
1.2 Sensors and Transducers 5

Fig. 1.2 Hall Effect sensor

1.2.3 Potentiometers

A potentiometer is an electromechanical device that contains a movable wiper arm

with the goal of maintaining electrical contact with a resistive surface; the wiper
is coupled mechanically to a movable linkage. It gives a voltage signal by divider
circuit when voltage is applied across the entire resistance within the potentiometer,
see Fig. 1.3. A variable potential difference can then be produced at a central wiper
arm relative to one of the resistor as the wiper is moved. The wiper is usually made
of a material such as beryllium [1].

1.2.4 Light Detection

Light sensors detect light emitted or given off from an object: such as LED, reflected
from surfaces, transmitted from electronics device, and so on. LED or light emitting
diode, is a solid-state semiconductor that emits light when current through it in the
forward direction. A photoelectric (see Fig. 1.4) sensor is an electrical device that
responds to the change in the intensity of the light falling upon it [16–22].
6 1 Introduction

Fig. 1.3 Potentiometers

Fig. 1.4 Light detection: data acquisition system

There are many sensing situations where space is too restricted or the environ-
ment too hostile even for remote sensors. Fiber optics is an alternative technology in
sensor “packaging” for such applications such as photoelectric sensing technology.
Moreover, fiber optics are flexible, transparent fiber made of glass (silica). It works
as a waveguide to transmit light [1].
1.3 Communications Cabling 7

1.3 Communications Cabling

Field wiring is the physical connection from the transducers/sensors to the hardware.
When the signal conditioning and/or data acquisition hardware is remotely located
from the PC/devices, then it is necessary to use field wiring that provides the
physical link. In this case, it is very important to estimate the effects of the external
noise, especially in industrial environments. In the next paragraph it provides an
estimation of this noise [1].

1.3.1 Noise

One characteristics of all electronics circuits is represented of noise: it is a random

fluctuation in an electrical signal generated by electronic devices. In communication
systems, the noise is an undesired random disturbance of a useful information signal.

1.3.1.1 Thermal Noise

Johnson–Nyquist noise or thermal noise is generated by the random thermal motion

of electrons. Thermal noise is approximately a white noise: the amplitude of the
signal can be described by a Gaussian probability density function.
The root mean square (RMS) voltage due to thermal noise vn , generated in a
resistance R (ohms) over bandwidth f (hertz), is given by
p
vn D 4kB TRf (1.7)

where kB is Boltzmann’s constant (joules per kelvin) and T is the resistor’s absolute
temperature (kelvin).

1.3.1.2 Shot Noise and Flicker Noise

Shot noise in electronic devices consists of unavoidable random statistical fluctu-

ations of the electric current in an electrical conductor. Moreover, Flicker noise,
also known as 1=f noise, occurs in almost all electronic devices, and results from a
variety of effects, though always related to a direct current [1, 23–25].
8 1 Introduction

1.4 Parameters

To properly design a data acquisition system, we must know some important param-
eters. The goal of this section is to describe major system parameters for a better
design in various field of the electronics, in particular, data acquisition systems,
microelectronics, and in the power management system for energy harvesting [23].

1.4.1 Noise

Each measurement generates noise as a combination of more signals. It is the

interference between two terminals. One factor, common-mode noise, indicates the
interferences that appear on both measurements inputs. The majority of common-
mode interference is attributable to 50 Hz (or 60 Hz) power frequency [26, 27].

1.4.2 Settling Time

The settling time of an electronic device is the time elapsed from the application
of an ideal step input to the time at which the value output has entered and
remained within a specified error range. Parameters that can describe settling time
are the following: propagation delay and time required for obtain output value
(Fig. 1.5) [23, 28, 29].

Peak Time

Slep Response
0.5

Overshoot
0.4
Steady State
desiderated error
amplitude 0.3

0.2

0.1
Settling time

0
0 2 4 6 8 10 12
Time (Sec)
Rise
Time

Fig. 1.5 Settling time

References 9

1.4.3 DC Input Characteristics

It indicates the value of offset voltages, offset currents, and bias current of electronic
devices [30–35].

References

1. Park, J., & Mackqy, S. (2003). Practical data acquisition for instrumentation and system
control Elsevier, Oxford.
2. Lacanette, K. (2003). National temperature sensors handbook. Annali di Matematica Pura ed
Applicata. National Semiconductor.
3. National Instruments. (1996). Data acquisition fundamentals, Application note 007.
4. National Instruments. (1996). Signal conditioning fundamentals for PC-based data acquisi-
tion, Application Note 048
5. Roundy, S. (2003). Energy scavenging for wireless sensor nodes with a focus on vibration to
electricity conversion. Ph.D Thesis, University of California.
6. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006). Micro seismic
electret generator for energy harvesting. Technical Digest PowerMEMS (pp. 133–136).
Berkeley, USA.
7. Sterken, T., Altena, G., Fiorini, P., & Puers, R. (2007). Characterisation of an electrostatic
vibration harvester, EDA Publishing Association.
8. Sterken, T., Baert, K., Puers, R., & Borghs, S. (2002). Power extraction from ambient vibration.
In Proceedings of the SeSens (Workshop on Semiconductor Sensors, Veldhoven, Netherlands)
(pp. 680–683).
9. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Electronics, 27(2), 803–815. ISSN: 0885-8993.
10. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning a
resonant energy harvester using a generalized electrical load. Journal of Smart Materials and
Structures, 19, 055003.
11. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
12. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
13. Szarka, G. D., Burrow, S. G., & Stark, B. H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
14. Maurath, D., Becker, P. F., Spreeman, D., & Manoli, Y. (2012). Efficient energy harvesting
with electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6)
15. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
16. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies—state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
17. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. In Proceedings of the World Congress on Engineering (Vol. 2)
10 1 Introduction

18. Zhu, D., Tudor, M. J., & Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
19. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy
harvesters—a classification approach. Micromachines, 4(2), 168–196. http://www.mdpi.com/
2072-666X/4/2/168. Accessed 20 Jan 2015.
20. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of applied electromagnetics
(6th ed.). Prentice Hall.
21. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
22. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
23. Taylor, J. (1986). Computer-based data acquisition system. Instrument Society of America,
USA.
24. Di Paolo Emilio, M. (2013). Data acquisition system, from fundamentals to applied design.
New York: Springer.
25. Roundy, S., Wright, P., & Pister. K. (2002). Micro-electrostatic vibration-to- electricity
converters. In Proceedings of ASME international mechanical engineering congress and
exposition (IMECE) (Vol. 220, pp. 17–22).
26. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: Self-sufficient energy
supply for micro systems. In Proceedings of the 16th international conference on thermo-
electrics (pp. 575–577).
27. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. Micro
IEEE, 21(3), 30–42.
28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008). A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
30. Razavi, B. (2002). Design of analog CMOS integrated circuits. McGraw-Hill
31. Razavi, B. (2008). Fundamentals of microelectronics. New York: Wiley.
32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University.
33. Razavi, B. (2002). Design of integrated circuits for optical communications. McGraw-Hill.
34. Hurst, P. J. (2001). Analysis and design of analog integrated circuits. New York: Wiley.
35. Spies, P. (2015). Handbook of energy harvesting power supplies and applications. CRC Press
Book, France.
Chapter 2
The Fundamentals of Energy Harvesting

2.1 What’s Energy?

Energy is the capacity to do work; in the physics field, work is something resulting
from the action of a force such as that of gravity. In Nature there are different types
of energy: the most classic case is solar energy and all that energies come from the
universe in the form of cosmic rays, X-rays, gravitational waves, dark matter, etc. A
system that produces energy can be represented from a kite that “floats” in the clouds
by means of wind, or a wave of light is passing through a space. According to the
energy conservation law, for example, one of the first laws of thermodynamics, the
total energy of a system is conserved, although it can be transformed into another
form. Two billiard balls can collide, for example, and energy transformations are
involve with sound and heat at the contact point: this phenomena is derived from
energy conservation law after the collision. In few words, all forms of energy can
be converted into another. This had already begun when the man (or woman) lit
the first fire by burning wood with the transformation of the chemical energy of the
molecules in the form of heat. The energy transfer is based on energy conservation.
Other examples, a battery that generates electrons from chemical reactions, a toaster,
the automobile, and many others. The sound is a form of kinetic energy: it is
caused from vibration of the air molecules described as mathematical models. This
vibration energy is transformed into electrical pulses that can be interpreted from the
human as sound wave. In some systems such as that for the production of nuclear
energy, the atoms are involved in multiple processes: the atoms of the nuclear fuel
are divided by releasing the creation of thermal energy which is capture as water
vapor to drive a kinetic energy generator. Subsequently, a motor turns it into a
current flow to provide power supply. Renewable energy (replenished naturally) is
generated from natural sources such as sunlight, wind, rain, tides, and geothermal
heat, which are renewable (that are replenished naturally). Alternative energy is a
term used for an energy source that is an alternative to the use of fossil fuels with a
low environmental impact [1–6]. In the International System of units (SI), the unit

© Springer International Publishing AG 2017 11

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_2
12 2 The Fundamentals of Energy Harvesting

of energy is the joule, named thanks to James Prescott Joule. It is a derived unit and
matched to the energy expenditure (or work) by applying a force of one Newton for
a distance of one meter. However, energy is also expressed in many other units that
are not part of the SI, as ergs, calories, British Thermal Units, kilowatt hours, and
kilocalories, they require a conversion factor when expressed in SI units. In classical
mechanics, from a mathematical point of view, energy is a conserved quantity. The
work, a form of energy, is a force over a given distance described by the following
equation:
Z
wD Fds (2.1)
C

Eq. (2.1) tells us that the work is equal to the line integral of the force along a
path C. In the energy field some terms are used: Hamiltonian and Lagrangian. The
total energy of a system can be expressed by Hamiltonian by using motion equation
of William Rowan Hamilton. Another energy-related concept is called Lagrange,
from Joseph-Louis Lagrange. This formalism is mathematically more convenient
than the Hamiltonian for non-conservative systems (such as friction systems). The
Lagrange is defined as the kinetic energy minus the potential energy.

2.2 Why Energy Harvesting?

An energy harvesting system captures the environmental energy and converts it into
electricity. There are many techniques about this field: for example, to capture the
energy lost or dissipated as heat, light, sound, vibration, or movement, then using
special electronic circuits to manage the collection energy and then transform it into
electrical signal.
This is important, because our existing electrical infrastructure is extremely
wasteful in its use of energy. For example, some today’s technologies used in the
production of electricity are not energy efficient. The old incandescent bulbs, now
they are no longer sold, transform into heat a good percentage of electricity.
An energy harvesting system can be described as in Fig. 2.1. The blocks are
described in the following points:
• Energy transducer used to convert ambient energy into electrical energy of
input. Environmental sources of energy available for the conversion may be
the following: heat (thermoelectric modules), light (solar cells), RF radiation
(antennas), and vibration (piezoelectric).
• Rectifier and super capacitor: a rectifier and an optional storage system for energy
management.
• Voltage regulator: a controller system for adapting the voltage level to the
requirements of the powered device.
2.3 Free Energy 13

Input Energy

Transducer Rectifier Capacitor

Load Voltage
Regulator

Storage

Fig. 2.1 General system of energy harvesting

• Optional energy storage element: depending on the requirements of application,

it is possible to use a battery as an energy storage element. In some systems
it can be activated at intervals of time, in others it is powered (or recharged)
permanently.
• Load: the impedance of the system to power. It may have different ways of energy
consumption making the whole system work in low-power mode.
The electrical energy obtained from an energy harvesting system is very small
(about 1 W/cm3 to 100 mW/cm3 ) and therefore has a working point in low-power
mode. A typical electronic load is constituted by a sensor, a microcontroller, and a
wireless transceiver (Fig. 2.2). Energy consumption is about of uA for the first two
components and a few mA for the transceivers. These considerations will be used in
the design of the energy harvesting system.

2.3 Free Energy

The main energy sources “freely used” are solar, mechanical, and thermal. The self-
powered devices are normally of small dimensions that belong to the category of
wearable devices or otherwise forming part of the Internet of Things (IoT) system.
A possible comparison can be made in terms of power density per unit volume. In
Table 2.1 are summarized the main sources of energy [7–13].
14 2 The Fundamentals of Energy Harvesting

Fig. 2.2 Typical electronic

load
Sensor

Microcontroller (MCU)

Transceiver

Table 2.1 Main energy Energy Category Harvested power

sources of energy harvesting
Human Vibration/Motion 4 W=cm2
Industry Vibration/Motion 100 W=cm2
Human Temperature 25 W=cm2
Industry Temperature 1–10 W=cm2
Indoor Light 10 W=cm2
Outdoor Light 10 mW=cm2
GSM/3G/4G RF 0.1 W=cm2
Wi-Fi RF 1 W=cm2

The light is a source of ambient energy available for low and high power
electronic devices. A photovoltaic system generates electricity by converting the
solar for a large number of applications in the various range of power. The sunlight
varies on the surface of the earth, depending on weather conditions and the position
expressed in terms of longitude and latitude. For each position there is an optimal
tilt angle and orientation of the solar cells in order to obtain the maximum radiation
for powering high power systems; these conditions are not suitable in the case of
small solar cells for the wearable electronics where there are not landmarks and the
design is done in according to the general case [14–20]. The sun radiates towards
the earth’s surface with a power density of at least 1350 J/m2 , with a total power on
earth about 170 * 109 MW.
As in almost all the transformations, kinetic energy is the base of the harvesting
in terms of movement of particles such as photons (sun) or generic waves. The
movement or deformation is converted into electrical energy in three main modes:
inductive, electrostatic, and piezoelectric [21–26].
2.4 Power Management Unit 15

Table 2.2 Vibration sources

Source Peak acceleration (m/S2 ) Frequency (Hz)
HVAC vents 0.2–1.5 4 50/60
Microwave oven 2.3 121
Dryer 3.5 121
Notebook with CD/DVD 0.6 75
Washing machine 0.5 109

The vibrations is the energy source for mechanical transducers and are charac-
terized by two parameters: acceleration and frequency. Table 2.2 visualizes a list of
peak accelerations and frequencies for different vibration sources in the industrial
field. From these data can be noted that the vibrations of industrial machines have
accelerations between 60 and 125 Hz. There is another possibility to use human
body as a source of vibrations. The vibrations associated with the human body have
accelerations with frequencies below about 108 Hz.
The Human Walking, for example, is one of the activities that have more
energy associated for the production of electrical signals. Two power modes can
be distinguished: active and passive. The active power of electronic devices occurs
when the user needs to do a specific work to power the device. The passive mode,
instead, is when the humans must not do any works than their daily activities: finger
movement, walking, heat of the body, etc.

2.4 Power Management Unit

The piezoelectric modules and the transducers operate with an output voltage of
the order of mV, which changes in according to the environmental conditions
and materials used. The electronic circuits, such as microcontrollers or wireless
transceivers which are very often used in power supplies energy harvesting, gener-
ally work with a supply voltage of between 1.8 and 5 V. They need constant power
to maximize their performance. The ripple oscillations declass the performance in
terms of parameters such as noise figure and accuracy. The property to suppress
that noise in a power supply line is expressed by the PSRR. PSRR is introduced to
indicate the amount of noise introduced from a power supply, it stands for “Power
Supply Rejection Ratio” expressed in dB as the ratio of the variation of the supply
voltage in an operational amplifier and the equivalent output voltage (differential).
PSRR is the main design parameter in modern SoC [27–32]. In order to limit the
oscillations, different circuits of power management such as boost converter are
used. The problem of the threshold voltage is reflected in the possibility of not
power a circuit if it doesn’t exceed 0.3 V. Various techniques (star-up circuits) are
used during the startup when the battery is not present in the system. After a level
transition, as soon as the converter provides stable voltage to the circuit, the start-
up circuits are disabled. These circuits are used primarily with thermogenerator.
16 2 The Fundamentals of Energy Harvesting

Another important aspect is the system impedance. With a certain power, the source
must provide maximum power for a given load. To adapt the impedance, Tracker
MPPT circuits such as those employed in the photovoltaic panels are used. In the
switching regulators, frequency variation is reflected in an input impedance change.
In this way, the tracker controls the input resistance to achieve maximum transducer
power. For a piezoelectric, to extract more energy implies a perfect layout of power
management. A transceiver requires a certain time and then a current to run a
data transmission in a specific time. These considerations can be expressed as the
following:

1
Ta D (2.2)
dr  D 1m
n

where dr is the data flow of data rate (bytes/s), D are the bytes of data to be
transmitted, n are the bytes of a data package, and m is the length of the package.
In the procedure for sending and receiving data, all necessary blocks are activated
unlike the sleep mode. The average current required by the transceiver is the
following:

Isleep Tsleep C IT TT
ID (2.3)
Ts

where Isleep is the current absorbed by the transceiver in the sleep mode with the
corresponding sleep time, IT is the current consumed during a transmission time TT ,
and Ts is the transmission period which is the sum of the sleep and transmission
time. The important parameters in the selection of a sensor for energy harvesting
applications are the current consumption in both active and passive mode, the
average power and the sleep time. Sensors can provide an analog or digital output.
The I2C/SPI bus provides a direct interface for transmitting data to digital circuits
such as microcontroller. The sensitivity of the sensor is the amount of variation
of the output signal in according to the variation of the measured parameter. A
conditioning circuit to manage the output voltage before being sent to a control
device is required. The response bandwidth of a sensor is expressed in Hertz and
is the maximum rate at which the sensor can work correctly. The microcontrollers
have different working operation mode in according to the current consumption.
In the active mode the consumption is obviously higher than passive mode with
some circuit parts turned off. In the active mode of low-power microcontrollers,
the current consumption is higher than passive mode and all the clocks are active,
while in the low-power consumption mode the CPU and some of the internal
clocks are disabled. In Fig. 2.3 is visualized a generic block diagram of a low-
power microcontroller, while Fig. 2.4 shown a typical current profile for a wireless
transceiver. In an general energy harvesting system, the transceiver is in standby
mode in most of the time to keep the average power consumption to a minimum.
When the data are to be transmitted, the transmission mode is activated and then the
maximum peak current is consumed [33–35].
2.5 Storage Systems 17

ANALOG DIGITAL I/O

PERIPHERALS
TIMERS
+
OP AMPS POWER UART
– MGMT
REGULATOR
SPI Bus

ADC
PORTS

CONTROLLER CORE
CLOCK
CPU JTAG FLASH RAM
SYSTEM

INTERRUPTS

Fig. 2.3 Block diagram of a low-power microcontroller

Standby Synthesizer Total transmitter Standby Time

enabled enabled

Fig. 2.4 Typical current waveform for a wireless transceiver

2.5 Storage Systems

In an energy harvesting system, the output voltage is not constant in time. It’s
very important to have the same amount of constant energy over time, a bit as in
photovoltaic devices. That means have the same average power in the time interval
in according to load system. An energy storage element is not necessary if the power
consumption of the electronic device is always less than the power generated by
energy harvesting generator that is actuated only when there is power generated.
18 2 The Fundamentals of Energy Harvesting

For the rest of the cases, an energy storage element is required such as a battery.
The goal is to present a way to compute the initial storage element charge before the
commissioning and the maximum amount of energy needed to store. First, you must
define the energy supplied by the transducer element of energy storage and energy
consumed by the load in a mathematical way must be defined. The average power
can be defined as measuring in the time interval T:
Z
1 T
ps D Ps .t/dt (2.4)
T 0

Considering Tli the time duration during which Ps is less than ps and Thl is the
opposite case, i.e., Ps greater than ps , we can define a deficit (minimum) energy for
the first case, while in the second case an excess (maximum) energy is defined.
The energy collected by the transducer and sent to the element of energy storage
will be within a certain margin:
Z
1 T
Emin  Ps .t/dt  Emax (2.5)
T 0

The goal for a wearable device, such as also a node of WSN, is to eliminate
the need to replace or recharge the battery. Therefore, it is necessary to ensure that
the battery is always maintained to the energy required by the electronic device. To
ensure this, the total energy must be greater than that consumed by the load, in other
words (with B the energy stored in the battery):
X
E>0 (2.6)
X
EB (2.7)

The energy available in the system is equal to the initial energy stored in the
energy collected by the transducer less battery energy consumed by the load and
any losses due to parasitic factors:

X Z T
E D B C Es  El  Pleak Tdt (2.8)
0

where Pleak is the leakage power of the storage element. In the case of a load with
different power consumption modes as displayed in Fig. 2.4, the calculation of the
required power is done by considering the following equation:

X
N
D Ti  (2.9)
iD1
References 19

Ti is the time interval where the consumption power is Pi . Generally the following
equation can be defined:
X
pl D Pi Ti (2.10)

To acquire data from remote locations, the sensor nodes designed for the Internet
of Things (IoT) must be able to function for as long as possible on a single battery
charge. In an ideal approach, it would not need a battery because its existence can
complicate the management of the system. One major problem is that the power is
difficult to “catch,” it comes as a very low level but with phase problems to resolve.
Accordingly, specialized techniques are required for the inputs, which include a
boost converter capable of handling the low-voltage sources, high impedance, and
other characteristics of many energy harvesting modules. Furthermore, circuits such
as boost converters may introduce high frequency noise that may disturb radio
communications.

References

1. Park, J., & Mackqy, S. (2003). Practical Data Acquisition for instrumentation and system
control. Elsevier, Oxford.
2. Lacanette, K. (2003). National temperature sensors handbook. Annali di Matematica Pura ed
Applicata. National Semiconductor.
3. National Instruments. (1996). Data acquisition fundamentals, Application note 007.
4. National Instruments. (1996). Signal conditioning fundamentals for PC-based data acquisi-
tion, Application Note 048
5. Taylor, J. (1986). Computer-based data acquisition system. Instrument Society of America.
6. Di Paolo Emilio, M. (2013). Data acquisition system, from fundamentals to applied design.
New York: Springer.
7. Roundy, S., Wright, P., & Pister. K. (2002). Micro-electrostatic vibration-to- electricity con-
verters. In Proceedings of asme international mechanical engineering congress and exposition
(IMECE) (Vol. 220, pp. 17–22).
8. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: self-sufficient energy
supply for micro systems. In Proceedings of the 16th international conference on thermo-
electrics (pp. 575–577).
9. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. Micro
IEEE 21(3), 30–42.
10. Roundy, S. (2003). Energy scavenging for wireless sensor nodes with a focus on vibration to
electricity conversion. Ph.D Thesis, University of California.
11. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006). Micro seismic
electret generator for energy harvesting. Technical Digest PowerMEMS (pp. 133–136).
Berkeley, USA.
12. Sterken, T., Altena, G., Fiorini, P., & Puers, R. (2007). Characterisation of an electrostatic
vibration harvester, EDA Publishing Association.
13. Sterken, T., Baert, K., Puers, R., & Borghs, S. (2002). Power extraction from ambient vibration.
In Proceedings of the SeSens (Workshop on Semiconductor Sensors, Veldhoven, Netherlands)
(pp. 680–683).
20 2 The Fundamentals of Energy Harvesting

14. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Electronics, 27(2), 803–815. ISSN: 0885-8993.
15. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning a
resonant energy harvester using a generalized electrical load. Smart Materials and Structures,
19, 055003.
16. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
17. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
18. Szarka, G. D., Burrow, S. G., & Stark, B.H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
19. Maurath, D., Becker, P. F., Spreeman, D., Manoli, Y. (2012). Efficient energy harvesting with
electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6), 1369–1380
20. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
21. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies—state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
22. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. In Proceedings of the World Congress on Engineering (Vol. 2)
23. Zhu, D., Tudor, M. J., Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
24. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy
harvesters—a classification approach. Micromachines, 4(2), 168–196. http://www.mdpi.com/
2072-666X/4/2/168. Accessed 20 Jan 2015.
25. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of Applied Electromagnet-
ics (6th ed.). Prentice Hall, USA.
26. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
27. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008). A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
30. Razavi, B. (2002). Design of analog CMOS integrated circuits. McGraw-Hill
31. Razavi, B. (2008). Fundamentals of microelectronics. New York: Wiley.
32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University.
33. Razavi, B. (2002). Design of integrated circuits for optical communications. McGraw-Hill.
34. Hurst, P. J. (2001). Analysis and design of analog integrated circuits. New York: Wiley.
35. Spies, P. (2015). Handbook of energy harvesting power supplies and applications. CRC Press
Book, France.
Chapter 3
Input Energy

3.1 Mechanical Energy

The possibility of avoiding the replacement of the batteries is very attractive

especially for the network WSN sensors, where the maintenance costs are high.
Other fields of application is the biomedical field where through piezoelectric
sensors can be implement touchable sensors. Recent research also includes the
conversion of energy from the occlusal contact during mastication by means of a
piezoelectric layer and the heartbeat (Fig. 3.1).
The mechanical energy is present in nature in the form of various examples, for
example, in vibrating structures or fluid is flowing along the structures. Wherever
there is a mass there is a high potential for energy harvesting applications. The
resulting energy must be oscillating in a periodic shape as those of a motor, or
causal as in most of the natural phenomena. Simple examples of energy harvesting
by using fluids are represented by windmills: exploiting the fluid mechanical
energy, electricity is generated by electromechanical generators. The main source
of mechanical energy for energy harvesting is vibrational. Any system, whatever it
is, is subjected to vibrations that somehow can be harvested to generate electricity.
Simple examples are body movements, as well as those of animals or other
vibrations resulting from movement of building structures [1–6]. In mechanical
vibrations, the mathematical expression for the state variable of q.t/ varies in time
following a harmonic motion: the latter is the most simple representation described
with trigonometric functions. The variable q.t/ represents a displacement, force, or
pressure angle, and can assume the following expression:

q.t/ D jAj cos.!t C / (3.1)

where jAj is the amplitude of the motion, ! the frequency in radians per second,
and  is the phase angle. To characterize a motion, all parameters are important:
first of all not only the amplitude but also the frequency that identifies the band

© Springer International Publishing AG 2017 21

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_3
22 3 Input Energy

Energy Harvesting

Application field Working principle

Solar Thermal RF Motion

Mechanical-
Photovoltaic Seebeck RF Piezo
Electrostatic

DC output
AC output

Fig. 3.1 Sources of energy for harvesting

of frequencies with one dominant and the other of a higher order. The vibrations
excited by humans are under the 2 Hz; those between 20 Hz and 20 kHz are audible
to the human ear and therefore are more likely to be caught. The lowest natural
frequency is called the fundamental frequency or dominant. It is useful to know
the natural frequencies of a structure to tune an energy harvesting system exactly
at a given frequency. The transducers measure the movement and turn it into an
electric signal in the time domain, and then with an FFT in the frequency domain.
The frequency analysis allows to determine the various frequencies and damping
of the system parameters. One or more transducers of movement measure the
output vibration. The signals measured in the time domain are then transformed
into the frequency domain, and modal analysis calculates the natural frequencies,
eigenvectors, and damping parameters of the system. The ability to extract energy
from human activities has been the subject of study for many years. It’s a fact: the
movement of the fingers (a few mW), limb movement (about 10 mW), exhalation
and inhalation, (about 100 mW), and walking (some W). The piezoelectric trans-
ducer offers a collection of higher power density than other electrostatic, especially
thanks to the advantages offered by the MEMS implementation. The maximum
power that is supplied to the energy harvesting devices depends on the frequency
and acceleration of the vibrating system, as well as by the size of the device. A
second-order model is used to analytically describe the process. From Fig. 3.2 a
3.1 Mechanical Energy 23

Fig. 3.2 Representation of

the vibration

mass is suspended with a constant k connected to a rigid frame, with d is indicated

the coefficient of damping. The frame is exposed in motion x.t/, thereby obtaining
a relative movement y.t/ of the mass [7–10]. The maximum power occurs at the
natural frequency of oscillation, namely:
p
!n D k=n (3.2)

The damping coefficient can be divided into electrical and mechanical: e and p .
The maximum power is generated at the frequency of resonance and can be
expressed in terms of excitation X:

e
Pmax D m!n3 X 2 (3.3)
4.e C p /2

This model shows how a highly damped system extracts energy over a wide
frequency band. A less damped system would extract more power, but in a smaller
frequency range.
The characterization of the vibrations can occur by means of sensors denomi-
nated accelerometers. Their advantage is to measure the absolute value and then
used directly on the structure. Piezoelectric accelerometers are the most used due to
the high dynamics, small size, and immunity to noise factors. Other techniques are
carried out with piezoresistive or capacitive techniques that can measure static accel-
eration, and electrodynamic accelerometers used to measure very low frequencies.
The accelerometers normally measure the linear acceleration, however, there are
additional sensors which measure the acceleration in the three degrees of freedom.
The block diagram of an accelerometer is displayed in Fig. 3.2, where a mass m
experiences a dynamic force given by the second law of Newton, transformed into
24 3 Input Energy

an electric signal by the piezoelectric effect. The piezoelectric effect is the ability of
a material to respond with a change of electricity in response to a mechanical event
(stress). A piezoelectric characteristic is the reversibility of the effect: namely, the
generation of electrical energy by means of a mechanical stress, and vice versa. The
piezoelectric effect is very useful in many applications are involving the production
and detection of sound, generation of high voltages, microbalance, and in optical
systems. There are many materials, both natural and artificial, which present a
series of piezoelectric effects. Some piezoelectric materials include natural berlinite
(structurally identical to quartz), brown sugar, topaz, and tourmaline. An example of
artificial piezoelectric material comprises barium titanate and lead titanate zirconate.
In recent years, due to growing environmental concerns regarding the toxicity in
devices containing lead and RoHS directives, there has been a push to develop lead-
free piezoelectric materials. To date, this initiative has led to the development of new
lead-free piezoelectric materials [11–17]. The analytical model for an accelerometer
can be described by the following equation:

y000 C 2!o y00 C !o2 D g (3.4)

where  is the damping coefficient linked to the inverse of the resonance fre-
quency !o :
p
!0 D k=m (3.5)

In particular, the piezoelectric sensors are used with high frequency sound in
ultrasonic transducers for medical imaging and non-destructive industrial controls.

3.2 Thermal Energy

In any part of the earth there is a thermal gradient and then we can use this to produce
energy. The thermal energy is available primarily in the industrial sector (machinery,
pipes, and vehicles), in buildings, and in the human body. The temperature gradient
exploits the Seebeck effect to generate electricity. To keep the thermal gradient is
need of a heat source on one side and a heat sink on the other. The thermal energy
is a byproduct of other forms of energy such as chemical and mechanical.
When the two ends of a conductor are at different temperatures (Fig. 3.3),
a potential difference is produced between the two ends. Seebeck had thought
to have created a new magnetic field mode but in reality it was an electrical
voltage. The magnitude of the electromotive force V generated between the two
junctions depends on the material and the temperature through the following linear
relationship as a function of the Seebeck coefficient S:

V D ST (3.6)
3.2 Thermal Energy 25

Fig. 3.3 Electronic electrons

dissemination from a cold to
hot zone e e e
T2 e e e e e T2<T1
e e e e e e e
e

- +
V

Instead, the maximum power collection is given by the following relation:

 
A 1 S2 .Th  Tc/2
Pe D (3.7)
l 4 m

where A is the section of the material,  the resistivity of the material, l the length
of the thermocouple and indicated with T are the temperatures of the hot (h) and
cold (c) zone. It’s necessary to measure the temperature of each side, minimizing
the local voltage drop. The temperature is measured by a temperature sensor such
as a thermocouple or RTD. The thermistors change the electrical resistance as
function of the temperature and represent a good compromise between various
sensors present on the market in terms of cost, accuracy, and response time. A
thermal resistor or thermistor changes its electrical resistance with the temperature.
There are thermistors with a positive temperature coefficient (PTC) and other with
a negative temperature coefficient (NTC). In mathematical terms, the resistance is
expressed in the following way:
.B.T0 T/
R.T/ D R0 e TT0 (3.8)

where R.T/ is the resistance at temperature T, B is a constant of the thermistor

sensor, and T0 and R0 are, respectively, the ambient temperature of 25 ı C and the
corresponding resistance. The temperature coefficient is expressed in the following
way:
1 dR B
˛D D 2 (3.9)
R dT T
The resistance of the equation is valid in a linear range between ln.R/ and
1=T, in all other intervals is used the Steinhart–Hart relation with a third-order
approximation:

1
D a C bln.R/ C dln3 .R/ (3.10)
T
26 3 Input Energy

a B b
R1 R2 +
R1
+
A C
+ Vout
R4 R3 −
Vin Rt

V D +
in

R2 Vout
− −

−
Fig. 3.4 (a) Wheatstone bridge and (b) voltage divider with the resistance of thermistor Rt

The thermistors are used in a Wheatstone bridge or voltage dividers solutions to

express the electrical voltage as a function of temperature change, or variation of
electrical resistance (Fig. 3.4).
In the Wheatstone bridge configuration where R3 is the thermistor and R1 D R4
and R2 D R0 , we have the following expression for the output:

Vout R1
D Vin (3.11)
RT .R1 C R0 /2

The heat generator of an assessment must be performed under certain tempera-

ture conditions, so it is necessary to establish the temperature of both sides through
two thermocoolers. The corresponding temperature is controlled by means of PID
adjusted to achieve a feedback and compensation control [18–23].

3.3 Electromagnetic Energy

The RF energy is currently used as cornerstones of transmitters around the world,

thinking of mobile phones, Radio, base stations, TV, etc. Obviously, the energy
levels are under about SAR levels. The ability to collect RF energy allows the
wireless charging of low-power devices, at the same time allows to set limits in
the use of the battery. The devices without battery, for example, can be designed
to operate in certain time intervals or when the super-capacitor has accumulated
sufficient charge to activate the electronic circuitry (Fig. 3.5).
The appeal of collecting RF energy is essentially derived from the fact that it
is free energy. Over the years, the number of broadcasting stations is increased
significantly, many market analysts estimate the number of mobile subscriptions
3.4 Space Radiation 27

Antenna

RF - to - DC Power Output
Conversion Conditioning

Fig. 3.5 RF energy harvesting

is growing. The mobile phones are an excellent source of RF energy and could be
a source for providing energy-on-demand for a variety of applications. Thinking of
how many wireless access points can find in a city center, all RF sources that we can
use to recharge, for example, our smartphones. The devices such as Powerharvester
receivers convert the RF energy continuously, working with standard 50  antennas
and provide the ability to maintain the RF-to-DC conversion efficiency in a wide
range of operating conditions, including variations of power in input and output
load resistance. The RF energy can be used to charge a wide range of low-power
devices. At close range, we can recharge GPS or wearable medical sensors and a
wide range of consumer electronics. Depending on the required power, the power
can be sent in a continuous mode, on a scheduled basis, or on-demand [24–30].

3.4 Space Radiation

The radiation is a form of energy composed of high speed particles. Typically there
are two types of radiation: ionizing and non-ionizing. The first they have a lot of
energy, enough to interfere with the atom and the electron to modify the state, unlike
the other radiation.
Ionizing radiation includes gamma rays, x-rays, protons, electrons, neutrons,
alpha and beta particles; and of non-ionizing radiation includes microwaves, visible
light, infrared, and radio frequency waves. The spatial radiation is ionizing and
consists of highly energetic charged particles.
The solar wind is a stream of charged particles that originated in the upper layers
of the sun. It consists mainly of electrons and protons and blows constantly from the
surface of the sun. The energy possessed by these charged particles is between 1.5
and 10 KeV. The average speed of these particles is about 145 km/s. This speed is
lower than the solar escape velocity of 618 km/s. However, some of the particles are
able to have sufficient energy to reach the terminal velocity of 400 km/s. So, they are
allowed to create the solar wind. At the same temperature, the electrons reach escape
velocity because of their smaller mass which helps to build up an electric field that
28 3 Input Energy

accelerates the charged particles further. The solar wind can travel up to the distance
of 75 astronomical units and the density can vary from 1 to 10 particles/cm3 .
Cosmic rays are high-energy radiation to which the origin of the universe. They
are mainly composed of high-energy protons and atomic nuclei ranging from the
lightest to the heaviest. They also contain high-energy electrons, positrons, and
other subatomic such as muons. The 90 % of cosmic ray nuclei are protons and
about 9 % are alpha particles. The main sources of cosmic rays are supernovae of
massive stars, active galactic nuclei, quasars, and gamma-ray bursts. The highly
charged particles of cosmic rays travel at nearly the speed of light. Most of galactic
cosmic rays have energies ranging from 100 MeV to 10 GeV. Since, cosmic rays are
electrically charged, they are deflected by magnetic fields, and their directions are
random without giving the exact idea of their origin [31–35].

3.5 Solar Radiation

The Sun is a star classified as “yellow dwarf” (surface temperature of about

5570 ı C) consists primarily of hydrogen (74 % of the mass) and helium (24–25 %
of the mass). The energy radiated by the sun derives from a nuclear fusion reaction:
every second approximately 6:2  1011 kg of hydrogen is transformed into helium,
with a mass loss of about 4.26103 kg; these are transformed into energy according
to Einstein’s relation E D mc2 , resulting in an energy of 3846  1026 J, which
corresponds to a radiated power Ptot D 3846  1026 W. It’s possible to approximate
the sun to a point source placed at a distance from the earth equal to the average
distance during the year (D D 149  106 km) that radiates energy in a manner
uniformly distributed throughout the solid angle (Fig. 3.6). Consequently, the energy
flow is constant in the unit of surface of radius D and its value is equal to one divided
by the entire surface of the sphere.

1400

1380

1360

W/m2
1340

1320

1300
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec

Fig. 3.6 Mean solar radiation during 1 year

3.5 Solar Radiation 29

The power of the incident solar radiation on a surface is called the solar radiation
and is measured in W=m2 . Some considerations about the solar radiation:
• It may be seem surprising that from such big numbers we’ve got to calculate a
value for the power available per square meter outside the atmosphere so close to
the typical values cited in the literature;
• Also, how is it possible to know with such precision the power produced in the
Sun?
• Actually it proceeds in exactly the opposite: it measures the power per square
meter available outside of the atmosphere and, by reversing the calculation, it is
estimated the power radiated by the sun.
The energy received on Earth from the Sun in a year is about 10,000 times the
current energy needs. If we consider a conversion efficiency of 10 % is that it would
require an area of solar panels equal to the area of England, that is one-thousandth
of the Earth’s surface exposed to the Sun (England area = 130.325 km2 , the Earth’s
surface exposed to the Sun D 127:796  106 km2 , and the ratio is 1/980.6).
Conventionally, we define the constant solar radiation outside the atmosphere
equal to 1360 W/m2 . It is interesting to analyze the spectrum of the radiation
emitted by the Sun shown in Fig. 3.7. In particular, some considerations are reported
below:
• The specter measured outside the atmosphere corresponds with high precision to
the theoretical emitted by a black body whose surface temperature is that of the
sun.

2.5
UV Visible Infrared

2 Sunlight top

1.5 Blackbody spectrum

1
Sea level
H2O

0.5 H2O
O2 Absorption Bands
H2O CO
2 H2O
O3 H2O
0
250 500 750 1000 1250 1500 1750 2000 2250 2500

Fig. 3.7 Spectral irradiance W/M2 =nm as function of wavelength (nm)

30 3 Input Energy

• The bulk of the emitted radiation is concentrated in the visible wave lengths
(approximately between 400 and 700 nm).
• The wavelength of the radiation is very important for the interaction with the
photovoltaic cell.
• The spectrum at sea level is significantly different: in general it is attenuated,
but mainly much more jagged and full of “holes”: these are due to the fact that
certain atomic species and the water present in the atmosphere completely absorb
the radiation of specific wavelengths.
As for the spectrum and the power available at sea level they vary depending
on the atmospheric conditions and of the atmosphere traversed thickness, which in
turn depends on the position of the sun in the sky. The minimum thickness is at its
zenith, that is when the Sun occupies a position vertically above the observer’s head
[30–35].

3.5.1 Photovoltaic Cell

The solar cells exploit the photoelectric effect, i.e., the effect for which a suitable
material, for example, a metal, emits electrons by light radiation. The mechanism
can be explained by assuming to represent the light like a beam of particles called
photons (alternate interpretation of the light wave phenomena).
Each photon, characterized by a certain wavelength, carries a well-defined
amount of energy according to the relation:

hc
ED (3.12)

where E is the energy carried, h is Plank’s constant (equal to 6.626 * 1034 J s), c is
the speed of light, and is the wavelength of the radiation. When a photon strikes
an electron of the metal, it is absorbed and the electron receives energy if the photon
energy is greater than the metal work function: the electron is free to break away;
otherwise the energy is dissipated as heat. Semiconductors are materials in which the
vast majority of external electrons bound to the atoms (that is, those that would be
available to conduct current) occupy the valence band, and only a minimal amount
occupy the conduction band. In semiconductors the valence band and the conduction
band are separated by a well-defined energy gap, for silicon, for example, 1.12 eV.
If we illuminate a silicon surface and the photons have energy higher than that of
the gap between the bands, an electron of the valence band can absorb the photon,
and it can work to acquire the energy and move to the conduction band to conduct
an electric current. A favorable situation is in a particular structure called junction.
It is the union of two semiconductor volumes, each of which has been separately
worked by introducing the impurities in the interior (doping). For example, silicon
is a tetravalent element (each atom is bonded to four other atoms equal, by means
3.5 Solar Radiation 31

of the four valence electrons available external orbital). Adding small amounts of a
pentavalent element (such as phosphorus or arsenic), the fifth electron not engaged
in a link goes to occupy the conduction band, by resulting in an excess of free
negative carriers. There is talk of doping n. Adding small amounts of tetravalent
elements (such as boron), instead, they are generated of unstable bonds with the
surrounding atoms that tend to trap an electron to stabilize, by leaving a positive
charge is not offset gap call, and then an excess of positive charges available to
conduct current. It is important to stress that although there is an excess of positive
and negative charges available to run, a total of the two volumes of doped material
are electrically neutral (the total number of positive charges inside them exactly
compensates the number of negative charges). When the two volumes of the excess
electrons of the n tends zone are contacted to diffuse in the p region; conversely
the excess of p tends to spread in the gaps area n area. This migration produces
an imbalance of electric charge that determines the birth of an electric field that
opposes the diffusion and attracts electrons to zone n and p gaps towards the area.
The equilibrium is reached when the number of charges that moves in one direction
due to diffusion is offset by an equal number of charges that moves in the opposite
direction due to the electric field (drift, Fig. 3.8).
In the single cell, the photocurrent generation coexists with the mechanisms that
regulate the normal flow of current in a p–n junction to vary the voltage at its ends.
The photocurrent has an intensity that depends on many factors: radiation, angle
of incidence, temperature, type of semiconductor, etc. The ideal model is shown in
Fig. 3.9.
The ID has the classic expression of the current of a diode:
qVc
ID D IS .e nKT  1/ (3.13)

where q is the electron charge, k is the Boltzmann constant, T is the temperature

in Kelvin, n is the ideality factor or emission coefficient (between 1 and 2, by
depending on the manufacturing process), and IS is the saturation current: the value
depends on the characteristics of the diode. The current delivered by the cell is
given by:
qVc
IC D IL  IS .e nKT  1/ (3.14)

The typical load output voltage of a cell is of the order of 0.5–0.6 V (not in
case very close to the classical voltage of a diode in conduction). Typical values of
photocurrent for a cell in silicon are of the order of 30 mA/cm2 . One of the most
popular formats for the cells is approximately square in shape and 125 mm side,
for a total surface of about 156 cm2 ; the short-circuit current is therefore at around
4 A. For the intermediate work points between the open circuit and short circuit the
characteristic of the cell varies according to a graph shown in Fig. 3.10.
32 3 Input Energy

DV

– – + +
– – + +
– – + +
P-Zone – – + + N-Zone
– – + +
– – + +
– – + +

–W1 W2 x
Electric field

ρ Charge density +

–W1 W2 x
–

E Electric field

–W1 W2 x

V DV
Voltage –W1 W2

Fig. 3.8 PN junction

The output power curve shown in Fig. 3.10 has a maximum (MPP—Maximum
Power Point) which correspond to the values of current and voltage IMPP and
VMPP. The Maximum Power Point (MPP) changes to vary of the characteristics
of the cell. The manufacturers characterize the cells through the CT temperature
coefficient that indicates the variation of delivered power as function of the
temperature. Typical values are between 0.2 and 0.5 %. The efficiency of the
photovoltaic cell is the ratio of the electric power output and the power incident on
the cell. The conversion efficiency varies greatly at different technology used, but in
general is rather low by ranging from 5 to 8 % for the cells of amorphous silicon to
20 % for multi-junction cells.
References 33

IC

IL ID

VC

Fig. 3.9 Model of a photovoltaic cell

I,P

ISC
IMPP

VMPP VOC V

Fig. 3.10 I/V characteristic of a photovoltaic cell and output power (dashed line)

References

1. Park, J., & Mackqy, S. (2003). Practical Data Acquisition for instrumentation and system
control. Elsevier, Oxford.
2. Lacanette, K. (2003). National temperature sensors handbook. Annali di Matematica Pura ed
Applicata. National Semiconductor.
3. National Instruments. (1996). Data acquisition fundamentals, application note 007.
4. National Instruments. (1996). Signal conditioning fundamentals for PC-based data acquisi-
tion, Application Note 048
5. Taylor, J. (1986). Computer-based data acquisition system. Instrument Society of America.
34 3 Input Energy

6. Di Paolo Emilio, M. (2013). Data acquisition system, from fundamentals to applied design.
New York: Springer.
7. Roundy, S., Wright, P., & Pister. K. (2002). Micro-electrostatic vibration-to- electricity
converters. In Proceedings of ASME international mechanical engineering congress and
exposition (IMECE) (Vol. 220, pp. 17–22).
8. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: self-sufficient energy
supply for micro systems. In Proceedings of the 16th international conference on thermo-
electrics (pp. 575–577).
9. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. Micro
IEEE, 21(3), 30–42.
10. Roundy, S. (2003). Energy scavenging for wireless sensor nodes with a focus on vibration to
electricity conversion. Ph.D Thesis, University of California.
11. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006). Micro seismic
electret generator for energy harvesting. Technical Digest PowerMEMS (pp. 133–136).
Berkeley, USA.
12. Sterken, T., Altena, G., Fiorini, P., & Puers, R. (2007). Characterisation of an electrostatic
vibration harvester, EDA Publishing Association.
13. Sterken, T., Baert, K., Puers, R., & Borghs, S. (2002). Power extraction from ambient vibration.
In Proceedings of the SeSens (Workshop on Semiconductor Sensors, Veldhoven, Netherlands)
(pp. 680–683).
14. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Electronics, 27(2), 803–815. ISSN: 0885-8993.
15. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning a
resonant energy harvester using a generalized electrical load. Smart Materials and Structures,
19, 055003.
16. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
17. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
18. Szarka, G. D., Burrow, S. G., & Stark, B.H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
19. Maurath, D., Becker, P. F., Spreeman, D., Manoli, Y. (2012). Efficient energy harvesting with
electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6), 1369–1380
20. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
21. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies—state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
22. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. In Proceedings of the World Congress on Engineering (Vol. 2)
23. Zhu, D., Tudor, M. J., Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
24. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy
harvesters—a classification approach. Micromachines, 4(2), 168–196. http://www.mdpi.com/
2072-666X/4/2/168. Accessed 20 Jan 2015.
25. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of Applied Electromagnet-
ics (6th ed.). Prentice Hall, USA.
26. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
References 35

27. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008). A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
30. Razavi, B. (2002). Design of analog CMOS integrated circuits. McGraw-Hill
31. Razavi, B. (2008). Fundamentals of microelectronics. New York: Wiley.
32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University.
33. Razavi, B. (2002). Design of integrated circuits for optical communications. McGraw-Hill.
34. Hurst, P. J. (2001). Analysis and design of analog integrated circuits. New York: Wiley.
35. Spies, P. (2015). Handbook of energy harvesting power supplies and applications. CRC Press
Book, France.
Chapter 4
Electromagnetic Transducers

4.1 Introduction

The study of RF signals implies a clear distinction between RF nonradiative and

radiative. The first is based on an inductive coupling, while the second uses the
transmission and reception of radio waves. The harvesting energy is a process where
the energy from the environment goes into a load. With the RF energy transfer,
however, it is the process that uses a dedicated RF source for the power wirelessly.
The transmission of RF signals transmitted in the collection process is involuntary
arising from various systems such as smartphones. The transfer of RF energy is
mostly used in inductive systems, according to the Wi-Fi standards. In these cases
two coils are placed in proximity to obtain RF transference, two coils in the vicinity
form an electrical transformer. In the circuit of Fig. 4.1 is displayed a general scheme
where two coils, one in reception and transmission in the other, form the transfer
system by using a capacitor to rectify the wave shape [1–10].
RF energy transfer over a distance is need to use the radiative transfer: an RF
source (not intentional) is connected to an antenna that emits radio waves. At a
certain distance, a receiving antenna picks up part of the converting waves into an
electrical signal and transferred to a load (Fig. 4.2). The power transfer between the
transmitting and receiving antenna is described by Friis equation.

4.2 Electromagnetic Waves and Antenna

The electromagnetic waves can travel in space or in a dielectric. The sound waves
are examples of mechanical waves, in contrast to the bright ones (photons) that
represent a type of electromagnetic waves. The electromagnetic waves are created
by the vibration of an electric charge: a wave with an electric and magnetic
component is created. The energy transported has a velocity in vacuum equal to that

© Springer International Publishing AG 2017 37

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_4
38 4 Electromagnetic Transducers

Rs Rr
2 2
1k
S Ls Lr
Cr Load
– Cs R
1 1

Fig. 4.1 General layout of an RF energy harvesting. The R block is used to rectify the signal

Antenna Antenna
Field Rectifier Load
T R

Fig. 4.2 Diagram block for RF energy harvesting

of light indicated with the letters c, in a medium speed it is less than c. The transport
mechanism exploits the phenomenon of absorption and re-emission of the wave
energy from the atoms of the material. When an electromagnetic wave affects atoms,
the energy is absorbed with vibration of the electrons that make up the material.
After a short period of vibrational motion, the vibrating electrons create a new
electromagnetic wave with the same frequency of the first electromagnetic wave.
These vibrations occur only for a very short time and the wave motion is delayed
through the medium. Once the electromagnetic wave energy is re-emitted by an
atom, it travels through a small region of space between atoms. Once the next atom
reached, the electromagnetic wave is absorbed, transformed into electron vibration,
and then re-emitted as an electromagnetic wave. As part of the energy harvesting, the
Friis’s equation used in telecommunications engineering provides us the transmitted
power of an antenna in ideal conditions to another antenna at a certain distance.
An antenna is an electrical device that converts electrical energy into radio waves,
and vice versa. It is usually used with a radio transmitter or radio receiver. In
the transmission, a radio transmitter provides a radio frequency oscillating electric
current (that is, a high frequency alternating current (AC)) to the antenna that
radiates the energy from the current as electromagnetic waves (radio waves). In
reception, an antenna intercepts a part of the power of an electromagnetic wave
to produce a small voltage at its terminals where is applied a receiver to amplify
the signal. The electromagnetic waves are described by Maxwell’s equations which
represent one of the most elegant and concise ways to affirm the fundamentals of
electricity and magnetism. From them have been developed most of the industry
working relationships [11–20]. The four equations of Maxwell describe the electric
and magnetic fields arising from the distribution of electrical charges, and how they
change over time. The formulation of the equations is a study of many scientists
and deep insight of Michael Faraday. The first equation expresses the Gauss’ law
for electric fields: the integral of the electric field output on a surface that encloses
a volume is equal to the internal total charge:
4.2 Electromagnetic Waves and Antenna 39

Z
!
 !
 q
EdA D (4.1)

0

The second equation is analogous to the magnetic fields:

Z
! ! 
BdA D 0 (4.2)

The first two equations of Maxwell are integrals of the electric and magnetic
fields on closed surfaces. The other two equations of Maxwell, discussed below,
are integral of electric and magnetic fields around closed curves which repre-
sent the work required to take a charge around a closed curve in an electric field,
and the similar in the magnetic field. The third is the Faraday’s law of induction and
the fourth is the Ampere’s law:
I Z
!
 !  d !
 ! 
Ed l D  BdA (4.3)
C dt
The first term is integrated around a closed line, usually a wire, and provides the
total variation of the voltage on the circuit that is generated by a variable magnetic
field.
I Z
!
 ! d !
 ! 
B d l D 0 .I C
0 EdA/ (4.4)
C dt
In differential form the equations for the electromagnetic waves become the
following:

r  E D 4 (4.5)
1 @B
rxE D  (4.6)
c @t
r BD0 (4.7)
4 1 @E
rxB D JC (4.8)
c c @t
where E is the electric field, B the magnetic field, J the current density, and 
the charge density. After the definition of the electromagnetic waves, now we can
evaluate the engineering or how to transmit them by means of antennas, and then
capture them to produce electric current. To begin with the derivation of the Friis
equation considering two antennas in free space (obstructions nearby) separated by
a distance R (Fig. 4.3).
Now, we suppose a transmission antenna omnidirectional, without loss, and the
receiving antenna is far from the transmission range. Pt is the transmission power
and p the power density (in watts per square meter) of the wave received on the
antenna at a distance R from the transmission antenna:
40 4 Electromagnetic Transducers

TX RX

Fig. 4.3 Transmission and reception layout

PT
pD Gt (4.9)
4R2
where the parameter Gt is the antenna gain. If the receiving antenna has an effective
capture area Aer , then the power received by the antenna is the following:

PT
Pr D Gt Aer (4.10)
4R2
where the effective capture area can be expressed in the following mode:

Aer D G (4.11)
4
And then the Friis’s formula, namely the expression for the power Pr can be
expressed as follows:

PT Gt Gr 2
Pr D (4.12)
.4R/2

To decide on the feasibility of RF energy harvesting, we need to assess the

power levels. Based on the state-of-the-art power consumption of sensors available
on the market, our goal is to power the level of about 100 W. For RF energy
harvesting, the most interesting systems to be explored in Europe are GSM900
(downlink: 935–960 MHz), GSM1800 (downlink: 1805.2–1879.8 MHz), and Wi-
Fi World (2.4–2.5 GHz). All these systems guarantee an excellent presence and size
of antennas very small about of 10–50 cm2 . Many studies have estimated power
density in the GSM900 band equal to a 0.01–0.3 Wcm2 range between 25 and
100 m from a base station. This implies dimensions of antennas of the order of
300–1000 cm2 to achieve power levels of 100 uW. The technological limitations
4.3 System Design 41

are related to the maximum power allowed to transmit and on the limitation of
transmission between transmitter power and antenna gain: their product is known
as EIRP, effective isotropic power [20–35].

4.3 System Design

The RF energy harvesting focuses on the receiving layout that needs some tricks to
be able to feed the load properly. The block diagram of receiving part is shown in
Fig. 4.4.
The general layout is composed of a receiving antenna, a rectifier connected
to an RF-side to the receiving antenna and on the DC side of a load. In general,
an impedance matching circuit is applied between the antenna and the rectifier.
In general there is a DC-DC conversion circuit connected to an energy storage
system (battery or capacitor) which is connected to the load. Considering the
equivalent antenna to a voltage source in series with a resistor, the equivalent circuit
is described in Fig. 4.5 where the load and the conversion part is combined in a load
single YL . The matching network is presented as an L network.

Fig. 4.4 Block diagram of receiving system

jA
R

S
+

– jB YL

IN

Fig. 4.5 Antenna equivalent circuit and L matching network

42 4 Electromagnetic Transducers

L
1 2

Vin Vout
C

Fig. 4.6 LC voltage boosting circuit

The input power is given by the following expression:

1 2
Pin D v Re.Yin / (4.13)
2 in
where Y is the admittance, i.e., the inverse of the impedance. In output, instead, the
power is given by the same expression calculated for YL . To connect the rectifier to
the antenna has been proposed a voltage boosting circuit: a classic example is the
LC circuit (Fig. 4.6).
In this case the gain is expressed in the following way:

vout 1p
gD D 1 C Q2 (4.14)
vs 2

where Q indicates the quality factor of the matching circuit. Whereas the incident
power is given by the following formula:
vs
Pinc D (4.15)
8R2
The output voltage is equal to:

1p
vout D .1 C Q2 /  8R2 Pinc (4.16)
2
The input voltage of the rectifier/multiplier is then dictated by the available power
from the antenna and by the Q factor of the matching circuit. To maximize vout ,
which will be beneficial for the RF-to-DC power conversion efficiency, as well as to
the output voltage DC level, there is need to design a rectifier/multiplier having a real
part of input admittance that is as low as possible. The easiest rectifier is constituted
by a single diode. In general, due to the rapid switching speed, it considers using a
Schottky diode. In Fig. 4.7 an example of circuit with Schottky rectifier diode; in the
circuit is also displayed the equivalent rectifier circuit. The source has an internal
resistance Rg while the Schottky diode is represented by a resistance that expresses
References 43

Fig. 4.7 Equivalent rectifier circuit

the conduction losses in the substrate, while Cj is the parasitic capacitance expresses
by the following relation of proportionality as function of the frequency:

1
Cj .f / / q (4.17)
f
1 

where  is the potential barrier. All these parameters can be determined by using
the data sheet of the single component. The antenna must be connected to a
rectifier/multiplier. Once the frequency band or frequency has been decided, there
is need to start determining the impedance of the input rectifier/multiplier. If we
want to design an antenna without by using an impedance matching circuit, it is
necessary to design an antenna input impedance equals to the complex conjugate
of the impedance of the rectifier input. In addition, we have to design an antenna
with a low resistive part of the input impedance. To achieve this antenna, we must
make the antenna electrically small. In addition, we have to design the antenna that
has a relatively high inductance, which means that we need a small loop antenna.
The RF energy harvesting systems use a recharge battery and an array of antennas
to provide a sufficient energy source to the load.

References

1. Park, J., & Mackqy, S. (2003). Practical Data Acquisition for instrumentation and system
control. Elsevier, Oxford.
2. Lacanette, K. (2003). National temperature sensors handbook. Annali di Matematica Pura ed
Applicata. National Semiconductor.
3. National Instruments. (1996). Data acquisition fundamentals, Application note 007.
44 4 Electromagnetic Transducers

4. National Instruments. (1996). Signal conditioning fundamentals for PC-based data acquisi-
tion, Application Note 048.
5. Taylor, J. (1986). Computer-based data acquisition system. Instrument Society of America.
6. Di Paolo Emilio, M. (2013). Data acquisition system, from fundamentals to applied design.
New York: Springer.
7. Roundy, S., Wright, P., & Pister. K. (2002). Micro-electrostatic vibration-to- electricity
converters. In Proceedings of ASME international mechanical engineering congress and
exposition (IMECE) (Vol. 220, pp. 17–22).
8. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: self-sufficient energy
supply for micro systems. In Proceedings of the 16th international conference on thermo-
electrics (pp. 575–577).
9. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. Micro
IEEE, 21(3), 30–42.
10. Roundy, S. (2003). Energy scavenging for wireless sensor nodes with a focus on vibration to
electricity conversion. Ph.D Thesis, University of California.
11. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006). Micro seismic
electret generator for energy harvesting. Technical Digest PowerMEMS (pp. 133–136).
Berkeley, USA.
12. Sterken, T., Altena, G., Fiorini, P., & Puers, R. (2007). Characterisation of an electrostatic
vibration harvester, EDA Publishing Association.
13. Sterken, T., Baert, K., Puers, R., & Borghs, S. (2002). Power extraction from ambient vibration.
In Proceedings of the SeSens (Workshop on Semiconductor Sensors, Veldhoven, Netherlands)
(pp. 680–683).
14. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Electronics, 27(2), 803–815. ISSN: 0885-8993.
15. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning a
resonant energy harvester using a generalized electrical load. Smart Materials and Structures,
19, 055003.
16. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
17. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
18. Szarka, G. D., Burrow, S. G., & Stark, B.H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
19. Maurath, D., Becker, P. F., Spreeman, D., & Manoli, Y. (2012). Efficient energy harvesting
with electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6), 1369–1380.
20. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
21. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies—state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
22. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. In Proceedings of the World Congress on Engineering (Vol. 2)
23. Zhu, D., Tudor, M. J., Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
24. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy
harvesters—a classification approach. Micromachines, 4(2), 168–196. http://www.mdpi.com/
2072-666X/4/2/168. Accessed 20 Jan 2015.
References 45

25. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of Applied Electromagnet-
ics (6th ed.). Prentice Hall, USA.
26. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
27. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008). A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
30. Razavi, B. (2002). Design of analog CMOS integrated circuits. McGraw-Hill
31. Razavi, B. (2008). Fundamentals of microelectronics. New York: Wiley.
32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University.
33. Razavi, B. (2002). Design of integrated circuits for optical communications. McGraw-Hill.
34. Hurst, P. J. (2001). Analysis and design of analog integrated circuits. New York: Wiley.
35. Spies, P. (2015). Handbook of energy harvesting power supplies and applications. CRC Press
Book, France.
Chapter 5
Piezoelectric Transducers

5.1 Introduction

The piezoelectric effect had its origin in the late 1800s, where the French Curie
discovered some crystals polarization effects subjected to mechanical stress. Apply-
ing an electric field there was also a deformation with implications in the field
of telecommunications. With the passage of time there were many solutions in
lead zirconate and materials PZT (zirconate-titanate). These latter have become the
dominant materials for a variety of applications such as actuators and ultrasonic
medical devices. Subsequently, also polymers such as polyvinylidene difluoride
(PVDF) have been found to have piezoelectric properties because of the stretching
of the molecules. Today, the most important applications of piezoceramics are in
medicine as ultrasonic devices, in the measurements of time (quartz) and the fuel
technology. New applications in micro-energy harvesting need further technical and
economic progress [1–15].

5.2 Materials

The single crystals such as quartz, LiNbO3 , GaPO4 , or Langasite (La3 Ga5 SiO14 )
are less commonly used as piezoelectric devices with respect to Pb (ZRX Ti1-x)
O3 (PZT), but there are, however, some commonly used applications involving
high frequencies or that require resistance to high temperatures. The process for
producing polycrystalline piezoceramic materials generally comprises two stages.
After preparation of the ceramic powder, it provides for the cooking of a mixture
of the oxide powder (calcination) and then milling into fine powder, the ceramic is
sintered to the desired shape. In the preparation of Pb (ZRX Ti1-x) O3 the oxide
powders of PbO, ZrO2 , and TiO2 are weighed in the appropriate proportions. In
the sintering process, the calcined powders are usually mixed in the desired shape.

© Springer International Publishing AG 2017 47

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_5
48 5 Piezoelectric Transducers

The final baking process at high temperatures (approx. 1200 ı C for 16 h for PZT)
enables the ceramic to reach its optimum density. The polycrystalline piezoceramic
materials must be polarized in an electric field in order to align the electric
dipoles for improving the piezoelectric properties of the material. Moreover, the
doping with small amounts of impurities can significantly improve the properties.
Currently, many research efforts are aimed for optimizing the properties by selecting
appropriate doping formulas. The most well-known difference is the distinction
between the so-called hard and soft piezoceramic actuators. A material with a
greater than 1 kV/mm field is called hard piezoelectric and a material with a pitch
between 0.1 and 1 kV/mm is called soft piezoelectric. For electrical connection
of the piezoelectric material, suitable electrode materials and related production
processes have to be chosen. The most commonly process uses silver-palladium,
which is sputtered or printed as a polymeric paste on the piezoelectric device. The
piezoelectric effect of any chosen material is limited by its Curie or phase transition
temperature. For the PZT, the Curie temperature varies between 250 ı C and 400 ı C
and is depending on its composition.

5.3 Model

A piezoelectric material is a transducer which converts electrical energy into

mechanical and vice versa. A schematic representation blocks can be displayed in
Fig. 5.1. The electric lock is defined by two parameters: the intensity of the electric
field E and the dielectric displacement D. That mechanical, instead, is represented
by the mechanical solicitation T and the mechanical stress S.
The ratio of the gate parameters is mathematically described by the constitutive
equations. The equations use the mechanical stress T and the electric field E as
independent variables and is called the d-formulation. The same formula applies in
the case of the material in energy harvesting area where T and D are the independent
variables. In this case the expressions of interest are the following:

E,D Piezoelectric T,S

Electrical element Mechanical

gate gate

Fig. 5.1 General layout of a piezoelectric material

5.3 Model 49

D
E D gT C (5.1)

T
S D sd T C gD (5.2)

where sd is the compliance (inverse of the Young modulus) measured with an

electric charge on the electrodes kept constant (expressed in m/N);
T is the
dielectric permittivity when a certain T (expressed in C/mV) is applied; g the
coefficient of piezoelectricity expressed in Vm/N [16–20]. Piezoelectric materials
are anisotropic materials for which constants are tensor and the electrical and
mechanical variables of the gate are expressed with the vectors. It is demonstrated
that the equations that describe the phenomenon are the following:

Qp D Cp Vp  d33 Fp (5.3)
1
zp D d33 Vp  Fp (5.4)
kp

where kp is the stiffness of the material, Fp is the force of activation, zp is the

displacement in the respective direction, C the electric charge, d33 is parameter
of the component D, Qp is the charge, and Vp the applied or generated voltage.
Whereas the electrical energy produced by the electromechanical response of the
piezoelectric material with respect to the mechanical energy supplied to the material,
the so-called coupling factor k is defined according to the following equation, in
other words the ratio between the electrical power stored and the input mechanical:

d2
k2 D (5.5)

0
T s

The conversion of the vibration energy into electrical energy power is a crucial
point for a successful design of energy harvesting devices. In general, a piezoelectric
energy harvesting system is often modeled as a vibrating system driven and damped.
This structure is constituted by a piezoelectric transducer coupled with the mechan-
ical structure and connected to an energy storage system by an energy harvesting
circuit. A mathematical level can be modeled with a set of N ordinary differential
equations [21–30]. Whereas an energy collection system can be described as a two-
port model with one degree of freedom (Fig. 5.2) due to the fact that an energy
harvesting device is often tuned to a certain natural frequency. In this case the
differential equations can be described by the following:

mu00 .t/ C bu0 .t/ C ku.t/ C AV.t/ D F.t/ (5.6)

Au0 .t/ C Cp V 0 .t/ D I.t/ (5.7)

where u is the displacement of the mass m and k is the overall stiffness of

the piezoelectric transducer and any other rigidity is connected. The effective
50 5 Piezoelectric Transducers

F(t)

mass

c Energy
k Harvesting
Circuit

Fig. 5.2 General layout of a piezoelectric oscillator

mass

c Energy
k Harvesting
Circuit

s(t)

Fig. 5.3 General layout of a piezoelectric oscillator with a basic solicitation

piezoelectric coefficient A and capacities Cp depend on the geometry of the

transducer and the load direction. Often, the energy collection system is applied over
a vibrating mechanism, and in this configuration the modeling must be expressed
whereas in Fig. 5.2 a basic solicitation. The system scheme is illustrated in Fig. 5.3.
The equations which govern the system are the following:

mq00 .t/ C bq0 .t/ C kq.t/ C AV.t/ D ms00 .t/ (5.8)

Aq0 .t/ C Cp V 0 .t/ D I.t/ (5.9)

In this case, q.t/ D u.t/  s.t/ represents the difference displacement of the
mass m and the excitation of the base. For a basic solicitation, the right side of the
equation expressed by ms00 .t/ is a d’Alembert force induced by the acceleration
of the ground. The equation of the system energy is obtained by multiplying the
above equation for u for the expression of the tension V.t/. Piezoelectric devices
are often modeled as current sources. The capacitance Cp of their inner electrodes
5.4 System Design 51

is considered in parallel with the load resistor R. Assuming that the internal current
generator is independent of the impedance of the external load, then the term V
can be eliminated from the above equation. This assumption is equivalent to the
statement that the connection is very weak or does not exist.

5.4 System Design

The voltage generated depends on the load R and the inductance L. It is assumed
that the system is driven by a force of external excitation, which is sinusoidal with a
frequency close to the natural frequency of the system with the piezoelectric element
loaded. To improve the performance of energy harvesting systems with piezoelectric
transducers can be required to apply more than one transducer on the structure.
For practical reasons, the transducers cannot be connected to different collection
circuits, but must be connected in series or in parallel (Fig. 5.4).
The piezoelectric transducers work as electrical generators and can be character-
ized by the equivalent circuit diagram of Fig. 5.4. Cp is the ability of the piezoelectric
transducer and the electric current Ip is resulting from mechanical excitation of the
piezoceramic [30–35]. For a piezo element of width b and length l, the electrical
current can be written as the following in the plane stress hypothesis:

d31 d

Ip .t/ D bl (5.10)
s11 C s12 dt

Considering a sinusoidal excitation:

Ip .t/ D Isin.!t/ (5.11)

The current and voltage values on the load can be expressed in conjugate complex
form as follows:

Fig. 5.4 Equivalent circuit

for a piezoelectric transducer
IP

V CP R
52 5 Piezoelectric Transducers

1
Ir D I (5.12)
1 C j!RCp
R
Vr D I (5.13)
1 C j!RCp

By varying the resistance R, it is possible to find the value for which there is the
maximum energy transfer. The power is expressed by the following equation:

I2R
PD (5.14)
1 C .R!Cp /2

where we can define:

1
RM D (5.15)
!Cp

To obtain the maximum power:

I2
Pmax D (5.16)
2!Cp

References

1. Park, J., & Mackqy, S. (2003). Practical Data Acquisition for instrumentation and system
control. Elsevier, Oxford.
2. Lacanette, K. (2003). National temperature sensors handbook. Annali di Matematica Pura ed
Applicata. National Semiconductor.
3. National Instruments. (1996). Data acquisition fundamentals, application note 007.
4. National Instruments. (1996). Signal conditioning fundamentals for PC-based data acquisi-
tion, Application Note 048.
5. Taylor, J. (1986). Computer-based data acquisition system. Instrument Society of America.
6. Di Paolo Emilio, M. (2013). Data acquisition system, from fundamentals to applied design.
New York: Springer.
7. Roundy, S., Wright, P., & Pister. K. (2002). Micro-electrostatic vibration-to- electricity
converters. In Proceedings of ASME international mechanical engineering congress and
exposition (IMECE) (Vol. 220, pp. 17–22).
8. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: self-sufficient energy
supply for micro systems. In Proceedings of the 16th international conference on thermo-
electrics (pp. 575–577).
9. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. Micro
IEEE, 21(3), 30–42.
10. Roundy, S. (2003). Energy scavenging for wireless sensor nodes with a focus on vibration to
electricity conversion. Ph.D Thesis, University of California.
11. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006). Micro seismic
electret generator for energy harvesting. Technical Digest PowerMEMS (pp. 133–136).
Berkeley, USA.
References 53

12. Sterken, T., Altena, G., Fiorini, P., & Puers, R. (2007). Characterisation of an electrostatic
vibration harvester, EDA Publishing Association.
13. Sterken, T., Baert, K., Puers, R., & Borghs, S. (2002). Power extraction from ambient vibration.
In Proceedings of the SeSens (Workshop on Semiconductor Sensors, Veldhoven, Netherlands)
(pp. 680–683).
14. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Electronics, 27(2), 803–815. ISSN: 0885-8993.
15. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning a
resonant energy harvester using a generalized electrical load. Smart Materials and Structures,
19, 055003.
16. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
17. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
18. Szarka, G. D., Burrow, S. G., & Stark, B.H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
19. Maurath, D., Becker, P. F., Spreeman, D., & Manoli, Y. (2012). Efficient energy harvesting
with electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6), 1369–1380.
20. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
21. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies—state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
22. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. In Proceedings of the World Congress on Engineering (Vol. 2)
23. Zhu, D., Tudor, M. J., Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
24. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy
harvesters—a classification approach. Micromachines, 4(2), 168–196. http://www.mdpi.com/
2072-666X/4/2/168. Accessed 20 Jan 2015.
25. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of Applied Electromagnet-
ics (6th ed.). Prentice Hall, USA.
26. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
27. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008). A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
30. Razavi, B. (2002). Design of analog CMOS integrated circuits. McGraw-Hill
31. Razavi, B. (2008). Fundamentals of microelectronics. New York: Wiley.
32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University.
33. Razavi, B. (2002). Design of integrated circuits for optical communications. McGraw-Hill.
34. Hurst, P. J. (2001). Analysis and design of analog integrated circuits. New York: Wiley.
35. Spies, P. (2015). Handbook of energy harvesting power supplies and applications. CRC Press
Book, France.
Chapter 6
Thermoelectric Transducers

6.1 Introduction

The discovery of the thermoelectric has been observed by Thomas J. Seebeck in

1821 after the deviation of a compass near two metallic conductors connected each
other at different temperatures. The degree of deflection was proportional to the
temperature difference. The reason was to study the difference of the electric field
that was created due to the temperature difference between the conductors. The
effect observed by the Seebeck is reversible as described by C. A. Jean Peltier in
1834: if we supply electrical energy in two connected conductors, a temperature
gradient occurs in the contact point; the thermal energy is transported from a point
of connection to the other, leading a cooling effect [1–15].

6.2 Seebeck and Peltier Effect

The Peltier effect is based on the production or absorption of heat at a junction

between two different conductors when an electrical charge flows through it.
The dQ=dt rate of heat produced or absorbed at a junction between the conductors
A and B is the following:
dQ
D .˛a  ˛b /I (6.1)
dt
where I is the electrical current and ˛ are the Peltier coefficients of the conductors.
The Seebeck effect is the production of electromotive force between junctions of
two different conductors. Two nodes connected back to back work with two different
temperatures, TH and TC and a tension between their free contacts:

V D S.Th  Tc / (6.2)

© Springer International Publishing AG 2017 55

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_6
56 6 Thermoelectric Transducers

S is the Seebeck coefficient. The Thomson effect is the production or absorption

of heat along a conductor with a temperature gradient T when the electric charge
flows through it. The heat dQ=dt produced or absorbed along a conductor segment
is the following:
dQ
D KJT (6.3)
dt
where J is the current density, K is the Thomson coefficient. The coefficients are
governed by the following relations:

˛ D TS (6.4)

with

˛ D ˛a  ˛b (6.5)

and
dS
KDT (6.6)
dT
Although these main thermoelectric effects have been known for a long time, it
is difficult to find explicit expressions in the literature for their three coefficients in
terms of the most fundamental physical quantity. The electrons in the wires occupy
energy levels in pairs of opposite spin. The lower levels are fully occupied and
higher levels are empty and the population of the level is determined by the Fermi-
Dirac statistics [16–25].
To move in the conductor an electron is occupying a certain level, it must
be dispersed to a vacuum level. For this reason, the low-energy electrons do not
contribute to the electric current because their neighboring levels are occupied. The
Fermi level is the energy at which the probability of occupation electron is 0.5.
Only electrons with energies near this level contribute to the current. The average
kinetic energy of the particles is calculated by adding all the velocity squared over
all directions in space in a solid angle of 4 determined by an angle , from 0 to ,
and azimuth of 2. Each speed is weighed by a probability distribution f0 .x; v; /:
Z
< v2 > f0 .x; v; /jvj2 d3 v (6.7)

where d3 v D 2v2dv sin. /d . The Maxwell–Boltzmann distribution is the

following:

m 3 mv2 =2KT
f0 .v/ D e (6.8)
2kT 2
6.4 Charges in a Semiconductor with a Temperature Gradient 57

where the exponential factor of last equation is determined by the following

condition:
Z 1
f0 4v 2 dv D 1 (6.9)
0

and it is applied to the calculation of < v 2 >. The average kinetic energy is the
following:
Z
< v 2 >D 4 f0 .v/v 4 dv D 3KT=m (6.10)

1 3
m < v 2 >D kT (6.11)
2 2

6.3 Potential

As example we can consider two pieces of N-type semiconductor (L and R) with

nL more than nR , where nL and nR are equal to the corresponding donors density.
Therefore, these densities are independent from temperature. If the two pieces are
brought in contact to form a junction, electrons will start to spread. The diffusion
creates a depletion region with electric field and the potential difference Vc between
the two pieces which stops the further diffusion of electrons. Vc is the potential for
contact expressible by the following formula:

kT
Vc D ln.nL =nR / (6.12)
e
where e is the electron charge and k the Boltzmann constant.

6.4 Charges in a Semiconductor with a Temperature

Gradient

The electrical current through a semiconductor with a temperature gradient can

be calculated by applying the transport equation of the Boltzmann. This equation
produces expressions for the currents that are similar to the linear equations of
Onsager, but with the advantage that there are no unknown linear coefficients. If the
density of the charge carriers is not dependent on position, then the electric current
will be:
 k 
2e
dT
Jq D  E  (6.13)
dx
58 6 Thermoelectric Transducers

where  is the electrical conductivity and E the electric field. By integrating above
equation and to equal Jq to zero:

k
VD T (6.14)
2e

Namely, a voltage of 43 V=ı C independent of the charge density and developed

between the hot and cold spots.

6.5 Thermoelectric Effect

The original Carnot cycle involves the transformation of heat between a hot bath
TH and a TC cold bath from a gas in a cylinder with a movable piston. The heat
flows in part from a hot bath to a cold bath, and in part is converted into mechanical
work in a closed cycle to four reversible stages. Carnot knew nothing of the chemical
potential, even on entropy. Nevertheless, he studied the cycle correctly without these
terms. At each stage of the macroscopic mechanical cycle, the work of the movable
piston is transformed to the thermal motion of the gas particles by elastic collisions
between the particles and the piston. Similarly, the electric work of a power source is
converted into a microscopic thermal heat junction by the acceleration of the charge
carriers in the electric field [26–30].
The electrons that pass through a junction will be slowed down or accelerated by
a contact potential difference Vc . Thus, they absorb or provide a quantity of heat at
the junction eVC , where Vc is calculated above with Peltier coefficients given by the
following equations:

˛a D kTln.nL / (6.15)
˛b D kTln.nR / (6.16)

6.6 Thomson Effect

The Thomson effect refers to the generation of heat by resulting from the passage
of a current along a portion of a single conductor on which is applied a differ-
ence of temperature T. Because of the temperature difference, absorbed heat per
unit of time is given by the following equation:

dQ
D ˇIT (6.17)
dt

where ˇ is the coefficient of Thomson. The origin of the effect is substantially the
same as for the Peltier effect. Here the temperature gradient along the conductor is
6.7 Thermoelectric Generator 59

responsible for differences in potential energy of the charge carriers. The Thomson
effect is not of primary importance in thermoelectric devices, but it should not be
overlooked by detailed calculations.

6.7 Thermoelectric Generator

A thermocouple usually consists—as the name suggests—of two different metals or

alloys. When the two junctions are at different temperatures, a low voltage—of the
order of tens of mV/K—is generated. This is called the Seebeck effect described in
this chapter and that can be used in temperature measurement and control.
In thermocouples, for the generation of energy, are partly replaced by semi-
conductor granules, contacted with a highly conductive metal strips. With such
materials, an order of magnitude greater are achieved and can also be used in
thin film devices for the measurement of temperature. Therefore the thermocouple
layers of micrometric thickness are deposited by a variety of techniques on a thin
substrate of electrically insulating support film. A thermoelectric converter obeys
the laws of thermodynamics; efficiency is defined as the ratio between the electrical
power supplied to the load and the absorbed heat to the hot junction (Figs. 6.1
and 6.2).
Expressions for the important parameters in the thermal power generation can
easily be deduced by considering the generator composed of a single thermocouple
and n- and p-type semiconductors as shown in Fig. 6.2. The thermocouples are

Fig. 6.1 General layout of a

thermocouple Heat

TH

n-type p-type

TC

L VC
60 6 Thermoelectric Transducers

I
Th Tc

N-type

Output L
power

Heat input
P-type Th Tc

N-type

Heat Output V
absorbed Heat

P-type

Fig. 6.2 Thermoelectric generator

constructed by two branches, one of n-type, a p-type material with the length Ln
and Lp , and constant cross sections An and Ap . The two branches are connected
to metal conductors of negligible electrical resistance. The heat is transferred only
by conduction along the branches of the thermocouple. The thermocouple is used
as a current generator by means of Seebeck or Peltier effect. The efficiency of a
generator is defined as the ratio between the energy provided to the load and the
heat absorbed by the junction. The load energy is related to the current and the
resistance itself; the transported heat instead is only linked by thermal conduction
and then the conductivity parameters of the two materials, with another dependence
from Peltier effect [31–35]. The generated current is given by:

.Sp  Sn /.Th  Tc /
ID (6.18)
R C Rl

where R is the resistance of the two semiconductors blocks of n and p.

6.8 Materials

Three parameters are used for classification of thermoelectric materials:  , electrical

conductivity; , thermal conductivity, and the Seebeck coefficient. The electrical
conductivity is given by the product of the concentration and mobility of charge
carriers. It is higher in metals, very low in insulators, with an intermediate
position taken by semiconductors. As measure of the potential usefulness of a
thermoelectric material is the figure of merit, the three parameters above mentioned
constitute the essential part of it. The Seebeck coefficient drops with increasing
concentration of carriers while electrical conductivity increases; consequently, the
electrical power factor (parameter proportional to the conductivity multiplied by
6.9 Figure of Merit 61

the Seebeck coefficient squared) has a maximum that is typically located around a
carrier concentration of about 1019 =cm3 . There are two components of the thermal
conductivity: vibration pattern and the electronic part. The latter also increases
with concentration of carriers and typically accounts for about one third of the
thermal conductivity. The maximum amount of energy falls into the region of
semiconductors. Therefore, the semiconductors are the materials of choice for
the further development of thermoelectric devices. Thermoelectric devices were
further classified with respect to the temperature ranges to which they can be
usefully employed. A positive direction of development has been the reduction of
the thermal conductivity, another search for the so-called electronic crystal glass
phonon, in which it is assumed that the crystal structures with weakly bound
atoms or molecules inside an atomic cage should conduct heat like a glass, but
conduct electricity like a crystal. Over the past decade, materials scientists were
optimistic in their belief that the low-dimensional structures such as quantum wells
(materials that are so thin as to be essentially two-dimensional 2D), quantum wires
(very small section and considered to be in one dimension 1D and referred to
as nano-wires), quantum dots that are confined in all directions, and superlattices
(a multi-layer structure of quantum wells) will provide a path for the achievement
of a significant improvement of figures of merit. There are also ongoing attempts to
improve the competitiveness of thermoelectric materials in directions different from
those of the figure of merit, such as reduction of costs, to the development of more
environmentally friendly materials.

6.9 Figure of Merit

Typically, an n-type and a p-type thermoelectric material are arranged thermally in

parallel and electrically in series as shown in Fig. 6.2. An effective thermoelectric
figure of merit for the two materials used in a module can be defined as follows:

.Sp  Sn /2
ZT D p p (6.19)
. p p C n n /2

where is the thermal conductivity and  the resistivity of the material. By means
of demonstrations, it can be shown that the maximum power is generated when the
external load resistance corresponds to the internal electrical resistance of the pair.
At this operating point, the power produced is given by the following expression:

..Sp  Sn /.Th  Tc //2

WD (6.20)
4R
62 6 Thermoelectric Transducers

At this point of maximum power, the power efficiency can be approximated as

follows:
ZT T 1
D D (6.21)
4 C ZTh C ZTm Th 2 C ZT4  T
2Th
h

In this case ZT is calculated as function of the average temperature.

References

1. Park, J., & Mackqy, S. (2003). Practical Data Acquisition for instrumentation and system
control. Elsevier, Oxford.
2. Lacanette, K. (2003). National temperature sensors handbook. Annali di Matematica Pura ed
Applicata. National Semiconductor.
3. National Instruments. (1996). Data acquisition fundamentals, Application note 007.
4. National Instruments. (1996). Signal conditioning fundamentals for PC-based data acquisi-
tion, Application Note 048
5. Taylor, J. (1986). Computer-based data acquisition system. Instrument Society of America.
6. Di Paolo Emilio, M. (2013). Data acquisition system, from fundamentals to applied design.
New York: Springer.
7. Roundy, S., Wright, P., & Pister. K. (2002). Micro-electrostatic vibration-to- electricity
converters. In Proceedings of ASME international mechanical engineering congress and
exposition (IMECE) (Vol. 220, pp. 17–22).
8. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: self-sufficient energy
supply for micro systems. In Proceedings of the 16th international conference on thermo-
electrics (pp. 575–577).
9. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. Micro
IEEE, 21(3), 30–42.
10. Roundy, S. (2003). Energy scavenging for wireless sensor nodes with a focus on vibration to
electricity conversion. Ph.D Thesis, University of California.
11. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006). Micro seismic
electret generator for energy harvesting. Technical Digest PowerMEMS (pp. 133–136).
Berkeley, USA.
12. Sterken, T., Altena, G., Fiorini, P., & Puers, R. (2007). Characterisation of an electrostatic
vibration harvester, EDA Publishing Association.
13. Sterken, T., Baert, K., Puers, R., & Borghs, S. (2002). Power extraction from ambient vibration.
In Proceedings of the SeSens (Workshop on Semiconductor Sensors, Veldhoven, Netherlands)
(pp. 680–683).
14. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Electronics, 27(2), 803–815. ISSN: 0885-8993.
15. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning a
resonant energy harvester using a generalized electrical load. Smart Materials and Structures,
19, 055003.
16. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
17. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
References 63

18. Szarka, G. D., Burrow, S. G., & Stark, B.H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
19. Maurath, D., Becker, P. F., Spreeman, D., & Manoli, Y. (2012). Efficient energy harvesting
with electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6), 1369–1380.
20. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
21. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies—state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
22. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. In Proceedings of the World Congress on Engineering (Vol. 2)
23. Zhu, D., Tudor, M. J., Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
24. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy
harvesters—a classification approach. Micromachines, 4(2), 168–196. http://www.mdpi.com/
2072-666X/4/2/168. Accessed 20 Jan 2015.
25. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of Applied Electromagnet-
ics (6th ed.). Prentice Hall, USA.
26. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
27. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008). A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
30. Razavi, B. (2002). Design of analog CMOS integrated circuits. McGraw-Hill
31. Razavi, B. (2008). Fundamentals of microelectronics. New York: Wiley.
32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University.
33. Razavi, B. (2002). Design of integrated circuits for optical communications. McGraw-Hill.
34. Hurst, P. J. (2001). Analysis and design of analog integrated circuits. New York: Wiley.
35. Spies, P. (2015). Handbook of energy harvesting power supplies and applications. CRC Press
Book, France.
Chapter 7
Electrostatic Transducers

7.1 Introduction

Micro-generators have applications in wireless sensor networks for security

systems, military applications, the monitoring of structural conditions, and personal
health systems; they require work to be done on a transducer so that electricity
can be generated. This requires a relative movement between the two ends of the
transducer (referred to the rotor and stator in the conventional generators) and
it is established by fixing the stator and allowing the movement of the rotor by
connecting it directly to the source, or through the use of an inertial mass, as shown
in Fig. 7.1. The inertial method is preferred because a single point of attachment is
necessary to collect energy from the source [1–15].
The system mainly includes an inertial mass of test, a spring and two shock
absorbers. One of these, De , is the mechanism of conversion between electric energy
and kinetic energy, and the other represents the parasitic mechanical damping.
The amount of energy that can be generated under specific operating conditions
in terms of amplitude and vibration is strongly dependent on the electrical damping.
In addition, the damping force must be tuned to enable the generator to continue
to work in maximum efficiency. One of the major design decisions is that the type
of transduction mechanism should be used. The electrostatic generators can make
use of the variable capacitor structures or piezoelectric materials. The choice of the
type of transducer for a particular application of generator is made on the basis of
the following two main criteria:
• Compatibility MEMS: The potential for widespread use of powered devices
means that they must be cheap to produce market. They are always embedded
with some electronic sensing and data transmission. Consequently, a very useful
characteristic of a micro-generator is its ability to be produced in large quantities
at low cost and that can be easily integrated with the electronics.

© Springer International Publishing AG 2017 65

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_7
66 7 Electrostatic Transducers

Fig. 7.1 Inertial generator

m
Zl z(t)

De Dp

• Controllability: Large-scale energy harvesting devices, such as photovoltaic

panels and wind turbines maximum, use of technical system about power point
tracking in order to collect as much energy as possible. An important aspect of
micro-generators is that, in order to maintain a high degree of efficiency, it should
be possible to apply techniques of monitoring (sensing).

7.2 Physical Phenomena

The electrostatic energy conversion mechanism of a transducer consists in the

physical coupling between electricity and mechanical means of an electrostatic
force. The electrostatic force is induced between opposite charges stored on two
opposite electrodes. The amount of charge Q accumulates on the electrodes is
a function of the potential difference V between the electrodes and the capacity
C according to the relationship Q D CV. The stored energy of the capacitor is
controlled by the following formula:

1 2
ED CV (7.1)
2
The physical principle of energy conversion cycle depends on the conditioning
circuit common to all energy harvesting generators and how the variable capacitor
is connected to the corresponding electrical circuits. In general we can distinguish
two types of connection: switching and continuous systems.
7.3 Switching System 67

7.3 Switching System

The switched connection between the transducer and the circuit involves a recon-
figuration of the system, through the operation of switch in different phases of the
generation cycle. The switched transducers can be further divided into two main
types: constant charge and constant voltage. In the constant charge if a variable
capacitor is pre-loaded to maximum capacity and then disconnected from any
external circuit before the capacitor geometry is modified by the motion, then the
extra energy will be stored in the electric field between the electrodes as a work
against the force electrostatic [16–25]. This energy can then be used to power a
circuit. The most common way in which this approach is implemented is shown in
Fig. 7.2.
The device is pre-charged to a low voltage in the first part of the cycle as shown
in the diagram QV of Fig. 7.2. The plates are arranged so that they can be separated,
thereby increasing the distance between them. The movement produces a constant
force between the two electrodes. The area outlined by QV diagram represents
the electricity generated. In the constant voltage mode, the capacitor is connected
to a constant voltage (possibly supplied by a battery), a reduction of capacity
between the electrodes caused by the relative motion of the plates would lead to
the charge to be removed from the condenser and pushed back into the voltage
source, thereby increasing the energy stored. If the plates are actuated in a sliding
motion, as indicated in Fig. 7.3, the force between the plates in the direction of
relative movement remains constant. The electronic circuitry to achieve a switching
of the transducer system can be quite different depending on the type of energy
conversion cycle. In Figs. 7.4 and 7.5 is shown an implementation of the circuit for
each mode of operation.
The switches are realized by MOSFET devices, which are operated by some
electronic control. The capacitor CV denotes the variable capacitor, while CR
denotes a storage capacitor (CRL: low voltage, CRH: high voltage). In the constant
voltage mode, the circuit must perform three different tasks: charge the variable

a b c
Q

+Q 2 Generate
+Q Pre- 3
+++++ +++++
+++++ +++++ charge

––– –– ––– – –
––– –– ––– – – Discharge
–Q
–Q
1
V

Fig. 7.2 Operation of an electrostatic transducer in constant charge mode. In (a) and (b) the
operational conditions and (c) the diagraph QV
68 7 Electrostatic Transducers

a b c
Q

2
+Q +Q
+++++ +++++ +++++ ++ Pre-
charge
––– –– ––– – – ––– –– ––– – –
Generate
–Q –Q
1 C increases
3
V

Fig. 7.3 Operation of an electrostatic transducer in constant voltage mode. In (a) and (b) the
operational conditions and (c) the diagraph QV

M1
Cv

L1
Control IC 1 2 (a)

M2 CR

Fig. 7.4 Electronic circuit for a switched electrostatic transducer in constant charge mode

D4

M3
CRH
Cv

D2
L2
Control IC 1 2
(b)

M4 D3 CRL

Fig. 7.5 Electronic circuit for a switched electrostatic transducer in constant voltage mode

capacitor to Vhigh voltage with the voltage constant and by reducing the capacity
from Cmax to Cmin ; moreover by transferring energy from the high-voltage to the
low-voltage store. In the other mode the associated circuit must implement the
following functions: charge the capacitor to a voltage level of Vlow by maintaining
7.4 Continuous Systems 69

constant the charge in the capacitor while the capacity is reduced from Cmax to Cmin ,
and the charge transfer from the variable capacitor to a store. The constant charging
phase is achieved by simply unplugging the variable capacitor by using electronic
or mechanical switches.

7.4 Continuous Systems

A third mode of operation is when the variable capacitor is constantly connected

to the load circuit, and this load circuit provides a bias voltage to the capacitor.
A simple example of this is a voltage source, a resistor, and a variable capaci-
tor connected in series. A capacitance change will always result in a charge transfer
between the electrodes through the load resistor by causing a work to be performed
in the load through a transfer of energy. A constant charge generator is equivalent
to a continuous generator with an infinite impedance load, while the constant
voltage generator corresponds to a continuous generator short-circuited. The use
of controlled switches complicates the implementation of the generator and the
circuitry necessary to control by consuming a minimal amount of power generated,
and therefore in some cases a continuous system is preferred over the other two
systems. The design can be based on a variable capacitor with a constant source or
time dependent [26–30]. A well-known example of the use of bias built-in sources
is given by piezoelectric generators: the capacitance between the electrodes of the
generator is therefore practically constant, but the bias voltage variations change as
a function of the displacement (Fig. 7.6).
If the piezoelectric materials are not available or are not applicable to the target
application, electrets materials can be used. In the silicon-based technologies it is
appropriate to use an SiO2/Si3N4 bilayer, similar to those used for the flash memory
cells. If the use of these layers is not compatible with the design of the capacitor,
then polymers such as Teflon or parylene are applied. In a continuous system, the

a b
Q
Q
Cpiezo
3

2
4

V
1
2 1

V 3

Fig. 7.6 Operation mode of a capacitor in continuous mode (a) or with a piezoelectric (b)
70 7 Electrostatic Transducers

a C1 b C2 C3
q1 q1 q2

Load Load Load

Cres Cres1

Fig. 7.7 Electrical scheme for the operation of the electrostatic transducer in continuous mode

variable capacitor is constantly connected to the circuit which includes the load.
A capacitance change will always result in a transfer of charge through the load
resistance. An advantage of a continuous system is that the transducer system can
be implemented without the use of switches. The use of switches requires some
extra circuitry to control them and valuable energy is consumed by the control
circuit. There are two basic schemes to achieve a continuous system. In a pattern,
a single variable capacitor is used in series with a voltage generator (provides the
bias voltage) and a load resistor (Fig. 7.7a). In this case, the charge flows through
the voltage source. An alternative method (Fig. 7.7b) implements two additional
capacitors with its capacity that varies in the opposite way. One of the advantages of
the latter method is that transduction is quite insensitive to parasitic capacitances.

7.5 Design

The simple design is a parallel-plate capacitor. The ability of a parallel-plate

capacitor is determined by the area A and the distance g of the plates and by the
dielectric Er of the material between the plates:

A
CD
0 Er (7.2)
g

where
0 is the vacuum permittivity. A variation of these parameters can determine
a variation of the capacitance of the capacitor with parallel plates. Since it cannot be
simple to vary the relative permittivity Er of the material from the kinetic movement,
the area A and the gap g are commonly used to provide a variable area of overlap
or a capacitor with variable gap. In general, gap-closing and overlap-area variable
capacitors can be classified by the direction of movement of the electrodes with
respect to the substrate surface. This movement can be either flat or off level. In
both cases the mobile and fixed electrodes must be electrically isolated from each
other. The manufacture of movable and fixed electrodes in a device layer is based on
silicon technology on insulator (SOI), which is widely accessible in semiconductor
and MEMS technologies [2, 31–35]. There are four main designs of electrostatic
transducers to implement miniature generators:
7.5 Design 71

Fig. 7.8 Four types of electrostatic transducers: (a) parallel plate—gap closing, (b) parallel
plate—variable overlap, (c) comb drive, (d) rolling rod

• Parallel-plate gap closing;

• Parallel-plate variable overlap;
• Comb drive;
• Rolling rod.
The device shown in Fig. 7.8a is a drawing to parallel plates, where they form
a capacitance between two parallel electrodes stacked one above the other. The
device is used in a constant charge mode and generates high-voltage spikes when
it is primed with a low voltage. The device of Fig. 7.8b is a drawing to parallel
plates and is generally combined with electrets and works in continuous mode. The
third type of transducer is still composed of multiple capacitors in parallel, but this
time the electrodes are on the same slice, forming one comb unit. The mass of
the generator can move in the plane when the movement occurs, by changing the
overlap of the electrodes. They may be operated continuously or in a switched mode.
It can be shown that the maximum output power possible from these systems is
proportional to the value of the test mass. If these systems are to be made in a cost-
effective way should be achievable by using batch production with MEMS silicon
standard. The device is shown in Fig. 7.8d and uses a proof mass that is not fixed
to the frame, and an outer mass can be used to not incur the problems associated
with the alignment and fastening of an external mass to the MEMS device. This
structure is composed of a metal rod on dielectric layers, which serves as a movable
electrode, and a series of strip electrodes beneath the dielectric as static electrodes.
72 7 Electrostatic Transducers

Fig. 7.9 Parallel

plate—variable overlap

The obtainable power density depends on the size of the generator, the test mass,
and the operating condition.
As an example, we can consider Fig. 7.9 as design corresponding to a parallel
plate—variable overlap.
The parameter g describes the fixed distance between the surfaces of the movable
and the fixed electrode. The capacitance C.x/ in function of the displacement x is
described by the following equation:

xmax C x
C.x/ D 2h (7.3)
g

where h is the height of the electrode. A similar procedure can be repeated for the
other design.

References

1. Park, J., & Mackqy, S. (2003). Practical data acquisition for instrumentation and system
control. Elsevier, Oxford.
2. Lacanette, K. (2003). National temperature sensors handbook. Handbook Ann. Mat. National
semiconcductor
3. National Instruments (1996). Data Acquisition Fundamentals, Application Note 007 (1996).
4. National Instruments (1996). Signal conditioning fundamentals for PC-based data acquisition
systems, Handbook National Instruments.
5. Taylor, J. (1986). Computer-based data acquisition system. Instrument Society of America,
USA.
6. Di Paolo Emilio, M. (2013). Data Acquisition System, from fundamentals to applied design.
New York: Springer.
7. Roundy, S., Wright, P., & Pister, K. (2002). Micro-electrostatic vibration-to-electricity convert-
ers. In Proceedings of ASME International Mechanical Engineering Congress and Exposition
IMECE2002 (Vol. 220, pp. 17–22).
8. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: Self-sufficient energy
supply for micro systems. In Proceedings of the 16th International Conference on Thermo-
electrics (pp. 575–577).
References 73

9. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. Micro
IEEE, 21(3), 30–42.
10. Roundy, S. (2003). Energy Scavenging for Wireless Sensor Nodes with a Focus on Vibration to
Electricity Conversion. PhD thesis, University of California.
11. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006). Micro seismic
electret generator for energy harvesting. In Technical Digest PowerMEMS 2006, Berkeley,
USA, November 2006 (pp. 133–136).
12. Sterken, T., Altena, G., Fiorini, P., & Puers, R. (2007). Characterisation of an electrostatic
vibration harvester. In DTIP of MEMS and MOEMS, Stresa, Italy, April 2007.
13. Sterken, T., Baert, K., Puers, R., & Borghs, S. (2002) Power extraction from ambient vibration.
In Proceedings of the SeSens, Workshop on Semiconductor Sensors, Veldhoven, Netherlands,
November 2002 (pp. 680–683).
14. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Electronics, 27(2), 803–815. ISSN: 0885-8993.
15. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning a
resonant energy harvester using a generalized electrical load. Smart Materials and Structures,
19, 055003.
16. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
17. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
18. Szarka, G. D., Burrow, S. G., & Stark, B. H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
19. Maurath, D., Becker, P. F., Spreeman, D., Manoli, Y. (2012). Efficient energy harvesting with
electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6), 1369–1380.
20. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
21. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies – state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
22. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. Proceedings of the World Congress on Engineering (Vol. 2).
23. Zhu, D., Tudor, M. J., & Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
24. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy harvesters
– a classification approach. Micromachines, 4(2), 168–196. http://www.mdpi.com/2072-666X/
4/2/168. Accessed 20 January 2015.
25. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of applied electromagnetics
(6th ed.). Prentice Hall, USA.
26. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
27. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008). A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
74 7 Electrostatic Transducers

30. Razavi, B. (2002). Design of analog CMOS integrated circuits. New York: McGraw-Hill.
31. Razavi, B. (2008). Fundamentals of microelectronics. London: Wiley.
32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University.
33. Razavi, B. (2002). Design of integrated circuits for optical communications. New York:
McGraw-Hill.
34. Hurst, P. J. (2001). Analysis and design of analog integrated circuits. John Wiley & Sons.
35. Spies, P. (2015). Handbook of energy harvesting power supplies and applications. CRC Press
book, France.
Chapter 8
Powering Microsystem

8.1 Power Conditioning

The power conditioning circuits play an essential role in an energy harvesting

system through various parameters such as the input impedance, at the same time
carries out processing functions such as power control and filtering. Advanced
techniques actively influence the behavior of harvesting devices, such as piezo pre-
biasing. The power limit of a system to be used was considerably reduced with
conditioning circuits that operate at lower levels of power, by reducing the losses to
increase the maximum efficiency of the harvesting system. The challenge is always
to optimize the energy and the associated conditioning circuits to cope with a system
where the correspondence of the power profiles and operation dynamics hours is
in some way optimized. Power supplies are often intermittent and the excitation
parameters may change over time. The studies give results measured on various
sources of vibrations and is possible to see how they differ from those ideals.
The purpose of the conditioning circuit is to avoid an oversized design, with a
storage system for providing a correspondence between the temporal profiles of the
power demand from the load source. In addition to the source of excitation, the
impedance is a conditioning factor that determines the operating conditions of
the system. While in general the source is not controllable, the input impedance
is the main control mechanism. The input impedance of the conditioning system
can be formed by a real part and an imaginary (resistive and reactive component),
and it is synthesized by the action of active and passive components of the converter,
usually controlled by the duty cycle of the active part and associated in the regulation
circuits. Whereas the harvesting system can be modeled by a combination of linear
and non-linear circuit elements, the maximum transmissible power is when the load
is the complex conjugate of the output impedance. However, it is not always possible
to work on this theory as a physical constraint, the voltage limitation or excursions
can precluded [1–15]. Furthermore, the energy consumption associated with the
resistive synthesizer and impedances of reactive load can become significant,

© Springer International Publishing AG 2017 75

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_8
76 8 Powering Microsystem

especially at low-power levels by causing a situation in which a different load can

produce more power. To understand the behavior of a circuit, it is often useful
to consider the waveforms of current and voltage in the frequency domain. For
a resistive component, voltage and current appearing in phase and represent the
power dissipated; in the case of phase out components, the fundamentals frequency
components appear as reactive components and are described for the displacement
factor, and represent the energy circulates between the source and the load that
can modify the frequency response of the system; the non-linearity of the input
impedance produces harmonic components that are described through the distortion
power factor. They modulate the flow of energy between the source and the load
to their particular frequency, the effect of which may or may not be significant in
terms of average power, by depending on the Q-factor of the mechanical system. The
impedance of the input resistive can be approximated by various types of circuits
in a discontinuous conduction mode or by configuring a feedback loop around a
converter to force the voltage to follow the current, or vice versa. An example is
shown in Fig. 8.1 where a circuit layout consists of a rectifier and a boost converter.
The rectifier has not a store capacitor and the mosfet device is activated by
depending on the input clock at various frequencies and duty cycles with the
inductor in discontinuous conduction, in this case the inductor current falls to zero
in every switching cycle. During switch on periods, the current ramp input can reach
a level determined by the input voltage, then the average current (Idcave ) follows the
input voltage. In the discontinuous conduction mode, the input resistance can be
approximated by the following equation:

Fig. 8.1 Resistive load with discontinuous mode converter

8.1 Power Conditioning 77

Idc
1 2

Vdc

HI
Load
Vin

LO

Idc

–
HI Current
1/Rin
Vdc + control/PWM

Fig. 8.2 Feedback control to force the phase of the current and voltage signals

2L
Rin D (8.1)
DT
where D and T are the duty cycle and the period of the switching waveform,
respectively, and L is the inductor value. Another approach to obtain an impedance
of resistive input is to employ a feedback to force the phase of the waveforms of
voltage and current; this technique is widely used for the correction of the power
factor in small-size power supplies. A block diagram of example system is shown
in Fig. 8.2.
The circuit is in current control mode, through a negative feedback on the PWM
controller in order to set the input current to the reference level. The reference
current is derived from the waveform of the input voltage: i.e., the instantaneous
voltage divided the desired input resistance Rin . In general, this circuit is difficult
to employ it for very low levels of micropowering system. However, an important
advantage is the operation of the converter with an inductor current in continuous
or discontinuous mode. The power conditioning circuits with some element of
the reactive input impedance determine the circulating energy between the power
electronics and the harvesting of energy at the fundamental frequency. Resonators
circuit can work in the “frequency tuning mode” to compensate for the reactive part
of the input block. In many situations occur when the transducer is described by
a mechanical energy system, common in piezoelectric devices. The linear circuits
known as rotators are able to synthesize the reactive impedance of input and
have found application in the suppression of the vibrations. However, they have
a significant consumption of energy to synthesize the reactive components. For the
energy harvesting, it is therefore necessary to base the layout around the switching
circuits, and an example is shown in Fig. 8.3. Since a reactive load implies a flow
of energy from the load to the source and vice versa, it is necessary to employ a
78 8 Powering Microsystem

+Vdc HI

Vin 1 2 Load

0 –Vdc LO

Iin

–
HI Current
1/Zin control /PWM
Vin +

Fig. 8.3 Layout with reactive impedance

V V
D
R
t t

Fig. 8.4 General layout of a half-wave rectifier

converter that can operate in four quadrants. The converter operates in the average
current control mode with the reference current derived from the input voltage.

8.2 Rectifier Circuit

The diode rectifiers circuits are widely used in electronic design, in particular
in power supply and demodulation systems. The main objective is the signal
conversion AC to DC that can be made with different circuit configurations, each
with their advantages and disadvantages, used in various industries. The rectifier
circuits are classified into two main groups, namely single-phase and three-phase.
These stages are very important for industrial applications and for the transmission
of energy into direct current (HVDC). The half-wave rectifier circuit is the simplest
form as visualized in Fig. 8.4. For the majority of power applications, this type
of circuit is not sufficient because the harmonic content of the rectifier’s output
waveform is very large and consequently difficult to filter. However, it is a very
simple way to reduce the power to a resistive load (Fig. 8.4).
The full-wave rectifier circuit (Fig. 8.5), however, uses full waveform.
This makes the rectifier most effective, and there is conduction on both half-wave on
the cycle of the sinusoid, the smoothing (literally “spread the signal,” or transform
the AC signal into DC) becomes much easier and more effective. The classical
8.2 Rectifier Circuit 79

~
v v
IN +

OUT
t ~ t
−

Fig. 8.5 General layout of a full-wave rectifier

circuit is based on four diodes in a bridge topology. The diodes can be replaced with
active elements to provide the switching and increase the efficiency. The choice
depends on the application of the circuit. While the full-wave circuits are mostly
used with the bridge configuration, half-wave circuits may offer a better solution
in some circumstances. For power applications are normally used power Schottky
diodes which require only a forward voltage of about 0.2–0.3 V.

8.2.1 Bridge Rectifier Circuit

The bridge rectifier (single phase) uses four rectifier diodes, each connected in
a closed-ring configuration (“bridge”) to produce the desired output. The main
advantage of this bridge circuit is that it requires a dual transformer. The diodes
(D1–D4 in Fig. 8.6) are arranged in series, only two diodes conduct current during
each half cycle of the sinusoidal waveform. During the positive half cycle, diodes
D1 and D2 are in conduction, while the diodes D3 and D4 are reverse biased and
current flows through the load as shown in Fig. 8.6. Conversely, during the negative
half cycle.
Bridge rectifiers integrated components are available in a range of different
voltages and dimensions that can be welded directly into a PCB circuit (Fig. 8.7).
Depending on the technology of the diode threshold voltage, each may vary in the
neighborhood of 0.6 V.
For a complete description, several parameters must to be considered: for
example, repetitive peak voltage (VRRM) and the Reverse Recovery Time. The
repetitive peak voltage (VRRM) is the actual value of maximum allowable reverse
voltage across the diode rectifier.
When switching from the conducting to the blocking state, a rectifier has stored
charge that must first be discharged before the diode blocks reverse current. This
discharge takes a finite amount of time known as the Reverse Recovery Time (trr ).
The smoothing capacitor placed in the output converts the wave into a DC
voltage. Generally for power circuits in current continues, the smoothing capacitor is
an electrolytic that has a capacitance value of the order of 100 F or more. However,
two parameters are important to consider when choosing a smoothing capacitor,
these are its operating voltage, which must be greater than the value of load output
80 8 Powering Microsystem

D4 D1

– +
+V

D2 D3 C
Load

0V
Wave with
capacitor
Ripple
Vdc
Wave without
capacitor
0V

Fig. 8.6 Example of bridge rectifier circuit

of the rectifier, and its capacitance value, which determines the amount of ripple
(variable part of the continuous signal) which will be superimposed on the DC
voltage. As a general rule, you can think of a ripple voltage of less than 100 mV peak
to peak. The amount of ripple voltage can be virtually eliminated by adding a filter to
the output terminals of the rectifier bridge. This low-pass filter is constituted by two
capacitors, generally of the same value, and an inductance in order to induce a path
of high impedance for the alternating ripple component. Another alternative layout
most practical and economical is to use a voltage regulator, such as an LM78xx
(where “xx” stands for the output voltage) that can reduce the ripple of over 70 dB
offering at the same time a constant output current of more than 1 A.
An important aspect of the rectifier circuits is the loss in the output voltage,
caused by the voltage drop of the diodes (about 0.6 V for silicon and 0.3 V for
the Schottky diodes). This reduces the output voltage limiting, therefore, that
available output. The loss of voltage is very important for low-voltage rectifiers
(for example, 12 V or less), but is insignificant in high voltage applications such as
HVDC. The rectifiers are also used for the detection of the amplitude modulated
signal relatively to radio signals. Another typical use of rectifier circuits is the
power supply design. A power supply can be divided into a series of blocks
(Figs. 8.8 and 8.9), each of which performs a particular function: a transformer
and a rectifier circuit for converting the AC signal into DC. The power supplies
are designed to produce less ripple that can cause several problems. For example,
in audio amplifiers, too much ripple looks like an annoying buzzing noise; in video
circuitry, an excessive ripple causes defects in the image; in digital circuits can cause
erroneous results of logic circuits.
8.2 Rectifier Circuit 81

–
˜
˜ + ˜ ˜ –
+

Case Style KBPM

PRIMARY CHARACTERISTICS
Package KBPM
IF(AV) 2.0 A
VRRM 50 Vto 1000V
IFSM 60 A
IR 5 µA
VF 1.1 V
TJ max 165°C
Diode variations In-Line

Fig. 8.7 Example of diode bridge of Vishay

IL
+ + +
Diode Voltage
Ac Line Filter V0 Load
rectifier Regulator –
– –

t t t t t

Fig. 8.8 Block diagram of a power supply

8.2.2 Zener Diode as Voltage Regulator

The zener diodes are widely used in reverse bias to produce a stabilized output
voltage. When connected in parallel to a variable voltage generator in reverse bias,
82 8 Powering Microsystem

Input Series Pass Output

Element
1 3

Current SOA
Generator Protection

Starting Reference Error

Circuit Voltage Amplifier

Thermal
Protection

GND
2

Fig. 8.9 General diagram of power supply

Fig. 8.10 General diagram RS

of a voltage regulator with
zener diode

Input DZ Output
VIN Vout

such as the rectifier diode circuits just discussed, the zener diode conducts when the
voltage reaches its reverse breakdown voltage. From that moment on, the relative
low impedance of the diode keeps a constant voltage at its terminals. In the circuit
shown in Fig. 8.10, an input voltage VIN is adjusted up to an output voltage VOUT
is stable. The breakdown voltage of the reverse bias of the diode DZ is stable
over a wide range of current and keeps VOUT relatively constant even if the input
voltage may fluctuate on a fairly wide range. Due to the low impedance of the
diode, the resistor RS is used to limit the current through the circuit. The value of
the resistance must satisfy two conditions: it must be small enough so that the
current through the zener diode is sufficient to maintain the conditions of reverse
breakdown; moreover, it must also be large enough for the current through DZ.
A small problem with stabilizer zener diodes circuits is the presence of electrical
noise in the attempt to stabilize the voltage. Normally this is not a problem for most
applications, but the addition of a decoupling capacitor of large capacitance through
the output of the zener may be needed to obtain a further stabilization of the voltage.
8.2 Rectifier Circuit 83

8.2.3 Considerations

The rectifier circuit is perhaps one of the most well-known circuits, formed by
diodes and capacitors. It’s widely used as the first stage in the conditioning circuit
in energy harvesting applications. However, the simple pattern as shown in the
preceding figures implicates some considerations. The classic circuit in Fig. 8.11
leads us to consider the Rs values, the classic case is having Rs tiny order of some
ohms, by resulting in a waveform shown in Fig. 8.12.
Generally, an impedance of the source will be comparable with that of the load,
then waveforms close to those of Fig. 8.12b. By a control of the waveforms, it can
be noted that the energy flow is directional uniform in peak rectifier. Namely, the
peak rectifier behaves as a non-linear resistive load. The peak rectifier is a simple
way to synthesize the impedance of the resistive input, by providing an important
adjustment function, which can be scaled to a wide range of power levels; but the
current harmonics may prove of falling in many of powering systems. However,

I1

Rs Is

Vdc

C RI
Vt

Vs

Fig. 8.11 Classic rectifier circuit

a b
i,v i,v

is peak
Vl

is vt
vt

a p /2 p wt a p /2 p wt

Fig. 8.12 Waveforms with (a) the load source impedance much smaller, and (b) the source
impedance close to that of the load
84 8 Powering Microsystem

I1

Cmass 1 Lc 2 Rc Ih

2
V Lcomp Cr RI
Rdamp Vh
–

Fig. 8.13 Vibration energy harvester

in some harvesting applications, the electronic characteristics minimize the impact

of current harmonics. In Fig. 8.13 is shown the equivalent circuit of a simplified
vibration energy harvesting resonant circuit with electromagnetic transduction and
peak rectifier.
The circuit appears as a filter by the time that the different components of
the rectifier harmonics see different output impedances, and especially different
resistive components. The fundamental current component is presented with a
component of resistive origin of Rcoil C Rdamp . A well-designed harvesting system
will have a resistance of the coil, which is some small fraction of the damping
mechanical. By introducing a distortion power factor, it enables simple conditioning
circuits of systems able to operate efficiently, with a half-wave peak rectifier that
produces more useful energy [16–25].

8.3 Piezoelectric Biasing

The energy harvesting systems that use piezoelectric transduction have output
impedance which is highly reactive due to the capacity of the piezoelectric material
and therefore the ideal complex conjugate load has a very low-power factor. Since a
typical piezo produces few milliwatts, the attempt to synthesize the optimal loading
with a circuit similar to that of Fig. 8.3 is likely to cause a system with greater power
generation capacity; so designers IC systems have turned to a series of non-linear
approaches to improve the power without large overhead costs. A typical approach
is shown in Fig. 8.14 and is called synchronous switched harvesting on inductor
(SSHI), which involves the flipping of charge polarity on the piezoelectric material
twice per cycle when the mechanical part reaches its maximum displacement by
means of a physical inductance.
Another biasing approach is shown in Fig. 8.15. The switches S1 and S4, as well
as S2 and S3, operate in pairs. When the piezoelectric material reaches its maximum
deflection, one of the pairs of switches is turned on, discharging the energy from
8.3 Piezoelectric Biasing 85

I1

Ih
2

U1
1
2 Co RI
Vpiezo Vh
Cpiezo L
–

1
Rs

Fig. 8.14 Circuit of synchronous switched harvesting on inductor (SSHI) with rectifier

2 2

S1 S2

1 1

Rs1
2 L6 1 Co
Vout
Cpiezo

2
2
S4
S3
1
1 Vpiezo
–

Fig. 8.15 Single-supply pre-biasing for piezoelectric

the piezoelectric capacitance in the capacitor through the series inductor (discharge
phase in Fig. 8.16), all very quickly with respect to the mechanical excitation
frequency of the system. As soon as the discharge phase is completed, the voltage
on the piezoelectric capacitor reaches zero and the opposite switch pair actives
and injects a certain piezoelectric charge on the capacitor of opposite polarity.
So it increases the force with which the transducer is able to oppose the relative
movement between the mass and the base, thereby by increasing the electrical
damping. The piezoelectric material is then moved to its extreme opposite, thereby
by increasing the voltage. The process is repeated with the generation of the
output voltage. The piezoelectric capacitor remains in open circuit during the entire
movement of the beam, while in SSHI the piezoelectric element is short-circuited
by the switching of the rectifier diode. To extract energy from a transducer, the
force produced by the transducer which opposes the motion must be in phase
with the speed of the transducer. Due to the impedance of the output capacitive,
a piezoelectric transducer to open circuit has strength and speed 90ı out of phase
86 8 Powering Microsystem

Fig. 8.16 Waveforms of the circuit of Fig. 8.15

and therefore has a power factor equal to zero. The discontinuity introduced in the
waveform (Fig. 8.16) effectively corrects the phase of the fundamental with nearly
unity power factor.

8.4 Voltage Control

The output voltage adjustment process is to maintain constant value in function of

load variations. The switching converters are power converters that allow to keep the
output power without losses, and then adjust the switched output typically modify
the input power supply by altering the impedance, usually through an automatic
feedback loop. The voltage control systems require a monotonic relationship
between the power and the impedance over the operating range of input. To increase
the power with a voltage source, the load resistance should be reduced; to increase
the power with a power generator, the load resistance must be increased. The power
response of an energy harvesting source has in general not only the regions in
which the resistance load increases the power, but also the regions in which the
opposite is true, in this way a simple feedback control is not possible when it
operates near the peak power region. A similar problem occurs with photovoltaic
panels. Because of the difficulty in the use of the power converter to perform a
feedback voltage control by controlling the impedance of the converter input, energy
harvesting applications can benefit from the alternative voltage control strategies.
A possible example is to separate the voltage regulation from that impedance
8.5 MPPT 87

Idc
1 2

Vdc

HI
Shunt Load
Vin

LO

Fig. 8.17 Powering conditioning with a shunt regulator

input, for example, using a converter with fixed input impedance and by using
a secondary shunt voltage regulator (Fig. 8.17). A possible layout is a flyback
converter operating in discontinuous open loop mode with a shunt regulator and
a low-power conditioning system [26–30].

8.5 MPPT

The power sources such as solar cells are commonly employed to work at a power
point (peak) by using a linear feedback system to ensure constant peak operation.
The type of layout is called MPPT also used for the energy harvesting. The
power converters tend to perform small adjustments to their working point (input
impedance), trying to find the point with the maximum output power or the point at
which the derivative respect to the operating point is zero power. The power can be
calculated from measured currents and voltages or deducible from other variables
when the voltage change is slow and compared to the capacitor charging rate. When
monitoring systems are used with the vibrations energy harvesting, it is important
to take note of the energy stored in the mechanical oscillator. When the dynamic
load is considered, it can be shown that the output power is a function of the
load impedance and then of impedance changing rate. The MPPT controllers are
in fact able to use all the power generated by the solar panel to charge the battery,
unlike traditional PWM controllers that send to the battery the current generated
by the panel. To understand this concept, we must first specify that the power of a
panel is the result of the following multiplication: Current supplied from the panel
for voltage generated by the panel. The working voltage generated by the panel
is typically around 16–18 V (not 12 V, as the battery voltage): this voltage surplus
is not considered in conventional voltage regulators, unlike the MMPT regulators.
We assume that the current generated by a photovoltaic panel is, in a certain
situation, 3 A: with a traditional PWM controller the current that is transferred to the
battery for charging is equal to 3 A. An MPPT controller instead analyzes the power
generated by the panel (P D V  I), and therefore also considers the panel voltage:
88 8 Powering Microsystem

if therefore we suppose that the panel voltage is 17 V at that time the power supplied
by the panel is 17 V  3 A = 51 W. If the battery voltage is 13 V, by considering the
maximum power of 51 W, the charging current to be transmitted to the battery is
51 W/13 V = 3.9 A. We note therefore that the battery will be charged with a current
of 3.9 A with the MPPT controller, instead of 3 A with a traditional controller, and
the charging will take place, therefore, at a faster rate of 30 %, with the same panel
and of current delivered. In practice, it is as if we used a panel of 130 W instead
of one about 100 W, then the higher cost of an MPPT controller is balanced by the
savings on the cost of the panel.

8.6 Architecture

In the conditioning system with passive circuits, such as the rectifier, the problems
are concentrated on the input impedance: a zero voltage to the storage capacitor,
the harvesting circuit looks like a short circuit. The problem is exacerbated if the
capacitor has a relatively large capacity and can take many minutes to load. In
addition to this it is necessary that during start-up the load does not require a lot of
start-up energy; to alleviate this is necessary to enter in the circuit active component
connections. The addition of active elements in the power conditioning to control the
converters presents a further problem of power supply. If the output is AC harvesting
system, a diode network can be used to provide power and start the main converter.
A simplified block diagram of this simple design concept is shown in Fig. 8.18. The
further advantage is the ability to initiate the harvester circuit when it has a peak
output voltage less than that required to power the active circuitry. By incorporating
a provision multiplier diode is possible to start when the harvester has a peak output
voltage less than that required to power the active circuitry.

Diode Multiplier

Harvester
Capacitor Load
Switch
Converter

Regulation
Control circuit (Voltage)

Fig. 8.18 Block diagram of a power conditioning circuit with parallel diodes for starting of the
main converter
8.7 DC-DC Systems 89

Idc
1 2 LDO

Vdc

HI
Shunt Load
vin (WSN)

LO

Fig. 8.19 Conditioning circuit with duty-cycled power demand

A further complication arises with the energy storage capacitor, to minimize the
value of the capacity to promote the rapid start-up will result in greater fluctuations
in voltage for a given load profile, and therefore requires a voltage regulation circuit.
In Fig. 8.19 a circuit example with a voltage regulator is shown.
The energy is stored in a capacitor with shunt adjustment followed by a low-
dropout regulator (LDO) by providing, for example, to the WSN load a constant
3.3 V. When switched on, the system takes a little more than 100 s to fully charge
the capacitor (6.8 V). The load can then perform a transmission operation of 12 s.
The capacity must be selected so that its voltage is to a minimum (3.3 V) at the
end of the transmission period, by avoiding voltage variations. WSN then goes
to low-power mode and the capacitor voltage recovers enough to get in a next
transmission period after 81 s. The duty cycle of WSN must always remain below
15 %. A switching converter can be used in place of the linear regulator; however, it
is not always the case that the circuital additional complexity justifies the efficiency
gains.

8.7 DC-DC Systems

In energy harvesting applications, the DC-DC converters play an essential role to

power electrical loads. The various energy transducers analyzed in the previous
chapters provide various output voltages, which are typically too high or too low to
power direct electrical loads. Therefore, the DC-DC converters have the objective
of providing a stable supply voltage.

8.7.1 Linear Regulators

The task of linear regulators is to establish a constant output voltage regardless of the
output current and input voltage. To check the adjustment element, an amplifier loop
90 8 Powering Microsystem

T1 +
IN IH
IH
R1 RL
Vout

LO

4
Vr
Vg1 2 –

V–
–
1 0
OUT
R2
3

V+
+
8
0

Vref
+
–

Fig. 8.20 Block diagram of a linear regulator

is used together with a more resistors network. The operational amplifier performs
the task by amplifying the signal, and then compare it with a reference voltage. In
Fig. 8.20 a classic example of the linear regulator is shown.
The loop system controls the voltage amplification that depends on the Vref
connected to the positive terminal of the operational amplifier. The disadvantage is
the power loss due to the voltage drop between input and output voltage. This means
that the efficiency is high when the difference between input and output voltage is
low. Furthermore, a linear regulator cannot have an output voltage larger than the
input. The advantage of this concept in comparison to the switching regulators is
the absence of the switching elements. Therefore, there is no ripple in the generated
output. In reference to Fig. 8.20 can be calculated the efficiency defined as:
 
Vout Iin  Vref  
Pout Vout Iout R2 Vout Vref
D D D D 1 (8.2)
Pin Vin Iin Vin Iin Vin Iin R2

By referring to efficiency equation, the maximum possible value is obtained by

minimizing the resistors R1 and R2. The efficiency depends on the ratio between
the output and input voltage.

8.7.2 Switching Regulators

The switching regulators are an alternative to linear regulators. They also fulfill the
task of generating a constant output voltage and independent from input voltage and
output current. The regulation element acts as a switch that can be in the state on
or off. The types of switching regulators can be divided into four main subtypes.
A buck converter delivers a lower output voltage and a boost converter with an
8.7 DC-DC Systems 91

output voltage higher of the input voltage. These two types have in common an
inductor as an energy storage element. For these two types of converter two different
modes of operation are defined: in the continuous conduction mode (CCM), the
current is always greater than zero, while in discontinuous conduction mode (DCM),
the current is zero for a certain period of time. The flyback converter uses a
transformer instead of an inductor and transfers the input energy during the off-
state of the switch to the output capacitor. This type of converter is mainly used
in powers of greater than 100 W output, for which it is not important state-of-the-
art applications for energy harvesting. Another important type of converter is the
charge pump: several switching transistors are used and only capacitors are used as
storage elements. A less common type of converter is Meissner, useful for energy
harvesting applications in which the energy transducer delivers output voltages less
than 500 mV. As in flyback converters, a transformer is used but in this case the
secondary winding is used to control the switching transistor.

8.7.3 Buck Converter

The step-down converter (buck) provides a lower output voltage of its input.
Basically, the input voltage is periodically turned on and off and a low-pass
filter transfers the average value of the converter output. The basic scheme of the
converter is shown in Fig. 8.21. While a typical electric circuit of a buck converter
is shown in Fig. 8.22. A voltage proportional to the output voltage is compared with
a reference voltage according to the desired output voltage. The difference between
these two signals (Verr ) is amplified. The output of the comparator is a PWM signal
of the T1 switching transistor.

8.7.4 Boost Converter

A step-up converter (boost) is used when in need a large voltage from an input
source. Regarding energy harvesting transducers, it is an important device for
thermogenerators and inductive generators, in which the output voltage is typically
less than 1 V. In Fig. 8.23 is shown the physical principle, while in Fig. 8.24 an
example of a circuit. In the first phase the switch is closed and the current of the
inductor ideally increases with a linear profile, therefore, the energy stored in the
inductor is increasing. In the second phase the switch is open and the energy stored
in the inductor is transferred to the exit where the capacitor C can be connected to
a second storage element. The variation of the current due to the open switch of the
inductor induces a voltage which is added to the input voltage. In this way, the two
phases are alternate periodically and the output voltage is always greater (or equal)
than converter input voltage.
92 8 Powering Microsystem

Switch = 0 L
1 2

RL
IL D C
+ VIN

Switch = 1
L
1 2 1 2

RL
D C
+ VIN IL

Fig. 8.21 Physical principle of a buck converter

T1 L1 +
IN HI 1 2 IH
D1 RL1
COUT Vout
C1 LO
0 R3
Vg1 0 –
0
0
4

V– 4

R4
6 2
V–

– –
7 1
OUT OUT
+ 5 + 3
V+

V+
8

+ Voltage reference

Fig. 8.22 Electrical layout of a buck converter

8.7.5 Buck-Boost Converter

The inverting converter is also called Buck-Boost because it refers to two types of
converter, a slightly more complex, which provides for the union of the two circuits
previously seen by using only one inductor in common with two switches and two
diodes, while the second that we will analyze in this paragraph can be seen as a
modification of one of the two. In this type of circuit two distinct meshes appear,
one that includes the switch placed in series and the inductor in parallel, and the
second created only by resistive and capacitive load. The diode is placed in series,
but in the opposite polarity to the previous configurations and constitutes a link
between the two circuital regions (Fig. 8.25).
8.7 DC-DC Systems 93

L D
1 2

RL
IL Switch = 0 C
VIN
+
–

L D
1 2

2
RL
IL C
VIN1 Switch = 1
+
– 1

Fig. 8.23 Physical principle of the boost converter

L1 D1 +
IN HI 1 2 HI

RL
COUT R1 Vout
C LO
T1 0
Vg1 –
0
0 0
4

V– 4

6 2
V–

– –
7 1
OUT OUT R2
5 3
V+

V+

+ +
8

Voltage reference
+
–

Fig. 8.24 Example of circuit layout of the boost converter

Even this circuit operates with two states of operation, of course scanned by the
switch. To the initial conditions of course, all passive components will be discharged
and free of stored energy, but when the power transistor is switched on, the current
will be forced to flow from the diode on the inductor which stores magnetic energy.
When the switch is opened, the inductor is forcing the current on the second mesh,
by going to charge the capacitor and supplying the load. So, by returning to the
switch closed again, the inductor will load again, while the capacitor will provide
enough energy in the output having still stored electrical energy. Of course with
this operating cycle the output current and thus the polarity of the voltage will be
opposite to that of input; hence the name of inverting [2, 31–35].
94 8 Powering Microsystem

M1
NMOS D1
VO
2

L1

Vin Vsw R1
+ + C1
– –
1

Fig. 8.25 Buck-boost converter

Fig. 8.26 Armstrong L1 L2

oscillator

u1 u2 C2

C3
R1

Q
C1

R2 R3

8.7.6 Armstrong Oscillator

For the energy harvesting applications, it may be advantageous to use as the

Meissner or Armstrong oscillator to build a switching converter. In this way it is
created a self-oscillating circuit in which the switching transistor is driven by a
winding of a transformer. Consequently, the adjustment circuit can be simplified,
since it is not necessary any more in the clock circuit with a driver to control
the switching transistor. This also helps to save energy in low load conditions
(Fig. 8.26).
8.8 Load Matching 95

8.8 Load Matching

Since there are some energy transducers such as a thermoelectric generator (TEG)
based on the Seebeck effect or a solar cells with a parasitic ohmic internal resistance,
it is useful to think of the load matching in order to obtain the maximum power.
A transducer is generally modeled as a real voltage generator. If the interface
circuit for the transducer absorbs current, the measure will be affected by an error
proportional to the voltage drop across the internal resistance of the generator,
therefore is important that the acquisition circuit has high input impedance. It is
also important that the first stage of the acquisition has low output impedance,
so as to provide the acquired signal to downstream circuits regardless of the
current absorbed by them (Figs. 8.27 and 8.28). These capabilities are defined
impedance matching and are typically made with an operational amplifier in buffer
configuration.
It can be important to realize a connection that makes the maximum voltage
across the load or to maximize the current or power level. In other situations it may
be important to minimize the deformation of the waveform on the load compared
to that provided by the generator. Hence we can distinguish four mode of matching
between the generator and load:

Fig. 8.27 Impedance matching

Fig. 8.28 General layout of matching

96 8 Powering Microsystem

1. Maximum voltage matching

2. Overcurrent matching
3. Maximum power matching
4. Matching of uniformity (minimum waveform distortion)
The condition for which there are no signals of reflections is that the voltage on
the load is in phase with that of the generator. Or if it is provided the following
condition:

Zu D Zg (8.3)

Taking into account some considerations, the energy matching condition results
from the following condition:

Zg D Zu (8.4)

The technique of “L matching” allows to do the matching of both real and

imaginary part of that impedance ZT of the transducer. It is implemented by using
two reactive components between the source and the load as in Fig. 8.29. The
reactive components may be inductors or capacitors and allow to adapt any load
impedance ZT to any generator impedance ZG . The use of reactive elements also
allows that there are no power losses in the matching network.

Fig. 8.29 L matching

8.9 AC-DC Systems 97

8.9 AC-DC Systems

Vibrations are a ubiquitous environment source used in energy harvesting systems.

Piezoelectric, electrostatic, and electrodynamic transducers provide AC power from
ambient vibrations. Therefore, AC-DC converters are needed in order to convert
their power into DC current which is required by the load of the energy harvesting
system. The AC-DC converter used for such transducers is normally a two-stage
power converter which consists of the following elements:
• AC-DC rectifier. The adjustment is usually done with a bridge rectifier of diodes
or voltage multipliers for the case of electrodynamic transducers to increase the
voltages. The current multipliers are used in piezoelectric transducers to increase
the low output current and rectifying the output power.
• DC-DC converter. After the rectification of the alternating current, it is necessary
to adapt the transducer voltage level to the load.
A rectifier is necessary when the piezoelectric transducers are employed in an
energy harvesting system. Voltage or current rectifiers-multipliers are an alternative
to full-wave rectifiers when is need to increase the voltage or current, respectively,
and correct the signal. There are three different rectifiers AC-DC: diode-resistor
pair rectifier, a diode-pair rectifier, and a synchronous rectifier (Figs. 8.30, 8.31,
and 8.32). A charge pump is used to provide a regulated output voltage to a filter

R1 C1
D1

V1 R2 C2 R3
+
–

Transducer

Fig. 8.30 Converters AC/DC, diode-resistor pair rectifier

R1 C1
D1

D2
V1 C2 R3
+

Transducer

Fig. 8.31 Converters AC/DC, passive full-wave rectifier

98 8 Powering Microsystem

4
V–
6 –
7
OUT
5 +

V+
8
R1 C1 M1
C2 R3
D1
V1
+
– M2
D2

4
V–
6
–
Transducer 7
OUT
+ 5

V+
8
Fig. 8.32 Converters AC/DC, synchronous full-wave rectifier

capacitor and the resistive load. The circuit shown in Fig. 8.30 is a half-wave
rectifier with a parallel connection of a resistance Rc element first of the diode D1.
The resistance is placed in that position because it reduces the charging time of
the output capacitor. The optimal value of Rc is what makes it possible to reach
the final voltage on the capacitor C2 without load in a minimum time. The circuit
shown in Fig. 8.31 is a full-wave rectifier that works as a voltage doubler. At the
resonance frequency, the piezoelectric element capacitance dominates its internal
resistance, and therefore the rectifier operates as a voltage doubler with a fixed
capacitor piezoelectric element. In the circuit shown in Fig. 8.32, the diodes are
replaced by transistors which operated in a synchronous mode in contrast with
the Fig. 8.30. In this solution, when no initial energy is available in the system, can
be started and then supplied power to the operational amplifiers to start synchronous
operation. The comparator U2 provides a high signal that turns on the transistor M2
when the input voltage has a negative value. The efficiency obtained by using a load
of 80 k in the circuits shown in Figs. 8.30, 8.31, and 8.32, is 34 %, 57 %, and 92 %,
respectively.

8.10 Electrical Storage Buffer

After the recovery and conversion of the energy, is needed to accumulate it through
an appropriate (storage) system. Typically, for a variety of electro-mechanical
applications, there are three main storage technologies:
• Fuel cells, mainly used for electric vehicles. A fuel cells is an electrochemical
device that allows to obtain electricity directly from certain substances, such as
oxygen or hydrogen, which takes place without any thermal combustion process.
The efficiency can be very high with a considerable mechanical stability; some
processes such as catalysis, however, pose practical limits to their efficiency
(Fig. 8.33).
8.10 Electrical Storage Buffer 99

– +

4OH–

4H+
2H2O

O2
2H2

Anode Cathode
Separator

Fig. 8.33 Fuel cell

• Electrochemical batteries, commonly used for electric vehicles, plug-in, and as

accumulators for stand-alone photovoltaic systems. The principle of operation of
an electrochemical battery is not much different than that of a fuel cell, since
the structure consists of two electrodes, each of which is affected by a redox
reaction, separated by a membrane permeable of ions. A difference of a fuel cell,
the electrochemical batteries are characterized by the presence within the cell
of reagents and products. There are various types of electrochemical batteries,
including lead-acid, the Nickel-Cadmium, and Lithium-ion battery.
• Supercapacitors, or special capacitors that have the characteristic of storing a
quantity of exceptionally high electrical charge respect to traditional capacitors;
in fact, while the capacitor has values of capacity at the order of mF, superca-
pacitors can reach up to over 5000 Farad. Therefore, they constitute the storage
devices characterized by high specific power and energy by far higher than
traditional capacitors.
The electrochemical capacitor observed from Fig. 8.34 is characterized by a very
similar construction to that of a battery; it has substantially two electrodes and an ion
permeable separator, placed between the electrodes which contains the electrolyte.
A storage system in supercapacitors must be suitably dimensioned in such a way
that this has long life and high efficiency; in this regard of particular importance,
the parameter known as Dept of Discharge (DOD) expresses how much energy has
been withdrawn from the supercapacitor as a percentage of total capacity. Typically,
the DOD of a supercapacitor should not be more than 20 % to obtain an optimal
operation. The supercapacitors may be classified according to the material used for
the construction of the electrodes (carbon, metal oxides, or polymers) and according
to the type of electrolyte (organic, aqueous, or solid); the choice of the type of
electrolyte will condition then all the dimension of the supercapacitor.
100 8 Powering Microsystem

Separator

Electode Electrolyte Electode

electric conductivity inonic conductivity electric conductivity

Fig. 8.34 Internal structure of a supercapacitor

8.10.1 Supercapacitors

Recent studies in the field of super-capacitors (SC), also known as ultracapacitors

or electrochemical capacitors, have developed most viable alternative for the energy
storage technologies. These devices in fact possess a greater number of charge-
discharge cycles than conventional batteries and can operate in a wide temperature
range. Conversely, disadvantages of these devices are the low energy density and
high self-discharge. In general, to maximize energy storage efficiency (Estorage ),
the total leakage power is kept low, charging the supercapacitors alternately or
according to their voltage series. To maximize energy efficiency on the load
(Edriving ), the residual energy is minimized by changing to a supercapacitor series
configuration so as to increase the voltage above the minimum of the dc-dc converter
input voltage. Experimental results show that for the charge is preferable the series
type when the input power is high due to low losses, while the individual type
is preferred when the input power is low. The series type is effective in reducing
the residual energy while the individual type can minimize the losses. The energy
stored is proportional to V2 , which means that two voltage conversions are usually
required: a transducer from energy to charge the super capacitor, the other from the
SC to supply the load. The residual energy below the minimum voltage of the dc-dc
converter or charge pump normally is not usable and grows linearly with the value
of the capacitance and quadratically with the voltage value. Another problem is the
leakage current that grows exponentially in the vicinity of the nominal voltage (2.5–
2.7 V); typically the supercapacitors have a voltage range from 0 to 2.7 V.
8.10 Electrical Storage Buffer 101

20
10F Supercapacitor
15 25F Supercapacitor
150F Supercapacitor
10

0
0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4

Fig. 8.35 Leakage power loss (mW) in the SC as function of the supercapacitor voltage (V)

The leakage current and power are given by the following expressions:
V.t/  V.t C t/
Il .t/ D C (8.5)
t
V 2 .t/
Pl .t/ D Il .t/V.t/ D C (8.6)
t
From these equations it is possible to derive the energy loss due to leakage:
Z t
El .t/ D Pl .t/dt (8.7)
0

The Fig. 8.35 shows how the power of the leakage increases with voltage across
of the supercapacitor, in particular when it is closer to the nominal voltage.
In addition, the following variables can be defined:
• Pcharge : power supplied by the transducer (at time t), assumed maximum and
converted to the appropriate voltage, which can be collected by the SC or less.
If it is not accepted, then it is dissipated as heat.
• Poverhead : sum of the conduction losses and switching losses.
• Eheld : energy possessed by the supercapacitor (at time t) less leakage and
switching losses.
• Eresidual : power unusable in SC with voltage below the minimum threshold of the
dc-dc converter. item PLOAD : power supplied to the load from the SC to the time t.
In terms of power is possible to define the charging power efficiency (c ) and the
driving power efficiency (d ) as follows:

Paccepted
c D (8.8)
Pchanged
Pload
d D (8.9)
Pdischarge
102 8 Powering Microsystem

In terms of energy is possible to define the charging energy efficiency, the energy
efficiency of storage, the driving energy efficiency, and overall end-to-end energy
efficiency:
Rt
Paccepted .t/dt
Echarging D R0 t (8.10)
0 Pcharge .t/dt
Rt
0 .Paccepted .t/ Pleak .t/  Poverhead .t//dt
Estorage D Rt (8.11)
0 Paccepted .t/dt
Rt
0 Pload .t/dt
Edriving D Rt Rt (8.12)
0 Pdischarge .t/dt C 0 Poverhead .t/dt C Eresidual .t/

There are three types of Energy Storage Elements (ESE) for energy harvesting
(EH) systems:
• Single Supercapacitor (SS): A Single Small Supercapacitor (SSS, about 1 F) can
charge quickly with a charging time long.
• Reservoir Supercapacitor Array (RSA): an array of SSS which allows to reduce
the losses of leakage.
• Dynamic Reconfigurable Supercapacitors: This type allows to reduce the leakage
and Eresidual and improves the efficiency of the output stage.

References

1. Park, J., & Mackqy, S. (2003). Practical data acquisition for instrumentation and system
control. Amsterdam: Elsevier.
2. Lacanette, K. (2003). National temperature sensors handbook. Handbook Ann. Mat. National
semiconcductor.
3. National Instruments (1996). Data acquisition fundamentals, application note 007 (1996).
4. National Instruments (1996). Signal conditioning fundamentals for PC-based data acquisition
systems, Handbook National Instruments
5. Taylor, J. (1986). Computer-based data acquisition system. Instrument Society of America,
USA.
6. Di Paolo Emilio, M. (2013). Data acquisition system, from fundamentals to applied design.
New York: Springer.
7. Roundy, S., Wright, P., & Pister, K. (2002). Micro-electrostatic vibration-to- electricity
converters. In Proceedings of ASME International Mechanical Engineering Congress and
Exposition IMECE2002 (Vol. 220, pp. 17–22).
8. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: Self- sufficient energy
supply for micro systems. In Proceedings of the 16th International Conference on Thermo-
electrics (pp. 575–577).
9. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. IEEE
Micro, 21(3), 30–42.
10. Roundy, S. (2003). Energy scavenging for wireless sensor nodes with a focus on vibration to
electricity conversion. PhD thesis, University of California.
References 103

11. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006). Micro seismic
electret generator for energy harvesting In Technical Digest PowerMEMS 2006, Berkeley,
USA, November 2006 (pp. 133–136).
12. Sterken T., Altena G., Fiorini P., & Puers, R. (2007) Characterisation of an electrostatic
vibration harvester, DTIP of MEMS and MOEMS, Stresa, Italy, April 2007.
13. Sterken, T., Baert K., Puers, R., & Borghs, S. (2002). Power extraction from ambient vibration.
In Proceedings of the SeSens, Workshop on Semiconductor Sensors, Veldhoven, Netherlands,
November 2002 (pp. 680–683).
14. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Electronics, 27(2), 803–815. ISSN:0885-8993.
15. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning a
resonant energy harvester using a generalized electrical load. Journal of Smart Materials and
Structure, 19, 055003.
16. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
17. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
18. Szarka, G. D., Burrow, S. G., & Stark, B. H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
19. Maurath, D., Becker, P. F., Spreeman, D., Manoli, Y. (2012). Efficient energy harvesting with
electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6), 1369–1380.
20. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
21. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies - state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
22. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. In Proceedings of the World Congress on Engineering (Vol. 2).
23. Zhu, D., Tudor, M. J., & Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
24. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy harvesters
- a classification approach Micromachines, 4(2), 168–196. http://www.mdpi.com/2072-666X/
4/2/168. Accessed 20 January 2015.
25. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of applied electromagnetics
(6th ed.). Upper Saddle River: Prentice Hall, Oxford.
26. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
27. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008) A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
30. Razavi, B. (2002). Design of analog CMOS integrated circuits. New York: McGraw-Hill.
31. Razavi, B. (2008). Fundamentals of microelectronics. London: Wiley
104 8 Powering Microsystem

32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University Press.
33. Razavi, B. (2002). Design of integrated circuits for optical communications. New York:
McGraw-Hill.
34. Hurst, P. J. (2001). Analysis and design of analog integrated circuits. New York: Wiley.
35. Spies, P. (2015). Handbook of energy harvesting power supplies and applications. Boca Raton:
CRC Press Book, France.
Chapter 9
Low-Power Circuits

9.1 Introduction

Most of the signals applied to the inputs of an electronic system are originated from
the devices called sensors which, on the basis of their input–output characteristic,
convert non-electrical quantities (for example, a temperature) into corresponding
electrical values in analog form (for example, a voltage). Furthermore, most of the
signals present at the outputs of an electronic system are used to drive devices called
actuators. For this reason, although the processing of the information takes place
typically in digital form, all electronic systems require the presence of circuits for
processing analog signals, if only to convert these analog signals into digital signals
(analog/digital conversion or A/D) and vice versa (D/A conversion). The main
operations that can be performed on analog signals are the amplification (which
is equivalent to multiplying the signal by a constant), the weighted sum of several
signals, the operations of derivation and integration in time, and the filtering in the
frequency domain [1–15].

9.2 Review of Microelectronics

The earliest electronic circuits were fairly simple. They were composed of a
few tubes, transformers, resistors, capacitors, and wiring. As more was learned
by designers, they began to increase both the size and complexity of circuits.
Component limitations were soon identified as this technology developed. The
transition from vacuum tubes to solid-state devices took place rapidly. As new types
of transistors and diodes were created, they were adapted to circuits. The reductions
in size, weight, and power use were impressive. Microelectronic technology today
includes thin film, thick film, hybrid, and integrated circuits and combinations of
these. Such circuits are applied in DIGITAL, SWITCHING, and LINEAR (analog)

© Springer International Publishing AG 2017 105

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_9
106 9 Low-Power Circuits

circuits. Because of the current trend of producing a number of circuits on a single

chip, you may look for further increases in the packaging density of electronic
circuits. At the same time you may expect a reduction in the size, weight, and
number of connections in individual systems. Improvements in reliability and
system capability are also to be expected.

9.2.1 Basic of Semiconductor’s Physics

Microelectronics is based on physics of structures. The main semiconductor ele-

ments used in the applications are silicon, boron, aluminum, etc. Silicon atom has
4 valence electrons and it’s important for electrical system. At temperature close
to 0 K, silicon crystal behaves as an insulator: its electrons are confined to their,
respectively, covalent bonds. At higher temperature, electrons gain thermal energy
and can be used as “free charge.” Bandgap energy, minimum energy to broke the
bond, of the silicon is 1.12 eV where 1 eV D 1:6  1019 J. The number of “free
electrons” that is possible to have in a semiconductor crystal depends on band
gap energy and temperature, it’s possible to define density of electrons (number
of electrons per unit volume) as the following:

3 Eg
n D 5:2  1015 T 2  e 2KT (9.1)

with k D boltzmann constant D 1:38  1023 J/K. For insulator materials Eg '
2:5 eV, instead, for conductor is about less than 1 eV. As example we can calculate
the density of electrons for silicon at T D 300 K:

3 1:721019
n D 5:2  1015 T 2  e 2KT  1010 e=cm3 (9.2)

The intrinsic semiconductors can’t be used for electrical application due to very high
resistance but also a small number of “free electrons” with respect to the conductor.
In an extrinsic semiconductor the situation is different: there is a p-Si and an n-Si.
In the first case majority carriers (holes) are p D NA , semiconductor is doped with
a density of NA (i.e., boron atom) and minority carriers n D p2i =NA ; in the other
case, instead, majority carriers (electrons) n D NA with semiconductor is doped
with a density of ND (i.e., phosphorus), minority carriers p D n2i =ND . The concept
of doping extends the flow of current in a semiconductor.

9.2.1.1 Drift

The drift is the movement of electrons charge due to an electric field: v D

E,  is “mobility” expressed in cm2 =V  s. If consider a piece of n-silicon
9.2 Review of Microelectronics 107

V
h
W

Fig. 9.1 Current flow in terms of charge density

(n D 1350 cm2 =V  s) of L D 1 m applied a voltage of V D 1 V, the velocity

of electrons (drift) is about v D n E D n VL ' 107 cm=s. For a p-silicon
p D 480 cm2 =s.
Now, we should calculate the current due to this drift v (Fig. 9.1):

I D v  W  h  n  q (9.3)

It can be written in another way:

Jn D n  E  n  q (9.4)

where J is the current density, W  h is the cross section, and n  q is the charge density
in column of the semiconductor. In presence of both electrons and holes:

J D Jn C Jp D q.n n C p p/E (9.5)

9.2.1.2 Diffusion

The mechanism of the diffusion is the movement of the charge from a zone of high
concentration to zone of low concentration (current). The mathematical relation that
explains this phenomena is the following:

dn
I DAqDn (9.6)
dx
108 9 Low-Power Circuits

where Dn is the “diffusion constant” in cm2 /s; in intrinsic silicon Dn D 34 cm2 =s

(for electrons) and Dp D 12 cm2 =s (for holes). A is the cross section, q is the charge,
and dn
dx
is the concentration (electrons) with respect to the direction “x” (Fig. 2.1).
From the last equation we can calculate the current density for electrons and holes:

dn
Jn D qDn (9.7)
dx
dp
Jp D qDp (9.8)
dx

9.2.2 PN Junction

From the doping we have obtained p-type and n-type semiconductors. A electric
field or a concentration gradient leads to the movement of the charges (electrons
and holes). We suppose to dope two adjacent pieces of semiconductors (Fig. 9.2);
in this configuration there will be a flow of electrons from n to p side and a flow of
holes in opposite direction (Fig. 9.3).
At the end of this process of “equilibrium,” an electric field (depletion region)
will emerge as indicated in Fig. 9.3. The junction reaches equilibrium once the

Cathode (n)

n p

Anode (p)

Si Si Si Si
–
P e B
Si Si Si Si

Fig. 9.2 Current flow in terms of charge density

n p n p n p
– – – – + + + + – – – + – + + + – – – + + – – + + +
– – – – – + + + + + – – – – + + + + – – – + + +
– – – – + + + + – – – + – + + + – – – + + – – + + +
– – – – – + + + + + – – – – + + + + + – – – + + + + +
– – – – + + + + – – – – + + + – – – – – + + +
– – – – – + + + + + – – – – + – + + + + – – – + + – – + + +
– – – – + + + + – – – + + + + – – – + + + + +
– – – – – + + + + + – – – – – + + + + – – – – – + + +
Free electrons Free holes Depletion region

Fig. 9.3 Evolution of charge concentration in a PN junction

9.2 Review of Microelectronics 109

electric field is strong enough to completely stop the diffusion current [16–25].
The existence of electric field due to depletion region suggests that PN junction
has a built-in potential defined in the following equation:
 
Dp pp
V.x2 /  V.x1 / D  ln (9.9)
p pn

Considering the Einstein’s equation, D


D KT
q
:
 
KT pp
jV0 j D ln (9.10)
q pn

where pn and pp are the concentrations at x1 and x2 , respectively. The PN junction

so realized is called diode. Having analyzed the PN junction in equilibrium, let us
observe how it works applying an external voltage (Fig. 9.4).

9.2.2.1 Reverse Bias

The external voltage rises the electric field of depletion region prohibiting the flow
of current. The device works as capacitor:

VR1
VR2

n p n p
– – – + + – – + + + – – – + + – – + + +
– – – + + + – – – + + +
– – – + + – – + + + – – – + + – – + + +
– – – + + + + + – – – + + + + +
– – – – – + + + – – – – – + + +
– – – + + – – + + + – – – + + – – + + +
– – – + + + + + – – – + + + + +
– – – – – + + + – – – – – + + +

– + + ++ – – – – + – – – – +– + + + +
– + + ++ – – – – + – – – – +– + + + +
– + + ++ – – – – + – – – – +– + + + +

VR1 VR2

Fig. 9.4 PN junction with external voltage

110 9 Low-Power Circuits

s
Cj0 q
si q NA ND 1
Cj D q I Cj0 D (9.11)
1 VR 2 NA C ND V0
V0

si is the dielectric constant of silicon.

9.2.2.2 Forward Bias

The external voltage decreases the electric field of the depletion area allowing
greater diffusion current (Fig. 9.4):
   
VR
ID D Is exp 1 (9.12)
VT

with VT , thermal voltage, and Is :

 
Dn Dp
Is D Aqn2i C (9.13)
NA Ln ND Lp

is the “reverse saturation current” and Ln and Lp are the electrons and holes
“diffusion length” (i.e., Ln D 20 m , Lp D 30 m)

9.2.3 Diode

The diode is a two terminal device with I-V characteristic indicated in Fig. 9.5. Some
application can be described in the following text:
• Wave rectifier: the circuit is visualized in Fig. 9.6: the ripple amplitude can be
calculated by:

Vp  VD;on
Vr  (9.14)
RL C1 fin

• Voltage regulation: due to significant variation of the line voltage, it is necessary

to be produced in the output of the diode a stabilized voltage. A possible outline
is visualized in Fig. 9.7.
• Limiting circuit: the circuit passes the input to the output, Vout D Vin and when
the input exceeds a “threshold” the output remains constant (Fig. 9.8).
9.2 Review of Microelectronics 111

Fig. 9.5 Characteristic I-V of the diode

9.2.4 Bipolar Transistor: Emitter Follower

The union of two junctions p–n (i.e., two diodes together) form the bipolar transistor
junction (BJT). Bipolar because the current is sustained by electrons and holes
(such as the diode). Compared to the diode, the BJT (three terminals) can be used
as a signal amplifier. Although the MOS technology (see next paragraph) is more
widespread, the technology bipolar remains significant (or predominant, in certain
cases) in several applications:
• Electronics vehicle
• Systems wireless
• Digital circuits ECL
• Draft discrete circuits.
The emitter follower circuit is particularly useful for applications where high
input impedance is required. It is typically used as a buffer in a wide variety
of areas. Emitter follower is a common collector transistor configuration. It can
easily be designed by circuit RC. The emitter follower is also known as a voltage
follower, or a negative current feedback circuit, with high input impedance and
low output impedance. The outline of the emitter follower is shown in Fig. 9.9 and
the corresponding equivalent small signal circuit in Fig. 9.10. We can calculate the
effective resistance seen from terminal B by:
112 9 Low-Power Circuits

D4 D1

– +

D2 D3 C1 RL

ripple Vdc

0V

Fig. 9.6 Wave rectifier

Fig. 9.7 Block diagram of voltage regulator with diodes

Rib D r C .ˇ C 1/RE ==RL ' .ˇ C 1/RE ==RL (9.15)

This value is relatively high. In general, RE ==RL is around k˝ and ˇ  100, so the
resistance seen from terminal B is in the hundreds of k˝. It is evident from Fig. 9.10
that input resistance depends on load resistance. Let us named the input resistance
as Rin :

Rin D RB ==Rib (9.16)

The effective resistance is the parallel between Rib and Rb . A common collector
configuration can be used as an amplifier in such a circuit, where a large input
9.2 Review of Microelectronics 113

Fig. 9.8 Limiting circuit

resistance is needed; a good application can be a pre-amplifier circuit. From the

voltage gain equation (A D V0 =Vi ) follows:

V0 RE ==RL Rin
AD D  (9.17)
Vs re C .RE ==RL / Rs C Rin

It becomes evident that as long as re  .RE ==RL ) and Rs  Rin the gain
approaches unity.
The emitter follower can be designed using the main steps described below:
• Choose a transistor: the transistor should be selected according to the system
requirements.
• Select emitter resistor: selecting a working point (for example, select an emitter
voltage of about half the supply voltage).
• Determine the base current: base current is the collector current divided by ˇ
(or hfe ).
114 9 Low-Power Circuits

power supply

R1
iC

RS +
vCE
–
+
VS
– R2 RE
v0
RL
iE i0

Fig. 9.9 Emitter follower (outline)

RS

ib rπ ic
+ ia
vs –
ie
RB
RE RL
io vo

Fig. 9.10 Emitter follower (small signal equivalent circuit)

• Determine the base resistor values: select the value of the resistor(s) to provide
the voltage required at the base.
• Determine the value of the input and output capacitor: The value of the
input/output capacitor should equal to the resistance of the input/output circuit
at the lowest frequency of operation.
When using the emitter follower circuit, there are two main practical points to
note:
• The collector may need decoupling: in some cases the emitter follower may
oscillate, in particular if long leads are present. One of the easier ways to prevent
9.2 Review of Microelectronics 115

Fig. 9.11 Emitter follower V1

(example with 2N3904)
12V

R2

10k

C1

2N3904
1.5µF

V2 Out
+
– R3 R1

10k 1k

this is to decouple the collector to ground with very short connections, or by

placing a small resistance between the collector and the power supply line.
• The input capacitance affects the RF: the base emitter capacitance may reduce
the high impedance of the input circuit if the signal is above 100 kHz.
One of the designs of an emitter follower is shown in Fig. 9.11, with a P-spice
simulation in Figs. 9.12 and 9.13. This configuration can be typically used in AC
coupling applications. It’s important that the frequency response and bias voltage
on the base are planned for. In the circuit mentioned here, there’s an added voltage
divider, consisting of R2 and R3 , and an AC coupling capacitor, C1 . Their component
values will need to be calculated. An important aspect of the emitter follower is
its relative immunity to temperature instability. When the current in the collector
increases (changing of ˇ), the voltage across the emitter resistor also increases. This
acts as a negative feedback, since it reduces the voltage difference between base and
emitter (VBE ), which drives the transistor to conduct more current, as such thermal
stability is maintained [26–30].
116 9 Low-Power Circuits

V[out]
7.2V
6.9V
6.6V
6.3V
6.0V
5.7V
5.4V
5.1V
4.8V
4.5V
4.2V
3.9V
136ms 144ms 152ms 160ms 168ms 176ms

Fig. 9.12 Simulation of emitter follower, output voltage

Fig. 9.13 Simulation of emitter follower, frequency response

9.2.5 MOS Transistor

Today the microelectronics is dominated by MOSFET devices. The physical

structure of the MOSFET is described in Fig. 9.14: on a substrate of monocrystalline
p-type, two junctions are made of type nC which are connected by two terminals
called drain and source (D and S in Fig. 9.14). In the area between the drain and
source is made to grow a layer of silicon dioxide as excellent insulator thick less
than 0.01 m.
The gate terminal is isolated from the Si substrate by a thin layer of SiO2 , and
therefore the gate current DC is null. The body terminal is generally connected to
that source. Drain and source are symmetrical.
Applying a sufficiently positive voltage between the gate and source, the
electrons are attracted towards the Si  SiO2 interface under the gate forming a
conductive channel between source and drain. In these conditions, if it is applied a
Vds > 0, a current may flow between the drain and source. The current between drain
and source is controlled by the voltage between gate and source, which controls
the formation of the channel. The electrical characteristics of the MOSFET depend
from L (gate length) and W (gate width), as well as technological parameters such as
oxide thickness and doping of body. Typical values of L and W are: L D 0:1  2 m,
W D 0:5  500 m. The range of gate oxide thickness is 3–50 nm.
9.2 Review of Microelectronics 117

S G D
D
Oxide Drain
Metal gate

Source

G
B
n+ n+
L Substrate (or body)
p

B S

Fig. 9.14 Inside outline (general) and symbol of Mosfet

There are four types of MOS transistors: two to channel n and two to channel p.
The MOSFET n-channel (nMOS) is formed on a p-type substrate:
• nMOS enrichment (enhancement) or normally off
• nMOS depletion (depletion) or normally on.
The MOSFET p-channel (pMOS) is formed on a substrate of n-type:
• pMOS enrichment (enhancement) or normally off
• pMOS depletion (depletion) or normally on.
In most MOSFET applications, an input signal is the gate (G) voltage Vg and the
output is the drain (D) current Id . The ability of MOSFET to amplify the signal is
given by the output/input ratio: the transconductance, gm D dI=dVgs .
In Fig. 9.15 reports the characteristic curve Id  Vds ; there are three regions of
work:
• Cutoff: in this case its necessary induce the channel, Vgs  Vt (Vt is the threshold
voltage) for nMOS.
• Triode: the channel must be induce and also keep Vds small enough so the channel
is continuous (not pinched off): Vds  Vgs  Vt .
• Saturation: in this mode need to induce the channel, Vds Vgs  Vt then ensure
that the channel is pinched off at the drain end. In Fig. 9.16 is visualized the plot
of Id versus Vgs for an enhancement type nMOS device in saturation.
In the saturation mode, this device is an ideal current source (Fig. 9.17). In reality,
there is to consider a finite output resistance (r0 ); the outline of Fig. 9.18 can be
described as in Fig. 9.17.
While the transconductance gm gives the variations of the drain current due to
variations of the voltage Vgs , there is another fundamental parameter of the Mosfet
that takes the name of the output conductance, which is the variation of the drain
current Ids due to variations of the voltage Vds .
The output conductance assumes a much greater importance, and the reason is
the effect of the so-called channel length modulation due to Vds : it is the effect
for which the effective length of the channel decreases with increasing Vds and in
118 9 Low-Power Circuits

Vds=Vgs–Vt

ID Triode region Saturation region

Vgs
Vgs ≤ Vtn (cutoff)

Vds

Fig. 9.15 Characteristic curve Id  Vds

ID

Vt Vgs

Fig. 9.16 Saturation, Id  Vgs

the saturation zone, the current increases with the Vds . From an analytical point of
view the effect of the modulation of the channel length can be written as follows:
Ids D k.Vgs  Vt /  2  .1 C Vds /. It is clear that provides a linear dependence of the
Ids vs Vds in according to named as parameter of the channel length modulation; k
is the transconductance factor proportional to the geometry of the Mosfet.
Example of application with Mosfet can be the common source amplifier
visualized in Figs. 9.19 and 9.20. The input signal is applied to the gate through
the coupling capacitor C1 . The output is on the drain and connected to the load
9.2 Review of Microelectronics 119

id
gate
+ drain

gmvgs
vgs

– source
–

Fig. 9.17 Mosfet model (ideal)

gate drain
+ +
id
gmvgs ro
vgs vds

– source –

Fig. 9.18 Mosfet model (not ideal)

Vdd
Vdd
RD
R1 C2 i0

+
C1
R + M1 RL
V0

–
+ Vin
–
V(t)

R2 Rs Cs

Fig. 9.19 Common source amplifier

by C2 , C1 , C2 , and CS which are the coupling capacitors and therefore can be

considered short circuits at the frequencies of the signal (center-band). The source
(S) is therefore to ground for the signals. R1 , R2 , RD , and RS form a bias network to
four resistors.
120 9 Low-Power Circuits

i0
iin D
R G + +
+ RL
rd
Vgs RD
Vin(t) R2 R1 V0
+
– V(t) gmVgs

– – –
S
RG

RL

Fig. 9.20 Common source amplifier (model)

9.2.6 Differential Amplifiers

The main part of analog-integrated-circuit design is the differential-pair or

differential-amplifier configuration. The differential amplifiers were introduced
in electronics to eliminate all or part of the problems of the amplifiers with direct
coupling. The goal is to create amplifiers characterized by an acceptable signal/noise
ratio (SNR), also in the presence of external/internal disturbances generated by
thermal variations due to the aging of electronics components. The differential-
pair of differential-amplifier configuration is widely used in IC circuit design. One
example is the input stage of an op-amp and the emitter-coupled logic (ECL).
This technology was invented in the 1940s in vacuum tubes; the basic differential-
amplifier configuration was later implemented with discrete bipolar transistors.
However, the configuration became most useful with the invention of the modern
transistor/MOS technologies.
The main features of differential amplifiers are as follows: (1) High input
resistance, so small voltage signals can be amplified without losses. (2) Temperature
drift is minimal. (3) Two input terminals, i.e., inverting and non-inverting inputs. (4)
It amplifies the difference between two input signals.
The differential amplifier works only on dual power supplies: it requires both
CVcc and Vcc voltages simultaneously. However, if the same circuit is connected
to a single power supply, its working becomes unstable and we do not get proper
amplification from the circuit. The differential amplifier (Fig. 9.21) provides a
number of advantages, making it one of the most useful circuit configurations,
particularly as input stage for high gain and DC amplifiers. The figure of merit
for differential amplifiers is called common-mode rejection ratio (CMRR), defined
as the ratio between difference mode gain (Ad) and common-mode gain (Ac).
Differential amplifiers is a special purpose amplifier designed to measure differential
signals, otherwise known as a subtractor. The CMRR can be calculated with the
following equation:
9.2 Review of Microelectronics 121

+15V

R1 Rc Rc R1
c vout vout' c

vi Q1 Q2 vi'
2N3904 2N3904
c c
A
R2 R2
Rr Rr
RE

Fig. 9.21 Example of a differential amplifier

Rc Rc 2RE C Rr C rtr
Ad D I Ac D I CMRR D (9.18)
2.Rr C rtr / 2RE C Rr C rtr 2.Rr C rtr /

where rtr is the transresistance, usually indicated also as re . The 741 (a common
op-amp chip) has a CMRR of 90 dB, which is reasonable in most cases. A value
of 70 dB may be adequate for applications insensitive to the effects on amplifier
output; some high-end devices may use op-amps with a CMRR of 120 dB or more.
The common-mode rejection ratio (CMRR) relates to the ability of the op-amp to
reject a common-mode input voltage. This is very important because common-mode
signals are frequently encountered in op-amp applications.
Common application of differential amplifiers is in the control of motors or
servos, as well as in signal amplification. In discrete electronics, a common arrange-
ment for implementing a differential amplifier is the long-tailed pair, typically found
as a differential element in most op-amp integrated circuits. A long-tailed pair can
be used as an analog multiplier with a differential voltage as one input and biasing
current as another. There are many differential amplifier ICs on the market today.
The AD8132 from Analog Devices (ADI) is a low-cost differential (or single-ended)
amplifier with one resistor for setting the gain. The AD8132 is a major improvement
on op-amps, especially for driving differential input ADCs or signals over long
lines. The AD8132 can be used for differential signal processing (gain and filter-
ing) throughout a signal chain, significantly simplifying the conversion between
differential- and single-ended components. Linear Technology also provides many
differential amplifiers for different applications. For example, its LTC6409 offers a
very high speed and low distortion, and is stable in a differential gain of 1.
122 9 Low-Power Circuits

9.2.7 Feedback

Amplifiers can be considered not perfectly linear. The gain (or amplification) of the
amplifier changes with power supply or temperature due to the variations of the
working point of the transistors. These and other real limitations of the amplifiers
can be minimized with the use of negative feedback. The good functionality of an
amplifier and sometimes limited by the presence of extraneous signals, such as the
hum of power supply and coupling with other amplifiers neighbors. In some specific
cases, the negative feedback can reduce these effects, while in other cases it does
not induce any improvement. An important example are the amplifiers and hi-fi
audio systems, which are powered with DC voltage obtained by rectifying AC power
supply and not perfectly stable.
In Fig. 9.22 is shown a possible feedback outline. The output signal of the
amplifier is always smaller than power supply voltages and its shape can be a
clear demonstration that the output is not linear. The result of this non-linearity
is manifested as a distortion of the output signal.
The use of negative feedback allows to realize amplifiers with better perfor-
mance. The main effects of feedback loop can be the following:
• greater stability of gain
• less distortion
• greater bandwidth
• reduction of the noise.
In general, the behavior of the individual block of negative feedback can be
described by transfer function (Fig. 9.23).
The G block is called forward, the block H is the feedback: the output signal
Vout (t) is sent back, through H, in input and is controlled by comparator. The
difference signal represents the error of the system, it acts as a input signal to the
block forward to have in output the corrected signal.

Fig. 9.22 Feedback outline

9.2 Review of Microelectronics 123

Open Loop
Σ Amplifier
Vin Vin - β.Vout Vout
+ G
–

Vf

(β.Vout) Vout
H

Feedback Network

Fig. 9.23 Negative feedback outline (general)

If the system is sufficiently fast, any changes of G block do not affect the output
signal. More generally, if the input signal changes in time, the negative feedback
system gives the necessary corrections to ensure that the output signal is a faithful
repetition (with amplification) of the output.
Otherwise a positive feedback can be used to design oscillators. Positive feedback
is responsible for the squealing of microphones when placed too close to the speaker
through which their input signals are amplified.
When the loop gain is positive and above 1, there will typically be exponential
growth, increasing oscillations or divergences from equilibrium.

9.2.8 Effects of Feedback

A disturbance or noise is an unwanted signal that is superimposed on the

input/output of a circuit. Disruptions can occur in various parts of the system,
their effect is greater in the input.
Some consideration about the effects of the noise in the loop systems:
• the open loop systems are not able to compensate for the disturbance but if noise
is present at the output the effect is smaller because the useful signal which is
superimposed has an amplitude greater.
• the closed loop systems are able to limit the effect of the disturbance when this
is present in the output but not when it is present at the input.
The effect of disorder can be evaluated through the ratio signal-to-noise (S/N or
SNR). The time that elapses between the moment in which it has the effect and the
moment in which this effect is taken into account to modify the system is called
“delay in the feedback loop.” When this delay is high, there may be problems of
stability. The Nyquist criterion allows to investigate the stability of the closed loop
124 9 Low-Power Circuits

system of the known transfer function in open loop and in particular its polar plot or
Nyquist plot. The Nyquist plot of a closed system in feedback is a representation in
the Gaussian plane of the value of frequency response function in open loop, G.j!/,
in terms of the real part and imaginary part, with the variation of the pulsation !.

9.2.9 Digital CMOS Circuits

It is virtually impossible to find electronic devices in our daily lives that do not
contain digital circuits, and with most of them having CMOS logic devices at their
heart. There are a large number of CMOS devices, divided in a number of families.
Using only very few components, it is possible to build fairly elaborate pulse and
signal generators.
Depending on the doping material used, there are mainly two types of metal-
oxide-semiconductor field-effect transistors (MOSFETs): the n-channel, or nMOS,
and p-channel, or pMOS. In N-type metal-oxide-semiconductor (NMOS) logic,
n-type MOSFETs are used to implement logic gates and other digital circuits. These
circuits are mostly used for switching due to their high speed nature, whereas PMOS
circuits are slow to transition from high to low state, and their asymmetric input
logic levels makes them susceptible to noise. However, metal-oxide-semiconductor
(CMOS) technology offers some attractive practical advantages over NMOS tech-
nology: high noise immunity and low static-power consumption. CMOS uses a
combination of p- and n-channel MOSFETs as building blocks, but here both low-
to-high and high-to-low output transitions are fast since the pull-up transistors have
low resistance when switched on, unlike the load resistors in NMOS logic. In
addition, the output signal swings the full voltage between the low and high rails.
This strong, nearly symmetric response also makes CMOS more resistant to noise
[31, 32].
In NMOS circuits the logic functions are realized by arrangements of NMOS
transistors, combined with a pull-up device that acts as a resistor. The concept of
CMOS circuits is based on replacing the pull-up device with a pull-up network
(PUN) that is built using PMOS transistors, such that the functions realized by
the PDN and PUN networks are complements of each other. Then a logic circuit,
such as a typical logic gate, is implemented as indicated in Fig. 9.24. It comprises
a network of NMOS transistors (pull-down network) and a network of PMOS
transistors (pull-up network), but each consisting of an equal number of transistors.
Each input variable requires an NMOS transistor in the pull-down network and a
PMOS transistor in the pull-up network.
9.2 Review of Microelectronics 125

VDD

Pull-up network
(PUN)

Vf

Vx1
Pull-down network
Vx2 (PDN)

Fig. 9.24 The structure of CMOS circuit

9.2.10 CMOS Inverter

A CMOS inverter (Fig. 9.25) is composed of two MOSFETs, with their gates
connected to the inverters input line, and their drains to the output line. The input
resistance of the CMOS inverter is extremely high, as the gate of an MOS transistor
is virtually a perfect insulator and draws no input direct current. Since the input
node of the inverter only connects to the transistor gates, the steady-state input
current is nearly zero. A single inverter can theoretically drive an infinite number of
gates (or have an infinite fan-out). A CMOS inverter dissipates a negligible amount
of power during steady-state operation, which occurs only during switching. This
makes CMOS technology useable in low-power and high-density applications.

9.2.11 Current Mirror

Current mirrors replicate the input current of a current sink or current source in
an output current, which may be identical or a scaled version. Current mirrors are
used to provide bias currents and active loads to circuits. Apart from some special
circumstances, a current mirror (Figs. 9.26 and 9.27) is one of the basic building
blocks of the operational amplifier; it is a circuit designed to keep the output current
126 9 Low-Power Circuits

Fig. 9.25 CMOS inverter

vcc R1

IR
IO
+

IB1 IB2

T1 T2

IE1 IE2

Fig. 9.26 Current mirror with BJT

constant regardless of loading. This type of topology may therefore be used in

order to create current generators. The main requirements a current mirror must
meet are:
• Output current independence of the output voltage.
• Wide range of output voltages at which the mirror is working properly.
• Low input voltage.
9.2 Review of Microelectronics 127

gmVGS goVDS
VGS VDS

Fig. 9.27 Current mirror with Mosfet (model)

The range of voltages within which the mirror works is called the “compliance
range,” and the voltage marking the behavior in active/linear region is called the
“compliance voltage.” There are also a number of secondary performance issues
with mirrors, such as temperature stability for example.

9.2.12 Ideal Current Mirror

A current mirror is usually and simply approximated by an ideal current source.

However, an ideal current source is unrealistic for several reasons:
• It has an infinite AC impedance, whereas a practical mirror has a finite
impedance;
• It provides the same current regardless of voltage, that is, there are no compliance
range requirements;
• It has no frequency limitations, whilst a real mirror has limitations due to the
parasitic capacitances of the transistors;
• The ideal source has no sensitivity to real-world effects like noise, power supply
voltage variations, and component tolerances.
The main part of the current mirror is a bipolar transistors (BJT) or a MOSFET.
Transistors in a current mirror circuit must be maintained at the same temperature
for precise operation. There are different types of current mirrors:
• Simple current mirror (BJT and MOSFET);
• Base current corrected simple current mirror;
• Widlar current source;
• Wilson current mirror (BJT and MOSFET);
• Cascode current mirror (BJT and MOSFET).
All of the circuits have compliance voltage that is the minimum output voltage
required to maintain correct circuit operation: the BJT should be in the active/linear
region and the MOSFET should be in the active/saturation region.
128 9 Low-Power Circuits

9.2.13 Current Mirror BJT/MOS

Current mirror circuits are usually designed with a BJT, such as an NPN transistor,
where a positively doped (P-doped) semiconductor base is sandwiched between
two negatively doped (N-doped) layers of silicon. These transistors are specifically
designed to amplify or switch current flow. In some current mirror design specifica-
tions, the NPN transistor works as an inverting current amplifier, which reverses the
current direction, or it can regulate a varying pulse current through amplification to
create output mirror properties. One of the reasons that BJTs are used for current
mirror design is due to the base-emitter (or PN part) of the transistor functioning
reliably like a diode. Diodes regulate both the amount of current that passes and
the forward voltage drop for that current. The basic current mirror can also be
implemented using MOSFET transistors (Fig. 9.28). In Fig. 9.28, M1 is operating
in the saturation or active mode, and so is M2 . In this setup, the output current IOUT
is directly related to IREF . The drain current of a MOSFET ID is a function of both
the gate-source voltage and the drain-to-gate voltage of the MOSFET given by a
relationship derived from the functionality of the MOSFET. In the case of transistor
M1 of the mirror, ID = IREF . Reference current (IREF ) is a known current and can be
provided by a resistor or by a “threshold-referenced” or “self-biased” current source
to ensure that it is constant and independent of voltage supply variations.

Fig. 9.28 Current mirror Vcc

with Mosfet

Iref

Iout

Vout

M1 M2
9.3 Low-Power MOSFET 129

9.3 Low-Power MOSFET

A field effect transistor (FET) works as a conduction channel in the semiconductor

with two ohmic contacts, drain and source, where the number of charge carriers in
the channel is controlled by a third contact, the gate. The most important FET is
the MOSFET. In a MOSFET of silicon, the gate is separated from the channel by
an insulating layer of silicon dioxide (SiO2 ). The charge carriers of the conductive
channel constitute a charge inversion, i.e., electrons in the case of a p-type substrate
(n-channel device) or holes in the case of an n-type substrate (p-channel device), that
are induced in silicon-insulator interface by the voltage applied to the gate electrode.
The MOSFETs are used both as discrete components that active elements in digital
and analog monolithic integrated circuits (IC).

9.3.1 General Characteristics of a MOSFET

The structure of a traditional metal-oxide-semiconductor (MOS) is obtained by

growing a layer of silicon dioxide (SiO2 ) on top of a silicon substrate, depositing
in turn a layer of metal or polycrystalline silicon. Since the silicon dioxide is a
dielectric material, its structure is equivalent to a planar capacitor, with one of the
electrodes replaced by a semiconductor. A transistor of metal-oxide-semiconductor
field effect (MOSFET) is based on the modulation of charge concentration by MOS
capacitor between an electrode and the gate located over the substrate (body) and
isolated from all other regions of the device to a dielectric layer which, in the case
of a MOSFET, is an oxide. Compared to the MOS capacitor, the MOSFET includes
two additional terminals (source and drain), each is connected to the single highly
doped regions and separated from the body region or substrate (substrate). Such
regions may be of the type p or n. If the MOSFET is an n-channel or nMOSFET,
then the source and drain are “nC” and the substrate is a region “p”. If the MOSFET
is a p-channel or pMOSFET, then the source and drain regions are “pC” and the
substrate is a region “n”. By applying a sufficiently positive voltage between gate
and source, the electrons are attracted towards the Si-SiO2 interface under the gate,
forming a conductive channel between source (S) and drain (D). In these conditions,
if a voltage is applied Vds > 0, a current will flow between drain and source
controlled by the voltage Vgs between gate and source, that controls the formation
of the channel. The electrical characteristics of the MOSFET depend on L (gate
length) and W (gate width), as well as the technological parameters such as oxide
thickness and doping of the body. Typical values of L and W are: L D 0:1  2 pm,
W D 0:5  500 m. The range of the gate oxide thickness is 3–50 nm. There are
four types of MOS transistors: the N-channel MOSFET (nMOS) is formed on a
p-type substrate: nMOS enrichment (enhancement) or normally OFF and nMOS
depletion (exhaustion) or normally on. The p-channel MOSFET (pMOS), instead,
130 9 Low-Power Circuits

Fig. 9.29 gm
Transconductance gm

VGS
VT

is constructed on an n-type substrate: pMOS enrichment (enhancement) or normally

off and pMOS depletion (exhaustion) or normally on.
The operation of the MOSFET Enhancement, or e-MOSFET, can be described
using its characteristic curves described above. When the input voltage (Vin / (gate)
is zero, the MOSFET does not conduct and the output voltage is equal to the supply
voltage. In this way, the MOSFET is “fully-OFF” and is in its cutoff region. When
input voltage is HIGH or equal to supply voltage (Vdd /, the operating point of the
MOSFET moves to the point A along the load line. The drain current ID increases
to its maximum value due to reduction of the channel resistance, and becomes
a constant value independent from Vdd and depending only on VGS . Therefore,
the device behaves as a closed switch with a minimum channel resistance (Rds ).
Similarly, when VIN is low or reduced to zero, the operating point of the MOSFET
moves from point A to point B along the load line. The channel resistance is very
high so that the transistor behaves like an open circuit and no current flows through
the channel. If the gate voltage of the MOSFET alternates between two values, high
and low, the MOSFET behaves like a sensor “single-throw single-pole” (SPST)
in the solid state. Referring to the Id  Vd characteristic of a MOSFET is also
possible to define a set of parameters. While the transconductance gm (Fig. 9.29)
gives the variations of the current Ids (output) as a function of the voltage Vgs ,
there is another fundamental parameter that takes the name of output conductance.
It expresses the variation of the current Ids as a function of voltage Vds . All two
parameters are used to characterize the MOSFET in the amplifier configuration.
Other parameters that characterize a MOSFET are the following: Rds (on), the
minimum resistance in conduction mode that can vary from a few ohms to a few
milliohms; Vgs (th) or VT , the voltage applied to the gate to conduct the Mosfet, is
usually greater than 4 V; QGD , the minimum amount of energy necessary to the gate
to switch-on the MOSFET; td (on) is the time taken to charge the input capacitance
of the device before drain current conduction can start; td (off) is the time taken to
discharge the capacitance after the after is switched off [33–35].
9.3 Low-Power MOSFET 131

9.3.2 Mosfet Power Control

By reason of the very high gate resistance (input), its high switching speed and
ease of manage, the MOSFET is ideal for layout with operational amplifiers or
standard logic gates. In this case, the input voltage of the gate-source should be
chosen properly, the device must have a low value of Rds (on) in proportion to
the input voltage. The power MOSFETs can be used to control the movement of
the DC motors or stepper motors brushless directly from the logic computer or by
using the PWM modulation. As a DC motor offers high starting torque proportional
to the current, the MOSFET in PWM mode can be used as a speed regulator which
provides a smoother and quieter operation of the engine. Since the engine load is
inductive, a simple diode is connected across the load to dissipate any electromotive
force generated by the motor when the MOSFET is in the “OFF” state. A network
locking formed by a zener diode in series with the diode can also be used to permit
faster switching and an improved control of the peak inverse voltage and the drop-
out time (Fig. 9.30).

9.3.3 Stage of Amplification

The main stages of amplifiers using Mosfet can be described in three configurations:
common source, common drain, and common gate. In a common-source amplifier,
for example, the input signal is the signal applied to the gate, extracted in
output from the drain as current or voltage with respect to ground. The stage

+VDD

Motor M

D ID
+5v

RG D1
G
TTL
S
R2

Fig. 9.30 Example of layout for power control

132 9 Low-Power Circuits

common-source has input resistance very high and aids in the functioning as
transconductor and fits perfectly also as a voltage amplifier. This simply means that
by reason of high output resistance, the voltage gain (the voltage ratio of the output
signal and the input level) depends on the load resistance. This dependence on the
load resistance limits its usefulness as a voltage generic amplifier. In a common
drain or source amplifier is applied the input signal with respect to ground on the
gate terminal of the MOSFET. The main application is related to the buffer, because
its input resistance is extremely high, while its output resistance is reasonably low.
Unlike a voltage buffer ideal, a source follower MOSFET (common-drain amplifier)
provides a gain that is always less than one. Although the common source is
capable of substantial power gains of the signal, the voltage gain limits its utility
in the application of small signal. The input port of a common gate amplifier has a
relatively small input resistance and applies the input signal in current. The output
response to the input current applied is significantly extracted as a current signal and
the gain is always less than unity. The common gate amplifier can be represented as
the dual tracker source or source follower.

9.3.4 Common Source

A common-source amplifier is a topology typically used as a voltage amplifier

or transconductance. As transconductance amplifier, the input voltage is seen as
a modulation of the current to the load. As a voltage amplifier, instead, the
input voltage modulates the amount of current that flows through the MOSFET.
However, the output resistance of the device is not high enough for a reasonable
transconductance amplifier (ideally infinite), or low enough for a discrete voltage
amplifier (ideally zero). Another serious drawback is the limited high frequency
response characteristic of the amplifier. Therefore, the output is often routed
through both a voltage follower (common-drain stage, CD ), or a current follower
(common gate, CG ) to obtain the output characteristics and frequency response more
favorable. The combination CS  CG is called a cascode amplifier. The bandwidth
of the common source amplifier tends to be low, because of the high capacitance
due to Miller effect. The limitation on bandwidth in this circuit derives mainly from
the coupling of the parasitic capacitance Cgd between gate and drain and the series
resistance RA .

9.4 Analog Circuits

An ideal amplifier is an electronic circuit which generates an output signal y.t/ given
by the product of the signal input x.t/ and a constant A. Since both x.t/ and y.t/ are
electrical quantities (voltages or currents), an amplifier can be considered a two-port
device: to the input port of a voltage or current signal is applied (possibly without
9.4 Analog Circuits 133

Output variables

Voltage Current

Voltage Amplifier Transconductance Amplifier

iOUT
+
V(V) vIN AvvIN vOUT vIN gmvIN

Input vOUT = AvvIN iOUT = gmvIN

variables

Transresistance Amplifier Current Amplifier

iOUT
+
I(A) iIN RmiIN vOUT iIN AiiIN

VOUT = RmiIN iOUT = AiiIN

Fig. 9.31 Types of amplifiers

disrupting the circuit that has generated such a signal), while the output port, the
amplifier forces a voltage or a current proportional to the input signal by a constant
amplification. By depending on the nature (voltage or current), the signals of these
amplifiers are classified into four categories as shown in Fig. 9.31.
Contrary to the ideals amplifiers, real circuits have input and finite output
resistance. This means that the input port of the actual voltage and transconductance
amplifiers is not designed as an open circuit and that the input port of the
transresistance amplifiers and current is not a short circuit. The actual amplifiers
work correctly only if the generated output signal (voltage or current) takes a value
within a specific range known as the amplifier output dynamic. It is also noted
that even in the case in which the signal applied to the amplifier input is included
inside its input dynamic range, the signal generated at the output port of a real
amplifier may not be able to drive properly the connected load by providing to
a further limitation of the output dynamics. This limitation does not only depend
on the desired output signal, but mainly by the value of the load. In addition to
the limitation of bandwidth, the actual amplifiers are affected by a limitation that
concerns the maximum slew rate, namely the maximum value of the time derivative
of the output voltage. The slew rate limitation is manifested when the output of an
amplifier is no longer able to chase the correct output voltage because it changes too
quickly. The electronic amplifiers based on semiconductor devices (transistors) are
heavily affected by non-idealities and their parameters are poorly controlled due to
134 9 Low-Power Circuits

x + ε y
A
–
βy

Fig. 9.32 Negative feedback

manufacturing tolerances. To get amplifiers with a behavior comes close to ideality

the principle of negative feedback is used. The principle of negative feedback can
be described by referring to the block diagram of Fig 9.32.
In this diagram, an amplifier (characterized by a high gain) is used to amplify
the difference between signals, that is, the error between the desired signal and the
output signal, set back from the entrance to block ˇ (a passive block, for example,
a voltage divider). Since we wanted to a negative feedback, the correction should
vary the output in such a way as to reduce the error. For this purpose, it is important
that the sign of the proportional term in output that is reported back at the entrance
is negative. By referring to the block diagram of Fig. 9.32, the output signal y can
be expressed as follows:

y D A.x  ˇy/ (9.19)

A 1 Aˇ
yD xD x (9.20)
1 C Aˇ ˇ 1 C Aˇ

It may also observe that the error signal

, at the entrance of the block A, is given
by the following equation:

1

D x  ˇy D x (9.21)
1 C Aˇ

9.5 Operational Amplifier

The operating principle of negative feedback can be used advantageously to design

highly accurate amplifiers. To be able to reconstruct the block diagram of Fig. 9.32,
first of all, is needed a differential amplifier, i.e., a block that amplifies the difference
between two input quantities (typically voltage) so as to realize the different block.
In addition it requires an amplifier with a high (it is not necessary, however, to
be accurate) differential amplification (A), so that the gain of Aˇ is much greater
than 1. Differential amplifiers are one of the most important analog circuits known
as operational amplifier or, more simply, as the op-amp (Fig. 9.33).
9.6 Power Supply and Rejection 135

Output variables

Voltage Current

Voltage amplifier Transconductance amplifier

+ +

+ – + –
V(V) vIN vIN
vOUT
R2 iOUT
R1 R

Input R2 1
variables vOUT = 1 + v iOUT = v
R1 IN R IN

Transresistance amplifier Current amplifier

+
+ iOUT
–
–
I(A) R2
vOUT
iIN R
iIN R1

R2
vOUT = RiIN iOUT = 1 + i
R1 IN

Fig. 9.33 Circuit diagrams with operational amplifiers

A simple and important circuit based on the use of op-amps is shown in Fig 9.34.
Such a circuit is known as a voltage follower, voltage buffer, or voltage follower: that
is, an ideal voltage amplifier with amplification equal to one. The voltage follower
importance is linked to the concept of load. If you want to generate a specific voltage
across a load connected to a port of a circuit, then to connect the load as shown at
the top of Fig. 9.35 is typically not possible, because it is the load introduces a
disturb effect on the voltage. However, by connecting a voltage follower with the
operational amplifier between the source and the load as shown in the lower part of
Fig. 9.35, the voltage source is not disrupted because the absorbed current from the
non-inverting op-amp is nothing with output voltage Vout D VIN .

9.6 Power Supply and Rejection

In the last paragraphs we have described the operational amplifiers like any other
electronic circuit, it requires a supply voltage continuously in order to work.
Although it is permissible to assume that these voltages are constant, often they are
actually disturbed with residual oscillations which should not influence the output
voltage of an op-amp, however, the output voltage of an actual operational amplifier
in practice shows an error proportional to the fluctuations of the supply voltage.
136 9 Low-Power Circuits

+ –
vIN vOUT

Fig. 9.34 Voltage follower

vo vL RL

vL π vO

+
iL

vo
–

vL RL

vL = vO

Fig. 9.35 Impedance mismatch with the voltage follower

The performance of an operational amplifier is often expressed in terms of rejection

of the supply voltage ratio (power supply rejection ratio), PSRR, defined as:

A
PSRR D (9.22)
Aps
9.7 Low Noise Pre-amplifiers 137

Vout
AD (9.23)
Vd
Vout
Aps D (9.24)
Vps

Since both A and Aps depend on the frequency, the PSRR of an operational
amplifier depends on the frequency. At low frequencies, PSRR can have values
greater than 120 dB, whereas at higher frequencies (order of 100 kHz) the PSRR
could be about 30–40 dB.

9.7 Low Noise Pre-amplifiers

Low noise amplifier (LNA) is an electronic amplifier used to amplify possibly very
weak signals. A pre-amplifier is an electronic device which amplifies an analog
signal. Generally is the stage which anticipates the high power amplifier.
In this example we design a simple pre-amplifier with background noise of
the order of 0.8 nV/Hz @ 1 kHz. The circuit diagram is shown in Fig. 9.36. The
main components used are LT1128 dell Linear Tech and the JFET IF3602 of
InterfetCorporation. LT1128 (Figs. 9.37 and 9.38) is an operational amplifier ultra
noise at high speed. Main characteristics are the following:
• Noise voltage: 0.85 nV/Hz @ 1 kHz
• Bandwidth: 13 MHz
• Slew rate: 5 V/uS
• Offset voltage: 40 uV.
The IF3602, instead, is a Dual-N JFET used as stage for input of the operational
amplifier.

R20

100k
0
V4
1Vac +
3 VDD
0Vdc –
R22
1 500
2
0 VDD
Q19 ee 7
R17 U1
5 R1 3
R19 5 + 6
–
600 2 V
C1
R18 LT1128
100n
5 4 VEE

5 R23 V2
6 Q20
–
+

VEE 500 VDD

gV
EE V3
7
VEE
–
+

VDD gV
R24
R2
5 0
500

Fig. 9.36 Low noise pre-amplifier

138 9 Low-Power Circuits

Fig. 9.37 Voltage noise LT1128

TOP VIEW ORDER PART ORDER PART

VO5 TRIM
NUMBER TOP VIEW NUMBER
8 VO5
1 VO5
VO5 TRIM 1 7 V+ TRIM 8
TRIM
– –IN 2 – 7 V+
LT1028CS8
–IN 2 6 OUT LT1028AMH LT1128CS8
+ +IN 3 + 6 OUT
LT1028MH
5 OVER- OVER-
+IN 3 LT1028ACH V– 4 5
COMP
4 COMP
LT1028CH S8 PART MARKING
V– S8 PACKAGE
(CASE) 8-LEAD PLASTIC SOL

H PACKAGE 1028
8-LEAD TO-5 METAL CAN 1128

TOP VIEW ORDER PART

LT1028AMJ8 NC 1 16 NC
NUMBER
TOP VIEW LT1028MJ8
NC 2 15 NC
VO5
1
VO5
8 TRIM
LT1028ACJ8
TRIM TRIM 3 14 TRIM
LT1028CJ8 LT1028CS16
–N 2 – 7 V–
LT1028ACN8 –IN 4 – 13 V+
+N 3 + 6 OUT
LT1028CN8 +IN 5 + 12 OUT
V– 4 5 OVER- LT1128AMJ8 V– 6 11 OVER-
COMP COMP
LT1128MJ8 NC 7 10 NC
J8 PACKAGE
8-LEAD CERAMIC DIP LT1128CJ8 9
NC 8 NC
NB PACKAGE LT1128ACN8
8-LEAD PLASTIC DIP
LT1128CN8 S PACKAGE
16-LEAD PLASTIC SOL

NOTE: THIS DEVICE IS NOT RECOM-

MENDED FOR NEW DESIGNS

Fig. 9.38 Pin-outs of LT1128

The op-amp is one type of differential amplifier (Fig. 9.39). The inputs of the dif-
ferential amplifiers consist of a VC and a V input, and ideally the op-amp amplifies
only the difference in voltage between the two, which is called the differential input
voltage. The operational amplifier can be realized with bipolar junction transistor
(BJT, as in the case of the LT1128) or MOSFET, which works at higher frequencies,
with an input impedance higher and a lower energy consumption. The differential
structure is used in those applications where it is necessary to eliminate the
undesired common components to the two inputs. In this way, in output are
eliminated eventual DC components on the input signal such as, the thermal drift.
References 139

Fig. 9.39 The internal structure of the LT1128

The noises of internal current and voltage of the amplifiers depend on the intrinsic
physical phenomena and are by their nature random, aperiodic, and uncorrelated.
Typically have a distribution of amplitude of Gaussian. The relationship between
the peak to peak value and the effective value of these components is statistical.
A possible qualitative rule is that the RMS value multiplied by 6 does not exceed
the peak to peak in the 99.73 % of cases.

References

1. Park, J., & Mackqy, S. (2003). Practical data acquisition for instrumentation and system
control. Oxford: Elsevier.
2. Lacanette, K. (2003). National temperature sensors handbook. Handbook Ann. Mat. National
semiconcductor.
3. National Instruments (1996). Data acquisition fundamentals, Application Note 007.
4. National Instruments (1996). Signal conditioning fundamentals for PC-based data acquisition
systems, Handbook National Instruments.
140 9 Low-Power Circuits

5. Taylor, J. (1986). Computer-based data acquisition system. Research Triangle Park: Instrument
Society of America.
6. Di Paolo Emilio, M. (2013). Data acquisition system, from fundamentals to applied design.
New York: Springer.
7. Roundy, S., Wright, P., & Pister, K. (2002). Micro-electrostatic vibration-to- electricity
converters. In Proceedings of ASME International Mechanical Engineering Congress and
Exposition IMECE2002 (Vol. 220, pp. 17–22).
8. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: Self- sufficient energy
supply for micro systems. In Proceedings of the 16th International Conference on Thermo-
Electrics (pp. 575–577).
9. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. IEEE
Micro, 21(3), 30–42.
10. Roundy, S. (2003). Energy scavenging for wireless sensor nodes with a focus on vibration to
electricity conversion. PhD thesis, University of California.
11. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006, November).
Micro seismic electret generator for energy harvesting. In Technical Digest PowerMEMS 2006,
Berkeley, USA (pp. 133–136).
12. Sterken, T., Altena, G., Fiorini, P., & Puers, R. (2007, April). Characterisation of an
electrostatic vibration harvester, DTIP of MEMS and MOEMS, Stresa, Italy.
13. Sterken, T., Baert, K., Puers, R., & Borghs, S. (2002, November). Power extraction from ambi-
ent vibration. In Proc. SeSens. Workshop on semiconductor sensors, Veldhoven, Netherlands
(pp. 680–683).
14. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Systems, 27(2), 803–815. ISSN: 0885-8993.
15. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning
a resonant energy harvester using a generalized electrical load. Smart Materials and Structure,
19, 055003.
16. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
17. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
18. Szarka, G. D., Burrow, S. G., & Stark, B. H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
19. Maurath, D., Becker, P. F., Spreeman, D., Manoli, Y. (2012). Efficient energy harvesting with
electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6), 1369–1380
20. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
21. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies–state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
22. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. In Proceedings of the World Congress on Engineering (Vol. 2).
23. Zhu, D., Tudor, M. J., & Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
24. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy harvesters
– a classification approach. Micromachines, 4(2), 168–196. http://www.mdpi.com/2072-666X/
4/2/168. Accessed 20 Jan 2015.
25. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of applied electromagnetics
(6th ed.). Upper Saddle River: Prentice Hall.
References 141

26. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
27. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008). A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
30. Razavi, B. (2002). Design of analog CMOS integrated circuits. New York: McGraw-Hill.
31. Razavi, B. (2008). Fundamentals of microelectronics. Chichester: Wiley.
32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University Press.
33. Razavi, B. (2002). Design of integrated circuits for optical communications. New York:
McGraw-Hill.
34. Di Paolo Emilio, M. (2015). Microelectronics from fundamentals to applied design. Cham:
Springer.
35. Di Paolo Emilio, M. (2014). Embedded system design for high speed data acquisition and
control system. Cham: Springer.
Chapter 10
Low-Power Solutions for Biomedical/Mobile
Devices

10.1 Introduction

Sensor networks have internal wireless modules solutions for the transmission
of the information to a remote computer and then a smartphone. The sensors
with a wireless connection can be supplied autonomously with energy harvesting
techniques, by using further battery to extend the life of the system. The worst case
scenario in wearable devices is the inability to take advantage of the mechanical
energy in patients with immobility in bed for various health reasons, where wearable
devices must rely on battery-powered solutions. There is no mechanical energy, the
intensity of light is too low to power a low-power device. Therefore only a part of
the head and sometimes the area of the wrists represent the only areas relatively
suitable for the harvesting of energy: thermal and light. The available low power is
connected to the level of internal illumination and the heat transfer from the person
that is considerably low, the latter determined by the natural convection around the
head. On one side there is provided a device that generates electricity from the
internal temperature difference of the human body, create the thermoelectric micro-
generators (TEG). They are able to generate a few microwatts of electrical power.
A BiCMOS device is made using standard materials to maintain the reduced cost.
The low thermal conductivity of these materials appears to be the main factor to
increase the output power. The materials used are the poly-Si and poly-SiGe. On the
other side, there is a piezoelectric device that converts the kinetic energy of a mass
free to move inside of a rigid frame into electrical energy due to the impact of this
mass with the two ends of the frame. The real-time monitoring of the patient’s vital
signs requires the use of wearable sensors and mobile devices [1–15].

© Springer International Publishing AG 2017 143

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_10
144 10 Low-Power Solutions for Biomedical/Mobile Devices

10.2 Design of Wearable Devices

The goal is to obtain an efficient energy harvesting by reducing the size of

the battery. The classical example is typical of pacemakers used in humans. If the
harvesting of vibrational energy in this case is realized, it is possible to reduce the
size of the pacemakers and to extend the life of the battery. They use internal sensors
to monitor the status of a patient, or are implanted to save life. Currently, the size of
a typical pacemaker is about 42 mm–51 mm–6 mm. Typically, the battery requires
about 60 % of the size of the pacemaker. The goal is to reduce the area and then set
the maximum size of the energy harvesting techniques such as 25 mm–25 mm–5 mm
with power consumption about 10 uW. Fortunately, the packaging of traditional
pacemaker eliminates the need to use biocompatible materials. The batteries and
the pacemaker circuits are encapsulated in a titanium case: a biocompatible material
with a sealed casing which ensures that there is no contact between the interior of
the body and the pacemaker batteries or circuits.
The humans constantly produce heat as a side effect of the metabolism. However,
only a part of this heat is dissipated in the environment as a flow of heat and infrared
radiation, the remainder is rejected in the form of water vapor. Furthermore, only
a small fraction of the heat flow can be used in collection of energy. The heat flux
can be converted into electrical energy using a thermoelectric generator (TEG). The
body has high thermal resistance; therefore, the heat flow is rather limited (Figs. 10.1
and 10.2). The main problem of these devices is the duration of the battery, even
though they have much longevity, the battery must be replaced: the patient must go
to hospital, he undergoes a surgical intervening, with possible complications, with a
few days of recovery, etc. Here comes the energy harvesting to create a device that is
fueled by the energy produced by the human body. In practice this would eliminate
the need an intervention for replacement of the battery. The objective can be to
develop a micro-generator TEG that is cheap, small, and low-power consumption,

a b
0.04 20
V (m/s) 15 a (m/s^2)
0.02
10

0 5
0
–0.02 –5
–10
–0.04
–15
–0.06 –20
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
t(s) t(s)

Fig. 10.1 Acceleration profile and heart rate speed [response]

10.2 Design of Wearable Devices 145

Fig. 10.2 Energy harvesting

for a pacemaker

as, for example, materials such as poly-Si and poly-SiGe, in BiCMOS configuration
(Bipolar Complementary Metal Oxide Semiconductor), which is a technology that
integrates CMOS and BJT on the same semiconductor chip. The advantage of this
procedure can be described in the following points:
• Two different technologies.
• Overall capacity of a BiCMOS gate is almost equal to that of the only BJT.
• The possibility to combine in the same integrated analog and digital electronics.
A TEG generator consists of n-type and p-type thermoelectric material. These
are electrically connected in series by means of metal bridges and are arranged to
use the best available area. The generator layer is positioned between the silicon
substrate and the heat stabilizer. A thermal gradient between the lower end of the
device and the environment at the upper end is creating a vertical heat flow, which
can be partially converted into electrical energy by the effects of thermoelectricity
[16–20]. Research into Body Area Network (BAN) and their applications are
experiencing growing interest, especially in monitoring the status of the human body
health, monitor their movements and activities. Wearable wireless sensors make
this possible, by collecting information on the individual or single community, to
anticipate their requests before they are expressed. Another possible application is
under Ambient-Assisted Living (AAL): by collecting physiological information in
real time to prevent hazardous events such as heart attack.
146 10 Low-Power Solutions for Biomedical/Mobile Devices

10.3 RF Solutions for Mobile

The radio frequency energy is nowadays a very interesting source for the power
supply in wireless sensor networks; through the wireless power distribution systems
it is possible to obtain an infrastructure ad hoc able to power networks of hundreds
or thousands nodes, with a single source of transmission. Today, the real challenge in
the research and development of these technologies is to recover power transmitted
from public telecommunications services, such as broadcast TV and radio or mobile
telephone communications.
The EH radio wave system is constituted by a system commonly called rectenna
composed of the receiving antenna, the matching network, and the rectifier circuit,
and a power management system that includes the dc-dc converter, the accumulation
system (supercapacitors), and the load (consumer). One drawback to the sensor
nodes is represented by the element of energy reserve, generally constituted by bat-
teries, because the maintenance for the replacement of batteries with a large number
of nodes is very expensive and also to aggravate the maintenance costs. Therefore,
there is need to make autonomous from an energetic point of view, the many sensor
nodes located in the various environments. The solution proposed to lengthen the
lifetime of the batteries or completely prevent its use is based precisely on the
recovery of energy from environmental sources, particularly from the airwaves.
The benefits to these technologies can indirectly arise from recent developments
in micro- and nano-technologies, which have led to the creation of ever-smaller
devices and with energy consumption which are reduced accordingly. The energy
harvesting system compared to traditional power systems is the great saving in
implementation and operating costs The rectenna constitutes the transducer of an
EH radio frequency system, and is therefore capable of converting the incident RF
signal on the antenna into a DC signal to be amplified and accumulated in the power
management circuit. The antenna is the transmitter of the radio frequency in the
EH system. When an electric field impacts with a certain intensity on it, an induced
voltage that is generated must be rectified through a rectifier circuit [21–30].

10.3.1 Ferrite Rod Antenna

As shown in Fig. 10.3, it is a ferrite core with a number of turns wound around
it. The purpose is to obtain a sort of small loop antenna (magnetic dipole) around
the ferrite core. This antenna, together with electronics that there is downstream,
can directly replace the batteries in specific applications, for example, for WSNs. It
constitutes a good solution to the low frequencies, in particular, for medium-wave
transmissions.
Figure 10.4 shows the antenna equivalent circuit also described in resonance
condition; RR is the radiation resistance of the loop, that is:
10.3 RF Solutions for Mobile 147

Fig. 10.3 Ferrite rod antenna

Lf RR R Rf Q2RR Q2R Q2Rf

C
Equivalent Circuit At Resonance
Fig. 10.4 Equivalent circuit of the ferrite rod antenna

 2
e nA
RR D 31; 200  (10.1)
2

where n is the number of turns, A is the loop section, e is the relative magnetic
permeability of the ferrite. Rf is the loss resistance of the rod, that is:

00 Af
Rf D 2f e 0
0 n2 (10.2)
 lf

where f is the operating frequency, 0 and 00 are the real and imaginary parts of
the magnetic permeability of the ferrite, respectively, Af is the cross sectional area
of the rod, and lf the length of the rod. The inductance of this antenna is instead
given by:

Af
Lf D e 0 n2 (10.3)
lf

By using the capacitor C in parallel is possible to convert the antenna impedance,

at the resonance frequency f0 , in a pure resistance whose value is greater of the
current loop resistance and Q is the quality factor defined as follows:

f0
QD (10.4)
f

where f is the bandwidth at 3 dB.

148 10 Low-Power Solutions for Biomedical/Mobile Devices

Fig. 10.5 Circular spiral inductor antenna

Fig. 10.6 Equivalent circuit of the circular spiral inductor antenna

10.3.2 Circular Spiral Inductor Antenna

The circular spiral antenna (Fig. 10.5) ensures a high power density and this is one
of the most used for the recovery of energy from radio waves. This antenna may be
supplied by a microstrip line and in this case it has the equivalent circuit of Fig. 10.6.
The total magnetic energy stored in the circular spiral is expressed through an
inductance defined as:
10.3 RF Solutions for Mobile 149

C1 C2 D1

L1 D2 C3 RL
R1

Fig. 10.7 Equivalent circuit of the folded dipole

   
N 2 Dm 2; 46 2
LD ln C 0; 2. / (10.5)
2 

where  is the magnetic permeability of the material,  is the fill-ratio, N the number
of windings, and Dm the mean diameter of the inductor.

10.3.3 Folded Dipole

The folded dipole, with respect to a common dipole =2, allows to have a radiation
resistance (approximately 300  for a folded dipole two-wire), a voltage Vout , and
a higher operating band. It is usually coupled with a voltage multiplier rectifier to
further increase the value of the dc voltage available.
The power density is the most important parameter for the RF energy harvesting;
the expected level of the received power depends on level of transmitted power and
the distance from the transmitter, while it appears to be independent of frequency
(Fig. 10.7).

10.3.4 Microstrip Antenna

The microstrip antenna (patch antenna) is the common antenna used for modern
cell (Figs. 10.8 and 10.9); it works well at high frequencies but has narrow band, so
that even small frequency deviations may cause large losses of the recovered power.
As the rectifier circuit can be used a single Schottky diode which allows to have a
higher conversion efficiency [31].
There are various configurations of microstrip antennas (L-shape, E-shape,
and U-shape), all of narrow-band type (narrow band). In Fig. 10.9 is shown the
equivalent circuit of a microstrip antenna composed of two inductors and a shunt
capacitance.
150 10 Low-Power Solutions for Biomedical/Mobile Devices

Fig. 10.8 Microstrip antenna

Fig. 10.9 Equivalent circuit L1 L2

of microstrip antenna
1 2 1 2

Cs

T T

10.4 Power Management

The sector of the energy harvesting recently had a growing interest, and there-
fore many studies have been directed towards the design of integrated circuits
for converters. In particular, the company of “Linear Technology” has recently
produced a wide range of ultra-low power integrated circuits for applications of
EH. These circuits may primarily to convert vibrational energy (piezo), solar and
thermal (thermopile, thermocouple) and have a high conversion efficiency. Some
of these may also be exploited for the radio wave energy conversion. These ICs
are particularly relevant for applications in industrial automation, transportation,
automotive, and wireless sensors. The LTC3108 of Linear Technology is a boost
converter with high efficiency which can operate with very low input voltages,
ranging from 20 mV to 0.5 V, and therefore can be used for energy harvesting system
in radio frequency.
The integrated LTC3108 can also work as AC Energy Harvester due to internal
rectifier circuit; in such case, it is possible to connect the source directly to the
circuit. The expected pattern is the following (Fig. 10.10), in which the input is
10.5 Ultra-Low Power 2.4 GHz RF Energy Harvesting and Storage System 151

Fig. 10.10 AC energy Rs C1

harvester C1 Vstore
Vs 50 1n C4
+
10m
– C2 Vout2_EN

U1
OUT
SW Vout
SC
50
VS1 Vout2
LTC3108

VS2 VLDO
C5

2.2μ
Vaux PGD
C3 GND
1μ

an antenna designed with a sinusoidal generator Vs D 225 mV at 102 MHz and a

resistance Rs in series from 50 ; C2 and SW pins are shorted to ground because it
is not expected the presence of the transformer.
The LTC3105 of Linear Technology is a high efficiency boost converter that
can operate with input voltages ranging from 225 mV to 5 V. It has integrated a
maximum power point controller (MPPC), which maximizes the energy that can be
extracted from any source, and a voltage regulator dissipative Low Drop Out (LDO).

10.5 Ultra-Low Power 2.4 GHz RF Energy Harvesting

and Storage System

An RF energy harvesting system optimizes energy management for many devices.

The circuit described in this paragraph is optimized for the power range typically
observed in the ISM 2.4 GHz band in order to collect the energy supplied by the
nearby Wi-Fi, Bluetooth, and other devices. In this layout, the rectified voltages
are low, then the operation of the power management circuit in 100 mV scheme
is critical. The RF energy harvesting system proposed can be thought in terms of
four main blocks, as shown in Fig. 10.11. A key feature is the boost converter that
represents the start to ultra-low input voltages, while maintaining high efficiency
to useful output voltages. Finally, we need a reserve of energy like a battery, large
capacitor, or supercapacitor to store energy from the RF input sources until sufficient
energy is available for a sensing operation [32–35].
The antenna is one of the most important components of an RF energy harvesting
system. It collects ambient RF power and transmits it to the rectifier; so it must
be designed to capture more energy. In Fig. 10.12 an example of a patch antenna
152 10 Low-Power Solutions for Biomedical/Mobile Devices

Fig. 10.11 Block diagram of the circuit ultra-low power 2.4 GHz

X0

Lt
y0

Ws W

Ls

Wg

Fig. 10.12 Antenna

consists of a U-groove structure. The antenna is printed on a substrate with a

thickness of 6 mm and a dielectric constant of 10.2. The radiating element has
dimensions of L D 0:155 D 19 mm  W D 0:118 D14.5 mm. Ls D10.15 mm,
Lt D1 mm, Wg D1 mm, x0 D L=2, Y0 D W=2. The dimensions of the ground plane
are 0:326  0:326 (40 mm  40 mm).
The rectifier is a single phase voltage doubler that uses the detector diodes Avago
HSMS-286C with the microstrip RF matching networks on the substrate. A typical
DC-DC boost converter connected an input capacitor through an inductor in an
output capacitor under the control of a transistor. High efficiency is possible with
a low-loss transistor with the inductor size and switching frequency of the transistor
carefully selected on the basis of the energy input (Fig. 10.13).
An NiMH battery can be used as energy storage as well as the supercapacitors. In
Fig. 10.14 an example of a charging circuit. Two delay stages are used, controlled by
U1 and U2 MOSFETs, to ensure sufficient stored energy from the boost converter
before the gating in the battery. The power switches for U1 and U2 are Q1 and Q2,
respectively. The diode D5 prevents any return trip of energy from the battery in the
charging circuit when the charger is inactive.
References 153

T1
0 2 2
1 2
Input SW1
330u 330u Output
1 1
0.1u

2p

Fig. 10.13 Boost configuration

Vbat HI

1M
1.6 v
battery
1M
Output dc-dc
150p 0.1u
10M
1nF

Fig. 10.14 Battery charging circuit

References

1. Park, J., & Mackqy, S. (2003). Practical data acquisition for instrumentation and system
control. Oxford: Elsevier.
2. Lacanette, K. (2003). National temperature sensors handbook. Handbook Ann. Mat. National
semiconcductor.
3. National Instruments (1996). Data Acquisition Fundamentals, Application Note 007.
4. National Instruments (1996). Signal Conditioning Fundamentals for PC-Based Data Acquisi-
tion Systems.
5. Taylor, J. (1986). Computer-based data acquisition system. Research Triangle Park: Instrument
Society of America.
6. Di Paolo Emilio, M. (2013). Data acquisition system, from fundamentals to applied design.
New York: Springer.
7. Roundy, S., Wright, P., & Pister, K. (2002). Micro-electrostatic vibration-to- electricity
converters. In Proceedings of ASME International Mechanical Engineering Congress and
Exposition IMECE2002 (Vol. 220, pp. 17–22).
8. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: Self- sufficient energy
supply for micro systems. In Proceedings of the 16th International Conference on Thermo-
Electrics (pp. 575–577).
9. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. IEEE
Micro, 21(3), 30–42.
10. Roundy, S. (2003). Energy scavenging for wireless sensor nodes with a focus on vibration to
electricity conversion. PhD thesis, University of California.
11. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006, November).
Micro seismic electret generator for energy harvesting. In Technical Digest PowerMEMS 2006,
Berkeley, USA (pp. 133–136).
12. Sterken, T., Altena, G., Fiorini, P., & Puers, R. (2007, April). Characterisation of an
electrostatic vibration harvester, DTIP of MEMS and MOEMS, Stresa, Italy.
154 10 Low-Power Solutions for Biomedical/Mobile Devices

13. Sterken, T., Baert, K., Puers, R., & Borghs, S. (2002, November). Power extraction from ambi-
ent vibration. In Proc. SeSens. Workshop on semiconductor sensors, Veldhoven, Netherlands
(pp. 680–683).
14. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Systems, 27(2), 803–815. ISSN: 0885-8993.
15. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning
a resonant energy harvester using a generalized electrical load. Smart Materials and Structure,
19, 055003.
16. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
17. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
18. Szarka, G. D., Burrow, S. G., & Stark, B. H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
19. Maurath, D., Becker, P. F., Spreeman, D., Manoli, Y. (2012). Efficient energy harvesting with
electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6), 1369–1380.
20. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
21. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies–state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
22. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. In Proceedings of the World Congress on Engineering (Vol. 2).
23. Zhu, D., Tudor, M. J., & Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
24. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy har-
vesters – a classification approach. Micromachines, 4(2), 168–196. http://www.mdpi.com/
2072-666X/4/2/168. Accessed 20 Jan 2015.
25. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of applied electromagnetics
(6th ed.). Upper Saddle River: Prentice Hall.
26. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
27. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008). A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
30. Razavi, B. (2002). Design of analog CMOS integrated circuits. New York: McGraw-Hill.
31. Razavi, B. (2008). Fundamentals of microelectronics. Chichester: Wiley.
32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University Press.
33. Gudan, K. (2015). Ultra-low power 2.4GHz RF energy harvesting and storage system with
25dBm sensitivity. San Diego: IEEE. ISBN 978-1-4799-1937-6.
34. Briand, D. (2015). Micro energy harvesting. Weinheim: Wiley.
35. Spies, P. (2015). Handbook of energy harvesting power supplies and applications. Boca Raton:
CRC Press.
Chapter 11
Applications of Energy Harvesting

11.1 Introduction

Energy harvesting makes use of ambient energy to power small electronic devices
such as wireless sensors, microcontrollers, and displays. Typical examples of these
environmental sources are sunlight and any artificial source such as vibration or
heat from engines or the human body. The energy transducers such as solar cells,
thermogenerators, and piezoelectric convert this energy into electrical energy. The
goal of each energy harvesting system is to replace the batteries used to power by
extending the charging intervals for the storage element. A first field of application is
the automation with self-powered switches. Further applications are the monitoring
systems for large industrial plants or structural monitoring of huge buildings.
Another promising market is the consumer area with purses, clothing which show
the energy transducers integrated in the form of solar cells or TEG or RF transmitters
to recharge consumer products such as mobile phones or audio players. A general
overview of energy estimated to be harvested in different application is summarized
in Tables 11.1 and 11.2, instead, compare the power consumption of different
microcontrollers state of the art in some modes of operation.

11.2 Building Automation

The building automation systems are networks of electronic devices for the control
of different functions. They manage lights, heaters, air conditioning systems, doors,
valves, and safety systems. The benefits of building automation have reduced energy
and maintenance costs, increased safety and comfort. Typical energy savings with
the aid of building automation systems are about 30 %. A typical reduction is
the wiring that may be reduced about 70 %. It also reduces operating costs, since
no energy is required for devices powered by energy harvesting. In addition, it

© Springer International Publishing AG 2017 155

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5_11
156 11 Applications of Energy Harvesting

Table 11.1 Power Applications Power requirement

consumption for some
devices Standy 10 nW
RFID tag 10 uW
GSM 1W
Bluetooth transceiver 10 mW
FW receiver 1 mW

Table 11.2 Activities of time and power consumption

Applications Switch-on time (ms) Power requirements (mW)
Radio 1 57
Microprocessor Permanent 1
Energy management Permanent 0.1
Radio + microprocessor, standby Permanent 0.003
Mobile phone Permanent 5

saves maintenance costs, because no batteries need to be replaced or recharged.

Wireless transceivers or transmitters and low-power sensors are the main devices
in the field of building automation devices powered from environmental sources.
Typical sensors used are those of temperature, humidity, and pressure. In addition,
light sensors, presence sensors, or motion are useful for the control of building
functions. Other application devices are sensors for controlling the position of valves
in the heating systems, ventilation, and air conditioning. The big advantage for all
these sensors is the slow variation of the physical parameters to be monitored. Thus,
the measurements and transmission of these data should only take place with a small
duty cycle, leading to a low overall power consumption of sensors and transmitters.
Wake-up times typical of such systems to make a reading of a sensor are 1, 10,
or 100 s. Most of the time, these sensors are in sleep mode. Therefore, even a
low energy consumption is obtained by reducing the amount of data sent or the
number of transmissions. The goal of optimization is always a minimum of energy
consumption to enable the energy harvesting with low cost hardware. The power
of transmission is typical, for example, 10 mW, the frequency band more or less at
868 MHz. In building automation, light as well as the movement and the heat can
be used to power electronic devices. In addition, the solar cells are used in buildings
in combination with a supercapacitor or a rechargeable battery back-up for night
operation. Sometimes, only a few hours of sunshine are sufficient to fully charge an
energy storage element and ensure an operation of a specific device for many hours
without illumination. As an additional safety feature, a signal can be transmitted
to report the correct functioning of the system. The typical brightness in homes is
between 100 and 500 lx. Most of the rules on health and safety work requiring a
minimum illumination of 500 lx in the workplace. The minimum illumination time
necessary to the day of the light will be about 30 % shorter with fluorescent light
of the same brightness. The thermal gradients are a further option to use in building
automation. Especially with radiators, air conditioning or hot water pipes. Due to the
11.4 Structural Health Monitoring 157

temperature around 20 ı C, the heat sinks must be used to maintain a thermal gradient
sufficient. When different temperature differences are present, the maximum power
point trackers (MPPT) are useful to match the power management for the maximum
power of the TEG. The vibrations can also be used in buildings to get electricity.
The problem is that only a small vibration in small frequencies are available, which
require large generators with large seismic masses. Typical 0.01 g and a frequency
range between 1 and 10 Hz are values available [1–20].

11.3 Environmental Monitoring

The idea of monitoring is to measure physical parameters such as the temperature.

The monitoring of the conditions can be used for all types of machines such as
motors, pumps, fans, or compressors to determine any faults. It employs different
types of sensors to measure operating parameters of the system. All these data
must be collected and analyzed by microcontrollers. Some monitoring systems of
the conditions provide an Ethernet connection to collect all the data and pass it
to a central computer. Most feasible are wireless sensor networks, which transmit
data by radio signals by eliminating the need of the installation of cables. The
galvanic isolation is required in some applications, as well as, for example, in high
power transmission systems. With the energy harvesting sensors, radio, actuators,
processors, and displays is possible to power without batteries or at least without the
need to recharge or replace them. Additionally, the installation in inaccessible areas,
remote or dangerous, and integration into machines becomes possible. In addition to
the transmission of measured values such as acceleration and temperature, directly
displays the sensors can be applied to allow monitoring of the parameters of interest
in the position of occurrence (Fig. 11.1).

11.4 Structural Health Monitoring

The structural health monitoring (SHM) is the discovery process of damage in

components and construction. The goal of SHM is to improve the safety and
reliability of aerospace, civil and mechanical infrastructure to predict and detect
damage before they reach a critical state. Especially extreme events such as earth-
quakes and typhoons, heavy snow and storm or simply the aging of materials, and
environmental degradation cause serious doubts on the integrity of any facilities.
Structural monitoring (SHM) is the characterization process of existing structures
to identify certain properties of the structure, in order to have information about the
actual structure in which power to develop analytical models for assessment of the
state of the structure or to assess changes in the structural behavior. The continued
operation of civil structures and infrastructures such as bridges and buildings
requires careful strategy to maintain an efficient structure and with a high level of
158 11 Applications of Energy Harvesting

Sensor
Node

Wireless
Sensor
Node
Server
Base station
Sensor
Node
Database

Sensor
Node

Fig. 11.1 WSN architecture

security but, at the same time, limiting maintenance costs. The problem becomes
even more relevant when necessary in the evaluation of the performance level of
potentially damaged by traumatic events structures, such as the seismic events.
At the same time, safeguarding the immense architectural and cultural heritage
represented by historic centers of many Italian cities, it requires careful monitoring
of his state of health and the structural effects of the increasing level of vehicular
traffic. The structures able to detect the operating temperatures, pressures which
could reduce the weight and costs of composite materials. The weak point would
connectors and cables for the passage of the sensor information to the outside of
the structure. Wireless transmissions with extremely miniaturized electronics could
solve this problem halfway, by leaving the question of power but with applicable
techniques related to RF and thermal. The large number of individual measuring
points and then large number of sensor nodes densely distributed in possibly random
configurations in the environment detection require the use of wireless sensors. The
cooperation between the sensor nodes is used as the local processing capability to
perform the merge data or other calculation functions.

11.5 Automotive

The application of energy harvesting techniques in the railway sector is a very

promising field. They were conducted studies and experiments on the potential of
energy recovery devices, with the aim to provide information on the electric power
actually generated by the use of harvester placed on board railway wagons. The
tire pressure monitoring (TPM) is a means to increase the safety and efficiency
of all types of vehicles by using tires inflated through monitoring the pressure.
11.6 Projects 159

Tire Pressure Monitoring systems were adopted in 1986 in a Porsche 959. The
technology has been used in top luxury range of vehicles such as the Audi A6,
Mercedes Benz S Class, and BMW7 Series to increase safety and the economy
maintenance. Indirect TPMS do not use pressure sensors. These systems measure
the air pressure in the wheel by monitoring individual wheel revolutions. Other
developments may also simultaneously detect under-inflation until all four tires
using vibration analysis of individual wheels or analysis of the effects of displace-
ment of the load during acceleration or cornering. These systems require additional
sensors suspension, which makes it more complex and expensive. To transfer the
pressure and other data from the wheel to the vehicle control unit uses radio
frequency (RF) communication or electromagnetic coupling. Due to the rotation
of the wheel, no wired communication may be employed. The power of the sensor
in the tire represents a challenge in direct TPM systems. Here batteries have the
same problems in many other applications, such as limited life, low temperature
range, weight, and additional costs. Due to the high amount of energy of vibration,
TPMS are a potential application field for the vibration energy harvesting. The small
required volume and low price facilitate the use of free self-powered systems. More
suitable for use in tires are piezoelectric and electrostatic principles, mainly because
of their small weight and size implementation related integration costs [21–30].

11.6 Projects

The system of Fig. 11.2 is designed to power a wireless sensor node that consumes
40 mW in active mode, with the average power of the load in according to the
average power generated. The coupling of the temporal power is achieved by
a supercapacitor and under-voltage lock-out which connects the load only when
sufficient power is available for a cycle of operation. The topology of the power
stage of the rectifier is shown in Fig. 11.3.
An advantage of the separation between start-up and main power circuits is that
the charge pump can quickly establish a provision to support the requirements of
the auxiliary circuitry, by allowing the rectifier to charge the super capacitor in
an optimal way. The thermoelectric generators have recently been used as power
supplies for strain gauges on planes. Strain gauges and the power supply are located
outside the cab, and then are subjected to strong temperature variations in the course
of a flight cycle. In order for the system to convert the greatest amount of latent heat
energy to the possible change of phase in electrical form material, the thermoelectric
generators are connected to a switching circuit capable of performing tracking the
maximum power point (Fig. 11.4). A critical aspect is the inversion of the polarity
between the ascent and descent of the aircraft, in this cases a rectifier is used in the
system. It may not be a passive rectifier because the voltage required to effectively
overcome the diode in the voltage state is not feasible for the TEG under these
thermal gradients. Therefore, the rectifier must be one active with MOSFET.
160 11 Applications of Energy Harvesting

Start-up circuit
Voltage
multipler HFET
(V)

Load circuit

Pgen Pin Pout Pstore Super Under- Pload

Sensor Boost rectifier Voltage Load
capacitor
lockout

Sensor

Pquies
Polarity Gate Linear
detection drive regulator

VDD = 1.5 V
Control circuit

Fig. 11.2 Block diagram of an energy harvesting circuit

OUT
Vin (t)

1 2

Load
+
–

PWM2
PWM1

Iin (t)

Fig. 11.3 Rectifier boost

11.7 Solar Infrastructure

Solar energy is derived from the electromagnetic waves produced by the sun. The
solar radiation has the advantage to be available almost fully used without the pro-
duction of polluting waste. The limits that may be encountered are discontinuities,
caused by the alternation of day and night and weather conditions, the other is the
low intensity which implies the need to have large areas of “energy storage.” There
are two different approaches for using solar energy. In the case of thermal utilization,
the collectors transform the sun rays into heat energy that can be used both for
producing hot water and to support the heating system. The sun’s rays are converted
into electrical energy using solar cells that are combined in modules within which a
number of cells are connected in series and parallel combinations in order to obtain
the voltage and the output current within the desired intervals. Also in this way
11.7 Solar Infrastructure 161

Sensor Vrect

Vteg

3
+
–

4 7
OUT

V– V+
6

Voltage Inverter

Fig. 11.4 TEG for airplane cabin

the power of a single module of practical utility reaches values (every single cell
supplies only a few watts). The photovoltaic panels are systems that use lenses to
concentrate the solar radiation on the cell, while still seeking to increase efficiency
and reduce costs. A photovoltaic system is called “island” when it does not connect
to any electrical service network and is therefore autonomous. Leverages on the
place of production the energy collected and then preserved by an accumulator. The
main components of a photovoltaic system “island” are generally PV array to gather
energy through photovoltaic modules arranged in favor of the sun. By depending on
the situation, the reference field is optimized to have a given voltage, usually either
12 or 24 V. Consequently, the majority of photovoltaic modules used in this type of
plant has output voltages equal to 12 or 24 V, the so-called electrical strings that form
the field are constituted of very few modules, up to the limit of the single module per
string. Photovoltaic systems “grid-connect” are connected to an existing electricity
distribution network and managed by third parties. Photovoltaic systems “hybrid”
remain connected to the electrical distribution network, but mainly use solar energy
and accumulator. Moreover, a control unit can be used by connecting the building
to the electricity grid for the supply [31–35].
162 11 Applications of Energy Harvesting

11.8 Wind Energy

The kinetic energy of wind is used to rotate a propeller or wheel, whose mechanical
energy is then converted into electrical energy. Its exploitation, relatively simple
and inexpensive, is implemented through wind machines divided into two distinct
groups depending on the basic module type used defined wind generator. The
kinetic wind energy utilization is very old: sailing and windmills constitute the most
significant examples. The currently most widely used wind turbines have become
integrated part of our landscape and the parts that make up a modern turbine are
essentially the following: (1) The machine blades are fixed on a hub and make up
in the rotor. The hub is connected to a rotating shaft that can be connected in turn,
by depending on the type of turbines, to a synchronous generator or to a multiplier
of revolutions and then to an asynchronous generator, currently most widespread
solution. (2) The power cables. The above components are located in the nacelle
positioned on a bearing support, so as to be easily adjustable according to the wind
direction. L’entire spacecraft is positioned on a tower, which can be lattice or conical
tubular. The most common system is to be varied the angle of incidence of the blade
to the wind through hydraulic or electrical systems. A control system continuously
checks the position of the blade and the power generated by changing continuously
the angle of incidence. This method represents a significant advantage over the
traditional system, which had a fixed angle of attack of the blades to the hub. The
advantage consists in having a higher range of functioning at different wind speeds
and lower vibrational stresses on the structure. The mini-wind turbine or even micro
wind has the potential to convert the wind’s kinetic energy into electrical energy,
usable for the operation of electrical appliances throughout the home. In a particular
way, for the installation of this type of wind power system, the low-power system
from 0 to 20 kw, it can be installed in stand-alone systems to isolated users, in the
best-known version of stand-alone, or in systems connected with the network, or the
best known grid-connected.

11.9 Conclusions

The energy harvesting is adopted in many fields of applications, from consumer

through automotive. Many of implementation scenarios differ with regard to the
properties of the environmental energy source and the needs of electrical users.
The economic benefits of harvesting became obvious when energy costs for battery
replacement or recharging should be considered as the main factor in the design.
Thus, energy harvesting is the preferred solution in applications where a single life
of the battery of a non-rechargeable battery or a charge cycle is not sufficient. In
applications with extreme conditions, it enables new solutions without the need to
compete with batteries. Many quantities of energy harvesting devices are currently
sold only in the building automation industry and in the consumer market. In the
References 163

consumer market solar cells represent the main choice, but more solutions in
terms of vibration and RF are entering the market. The rapidly expanding of
wearable devices is set to revolutionize the technology and current processes in
all areas, by creating new market opportunities and new business models. The
market will continue to evolve as several start-ups and established companies
have a strong interest in research and development, in particular for implementing
multiple functions on a single device. The development of IoT promises to be
exciting and have a great economic impact due to the development of semiconductor
devices, and to the advancement of wireless technology, by allowing smaller
devices and more efficient at the same time. IoT is a fundamental requirement for
power management: mobile devices obviously require batteries, but the possibility
of replacing them completely or restrict the replacement/charging is a factor of
considerable importance given the advent of further devices connected in the near
future. Thanks to new small autonomous devices that can retrieve small amounts of
energy arising from vibrations from the motorcycle and the electromagnetic waves
in the ether (such as those of mobile phones or TV). Several prototypes have been
designed from shoes with electromagnetic recovery to tensile structures in the road
coverage that can capture the energy derived from the traffic drafts; even the clothing
capable of incorporating the energy produced by the movement of legs and arms,
up to road surfaces able to absorb the vibrations of the vehicles and transform
them into electrical energy. Mechanical vibrations, which are continually subject
the streets and sidewalks, generate differences in electrical potential in piezoelectric
devices “drowned” in areas subject to the passage of vehicles or pedestrians. The
goal is to suppy in a sustainable way the lights sensors for traffic monitoring, street
lighting, illuminated signs. A futuristic scenario is from one side to do devices
without batteries, in many research centers are doing studies to optimize the power
consumption in processors and electronic devices, and on the other side to recover
the energy already available in the surrounding environment to be able to store and
reuse for the self-supply in energy-efficient technologies such as wireless networks,
sensors, LCD screens, MP3 players, GPS systems, and mobile phones.

References

1. Park, J., & Mackqy, S. (2003). Practical data acquisition for instrumentation and system
control. Oxford: Elsevier.
2. Lacanette, K. (2003). National temperature sensors handbook. Handbook Ann. Mat. National
semiconcductor
3. National Instruments (1996). Data acquisition fundamentals, Application Note 007.
4. National Instruments (1996). Signal conditioning fundamentals for PC-based data acquisition
systems, Handbook National Instruments.
5. Taylor, J. (1986). Computer-based data acquisition system. Research Triangle Park: Instrument
Society of America.
6. Di Paolo Emilio, M. (2013). Data acquisition system, from fundamentals to applied design.
New York: Springer.
164 11 Applications of Energy Harvesting

7. Roundy, S., Wright, P., & Pister, K. (2002). Micro-electrostatic vibration-to- electricity
converters. In Proceedings of ASME International Mechanical Engineering Congress and
Exposition IMECE2002 (Vol. 220, pp. 17–22).
8. Stordeur, M., & Stark, I. (1997). Low power thermoelectric generator: Self- sufficient energy
supply for micro systems. In Proceedings of the 16th International Conference on Thermo-
Electrics (pp. 575–577).
9. Shenck, N., & Paradiso, J. (2001). Energy scavenging with shoe-mounted piezoelectrics. IEEE
Micro, 21(3), 30–42.
10. Roundy, S. (2003). Energy scavenging for wireless sensor nodes with a focus on vibration to
electricity conversion. PhD thesis, University of California.
11. Tsutsumino, T., Suzuki, Y., Kasagi, N., Kashiwagi, K., & Morizawa, Y. (2006, November).
Micro seismic electret generator for energy harvesting. In Technical Digest PowerMEMS 2006,
Berkeley, USA (pp. 133–136).
12. Sterken, T., Altena, G., Fiorini, P., & Puers, R. (2007, April). Characterisation of an
electrostatic vibration harvester, DTIP of MEMS and MOEMS, Stresa, Italy.
13. Sterken, T., Baert, K., Puers, R., & Borghs, S. (2002, November). Power extraction from ambi-
ent vibration. In Proc. SeSens. Workshop on semiconductor sensors, Veldhoven, Netherlands
(pp. 680–683).
14. Szarka, G., Stark, B., & Burrow, S. (2012). Review of power management for energy harvesting
systems. IEEE Transactions on Power Systems, 27(2), 803–815. ISSN: 0885-8993.
15. Cammarano, A., Burrow, S. G., Barton, D. A. W., Carrella, A., & Clare, L. R. (2010). Tuning
a resonant energy harvester using a generalized electrical load. Smart Materials and Structure,
19, 055003.
16. Guyomar, D., Badel, A., Lefeuvre, E., & Richard, C. (2005). Toward energy harvesting using
active materials and conversion improvement by nonlinear processing. IEEE Transactions on
Ultrasonics, Ferroelectrics, and Frequency Control, 52, 584–595.
17. Mitcheson, P. D., Stoianov, I., & Yeatman, E. M. (2012). Power-extraction circuits for
piezoelectric energy harvesters in miniature and low-power applications. IEEE Transactions
on Power Electronics, 27, 4514–4529.
18. Szarka, G. D., Burrow, S. G., & Stark, B. H. (2012). Ultra-low power, fully-autonomous boost
rectifier for electro-magnetic energy harvesters. IEEE Transactions on Power Electronics,
28(7), 3353–3362. doi:10.1109/TPEL.2012.2219594.
19. Maurath, D., Becker, P. F., Spreeman, D., Manoli, Y. (2012). Efficient energy harvesting with
electromagnetic energy transducers using active low-voltage. IEEE Journal of Solid-State
Circuits, 47(6), 1369–1380.
20. Beeby, S. P., Tudor, M. J., & White, N. M. (2006). Energy harvesting vibration sources for
microsystems applications. Measurement Science and Technology, 17, R175–R195.
21. Khaligh, A., Zeng, P., & Zheng, C. (2010). Kinetic energy harvesting using piezoelectric and
electromagnetic technologies–state of the art. IEEE Transactions on Industrial Electronics,
57(3), 850–860.
22. Paulo, J., & Gaspar, P. D. (2010). Review and future trend of energy harvesting methods for
portable medical devices. In Proceedings of the World Congress on Engineering (Vol. 2).
23. Zhu, D., Tudor, M. J., & Beeby, S. P. (2010). Strategies for increasing the operating frequency
range of vibration energy harvesters: A review. Measurement Science and Technology, 21,
022001-1–022001-29.
24. Cepnik, C., Lausecker, R., & Wallrabe, U. (2013). Review on electrodynamic energy harvesters
– a classification approach. Micromachines, 4(2), 168–196. http://www.mdpi.com/2072-666X/
4/2/168. Accessed 20 Jan 2015.
25. Ulaby, F. T., Michielssen, E., & Ravaioli, U. (2010). Fundamentals of applied electromagnetics
(6th ed.). Upper Saddle River: Prentice Hall.
26. Roundy, S., Wright, P. K., & Rabaey, J. M. (2003). A study of low level vibrations as a power
source for wireless sensor nodes. Computer Communications, 26(11), 1131–1144.
27. Sazonov, E., Li, H., Curry, D., & Pillay, P. (2009). Self-powered sensors for monitoring of
highway bridges. IEEE Sensors Journal, 9, 1422–1429.
References 165

28. Toh, T. T., Mitcheson, P. D., Holmes, A. S., & Yeatman, E. M. (2008). A continuously
rotating energy harvester with maximum power point tracking. Journal of Micromechanics
and Microengineering, 18, 104008-1-7.
29. Howey, D. A., Bansal, A., & Holmes, A. S. (2011). Design and performance of a centimetre-
scale shrouded wind turbine for energy harvesting. Smart Materials and Structures, 20,
085021.
30. Razavi, B. (2002). Design of analog CMOS integrated circuits. New York: McGraw-Hill.
31. Razavi, B. (2008). Fundamentals of microelectronics. Chichester: Wiley.
32. Sedra, A. S., & Smith, K. C. (2013). Microelectronic circuits. Oxford: Oxford University Press.
33. Gudan, K. (2015). Ultra-low power 2.4GHz RF energy harvesting and storage system with
25dBm sensitivity. San Diego: IEEE. ISBN 978-1-4799-1937-6.
34. Briand, D. (2015). Micro energy harvesting. Weinheim: Wiley.
35. Spies, P. (2015). Handbook of energy harvesting power supplies and applications. Boca Raton:
CRC Press.
Index

A Common source MOS, 131

AC-DC, 76 Conditioning, 76
active sensors, 2 Conductors, 55
Amplifier, 131 Constan-charge, 69
amplifier, 120 Continuos system, 67
Antenna, 41, 144 Converter, 76
Applications, 155 Cosmic rays, 27
Armstrong oscillator, 94 Current Mirror, 125
Atoms, 37
Automotive, 158
D
DAQ, 1
B DC-DC, 41, 76
BAN, 146 Diffusion, 57
Bandwidth, 132 digital, 1
Battery, 143 Diode, 30, 42, 110
BiCMOS, 145 Donors, 57
BJT, 111
Boltzmann, 57
Boost converter, 91 E
Bridge rectifier, 79 Earth, 28
Buck converter, 91 Efficiency, 60
Buffer, 98 electromagnetic wave, 37
Building automation, 155 electrostatic transducer, 65
Emitter follower, 113
energy, 1, 11
C Entropy, 58
Capacitor, 66 Enviromental monitoring, 157
Carnot, 58 Ethernet, 157
Channel, 129 Excitation piezo, 51
Circular spiral antenna, 148
CMOS Inverter, 125
Comb drive, 70 F
Common drain MOS, 131 Feedback, 122
Common gate MOS, 131, 132 Ferrite, 146

© Springer International Publishing AG 2017 167

M. Di Paolo Emilio, Microelectronic Circuit Design for Energy
Harvesting Systems, DOI 10.1007/978-3-319-47587-5
168 Index

FET, 129 Mobile, 26, 146

field wiring, 7 Mosfet, 69, 116, 129
Figure of merit, 61 MPP, 32
flicke noise, 7 MPPT, 87
Folded dipole, 149 Multi-layer, 61
Forward Bias, 110
Friss, 37
Full-wave circuit, 78 N
NMOS, 124
nMosfet, 129
G noise, 7
Gap closing, 70 Nyquisit, 124
Gradient, 57
O
H Operational amplifier, 134
Half-wave circuit, 78 Output response, 132
HVAC, 15
P
Parallel plate, 70
I passive sensors, 2
inertial motion, 65 Peacemakers, 144
IoT, 162 Peltier, 55
Photovoltaic, 160
Photovoltaic cell, 30
J piezoceramic, 47
Junctions, 55 Piezoelectric, 22, 69
piezoelectric, 47
Piezoelectric frequency, 49
L piezoelectric model, 48
L network, 41 Plank, 30
langasite, 48 PMOS, 124
light sensors, 5 pMosfet, 129
Linear regulators, 89 polycrystalline, 48
LNA, 139 potentiometer, 5
load, 13 Power management, 150
Load matching, 95 power management, 15
Low noise pre-amplifier, 139 PSRR, 15, 136
PZT, 47
M
magnetic field, 4 Q
magnetic sensors, 4 Q factor, 42
Matching circuit, 42
Maxwell, 37
Mechanical, 65 R
mechanical stress, 48 Rectenna, 146
memory board, 3 Rectifier, 42, 76
MEMS, 22, 66 rectifier, 13
micro-generators, 66 Rectifier circuits, 78
Microelectronic, 105 Reverse Bias, 109
Micropowering, 78 RF harvesting, 151
Microstrip, 149 Rolling rod, 70
Miller, 132 RTD, 3
Index 169

S thermocouples, 3
Schottky, 42 Thermodynamics, 59
Seebeck, 24, 55 Thermoelectric, 55
semiconductor, 106 Thermoelectric generator, 59
Semicondutors, 57 Thermogenerators, 155
sensors, 1 thin film, 105
settling time, 8 TPM, 158
SHM, 157 transconductor, 132
shot noise, 7 transducers, 1
Silicon, 69
silicon, 106
Silicon dioxide, 129 V
SNR, 120 Variable overlap, 70
Solar, 88 Vibration, 21
Solar infrastructure, 160 vibration, 11
Solar radiation, 27 Vibrations, 49
Solar wind, 27 Voltage control, 86
Space radiation, 27 voltage regulator, 13
storage, 18
Sun spectrum, 28
Supercapacitor, 26 W
Supercapacitors, 98 Wearable, 143
Switched system, 67 Wi, 37
Switching regulators, 90 Wind infrastructure, 160
Wireless sensors, 157
T WSN, 18
Teflon, 69
TEG, 143, 155
Thermal, 22 Y
thermal, 12 Young, 49
Thermal gradient, 24
Thermal noise, 7
thermistors, 3 Z
Thermocouple, 59 Zener diode, 81