The ESD Control Program Handbook
The ESD Control Program Handbook
Jeremy M Smallwood
Electrostatic Solutions Ltd
Southampton, UK
This edition first published 2020
© 2020 John Wiley & Sons Ltd
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in
any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by
law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/
permissions.
The right of Jeremy M Smallwood to be identified as the author of this work has been asserted in accordance
with law.
Registered Offices
John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
Editorial Office
The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
For details of our global editorial offices, customer services, and more information about Wiley products visit us at
www.wiley.com.
Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that
appears in standard print versions of this book may not be available in other formats.
10 9 8 7 6 5 4 3 2 1
To Jan, who has often put up with my grumpy non-communication while I’ve been writing
this book. To Alia, now making her own life journey. To Caroline, who tragically died so
young.
To the subject of electrostatics and ESD, that has kept me occupied, perplexed and challenged
for many hours.
To the many people who have attended my courses and asked so many awkward questions
that helped me understand while trying to explain. To my fellow ESD practitioners whose
opinions and expertise give us many interesting, and sometimes heated, debates – and who
mainly agree the answer to most electrostatic questions is – “it depends.”
vii
Contents
Introduction xxvii
Foreword xxxiii
Preface xxxvii
Acknowledgments xxxix
4.7.9.1 Basic Requirements for Grounding Personnel Handling ESDS Devices 111
4.7.9.2 Grounding Personnel via a Wrist Strap 112
4.7.9.3 Grounding Personnel via Footwear and Flooring 112
4.7.9.4 Grounding Seated Personnel 113
4.7.10 Grounding ESD Control Equipment 113
4.7.10.1 General Considerations 113
4.7.10.2 Work Surfaces 113
4.7.10.3 Floors 116
4.7.10.4 Carts, Racks, and Other Floor Standing Work Surfaces 117
4.7.10.5 Seats 118
4.7.10.6 Tools 119
4.7.10.7 Gloves and Finger Cots 120
4.7.10.8 ESD Control Garments 121
4.7.11 What If a Conductor Cannot Be Grounded? 122
4.8 Habit 5: Protect ESDS Using ESD Packaging 122
4.8.1 Don’t Take Ordinary Packaging Materials into an EPA 122
4.8.2 The Basic Functions of ESD Packaging 123
4.8.3 Open ESD Packaging Only Within an EPA 123
4.8.4 Don’t Put Papers or Other Unsuitable Material in a Package with an ESDS
Device 124
4.9 Habit 6: Train Personnel to Know How to Use ESD Control Equipment and
Procedures 124
4.9.1 Why Train People? 124
4.9.2 Who Needs ESD Training? 125
4.9.3 What Training Do They Need? 126
4.9.4 Refresher Training 126
4.10 Habit 7: Check and Test to Make Sure Everything Is Working 126
4.10.1 Why Do We Need to Check and Test? 126
4.10.2 What Needs to Be Tested? 127
4.10.3 ESD Control Product Qualification 127
4.10.4 ESD Control Product or System Compliance Verification 127
4.10.5 Test Methods and Pass Criteria 127
4.10.6 How Often Should ESD Control Items Be Tested? 128
4.11 The Seven Habits and ESD Standards 129
4.12 Handling Very Sensitive Devices 129
4.13 Controlling Other ESD Sources 130
References 131
Further Reading 132
11.6.4 Concentric Ring Electrodes for Packaging Surface and Volume Resistance
Measurement 352
11.6.4.1 Conversion of Surface Resistance to Surface Resistivity for a Concentric Ring
Electrode 354
11.6.4.2 Conversion of Volume Resistance to Volume Resistivity 354
11.6.5 Two-Point Probe for Packaging Surface Resistance Measurements 354
11.6.6 Footwear Test Electrode 354
11.6.7 Handheld Electrode 355
11.6.8 Tool Test Electrode 356
11.6.9 Metal Plate Electrode for Volume Resistance Measurements 356
11.6.10 Insulating Supports 357
11.6.11 ESD Ground Connectors 357
11.6.12 Electrostatic Field Meters and Voltmeters 357
11.6.12.1 Electrostatic Field Meter–Based Instruments 357
11.6.12.2 Noncontact Electrostatic Voltmeters 358
11.6.12.3 Contact Electrostatic Voltmeters 359
11.6.13 Charge Plate Monitors (CPM) 359
11.7 Common Problems with Measurements 361
11.7.1 Humidity 361
11.7.2 Accidental Measurement of Parallel Paths 361
11.8 Standard Measurements Specified by IEC 61340-5-1 and ANSI/ESD S20.20 362
11.8.1 Resistance to Ground 362
11.8.1.1 Resistance to Ground from a Work Surface or Floor 362
11.8.1.2 Compliance Verification of Seating in the EPA 364
11.8.1.3 Qualification of Seating 364
11.8.2 Point-to-Point Resistance 366
11.8.2.1 Point-to-Point Resistance of Work Surface 366
11.8.2.2 Point-to-Point Resistance Measurements on ESD Garments 367
11.8.2.3 Simple Point-to-Point Measurement on the Garment Fabric 368
11.8.3 Personal Grounding Equipment Tests 371
11.8.3.1 End-to-End Resistance of a Ground Cord 371
11.8.3.2 Measurement of Wrist Straps and Cords as Worn 372
11.8.3.3 Measurement of Personnel Grounding Through Footwear to Foot Plate
Electrode 374
11.8.3.4 Measurement of Resistance from Person to Ground 374
11.8.3.5 Resistance to Ground of an Earth Bonding Point 377
11.8.4 Surface Resistance of Packaging Materials 377
11.8.4.1 Surface Resistance of Packaging Measured Using a Concentric Ring
Electrode 378
11.8.4.2 Point-to-Point Resistance of Small Packaging Items 379
11.8.4.3 Point-to-Point Resistance of Packaging Using 2.5 kg Resistance Measurement
Electrodes 380
11.8.5 Volume Resistance of Packaging Materials 381
11.8.6 ESD Shielding of Bags 382
11.8.7 Evaluation of ESD Shielding of Packaging Systems 383
xxii Contents
Index 477
xxvii
Introduction
Electrostatic discharge (ESD) can damage or destroy many types of modern electronic com-
ponents, or modules or assemblies containing electrostatic discharge–susceptible (ESDS)
components.
In electronics manufacturing, sensitivity to ESD became a general concern after the
adoption of metal-oxide-semiconductor (MOS) transistor technology exacerbated by
decreasing internal physical size of semiconductor component features and the rise of
integrated circuits (ICs). The first Electrical Overstress/Electrostatic Discharge Sympo-
sium in the USA was organized in 1979 (Reliability Analysis Center 1979). The 1980
Symposium (Reliability Analysis Center 1980) shows papers on topics as diverse as theory
and practice, device failure analysis studies, failure mechanisms and modeling, design
of device protective networks, and implementing ESD controls, facility evaluation and
effective training. Standards and technical handbooks also emerged around this time. The
standards gave requirements for an ESD control program, while the technical handbooks
gave technical data and tutorial material useful for educating the user and developing the
ESD control program. ElectroStatic Attraction (ESA) of contaminant particles is a problem
for manufacturers of semiconductor devices and displays. For operating electronic systems,
ESD provides a source of electromagnetic interference (EMI) that can result in system
crash, malfunction or data corruption.
Thus, the issues of ESD in electronic components and systems give two areas of interest.
Issues of ESD control during electronic component, assembly and system manufacture are
largely concerned in preventing damage in the unpowered non-operational state and ensur-
ing that product reaches the customer in good condition without compromise to appearance
or reliability. This area can itself be further subdivided into
● Electrostatic and ESD issues affecting product yield and quality during semiconductor
wafer scale fabrication
● ESD issues affecting product yield and quality during component, assembly and system
manufacture and assembly, sometimes known as “factory issues”
● Design of semiconductor devices to withstand ESD up to target levels
ESD interference and damage during working electronic system operation is generally
viewed as part of ElectroMagnetic Compatibility (EMC) and the responsibility of a different
xxviii Introduction
community. In some areas (e.g. Europe) electronic products are subject to ESD immunity
test as part of their evaluation for fitness to be placed on the market (and qualification for
CE marking) (Williams 2007).
Nevertheless, there is some overlap and often confusion between these areas. EMI caused
by ESD in the manufacturing process can cause interference to product testers and lead
to rejection of product and hence reduction of yield. In ESD test in EMC, ESD applied to
exposed circuit connectors can lead to component or system hardware failure and may lead
to requirements for ESD robustness of components that connect to the outside world.
This book is largely concerned with the development and maintenance of an ESD control
program for protection of ESD susceptible components and assemblies during electronic
system and assembly manufacture. This is the so-called “factory issues” area of ESD control.
It is intended that the book can be used as a handbook or practical guide for personnel work-
ing in ESD control in electronics or other companies that handle unprotected ESD suscepti-
ble parts. At the same time, sufficient background information and technical explanation is
given to enable the user to understand the principles and practice of effective ESD control.
Personnel working in this field can have a wide variety of technical background and do
not necessarily have strong electronics or electrical understanding. Many will not have had
opportunity to attend courses on ESD control other than basic ESD awareness. It is sur-
prising that at the time of writing very few University electronics related courses offer any
modules on ESD control. Conversely there are as yet few industry courses available that
deal with the subject in any depth other than basic ESD awareness. Worldwide, there are
still only a very few courses and qualifications available to those who wish to obtain a good
grounding in the field.
So, I have attempted to present the subject with a minimum of theory to make it accessi-
ble to those who do not have a strong relevant theoretical background. This is balanced by
sufficient description of background theory for understanding the material presented, with
references and a bibliography of further reading for those who wish to go into the subject in
greater depth. The intention is to reveal and clarify the principles behind an area often con-
sidered a mysterious “black art.” In many ways, I have tried to write the book I would have
liked to have found when I started learning about ESD control in the electronics industry.
A widespread current approach to development of an ESD control program is to com-
ply with the requirements of an ESD control standard such as ANSI/ESD S20.20 or IEC
61340-5-1. It is often thought that this is sufficient to ensure that product ESD damage is
brought under control. While this can be successful, if applied with insufficient knowl-
edge it can lead to a program that is not well optimized or fails to address all the ESD
threats (Smallwood et al. 2014; Lin et al. 2014). With knowledge and understanding an
optimized and effective ESD control program may be achieved and maintained, often with
lower costs. Nevertheless, compliance with an ESD control standard is advantageous and
can help demonstrate, especially to customers, the seriousness with which ESD control
is treated in the facility. Development of an ESD control program in compliance with the
most widely used and respected ESD control standards 61340-5-1 and S20.20 is therefore
discussed in some length. Properly specified ESD control programs compliant with these
standards are held to be adequate to protect ESD sensitive devices with withstand voltages
down to 100 V Human Body Model (HBM), while also addressing basic ESD risks due to
charged metal objects and charged devices.
Introduction xxix
ESD susceptible components become ever more sensitive to ESD damage as time goes
on, due to component technology developments. The need for development of ESD control
programs through knowledge and understanding rather than rote application of standard
techniques will grow as a greater number of more ESD sensitive components are handled
in electronic manufacturing, assembly, and maintenance processes in the future. In parallel
with the development of ESD control techniques and standards, a massive research effort
has supported on-chip ESD protection networks aimed at reducing device ESD suscepti-
bility, with target withstand voltages of 2 kV HBM, 200 V Machine Model (MM) and 500 V
Charged Device Model (CDM) (Industry Council 2011). In the early 2000s The Industry
Council on ESD Target Levels was formed with members from IC manufacturing and elec-
tronics assembly companies, and independent consultants in the industry. In the face of
increasing difficulty in achieving the target ESD withstand levels, and the belief that mod-
ern electronic manufacturing companies have ESD control programs routinely achieving
the standard protection levels, they recommended reduction of on-chip target protection
levels to 1 kV HBM, 30 V MM and 250 V CDM (Industry Council 2011, 2010a,b). At the
same time, many discrete components and ICs exist that for various reasons do not have
on-chip ESD protection or otherwise have lower ESD withstand voltage than these levels. It
seems likely that this will be the first of many reductions in target level driven by technology
changes and the assumption that industry can handle lower ESD withstand components.
While this book is primarily intended to support the industry factory practitioner, I hope
that this book will encourage and enable Universities and Further Education organizations
to offer courses and modules on ESD control for personnel who wish to make a career in
electronics production and related fields.
This book does not attempt to address electrostatics and ESD control in semiconductor
wafers and device manufacture, or device design for ESD protection. The former may be
still as yet inadequately covered by the very few books available on the subject but is better
discussed in a book more focused on this technology area such as Welker (2006). The topic of
device design is best covered by specialist books such as Amerasekera and Duvvury (2002)
and Wang (2002).
ESD immunity of operating electronic systems is left to be treated in other books as part
of EMC issues, except for some discussion confined to areas of overlap with the ESD factory
issues topic. This field is more concerned with the design of electronic systems for immunity
to ESD than it is with ESD control (Ind. Co. White Paper 3, Johnson and Graham 1993;
Montrose 2000; Williams 2007).
While electrostatic control is used in other industries such as explosives and flammable
materials handling (the latter known in Europe as “ATEX”), these are typically governed by
other standards or regulations. They are only mentioned in this book to draw attention to
possible confusion areas and help avoid mistakes, for example in equipment specification
and sourcing.
While this book could be read “cover to cover” it is probably more likely that the reader
will “dip into” specific chapters as the need arises to learn about different topics while work-
ing in ESD control. The book has been written with this in mind. For those who wish to go
deeper into the subject, lists of references are provided with each chapter.
Every specialist subject has its own set of specialist terms or uses specific terms in spe-
cific ways. Chapter 1 defines and introduces the reader to the commonly used terminology
xxx Introduction
in ESD control. Whilst this chapter forms a general introduction to the key concepts and
terminology in the field, it is also likely to be used to revise or clarify the meaning of terms
during reading of other chapters. This is why the definitions and terminology has been
provided together in one chapter rather than being defined and explained as required dur-
ing the remainder of the book. Chapter 2 then explains in more detail the principles that
underlie ESD control work.
Chapter 3 discusses ESD susceptible devices, and how ESD susceptibility of a compo-
nent is measured. The range of ESD susceptibility of components, and current trends in
ESD susceptibility are reviewed. The topic of failure analysis as it is applied to ESD failed
components is outlined. Some case studies of ESD failures from the literature are briefly
described.
Chapter 4 describes the “seven habits of a highly effective ESD program.” This is a way
of explaining the essential activities of an effective ESD control program, that the author
has used in ESD training work for many years. If these activities are effectively and habitu-
ally implemented, it is likely that an ESD control program will be, and remain, effective. If
any one of them is neglected, it is likely the effectiveness of the ESD control program will
eventually suffer.
Most basic ESD control techniques and standards mainly address manual handling of
ESD susceptible devices, components, and assemblies. Chapter 5 extends the discussion to
ESD control in automated systems, processes and handling, which form a major part of
modern electronic manufacture.
Chapter 6 explains the approach and requirements given by the IEC 61340-5-1 and
ANSI/ESD S20.20 ESD control standards at the time of writing. These standards are
continually updated as time goes on, and so the reader is advised check for current versions
available at the time of reading.
Chapter 7 gives an overview of the equipment and furniture commonly used in ESD con-
trol and commonly specified for use in an electrostatic discharge protected area (EPA) to
control common ESD risks. The chapter explains how these often work together as part of
a system and must be specified with that in mind.
ESD protective packaging is one of the most misunderstood areas of ESD control. ESD
packaging is now available in an extraordinary range of forms from bags to boxes and bubble
wrap to tape and reel packaging for automated processes. The principles and practice of ESD
protective packaging are explained in Chapter 8. This is a deep and constantly developing
subject in itself, and this chapter can barely do more than give an introduction to it.
The thorny question of how to evaluate an ESD control program is addressed in Chapter
9 with a goal of compliance with a standard as well as effective control of ESD risks and
possible customer perceptions.
Whilst evaluating an existing ESD control program provides challenges, developing an
ESD control program from scratch provides others. Chapter 10 gives an approach to this.
ESD control product qualification and compliance verification is an essential part of an
ESD control program. Standard test methods have been developed and specified to go with
compliance with ESD control standards. These are explained in Chapter 11. The ESD con-
trol program may also need to use control measures and equipment that are not currently
specified in the standards. Some examples of test methods that may be used with these are
also given in this chapter.
Introduction xxxi
ESD Training has long been recognized as essential in maintaining effective ESD control.
Chapter 12 discusses this in more detail. It describes some demonstrations and techniques
which the author has used to help trainees understand static electricity, ESD and static
control in practice.
Finally, Chapter 13 attempts to look at where ESD control may go in the near future.
References
Amerasekera, A. and Duvvury, C. (2002). ESD in Silicon Integrated Circuits, 2e. Wiley. ISBN:
0 470 49871 8.
Industry Council on ESD Target Levels (2010a) White paper 2: A case for lowering
component level CDM ESD specifications and requirements. Rev. 2.0. http://www
.esdindustrycouncil.org/ic/en/documents/6-white-paper-2-a-case-for-lowering-
component-level-cdm-esd-specifications-and-requirements [Accessed: 10th May 2017]
Industry Council on ESD Target Levels (2010b) White paper 3: System Level ESD Part I:
Common Misconceptions and Recommended Basic Approaches. Rev. 1.0 http://www
.esdindustrycouncil.org/ic/en/documents/7-white-paper-3-system-level-esd-part-i-
common-misconceptions-and-recommended-basic-approaches [Accessed: 10th May
2017]
Industry Council on ESD Target Levels (2011) White paper 1: A case for lowering
component level HBM/MM ESD specifications and requirements. Rev. 3.0. Available
from: http://www.esdindustrycouncil.org/ic/en/documents/37-white-paper-1-a-case-for-
lowering-component-level-hbm-mm-esd-specifications-and-requirements-pdf
[Accessed: 10th May 2017]
Johnson, H. and Graham, M. (1993). High Speed Digital Design – A Handbook of Black Magic.
Prentice Hall. ISBN: 0 13 395724 1.
Lin N, Liang Y, Wang P. (2014) Evolution of ESD process capability in future electronics
industry. In: 15th Int. Conf. Elec. Packaging Tech.
Montrose, M. (2000). Printed Circuit Board Design Techniques for EMC Compliance, 2e. Wiley
Interscience/IEEE Press. ISBN: 0 7803 5376 5.
Reliability Analysis Center (1979). Electrical Overstress/Electrostatic Discharge Symposium
Proceedings. EOS-1. Griffiss AFB, NY: Reliability Analysis Center.
Reliability Analysis Center (1980). Electrical Overstress/Electrostatic Discharge Symposium
Proceedings. EOS-2. Griffiss AFB, NY: Reliability Analysis Center.
Smallwood J., Taminnen P., Viheriaekoski T. (2014) Paper 1B.1. Optimizing investment in
ESD Control. In: Proc. EOS/ESD Symp. EOS-36.
Wang, A.Z.H. (2002). On-Chip ESD Protection for Integrated Circuits. Klewer Academic Press.
Welker, R.W., Nagarajan, R., and Newberg, C. (2006). Contamination and ESD Control in
High-Technology Manufacturing. Wiley-Interscience/IEEE Press. ISBN-10: 0 471 41452 2
ISBN-13: 978 0 471 41452 0.
Williams, T. (2007). EMC for product designers, 4e. Newnes. ISBN-13: 978-0750681704
ISBN-10: 0750681705.
xxxii Introduction
Further Reading
Foreword
I was quite flattered when Dr. Jeremy M Smallwood asked me to write this foreword
for his book. I view this as a significant honor. Dr. Smallwood (Jeremy) and I have
worked together on standards for electrostatics since the mid-1990s after the IEC formed
Technical Committee 101 – Electrostatics. Both of us had considerable time invested in
the standards process on our sides of the Atlantic: Jeremy with BSI and myself with the
ESD Association (ESDA). In the early 1990s, several things occurred: The IEC formed
Technical Committee-TC101 – Electrostatics. Most of the original members were from
the CENELEC committee that produced the electrostatics document CECC00015 along
with delegates from other non-European countries; and the ESDA became a recognized
American National Standards (ANSI) development body, thus able to officially represent
the United States to the IEC. I was appointed the first lead delegate from the United
States National Committee, and Jeremy was appointed delegate from the UK National
Committee. I had the pleasure of hosting an early IEC TC101 working group meeting
in Austin, Texas, in 1996, following our annual EOS/ESD Symposium held that year in
Orlando, Florida. That was the first time I met Jeremy. While the early years of deliberation
on standards for electrostatics were contentious at times, the committee eventually came
together to form a cohesive group and has prepared important standards, recognized
around the world. Jeremy had considerable input and influence during the formative years
of TC101, including his long-term appointment as the TC101 Chairman. Currently, he is
the lead delegate from the UK and participates actively on many working groups.
As a past president of the ESDA, I was delighted to present Jeremy with the ESD
Association’s Industry Pioneer Award in 2010 in recognition of his contribution to the
science of electrostatics, and in particular, for his role in standards development.
There are many good books on the fundamentals of electrostatics by well-known authors
and practitioners such as Professor Niels Jonassen, University of Denmark and Professor
A.D. Moore (University of Michigan). There are other books that deal with specific issues in
electrostatics and these can be found in the Bibliography of this book. However, this is one
of the few that covers all the aspects of the modern standardization process for electrostatic
control in the manufacturing of electronics and other materials sensitive to electrostatic
xxxiv Foreword
influences. This book is designed to assist a novice to the world of electrostatics as well as
the expert practitioner. I believe the book could be a useful reference to anyone who has to
deal with electrostatics in any field and in particular, electronics manufacturing operations.
The principles and standards discussed herein may be applied to any manufacturing area
and process. This book would also be useful as a text for a college level course dealing with
electronic design when the subject matter turns to reliability and sensitivity of electronic
parts and assemblies.
Dr. Smallwood uses his considerable experience in standards development and practical
experience to guide the reader through the maze and tangle often associated with applica-
tion of standards to a manufacturing process. Since electrostatics is a natural phenomenon
and around us all the time, one would think standardization is an improbable task (maybe
leaning toward impossible). The practical approach taken by Jeremy in Chapters 9 and 10
helps sort out the implementation and program management processes.
Chapter 12 on Training should be extremely useful to anyone that has to set-up, run and
maintain an ESD control program. Jeremy has provided some very useful “tips” on training
considerations, garnered from his years of experience in providing basic to advanced
classes on the “art and science” of electrostatics.
Another important section is Chapter 6 which covers standards very well. Since the early
1990s, the understanding of how to deal with the “mysteries” of electrostatics in industry
has increased dramatically, with the result being able to develop very useful and practical
standards, standard practices, advisories and operating guides. While we must appreciate
and understand that static electricity cannot be prevented, we now know how to live with
it in the manufacturing world and generally can provide mitigation techniques to resolve
most issues before serious problems occur. That being said, fires and explosions, product
damage and loss of life occur every year due to static electricity happening when least
expected or when proper procedures are not followed. Simply having a bad connection
to ground (earth) in the wrong place at the wrong time can (and does) have catastrophic
results more often than most will realize.
The first step in resolving or preventing electrostatic issues is to develop an under-
standing of the phenomenon. Chapters 1–5 provide an excellent introduction to the basic
considerations involved in electrostatics and will be a great starting point for the novice in
this subject. Even the experienced practitioner will find these chapters useful for review
and provide a reminder of what they have “forgotten.” Overall the book is written for
the non-technical person to be able to understand the electrostatic phenomenon but the
“expert” in the field will also find it useful. The extensive bibliography provides a great
resource for anyone needing details on any of the related subjects.
Foreword xxxv
I am confident that you will find this book as entertaining, enlightening and useful as I
have. Happy reading and stay safe.
Preface
Although the subject of ESD has been a concern to the electronics industry since the 1970s,
there have been relatively few books written on the subject from the point of view of the
person who has to put together, or evaluate, maintain, and update an effective ESD control
program. In my work as a consultant and trainer I have met many people in this position.
Some of these, when I met them, had recently had responsibility for the ESD program
thrust upon them with little or no experience of the subject. Others had some knowledge
but were confused by the array of facts and myths they had learned about the subject.
Still others had developed true expertise over a considerable experience in the subject and
developing and running their own company ESD Programs. Of course, I have learned from
them all – as a trainer and consultant I find that I learn most from trying to explain my
subject to others. Doing so often challenges my own understanding and causes me to think
things through with greater clarity.
In the process, I started trying to simplify the presentation of the principles of ESD preven-
tion. With a nod to Steven Covey’s “Seven Habits of a Highly Effective People” the “Seven
Habits of a Highly Effective ESD Program” was born. Why “Habits”? Because if these things
are done habitually when handling ESD susceptible components and assemblies, we are
well on the way to having an effective ESD control program.
In the mid 1990s, then working for ERA Technology Ltd., I started participating in devel-
opment of British Standards at BSI in Chiswick, London. Through this work I soon found
myself participating also in international standards through BSI’s participation in Interna-
tional Electrotechnical Commission (IEC) Technical Committee101 “Electrostatics” which
was formed in the mid 1990s. Standards work was an eye-opener and I found it simultane-
ously highly stimulating and very frustrating. Stimulating, because I found myself talking
with experts from around the world whose knowledge of their field could be exceptionally
deep, and their experience widely varied and practical. At BSI we could argue for hours
about technical issues, and how best to write a standard test method, that could be under-
stood and reproduced by anyone with reasonable technical ability and expected to produce
results which would agree with others who had done the same. In international standards,
we would first have to agree on a test method acceptable to the participating National Com-
mittees. There could be several to choose from already in use amongst the standards from
the participating countries. Naturally each National Committee expert would favor a partic-
ular approach, especially if it was already adopted within their country. Mostly, of course,
these would produce different results due to different conditions and methodologies.
xxxviii Preface
We would discuss these at length amongst a group of international experts whose cultures
and experiences could be very different, and English often not their mother tongue.
The resulting methodology would have to be acceptable in the very different climates,
conditions, and working practices in Japan, UK, USA, Canada, Scandinavia, France,
Germany, Italy or wherever the participating experts were from. In most cases the end
product would have to be translatable into the expert’s mother language for publication.
We found that some common ways of writing in English can be difficult to translate or may
be unclear in other languages. English phrases or words can even mean different things to
an American or a UK English speaker – leading to long discussions about the best wording
for a single sentence!
In the early 2000s the ESD Association standards were becoming widely accepted in
the electronics industry, and their use spreading from North America to other areas of the
world. Encouraged by ESDA standardization experts working with IEC TC101, the decision
to rewrite IEC 61340-5-1 with an unofficial harmonization with ANSI/ESDA S20.20 was a
landmark decision by the IEC TC101 Working Group 5 revising this standard. Subsequent
further harmonization has simplified the task of the ESD Coordinator, especially in
multinational companies.
As time goes on, the components commonly handled in electronics facilities are becom-
ing more susceptibility to ESD damage. The variety of facilities and processes in which
they are handled, stored or transported grows greater. This means that there is an increas-
ing necessity for the person developing, implementing and maintaining an ESD control
program to understand, analyze and specify effective protection against ESD risks.
Part way through writing this book, I realized I was trying to write the book I would have
liked to find when I first got into the subject of ESD control in the electronics industry. This
book aims to help the reader understand the principles and practice of ESD control to the
point where they can make the decisions needed to develop an effective and optimized ESD
control program compliant with the current ESD control standards. To do this one needs to
understand the purpose of ESD control equipment and materials, and how to specify and
test that it does the job intended. If the reader wishes to improve their knowledge further,
the references and bibliography given should give them a good starting point. Perhaps
most importantly, I hope this book will help the reader find that an initially mysterious set
of practices is actually based on sound engineering principles that they can learn to apply
with confidence.
xxxix
Acknowledgments
I would like to acknowledge my debt to all the experts in the field of ESD control and stan-
dardization who through discussions and published work have contributed to the current
state of my understanding. I am also indebted to all my clients and course attendees who
have challenged me to clarify, explain and justify effective ESD control techniques applied
in many different situations.
I would especially like to thank David E Swenson for his comments on the text and for
contributing photographs and other material, and for writing the Foreword, as well as for
many enlightening discussions over the years. Dave performed the extraordinary feat of
reading and commenting on almost all the draft Chapters at least once. This helped enor-
mously in picking up my mistakes and typographic errors, adding or clarifying important
points and generally improving my work.
Several other friends and colleagues have also very kindly read and commented on
Chapters of this book and encouraged me in this work. Special thanks are due to Rainer
Pfeifle, Charvakka Duvvury and Christian Hinz who each reviewed and commented on
various Chapters in detail. Bob Willis also contributed comments, and Charles Cawthorne
kindly provided me some photographs from his own ESD training materials. Lloyd Lawren-
son kindly allowed me to use Kaisertech facilities for some of my photography. I’m indebted
to Lisa Pimpinella of the ESD Association for arranging permission for me to include figures
from the 2016 ESD Association Electrostatic Discharge (ESD) Technology Roadmap.
Last, but definitely not least, I would like to thank my wife Jan for her good-natured tol-
erance of my absent mindedness and lack of communication when engrossed in my work,
and my daughter Alia for helping improve some of my photographs in preparation for pub-
lication in this book.
1
As with any specialist subject there are many terms that are the “jargon” of the subject
that can be confusing to the newcomer. There are also terms that have specific meanings
in the context of these standards but may have different meanings in common parlance.
The intention here is not to give strict and rigorous academic definitions, but to assist the
newcomer to the field to understand the following chapters.
Sometimes a range of meanings is common in different industries. For example, the terms
conductive, static dissipative, insulative or insulating, and antistatic can mean many differ-
ent things to different people from different industry areas or in the context of different
standards or electrostatic discharge (ESD) control product types. In most cases, only the
meaning common in ESD control, and in particular in the context of the IEC 61340-5-1 and
ANSI/ESD S20.20 and related standards, is emphasized here.
The task of supervising an ESD control program is often given to personnel from many
technical and educational backgrounds. For this reason, the minimum of prior technical
knowledge is assumed in this book.
Despite this, some of the terms used in this document are defined with basic mathemati-
cal relationships given where appropriate. This is because simple mathematics often helps
to clarify the subject and, in some cases, may be essential to helping the user understand
how to specify aspects of an ESD control program. In many cases, these aspects, and their
practical importance and application, are discussed further in Chapter 2.
In electrostatics and ESD work, we often have to deal with very large or very small numbers.
For example, the resistance of a material may be measured to be 10 000 000 000 Ω. For conve-
nience and clarity, we use scientific notation and SI unit prefixes as shorthand for numbers
(http://physics.nist.gov/cuu/Units/prefixes.html).
In scientific notation, the number is rewritten in the form a × 10b , where a lies between 1
and 10, and b is the number of decimal places a must be shifted to get the full number. This
is probably most easily understood by examples of resistance and capacitance (Table 1.1).
Sometimes, when the number a is simply 1, it is omitted.
1.2.2 Ions
Ions are very small particles having a small electrostatic charge. They are naturally present
in the air but may also be generated deliberately or accidentally around objects at high
voltage.
Charges are naturally present in atoms. Protons in the atomic nucleus have positive
charge, and electrons in the atom have negative charge. A negative ion is formed when a
particle gains one or more electrons. A positive ion is formed when a particle loses one or
more electrons. Ions may consist of free electrons, single atoms, many atoms, or molecules
(Wikipedia 2018). Sometimes these ions may become attached to larger particles.
V = QEs
The energy taken to move a charge between two points is the same, no matter what route is
taken between the points. Potential difference is measured in volts (V) and is often referred
to as voltage. The unit volt (V) is equivalent to joules per coulomb. Voltage is a measure of
the potential energy at a point and is perhaps analogous to pressure in a fluid system or
height in a gravitational system.
Engineers often talk about the potential of (for example) a conductor (see Section 1.7.3
for a discussion of conductors and insulators), as a synonym to voltage. This is not strictly
correct as potential is strictly the work done in bringing a charge from infinity to the place
of measurement (Jonassen 1998).
A voltage or potential difference at a place of measurement must always be referred to
another place. In practice, the potential difference is usually quoted with reference to the
potential of the earth (also referred to as ground; see Section 1.5), which is defined for con-
venience as zero volts. If this other place is not specifically stated, it is usually ground (the
earth).
All points in space surrounding a charge have a voltage (potential) – typically this voltage
will be different from its neighboring points. For a conducting surface, if it is not initially
4 1 Definitions and Terminology
an equipotential, voltage differences cause charge (current) to flow until the voltage around
the surface is eventually equal. So, an electrically conducting surface in equilibrium is an
equipotential surface.
F = qE
If equal positive and negative charge are sufficiently close together, from a distance their
electric field effects cancel, and no external field is noticed. The charges are said to be neu-
tralized.
Electrostatic fields and potentials around an object are not easy to visualize. One way of
doing so is by use of field and equipotential lines. A field line represents the path a small
charge would take, if it were free to move under the influence of the force due to the electro-
static field alone. Field lines always leave a conductor at a right angle (90∘ ) to the surface.
In Figure 1.1 a charged spherical conducting object has a potential V. Each point in the
surrounding space can also be assigned a potential, according to the work required to move
a unit charge to that position. If all the positions of equal potential are marked, an equipo-
tential line (or in three dimensions a surface) is marked out. A system of equipotential
surfaces could be marked, forming contours of potential showing the presence of the peak
in potential rather like the contours on a map showing the presence of a hill. Equipotential
lines are always at a right angle (90∘ ) to the field lines.
Equipotential lines are like contour lines on a map of an area of the earth’s surface. Height
is a form of potential energy. If a ball is released on a smooth hillside, it will roll down the
hill perpendicular to the contour lines. Similarly, if a same polarity charge (e.g. a positive
charge, next to a high positive potential) is present in the electrostatic field, it will move
away from the peak in potential, in a path perpendicular to the equipotential lines. These
paths form lines of electrostatic field. The intensity of the field is given by how close together
the field and equipotential lines are.
1.2 Charge, Electrostatic Fields, and Voltage 5
Q, V
Charged
Equipotential line is conductor
at right angle to field
line
The electric field E (vector, as it has magnitude and direction) is the gradient of voltage
V over a distance s. Electric field, therefore, has the units volts per meter (V/m).
−dV
E=
ds
In Figure 1.1 if the charged sphere is very small, it is effectively a point of charge. The
electrostatic field around the charge Q at a distance r from this point is proportional to the
charge present, according to Coulomb’s law (Cross 1987)
Q
E∝
r2
From this equation, in this case the field strength decreases rapidly with the distance from
the charge, with 1/r 2 . This is also indicated by the spreading of the field lines with distance
from the charge. Field lines can be considered to begin and end on electrostatic charges,
and so a high density of field lines at a surface implies a high charge density as well as high
electrostatic field.
For other shapes of charge patterns, the equipotentials will not in general be spherical,
and field lines will in general be curved rather than straight lines. Field lines are always
perpendicular to the equipotentials and are always perpendicular to conducting surfaces as
these are also equipotentials.
1.2.8 Permittivity
Coulomb’s law shows that the field due to a point charge is proportional to the charge and
inversely proportional to the distance from it squared (Cross 1987).
Q
E∝
r2
Permittivity (dielectric constant), 𝜀, was defined to give a convenient constant of propor-
tionality in this relation.
1 Q
E=
4𝜋𝜀 r 2
For air, the permittivity is very close to the permittivity of free space 𝜀0 (vacuum)
𝜀 = 𝜀0 = 8.8 × 10−12 Cm−1 . Other materials have different permittivity and affect field
strengths correspondingly. In general, a dielectric material has a permittivity greater than
air. This is conveniently expressed as a relative permittivity 𝜀r , and the permittivity is given
by
𝜀 = 𝜀r 𝜀0
Polymers often have relative permittivity in the range 2–3 and many other materials in the
range 2–10. Materials such as ceramics can have far higher permittivity.
∫0
Q= I dt
and so
dQ
I=
dt
1.5 Earthing, Grounding, and Equipotential Bonding 7
In many practical cases, one of the conductors concerned may already be electrically con-
nected to an electrical earth or can be conveniently connected to earth. The earth is often
defined as our 0 V reference point in electricity power distribution, electrostatics, and ESD
control. So, it is often convenient and is common practice to electrically connect all conduc-
tors to earth. Earth is also known as ground, and earthing is also known as grounding.
The terms earthing and grounding can have different meanings and requirements in dif-
ferent contexts or industries. An electrical engineer may require an earth resistance less
than an ohm. A plant engineer may earth bond two items of plant, requiring a resistance
less than 10 Ω. An electromagnetic compatibility (EMC) engineer may require an extremely
low impedance to be maintained from direct current (DC) to hundreds of MHz or even GHz.
To an ESD control practitioner, a resistance to ground <109 Ω at dc may be sufficient.
In practice in ESD control, there are various types of ground that can be used. In the ESD
standards IEC 61340-5-1:2016a and ANSI/ESD S20.20-2014, the term grounding is used to
mean any of the following:
● Connection to electrical earth (the safety earth wire of a mains electrical system)
● Connection to a functional earth (e.g. an earth rod driven into the ground)
● Connection to an equipotential bonding system
1.7.1 Resistance
Electrical resistance is the ratio between the dc voltage applied to a circuit or material and
the current flowing through it, given by Ohm’s law.
V
R=
I
1.7 Resistance, Resistivity, and Conductivity 9
Resistance
meter
Electrodes
Material surface
Resistance meter
Electrode area A
Material
designed for volume resistivity measurement so that conversion from volume resistance to
volume resistivity is a simple calculation. In ESD work, conversion to volume resistivity
is often not needed. The volume resistance obtained with defined standard electrodes is
used directly, saving the effort of calculation. Examples of surface and volume resistance
measurement methods are given in Chapter 11.
The conductivity, 𝜎, of the material is simply the inverse of its resistivity.
1
𝜎=
𝜌v
The units of conductivity are siemens per meter (Sm−1 ).
The resistivity of materials can vary by many orders of magnitude from 10−8 Ωm (e.g.
copper) to more than 1015 Ωm (e.g. mica, quartz, polytetrafluoroethylene, polyethylene).
● A conductor is a material that allows charge to move away quickly enough to avoid sig-
nificant electrostatic charge build up.
● An insulator is any material that is not a conductor, in other words, a material that does
not allow charge to move quickly enough to avoid charge build up.
Conductors are easily maintained at a low voltage by connecting them to earth (ground).
However, an insulator in electrostatic terms cannot be maintained at a low voltage by
installing a ground connection. The charge on the material simply does not move to the
ground connection quickly enough to be conducted away in the desired timescale.
Materials or equipment are often defined as conductors or insulators based on either their
measured resistance or a charge decay time. This is discussed further in Chapter 2.
Table 1.2 shows how the terms insulating, dissipative, conductive, and antistatic are widely
used in ESD control. Take care when using these terms, because they may be defined differ-
ently in different contexts and may mean different things to different people. When defined
in the standards, the precise definition can change as the standards evolve into new editions.
The situation becomes worse if usage of these terms in other industries and for specific
products is considered (Table 1.3). In general, these words should be considered unreliable
in meaning unless specified by standards as part of an ESD control system.
12 1 Definitions and Terminology
Table 1.2 Example of how meanings of conductive, static dissipative, insulative, and antistatic can
vary with context in ESD control in electronic manufacturing.
≥1011 Ω
ESD protective Surface resistance Surface and
≥1011 Ω
packaging volume resistance
Table 1.3 Example of how meanings of conductive, dissipative, and insulative can vary with context
in static control in other industries (IEC 60079-32-1:2013).
R1 R2 Rn
1.8 Capacitance
C1 C2 Cn
1.11 Relative Humidity and Dew Point 15
1.9 Shielding
The term shielding is used in ESD control in a different way to other disciplines, espe-
cially EMC and radio frequency work. Shielding definitions and tests used in ESD control
are often highly specific to the standards used. Typically, the term is used to describe the
attenuation of electrostatic fields or electrostatic discharge energy applied to the outside of
a protective package, measured at the inside of the package. This is discussed further in
Chapter 8.
If the temperature is lowered sufficiently, the water vapor pressure eventually becomes
equal to the saturated vapor pressure, and the air becomes saturated. Any further reduction
in temperature may result in moisture condensing from the air on to surfaces in contact with
it or in fog forming. This temperature is called the dew point.
References
Cross, J.A. (1987). Electrostatics Principles, Problems and Applications. Adam Hilger. ISBN:
0-85274-589-3.
EOS/ESD Association Inc. (2014) ANSI/ESD S20.20-2014. ESD Association Standard for the
Development of an Electrostatic Discharge Control Program for – Protection of Electrical
and Electronic Parts, Assemblies and Equipment (excluding Electrically Initiated Explosive
Devices). Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015a) ANSI/ESD STM 11.11-2015. ESD Association Standard for
Protection of Electrostatic Discharge Susceptible Items – Surface Resistance Measurement of
Static Dissipative Planar Materials. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015b) ANSI/ESD STM 11.12-2015. ESD Association Standard for
Protection of Electrostatic Discharge Susceptible Items – Volume Resistance Measurement of
Static Dissipative Planar Materials, Rome, NY, EOS/ESD Association Inc.
International Electrotechnical Commission. (2012) IEC/TR 61340-1: 2012. Electrostatics – Part
1: Electrostatic phenomena — Principles and measurements. Geneva, IEC.
International Electrotechnical Commission. (2013) PD/IEC TS 60079-32-1. Explosive
atmospheres Part 32-1. Electrostatic hazards, guidance. Geneva, IEC.
International Electrotechnical Commission. (2015) IEC 62631-3-2. Dielectric and resistive
properties of solid insulating materials - Part 3-2: Determination of resistive properties (DC
methods) - Surface resistance and surface resistivity. Geneva, IEC.
International Electrotechnical Commission. (2016a) IEC 61340-5-1: 2016. Electrostatics – Part
5-1: Protection of electronic devices from electrostatic phenomena - General requirements.
Geneva, IEC.
International Electrotechnical Commission. (2016b) IEC 61340-2-3:2016. Electrostatics.
Methods of test for determining the resistance and resistivity of solid planar materials used to
avoid electrostatic charge accumulation. Section 3: Methods of test for determining the
resistance and resistivity of solid planar materials used to avoid electrostatic charging. Geneva,
IEC.
International Electrotechnical Commission. (2016c) IEC 62631-3-1. Dielectric and resistive
properties of solid insulating materials - Part 3-1: Determination of resistive properties (DC
methods) - Volume resistance and volume resistivity - General method. Geneva, IEC.
Jonassen, N. (1998). Electrostatics. Chapman & Hall. ISBN: 0 412 12861 6.
Lawrence, M.G. (2005). The relationship between relative humidity and the dewpoint
temperature in moist air. A simple conversion and applications. Bull. Am. Meteorol. Soc.:
225–233. https://doi.org/10.1175/BAMS-86-2-225 [Available from htt.s://journals.ametsoc
.org/doi/pdf/10.1175/BAMS-86-2-225. Accessed 15th Aug. 2018.].
Wikipedia (2018) Ion, viewed 17 October 2018, [Available from https://en.wikipedia.org/wiki/
Ion]
17
2.1 Overview
ESD stands for electrostatic discharge or, according to some, electrostatic damage. This
chapter provides the basis of how static electricity arises and can lead to ESD in the real
world. It also provides the principles that underlie ESD control techniques and equipment
design.
Electrostatic charge can build up in a variety of ways. The charged object has an electro-
static field that could conceivably lead to an ESD event in several ways:
● Direct breakdown of sensitive parts due to high electric field
● Generation of an ESD event directly subjecting the part to discharge currents
● Generation of an ESD event subjecting a part to induced transient electric or magnetic
fields, or some other stress
At the root of any ESD event there is an object or surface that has a voltage that is different
to its surroundings. Without this voltage difference, no electric field is present, and no ESD
current can flow. Hence, the objective of ESD prevention measures has been to keep surface
voltages and electric fields to a low level, below which damaging ESD cannot occur.
A review of the explanation of electrostatic charge build-up and ESD sources included
here quickly reveals many ways in which ESD risks can be generated in the real world.
These are summarized in brief in this chapter.
In practice, the amount of charge imbalance required to give strong electrostatic effects
is surprisingly small. The limit of the amount of charge that can be built up on a surface is
governed by the electrical breakdown field strength of air, around 3 × 106 V m−1 . The sur-
face charge density required to give this field is only 2.64 × 10−5 Cm−2 (Cross 1987). This is
equivalent to about 1.7 × 1014 electrons m−2 , or 8 atoms per million on the surface acquiring
or losing an electron!
One common way in which static electrical charge imbalances can arise is when two
materials make contact and then are separated. While in contact, electrons move from one
material to the other at points of contact; this material gains a net negative charge, and the
donor material gains a net positive charge. When the objects are separated, the negatively
charged object can take its charge with it, leaving an equal positive charge on the other
object. Although it is really charge separation that takes place, it is common to refer to the
“generation” of static electrical charge.
Glass
Mica
Human hair
Nylon
Wool
Lead
Silk
Aluminum
Paper
Cotton
Steel
Wood
Epoxy-glass
Copper
Stainless steel
Acetate rayon
Polyester polyurethane
Polyethylene
Polypropylene
PVC
Silicon
PTFE Charges to negative polarity
dQ/dt = I
Charge R C
generation Resistance Capacitance
(current source)
For a charge generation rate of 1 μA, a resistance of 1010 Ω would yield a voltage of 10 kV.
In practice, electrostatic sources rarely generate charge at this rate or on a steady current
basis unless there is steady movement involved (e.g. in a conveyor system).
The rate of electrostatic charge generation is affected by many factors. Some of the key
factors are as follows:
The many factors involved in triboelectric charge separation make it a highly unpredictable
phenomenon.
CV = Q
Q
C=
V
In practice, capacitance is usually a variable that depends on the materials and nearby
objects and on the proximity to earth. Objects move around in daily life, and so their capac-
itance changes.
As an example, we can consider the human body. It is, in electrostatic terms, a conducting
object, being mainly composed of water, which is a conducting material. Even if we neglect
the nearby objects and earth, the human body can be approximated as a sphere that has a
similar surface area. The “free space” capacitance of a sphere is given by 4𝜋𝜀0 r, where r is
the radius and 𝜀0 is the permittivity of free space, 8.8 × 10−12 Fm−1 . Typically, a 1 m radius
sphere gives a useful approximation and has a “free space” capacitance around 110 pF.
2.3 Electrostatic Charge Build-Up and Dissipation 21
Conductor
area A
Nearby objects and earth increase this value. In fact, the human feet, on the earth, approx-
imate two parallel plate capacitors in parallel with the free space capacitance. Each capac-
itor is made up of two electrodes: the sole of the foot and the earth. These are separated by
a layer of material (the shoe sole, typically an insulating polymer of relative permittivity 𝜀r
around 2.5). Each foot capacitance varies from moment to moment as the feet are lifted and
replaced on the ground during walking. Each foot capacitance can be modeled as a parallel
plate capacitor (Figure 2.2), with plates of area A separated by a distance d and capacitance
C given by
A
C = 𝜀0 𝜀r
d
In general, if either the area A or the distance of separation d are changed, then the capaci-
tance will change. If the charge is held constant, increasing the capacitance will decrease the
body voltage, and reducing capacitance will increase body voltage. Reducing capacitance
can be achieved by reducing the area (e.g. standing on tip-toe) or increasing the separation
distance (e.g. raising the foot from the floor).
The previous equation shows that if the charge on conductor is unchanged and the capac-
itance changes, then the voltage of the conductor changes. For example, if a person’s body
capacitance changes between 50 and 150 pF while walking and the charge on their body is
constant at 5 nC, their body voltage will vary between 100 V (at 50 pF) and 33 V (at 150 pF).
If a printed circuit board (PCB) conductor has a capacitance of 20 pF and charge 5 nC when
resting close to a large earthed machine part, its voltage will be 250 V. If its capacitance is
reduced to 5 pF when far away from this machine part, its voltage will rise to 1000 V.
It can be useful to have an idea of the approximate capacitance of everyday objects, espe-
cially when estimating the possible effect of ESD to or from such items. Table 2.2 gives some
examples (IEC 61340-1).
The variable ratio between charge and voltage behaves similarly for nonconductors.
When seated in a chair, the body generates charge on the clothes surfaces in contact with
the chair. This forms a large area of charged material with a small distance between the
charges (the two surfaces are in contact). The person’s body voltage is low in this situation
even though their clothing may be highly charged. On rising from the chair, the person can
take much of the separated charge with them. The effective “capacitance” between the body
and the chair is rapidly reduced (separation is rapidly increased), and a high body voltage
quickly results if the charge cannot dissipate to ground. It is common to feel a shock on
touching something metal after rising from a chair or car seat – voltages over 10 kV have
been measured on people after getting out of a car seat (Pirici et al. 2003; Andersson et al.
2008).
22 2 The Principles of Static Electricity and Electrostatic Discharge (ESD) Control
A corollary of this is that the voltage and field surrounding a charged object may be sup-
pressed by the presence of a nearby conducting object. If the capacitance of the system is
increased, the voltage is decreased.
As an example, a charged garment that fits snugly to the body has voltage suppressed
due to the proximity of its surfaces to the body. Even if the garment is highly charged, the
external field may be limited due to this. If the garment flaps open, the body and garment
surfaces move apart. “Capacitance” is reduced, and a high voltage and electrostatic field
appears outside the garment.
𝜏 = RC
In a time 𝜏, the voltage will decay to about 37% of its initial value.
In the example, if the charge current is suddenly halted at time t = 0 with initial voltage
V 0 , the voltage V on the capacitor reduces as
−t
V = V0 exp
𝜏
An electrostatic field meter monitoring the material surface would measure this exponen-
tial decay of voltage. The product of a material’s resistivity 𝜌 and permittivity 𝜀0 𝜀r gives a
physical time constant for the material.
𝜏 = 𝜌𝜀0 𝜀r
This behavior has important practical implications. If we consider a situation in which the
capacitance is fixed at 100 pF (the order of magnitude of capacitance of a person) and the
charging current 100 nA, we can consider the effect of different resistances. With a resis-
tance of 1 GΩ, the voltage generated is only 100 V, and on cessation of the current, the
2.3 Electrostatic Charge Build-Up and Dissipation 23
Initial
voltage
100%
Voltage
curve
Decayed
voltage
0 Time τ t
voltage will fall to 37% of its initial value within 109 × 10−10 = 0.1 seconds. The effect of
a short duration charging current of this magnitude is unlikely to be noticed.
If the resistance is increased 10 GΩ, not only is the voltage generated increased to 1 kV,
but on cessation of the current, the voltage will take 1010 × 10−10 = 1 seconds to fall to 37%
of its initial value. The presence of this voltage may or may not be noticeable or cause a
problem, depending on the circumstance.
If the resistance is increased 100 GΩ, not only is the voltage generated increased to 10 kV,
but on cessation of the current, the voltage will take 10 seconds to fall to 37% of its initial
value. The presence of this voltage for such a long time could lead to the person experiencing
shocks on touching something or discharging to cause some problem.
In ESD control, a different definition of charge decay time is usually used in standard
measurements, and often the time for charge to reduce to one-tenth of its initial value is
measured (Figure 2.3). This value is theoretically equal to 2.3𝜏.
In practice, the charge decay time is often measured from the starting voltage down to
a certain threshold voltage, e.g. 100 V. Polymers may have time constants of many tens or
hundreds of seconds, or even days under clean dry conditions.
In practice, the simple model does not always correspond well with material behavior.
Measured charge decay curve may depart considerably from the ideal exponential, and the
measured time “constant” varies with measurement conditions. Often with high resistance
materials the decay time lengthens as the surface voltage drops and may become very long
at low voltages.
Table 2.3 The effect of humidity on typical electrostatic voltages (MIL HDBK 263).
indoors and heated, very low relative humidity can result. Hence, ESD problems can be
seasonal and occur often in winter. Even in a room where the relative humidity is con-
trolled, dry local microclimates can form where there are heat sources such as equipment,
especially if air circulation is restricted.
A view of the effect of relative humidity on static electricity in daily life is indicated by
the following typical voltages (Table 2.3) given by MIL HDBK 263 as observed at different
ambient humidities. These are indicative and cannot be used to predict voltages occurring
in real situations.
0V
For larger and different shaped objects, the field line pattern can be very different and the
fall-off in field strength can be much less rapid. In practice, the field lines start and finish
on conductors at different voltages in the region, and field lines may be more or less curved
at any region in space between them.
The electrostatic field between two large flat parallel plates, well away from the plate
edges, is uniform (Figure 2.5), and the field lines are parallel between the electrodes. Away
from the plate edges, the field E is uniform and is easily calculated from the voltage differ-
ence between the plates V and the distance between them d.
V
E=
d
If a conductor is placed in an electrostatic field, it has the effect of drawing the field to
itself with field lines always emerging at right angles to the conductor surface. In response,
charges on the conductor are redistributed until the voltage is the same all over the con-
ductor surface. One of the consequences of this is that any instrument we use to mea-
sure an electrostatic field inevitably changes the field it is measuring. Figure 2.6 shows
how this happens with the electrostatic field between a metal plate at voltage V and a
grounded electrostatic field meter. The field meter actually sees a field higher than the V/d
value. The same effect happens of course for any component, PCB or any other electrostatic
discharge–sensitive (ESDS) device brought into the field. The density of the field lines at the
2.4 Conductors in Electrostatic Fields 27
Distance d
Metal
plate
Field E
Voltage
V
Figure 2.6 Electrostatic field between a field meter and metal plate at voltage V .
surface are related to the field strength and surface charge density induced at the surface.
A high concentration of field lines indicates a high field strength.
Field lines tend to congregate at the tip or edge of an object, and the electrostatic field
strength becomes more intense in these regions. Discharges tend to occur preferentially
from high field strength regions at sharp edges on objects. This is used to an advantage, for
example, in using sharp pins to produce intense fields and corona discharges as a source of
ions in an ionizer for charge neutralization.
For a charged insulating surface, the situation is even more complicated. A charged insu-
lator will normally have a highly variable charge density over its surface. The surface voltage
is highly dependent on the surface charge density and presence of other materials nearby.
Conductive
container surface
has equal voltage
No field
within container
+V
Vm
Charged Metal
object plate
Q = Cm Vm = Cg (V − Vm )
Cg V
Vm =
(Cg + Cm )
2.4 Conductors in Electrostatic Fields 29
Note that the metal plate would have a capacitance and voltage with respect to ground,
even if the field meter were not present. Although there is now a voltage on the plate, the
net charge remains zero (+Q −Q), and it is not charged!
This changing voltage on an object happens in practice if any conductive object passes
through an electric field. If, for example, an integrated circuit passed into an electric field
arising from a charged garment surface, it could acquire a voltage in this way. If it were
subsequently grounded in this state, an ESD event would happen.
Charge
Charge Q
Q Conductive
container
Induced voltage differences can also lead to breakdown over small gaps between nearby
conductors in a field, if the voltage difference exceeds the gap breakdown voltage. This can
also lead to ESD risks.
waveform is highly dependent on the source and “load” circuit characteristics and can have
unidirectional or oscillatory waveforms (see Section 2.7).
The energy E stored in a capacitor C charged to voltage V is easily calculated using this
simple formula
E = 0.5CV 2
In the absence of significant series resistance, it is often reasonable to assume that all this
energy is transferred to the discharge.
The electrical breakdown field strength of about 3 MV m−1 is valid for normal air pressure
and rather large distances (e.g. for a gap of 10 mm and large diameter or flat electrodes,
the breakdown voltage would be about 30 kV). The relationship between breakdown field
strength and air pressure is given by Paschen’s law (Kuffel et al. 2001) and is nearly linear for
larger gaps and uniform fields. At smaller gap distances d the breakdown voltage reaches a
minimum (known as the Paschen minimum). As breakdown voltage V b is also dependent on
atmospheric pressure P, the Paschen curve is usually plotted as breakdown voltage against
the product Pd (Figure 2.11). For air, according to Paschen’s law, below about 350 V no
breakdown occurs (minimum Pd 0.55 Torr cm, or 7 μm at 1 atm), and ESD can happen only
with direct metal-to-metal contact. There is evidence that in practice discharges can occur
through small gaps below the Paschen minimum voltage, possibly due to field emission
(Wallash and Levitt 2003).
Breakdown
voltage Vb
Paschen
Minimum minimum
breakdown
voltage Vb
Figure 2.11 The relationship between breakdown voltage and spark gap Pd (Paschen curve).
32 2 The Principles of Static Electricity and Electrostatic Discharge (ESD) Control
50
–0
–50 0 50 100 150 200 250 300 350 400 450
–50
–100
–150
Current (mA)
–200
–250
–300
–350
–400
–450
Time (ns)
Figure 2.12 Discharge from negatively charged (>20 kV) insulating surface.
Norberg and Lundquist 1991; Smallwood 1999; Landers 1985). The power dissipation and
energy of a brush discharge is not easy to calculate.
Any object that is at a different voltage from an ESDS device can be a source of ESD if
the object can touch the device or come close enough for a discharge to jump a small air
gap between them. The ESD that occurs may be more or less damaging or problematic
according to its characteristics. Different ESD sources produce waveforms with very differ-
ent characteristics in terms of parameters such as peak current, duration, energy and charge
transferred to the device, and frequency spectrum. Even an apparently similar source can
give widely different ESD waveforms under different circumstances. Some examples of real
ESD waveforms are given next – these may or may not be representative of ESD produced
from similar sources in other real situations, which may be highly variable.
2.6 Common Electrostatic Discharge Sources 33
60
50
Current (mA)
40
30
20
10
0
–100 0 100 200 300 400 500 600
–10 Time (ns)
1.8
1.6
1.4
1.2
Current (A)
1.0
0.8
0.6
0.4
0.2
–0.0
–50 0 50 100 150 200
–0.2
Time (ns)
Figure 2.13 Example of waveform of discharge from the author charged to 500 V and discharging
via skin of a finger (above) and small metal object (coin, below).
34 2 The Principles of Static Electricity and Electrostatic Discharge (ESD) Control
2.0
1.5
Current (A)
1.0
0.5
0.0
–5 0 5 10 15 20 25 30
Figure 2.14 ESD waveform from screwdriver blade charged to +530 V. Charge transferred 0.03 nC.
3.0
2.5
Current (A)
2.0
1.5
1.0
0.5
0.0
–5 0 5 10 15 20 25 30
–0.5
Figure 2.15 ESD waveform from a160 × 180 mm metal plate charged to 550 V. Charge transferred
2.5 nC.
2.6 Common Electrostatic Discharge Sources 35
0.2
0.1
0
–5 –5 –5 10 15 20
–0.1
ESD current (A)
Time (ns)
–0.2
–0.3
–0.4
–0.5
0.4
0.2
0
–5 0 5 10 15 20
–0.2
ESD current (A)
–0.4
–0.6
–0.8
–1 Time (ns)
–1.2
Figure 2.16 ESD waveforms from charged integrated circuits: (above) 32-pin plastic-leaded chip
carrier and (below) 24-pin dual-inline package.
36 2 The Principles of Static Electricity and Electrostatic Discharge (ESD) Control
0.5
0
–2 0 2 4 6 8 10 12 14 16
Time (ns)
–0.5
ESD Current (A)
–1
–1.5
10
0
–2 0 2 4 6 8 10 12 14 16
–2
Time (ns)
ESD current (A)
–4
–6
–8
–10
–12
–14
Figure 2.17 ESD waveform from a printed circuit board (above) charged to 1 kV (below) field
induced charged by insulator 40 mm away.
2.7 Electronic Models of ESD 37
20
Current (A)
1.5
1.0
0.5
0.0
–20 0 20 40 60 80 100 120
–0.5 Time (ns)
Figure 2.18 ESD waveform from a charged automotive module taken out of a polythene bag.
Charge transferred 35 nC.
6
5
4
3
2
Voltage (kV)
Time (s)
1
6
–1 0 5 10 15
–2
–3
–4
–5
made to the module in this state, a discharge can occur at the termination at which contact
is made.
2
Current (A)
1
Time (ns)
0
–50 0 50 100 150 200 250 300
–1
–2
–3
(a)
0
Current (A)
–2
–3
–4
(b)
Figure 2.20 ESD waveform from a charged automotive wiring loom cable lying against an earthed
metal plate. Positive (above) and negative (below) charging polarity.
Cs
Capacitance Victim device
Rd
The discharge is usually initiated by a breakdown of an air gap or some other insulating
medium. At low voltages, it can also be initiated by contact or near-contact between two
conductors. The discharge can itself have significant impedance RESD that can affect the
waveforms produced and the energy delivered into the victim device. Often, however, this
is negligible compared to the other impedances in the circuit, especially for larger ESD
events.
2.7 Electronic Models of ESD 39
After the discharge commences, the current flows through some circuit that includes
some elements of resistance Rs and inductance Ls . These are normally due to the resistance
and electrical properties of the materials in the current path.
In the case of ESD to a victim device, the device also has impedance, modeled in this
simple circuit by a resistance Rd . In practice, a nonlinear impedance would be more typi-
cal of a semiconductor device. The impedance of the spark channel is highly variable and
nonlinear.
For simplification, the total circuit resistance R is assumed to be linear and is the sum of
the circuit resistances.
R = Rs + RESD + Rd
For derivation of the equations for this and the following equations, the reader is referred to
other texts (e.g. Agarwal and Lang 2005, https://en.wikipedia.org/wiki/RLC_circuit). This
equation has two roots α, β given by
( ) √( )
R R2 1
𝛼, 𝛽 = − ± −
2Ls 4L2s Ls Cs
The waveform shape takes very different forms depending on the circuit component val-
ues. If the total circuit resistance is large and dominates the discharge path impedance,
the waveform has a unidirectional shape, simulated in Figure 2.22 using model component
values given for human-body model ESD (see Table 3.12). This occurs when
R2 1
2
≫
4Ls LC
I (Rd)
260mA
240mA
220mA
200mA
180mA
160mA
140mA
120mA
100mA
80mA
60mA
40mA
20mA
0mA
0ns 60ns 120ns 180ns 240ns 300ns
Figure 2.22 Simulated overdamped device current waveform IESD for dominant circuit resistance:
Rs = 1500 Ω, Rd = 10 Ω, RESD = 0 Ω, Ls = 10 000 nH, C s = 100 pF, V ESD = 500 V.
40 2 The Principles of Static Electricity and Electrostatic Discharge (ESD) Control
The discharge current rises rapidly to a peak I p that, when inductance is small, approaches
the value and polarity near that predicted by Ohms law.
Ip ≈ VESD∕
Rs
Thereafter, the current drops nearly exponentially with decay time approaching Rs CESD .
At the other extreme, if the circuit resistance is insignificant compared to the inductive
and capacitive impedance, the waveform is quite different. This occurs when
R2 1
≪
4L2s LC
The waveform rises to a peak and then oscillates negative and positive about zero. The over-
all amplitude decays exponentially with time, simulated in Figure 2.23 using values given
for machine model ESD (Table 3.12).
Between the two extremes, the waveform duration decreases and is minimum around the
point of critical damped waveform, where the waveform changes between the two different
shape types. This is simulated in Figure 2.24 using model values close to those given for the
charged device model (Table 3.12). This occurs at the condition
R2 1
=
4L2s LC
Practical ESD sources often require the addition of more components (e.g. additional capac-
itors) to better represent additional charge storage (e.g. metal parts) and other features that
may be present. These may contribute further current peaks or modify the shape of the
waveform (Verhage et al. 1993).
The stored energy EESD in a conductive ESD source is given by
2
EESD = 0.5Cs VESD
I (Rd)
6.3A
5.4A
4.5A
3.6A
2.7A
1.8A
0.9A
0.0A
–0.9A
–1.8A
–2.7A
–3.6A
0ns 60ns 120ns 180ns 240ns 300ns
Figure 2.23 Simulated underdamped device ESD current waveform for the case of low circuit
resistance (dominant inductive and capacitive impedance): RESD = 10 Ω, Rd = 10 Ω, Ls = 750 nH,
C s = 200 pF, V ESD = 500 V.
2.8 Electrostatic Attraction (ESA) 41
I (Rd)
11A
10A
9A
8A
7A
6A
5A
4A
3A
2A
1A
0A
–1A
0.0ns 1.0ns 2.0ns 3.0ns 4.0ns 5.0ns
Figure 2.24 Simulated device ESD current waveform for near critical damping: RESD = 20 Ω,
Rd = 10 Ω, Ls = 2.5 nH, C s = 10 pF, V ESD = 500 V.
All this energy is dissipated in the total circuit resistance R. Only a fraction of this is the
energy dissipated in the device Ed .
EESD Rd
Ed =
R
In a source such as the charged human body that has significant resistance, most of the
stored energy is dissipated in the circuit (body) resistance, and only a small fraction is dis-
sipated in the victim device. In contrast, a charged metal object is a low-resistance ESD
source, and most of the stored energy can be dissipated in the victim device. This is one
reason why some components may be damaged by a lower voltage with a metal ESD source
compared to a charged person. In general, the likelihood of ESD damage to a component by
ESD from a source will depend on the susceptibility of the device to ESD current, voltage,
energy, or other parameter of the discharge. This is further discussed in Chapter 3.
to be transported to a wafer within the field. Particle contamination can then cause loss of
product yield (Welker et al. 2006).
Other processes in which product cleanliness is important can include
● Manufacture of flat-screen displays. Loss of even a small number of pixels due to contam-
ination can result in rejection of the product.
● Packing of consumer products where dust or particle contamination can mar the appear-
ance of the product before purchase.
● Assembly of optical systems where performance can be reduced by contamination.
● Assembly of medical systems where infection of the user may be a risk.
vd = μE
The mobility of the ion is dependent on the ion size. In air, charges bind to water, nitrogen,
and other molecules or particles and form small or large ions. Small ions have mobility in
the range 1−2 × 10−4 m2 V−1 s−1 (Jonassen 1985). Large ions have mobility in the range
8 × 10−7 to 3 × 10−8 m2 V−1 s−1 .
The number of air ions present can be increased using an ionizer. These produce air
ions by various means such as corona discharge, radioactive, or X-ray ionization of the
air. Radioactive and X-ray ionization sources provide both polarity ions by splitting air
molecules into positive and negative ions.
Corona discharge sources use a high voltage applied to a sharp electrode (e.g. needle) to
produce ions of one polarity. A nearly balanced ion source can be produced by this method
by using an alternating current (AC) high voltage or two separate sources of opposite
polarity.
A charged surface produces an electrostatic field surrounding it that repels like polarity
ions and attracts opposite polarity ions. That is, a negatively charged surface repels negative
ions and attracts positive ions. A positively charged surface attracts negative ions and repels
positive ions. Opposite polarity ions will drift to the charged surface at a rate proportional
to the field strength and in numbers proportional to the ion concentration. An opposite
polarity charge on reaching the charged surface neutralizes an equal charge, reducing the
net surface charge and electrostatic field. The ion drift represents a neutralizing current,
limited by the ion concentration and field strength.
2.8 Electrostatic Attraction (ESA) 43
2.8.3 Ionizers
Ionizers are devices used to generate air ions for neutralization of surface charges on
charged materials and objects (Jonassen 1985, 1986). These are available in various types
based on passive, radioactive, electrical, and other principles of operation.
Passive ionizers rely on high electric fields developed around sharp points or edges
on earthed conductors to generate air ions by corona discharge. These will always
generate ions of the opposite polarity to the voltage producing the field at the point or
edge. Unfortunately, corona discharges do not occur below a threshold field strength,
and this means there is a threshold voltage, known as the corona inception voltage,
below which ions are not produced, and neutralization does not occur. This threshold
voltage may be several kilovolts (kV). This means that passive ionizers cannot be used
to reduce voltages to below this threshold, and they are seldom useful in ESD control in
electronics manufacture. They do find major applications in industrial processes such as
plastic film manufacturing, printing, copiers, and other processes involving insulating
materials.
Electrical ionizers also use the high electrostatic fields at points, usually needles, to
generate air ions. In this case, however, a power supply is used to raise the needle to a
voltage above the corona inception voltage to ensure that sufficient ions are produced.
A balanced ion source with equal numbers of positive and negative ions is usually required.
This can be produced either by using an AC driving voltage or by using two sets of nee-
dles, one at positive voltage and the other at negative voltage. This can produce an ion
stream that is nearly balanced, but it is difficult to produce an exact balanced ion stream
by this method.
Radioactive ionizers use a radioactive source to ionize the air by impact of radioactive par-
ticles with molecules in the air. Exactly balanced ion streams are produced by this means, as
the molecule splits into two ions, one positive and one negative ion. The rate of production
of ions is small and limited by the level of radioactivity of the ionizing material.
Offset voltage
100
0
25
–5 –100 0 5 10 15 20
–300
–400 Decay time
–500
–600
–700
–800
–900
–1000
–1100
Figure 2.25 Electrostatic voltage on a CPM plate reducing during charge neutralization showing
decay time and offset.
Charge neutralization by ionizers is typically quite a slow process and can take tens of sec-
onds or longer, depending on the electrostatic field strength and ion density in the region
surrounding the charged surface. This charge density, and the speed of charge neutraliza-
tion, can be affected by many factors.
A poorly maintained ionizer can become seriously out of balance. This can happen in
service due to erosion or contamination of the ionizer needles. If this occurs, then sur-
faces within the ion cloud may become charged to hundreds or even thousands of volts.
Monitoring this effect over time and performing appropriate maintenance of an ionizer to
prevent the imbalance from reaching an excessive level is an important aspect of ionizer
maintenance (Simco Ion TN-003 2019).
Considering these conditions shows why ESD damage is a probabilistic phenomenon. Only
a small fraction of ESD events may cause ESD damage, because
● The capacity of the ESD source to produce damage varies with the characteristics of the
source.
● The magnitude of the source voltage and energy will vary tremendously with the condi-
tions that cause it.
● The variation in discharge paths will cause great variation in characteristics of the dis-
charge such as peak current, waveform, rise and fall times and duration, and power and
energy deposited in the ESDS device.
● The susceptibility of the ESDS device to peak current, power, energy, waveform, and other
parameters of the discharge through it varies between devices.
● The ESD current path through the ESDS component may be through parts that have
different damage thresholds.
● The ESD event may not be sufficiently strong to exceed the damage threshold of its path
through the component.
So, ESD may occur many times during handling or processing a component, without ESD
damage occurring. The probability of ESD damage occurring may be one in hundreds or
thousands of ESD events. Combine this with the fact that damage will be discovered often
at a much later manufacturing stage, and may not be identified as due to ESD, it is small
wonder that many people get the impression that ESD damage is something that does not
happen to them!
Nevertheless, each time that a conductor contacts an ESDS component is an opportunity
for ESD to occur. Given sufficient ESD opportunities and insufficient ESD control, ESD
damage becomes inevitable.
● ESDS are handled only within an electrostatic discharge–protected area (EPA; see
Chapter 4) or under ESD-controlled conditions. This ensures that when ESDS devices
are handled, the ESD risks are controlled to an acceptable level.
References 47
● Outside the EPA in unprotected areas (UPAs), ESD protective packaging is used to protect
the ESDS device. The packaging is designed to prevent ESD sources outside the package
from having a significant effect on the ESDS device within the package. It also provides
a safe region within the package in which ESD risks are controlled.
Within the EPA, ESD risks are controlled by eliminating as far as is possible the sources of
ESD likely to damage the ESDS.
● Conductors that may touch an ESDS, especially metal items and people, are grounded
wherever possible. This ensures that as far as possible, electrostatic voltages on conduc-
tors are the same and near zero. This is to prevent them from becoming charged and a
source of significant ESD to ESDS devices.
● Where conductors cannot be grounded and might contact an ESDS, the voltage difference
between the conductor and the ESDS device must be reduced to a sufficiently low level
to prevent significant ESD risk.
● Electrostatic fields that may occur near ESDS devices are eliminated or reduced to a low
level. Insulators that may become charged and the source of electrostatic fields are, where
possible, removed from the vicinity of ESDS devices. This is to reduce the risk of induced
charging of isolated ESDS devices or other ungrounded conductors.
● Items that may contact an ESDS device are preferably made from materials that have
appreciable electrical resistance. If ESD occurs, this helps to reduce discharge current to
a safe level and absorb much of the discharge energy within the material rather than in
the ESDS device.
These measures do not eliminate ESD but help reduce the numbers of ESD occurring, the
magnitude of any ESD that occurs, and the likelihood of it damaging the ESDS. Reducing
the number of times contact is made with an ESDS device can also help reduce the likeli-
hood of ESD damage occurring. So, the simple measures of not handling the ESDS device
any more than necessary and reducing to minimum contact between the ESDS device and
other conductors that might be at different voltage can make a useful contribution to reduc-
ing ESD risk. If the materials are resistive rather than low resistance where they contact
ESDS devices, then a further reduction of risk of ESD damage is made. This is due to reduc-
tion of peak current in the discharge and absorption of energy by the resistance of the
material.
Suitably specified ESD protective packaging can reduce the risk of ESD damage occurring
in the UPA to an insignificant level. The packaging must be specified to address the ESD
risks appropriate to the ESDS component or item.
References
Agarwal, A. and Lang, J.H. (2005). Foundations of Analog and Digital Electronic Circuits.
Morgan Kaufmann, ISBN 1-55860-735-8.
Andersson, B., Fast, L., Holdstock, P., and Pirici, D. (2008). Charging of a person exiting a car
seat. Electrostatics 2007. J. Phys. Conf. Ser. 142: 012004.
Bailey, A.G., Smallwood, J.M., and Tomita, H. (1991). Electrical discharges from the human
body. In: Electrostatics –Inst. Phys. Conf. Se. 118 Sec. 2. Inst.Phys.
48 2 The Principles of Static Electricity and Electrostatic Discharge (ESD) Control
Barnum, J.R. (1991). Sandia’s severe human body electrostatic discharge tester (SSET). In: Proc.
EOS/ESD Symp. EOS13, 29–30. Rome, NY: EOS/ESD Association Inc.
Cross, J.A. (1987). Electrostatics Principles, Problems and Applications. Adam Hilger. ISBN
0-85274-589-3.
Department of Defense. Military Handbook. (1994) Electrostatic Discharge Control Handbook
for protection of electrical and electronic parts, assemblies and equipment (excluding
electrically initiated explosive devices) (metric). MIL HDBK-263B. Washington DC,
Department of Defense.
International Electrotechnical Commission. (2013) PD/IEC TS 60079-32-1. Explosive
atmospheres Part wp2-1. Electrostatic hazards, guidance. Geneva, IEC.
Jonassen N. (1985) The physics of air ionization. In: Proc. EOS/ESD Symp. EOS-7 Minneapolis
USA. Rome, NY, EOS/ESD Association Inc. pp. 59–66.
Jonassen, N. (1986). The physics of air ionization. In: Proc. EOS/ESD Symp. EOS-8, 35–40.
Rome, NY: EOS/ESD Association Inc.
Jonassen, N. (1998). Human body capacitance – static or dynamic concept? In: Proc. EOS/ESD
Symp. EOS-20, 111–117. Rome, NY: EOS/ESD Association Inc.
Kelly MA, Servais G E, Pfaffenbach T V. (1993) An Investigation of Human Body Electrostatic
Discharge. 19th International Symposium for Testing & Failure Analysis Los Angeles,
California, USA. Russell Township, OH, ASM International.
Kuffel, E., Zaengl, W.S., and Kuffel, J. (2001). High Voltage Engineering. Newnes ISBN 0 7506
3634 3.
Landers, E.U. (1985). Distribution of charge and fieldstrength due to discharge from insulating
surfaces. J. Electrostat. 17: 59–68.
Norberg, A. (1992). Modelling current pulse shape and energy in surface discharges. IEEE
Trans. Ind. Appl. 28 (3): 498–503.
Norberg, A. and Lundquist, S. (1991). A distributed RC transmission line model for electrostatic
discharges from insulator surfaces. In: Inst. Phys. Conf. Se. 118, 269–274. Electrostatics 1991.
Norberg, A., Szedenik, N., and Lundquist, S. (1989). On the pulse shape of discharge currents.
J. Electrostat. 23: 79–88.
Pirici, D., Rivenc, J., Lebey, T. et al. (2003). A Physical model to explain electrostatic charging in
an automotive environment: Correlation with experimental approach. In: Proc. EOS/ESD
Symp. EOS-25, 161. Rome, NY: EOS/ESD Association Inc.
Simco Ion (2019) Emitter point maintenance. Technical note TN-003. Available from: https://
technology-ionization.simco-ion.com/DesktopModules/Bring2mind/DMX/Download.aspx?
command=core_download&entryid=112&language=en-US&PortalId=0&TabId=145
[Accessed 16 April 2019]
Smallwood, J.M. (1999). Simple passive transmission line probes for electrostatic discharge
measurements. In: Inst. Phys. Conf. Se. 163, 363–366. Electrostatics 1999, Inst.Phys.
Tamminen P, Viheriäkoski T, Ukkonen L, Sydänheimo L (2015) ESD and Disturbance Cases in
Electrostatic Protected Areas. In: Proc EOS/ESD Symp. 5B.2, Rome, NY, EOS/ESD
Association Inc.
Verhage, K., Roussel, P.J., Groeseneken, G. et al. (1993). Analysis of HBM ESD Testers and
specifications using a 4th order lumped element model. In: Proc. EOS/ESD Symp, 129–137.
Rome, NY: EOS/ESD Association Inc.
Further Reading 49
Viheriäkoski T, Peltoniemi T, Tamminen T (2012) Paper 4A3. Low Level Human Body Model
ESD. In: Proc. EOS/ESD Symp. Tucson Ariz. USA. Rome, NY, EOS/ESD Association Inc.
Wallash A, Levitt L. (2003) Electrical breakdown and ESD phenomena for devices with
nanometer-to-micron gaps. In: Proc. SPIE 4980, Reliability, Testing, and Characterization of
MEMS/MOEMS II. San Jose, CA, SPIE.
Wallash A, Smith D. (1998) Paper 4B.6. Electromagnetic Interference. (EMI) damage to giant
magnetoresistive (GMR) recording heads. In: Proc. EOS/ESD Symp., Rome, NY, EOS/ESD
Association Inc. pp. 368–74.
Welker, R.W., Nagarajan, R., and Newberg, C.E. (2006). Contamination and ESD Control in
High-Technology Manufacturing. Wiley ISBN-13: 978-0-471-41452-0.
Further Reading
EOS/ESD Association Inc. (2015) ANSI/ESD STM3.1-2015. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Ionization. Rome, NY, EOS/ESD
Association Inc.
International Electrotechnical Commission. (2017) IEC 61340-4-7:2017. Electrostatics - Part 4-7:
Standard test methods for specific applications – Ionization. Geneva, IEC.
51
Figure 3.1 ESD-sensitive devices range from individual transistors or diodes to PCBs or modules.
change. They may still pass functional tests to fail at a later stage. This is called a latent
failure.
It is not just individual components that may be susceptible to ESD. Any printed circuit
board (PCB), module, or assembly containing ESDS is likely to be itself an ESDS device if
not in some way protected against ESD. In some cases, the susceptibility to ESD may be
limited to remaining contact points such as flying leads, connectors, or limited exposed
conductors.
Fully assembled systems that include ESDS, protected within an enclosure or housing
that acts as an effective barrier to ESD, are normally no longer considered an ESDS device.
In many cases, these systems must be tested for ESD immunity during operation, using
standard ESD immunity tests required by electromagnetic compatibility (EMC) regulations
or requirements. These system ESD immunity tests are aimed at ensuring functionality
of the system is not compromised during operation by ESD occurring in the uncontrolled
operating environment or during consumer operations.
Working electronic systems tend to suffer partial or temporary (soft) failures due to ESD.
The system may recover almost immediately, or it may suffer effects such as program crash
or malfunction, spontaneous reset, or data corruption. Often effects such as program crash
or malfunction may be completely recovered by restarting the system. Data corruption
effects may lead to corrupted data remaining in the system.
Often the remaining connector terminations and other potential ESD entry points will
have been designed for immunity to expected ESD threats. Sometimes, a risk of ESD damage
3.2 Measuring ESD Susceptibility 53
can remain from strong ESD to components within the system connected to lines emerging
on connectors or accessible via user interfaces such as keyboards or touch screens. Severe
ESD to system parts such as cable discharge to connectors can sometimes result in physical,
hard damage to the system. For this reason, some ESD protection may need to be built into
the system.
Device
C
and/or
Vesd spark gap
54 3 Electrostatic Discharge–Sensitive (ESDS) Devices
Table 3.1 Typical ESD model simulation component values for common sources.
Charging resistor
100MΩ
Rs
Test
High voltage Device load
CESD under
generator 500 Ω
test or
0Ω
Table 3.2 ESD model component values to achieve specified waveforms in standards.
Human body model (HBM) IEC 60749-26 ANSI/ESD/JEDEC JS-001 1500 100
Machine model (MM) IEC 60749-27 ANSI/ESD/JEDEC STM 5.2 200
Human metal model (HMM) IEC 61000-4-2 ANSI/ESD S5.6 330 150
These are defined in terms of waveform characteristics when discharged into a defined
load, rather than model component values. The waveforms defined by these standards are
given in Sections 3.2.4–3.2.8. Typical model component values used to create these wave-
forms are given in Table 3.2.
Current
Ip
0.9 Ip Ir
0.36 Ip
0.1 Ip td
tr Time
Table 3.3 ANSI/ESDA/JEDEC JS001-2017 HBM waveform parameters with 0 Ω calibration load.
After pulses have been applied to all pin combinations at the test level, the device is tested
for failure. A failure is concluded when the device no longer meets the required parametric
and functional parameters. Devices are classified after test, according to their failure voltage
level (Table 3.4).
The peak current in an HBM discharge is mainly determined by the series 1.5 kΩ
resistance. Stray inductance and circuit resistance have relatively little effect. In this model,
the device has little effect on the waveform, providing it has low impedance compared to
the series resistance. In this case, much of the energy stored in the capacitor is dissipated
in the series resistor and not in the device.
The rather long rise time limit of 25 ns allowed equipment manufacturers to build testers
(that typically have high stray capacitance and slow the rising edge) for testing high pin
count devices.
0Z <50 V
0A 50 to <125 V
0B 125 to <250 V
1A 250 to <500 V
1B 500 to <1000 V
1C 1000 to <2000 V
2 2000 to <4000 V
3A 4000 to <8000 V
3B ≥8000 V
Current
Ipeak
0.9 Ipeak
I at 30 ns
I at 60 ns
0.1 Ipeak
tr Time
30 ns 60 ns
1 2 2
2 4 4
3 6 8
4 8 15
x User defined User defined
1 2 7.5 0.8 4 2
2 4 15 0.8 8 4
3 6 22.5 0.8 12 6
4 8 30 0.8 16i 8
The rise time is measured between 10% and 90% of the value of the first current peak.
The current is measured at 30 and 60 ns from the time at which the current reaches
10% of the first peak value.
Nevertheless, the 61 000-4-2 waveform has sometimes been used to test ESD withstand
of ESDS devices where particularly severe test conditions are required, especially for com-
ponents that have pins that are likely to emerge directly to system connectors. Recently the
IEC 61000-4-2 waveform has been adopted for component tests as the human metal model
(HMM) (ANSI/ESD S5.6 (EOS/ESD Association Inc. 2009)). This practice has arisen from
demand that components that have pins that are likely to emerge directly to system con-
nectors should be tested for susceptibility using the system test waveform (Ashton 2008;
Industry Council 2010a,b).
Other system test models exist, for example ISO 10605 “Road vehicles – Test Methods for
electrical disturbances from electrostatic discharge,” which is intended for use in evaluating
ESD susceptibility of electronics modules for vehicle use (International Organization for
Standardization 2008). This uses combinations of 150 or 330 pF capacitances with 330 or
2000 Ω resistance and test voltages up to 25 kV air discharge or 15 kV contact discharge.
The IEC 60749-27 standard defines MM current waveforms for device test purposes. The
waveform parameters are defined in terms of peak current and period (Figures 3.6 and 3.7),
and they give required values for qualification of the MM test equipment measured with
0 and 500 Ω calibration loads (Table 3.7). With a device load, waveforms will typically be
different from those obtained with the calibration loads as the device under test adds sig-
nificant impedance and is likely to act as a nonlinear load.
Current
Ip1
tpm
t0 t1 t2 t3 Time
Ip2
Current
Ip
I100
100 ns Time
Figure 3.7 IEC 60749-27 MM waveform definition with 500 Ω calibration load.
Table 3.7 IEC 60749-27 MM waveform parameters with 0 and 500 Ω calibration loads.
A 200 pF capacitor simulates charge storage on the conductive object. There is no defined
additional series resistance or inductance, but in practice stray inductance and resistance
are always present in the discharge circuit. The stray inductance and circuit resistance,
including the impedance of the device subjected to the ESD, determines the peak current.
The fact that these are not well defined makes the MM ESD withstand test more prone to
variation than HBM. As the circuit resistance is low, most of the stored energy is dissipated
in the device.
As with HBM, device evaluation typically is done using three samples of the device. Each
sample is tested with negative and positive polarity ESD with a minimum pulse interval of
300 ms. The test starts at the lowest voltage level and is increased progressively, providing
the devices do not fail. Pulses are applied to each pin in turn, with each power pin in turn
grounded. Pins that are not grounded or under test are left floating. Pulses are also applied
to all nonsupply pin pair combination.
After pulses have been applied to all pin combinations at the test level, the device is tested
for failure. A failure is concluded when the device no longer meets the required parametric
and functional parameters. Devices are classified after test, according to their failure voltage
level (Table 3.8).
The MM waveform is highly dependent on the load and can in practice be unidirectional
or oscillatory, whereas the HBM waveform is largely defined by the series resistance and
is always unidirectional. The peak current, for a given ESD voltage, is typically an order of
magnitude greater for MM than for HBM.
any air gaps or dielectrics between the device and the ground plane. The series resistance
and inductance in the circuit are the resistance and inductance of the tester, of the spark,
and within the device. These determine the peak current and waveform in the discharge.
Two versions of the CDM test exist (Brodbeck and Kagerer 1998). In the field-induced
charged device model (FICDM) test, the device is placed on a metal plate that can be
raised to the test voltage. A thin layer of insulator separates the device from the plate
(Figure 3.8). After the plate has been raised to the test voltage, a metal “pogo pin” is
touched to the device leg to initiate discharge. The FICDM simulates real-world events
well but is time-consuming and expensive to perform.
The first and second peak current, rise time, full-width, and half-maximum height dura-
tion of the waveform are key parameters (Figure 3.9). The bandwidth of the measuring
oscilloscope also affects the measurement results, and tables are given for use with 1 and
6 GHz oscilloscopes.
The system is calibrated using a small or a large verification module in the place of the
device. These are metal discs, with the smaller giving a capacitance of 6.8 pF ±5% and the
larger giving 55 pF ±5%. The parameters for small (Table 3.9) and large (Table 3.10) verifi-
cation modules using a 1 GHz oscilloscope are given here as examples. The test condition
Fast digital
storage
oscilloscope
50 Ω input
50 Ω coaxial cable
Metal ground plane
High-voltage plane
Current
Ip
0.9 Ip
FWHM
0.5 Ip
0.1 Ip
tr Time
Ip2
Figure 3.9 ANSI/ESDA/JEDEC JS-002 field-induced charged device model calibration waveform.
62 3 Electrostatic Discharge–Sensitive (ESDS) Devices
TCxxx denotes the voltage stress level but does not correspond to the actual voltage applied
to the field plate. This is because the plate voltage must be adjusted to give the required
peak current values for each test condition during the calibration process.
The CDM test is quite sensitive to changes in experimental conditions including atmo-
spheric humidity and contamination or oxidation of the verification module surfaces. At
higher 1 kV and above, corona discharge may limit the voltage achievable on the device
before discharge.
After test, devices are classified according to their failure level (Table 3.11).
A socketed device test (SDM) is easier to perform, but the result does not correlate well
with FICDM results. In the SDM method, the device is placed in a test socket and subjected
to discharges via a relay network. This technique has advantages for devices that have heat
sink fins or uneven packages and so do not lend themselves to FICDM test.
Table 3.12 ESD model parameters compared (Gieser and Ruge 1994) assuming a
10 Ω load in place of the device.
At a time when the HBM, CDM, and MM models were under development, Gieser and
Ruge (1994) compared the characteristics of the models (component values and character-
istic waveforms may be slightly different from current model versions).
Geiser and Ruge noted the following:
● In the HBM discharge, most of the energy stored on the capacitor is dissipated in the
series resistance.
● The series resistor and load form a voltage divider circuit. The applied voltage results in
a voltage on the device that could stress oxide layers.
● If the stray component impedance is low, then the HBM discharge current is largely deter-
mined by the series resistance.
● The peak current achieved by HBM is achieved at much lower ESD voltage in MM and
CDM. For the same ESD voltage, MM and CDM currents are about 10 times the HBM
value.
● The rise and decay times for CDM are around two orders of magnitude faster than for
HBM and MM.
64 3 Electrostatic Discharge–Sensitive (ESDS) Devices
● Assuming a resistive load, peak power dissipated in the load increases with the square
of current. For CDM, peak power is in the kW range. MM is lower, in the 500 W region.
HBM peak power is only in the watt region.
● The higher capacitance of MM stores about 20 times the energy of the CDM case for the
same test voltage.
● In practice, the real HBM, MM, and CDM waveforms achieved in any tester are influ-
enced by interactions with stray “parasitic” components, as well as the actual load
impedance, that may be far from purely resistive.
According to Smedes (2009), at 2 kV the HBM peak current is 1.3 A, and the duration about
200 ns. At 200 V MM, the peak current is 3.8 A with a pulse width around 30 ns. A 500 V
CDM ESD has peak current around 2–10 A and duration around 1 ns, depending on the
device capacitance.
Failure modes for HBM and MM are generally found in the on-chip protection cir-
cuits, whereas CDM failures are usually gate oxide damage. The duration of the CDM
pulse (<1 ns) is often less than required to trigger the protection circuits (Amerasekera
and Duvvury 1995). In CDM ESD the direction of discharge current in internal circuit
elements may be opposite to that of HBM and MM as the charge is stored internally
and discharges to the outside world. The protection circuits may not be designed for this
situation.
There is good correlation between HBM and MM damage stress levels, with the HBM
withstand voltage about 12 times greater than MM (Kelly et al. 1995). More generally HBM
withstand voltage is 10–20 times greater than MM. There is little correlation between CDM
damage stress levels and the other models.
change in leakage current, and a leakage current measurement can be done after each test
to evaluate device damage.
Correlation between TLP, HBM, and MM results have been studied and some empirically
derived correlations established, but these may not be universally applicable.
Application of TLP to CDM and system-level ESD tests is less clear. A very fast transmis-
sion line pulse (VFTLP) method has been proposed to address this for CDM, in which the
pulse width is reduced to 1–5 ns with rise time around 200 ps. This has provided improved
results. Nevertheless, VFTLP remains a stress applied between two pins, whereas CDM is
a single-pin connection stress, and the device internal current flow is therefore different.
3.2.11 The Relation Between ESD Withstand Voltage and ESD Damage
There are essentially two main types of ESD susceptibility in devices – energy and volt-
age susceptibility (Baumgartner ESD TR50.0-03-03). Nevertheless, there are various ESD
parameters that affect the likelihood of damage.
These parameters may have different effects on different components, and real-world ESD
can have widely varying characteristics. Because of this, devices that have similar ESD with-
stand voltages may have different susceptibility to damage from real-world ESD. Changing
the ESD parameters can have a large impact on ESD risk even though the ESD source volt-
age is not changed. As an example, a 250 V CDM device may risk damage if it contacts a
metal part when the voltage difference between them is 250 V. The damage is often due to
the fast high-current transient that flows in the charged device ESD event causing an over-
voltage of either device internal part. If, instead of contacting metal, the device contacts
a high-resistance material, the ESD current is greatly reduced and the discharge slowed
by the material resistance. In this circumstance, the device may not be damaged even at
voltages far higher than 250 V.
In practice, it is doubtful whether it is appropriate to measure ESD withstand thresholds
in terms of ESD source voltage as this is only indirectly related to the parameter caus-
ing damage to the device. The damage sustained is often due to energy dissipation in the
device or discharge current, and protection devices divert and must withstand these cur-
rents. The current at which a device fails is often approximately equivalent for HBM and
MM (Amerasekera et al; 2002). Even in CDM the failures are due to high current levels,
which cause internal voltage stresses (Brodbeck and Kagerer 1998).
In a practical situation, the source voltage is only one of several factors affecting the ESD
current and energy in the device. Because of this, it is difficult to directly relate measured
ESD withstand voltages to ESD risk in a factory situation. Nevertheless, at the time of writ-
ing, it seems unlikely that specification of ESD susceptibility in any terms other than source
voltage will become generally accepted in the foreseeable future.
66 3 Electrostatic Discharge–Sensitive (ESDS) Devices
Reiner (1995) suggested that a high-resistance silicon melt ball could form around the site
of gate oxide damage caused by CDM. This caused unstable leakage currents of the order
1 μA at 5 V. Subsequent heating could convert some of this material into higher conductivity
crystalline silicon and change to permanent damage.
Hellstrom (1986) analyzed ESD failures in bipolar and metal-oxide silicon (MOS) devices.
They observed latent damage through “birth,” growth, and completion to component fail-
ure.
Baumgartner commented that most experts agree that there are latent defects caused by
ESD in some technologies, but they disagree as to whether the device goes on to become a
latent failure. Latent damage is likely to occur in a narrow window of ESD strength between
no damage and complete failure. This window may be narrow, and the probability of ESD
strength lying within the window may be correspondingly small (Beal et al. 1983). Failure
analysis data suggests that latent failures are statistically insignificant. Infant failures of
components may be more likely than latent failures.
Tunnecliffe et al. (1992) subjected 32 enhancement mode and 32 depletion mode field
effect transistors (FETs) to HBM ESD of ±100 and ±200 V. Latent damage was identified, but
they concluded that the risk to reliability was small as the latent failure window is narrow. It
is perhaps more likely that the device would either be catastrophically damaged or remain
undamaged.
In contrast, Gammill and Soden (1986) found that latent failures occurred in comple-
mentary metal-oxide silicon (CMOS) integrated circuits (ICs) both as a field failure and in
a life test experiment. McKeighan et al. (1986) found that reversible charge–induced surface
inversion caused CMOS switches to fail to turn on when addressed. A factor influencing this
behavior included use of a highly insulating ceramic package. The charge-induced failures
occurred during board handling or assembly or were field induced.
Some workers have found that degraded devices could fail parametrically or functionally
when subjected to ESD (Taylor and Woodhouse 1986; Enoch et al. 1983) and recover after
further events, becoming fully functional. Shorting of a p-n junction did not recover on
further ESD event, but a dielectric short due to ESD could recover. Krakauer and Mistry
(1989) found that an ESD event lower than the threshold for catastrophic failure could cause
charge carriers to be injected into oxide to become trapped there. This could cause voltage
threshold shift and changes to MOSFET drain current.
Cook and Daniel (1993) found that ESD applied to ASIC power pins could cause latent
damage that compromised device latch-up immunity.
Anand and Crowe (1999) studied latent failures in Schottky diodes that caused loss of
microwave transceiver sensitivity. The microwave diodes they studied had a junction area
of 5–8 μm diameter. Using a reverse leakage test, they identified damaged devices that had
higher leakage current than good devices. Faulty devices were not identified by the standard
test procedure because the device specification did not include reverse leakage current.
Chen et al. (2009) found that one effect of reverse applied HBM and MM ESD on GaN
light-emitting diode (LED) devices caused an increase in leakage paths that could have a
cumulative and latent damage effect.
Smedes and Li (2003) reported that ESD can cause latent damage as well as permanent
damage to interconnect structures, reducing electrothermomigration lifetimes by a factor
of 100. At low level, ESD metal leads can melt and in cooling change their grain boundary
68 3 Electrostatic Discharge–Sensitive (ESDS) Devices
structure. This can then lead to reduction in electromigration lifetime due to an increase in
the metal resistance.
Laasch et al. (2009) found that ESD protection diodes stressed with multiple pulses suf-
fered cumulative latent damage in the form of a metal alloy front that progressed with each
pulse and eventually could short the p-n junction.
Sylvania (2009) gave examples of partial failure of an LED in a series-connected LED
“coupon” due to damage to one of the LEDs. In the examples, one of the LEDs in the string
is degraded causing reduced light output while remaining sufficiently conductive to allow
the remaining LEDs to function. The operating life of the unit may be dramatically reduced.
Dhakad et al. (2012) described detection of a functional CDM ESD failure in an advanced
CMOS IC where no obvious physical damage was present. The failing device was identified
as a single transistor that had suffered gate oxide charge trapping degradation.
Overall, these studies suggest that while latent damage has been demonstrated, it appears
to be usually a low-probability risk. Nevertheless, it can be a significant concern, espe-
cially in high-reliability applications or where a high cost or severe consequences make
a risk of failure unacceptable. They also suggest that life test studies of advanced technol-
ogy high-speed semiconductor devices stressed with HBM or CDM at levels just below their
threshold values might give some insight into their reliability in the field (Duvvury C. Pri-
vate communication).
Some component pins are particularly difficult to protect with on-chip protection net-
works due to their specialist function. Examples include radio frequency (RF) and high
data rate pins or some types of analog input/output pins. ESD protection networks increase
capacitance and reduce the quality factor of RF circuits. Added capacitance can degrade
the performance of RF circuits. Simple low-capacitance protection networks must be used
in these capacitance-sensitive circuits. Higher-frequency operation generally reduces the
allowed “capacitance budget” for ESD protection. The capacitance of HBM ESD protec-
tion networks typically increases with ESD withstand voltage, so reducing the capacitance
budget can force a reduction in ESD withstand voltage of the protection network.
Not all pins on a device have the same ESD withstand voltage. For example, the Burr
Brown DAC8043 data sheet gives detailed information about the ESD performance of the
device (Burr Brown 1993). This device has mixed digital and analog functions. The digital
pins are stated as withstanding ±2500 V HBM. The analog pins are stated as withstanding
only 1000 V HBM. The data sheet goes on to say that two pins, VREF and RFB , “show some
sensitivity.” What this means is not clear, but presumably they did not survive 1000 V HBM.
On-chip ESD protection typically uses a system of clamp circuits to divert ESD current
away from the sensitive internal circuitry (Figure 3.10). The clamp components can be var-
ious types of components such as p-n or p-i-n diodes, bipolar or MOS transistors, or silicon
controlled rectifiers (SCRs) (Amerasekera et al. 2002).
Operation depends on the design clamp style used. If clamp 2 is a reverse diode, then
only clamp 1 and the power supply ESD clamp turn on for positive stress with respect to
ground. If clamp 2 is a breakdown NMOS device with characteristics shown in Figure 3.11,
then it is clamp 2 that offers protection (Duvvury C. Private communication). For a negative
transient, clamps 2 and 4 and the power supply clamp turn on.
The clamp devices typically have V-I curves that have some, or all, of the elements
shown in Figure 3.11. During an ESD event, the voltage across the device increases until
the device turns on at V t1 and the current through the device increases. If the current
reaches a level I t2 , a second breakdown and thermal failure can occur. So, ESD protection
clamp circuits are also ESD susceptible as there is limit to the current and energy they can
handle. The ESD withstand voltage in an HBM test or failure during a real ESD event is
determined by the maximum current flowing during the test rather than the source ESD
voltage.
+Ve
Output supply
ESD driver ESD
I/O clamp clamp
pin 1 3 Power
supply
Rs Rin ESD
ESD ESD Device clamp
clamp clamp input
2 Output 4 circuit
driver Ground
Figure 3.10 Typical clamp arrangement used in on chip ESD protection of an input/output (I/O)
device pin.
70 3 Electrostatic Discharge–Sensitive (ESDS) Devices
Holding
voltage
Vsp
Vt1
Turn on
It1
Source: Based on IEC 61340-5-2/TS: 1999, component data sheets, and other sources.
3.3 ESD Susceptibility of Components 71
given in IEC 61340-5-2:1998 with additions from other sources and data sheets. While this
gives a general idea of the range of HBM ESD withstand voltage that might be expected for
various technologies, it cannot be relied on for specific components.
In the 1980s and 1990s the Reliability Analysis Center published compilations of ESD
susceptibility of a range of components including discrete and passive devices (Reliability
Analysis Centre 1989a,b, 1995).
Thermal secondary breakdown, metallization and interconnect melting, and bulk break-
down are dependent on energy dissipated in the discharge through the device.
Breakdown of thin dielectric layers, gaseous arc discharge, and surface breakdown are
due to excess voltage stress of some part of the device. In a high-resistance capacitive struc-
ture such as a MOS capacitor or MOSFET gate, conduction of sufficient charge into the
structure can raise the voltage to the breakdown level.
According to Analog Devices (2014), most ESD failures occur due to conductor or resistor
melting, dielectric damage, or junction damage or contact spiking.
Conductor or resistor melting occurs in thin metal or polysilicon interconnects and
thin-film, thick-film, or polysilicon resistors. The high current flowing due to ESD causes
local heating in the conductor or resistor material, leading to fusing and open circuit. This
type of damage is commonly due to HBM ESD because a charged person represents a
high energy ESD source. For thick- and thin-film resistors, partial melting and a change in
resistance value are possible. This can lead to parametric failure of the IC. These damage
mechanisms are largely found by HBM ESD testing rather than any vast evidence gathered
from field failures. In contrast, for CDM ESD the field failures correlate well with observed
damage for sensitive devices (Duvvury C. Private communication).
Dielectric breakdown occurs when a thin insulating dielectric layer such as silicon diox-
ide or nitride is stressed by an applied voltage that exceeds its time dependent dielectric
breakdown strength. This results in punch through of the dielectric. results. This type of
failure is typical of CDM damage because the fast rise time results in high voltages occur-
ring within the chip. The high current that flows on breakdown can lead to formation of a
melt filament of silicon.
Junction damage and contact spiking can occur when a p-n junction is subjected to
avalanche breakdown and secondary breakdown. Avalanche breakdown occurs when
the reverse breakdown voltage of a reverse biased p-n junction is exceeded. Secondary
breakdown can then occur at a point where the junction material is sufficiently hot. The
high current that flows is channeled through this site and causes high adiabatic local
heating, increasing current flow. Ultimately, the silicon can melt if the temperature exceeds
1415 ∘ C. If heating is sufficient, adjacent contact metal can melt and migrate. The junction
is shorted by resistive material.
3.4 Some Common Types of ESD Damage 73
3.4.3 MOSFETs
According to Infineon (2013), there are three basic ESD failure modes in MOSFETs. These
are junction damage, gate oxide damage, and metallization burnout. The most common
failure in HBM is junction damage caused by an ESD transient of sufficient ESD energy
and duration. A damaged junction often shows high reverse bias leakage or a short circuit.
Gate oxide damage is, however, the main category of ESD damage. This occurs when
the gate is subjected to ESD causing the gate oxide to break down. This happens when
the voltage across the thin gate oxide exceeds the breakdown field strength. For thin oxide
74 3 Electrostatic Discharge–Sensitive (ESDS) Devices
layers, the breakdown voltage can be small, although for short duration ESD, a much higher
ESD voltage may be required to achieve breakdown.
Another common failure mode is that ESD may lead to trapped charge in the gate
oxide. This can cause a shift in the gate threshold voltage and functional failure. This can
sometimes anneal itself after a few hours, and so trapped charge effects are not always
permanent (Duvvury C. Private communication).
Metallization burnout can also occur, although this is often a secondary effect occurring
after initial junction or gate oxide breakdown.
A MOS gate is a capacitive structure. If the gate is charged to over the breakdown voltage
of the dielectric, breakdown occurs, and the gate is likely to be damaged. An HBM or MM
test of this type of structure simply takes the charge in one capacitor (the ESD source) and
dumps it into another (the MOS gate) (Figure 3.12). The final voltage on the gate can be
calculated if the gate capacitance is known. Hence, the MM or HBM ESD withstand can be
estimated if the MOSFET gate capacitance and breakdown voltage is known (International
Rectifier n.d, Application note AN-986).
The initial charge QESD in the ESD source capacitance CESD at a voltage V ESD is
QESD = CESD VESD
During the ESD event, this charge is shared between the source capacitance and gate
capacitance Cg in parallel. The charge Qbr required to achieve the gate breakdown
voltage V gbr is
Qbr = (CESD + Cg )Vgbr
The threshold ESD voltage at which the gate reaches this breakdown voltage can be found
when QESD equals Qbr .
CESD VESD = (CESD + Cg )Vgbr
(CESD + Cg )
VESD = Vgbr
CESD
So, the effective ESD withstand voltage is heavily dependent on the gate capacitance. If
Cg ≪ CESD , then V ESD is around the gate breakdown voltage V gbr . For small devices, this
can be a few volts or tens of volts. If Cg ≫ CESD , then V ESD is around V gbr Cg /CESD .
device and another object, and electrostatic field–induced charged device ESD occurs when
they touch. Control of electrostatic fields is normally essential where charged device ESD
is a concern.
Some types of components have closely spaced unpassivated conductors on a surface or
with an air gap. Components containing isolated conductors at small distances apart will
have voltage differences induced by changing electrostatic fields. These voltage differences
could produce breakdown and discharges between them. Sparking can cause damage to the
electrodes and transfer of material across the gap. Electromigration of materials under high
electrostatic fields between conductors is also possible.
Photomask reticles used in semiconductor device manufacture have this type of structure.
The mask consists of very fine metallic tracks with very fine (μm or nm) etched gaps on the
surface of a quartz substrate. Electrostatic fields, for example due to charge on the reticle
enclosure, cause voltage differences between the metallic tracks. These voltage differences
between tracks can lead to breakdown, ESDs, and damage to the reticle at voltages below
the Paschen minimum breakdown voltage (Rider and Kalkur 2008; Rider 2016). Damage to
reticles leads to damaged components in the semiconductor wafer and reduction of compo-
nent yield. Englisch et al. (1999) proposed a “canary” reticle test structure that could be used
to simulate and evaluate the effects of electrostatic field exposure during reticle handling.
Other devices such as surface acoustic wave (SAW) filters that have closely spaced elec-
trodes can also suffer damage (Mil HDBK 263B). Wallash and Honda (1997) reported that
magnetoresistive (MR) recording heads could be damaged by electrostatic fields, due to
breakdown of small gaps by voltage differences induced by the field.
Sangameswaran et al. (2009) found that MEMS capacitive switches could suffer both
dielectric and air breakdown.
Wallash and Levitt (2003) found that ESD could still occur at very small submicron gaps
less than the Paschen minimum by field emission. This could be a threat for photolitho-
graphic reticles, magnetic recording heads, MEMSs, and field emission displays.
point of the semiconductor. This can result in changes in junction characteristics or short
circuits. This is called thermal secondary breakdown.
This failure mechanism is dependent on the power dissipated in the junction. Junctions
with high breakdown voltage and low leakage current may be more susceptible to ESD
damage. Where hot spots do not develop to failure, migration of material under the elec-
tric field may cause a partially developed filament short circuit. This may increase leakage
current.
When forward biased, the junction has low voltage drop, and the power developed by
the ESD current is spread through the body of the device, and failure of the junction is less
likely.
For most junction transistors, the emitter-base junction is more susceptible than the
collector-base junction due to smaller junction dimensions. Junction FETs that have a high
gate-source breakdown voltage and low leakage can be particularly susceptible. Schottky
diodes and Schottky TTL components are more sensitive to ESD because they have thin
junctions and metal present that may be carried through the junction.
Not all p-n junctions are susceptible to ESD damage. Transient suppressor diodes, zener
diodes, power rectifier diodes, power bipolar transistors, and thyristors can be very robust
to ESD.
dead, short, or with low forward or reverse voltage. Avago Technologies’ InGaN LEDs are
classified as Class 1x and Class 2 (HBM ESD withstand voltage 250–4000 V).
Talbot (1986) found that the ESD withstand voltage of nine various color LEDs mostly var-
ied from 4–15 kV, although two “low current” types showed damage at 100–200 V. Devices
that had been subjected to reverse breakdown often functioned normally in the forward
direction. There was a drop in light output dependent on the ESD voltage. Failure mecha-
nisms included junction burnout, nitride punch through, and metallization burnout.
Chen et al. (2009) found four different effects of reverse applied HBM and MM ESD
on GaN LED devices, affecting leakage paths in the devices. Low-level ESD below 650 V
reduced leakage but above 700 V leakage was increased by three to five orders of magnitude.
Further application of ESD caused unstable behavior. The device still glowed in forward
bias. When V-I measurements were made, the device was destroyed. They concluded that
the increase in leakage paths could have a cumulative and latent damage effect.
3.4.10 MEMS
MEMS devices are used in modern electronics systems. Some unprotected MEMS have been
found to been highly sensitive to ESD damage (Walraven et al. 2000, 2001). Sangameswaran
et al. (2008, 2009, 2010a,b) found HBM withstand voltages as low as 40 V. Traditional elec-
trical characterization was found to overestimate ESD robustness. Capacitive switch MEMs
devices were found to suffer dielectric or air breakdown and mechanical failures in response
to ESD. ESD failures were often detectable only through mechanical tests. Atmospheric gas,
pressure, temperature, and humidity were all influential to ESD-induced breakdown.
and energy levels in the discharge. Discharge to connector or device pins gave the greatest
threat, with the least threat for ESD to long PCB traces between devices. The threat was
worse for PCBs with no ground plane and less for PCBs with ground and power planes.
Floating PCBs represented a lesser threat than those connected to an external ground.
Often modules and assemblies are designed with protection network circuits on the con-
nector pins emerging to the outside world. These by design will give a level of protection
against ESD entering the assembly by this route. They will not, however, protect against
ESD arising from direct contact with other parts of the assembly.
Shaw and Enoch (1985) found that type 74 373 octal latch ICs failed in the voltage range
250–2500 V due to charged PCB ESD transients, compared to 1000 to >4000 V for HBM and
CDM testing. They found the charged PCB damage voltages were not related to CDM or
HBM failure levels but were dependent on the capacitance of the PCB.
Olney et al. (2003) found that ICs that are relatively robust as components could be dam-
aged by charged board ESD (see Figure 2.19). As PCB capacitance is higher than device
capacitance, the energy stored for a given voltage is much greater. They commented that
charged board ESD damage could be mistaken for EOS damage and that this should be
considered before a conclusion of EOS is drawn in failure analysis. They gave guidelines on
how to avoid charged board ESD failures.
Paasi et al. (2003) studied the behavior of charged PCBs to evaluate ESD sensitivity of
devices on the board with respect to discharge current and energy. They concluded that
energy-sensitive devices could be at risk at lower charge or voltage levels when on a PCB
than before assembly. They based their evaluation on HBM and MM device data. They
pointed out that the capacitance, voltage, and stored energy of a PCB vary with movement
thorough the manufacturing line. For a given PCB charge, a low capacitance high voltage
condition gives higher stored energy and ESD current than a high capacitance low voltage
condition and can represent a higher risk of ESD damage.
Gärtner et al. (2014) concluded that a CBE was more likely in a PCB production line than
a charged device ESD event. As peak discharge current is a critical parameter in CDM dam-
age, they used it to evaluate CBEs. They showed that the peak current is not significantly
higher than for a single device at the same voltage level. The total charge transferred is,
however, significantly higher giving a comparatively long (10–50 ns) discharge. This seemed
to represent a higher stress to devices than charged device ESD. Case studies have shown
that devices can be damaged at a voltage level at which devices would have survived in the
CDM test. Stresses encountered in the real world are typically lowered due to reduced PCB
capacitance due to greater gap between the PCB and ground. Stresses are also reduced for
field-induced stresses for a parallel external field compared to a perpendicular field.
connection to cables. They may be assembled into the system in a facility that has little or
no ESD control.
Many system components or modules are housed within polymer enclosures or potted
with only connectors or flying leads exposed to the outside world. While these enclosures
prevent direct contact and ESD to the internal circuitry, the enclosure may become itself
charged during handling and transport, especially if packaged within plastic packaging.
The internal circuits can attain high voltages by induction especially when packaging is
removed. ESD can then occur when connection is made to the connectors or flying leads.
3.5.2 The Relationship Between System Level Immunity and Component ESD
Withstand
There is, however, some overlap between the topics of system-level ESD immunity and com-
ponent ESD susceptibility. It has often been assumed, mainly erroneously, that system-level
immunity depends on component ESD withstand (Ind. Co. 2010a, 2012). This has led to
system designers placing ESD withstand requirements on components used in the system.
3.5 System-Level ESD 81
System ESD failures can be classified as hard or soft. Hard failures are those in which
physical damage has occurred that is not recoverable. Often, system-level failures are soft
failures that represent upset or temporary malfunction of the system in a way that is recov-
erable.
Ind. Co. (2010a) found that there is rarely correlation between system-level ESD immu-
nity and component HBM ESD withstand level. Component ESD withstand data is obtained
with the component in an unpowered state and represents hard failures of the component.
System failures are often soft failures and occur with the system in a powered state. The
ESD waveforms used in obtaining component ESD withstand data have significant differ-
ences with those used in system ESD immunity evaluation, and the test environment is very
different. In practice, system ESD immunity is dependent on the system design including
PCB design and on-PCB ESD protection, as well as individual component response to ESD
transients. Component ESD tests do not reflect the conditions that occur for devices during
system-level ESD events.
There may be some components, e.g. those that are directly connected to external
connector pins, that do require some ESD withstand capability. The Industry Council on
ESD Target Levels discussed this area in its White Paper 3 and proposed a system-efficient
electrostatic discharge design (SEED) approach to understanding system-level ESD needs
regarding ESD robust components.
design can be achieved by providing the interactions between the ESD stress and the com-
plete system design are understood and addressed. The SEED approach recognizes that
The SEED approach was proposed as a better system design philosophy, optimizing the
balance between system cost and performance and reducing design effort. The philosophy
and approach to implementation of ESD robust system designs and state of the art was
further described in Part 2 of White Paper 3 (Ind. Co. 2012). A more comprehensive dealing
of SEED is presented by Duvvury and Gossner (2015).
References
Amerasekera, A. and Duvvury, C. (1995). ESD in Silicon Integrated Circuits, 1e. Wiley. ISBN:
0471954810.
Amerasekera, A., Duvvury, C., Anderson, W. et al. (2002). ESD in Silicon Integrated Circuits, 2e.
Wiley. ISBN: 0 470 9871 8.
Analog Devices (2014). Reliability Handbook. UG-311 Rev. D. Available from: http://www
.analog.com/media/en/technical-documentation/user-guides/UG-311.pdf [Accessed: 10th
May 2017]
Anand, Y. and Crowe, D. (1999). Latent failures in Shottky barrier diodes. In: Proc. of EOS/ESD
Symp. EOS-21, 160–167. Rome, NY: EOS/ESD Association Inc.
Ashton, R. (2008). Reliability of IEC 61000-4-2 ESD testing on components. E E Times Available
from: http://www.eetimes.com/document.asp?doc_id=1273265 [Accessed: 10th May 2017]
Avago Technologies. (2017) Premium InGaN LEDs - Safety Handling Fundamentals ESD
Electrostatic Discharge Application Note 1142. Available from: http://www.avagotech.com/
docs/AV02-0160EN [Accessed: 10th May 2017]
Baumgartner, B. (n.d.) ESD TR50.0-03-03. Voltage and Energy Susceptible Device Concepts,
Including Latency Considerations. Rome, NY, EOS/ESD Association Inc.
Beal, J. Bowers, J. Rosse, M. (1983) A study of ESD latent defects in semiconductors. In: Proc.
EOS/ESD Symp. EOS-5. Rome, NY, EOS/ESD Association Inc.
Boxleitner, W. (1990). ESD stress on PCB mounted ICs caused by charged boards and
personnel. In: Proc. EOS/ESD Symp. EOS-12, 54–60. Rome, NY: EOS/ESD Association Inc.
References 83
Brodbeck, T. and Kagerer, A. (1998) Paper 4A.7). Influence of the device package on the results
of CDM tests – consequences for tester characterization and test procedure. In: Proc.
EOS/ESD Symp, 320–327. Rome, NY: EOS/ESD Association Inc.
Brundrett, G.W. (1976). A review of the factors influencing electrostatic shocks in offices. J.
Electrostat. 2: 295–315.
Burr Brown. (1993) DAC8043 CMOS 12-Bit serial input multiplying digital to analog converter.
www.ti.com.cn/cn/lit/ds/symlink/dac8043.pdf [Accessed: 10th May 2017]
Chase, E.W. (1982). Electrostatic discharge (ESD) damage susceptibility of thin film resistors
and capacitors. In: Proc. EOS/ESD Symp. EOS-4, 13–18. Rome, NY: EOS/ESD Association
Inc.
Chen, N.C., Wang, Y.N., Wang, Y.S. et al. (2009). Damage of light-emitting diodes induced by
high reverse-bias stress 97-B-016. J. Cryst. Growth 311: 994–997.
Cook, C. and Daniel, S. (1993). Characterisation of new failure mechanisms arising from power
pin stressing. In: Proc. EOS/ESD Symp. EOS-15, 149.
Danglemeyer, T. (1999). ESD Program Management, 2e. Clewer: Springer. ISBN: 0-412-13671-6.
Department of Defense. 1994 Military Handbook. Electrostatic discharge control handbook for
protection of electrical and electronic parts, assemblies and equipment (excluding
electrically initiated explosive devices) MIL-HDBK-263B 31st July 1994
Dhakad, H., Gossner, H., Zekert, S., Stein, B., Russ, C. (2012). Paper 3A.1. Chasing a latent
CDM ESD failure by unconventional FA methodology. Proc. EOS/ESD Symp.
Duvvury C, Gauthier R. (2011) IC Technology Scaling Effects on Component Level ESD. Ch. 6
in Industry Council on ESD Target Levels (2011) White paper 1: A case for lowering
component level HBM/MM ESD specifications and requirements. Rev. 3.0. Available from:
www.esdindustrycouncil.org. [Accessed: 10th May 2017]
Duvvury, C. and Gossner, H. (2015). System Level ESD Co-Design. Wiley – IEEE. ISBN:
978-1-118-86190-5.
Duvvury C., Ashton R., Righter A., Eppes D., Gossner H., Welsher T. and Tanaka M, (2012)
Discontinuing Use of the Machine Model for Device ESD Qualification. In Compliance
Magazine, July 2012. Available from: https://incompliancemag.com/article/discontinuing-
use-of-the-machine-model-for-device-esd-qualification [Accessed 6th March 2019]
Englisch, A., van Hesselt, K., Tissier, M., and Wang, K.C. (1999). CANARY: a high-sensitive
ESD test reticle design to evaluate potential risks in wafer fabs. In: Proceedings of the SPIE,
19th Annual Symposium on Photomask Technology, BACUS, vol. II, 886–892. Available from:
http://dx.doi.org/10.1117/12.373381 [Accessed: 10th May 2017].
Enoch, R.D., Shaw, R.N., and Taylor, R.G. (1983). ESD sensitivity of NMOS LSI circuits and
their failure characteristics. In: Proc. EOS/ESD Symp. EOS-5, 185–197. Rome, NY: EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2006) Trends in Semiconductor Technology and ESD Testing.
White paper II. ISBN: 1-58537-116-5
EOS/ESD Association Inc. (2009). ESD S5.6-2009. ESD Association Standard Practice for
Electrostatic Discharge Sensitivity Testing – Human Metal Model (HMM) – Component Level.
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016) ESD Association Electrostatic Discharge (ESD) Technology
roadmap – revised 2016. Available from: https://www.esda.org/assets/Uploads/docs/
2016ESDATechnologyRoadmap.pdf [Accessed: 10th May 2017]
84 3 Electrostatic Discharge–Sensitive (ESDS) Devices
EOS/ESD Association Inc., JEDEC. (2012). ANSI/ESD STM5.2-2012. ESD Association Standard
Test Method for Electrostatic Discharge (ESD) Sensitivity Testing – Machine Model
(MM) – Component Level. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc., JEDEC. (2017). ANSI/ESDA/JEDEC JS-001-2017. ESDA/JEDEC
Joint Standard for Electrostatic Discharge Sensitivity Testing – Human Body Model
(HBM) – Component Level. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc., JEDEC. (n.d.) ANSI/ESDA/JEDEC JS-002-2014. ESDA/JEDEC joint
standard for electrostatic discharge sensitivity testing – Charged Device Model (CDM) – Device
Level. ISBN: 1-58537-276-5 Rome, NY, EOS/ESD Association Inc.
Gammill, P.E. and Soden, J.M. (1986). Latent failures due to electrostatic discharge in CMOS
integrated circuits. In: Proc. EOS/ESD Symp. EOS 8, 75–80. Rome, NY: EOS/ESD Association
Inc.
Gärtner, R., Stadler, W., Niemesheim, J., Hilbricht, O. (2014) Do Devices on PCBs Really See a
Higher CDM-like ESD Risk? In: Proc. EOS/ESD Symp. Rome, NY, EOS/ESD Association Inc.
Gieser, H. (2002). Test Methods. In: ESD in Silicon Integrated Circuits, 2e (eds. A. Amerasekera
and C. Duvvury). Wiley. ISBN: 0 470 9871 8.
Gieser, H. and Ruge, I. (1994). Survey on electrostatic susceptibility of integrated circuits. In:
Proc. ESREF Symp, 447–455. Rome, NY: EOS/ESD Association Inc.
Hellstrom, S., Welander, A., and Eklof, P. (1986). Studies and revelation of latent ESD failures.
In: Proc. EOS/ESD Symp. EOS-8, 81–91. Rome, NY: EOS/ESD Association Inc.
Industry Council on ESD Target Levels (2010a) White paper 3: System Level ESD Part I:
Common Misconceptions and Recommended Basic Approaches. Rev. 1.0 http://www
.esdindustrycouncil.org/ic/en/documents/7-white-paper-3-system-level-esd-part-i-
common-misconceptions-and-recommended-basic-approaches [Accessed: 10th May 2017]
Industry Council on ESD Target Levels (2010b) White paper 2: A case for lowering component
level CDM ESD specifications and requirements. Rev. 2.0. http://www.esdindustrycouncil
.org/ic/en/documents/6-white-paper-2-a-case-for-lowering-component-level-cdm-esd-
specifications-and-requirements [Accessed: 10th May 2017]
Industry Council on ESD Target Levels (2011) White paper 1: A case for lowering component
level HBM/MM ESD specifications and requirements. Rev. 3.0. Available from: http://www
.esdindustrycouncil.org/ic/en/documents/37-white-paper-1-a-case-for-lowering-
component-level-hbm-mm-esd-specifications-and-requirements-pdf [Accessed: 10th May
2017]
Industry Council on ESD Target Levels (2012) White paper 3: System Level ESD Part II:
Implementation of Effective ESD Robust Designs. Rev. 1.0 http://www.esdindustrycouncil
.org/ic/en/documents/36-white-paper-3-system-level-esd-part-ii-effective-esd-robust-
designs [Accessed: 10th May 2017]
Infineon. (2013). Preventing ESD Induced Failures in Small Signal MOSFETs. Application Note
AN-2013-04 V2.0 http://www.infineon.com/dgdl/Infineon+-+Application+Note+-+
PowerMOSFETs+-+Small+Signal+-+Preventing+ESD+Induced+Failures+in+Small+
Signal+MOSFETs.pdf?fileId=db3a30433dfcb54c013dfe36f38d0295 [Accessed: 10th May
2017]
International Electrotechnical Commission. (1999) IEC 61340-5-2/TS:1999. Electrostatics – Part
5-2: Protection of electronic devices from electrostatic phenomena - User guide. Geneva, IEC.
References 85
Paasi, J., Salmela, H., Tamminen, P., and Smallwood, J. (2003). ESD sensitivity of devices on a
charged printed wiring board. In: Proc. EOS/ESD Symp. EOS-25, 143–150. Rome, NY:
EOS/ESD Association Inc.
Reiner, J.C. (1995). Latent gate oxide defects caused by CDM ESD. In: Proc. EOS/ESD Symp.
EOS-17, 311–321. Rome, NY: EOS/ESD Association Inc.
Reliability Analysis Centre (1989a) Electrostatic Discharge susceptibility data of microcircuit
devices Vol. I. VZAP-2 Reliability Analysis Center P.O. Box 4700 Rome, NY 13440-8200.
Reliability Analysis Centre (1989b) Electrostatic Discharge susceptibility data of
discrete/passive devices Vol. II. Reliability Analysis Center P.O. Box 4700 Rome, NY
13440-8200.
Reliability Analysis Centre (1995) Electrostatic Discharge susceptibility data of discrete/passive
devices. VZAP-95 Reliability Analysis Center 201 Mill St, Rome, NY 13440.
Rider, G.C. (2016). Electrostatic risk to reticles in the nanolithography era. J.
Micro/Nanolithogr. MEMS MOEMS 15 (2): 023501.
Rider, G. C., Kalkur, T. S. (2008) Experimental quantification of reticle electrostatic damage
below the threshold for ESD. Proc. SPIE 6922, Metrology, Inspection, and Process Control for
Microlithography XXII, 69221Y
Sangameswaran, S., De Coster, J., Linten, D. et al. (2008). ESD reliability issues in
michromechanical systems (MEMS): a case study on micromirrors. In: Proc. EOS/ESD
Symp, 3B.1-1–3B.1-9. Rome, NY: EOS/ESD Association Inc.
Sangameswaran, S., De Coster, J., Scholz, M. et al. (2009). A study of breakdown mechanisms
in electrostatic actuators using mechanical response under EOS-ESD stress. In: Proc.
EOS/ESD Symp, 3B.5-1–3B.5-8. Rome, NY: EOS/ESD Association Inc.
Sangameswaran, S., De Coster, J., Linten, D. et al. (2010a). Investigating ESD sensitivity in
electrostatic SiGe MEMS. J. Micromech. Microeng. 20 (5): 055005.
Sangameswaran, S., De Coster, J., Chermin, V. et al. (2010b). Behaviour of RF MEMS switches
under ESD stress. In: Proc. of the EOS/ESD Symp, 443–449. Rome, NY: EOS/ESD Association
Inc.
Shaw, N.R. and Enoch, R.D. (1985). An experimental investigation of ESD damage to integrated
circuits on printed circuit boards. In: Proc. EOS/ESD Symp. EOS-7, 132–140. Rome, NY:
EOS/ESD Association Inc.
Smallwood, J.M. (2004). Static electricity in the modern human environment. In:
Electromagnetic Environments and Health in Buildings (ed. D. Clements-Croome). Taylor &
Francis. ISBN: 0 415 31656 1.
Smedes, T. (2009). ESD testing of devices, ICs and systems. Microelectron. Reliab. 49: 941–945.
Smedes, T. (2011) Machine Model – Correlation between HBM and MM ESD. Ch. 3 in Industry
Council on ESD Target Levels (2011) White paper 1: A case for lowering component level
HBM/MM ESD specifications and requirements. Rev. 3.0. Available from: http://www
.esdindustrycouncil.org/ic/en/documents/37-white-paper-1-a-case-for-lowering-
component-level-hbm-mm-esd-specifications-and-requirements-pdf [Accessed: 10th May
2017]
Smedes, T. and Li, Y. (2003). Paper 2A.6). ESD phenomena in interconnect structures. In: Proc.
EOS/ESD Symp. EOS-25, 108–115. Rome, NY: EOS/ESD Association Inc.
References 87
Stadler, W., Brodbeck, T., Gartner, R., and Gossner, H. (2006). Cable discharges into
communication interfaces. In: 2006 Electrical Overstress/Electrostatic Discharge Symposium,
144–151. IEEE.
Stadler W., Niemesheim J., Stadler A., Koch S., Gossner H. (2017) Paper 3A1. Risk Assessment
of Cable Discharge Events. In: Proc. EOS/ESD Symp. EOS-39. Rome, NY, EOS/ESD
Association Inc.
Sylvania. (2009) ESD protection for LED systems. Application note. LED093. Available from:
http://assets.sylvania.com/assets/documents/ESD.a9e90e9d-c91e-4ea3-9ebe-0be6e5570cd7
.pdf [Accessed: 30th October 2017]
Szwarc, J. (2008) ESD Sensitivity of Precision Chip Resistors Comparison between Foil and
Thin Film Chips. Vishay Available from: http://www.vishaypg.com/docs/60106/esdsensi
.pdf [Accessed: 10th May 2017]
Talbot, J.W. (1986). The effect of ESD on III-Vmaterials. In: Proc. EOS/ESD Symp. EOS-8,
238–245.
Talebzadeh, A., Patnaik, A., Moradian, M. et al. (2015, 2015). Dependence of ESD charge
voltage on humidity in data Centers: part I-test methods. ASHRAE Trans. 121: 58.
Taminnen, P., Sydänheimo, L., Ukkonen, L. (2014) Paper 9A.2 ESD Sensitivity of 01005 Chip
Resistors and Capacitors In: Proc. EOS/ESD Symp. Rome, NY, EOS/ESD Association Inc
Taylor, R.G. and Woodhouse, J. (1986). Junction degradation and dielectric shorting: two
mechanisms for ESD recovery. In: Proc. EOS/ESD Symp. EOS-8, 92. Rome, NY, EOS/ESD
Association Inc.
Tunnecliffe, M., Dwyer, V., and Campbell, D. (1992). Parametric drift in electrostatically
damaged MOS transistors. In: Proc. EOS/ESD Symp. EOS-14, 112–120. Rome, NY, EOS/ESD
Association Inc.
Viheriäkoski, T., Kokkonen, M., Tamminen, P., Kärjä, E., Hillberg, J., Smallwood, J. (2014) 4B.2
Electrostatic Threats in Hospital Environment. In: Proc. EOS/ESD Symp. EOS 36
Vishay (2011) Resistor Sensitivity to Electrostatic Discharge (ESD). Vishay Document 63129.
Available from: http://www.vishaypg.com/docs/63129/esd_tn.pdf [Accessed: 10th May
2017]
Wallash, A.J. (1996). Field induced charged device model testing of magnetoresistive recording
heads. In: Proc. EOS/ESD Symp. EOS-18. 4B.2, 8–13.
Wallash, A. and Honda, M. (1997). Field induced breakdown ESD damage of Magnetoresistive
recording heads. In: Proc. EOS/ESD Symp. EOS-19, 382–385. Rome, NY: EOS/ESD
Association Inc.
Wallash, A., Levitt, L. (2003) Electrical breakdown and ESD phenomena for devices with
nanometer-to-micron gaps. Proc. SPIE 4980 Reliability, Testing, and Characterization of
MEMS/MOEMS II, 87 doi: http://dx.doi.org/10.1117/12.478191
Walraven, J.A., Soden, J.M., Tanner, D.M. et al. (2000). Electrostatic discharge/electrical
overstress susceptibility in MEMS: a new failure mode. In: Proceedings of the SPIE 2000, vol.
4180, 30–39.
Walraven, J. A., Soden, J. M., Cole, E. I., Tanner, D. M., Anderson, R. R. (2001). Paper 3A.6
Human Body Model, Machine Model, and Charged Device Model ESD testing of surface
micromachined microelectromechanical systems (MEMS). In: Proc. EOS/ESD Symp. EOS-23.
Rome, NY, EOS/ESD Association Inc
88 3 Electrostatic Discharge–Sensitive (ESDS) Devices
Wang, A.Z.H. (2002). On-Chip ESD Protection for Integrated Circuits. Klewer Academic Press.
Williams, T. (2001). EMC for Product Designers, 3e. Newnes. ISBN: 0 7506 4930 5.
Wilson, N. (1972). The static behaviour of carpets. Text. Inst. Ind. 10 (8): 235.
Further Reading
EOS/ESD Association Inc. (2013b). Technical Report for the Protection of Electrostatic Discharge
Susceptible Items – System Level Electrostatic Discharge (ESD) Simulator Verification ESD
TR14.0-02-13. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2014a). Technical Report for Electrostatic Discharge (ESD) Sensitivity
Testing – Very Fast – Transmission Line Pulse (TLP) – Round Robin Analysis ESD TR5.5-03-14.
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2014b). Technical Report for Relevant ESD Foundry Parameters for
Seamless ESD Design and Verification Flow ESD TR22.0.01-14. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2014c). Technical Report for ESD Electronic Design Automation
Checks ESD TR18.0-01-14. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015a). Standard Practice for Electrostatic Discharge Sensitivity
Testing – Near Field Immunity Scanning - Component/Module/PCB Level ANSI/ESD
SP14.5-2015. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015b). Technical Report for ESD Process Assessment Methodologies
in Electronic Production Lines – Best Practices used in Industry ESD TR17.0-01-15. Rome, NY,
EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016a). Standard Test Method for Electrostatic Discharge (ESD)
Sensitivity Testing – Transmission Line Pulse (TLP) – Component Level ANSI/ESD
STM5.5.1-2016. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016b). Technical Report for Electrostatic Discharge Sensitivity
Testing – Charged Board Event (CBE) ESD TR25.0-01-16. Rome, NY, EOS/ESD Association
Inc.
EOS/ESD Association Inc. (1999) ESD Association Technical Report - Can Static Electricity be
Measured? ESD TR50.0-01-99
International Organization for Standardization. (2008) Road vehicles – Test Methods for
electrical disturbances from electrostatic discharge. ISO 10605:2008/Amd.1:2014(en)
King, W.M. (1979). Dynamic waveform characteristics of personnel electrostatic discharge. In:
Proc. of the EOS/ESD Symp. EOS-1, 78.
Lin, D.L., Strauss, M.S., and Welsher, T.L. (1987). On the validity of ESD threshold data
obtained using commercial human-body model simulators. In: Proceedings of the 25th
International Reliability Physics Symposium, 77. IEEE.
Lin, N., Liang, Y., Wang, P., and Pelc, T. (2014). Evolution of ESD process capability in future
electronics industry. In: 15th Int. Conf. Elec. Packaging Tech, 1556–1560. IEEE.
McAteer, O.J., Twist, R.E., and Walker, R.C. (1980). Identification of latent ESD failures. In:
Proc. of the EOS/ESD Symp. EOS-2, 54–57. Rome, NY: EOS/ESD Association Inc.
McAteer, O.J., Twist, R.E., and Walker, R.C. (1982). Latent ESD failures. In: Proc. of the
EOS/ESD Symp. EOS-4, 41–48. Rome, NY: EOS/ESD Association Inc.
Paasi, J., Smallwood, J., and Salmela, H. (2003) Paper 2B4). New methods for the assessment of
ESD threats to electronic components. In: Proc. of the EOS/ESD Symp, 151–160. Rome, NY:
EOS/ESD Association Inc.
Smallwood J, Paasi J. (2003) Assessment of ESD threats to electronic devices. VTT Research
Report No BTUO45-031160
Smallwood J., Taminnen P., Viheriaekoski T. (2014) Paper 1B.1. Optimizing investment in ESD
Control. In: Proc. of EOS/ESD Symp. EOS-36. Rome, NY, EOS/ESD Association Inc.
90 3 Electrostatic Discharge–Sensitive (ESDS) Devices
Strauss, M.S., Lin, D.L., and Welsher, T.L. (1987). Variations in failure modes and cumulative
effects produced by commercial human-body model simulators. In: Proc. of EOS/ESD Symp.
EOS-9, 59–63.
Viheriäkoski T, Peltoniemi T, Tamminen T, (2012) Paper 4A3. Low Level Human Body Model
ESD. In: Proc. of EOS/ESD Symp. Rome, NY, EOS/ESD Association Inc.
Vinson, J.E. and Liou, J.J. (1998). Electrostatic discharge in semiconductor devices: an
overview. Proc. IEEE 86 (2): 399–420.
Voldman, S. (2009). ESD Failure Mechanisms and Models. Wiley. ISBN: 978-0-470-1137-4.
Vollman, S., Hui, D., Warriner, L. et al. (1999). Electrostatic discharge (ESD) protection in
silicon-on-insulator (SOI) CMOS thechnology with aluminium and copper interconnects in
advanced microprocessor semiconductor chips. In: Proc. of the EOS/ESD Symp. EOS-21,
105–115. Rome, NY: EOS/ESD Association Inc.
91
Habit: “settled or regular tendency or practice, especially one that is hard to give up”
(Oxford Dictionary 2017)
For effective electrostatic discharge (ESD) control, we need to set up practices that reduce
the ESD risk to our electrostatic discharge–sensitive (ESDS) devices to an acceptable level.
These practices can be ways of working but also involve using certain ESD-protective equip-
ment to reduce ESD risk. If we can establish and maintain these practices so well, they
become a habit, and then our ESD control program is likely to remain effective.
Many ESD threats occur while handling ESDS devices, for example during assembly pro-
cesses. To protect against this sort of threat, we can set up a permanent or temporary ESD
protected area (EPA) in which the ESD threats are controlled so that we can handle the
devices and assemblies conveniently and relatively free from ESD risk.
Other ESD risks occur when an ESDS device is stored or transported in an uncontrolled
unprotected area (UPA) where static electricity can build up and ESD sources arise. In this
situation, we use ESD-protective packaging to enclose and protect the ESDS device from
damage.
Of course, we need to know how effectively we are adhering to our established ESD pro-
gram practices. Also, equipment is likely to fail from time to time under the day-to-day
wear and tear it experiences during use. We need to detect when equipment fails, falls out
of specification, or requires maintenance. For these reasons, a habit of checks and tests is
required.
Finally, we need to be sure that everyone who is concerned with ESD control or must use
the provisions of the ESD control program understands what they must do and not do. They
may need to know what equipment to use and even how to check it is functioning correctly.
They need to know what procedures to follow. It can be of great benefit if they watch out
for noncompliance and correct them as they go. So, training will be needed to make sure
they have the knowledge they need to fulfill these roles.
This chapter explores these habits, the reasons for them, and how to decide what should
be included in our habitual practices. Many of them are incorporated into the design and
specifications of equipment and materials designed for use in ESD control. We can enact
The ESD Control Program Handbook, First Edition. Jeremy M Smallwood.
© 2020 John Wiley & Sons Ltd. Published 2020 by John Wiley & Sons Ltd.
92 4 The Seven Habits of a Highly Effective ESD Program
some aspects of the “habits” by specifying special equipment and materials that are used in
EPAs where ESD risks are carefully controlled.
There are two key strategies that form the basis of successful ESD protection practice.
● Handle unprotected ESDS devices only in an area in which the ESD risks are reduced to
an insignificant level.
● In uncontrolled (unprotected) areas, protect ESDS devices by enclosing them within
ESD-protective packaging that protects the ESDS devices from ESD risks.
These two strategies should be applied to all aspects of handling, storage, and transport
of ESDS.
ESDS devices can be of many types and forms, from minute individual semiconductor
devices such as transistors, diodes, or integrated circuits, to printed circuit boards, mod-
ules, or system components. They usually contain semiconductor devices of some sort,
although other types of device (e.g. some types of resistors and capacitors) can have some
ESD sensitivity. ESDS devices and their failure modes are discussed in more detail in
Chapter 3.
The key factor in identification of an ESDS device is understanding that the item would
be at some risk of ESD damage if handled in an UPA without precautions. An item is an
ESDS device if it satisfies two criteria.
It follows that if either of these criteria is not met, the item can be considered as not
being ESDS.
In building an electronic system, the risk of ESD damage and susceptibility to ESD often
change considerably with build state. Let us take as an example the construction of a simple
product that consists of a printed circuit board (PCB) within a housing. Many of the com-
ponents that go onto the PCB are likely to be ESDS. The populated PCB is also likely to be
ESDS and should be handled as such. However, once built into and protected by its housing,
the final product may well be quite immune to normal ESD sources due to the protective
barrier provided by the housing. In many cases, electromagnetic compatibility (EMC) reg-
ulations may require testing and demonstration of ESD immunity of the working system
for market acceptance. There may, however, be some residual ESD susceptibility dependent
on its design, for example susceptibility to ESD to connector pins, e.g. from connection of
charged cables.
4.4 Habit 1: Always Handle ESDS Components Within an EPA 93
In some cases, the build stage at which the product can be considered no longer suscep-
tible to ESD damage is not clear. This will then have to be decided by some evaluation of
the ESD risks and susceptibility.
An item such as a PCB may be subject to ESD occurring to almost any part of it that may
be touched by a person, tool, or other conductive item. If the same PCB is encapsulated or
potted to become a module or subassembly, the number of ways in which ESD can occur to
it, and hence ESD risk, is much reduced as the encapsulation can form an effective barrier
to ESD to most parts of the assembly. This may not mean that the module is immune to ESD
damage or needs no ESD protection. The module may have flying leads or connectors to the
PCB within. The module may be susceptible to ESD to these leads or connections, unless
designed for immunity. Tribocharging of the module surface can lead to high induced volt-
ages occurring on the PCB within, and these can discharge if the leads or connector pins
touch something conductive.
A product that is at a built state at which it is no longer ESDS can become susceptible
to ESD damage again if modified or disassembled in some way. For example, a desktop
computer would not normally be considered ESDS once it is fully assembled to the state
at which the user would normally receive it. If, however, the covers are removed, PCBs
containing ESDS parts may become accessible to touch, and appropriate ESD control pro-
cedures should be in place while operating on these PCBs. Once the covers are replaced
and no ESDS parts are accessible, the system can again be considered not susceptible to
ESD damage.
in response to electrostatic attraction from the high electrostatic fields near these surfaces.
When taking off fleece or other clothing made of man-made fibers, we may hear the crack-
ling of small discharges, and, in low light conditions, we may see them as tiny flashes.
While we are insensitive to ESD and electrostatic fields, as we have seen, many electronic
components are not. We must protect them against electrostatic effects, either by handling
them within an EPA or by enclosing them in ESD-protective packaging. This section is
about EPAs and the habits we need to develop to make them effective.
We can make EPAs in many different forms. They can be temporary, or they can be fixed
facilities. A field service kit can be used to provide a temporary EPA for field use. A fixed
facility may be a single workstation or may enclose many workstations or a whole room or
work area (Figure 4.1). A machine, or part of a machine, where unprotected ESDS devices
are handled may need to be part of an EPA.
So, what do we need to have to make an effective EPA? There are two basic aspects.
The need for a clear boundary is because we must be clear whether we are inside the
EPA or outside it. Outside the EPA boundary, the ESDS device should never be taken out
It is often a good idea to restrict entry into the EPA to certain points rather than having
an extended open boundary. This can minimize the amount of boundary marking required.
A physical barrier (temporary or permanent) can go a long way to reduce unauthorized
entry into an EPA.
EPA entry points should be marked with clear signage showing personnel approaching
the entrance that they are about to enter an EPA (Figure 4.2). Signs should be obvious and
eye-catching. Signs at around eye-level may be more easily noticed than signs positioned at
low level or above an entrance.
It can also be a good idea to post signs visible to personnel leaving the EPA, warning that
they are about to leave the EPA.
There is a risk of ESD occurring whenever an ESDS device contacts another conductor at
a different voltage. The strength of the ESD occurring can be minimized by decreasing the
likely peak current, energy, and charge transferred in any ESD that may occur. The way we
evaluate and control ESD risk is further discussed in Habits 2–5.
One common way of deciding what ESD measures should be implemented is to adopt
the requirements of an ESD control standard (see Chapter 6). These normally list a range
of control measures that address the most common ESD threats. A detailed evaluation of
ESD risks is often not attempted. This approach has the advantage that it is easily achieved
and needs the minimum of expertise. The ESD control program that results may be easily
accepted by customers and found compliant with the standard by auditors.
This approach can, however, have disadvantages.
● ESD control measures may be included that address risks that do not in fact exist within
the processes and facility.
● ESD risks may exist that are not addressed by the standard ESD control measures.
● The ESDS device is unusually sensitive, having ESD withstand voltage lower than the
design withstand voltage level of the standard.
For these reasons, it may be necessary or preferable to undertake some level of evaluation
of the ESD risks in a process and facility in preparation for determining the ESD control
measures that are required. With fuller knowledge of the specific ESD risks, ESD control
measures can be specified that address these more effectively, efficiently, and completely.
Determination of ESD protection measures is further discussed in Chapter 10.
4.4.7 Who Will Decide What ESD Protection Measures Are Required?
An ESD control program that has no one leading and responsible for it is likely to fail
through lack of attention and maintenance. Someone will need to develop and implement
the ESD control program. Someone will need to document it, maintain it, test it, and train
people to work within it.
4.5 Habit 2: Where Possible, Avoid Use of Insulators Near ESDS 99
These functions may of course involve several people taking different roles and respon-
sibilities. Some companies have a committee of personnel within a site and, for a multisite
organization, on different sites, working on a company ESD program.
Nevertheless, it is advisable to have a person responsible for coordinating, implementing,
and maintaining the ESD control program at each site. In the main ESD control standards
current at the time of writing, it is a requirement to have such a person, known as the ESD
Coordinator. They do not have to do everything themselves, but they do have to make sure
it gets done. So, they must have the necessary authority, backup, and resources to fulfill
the role.
In some organizations, a committee rather than a person fulfills this role. Usually, the
tasks required in implementing and maintaining the ESD control program are delegated
wholly or in part to other people in the organization. For example, specially trained techni-
cians may do routine testing of the equipment in the EPA, and a designated trainer may do
some or all the training.
● The voltage is unlikely to be the same at every point on the surface of the insulator. When
voltage differences occur, electrical currents will not flow quickly enough in response to
prevent the differences.
● If ESD from the insulator occurs, only a small amount of the charge and energy stored
on the insulator may be delivered during the discharge.
dQ/dt = I
Charge
generation R or ρ C or ε
(current
source)
Typically, we wish to prevent a risk by either keeping the voltage produced below a cer-
tain value at which some hazard may occur or making sure that any voltages arising are
dissipated quickly before they have chance to cause a hazard.
In an EPA where manual handling is the norm, things don’t usually happen very quickly.
If there are no strong continuous electrostatic charging mechanisms present, it will often be
sufficient to make sure that charge and voltages produced during normal contact between
materials is dissipated within a few seconds. So, materials will often be acceptable if the
decay time 𝜏 = 𝜌𝜀r 𝜀0 is of the order of a few seconds. Given that many materials have
relative permittivity 𝜀r around 2–3, and 𝜀0 = 8.8 × 10−12 Fm−1 , then setting an upper limit
of material volume resistivity of 100 GΩm (1011 Ωm) gives theoretical decay times around
1.8–2.6 seconds. This approximately correlates with the upper limits usually chosen for
ESD-protective packaging materials, although these are usually expressed as surface or vol-
ume resistance rather than resistivity (see Section 1.7).
Table 4.1 gives some examples of essential and nonessential insulators that can be found
in many facilities. Some types of insulator are clearly essential (e.g. the PCB and compo-
nents). Others are less easily categorized (e.g. parts of test jigs and production papers). In
some cases, these may be essential, and in others nonessential.
Metal
plate Metal
plate E=V/d
V=1000V Field E 0V
10 000 2m
5 000 1m
2 500 50 cm
1 500 30 cm
500 10 cm
125 2.5 cm
50 1 cm
If the field meter plate is removed, leaving the field meter in position, the electrostatic
field lines terminate at the field meter instead of the plate (Figure 4.5). We assume that the
electrostatic field meter is constructed as an earthed conductor at 0 V. The field lines focus
on the field meter and so a higher nonuniform electrostatic field exists at the field meter in
this case. Because of this, we can no longer assume that as we change the distance between
the field meter and the plate, the field will vary with 1/d.
In practice, the assumption that the field decreases as 1/d often gives reasonable agree-
ment with experiment (Stadler et al. 2018). Many field meters are calibrated to give a reading
of surface voltage when at a set distance of 2 or 2.5 cm (1 in.) from a large flat metal sheet at
a given voltage. In Figure 4.6 the voltage reading taken with the field meter at 2 cm distance
from a flat metal plate at 1 kV. The field (voltage) readings at other distances are shown as
a percentage of the field (voltage) reading at other distances. It is interesting to note that
the field at 30 cm distance is only 7% of the value at 2 cm distance. Some standards require
4.5 Habit 2: Where Possible, Avoid Use of Insulators Near ESDS 103
Distance d
Metal
plate
Field E
V=1000V
Figure 4.5 The electrostatic field between a field meter and a charged plate varies with distance.
Figure 4.6 Electrostatic field meter reading variation with distance from a charged metal plate at
0.5 kV, expressed as a percentage of the value measured at 2.5 cm distance.
that any insulator that has a surface voltage >2 kV measured in this way should be kept at
least 30 cm from any ESDS device. This practice ensures that the electrostatic field from the
charged insulator experienced by the ESDS device is less than 7% of its close-up value.
Alternatively, standards often give a limit to the electrostatic field at the position of the
ESDS, e.g. 5 kV m−1 . As most field meters are calibrated in terms of voltage, it is not imme-
diately obvious how to measure this. However, any field meter calibrated in terms of voltage
can be easily calibrated to find this field limit, using a metal plate raised to a set voltage to
give the field as in Figure 4.5. For example, if a field meter is set up at 0.020 m from a metal
plate, and 100 V is applied to the plate, the electrostatic field is 100/0.020 = 5000 V m−1 . The
reading observed on the meter with this setup will depend on the design operational and
calibration conditions of the meter. This does not matter, and it does not matter what the
104 4 The Seven Habits of a Highly Effective ESD Program
number is – it represents the field of 5 kV m−1 . Anything giving a reading above this value
is producing a field above 5 kV m−1 , and anything less is a field <5 kV m−1 .
the ESDS is not likely to be moved into a position where the field may be significant.) If the
insulator is not likely to be handled or moved or become charged, the risk of a field arising
may be negligible.
The ESD risk can in principle be controlled in several ways.
The ESD control measure should be chosen as appropriate to the situation. Some possible
examples are given in Table 4.3.
Electrostatic charging of the item should preferably be measured under worst-case
conditions, which usually means under low-humidity (≪30% rh) ambient atmosphere. In
practice, measurements may have to be done under ambient atmospheric conditions, as
a humidity-controlled facility is usually not available. Nevertheless, an initial evaluation
done under ambient condition at higher humidity may give useful first evaluation and
should then be followed up and repeated when the weather conditions give lower humidity.
The question then arises – what level of charging can be considered negligible (Swenson
2012)? Unfortunately, this may not be easy to answer and depends on the withstand voltages
of the ESDS being handled and other factors. For example, if it is charged device model
(CDM) damage to the ESDS device that is of concern and if the voltages produced are lower
than the CDM withstand of the ESDS device, they can be considered negligible. The voltage
induced on a conductor can never exceed the voltage of the electrostatic field source. In
practice, higher voltages may also be negligible, but evaluation of this may be more difficult.
Table 4.3 Some examples of essential insulators and possible ways of dealing with them.
Product components and parts Use an ionizer to reduce charge levels to acceptable value.
Parts of test jigs and fixtures that must Treat with an antistat on regular basis and reduce
be made from insulating materials charging with humidity control.
Use an ionizer to reduce charge levels to acceptable value.
Papers that are required to be present Keep in static dissipative document holders.
and used during the process If required to be removed from holders (e.g. for
annotation), do this on a separate work area a minimum
of 300 mm from workstation where ESDS devices are
handled.
Use computer-based document displays designed to be
ESD safe.
Computing equipment on the Position the equipment on a separate part of the
workstation workstation or well away from the likely position of
ESDS.
106 4 The Seven Habits of a Highly Effective ESD Program
Standards may also give requirements that can be used to evaluate fields from charged
insulators. For example, the IEC 61340-5-1:2016 standard gives requirements that the elec-
trostatic field at the position of the ESDS must be <5 kV m−1 . Also, insulators charged to
>125 V must be kept at least 2.5 cm from the ESDS, and if charged to >2 kV must be kept
>30 cm from the ESDS. If these conditions are fulfilled, the electrostatic fields and voltages
can be considered negligible for the purposes of this standard.
Figure 4.7 Ionizer charge decay curves showing decay time and offset voltage for positive (above)
and negative (below) charge neutralization.
4.7 Habit 4: Ground Conductors, Especially People 107
can be different for positive and negative ions, the charge neutralization rates can also be
different, even for a well-balanced ion stream. For a poorly balanced ion stream the differ-
ence in ion concentration will also contribute to a difference in charge neutralization rate
for each polarity. So, neutralization of one polarity can be significantly slower than the other
polarity. As a result, as Figure 4.7 shows, it can take a significant time for a surface charge to
be neutralized to a low level. In Figure 4.7, the negative polarity voltage is reduced to −100 V
in <5 seconds. The positive polarity voltage requires about double this time to reduce to
+100 V. In an assembly process, it may be necessary to wait several seconds until the charge
has been reduced to a sufficiently low level to reduce ESD risk to an acceptable level.
The voltage decay time also varies with the position of the charged object relative to the
ionizer. How it varies will depend on the type of ionizer used. This can be an important
factor in choosing an ionizer for a process role. Typically, the charge decay time will increase
as the distance from the ionizer increases. This is because the ion concentration in the air
reduces as the ions spread out by mutual repulsion of like charges, and opposite polarity
charges attract and recombine forming neutral particles.
Many ionizer types can also be quite directional in their effectiveness. For example, a
fan ionizer blows ions in one general direction with the fan airstream. Its effectiveness can
dramatically reduce outside the airstream.
Most electrical ionizers exhibit an offset voltage due to a small imbalance of the posi-
tive and negative ion density that they produce. The ionizer offset voltage does not usually
cause any problems in neutralizing insulators. Small charge levels giving a few tens of Volts
on insulators do not usually cause ESD risk except in handling extremely sensitive compo-
nents. Standards often specify a maximum acceptable offset voltage or may leave it to the
user to define an appropriate maximum for their application.
having resistance <1 MΩ (106 Ω). Dissipative is thought to apply to materials having resis-
tance between 1 MΩ and 100 GΩ (1011 Ω). Insulative is thought to apply to materials having
resistance over 100 GΩ (1011 Ω). Beware – these definitions are not universal. Within the
61340-5-1 and S20.20 systems of standards, these terms have standardized definitions only
in some contexts. One is ESD-protective packaging materials (see Chapter 8), where an
insulator is a material having surface or volume resistance greater than or equal to 100 GΩ.
Even in this specific topic, the definition of a “conductive” packaging material has changed
over recent years with updating standards. Because of this lack of clarity, it is unwise to
specify materials in terms of conductive or dissipative. Instead, measurable parameters
such as an acceptable range of surface resistance should be specified.
● Under quasistatic conditions with no current flowing through the conductor, the voltage
will be the same at every point on the surface of the conductor.
● If ESD from the conductor occurs, almost all the charge and energy stored on the con-
ductor could be delivered from the conductor during the discharge
The first point arises from the fact that if a voltage difference occurred, a current would
flow until no voltage difference is present. Equilibrium could be attained relatively quickly.
In practice, the timescale in which it happens depends on the material characteristics,
namely, resistivity and permittivity.
The second point arises from the fact that when a discharge is initiated from the material,
the voltage at the point of discharge changes quickly as charge is conducted away. Voltage
differences then occur across the material that cause currents to flow until the voltage across
the material is again equalized. This can be the point at which charge stored on the material
is exhausted. Because of this, a charged conductor is often a potent source of ESD.
the order of a second or so, which may be acceptable. Higher-capacitance items will need a
lower resistance to ground to keep charge and voltage decay times acceptably short.
This situation may be different in automated processes, partly because continuous charge
generation processes may be present. Machine movements may be much faster than in
manual processes and so shorter charge decay times may be necessary to avoid significant
charge buildup and ESD risk.
need to have equipotential bonding of the conductors in the area. It would be perfectly
possible to have successful control of voltage differences between conductors by equipo-
tential bonding in an aircraft or other situation with no contact with earth. For this reason,
in modern ESD control standards, equipotential bonding is treated equally with grounding.
The term grounding is often used to include equipotential bonding as an alternative to other
grounding methods, meaning connecting an item to the designated ESD ground.
In many EPAs, mains electrical safety earth is present, and many types of equipment are
already connected to it for electrical safety. So, it often is most convenient to use mains
electrical safety earth as the EPA earth for ESD control purposes. All items of noninsulative
EPA materials and equipment are then electrically bonded to this.
In some facilities, mains electrical safety earth may not be available or for some reason it
may be undesirable to bond to this earth. A separate “functional” ground such as an earth
rod sunk into the ground may be used.
It is normally undesirable to have two different and separate grounds present in an EPA.
If this occurs, they may be at different voltages and could become a serious source of ESD
risk. All earths (grounds) in an EPA should be electrically bonded together to make sure
that no significant voltage differences can occur between them.
A basic requirement for grounding personnel is to control body voltage to make sure that
it does not reach a level where damaging ESD could occur when ESDS device are touched.
Current practice is to keep body voltage lower than the human body model (HBM) with-
stand of any ESDS devices that are handled. Current standards aim to safely handle 100 V
HBM devices, and so the body voltage is controlled to less than 100 V. When devices of lower
withstand voltage are handled, body voltage may need to be maintained below a lower level.
Considering the simple electrical model of Figure 4.3 for a person, the main way in which
body voltage can be controlled is by providing a ground path with sufficiently low resistance
from the person’s body to ground through an appropriate grounding system. The two main
grounding systems used are
● Wrist strap connected to the person’s body and to an earth bonding point provided for the
purpose
● ESD control footwear when standing on an appropriately specified ESD control floor
Other systems are occasionally used, with the principles remaining the same. The max-
imum resistance from body to ground is usually specified so that under all circumstances
expected, body voltage generated is kept below a specified level.
It is important to realize that changing the footwear or flooring types for others of the
same resistance would not necessarily give the same body voltage performance as the
charge generation properties would be different. So, once a footwear-flooring combination
has been evaluated and selected, it is necessary to use only this combination, or another
that has likewise been shown to give the required performance. Unfortunately, the
resistance of the footwear and the flooring measured separately have been shown to be a
poor predictor of the footwear and flooring performance in combination (Smallwood et al.
2018).
If several floor types are to be used with the same footwear, then the performance of the
footwear must be evaluated with each type of floor. Similarly, if several types of footwear
are to be used, then each type of footwear must be evaluated in combination with the floors
with which they will be used. Quite different performance can be found with different floor
materials and at different atmospheric humidity (Figure 4.8). Most standard ESD control
programs require the maximum body voltage produced during a walk test to be less than
100 V. This may be reduced when handling components less than 100 V HBM withstand
voltage.
(a)
(b)
Figure 4.8 (a) Footwear 10 MΩ, “dissipative” floor 10 MΩ resistance to ground, 15% rh;
(b) footwear 10 MΩ, “conductive” floor 900 kΩ resistance to ground, 15% rh; (c) footwear 10 MΩ,
“standard vinyl tile” floor, 15% rh; (d) footwear 10 MΩ, “standard vinyl tile” floor, 50% rh. Body
voltage generated using ESD control footwear with different types of flooring. Source: D. E.
Swenson.
from any noninsulative material or object that is placed on it. This may include tools and
components including ESDS devices.
Any isolated (nongrounded) conductor that is placed on the work surface can be expected
to initially be at a different voltage. An ESD event will occur as the conductor approaches
4.7 Habit 4: Ground Conductors, Especially People 115
(c)
(d)
or touches the work surface. If the conductor is part of an ESDS, this gives a risk of charged
device ESD damage. For this reason, where the ESDS devices handled are susceptible to
charged device ESD damage, the work surface should be chosen to have a high surface resis-
tance to limit the peak ESD current. Where the ESDS devices handled are not susceptible
to charge device ESD, metal or low-resistivity surfaces may be used.
Work surfaces are usually either hardwired or connected via earth bonding plugs to the
EPA ground system. A work surface material with point-to-point resistance and resistance
from surface to ground less than 1 GΩ is normally considered adequate for ESD control in
116 4 The Seven Habits of a Highly Effective ESD Program
most EPAs. A minimum point-to-point resistance may be specified for safety or charged
device ESD damage prevention. (See Section 4.7.5.3)
4.7.10.3 Floors
ESD control floors are often provided to give a convenient way of grounding personnel as
well as carts, racks, chairs, and other free-standing equipment on the floor. The ESD control
function of the floor is often misunderstood – it must operate as a system with all the items
that it is intended to ground. So, in specifying the characteristics of a floor, it is necessary
to consider the items that will be grounded by it. Typical grounding systems using the floor
include the following:
● An operator’s body grounded through footwear and flooring
● A cart (trolley) grounded from its surface through the chassis, wheels, and floor
● A rack grounded from the shelf surface through the frame, feet, and floor
● A seat grounded from the seat surfaces through the frame, feet, or wheels and floor
The resistance to ground of the system includes the resistance of all the parts of the ground
path including the item being grounded, the floor, and the resistance of the contact between
them. So, it might be expected that if the resistance of the individual parts of the ground path
are measured, the total resistance to ground over the system would be the sum of the parts.
Unfortunately, this is not generally true, largely because the contact resistance between
the floor and the item being grounded can be higher or lower than expected and can vary
considerably. The contact resistance with the floor generally depends strongly on the area
and pressure of the contact. The contact areas and pressures between the item and the floor
can vary considerably. These pressures and areas are generally quite different from the pres-
sure and area of a measurement electrode standing on the floor. So, the contact resistance
between a shoe, wheel, or equipment foot and the floor are likely to be very different than
that of a measurement electrode. Contamination or coatings on the contacting surfaces (e.g.
dirt or polish) can also make a big difference to the effective contact resistance.
ESD control standards typically specify a maximum resistance from the floor surface to
ground of 1 GΩ. Many do not specify a lower limit, as there is no minimum resistance
requirement for ESD control purposes. In some applications, a minimum floor resistance
to ground may be desirable, e.g. for safety in the presence of high voltages. The minimum
body to ground resistance of the footwear and flooring system should be measured in qual-
ification tests with the person standing on the ESD control floor wearing the footwear with
which it will be used. This should be considered an essential part of the floor specification
process.
Similarly, to a first approximation, the resistance to ground from the surface of an item
grounded through feet or wheels through the floor can be assumed to be the sum of the
resistance from surface to feet (wheels) and the floor surface to ground. In practice, this is
often not correct for various reasons, including that the contact between feet (wheels) and
floor surface can have significant resistance that is difficult to predict.
Nevertheless, if the resistance to ground of an item that is grounded via a floor is required
to be below a certain value, the floor resistance to ground should also be below that value.
It is often best to select a lower resistance from floor surface to ground to give some margin
for added resistance from contact resistance. For example, if the resistance from body to
4.7 Habit 4: Ground Conductors, Especially People 117
ground of a person standing on a floor is required to be below 35 MΩ, the floor should be
selected to give installed resistance to ground less than 35 MΩ. In practice, selecting a target
value of, say, 10 MΩ gives some “headroom.”
It is perfectly possible to have an effective EPA that does not have an ESD control floor,
if one is not needed for grounding personnel or equipment. For example, a single worksta-
tion EPA in which a standing operator is grounded via a wrist strap may not need an ESD
control floor.
Two such workstations nearby in the same room would effectively be two separate EPAs
with uncontrolled space between them (Figure 4.9). An operator needing to transport an
ESDS from one workstation to the other would need to consider using ESD-protective pack-
aging to protect the ESDS. The operator would not be grounded when moving between the
workstations, and this would lead to ESD risk. In contrast, if an ESD floor is provided, link-
ing the two workstations, the operator can remain grounded when moving between the
workstations. The two workstations can now be part of the same EPA, and moving an ESDS
from one to the other does not take it through uncontrolled space. The ESD risk is greatly
reduced, and the handling of the ESDS is made more convenient.
Figure 4.9 (left) Two EPA workstations separated by a UPA, (right) joined as single EPA using ESD
control floor.
118 4 The Seven Habits of a Highly Effective ESD Program
Wheels and feet are often prone to contamination from an accumulation of dust and dirt.
This increases the resistance to ground over time and may ultimately cause it to exceed the
specified resistance requirements. Regular cleaning may be required to bring the equipment
back within specification.
Drag chains are sometimes used to ground carts, but these can be particularly prone to
dirt collection and can be unreliable.
An example of a carts is shown in Figures 7.9.
4.7.10.5 Seats
It is a common misconception that ESD control seats are intended to ground personnel
sitting in them. Although seats are sometimes used to do this, many are not designed for this
purpose. They are generally designed to prevent the seat from becoming highly charged and
an ESD risk (Figure 4.10). ESD control seats also reduce the charging of personnel sitting
on the seat.
Like carts and racks, seats designed for EPA use must have a continuous conducting path
from the various parts through the legs and feet or wheels to the floor.
Wheels and feet are often prone to contamination from an accumulation of dust and dirt.
This may increase the resistance to ground over time and may ultimately cause it to exceed
the specified resistance requirements. Regular cleaning may be required to bring the equip-
ment back within specification.
4.7.10.6 Tools
Ordinary tools often have insulating handles and metal bits or other parts that are electri-
cally isolated. These can reach a high voltage by tribocharging or induction and cause risk
of ESD to any ESDS devices that they may contact.
Hand tools designed for EPA use typically have noninsulative handles that electrically
connect the bit and any other metal parts to the user’s hand. Any charge developed on the
tool is dissipated safely to the user’s hand and via their body and personal grounding to
ground.
If a glove is worn, the glove must also be noninsulating to allow the tool to be grounded
through the glove to the operator’s hand (Figure 4.11).
Typically, the important characteristics for an ESD-protective tool include the resistivity
of the material that may contact the ESDS devices and the resistance to ground via the nor-
mal ground path. A surprisingly high resistance to ground may be tolerable. For example,
a tool that has capacitance 10 pF would have a charge decay time around one second if
grounded with a resistance to ground of 1011 Ω. This would in most cases be sufficient to
ensure insignificant ESD risk.
If low CDM ESD withstand voltage devices are to be handled, the resistivity of the mate-
rial of any part of a tool that may contact an ESDS may need to be sufficiently high to limit
ESD currents during any contact discharge (Wallash 2007).
Figure 4.11 A hand tool designed for EPA use is grounded via the user’s hand. If a glove is used, it
must also be noninsulative to maintain the ground path.
120 4 The Seven Habits of a Highly Effective ESD Program
eliminate the possibility that an ESDS device may become charged when handled due to
tribocharging. The amount of charging of any item handled will depend on the choice
of material used in the outer surface of the glove that comes into contact with the items
handled.
Figure 4.13 (left) ESD control garments should cover the clothing of the arms and body, (right)
incorrectly worn.
122 4 The Seven Habits of a Highly Effective ESD Program
Some types of garments are designed to maintain contact with the wearer’s skin, usually
at the wrists. This can be an effective way of grounding the garment material via the wearer’s
body (which must always be grounded).
An ESD garment should not contain any exposed ungrounded conductors of any signifi-
cant size and capacitance that could possibly become charged and a source of ESD risk.
ESD training. So, ESD training can be an excellent investment. ESD training is discussed in
greater depth in Chapter 12. An overview is given here.
For those involved in implementing ESD control, training and education can elevate the
topic from “magic” to application of sound engineering principles. This helps them lay the
foundation of an effective and efficient ESD control program through understanding ESD
controls and systems.
touch certain items or to avoid certain actions such as placing cleaning equipment in certain
places or unplugging ground points to plug in cleaning equipment.
Visitors may need a simple set of instructions as to what they may or may not do, as well
as instruction on using personal grounding equipment or garments.
Maintenance and facilities personnel may need specific instructions on how to undertake
maintenance and other work in EPAs without causing ESD risk. There may be contention
between ESD requirements and safety requirements (e.g. electrician’s use of insulating
footwear) that requires specific instructions to resolve. Safety requirements should always
take precedence over ESD requirements.
For nonstandardized items, test methods and criteria must be defined during the ESD
program planning process. Nonstandard test methods and pass criteria can also be used
for items that are covered by standard methods and criteria. Modern ESD control stan-
dards such as IEC 61340-5-1 and ANSI/ESD S20.20 allow nonstandard test methods to be
used, providing correlation of the results of the nonstandard test method with the standard
test method are demonstrated (see Chapter 6). In most cases, it may be easier to adopt the
standard test method.
Different test methods will usually give different results for various reasons. So, use of a
different test method will usually require changing the pass criteria. A specified pass cri-
terion should be applied only to results obtained using the test method with which it was
intended to be used or a test method that has been demonstrated to reliably give the same
results.
● The likely consequence and ESD risk to ESDS products associated with failure
● Likely permanence and ruggedness of the item
● Risk of performance change through contamination or other factors
● Failure history found by experience
● Risk of accidental human intervention (e.g. unplugging ground cord)
● Known item lifetime limitations
● A wrist strap is essential for grounding a seated person who handles ESDS devices. The
consequences of failure would be that the person would be ungrounded and a serious ESD
risk to the ESDS handled. Because of this, wrist straps are normally tested each working
day before use. Some organizations that handle high-cost items or require low ESD failure
rates test even more frequently. Some organizations use an automated test system (e.g.
turnstile access control) to exclude personnel from the EPA unless they pass wrist strap
and ESD footwear tests.
● An ESD control floor that has been installed for some years in a clean area and has been
regularly tested and found to give stable performance might have the frequency of testing
reduced accordingly. A newly installed floor, or one in an area subject to dirt or other
contamination, might need to be tested frequently to monitor performance changes. This
would be especially important if it were used to ground personnel handling high-value
ESDS devices or where a low failure rate is required.
● A permanently installed workstation surface mat that has hardwired ground connections
and is in a clean process area might need relatively infrequent testing. A workstation mat
that is grounded by plug-in connectors to a mains electrical point might need more fre-
quent testing to ensure it has not been inadvertently disconnected. Similarly, workstation
surface in a process that has a high risk of contamination would need frequent testing.
4.12 Handling Very Sensitive Devices 129
● An ESD control chair grounded through wheels to the floor might need frequent testing
due to the risk of grounding failure due to contamination buildup on the wheels. This
would be especially true in a dusty area. In a clean room facility, a longer time between
tests could be adequate.
It is not surprising that many of the aspects of ESD control described in this chapter form
the basis of modern ESD control standards. These have been crystallized over many years
in systems of requirements for ESD control equipment and procedures. The requirements
of two ESD control standards, IEC 61340-5-1 and ANSI/ESD S20.20, prevalent at the time
of writing, are discussed in Chapter 6.
The ESD control standards evolve over time, reflecting developments in ESD control
practice. When preparing for compliance with a standard, obtain copies and refer to the
standards for up-to-date compliance requirements.
At the time of writing, standard ESD control programs are designed to handle devices
with ESD withstand voltage down to 100 V HBM and 200 V CDM (IEC 61340-5-1:2016,
ANSI/ESD S20.20-2014). For devices down to about 500 V HBM and 250 V CDM (Industry
Council on ESD Target Levels 2011), it is relatively easy to implement an effective ESD
control program using standard ESD control measures alone. These basic ESD control
measures can be summarized as
As time goes on, more organizations are handling components and assemblies with ESD
withstand below this level and some below 100 V HBM. These very sensitive components
are sometimes known as “Class 0” devices. The 2016 ESD Roadmap (EOS/ESD Association
2016) predicts that the proportion of components in this range is likely to increase into
the future (see Chapter 13). Handling these components requires a greater level of care
and may require additional nonstandard ESD control measures. Multiple redundant control
measures may also be used to ensure that control is maintained if a single measure fails.
When handling low ESD withstand voltage devices, specific ESD risks may need to be
evaluated and addressed with specific control techniques. The technical requirements of
standard ESD control programs may need to be tightened.
130 4 The Seven Habits of a Highly Effective ESD Program
As an example, for handling 50 V HBM components, the body voltage developed on per-
sonnel should be limited to 50 V. In addition, it becomes more important that the body
voltage developed during normal activities is verified (by ESD control product qualification
and/or compliance verification). This means that the wrist band cords may need to have
specification of a reduced upper limit of resistance. It becomes even more important to
ensure that body voltage developed when grounded by ESD control footwear and flooring
remains below 50 V under all working conditions. For reduced CDM withstand voltage,
the electrostatic fields and voltages developed on insulators near the ESDS, and voltages
developed on the ESDS itself, may need to be controlled to a lower voltage level.
For handling low CDM ESD withstand devices, it becomes important that voltages devel-
oped on the device by induction, conduction, or triboelectrification remain low. At the same
time, the possibility of contact between the ESDS and highly conductive materials should be
eliminated where possible. Materials contacting the ESDS should, where possible, be static
dissipative (>10 kΩ surface resistance as defined in current ESD packaging standards; see
Section 8.5).
Standard ESD control programs largely address the most common ESD sources, namely,
charged personnel, charged metal objects, and charge devices. Other ESD sources occur
that are not necessarily well controlled by standard ESD control measures.
PCBs, assemblies, and modules can become charged and then discharge when they touch
another conductor. The ESD susceptibility of these items is not usually tested. Some studies
have found that damage previously thought to be due to electrical overstress can result from
charged board ESD or charged cable ESD (Olney et al. 2003). Cables, PCBs, and assemblies
can have rather high capacitance due to their size and so can store relatively high energy
prior to discharge.
Charged cable ESD can arise when a cable is connected to a PCB, module, or assembly
at a different voltage. This can occur if the cable is charged or the ESDS is charged or both.
Cables can become charged by triboelectrification by contact with external surfaces (work
surfaces or equipment) or packaging. High voltages can be induced on cable conductors
by nearby electrostatic field sources. ESD can occur to connector pins when the cable is
connected to an ESDS. A similar discharge occurs if the ESDS is charged when connected
to the cable.
PCBs or assemblies are often potted in resin or enclosed within a plastic housing. Potted
modules sometimes have flying lead terminations for power and I/O connections. When
the housing becomes charged, high voltages can be induced on the PCB or assembly within.
Similarly, electrostatic fields from nearby charged insulators can induce high voltages on
the PCB or assembly. A discharge can occur when a cable is connected or a flying lead
touches a metal item.
References 131
Evaluation and control of these ESD risks often requires analysis of the circumstances
and evaluation of the possible ESD susceptibility of the ESDS. Simple specific ESD control
measures can often be chosen to address specific ESD risks.
References
EOS/ESD Association Inc. (2014) ANSI/ESD S20.20-2014. ESD Association Standard for the
Development of an Electrostatic Discharge Control Program for – Protection of Electrical and
Electronic Parts, Assemblies and Equipment (excluding Electrically Initiated Explosive Devices).
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016) ESD Association Electrostatic Discharge (ESD) Technology
roadmap – revised 2016. Available from: https://www.esda.org/assets/Uploads/docs/
2016ESDATechnologyRoadmap.pdf [Accessed: 10th May 2017] Rome, NY, EOS/ESD
Association Inc.
Industry Council on ESD Target Levels (2011) White paper 1: A case for lowering component
level HBM/MM ESD specifications and requirements. Rev. 3.0. Available from: http://www
.esdindustrycouncil.org/ic/en/documents/37-white-paper-1-a-case-for-lowering-
component-level-hbm-mm-esd-specifications-and-requirements-pdf [Accessed: 10th May
2017] Industry Council on ESD Target Levels
International Electrotechnical Commission. (2013) PD/IEC TS 60079-32-1. Explosive
atmospheres Part 32-1. Electrostatic hazards, guidance. ISBN 978-2-8322-1055-0, Geneva, IEC.
International Electrotechnical Commission. (2015) IEC 61340-5-3:2015. Electrostatics - Part 5-3:
Protection of electronic devices from electrostatic phenomena - Properties and requirements
classification for packaging intended for electrostatic discharge sensitive devices. Geneva, IEC.
International Electrotechnical Commission (2016) IEC 61340-5-1: 2016. Electrostatics – Part
5-1: Protection of electronic devices from electrostatic phenomena - General requirements.
Geneva, IEC.
Olney, A., Gifford, B., Guravage, J., and Righter, A. (2003). EOS-25. Real-world charged board
model (CBM) failures. In: Proc. EOS/ESD Symp., 34–43. Rome, NY: EOS/ESD Association
Inc.
Oxford Dictionary. Available from: https://en.oxforddictionaries.com/definition/habit
[Accessed: 12th May 2017]
Smallwood, J., Swenson, D.E., and Viheriäkoski, T. (2018). Paper 1B.1. Relationship between
footwear resistance and personal grounding through footwear and flooring. In: Proc.
EOS/ESD Symp. EOS-40. Rome, NY: EOS/ESD Association Inc.
Stadler, W., Niemesheim, J., Seidl, S. et al. (2018). The risks of electric fields for ESD sensitive
devices. Paper 1B.4. In: Proc. EOS/ESD Symp. EOS-40. Rome, NY: EOS/ESD Association Inc.
Swenson, D.E. (2012). Electrical fields: What to worry about? Paper 3B.6. In: Proc. EOS/ESD
Symp. EOS-34. Rome, NY: EOS/ESD Association Inc.
132 4 The Seven Habits of a Highly Effective ESD Program
Wallash, A. (2007). A study of “Soft Grounding” of tools for ESD/EOS/EMI control. 2B8-1. In:
Proc. EOS/ESD Symposium EOS-07, 152–157. Rome, NY: EOS/ESD Association Inc.
Further Reading
EOS/ESD Association Inc. (2016) ESD TR20.20-2016. ESD Association Technical Report -
Handbook for the Development of an Electrostatic Discharge Control Program for the Protection
of Electronic Parts, Assemblies and Equipment. Rome, NY, EOS/ESD Association Inc.
International Electrotechnical Commission. (2018) IEC TR 61340-5-2. Electrostatics – Part 5-2:
Protection of Electronic Devices from Electrostatic Phenomena - User Guide. ISBN
978-2-8322-5445-5 Geneva, IEC.
133
Automated Systems
So, if the machines used are not completely up to date, there is a chance that the ESD
control may not be up to the standard required for handling the most sensitive of current
components.
The ESDS devices of concern may be printed circuit boards (PCBs) as well as devices.
Charged board ESD has been increasingly recognized as a significant source of damage. In
recent years, it has been realized that damage mistaken for electrical overstress (EOS) has
actually been caused by charged board ESD (Olney et al. 2003).
Third, adaption or modification of machines or processes can easily inadvertently intro-
duce ESD risks. As in manual processes, failures of ESD control equipment can also result
in ESD risks. Standards are not yet available for ESD control in automated equipment,
although the ANSI/ESD SP10.1 gives some guidance on standard practices. In an auto-
mate process, automated ESD generators can arise that can provide automated damage of
devices! If this occurs, the ESDS damage rate can be high.
The automated environment also brings some special challenges. Many components
designed for automated handling are extremely small. The packaging for these has
correspondingly small features, and this can make it difficult or impossible to test the
packaging with standard test methods. Some components are so small that electrostatic
charging can result in ejection of the component from the packaging in an uncontrolled
manner by electrostatic forces (see Section 5.8).
Access into automated processes during operation may be highly restricted. So, it
can be difficult to observe and diagnose potential ESD issues and make measurements
during operation of the process. Special test and measurement methods may be required,
either due to the form of the items measured or due to the restrictions of the measure-
ment environment in AHE. Stepped process simulation can help enable measurements
to be made.
AHE systems often work within interlocked protective enclosures. Making observations
or measurements on operating systems can be difficult.
● ESD can usually occur only where contact occurs between an ESDS device and a conduc-
tor. The only exception is for ESDS devices where electrostatic fields could cause voltage
differences that can cause damage within the ESDS device.
● ESD can occur only where there is a sufficient voltage difference between the ESDS
device and the contacting conductor. The ESD risk increases with the voltage difference.
The voltage difference that is “sufficient” is dependent on the circumstances and the
component sensitivity.
The first point focuses our attention on parts of a process where an ESDS device makes
contact with other conductors. These could be personnel, machine parts, tools, or other
ESDS devices. Sufficiently far away from the vicinity of the ESDS device, where the risk of
contact and the influence of electrostatic fields on it are insignificant, we do not usually
need to take ESD control measures. It also makes clear that if we can reduce the number of
occasions on which the ESDS makes contact with a conductor, we will reduce the number
of opportunities for ESD to occur. This can in some circumstances reduce ESD risk and the
burden of controlling it.
The second point highlights a need to control voltages on the ESDS device and on
conductors that could contact the ESDS. The voltage difference between the ESDS device
and any conductor that contacts it must be minimized, or at least kept below a risk
threshold level. These voltage differences can be from two causes – triboelectrification
(contact charging) and induction (voltages induced by nearby electrostatic fields)
(see Chapter 2).
Within the critical region, ESD risks should be analyzed using visual observations,
measurements, statistical failure data, or any other available relevant techniques and
information. Critical parts of the process are where ESDS devices make contact with other
items, especially when under the influence of electrostatic fields. Where contact is made,
voltage differences between the ESDS and the contacting item must be controlled. Ideally
ESDS devices should make contact only with static dissipative (see Section 5.2) materials.
Chapter 9 looks in more detail at evaluation of ESD risks.
● Conductors that make contact with ESDS devices must be grounded, where possible.
● Conductors that cannot make contact with ESDS devices need not be grounded.
● If a conductor makes contact with ESDS devices and it cannot be grounded, the ESD risk
must be evaluated. Other ESD controls must be devised if necessary.
Once the ESD risk points have been identified, they can be prioritized and suitable ESD
control measures developed. The control measures will need to be documented and a com-
pliance verification program developed. This may require test methods and equipment to
be defined with suitable pass criteria and frequency of testing.
Paasi (2004) suggested it is useful to think in terms of “intimate” and “proximity” mate-
rials in AHE systems, as defined in ESD-protective packaging materials. From this view,
intimate parts, materials, and surfaces are those that could come into direct contact with
ESDS devices or PCBs during normal operation. Intimate parts of AHE include conveyor
belts and rollers, supports, racks, and other parts that come into direct contact with the
ESDS device. Tape and reel packaging is also of concern. Materials that come into contact
with ESDS devices should be selected to minimize tribocharging of the ESDS device being
handled.
Proximity parts, materials, and surfaces are in the critical region surrounding intimate
parts, containing objects, materials, and surfaces that do not normally come into contact
with the ESDS device. By analogy with packaging, intimate and proximity items should, if
possible, not be made of insulating materials and should be grounded. This is to prevent
electrostatic fields due to charged materials inducing voltages on ESDS devices within the
critical path.
The materials, techniques, and equipment used to achieve these objectives are discussed
in greater detail in Section 5.7.
● Wire bundles in the critical region should be shielded, e.g. with a grounded metal braid.
● Device pickup mechanisms such as vacuum cups, nozzles, and grippers should be made
of conducting materials and should be grounded. The contact area and velocity should
be minimized to reduce tribocharging of device packages.
● ESD grounding points should be directly connected to the equipment ground. SP10.1
recommends a resistance to ground of 1 Ω or less. Lower resistance might be necessary if
high currents (e.g. for motors) or fault currents might be carried.
● Conductors that are relied upon to provide a ground path must be sufficiently robust to
prevent accidental disconnection. The conductors should be braided wire where possible.
● For anodized surfaces, ensure the underlying substrate is grounded to the machine earth.
Methods of grounding moving parts suggested by SP10.1 include flexible conductors such as
braided cables, metal bushes, graphite or beryllium copper commutators, and conductive
greases within the bearings. One problem with moving parts is that resistance to ground
measurements made when stationary may not represent the resistance to ground when
moving. Oil films and other effects can result in intermittent loss of contact that leads to
charge/discharge behavior.
found that all but one of eight YAC handler systems they examined had multiple failures
of internal ground wires. They found that anodized soak boats suffered wear of the
anodizing that allowed metal-to-metal contact between ESDS devices and soak boat
substrate metal.
As with manual processes, all ESD control measures must be checked at regular intervals
to detect any failures or damage that could cause ESD risk. This means that an inspection
regime must be defined, with checks and tests performed at appropriate intervals designed
to detect the failures. In some cases, it may be necessary to implement maintenance proce-
dures to prevent failures.
Tests may be required using standard or custom measurements, with suitably chosen pass
criteria. This compliance verification test program must be documented and established as
part of the ESD control procedures.
to ground can be specified from the view of maintaining low voltage under condition
of maximum charge generation current flow. For circumstances where current flow is
low, maximum resistance to ground can be derived from charge and voltage decay time
considerations.
In some cases, a minimum resistance to ground may also be specified for safe limitation of
current under a fault condition, or to limit ESD current when charged ESDS devices contact
the grounded material. For safety considerations, adding discrete resistance on the ground
path may be acceptable. For reduction of charged device ESD current, the resistance must
be inherent in the contacting material (see Section 4.7.5.2 and 5.7.5).
Where an insulator can become charged and could make contact with a conductor, an
ESD control measure is likely to be essential. Control techniques could include, for example
(Paasi 2004, SP.10), the following:
● Replacement or coating of the insulator with a conducting material, and grounding it
● Prevention of charging by some means (e.g. prevention of rubbing)
● Relocation outside the contact zone
● Shielding or suppression of the field from the insulator by positioning a grounded con-
ductor between it and the ESDS device
● Use of an ionizer to neutralize charge
● Use of topical antistats to reduce charging
Replacing the insulator with a grounded conductor, moving it outside the critical contact
zone, or preventing charging can be among the most effective solutions, if they are possi-
ble. Use of an ionizer is discussed in Section 5.7.9. Prevention of insulating surfaces from
rubbing can be a useful low-cost way of preventing an insulator from becoming charged.
This may require adjustment of the position of adjacent parts of equipment. It can be a par-
ticularly useful technique with cables, pneumatic lines, or other flexible insulating lines.
Perhaps the easiest but least reliable technique is to use a topical antistat to reduce electro-
static charging. These often effect a temporary solution but can quickly become ineffective.
Tan (1993) used antistats to confirm that electrostatic charging of insulating spacers was the
cause of some charged device ESD damage. He found that the lifetime of the antistat was
much shorter in a varying (hot and cold) temperature environment than the accelerated
life test had suggested, lasting only one day in practice. Often antistats rely on atmospheric
humidity for their action, and so they can become ineffective under low-humidity con-
ditions. In an enclosed space within AHE with heat sources present, local low-humidity
regions can easily be established.
The effect of electrostatic field sources can also be reduced or eliminated by shielding the
ESDS device from the field. This may be done by inserting a grounded conductor between
the field source and the ESDS critical path. Tan (1993) gives several examples of using metal
flanges, washers, copper tubes, or self-adhesive metal tapes for shielding purposes.
It can be difficult to prevent charging by contact in this way. Choice of contacting mate-
rials can make a difference to the charge generation levels. Often the charge accumulated
on the ESDS must be neutralized to bring it to an acceptable level before contact is made
with another conductor.
Kim et al. (2012) found that triboelectrically generated charge on liquid crystal display
(LCD) screens during production could not be adequately controlled by ionization (see
Section 5.7.9). The glass panel substrate generated high charge levels and held charge for
long periods. More than 40 material contact and separation actions occurred during the
production process, including friction from photoresist coatings applied to the glass sur-
face, deionized water spray rinses used to clean the surface, contact with rollers or belts,
and pressure and separation of glass panels from vacuum chucks. Control measures they
used to reduce ESD risks included increasing surface roughness and minimizing separa-
tion speed and vacuum pressure to control the charge generation of glass substrates. They
found that using insulative instead of dissipative or conductive lift pin materials prevented
ESD events. Specification of adequate separation distance and insulation thickness was also
used to prevent air discharge between the glass and metal objects.
5.7.5 Using Resistive Contact Materials to Limit Charged Device ESD Current
Charged device ESD risk is greatest when the charged device makes contact with a low resis-
tance conductor. This results in a high peak current discharge (so called hard discharge).
One means of reducing this risk is to ensure that the device contacts a resistive material,
rather than a highly conductive material. In this case, the peak discharge current can be lim-
ited by including sufficient contacting material resistance. Placing a resistor in the ground
path is ineffective for this purpose (see Section 4.7.5.2).
There is some debate as to whether direct current (DC) resistance measurements on poly-
meric ESD control materials provide a realistic guide to ESD risk (Viheriäkoski et al. 2017).
They found the “ESD resistance” due to frequency dependent impedance of the materials
could give much lower ESD current and charged device ESD risk than would be expected
from DC resistance measurements.
5.7.6 Anodization
Anodization is often used as a passivation and protection of machine surfaces. Although
considered insulative by some, it can be used to provide a noninsulating surface coating
on an aluminum alloy substrate (Bellmore 2001). The anodized layer may be 5–40 μm
(0.0002–0.0015 in.) thick. Different anodizing processes can produce a wide range of
surface resistance from <1 MΩ to over 100 GΩ.
If done after screw holds and threads are made, anodization can prevent good contact
between metallic machine parts (Yan et al. 2009). The resistance through the joints may
be increased to several ohms. Thick anodization can result in higher resistances. It can be
difficult to remedy this after the machine has been constructed.
Anodization can be used to provide a hard-wearing static dissipative surface coating to
reduce the risk of charged device ESD damage when device leads touch the surface (Yan
et al. 2009; Smallwood and Millar 2010; Millar and Smallwood 2010).
5.7 Materials, Techniques, and Equipment Used for ESD Control in AHE 145
Millar and Smallwood (2010) found that anodized soak boats experienced wear or dam-
age to the anodized layer. Direct contact could then be occasionally made between the leads
of devices in the boats and the underlying metal soak boat material. Smallwood and Millar
(2010) measured anodized soak boat surface resistance using standard and nonstandard
electrodes as well as resistance from the boat metal substrate through the surface and a
supporting dissipative rail to ground.
The influence of applied test voltage was also measured. They found that the resistance
of the anodized layer varied dramatically with the applied voltage and test electrodes used.
At 1000 V the resistance through the anodized layer was <1 Ω, but at 100 V this increased
to over 20 GΩ. Surface resistance from point to point was <1 GΩ using ESD S4.1 electrodes,
but 30–96 GΩ using an S11.13 2-pin measurement. Other methods also gave varying results.
As variation in resistance could be expected to cause variations in charge decay time of
the boat standing on the rail, this was also measured. The charge decay time varied with
the remaining voltage on the boat, increasing as the voltage reduced. At 1000 V, the voltage
disappeared immediately when the applied voltage source was removed. Decay times could
be as long as 20 seconds at 100 V applied voltage.
Their results showed that large soft electrodes with high contact pressure such as S4.1
gave lower resistance values. The surface may have small regions of high conductivity,
damage, or pin holes that show low resistance with these electrodes. Small area electrodes
with hard surfaces are less likely to make good contact through low-resistance or damaged
regions.
These results showed that when measuring the performance of an anodized layer, it is
important to use test voltage and electrodes that simulate the working condition as far as is
possible.
5.7.7 Bearings
Many moving machine parts are supported on bearings. The grounding of these machine
parts is often through the bearing. If measured when stationary, the resistance through
the bearing may be low. When the machine is running, the resistance through the bearing
may be much higher or intermittent due to the oil film that separates the moving bearing
surfaces. This problem can be avoided by fitting a bearing that is lubricated by a conducting
(noninsulative) lubricant. Alternatively, a parallel ground path can be established through
a commutator or other mechanism.
contact with another conductor. Where a device has an insulating package material, the
package material that makes contact with the vacuum cup will become charged. If the vac-
uum cup material is also made of insulating material or is not grounded, charge will build
up on the cup during successive operations. So, the cup should be made of a static dissipa-
tive material and grounded. Checking of the grounding of cups should be specified as part
of compliance verification planning.
Nevertheless, some contact charging of the device package cannot be avoided (Yan et al.
2009). The residual charge on the device does not produce voltage on the device until separa-
tion between the cup and the device occurs. This may occur only when the device is a small
distance from making contact with a conductor. Even highly efficient ionizers are unlikely
to have much effect in the small time between release of the device and contact with the
conductor. In this situation, specification of resistive contacting materials, if possible, may
be necessary to avoid damaging ESD.
ESD-protective packaging used in automated equipment often takes special forms such as
tape and reel or JEDEC trays (see Chapter 8). These may be specifically designed to handle
the devices used in the processes. Their shape and form may pose difficulties in measure-
ment for verification of the packaging (see Section 5.9.2).
Components are supplied in tape and reel for placement on PCBs in a pick and place
machine. ESDS devices should be supplied in ESD-protective packaging, but non-ESDS
components are often supplied in tape and reel made from insulating materials. If ESDS and
non-ESDS components are placed in the same operation, electrostatic fields from insulating
tape and reel packaging can become a risk to the ESDS devices. One way of reducing this
risk may be to separate ESDS and non-ESDS components in the placement machine.
Very small devices and their packaging can become sufficiently highly charged that
they may be ejected from the packaging in an uncontrolled manner by electrostatic forces
(Swenson 2018).
● Resistance to ground from parts of equipment, surfaces, conveyor belts, or other items
● Electrostatic field and voltage measurements on essential insulators or ungrounded con-
ductors
● Voltage measurements on ESDS devices or small conductors
● Charge measurements on conductors or insulators
● Voltage decay time and offset voltage due to neutralization by an ionizer
● Detection of ESD using radiated electromagnetic interference (EMI) from ESD
148 5 Automated Systems
nonflat surfaces by calibrating the meter in use with a target of the same shape and form
raised to a known voltage.
Voltages on ESDS devices or other moving items typically change with the position of the
item due to changes in capacitance and charge levels. If the item is moving quickly or in a
confined space, it can be difficult to mount a voltmeter in a way that can measure the voltage
on it. In a fast-moving system, the response speed of the voltmeter needs to be fast enough
to follow the voltage changes accurately. A data logger or storage oscilloscope operating in
roll mode may be needed to record voltage changes as they happen. In practice, it is likely
to be the voltage on the conductor immediately before contact with another conductor that
is relevant for evaluating charged device ESD risk.
Voltage levels on insulative parts of an item are even more variable as the charge density
on the surface of the insulator varies from point to point across the surface. There is no
single value that can be ascribed to the surface of an insulator such as a device package
material. Voltages measured are typically the net effect of charge density over a region on
the surface and will change with proximity to conductors including any voltage measuring
equipment used. This can be a useful measurement in ESD risk evaluation as in practice it
is often also the net effect of charges inducing a voltage on a conductor that causes ESD risk.
5.9.5 Measurement of the Voltage Decay Time and Offset Voltage Due
to Neutralization by an Ionizer
Standard measurement of the effectiveness of ionizers is done using a charged plate monitor
(CPM) in which the measurement plate is 150 × 150 mm (see Section 11.6.13). The positive
and negative polarity charge decay time and offset voltage are measured. This size of plate
approximates the size of a semiconductor wafer or PCB. Many devices or machine parts
that are to be neutralized in AHE are much smaller than this. It is questionable whether a
CPM plate this size is representative of charge neutralization on small devices. Many non-
standard CPM instruments are available that have smaller plates a few tens of millimeters
in dimensions. It is possible that these give more representative indication of neutralization
of similarly sized devices.
As time goes on, devices having low HBM and CDM ESD withstand voltages are becoming
more common (ESDA 2016a). Some have HBM withstand voltage less than 100 V HBM and
200 V CDM withstand voltage.
The ESD TR10.0-01-02 discusses measurement and control issues in handling very
sensitive devices in AHE. This document defines hard ground as a having resistance
to ground of 1 Ω or less, and soft ground as between 1 kΩ and 1 GΩ. Machine parts
typically require hard grounding, but an ESDS device making contact with this low
resistance may risk charged device ESD damage. AHE parts that may contact ESDS
may require soft grounding. Charged device ESD damage is typically due to high
ESD peak current flow, and it is the resistance (or impedance, see Viheriäkoski et al.
2017) of the material at the point of contact that typically limits ESD current in the
discharge.
When handling low CDM ESD withstand voltage devices in AHE, the voltages in the
critical region and on the device need to be kept to a low level. This means that if ionizers are
used for voltage reduction, the ionizer may need to have greater capacity to reduce voltages
in a shorter time, as well as having low-offset voltage (see Section 5.7.9).
CPMs used for characterizing performance of ionizers should be as representative as
possible of the situation that they are simulating. In the AHE environment, this may
mean use of a small low capacitance CPM plate. For simulating neutralization of low
CDM ESD withstand voltage devices, it may be better to measure decay time over a
lower voltage range (e.g. 200–20 V) than the standard CPM measurement (typically
1000–100 V).
References
Bellmore, D. (2001). Anodized aluminium alloys – insulators or not? In: Proc EOS/ESD Symp.
EOS-23, 141–148. Rome, NY: EOS/ESD Association Inc.
Bellmore, D.G. (2004). Paper 4A.6. Characterizing automated handling equipment using
discharge current measurements. In: Proc EOS/ESD Symp. EOS-26. Rome, NY: EOS/ESD
Association Inc.
ESD Association. (2015). ESD TR17.0-01-15. Technical Report for ESD Process Assessment
Methodologies in Electronic Production Lines – Best Practices used in Industry, Rome, NY,
EOS/ESD Association Inc.
ESD Association. (2016a) ESD Association Electrostatic Discharge (ESD) Technology
roadmap – revised 2016. Available from: https://www.esda.org/assets/Uploads/docs/
2016ESDATechnologyRoadmap.pdf [Accessed: 10th May 2017] Rome, NY, EOS/ESD
Association Inc.
ESD Association (2016b) ANSI/ESD SP10.1-2016. Standard practice for protection of
Electrostatic Discharge Susceptible Items – Automated handling Equipment (AHE), Rome, NY,
EOS/ESD Association Inc.
References 153
Further Reading
Bellmore, D.G. and Bernier, J. (2005). Characterizing automated handling equipment using
discharge current measurements II. In: Proc. EOS/ESD Symp., 195. Rome, NY: EOS/ESD
Association Inc.
Dangelmeyer, T. (1999). ESD Program Management, 2e. Clewer. ISBN: 0-412-13671-6.
EOS/ESD Association Inc. (2016). ESD TR20.20-2016, ESD Association Technical Report -
Handbook for the Development of an Electrostatic Discharge Control Program for the Protection
of Electronic Parts, Assemblies and Equipment. Rome, NY, EOS/ESD Association Inc.
ESD Association. (2002). ESD TR10.0-01-02. Technical Report - Measurement and ESD Control
Issues for Automated Equipment Handling of ESD Sensitive Devices Below 100 Volts, Rome,
NY, EOS/ESD Association Inc.
ESD Association. (2006). ANSI/ESD STM4.1-2006. ESD Association Standard for the Protection
of Electrostatic Discharge Susceptible Items – Worksurfaces - Resistance Measurements, Rome,
NY, EOS/ESD Association Inc.
ESD Association. (2014). ANSI/ESD S20.20-2014. ESD Association Standard for the
Development of an Electrostatic Discharge Control Program for – Protection of Electrical and
Electronic Parts, Assemblies and Equipment (excluding Electrically Initiated Explosive Devices),
Rome, NY, EOS/ESD Association Inc.
ESD Association. (2015) ANSI/ESD STM11.13-2015. ESD Association Standard Test Method for
the Protection of Electrostatic Discharge Susceptible Items – Two-Point Resistance
Measurement, Rome, NY, EOS/ESD Association Inc.
Halperin, S., Gibson, R., and Kinnear, J. (2008). Paper 2B-21. Process capability & transitional
analysis. In: Proc. EOS/ESD Symp. EOS-30. Rome, NY: EOS/ESD Association Inc.
International Electrotechnical Commission. (2015) IEC 61340-5-3:2015. Electrostatics.
Protection of electronic devices from electrostatic phenomena. Properties and requirements
classifications for packaging intended for electrostatic discharge sensitive devices, International
Electrotechnical Commission, Geneva.
International Electrotechnical Commission. (2016a). IEC 61340-2-3:2016. Electrostatics. Part
2-3: Methods of test for determining the resistance and resistivity of solid materials used to avoid
electrostatic charging, International Electrotechnical Commission, Geneva.
International Electrotechnical Commission (2016b) IEC 61340-5-1: 2016. Electrostatics – Part
5-1: Protection of electronic devices from electrostatic phenomena - General requirements,
International Electrotechnical Commission, Geneva.
International Electrotechnical Commission. (2018) IEC TR 61340-5-2:2018. Electrostatics – Part
5-2: Protection of electronic devices from electrostatic phenomena - User guide. International
Electrotechnical Commission, Geneva.
Kietzer, G. (2012). Paper 2B.2. ESD risks in the electronics manufacturing. In: Proc. EOS/ESD
Symp. EOS-34, 202. Rome, NY: EOS/ESD Association Inc.
Kim, D.-S., Lim, C.-B., Yoon, S.-H. et al. (2013). Paper 7B.1. Electrostatic control and its analysis
of roller transferring processes in FPD manufacturing. In: Proc. EOS/ESD Symp. EOS-35.
Rome, NY: EOS/ESD Association Inc.
Koh, L.H., Goh, Y., and Lim, S.H. (2013). Paper 1B.3. Reliability assessment of high
temperature automated handling equipment retrofit for CDM mitigation. In: Proc. EOS/ESD
Symp. EOS-35. Rome, NY: EOS/ESD Association Inc.
Further Reading 155
Koh, L.H., Goh, Y.H., and Wong, W.F. (2017). Paper 1B.2. ESD risk assessment considerations
for automated handling equipment. In: Proc. EOS/ESD Symp. EOS-39. Rome, NY: EOS/ESD
Association Inc.
Paasi, J., Tamminen, P., Kalliohaka, T. et al. (2002). ESD control tools for surface mount
technology and final assembly lines. In: Proc. EOS/ESD Symp. EOS-24, 250–256. Rome, NY:
EOS/ESD Association Inc.
Paasi, J., Tamminen, P., Salmela, H. et al. (2005). ESD control in automated placement process.
In: Proc. EOS/ESD Symposium EOS-27, 203. Rome, NY: EOS/ESD Association Inc.
Steinman, A. (2010). Paper 3B3. Measurements to establish process ESD compatibility. In: Proc.
EOS/ESD Symp. EOS-32. Rome, NY: EOS/ESD Association Inc.
Steinman, A. (2012). Paper 2B.4. Process ESD capability measurements. In: Proc. EOS/ESD
Symp. EOS-34. Rome, NY: EOS/ESD Association Inc.
Steinman, A. (2014). Paper 1B.3. Measuring handler CDM stress provides guidance for factory
static controls. In: Proc. EOS/ESD Symp. EOS-36. Rome, NY: EOS/ESD Association Inc.
Tamminen, P. and Viheriäkoski, T. (2007). Paper 3B.3. Characterization of ESD risks in an
assembly process by using component-level CDM withstand voltage. In: Proc. EOS/ESD
Symp. EOS-29, 202–211. Rome, NY: EOS/ESD Association Inc.
Tamminen, P. and Viheriäkoski, T. (2011). Product specific ESD risk analysis. In: Proc.
EOS/ESD Symp. EOS-33, 97. Rome, NY: EOS/ESD Association Inc.
Welker, R.W., Nagarajan, R., and Newberg, C. (2006). Contamination and ESD Control in
High-Technology Manufacturing. Wiley-Interscience/IEEE Press. ISBN-10: 0 471 41452 2,
ISBN-13: 978 0 471 41452 0.
Yan, K.P., Gaertner, R., and Wong, C.Y. (2010). Paper 3B.1. ESD protection program at
electronics industry – areas for improvement. In: Proc. EOS/ESD Symp. EOS-32. Rome, NY:
EOS/ESD Association Inc.
Yan, K.P., Gaertner, R., and Wong, C.Y. (2013). Semiconductor back end manufacturing
process – ESD capability analysis. In: Proc. EOS/ESD Symposium EOS-35, 30. Rome, NY:
EOS/ESD Association Inc.
157
6.1 Introduction
The facilities equipment and materials needed for use in electrostatic discharge (ESD) pre-
vention are specified in various standards worldwide. The two main standards discussed in
this book are IEC 61340-5-1 and ESD Association ANSI/ESD S20.20. IEC 61340-5-1 has also
been adopted as a national standard in many countries, in Europe becoming the European
Norm EN61340-5-1. Individual countries may have their own versions, and, in the United
Kingdom, this is BS EN 61340-5-1.
For clarity, the 61340-5 series terminology is used, although the ESD Association (ESDA)
series documents are in many cases nearly identical and are cross-referenced. In many
cases, International Electrotechnical Commission (IEC) documents were based on ESDA
standards that were published earlier. Where appropriate, some differences with the ESDA
standards are noted.
separate documents IEC 61340-5-2 and ESD TR20.20 (International Electrotechnical Com-
mission 2018; EOS/ESD Association Inc. 2016b). These user guides contained no require-
ments for compliance but gave an increasing amount of useful information and guidance
to help develop and maintain an effective ESD control program.
These documents that do not contain requirements for compliance, but often contain
guidance or information of a different type, are called “Technical Reports.” In the IEC and
ESDA systems they are designated by “TR” in the document number, for example ESD
TR20.20 or IEC TR 61340-5-2. They can be extremely useful sources of information of many
types. The references at the end of this chapter are a list of many of these.
Argentina O-Member
Australia O-Member
Austria P-Member
Belgium P-Member
Bulgaria O-Member
China P-Member
Czech Republic P-Member
Denmark O-Member
Egypt O-Member
Finland P-Member
France P-Member
Germany P-Member
Hungary P-Member
Ireland P-Member
Israel O-Member
Italy P-Member
Japan P-Member
Korea, Republic of P-Member
Netherlands P-Member
Norway O-Member
Poland P-Member
Portugal O-Member
Romania O-Member
Russian Federation P-Member
Serbia O-Member
Spain P-Member
Sweden P-Member
Switzerland P-Member
Thailand O-Member
Turkey O-Member
Ukraine O-Member
United Kingdom P-Member
United States of America P-Member
Kingdom. All these steps and procedures take an amount of elapsed time. It is often several
years between commencement of a project at IEC and publication of an IEC standard and
then several further months before publication as BS EN in the UK or other national stan-
dard!
6.4 The IEC and ESDA Standards 161
Table 6.2 Near-equivalent ESD control program standards in the IEC 61340-5-1 and ANSI/ESD
S20.20 series of documents.
Table 6.3 Near-equivalent test method documents in the IEC 61340-5-1 and ANSI/ESD S20.20
series.
IEC 61340-5-1 was adopted by CENELEC and replaced the first edition of EN61340-5-1:
2001 in Europe.
A new version of ANSI/ESD S20.20 was published in 2014. This was followed by a new
version 61340-5-1 in 2016. These versions were current at the time of writing, and the fol-
lowing sections summarize the requirements of these documents. The documents continue
to be updated in the light of technology and industry changes and the development of ESD
protection and control knowledge and understanding as time goes on. The reader should
of course check current versions of the standards for up to date information on them.
initiated explosive devices, flammable liquids, gases, and powders are specifically excluded
by the scope. Some of these nonelectronic applications may be covered by specific national
regulations or other standards such as IEC 60079-32-1 International Electrotechnical Com-
mission. (2013a).
● Compliance verification
● Grounding and bonding systems
● Personnel grounding
● EPA requirements
● Packaging systems
● Marking
The ESD Control Program Plan is the main document for implementing and verifying
the ESD program and must be applied to all relevant areas. It should conform to internal
quality requirements.
IEC 61340-5-1 requires that the ESD control program documents the lowest ESD with-
stand voltage that can be handled by the program. As it can be difficult to find the ESD
withstand data on all components, it may be easiest to state the standard ESD withstand
voltages of 100 V HBM and 200 V CDM in the ESD Control Program Plan.
Initial and recurrent training must be provided to all personnel who handle or come into
contact with ESDS devices. Initial training must be provided before personnel handle ESDS.
Test methods may be used that differ from those given in the standard, but evidence must
be provided in a tailoring statement that the results correlate with the reference standards.
Records must be kept providing evidence of compliance with the technical requirements
of the ESD Program.
Wrist band Interior ≤105 Ω Not specified Interior ≤105 Ω Wrist strap
(not worn) Exterior >107 Ω Exterior >107 Ω system test is
IEC 61340-4-6 ANSI/ESD S1.1 used
Wrist strap cord < 5 × 106 Ω or Wrist strap 0.8 × 105 ≤ R ≤
user defined system test is 1.2 × 106 Ω
IEC 61340-4-6 used ANSI/ESD S1.1
Wrist strap Not specified < 5 × 106 Ω Rg < 2 Ω Rg < 2 Ω
connection point
ANSI/ESD S6.1 ESD TR53
(not monitored)
grounding and
bonding systems
Wrist strap system Not specified < 3.5 × 107 Ω < 3.5 × 107 Ω < 3.5 × 107 Ω
test IEC 61340-4-6 ANSI/ESD ESD TR53 wrist
S1.1 Sec. 6.11 strap
Continuous Not specified Not specified User defined Manufacturer
monitors defined limit
ESD TR53
continuous
monitors
Footwear < 108 Ω Person-footwear < 109 Ω Rg < 109 Ω
IEC 61340-4-3 system test is ANSI/ESD STM ESD TR53
used 9.1 Footwear
ANSI/ESD STM section
9.2
Person, footwear Body voltage Rg < 109 Ω Body voltage Footwear
and flooring <100 V AND peak <100 V and Rg < 109 Ω
IEC 61340-4-5
system test Rg < 109 Ω floor Rg < 109 Ω
Periodic ESD TR53
IEC 61340-4-5 voltage test ESD STM97.1 Footwear
and 97.2 section
Flooring
Rg < 109 Ω
ESD TR53
Flooring section
Person footwear Not specified Rg < 109 Ω
system test
IEC 61340-5-1
Annex A
170 6 ESD Control Standards
Bench surface, Rp-p < 109 Ω Rg < 109 Ω Rp-p < 109 Ω N/A
racks, trolleys
IEC 61340-2-3 IEC 61340-2-3 ANSI/ESD STM4.1 R < 109 Ω
etc. surface g
Rgp < 109 Ω Rgp < 109 Ω ESD TR53
IEC 61340-2-3 ANSI/ESD STM4.1 worksurface
ESD TR53 mobile
Rg < 109 Ω equipment
ANSI/ESD STM4.1
< 200 V
ANSI/ESD
STM4.2:2012a
Floor Rgp < 109 Ω Rg < 109 Ω Rg < 109 Ω Rg < 109 Ω
IEC 61340-4-1 IEC 61340-4-1 ANSI/ESD STM7.1 ESD TR53 Flooring
section
Rgp < 109 Ω
ANSI/ESD STM7.1
Rp-p < 109 Ω
ANSI/ESD STM7.1
Chair Rgp < 109 Ω Rgp < 109 Ω Rgp < 109 Ω Rg < 109 Ω
IEC 61340-2-3 ESD STM12.1 ESD TR53 seating
Static control Rp-p ≤ 1011 Ω or Rp-p ≤ 1011 Ω or Rp-p < 1011 Ω Rp-p < 1011 Ω
garment user defined user defined ANSI/ESD ESD TR53
IEC 61340-4-9 IEC 61340-4-9 or STM 2.1 garments
or user defined user defined
Groundable Rgp < 109 Ω Rgp < 109 Ω Rgp < 109 Ω Rgp < 109 Ω
static control
IEC 61340-4-9 IEC 61340-4-9 ANSI/ESD ESD TR53
garment
STM 2.1 garments
Groundable Not specified Not specified Rg < 3.5 × 107 Ω Rg < 3.5 × 107 Ω
static control
ANSI/ESD ESD TR53
garment
STM 2.1 garments
system
6.5 Requirements of IEC 61340-5-1 and ANSI/ESD S20.20 Standards 171
6.5.12.6 Insulators
All nonessential insulators such as cups, packaging, and personal items must be removed
from any workstation where unprotected ESDS devices are handled. (In ANSI/ESD S20.20
they must be removed from the EPA.)
Where process essential insulators are present, the ESD threat must be evaluated.
● The electrostatic field must not exceed 5 kV m–1 at the position where ESDS devices are
handled.
● Where the surface potential of an insulator exceeds 2 kV, it must be kept at least 30 cm
from any ESDS device.
● A new requirement that where the surface potential of an insulator exceeds 125 V, it must
be kept at least 2.5 cm from ESDS devices.
Any ESD risk must then be mitigated by some means e.g. using ionizers.
Table 6.9 Requirements for soldering and desoldering hand tools (ANSI/ESD S20.20 only).
Packaging – 104 Ω ≤ Rs < 1011 104 Ω ≤ Rv < 1011 104 Ω ≤ Rs < 1011 104 Ω ≤ Rv < 1011
Static IEC 61340-2-3 IEC 61340-2-3 STM11.11 STM11.12
dissipative
STM11.13
Packaging – Rs ≤ 104 Ω Rv < 104 Ω Rs < 104 Ω Rv < 104 Ω
conductive IEC 61340-2-3 IEC 61340-2-3 STM11.11 STM11.12
STM11.13
Packaging – Rs ≥ 1011 Ω Rv ≥ 1011 Ω Rs ≥ 1011 Ω Rs ≥ 1011 Ω
insulator IEC 61340-2-3 IEC 61340-2-3 STM11.11 STM11.12
STM11.13
“The user should determine
whether or not a specific packaging
configuration provides for a
Electrostatic Rs < 103 Ω Rv < 103 Ω
reduction of electric field strength at
(Electric) field IEC 61340-2-3 IEC 61340-2-3 the position in the package where
shielding
sensitive items are contained.”
In ANSI/ESD S20.20, where ESDS devices are placed on packaging and work performed
on them, the packaging is considered to be a work surface and work surface requirements
for resistance to ground apply.
6.5.14 Marking
Marking of ESDS devices, systems, or packaging must be according to customer contracts,
purchase orders, or other documentation. Where these are not defined, then the need for
marking should be considered in the ESD Control Program Plan. Where marking is needed,
it must be documented in the ESD Program Plan.
References
British Standards Institute (1984) BS 5783:1984. Code of practice for handling of electrostatic
sensitive devices. London, BSI.
British Standards Institute (1992) BS EN 100015-1:1992. Basic specification. Protection of
electrostatic sensitive devices. Harmonized system of quality assessment for electronic
components. Basic specification: protection of electrostatic sensitive devices. General
requirements. London, BSI.
British Standards Institute (1994a) BS EN 100015-2:1994. Basic specification. Protection of
electrostatic sensitive devices. Requirements for low humidity conditions. London, BSI.
British Standards Institute (1994b) BS EN 100015-3:1994. )Basic specification. Protection of
electrostatic sensitive devices. Requirements for clean room areas.London, BSI.
British Standards Institute (1994c) BS EN 100015-4:1994. Basic specification. Protection of
electrostatic sensitive devices. Requirements for high voltage environments. London, BSI.
British Standards Institute (2001) BS EN 61340-5-1:2001. Electrostatics. Protection of electronic
devices from electrostatic phenomena. General requirements. London, BSI.
British Standards Institute (2016) BS EN 61340-5-1:2016. Electrostatics. Protection of electronic
devices from electrostatic phenomena. General requirements. London, BSI.
Department of Defense. (1980a) MIL-STD-1686. Standard Practice. Electrostatic discharge
control program for protection of electrical and electronic parts, assemblies and equipment
(excluding electrically initiated devices). Washington, D.C., DoD.
Department of Defense. (1980b) MIL-HDBK-263. Military handbook. Electrostatic discharge
control handbook for protection of electrical and electronic parts, assemblies and equipment
(excluding electrically initiated devices. Washington, D.C., DoD.
EOS/ESD Association Inc. (1999) ANSI/ESD S20.20-1999. ESD Association Standard for the
Development of an Electrostatic Discharge Control Program for – Protection of Electrical and
Electronic Parts, Assemblies and Equipment (excluding Electrically Initiated Explosive Devices).
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2003) ESD SP9.2-2003. ESD Association Standard for the Protection
of Electrostatic Discharge Susceptible Items – Footwear – Foot Grounders Resistive
Characterization (not to include static control shoes). Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2006a) ANSI/ESD STM4.1-2006. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Worksurfaces - Resistance
Measurements. Rome, NY, EOS/ESD Association Inc.
References 175
EOS/ESD Association Inc. (2006b) ESD TR53-01-06. Technical Report for the protection of
electrostatic discharge susceptible items - Compliance Verification of ESD Protective Equipment
and Materials. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2012a) ANSI/ESD STM4.2-2012. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – ESD Protective Worksurfaces – Charge
Dissipation Characteristics. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2012b) ANSI/ESD STM11.31-2012. ESD Association Standard Test
Method for Evaluating the Performance of Electrostatic Discharge Shielding Materials – Bags.
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2013a) ANSI/ESD STM12.1-2013. ESD Association Standard Test
Method for the Protection of Electrostatic Discharge Susceptible Items - Seating - Resistance
Measurement. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2013b) ANSI/ESD STM7.1-2013. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Floor Materials - Resistive
Characterization of Materials. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2013c) ANSI/ESD S1.1-2013. Standard for protection of Electrostatic
Discharge Susceptible Items - Wrist Straps. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2013d) ANSI/ESD STM2.1-2013. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Garments. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2014a) ANSI/ESD S6.1-2014. Standard for the Protection of
Electrostatic Discharge Susceptible Items – Grounding. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2014b) ANSI/ESD STM9.1-2014. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Footwear - Resistive Characterization.
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2014c) ANSI/ESD S20.20-2014. ESD Association Standard for the
Development of an Electrostatic Discharge Control Program for – Protection of Electrical and
Electronic Parts, Assemblies and Equipment (excluding Electrically Initiated Explosive Devices).
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015a) ANSI/ESD STM3.1-2015. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Ionization. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2015b) ANSI/ESD STM11.11-2015. ESD Association Standard for
Protection of Electrostatic Discharge Susceptible Items – Surface Resistance Measurement of
Static Dissipative Planar Materials. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015c) ANSI/ESD STM11.12-2015. ESD Association Standard for
Protection of Electrostatic Discharge Susceptible Items.Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015d) ANSI/ESD STM11.13-2015. ESD Association Standard Test
Method for the Protection of Electrostatic Discharge Susceptible Items – Two-Point Resistance
Measurement. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015e) ANSI/ESD STM97.1-2015. ESD Association Standard Test
Method for the Protection of Electrostatic Discharge Susceptible Items – Floor Materials and
Footwear – Resistance Measurement in Combination with a Person. Rome, NY, EOS/ESD
Association Inc.
176 6 ESD Control Standards
Further Reading
EOS/ESD Association Inc. (1995a) ESD ADV53.1-1995. Advisory for Protection of Electrostatic
Discharge Susceptible Items - ESD Protective Workstations. Rome, NY, EOS/ESD Association
Inc.
EOS/ESD Association Inc. (1995b) ESD ADV11.2-1995. Advisory for the Protection of
Electrostatic Discharge Susceptible Items - Triboelectric Charge Accumulation Testing. Rome,
NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (1999a) ESD TR13.0-01-99. Technical Report - EOS Safe Soldering
Iron Requirements. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (1999b) ESD TR15.0-01-99. Standard Technical Report for the
Protection of Electrostatic Discharge Susceptible Items-ESD Glove and Finger Cots. Rome, NY,
EOS/ESD Association Inc.
EOS/ESD Association Inc. (1999c) ESD TR50.0-01-99. Technical Report - Can Static Electricity
be Measured? Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (1999d) ESD TR50.0-02-99. Technical Report - High Resistance
Ohmmeter Measurements. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2000a) ESD TR2.0-01-00. Technical Report - Consideration For
Developing ESD Garment Specifications. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2000b) ESD TR2.0-02-00. Technical Report - Static Electricity
Hazards of Triboelectrically Charged Garments. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2001) ESD TR1.0-01-01. Technical Report - Survey of Constant
Monitors for Wrist Straps. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2002a) ESD TR3.0-01-02. Technical Report - Alternate Techniques for
Measuring Ionizer Offset Voltage and Discharge Time. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2002b) ESD TR4.0-01-02. Technical Report - Survey of Worksurfaces
and Grounding Mechanisms. Rome, NY, EOS/ESD Association Inc.
178 6 ESD Control Standards
EOS/ESD Association Inc. (2002c) ESD TR10.0-01-02. Technical Report - Measurement and ESD
Control Issues for Automated Equipment Handling of ESD Sensitive Devices Below 100 Volts.
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2003) ESD TR50.0-03-03. Technical Report - Voltage and Energy
Susceptible Device Concepts, Including Latency Considerations. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2004) ESD TR55.0-01-04. Technical Report - Electrostatic Guidelines
and Considerations For Cleanrooms and Clean Manufacturing. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2005) ESD TR3.0-02-05. Technical Report - Selection and Acceptance
of Air Ionizers. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2011a) ANSI/ESD SP15.1-2011. Standard Practice for the Protection
of Electrostatic Discharge Susceptible Items – In-Use Resistance Measurement of Gloves and
Finger Cots. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2011b) ESD TR7.0-01-11. Technical Report for the Protection of
Electrostatic Discharge Susceptible Items – Static Protective Floor Materials. Rome, NY,
EOS/ESD Association Inc.
EOS/ESD Association Inc. (2012) ANSI/ESD S11.4-2012. Standard for the Protection of
Electrostatic Discharge Susceptible Items - Static Control Bags. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2015a) ANSI/ESD S13.1-2015. Provides electrical
soldering/desoldering hand tool test methods for measuring current leakage, tip to ground
reference point resistance, and tip voltage. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015b) ESD TR17.0-01-15. Technical Report for ESD Process
Assessment Methodologies in Electronic Production Lines – Best Practices used in Industry.
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015c) ESD TR53-01-15. Technical Report for the Protection of
Electrostatic Discharge Susceptible Items – Compliance Verification of ESD Protective
Equipment and Materials. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016a) ANSI/ESD SP3.3-2016. Standard Practice for the Protection
of Electrostatic Discharge Susceptible Items – Periodic Verification of Air Ionizers. Rome, NY,
EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016b) ANSI/ESD SP3.4-2016. Standard Practice for the Protection
of Electrostatic Discharge Susceptible Items – Periodic Verification of Air Ionizer Performance
Using a Small Test Fixture. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016c) ESD SP10.1-2016. Standard practice for protection of
Electrostatic Discharge Susceptible Items – Automated handling Equipment (AHE). Rome, NY,
EOS/ESD Association Inc.
EOS/ESD Association Inc. (2017a) ESD ADV1.0-2017. ESD Association Advisory for
Electrostatic Discharge Terminology – Glossary. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2017b) ANSI/ESD S8.1-2017. Draft Standard for the Protection of
Electrostatic Discharge Susceptible Items – Symbols – ESD Awareness. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2019) ESD Fundamentals Part 6: ESD Standards. Available from:
https://www.esda.org/index.php/about-esd/esd-fundamentals/part-6-esd-standards
(Accessed 26th January 2019).
179
7.1 Introduction
Electrostatic discharge (ESD) control equipment can form a considerable part of the
investment cost of an ESD control program. It is important to be sure that the equipment
will do the task intended, for the intended life of the equipment. There may be many types
of similar equipment on the market, and the type chosen should be appropriate for use
in the facility and processes intended. Cleaning, care, and maintenance may be needed to
ensure that the equipment continues to work correctly over its life.
Unless equipment is single use and disposable, it is important that any failures of equip-
ment are detected, with failed equipment taken out of use for maintenance or replaced.
So, most equipment will require compliance verification testing at regular intervals.
Appropriate test methods and pass criteria must be defined. This is discussed further in
Chapter 9, and some test methods are discussed in Chapter 11.
standards. The standards may be to do with ESD control or any other aspect such as safety.
For example, footwear that is personal protective equipment (PPE) may, in Europe, need to
be compliant with a standard such as ISO 20345.
In some cases, it may be advisable to obtain one or more specimens of items for evalua-
tion or test. This may be done in-house or by a third party. As they are done only initially to
qualify the product before use, the qualification tests may be more in-depth than compli-
ance verification tests. They often are designed to check individual characteristics of parts
of the equipment in detail. For example, in qualifying an ESD control chair, it may be desir-
able to check that the seat back and arms as well as the seat are made of static dissipative
material and connected reliably together and to a groundable point, feet, or wheels. It may
be necessary to establish that more than one foot or wheel is made of noninsulating material
and capable of establishing a ground path to the floor.
In qualification tests, it is sometimes desirable to test the item isolated from its normal
working situation in an uninstalled condition. For example, a work surface point-to-point
surface resistance may be tested as well as the resistance from the surface to a groundable
point. The latter test does not require connection to ground. Often, it will be desirable to
qualify the item under “worst case” operating conditions. This usually means performing
tests under dry atmosphere <30% rh.
Conversely, in some qualification tests, it is desirable to test the operation of the equip-
ment as a system with other ESD control equipment. A good example of this is measurement
of body voltage generated on a subject wearing ESD control footwear and the flooring with
which it will be used. This test can be done only with representative samples of the footwear
and flooring under qualification (see Section 7.5.4).
Where compliance with a standard is required, testing should be done according
to the test methods required by the standard, and compliance with the standard pass
criteria should be established. Standards will often specify qualification testing under dry
atmosphere conditions. IEC 61340-5-1 and ANSI/ESD S20.20 usually require qualification
testing at 12 ± 3% rh (International Electrotechnical Commission 1998, 2016b; EOS/ESD
Association Inc 2014b).
7.1.2 Use
The intended use of equipment can be an important consideration, first in whether it is
needed at all and second in selecting its required characteristics. ESD control equipment
often acts as part of a system with other equipment. Careful thought should be given to
how the equipment is used and if it will help to give a convenient working environment
to the personnel in the area. Analysis of this can affect the combination of the ESD control
equipment selected. ESD control practices that are selected to be convenient to the operator
are more likely to be reliable in practice. This is discussed further in Chapter 10.
surfaces may need different cleaning procedures and materials than those in unprotected
areas, selected to preserve and enhance their ESD control properties.
Some equipment such as ionizers may need regular maintenance to keep their
performance in the long term.
Equipment that relies on grounding through feet or wheels often collects dirt on these
contact points, which can make the ground path fail or intermittent. Cleaning the wheels
or feet can bring the equipment back into compliance.
It is undesirable to have more than one earth present in a facility. If there are multiple
earths and they are not connected together, they are likely to be at different voltages due
to earth currents or other phenomena. Often, there can be a variety of equipment in the
EPA that is already connected to mains protective “safety” earth. In this case, it is usually
convenient to also use mains protective earth as the ESD control earth. More information
on electrical power systems can be found in standards such as IEC 60364-1 and national
electrical codes of practice and regulations (International Electrotechnical Commission
2005).
If there is no mains protective earth available in the EPA, it can be convenient to use a
separate earth rod buried in the earth as the ESD control earth.
It is preferable to connect individual equipment (e.g. workstations) separately to ground
in “star” connection form (Figure 7.1) rather than “chain” or series connection (Figure 7.2).
This is because with the “chain” connection, failure of one connection can lead to several
ESD control items being disconnected from earth.
An exception to this is where an ESD control item (e.g. ESD floor) is designed to provide
a ground path for equipment (e.g. chairs or carts/trolleys) or personnel grounded through
it as a grounding system.
A floor also acts as a means of conjoining workstations or areas to make a single EPA
with no uncontrolled area between them. This can improve the convenience of handling of
electrostatic discharge–sensitive (ESDS) by eliminating the need for protective packaging
for transport between workstations.
The floor operates as a system with the items that it is intended to ground. So, in specifying
the characteristics of a floor, it is necessary to consider the items that will be grounded by
it. Typical systems using the floor include
Floors and floor covering materials of course have many non-ESD-related functions. They
must withstand the physical wear and tear expected for the lifetime required of the floor.
This may include considerations like use of forklift trucks and vehicles, or withstanding use
of chemicals. Floor materials for use in clean areas may need to be selected for low contri-
bution to particulate or other contaminants in the area. These factors must be included in
the overall selection of the floor material.
Floor mats can be an easy and portable solution, especially for small areas, for example
around a single workstation. They can also provide a replaceable “sacrificial” surface
around an area where high contamination or solder damage is expected. Comfort floor
mats may be provided where operators are standing for long periods at workstations.
Some types of mat can tend to curl and ruck and form a trip hazard. They are often con-
nected to ground via a wire attached to a bonding point, and this may act as a trip hazard
or be prone to accidental disconnection.
Paints and coatings can be applied to existing floors such as concrete, giving easy applica-
tion and coverage of large areas. They have intermediate lifetime and tend to wear, requiring
reapplication. Some materials may not be suitable for clean areas due to their carbon load-
ing or tendency to abrade and shed particles.
Floor finishes such as topical antistats are treatments that are easy to apply and can be
applied to carpets. Their effect is usually temporary and requires regular refreshment. Their
effect often depends on moisture from the air, and they may become ineffective in low rel-
ative humidity atmospheric conditions. Some finishes may tend to be slippery and cause
slip hazard to personnel. Some may be susceptible to easily being washed off or worn away.
Careless application and maintenance may result in unreliable performance. Some finishes
may not be usable in clean room areas due to contamination issues.
installation conditions. For this reason, it can be advisable to test the proposed material
as a small installed area before deciding to install a large area of material. Some materials
(e.g. treatments and coatings) can by their nature be tested only in the installed condition.
There may also be concerns over whether a treatment could affect other aspects such
as the appearance of a treated material. In this case, it may be advisable to test a small
area before treating the full proposed application area. Qualification test on an installed
floor material will usually have to be done under ambient conditions. If possible, some
tests should be done under worst-case (e.g. low-humidity) conditions. Testing typically
includes
● Resistance to ground of the installed material, made at several points on the surface
● Resistance from body to ground (or groundable point) of a person wearing selected ESD
control footwear intended to be used in practice and standing on the surface
● A walk test on the specimen by a subject wearing the footwear intended to be used
in practice to test the body voltage generation of the selected footwear-flooring
combination(s)
ESD control standards may require these tests to be done as part of their ESD control
product qualification requirements. Where compliance with a standard is required, testing
should be done according to the standard, and compliance with the standard pass criteria
should be established.
It is important to use a cleaning regime that will preserve and maintain the floor elec-
trostatic characteristics. Use of incorrect cleaning materials or polish can leave a coating
of wax or another contaminant material on it. This can seriously affect the surface charge
generation characteristics as well as the surface contact resistance of the floor. Guidance
should be obtained from the material manufacturer or supplier regarding suitable cleaning
methods and materials.
Where water or liquid cleaners have been used, leave to dry well before retest of the floor,
as the presence of residual moisture can dramatically affect the measured resistance value.
cleaning action while being quick in drying. The electrode should be allowed to dry com-
pletely before further measurements are made.
If areas of excessive resistance to ground are found on a floor, the area should be first
cleaned to check whether it may be due to contamination. If this does not restore the floor
performance, an evaluation should be made of the size of the area affected, whether the
performance has changed, whether low humidity may be a factor, and whether there may
be other areas in the process of deterioration. Depending on the results of this evaluation,
remedial action may need to be considered.
Small areas of high resistance can sometimes be due to a previously undetected problem
in a floor material or installation. If the area affected is very small and the effect on ESD con-
trol is minimal, then the position of the problem area may be noted, and the noncompliance
might be acceptable as insignificant. A key question here is whether the noncompli-
ant area prevents reliable grounding of an ESD control item, leading to ESD risk to
ESDS devices.
Figure 7.3 A typical earth bonding (grounding) plugs that can be used to connect to mains
protective earth.
190 7 Selection, Use, Care, and Maintenance of Equipment and Materials for ESD Control
If practical, the best option for both safety and ESD control is to prevent personnel from
touching the ESDS system while high voltages are present. Handling the system with no
voltages present is safe, providing there are no built-in energy sources such as high-voltage
batteries or charged capacitors.
When working in high-voltage systems, personnel may be required to wear high-voltage
protection gloves. These are an effective barrier to human body ESD at moderate levels
as well as essential protection for electrical safety, although ESD risk due to triboelectric
charging of the gloves may need evaluation.
The wrist band is a flexible band like a bracelet or watch strap that is worn in contact
with the wearer’s skin and must make good electrical contact with the skin. These bands
are normally worn on the wrist but are occasionally worn in contact with other parts of the
arms or legs. A ground cord makes electrical connection between the wrist band and an
earth bonding point. The earth bonding point provides a point where the ground cord can
be connected directly to EPA ground (earth).
Wrist bands come in various forms such as knitted or woven fabric, expanding metal
bracelet, or resin bracelet. Stick-on patches are also available. Wrist bands usually include
a hypoallergenic metal plate against the skin to promote good skin contact. The band inner
surface is also usually conductive to give electrical contact around nearly all the inner
surface of the band. The wrist band has a quick-release connection point for the ground
cord. This is designed to give a reliable connection strong enough to prevent accidental
disconnection but light enough for easy release. The disconnect force is normally in the
range 13–36 N.
The ground cord is an insulated wire with a connector on each end, used to connect the
wrist band groundable point to an earth bonding point. The ground cord includes various
features for safety and other considerations. The cord usually contains a current limiting
resistor at the wrist band end. Some types include a resistor at both ends. The cord wire must
be designed to be flexible and withstand constant flexing and dragging over equipment and
workstation edges. They are available in a range of length and colors and straight or coiled
wire construction. The earth bonding point connector is commonly a snap connector or
4 mm plug, although any suitable connector may be used.
The wrist strap system earth bonding point can be any suitable and reliable direct elec-
trical connection point to EPA earth. Dedicated earth bonding points are often provided on
workstations or equipment. It is not good practice to clip a wrist strap ground cord to a work-
station mat or other item that might introduce considerable addition resistance between
the ground cord and earth. Dedicated earth bonding points may include some resistance
providing the total wrist strap system resistance to ground does not exceed required upper
limits. Current standards IEC 61340-5-1 and ESD S20.20 specify an upper limit for the wrist
strap system resistance to ground of 35 MΩ although this may be changed according to the
user’s requirements.
Capacitance constant monitors use a single-wire wrist strap system. They detect the
person-earth capacitance of the person wearing the wrist strap using an alternating current
(AC) sensing signal applied to the system. Capacitance changes outside the set limits of the
monitoring system cause an alarm. Some disadvantages are that the monitor must be set
up for the individual using it and that false alarms may occur due to capacitance changes
rather than failure of the wrist strap system.
Impedance constant monitors are single-wire systems that use similar principles to the
capacitance monitor but detect circuit impedance changes rather than capacitance changes.
This eliminates the need for adjustment and can reduce false alarms.
Resistance constant monitors use a two-wire wrist strap system, measuring the resistance
of an out and return circuit through the wrist strap system and the wearer’s body. An alarm
sounds if the direct current (DC) resistance of this loop goes outside the accepted range.
The system may use a constant or pulsed DC voltage up to 16 V to make the measurement.
In most monitor designs of this type, the maximum voltage is evident only during an alarm
condition. The voltage is generally considerably less under normal conditions when the
person’s resistance is well under the set limit of the monitor. Some users may experience
skin irritation when a DC voltage is used.
Body voltage constant monitors use a two-wire system to measure the body voltage of
the user. This system may not detect an open circuit that results in apparently zero body
voltage. In some systems, an impedance or resistance monitor is incorporated to detect this
situation.
Each type of constant monitoring system requires an appropriate tester to verify the oper-
ation of the system. The single- and dual-wire wrist strap systems used with the monitors
can also be tested using resistance measurements.
The intended reliability and durability of the wrist strap system should be considered.
The wrist band should be comfortable for the length of time the user is expected to use it.
Discomfort might encourage users to remove the strap and forget or omit to wear it while
handling ESDS.
Many wrist straps include a metal plate that contacts the skin under the ground cord
bonding plug, improving contact reliability. The rest of the inside of the band is normally
also conductive to give good all-round skin contact.
The length of ground cord, and whether it should be retractable, should be considered.
The type of earth bonding point connector should normally be standardized throughout the
facility so that there is no difficulty in using it in different areas as required.
If constant monitoring is to be used, it may be necessary to select a wrist strap system that
is designed specifically for use with the monitoring system.
Figure 7.5 Examples of proprietary portable wrist strap testers. Source (left): D E Swenson.
The wrist band and cord are normally tested as a system as worn, using a proprietary tester
(Figure 7.5). These are available as portable or wall-mounted testers, some of which can also
test footwear. The tester does a simple resistance measurement to verify the resistance from
the wearer’s body through to the cord groundable point is within acceptable maximum and
minimum limits. These limits are specified in the standards, although the user can select
different pass criteria for their specific circumstances. The resistance of the skin to band
contact is also tested in this measurement and can be highly variable. For a person with dry
skin, this resistance can be high and result in failure of the test.
In practice, these tests are usually done by personnel for themselves using proprietary
wrist strap checking instruments. These typically have built-in resistance checking, and
display pass and fail as colored lights. Some sophisticated systems may use this to control
access to the EPA via an entrance turnstile and automatically maintain compliance data.
In other facilities, the user may be required to record their wrist strap checking pass result
and relies on them to take remedial action if a failure is indicated.
With these instruments, the pass criteria are effectively provided by the instrument and
may or may not be adjustable. It is important to ensure that the pass criteria are aligned with
the ESD control program requirements. In current “off-the-shelf” testers, the upper limit is
often set at 35 MΩ with a lower limit of about 750 kΩ. These limits, and the possibility of
adjusting them if necessary, should be checked when qualifying a tester for use in an ESD
program.
It is also possible and acceptable to measure the wrist strap resistance from body to the
cord groundable point using a suitable resistance meter. The resistance meter must be
selected to be safe for use in this way, with limited output current to prevent electrical
shock risk. The measurement can be done with many types of instrument at 10–100 V test
voltage. The voltage used may depend on compliance with a standard or safety regulations.
Contact to the body should be made via a handheld electrode giving a reasonably large
skin contact area. Holding a small-point probe may not give adequate contact area for
reliable measurements.
7.5 Personal Grounding 197
It can be useful to keep failure data for the types of wrist straps in use, including for
example identification of skin contact, ground cord or connector failure, as well as manu-
facturer and type data. This allows the reliability of different wrist strap systems and their
components under operating conditions to be compared. The more reliable components
can be identified and selected for future purchase.
Testing the EPA wrist strap earth bonding points must also be done periodically to ensure
that any failure of these is discovered sufficiently quickly. This can be done using a sim-
ple resistance to ground test (see Section 11.8.3.5). A maximum resistance to ground for
earth bonding points, and a suitable test method, should be specified in the ESD program’s
Compliance Verification Plan.
The frequency of wrist strap testing can be determined according to factors such as the
level of use and established reliability of the wrist strap system. ESDS sensitivity and value
of the product handled, and possible consequences and cost of product failure, are also
relevant factors. If a wrist strap fails, any ESDS handled between failure and detection of
the failure might be damaged.
As personal grounding is of major importance in ESD control, it is common for the wrist
strap system to be tested at least each working day before the wearer commences handling
ESDS devices. In some ESD programs, wrist straps are tested every time the wearer enters
the EPA. When it is considered highly important to avoid risk of product damage and is
wished to immediately detect any wrist strap system failure, constant monitoring can be
specified.
Figure 7.8 A wide variety of shoe and work boot styles are available.
Electrical properties of footwear are classified under ISO 20345 as electrically “insu-
lating,” “antistatic,” or “conductive” depending on their electrical resistance. Antistatic
footwear has resistance between 100 kΩ and 1000 MΩ, and conductive footwear has
resistance less than 100 kΩ measured using ISO 20344 (International Organisation for
Standardisation 2011a,b). This test method is like, but slightly different from, test methods
used for ESD control footwear. The conditioning and test atmosphere conditions may also
be different. The result is that most types of antistatic and conductive footwear might be
suitable for EPA use, but they are not tested using the test methods and conditions required
for use in EPAs. PPE footwear that has not been qualified for EPA use should be tested and
qualified using appropriate tests.
or EOS/ESD Association Inc 2015). If the footwear is to be used with several types of floor
material, it should be qualified for use with all of them.
Qualification tests should, where possible, be tested with the flooring under the
worst-case conditions that can be envisaged during operation. This usually means testing
under low-humidity atmospheric conditions in a laboratory. Where a minimum resistance
for electrical safety is required, it may be necessary to establish performance under damp
conditions.
Where qualification with an existing floor is required, it is often not possible to test
the footwear-flooring combination in the laboratory. Qualification should be done under
the worst-case conditions practical under the circumstances. Confirmation of perfor-
mance under low-humidity atmospheric conditions when these arise is advisable. These
conditions often occur in the winter in colder climates.
The footwear should also be checked with the intended compliance verification test
equipment to make sure they reliably pass the intended compliance verification tests.
Where compliance with a standard is required, testing should be done according to the
standard and compliance with the standard pass criteria should be established.
Sampling and test of incoming ESD control footwear is advisable.
measurement plate. Proprietary testers are commonly available. These are often pro-
vided at the entrance to EPAs so that personnel can test their footwear before entry
to the EPA.
A disadvantage of many testers is that they only give a pass/fail indication rather than
a resistance value. Some automated data logging types also record the resistance value
measured. If necessary, a resistance measurement may be made using a suitable resistance
meter (see Section 11.8.3.3). The pass criteria of the footwear tester should be aligned with
the requirements of the ESD Control Program Plan documents.
The simplest EPAs for field work may consist of little more than a portable workstation
surface, personal grounding, and a connection to ground. In a simple case like this, the
boundary of the work surface may define the boundary of the EPA.
The activities and processes and types of ESDS handled, along with whether it is a tempo-
rary or permanent facility, will probably largely determine the type of work surface selected.
For a field service workstation mat, compact storage and portability in a tool kit may be an
overriding consideration.
Durability considerations include the hardness, abrasion, and tear resistance of the
surface material. Aspects such as chemical (e.g. solvent), heat, and solder resistance may
also be a concern.
One significant point will be to decide whether a soft and cushioning surface is needed for
handling the ESDS device or a durable surface is needed for handling heavy or sharp items.
Even on a permanent workstation there may be a reason to use a temporary work surface.
Some ESD programs use a replaceable mat surface as a sacrificial cover where chemical
contamination or physical damage is a problem. Others use a mat to modify the surface on
some workstations, e.g. use of a cushioning mat on workstations that have a hard surface.
For mats, curling with age or due to storage in rolled or folded form may be a concern.
Where items such as cutting mats or trays are placed on the workstation surface, they
should comply with the requirements for worksurfaces.
If the workstation is in a clean room, this may restrict the work surface type to clean
room–compatible materials.
In workstations that use high voltages or handle ESDS devices containing batteries or
power sources, it may be necessary to use a work surface that has a specified minimum resis-
tance. The minimum resistance may be specified for safety, power drain, or short-circuit
elimination considerations. Where electrical safety is a concern, appropriate tests and speci-
fications must be selected that comply with local regulations. DC resistance measurements
as made for ESD control purposes may be inadequate for safety testing, especially where
high-voltage AC is present.
In a test workstation, proximity or contact with a conductive work surface could adversely
affect the operation of the ESDS under test.
A minimum resistance of the work surface may also be required to control a risk of
charged device ESD to ESDS devices placed on the surface (see Section 4.7.5). A metal
tray placed on a workstation surface can give charged device ESD risk if an ESDS device
is placed upon it.
Appearance can be a consideration in selecting a work surface. Colors can be used to
identify workstations for certain operations or for corporate identity identification.
common example is a metal tray placed on the bench surface can introduce charged device
ESD risk to ESDS devices placed on the tray.
unprotected area shelves or placing of insulative items on EPA shelves, causing ESD risks.
EPA racks and shelves should be maintained a suitable distance from any unprotected
area activities or storage. A distance of 0.5 m will normally reduce any electrostatic
fields from charged insulative materials to a sufficiently low level to minimize ESD
risks to all but extremely sensitive ESDS devices. A greater distance may give additional
confidence.
Factors indicating that racks may be better specified as EPA might include
● It would be convenient to store ESDS devices without packaging them in ESD protective
packaging.
● The items are to be taken to workstations in the same EPA after storage.
● The shelf can be sited for access without leaving an EPA. (If the floor between EPA work-
stations and the shelf is not an ESD control floor, transport to the shelf is effectively
through an unprotected area.)
● Non-ESDS and insulative items can be stored on a separate shelf unit.
Drawer systems and carousels used to store unprotected ESDS can often be considered to
have the function of ESD control packaging and can be specified similarly.
The care and maintenance aspects are like those for workstations.
7.7 Storage Racks and Shelves 209
Some types of moveable shelf or rack systems can be difficult to distinguish visually from
similar systems not designed for EPA use. An example of this may be metal racks that have
plastic components joining the shelves to the frames. This can lead to unsuitable units being
inadvertently substituted for EPA shelf units during a work area rearrangement.
Mobile equipment and carts can be grounded through noninsulative wheels, a drag chain,
or some other means to an ESD control floor. These can provide a convenient means of
grounding the cart if it is standing on the ESD control floor and has the advantage that it
needs no operator action. Qualification tests should demonstrate that grounding is effec-
tive when the equipment is standing on all the ESD control floors with which the cart will
be used.
Where there is no ESD control floor (e.g. at a stand-alone workstation), carts and mobile
equipment should be grounded via an earth bonding wire before and during loading and
unloading.
For mobile PCB racks or other items that are specified like ESD control packaging, the
exposed surfaces should be tested to confirm that they are static dissipative or conductive.
Resistance to groundable point tests should be made to ensure that the surfaces are bonded
to the groundable points.
Any covers, doors, or sides to the cart or mobile equipment should be tested to make
sure they do not have exposed insulating surfaces that could charge up and give significant
electrostatic fields.
Where compliance with a standard is required, testing should be done according to the
standard, and compliance with the standard pass criteria should be established.
7.9 Seats
7.9.1 What Is an ESD Control Seat for?
An ordinary chair not designed for ESD control usually has wheels or feet, structural parts,
and cover material made of insulating materials. The covering material and structural part
may become very highly charged in contact with the user’s outer clothing and by movement
of the chair over the floor. A person sitting on a highly charged seat might, if inadequately
grounded, rise to high induced voltage and be a source of ESD risk. The design of an ESD
control seat addresses these problems (see Section 4.7.10.5).
Wheeled seats can get charged by rolling action of the wheels on the floor. As well as
being a strong source of electrostatic fields, ordinary chairs can generate internal ESD that
can radiate electromagnetic interference (EMI) and interfere with electronic equipment
(Smith 1993, 1999).
An ESD control seat has a noninsulating covering material that can dissipate the charge
generated by contact with the user’s clothes. The structure of the chair is designed so that
an electrical connection is provided between the covering material and exposed chair parts
through to ground, usually via the feet or wheels or a grounding means such as a drag chain.
A common misconception is that an ESD seat is designed to ground personnel sitting on
the seat. In most ESD programs and current ESD standards, this is not so. One reason for
this is that it can be difficult to ensure reliable connection between the user’s body and the
seat through layers of intervening clothing. It would be difficult in many cases to ensure
that a sufficiently low-resistance ground connection and low body voltage is maintained.
Sitting personnel must normally be grounded via a wrist strap (see Section 7.4). Neverthe-
less, personnel have sometimes been successfully grounded via seating. This is discussed
further in Section 7.9.8.
Where wheels or feet are used for grounding to an ESD control floor, it is advisable to
make sure that two or more noninsulative wheels or feet are provided. This will provide a
level of redundancy to ensure that grounding is maintained even if a patch of contamination
on a floor, wheel, or foot prevents good contact.
ESD control seats can be almost indistinguishable in appearance from ordinary seats. If
confusion with non-ESD control seating could be possible during use, consider identifying
the seats with appropriate markings.
in many cases to ensure that a sufficiently low-resistance ground connection and low body
voltage is reliably maintained.
The resistance to ground through many types of seat is rather high for grounding person-
nel, with up to 109 Ω being allowed by current standards. In contrast, the limit required for
wrist strap grounding is set at 35 MΩ. Current standards require that sitting personnel must
normally be grounded via a wrist strap.
If seating is to be used for grounding personnel, these potential problems must be
addressed. The first difficulty is to ensure that reliable connection can be established
between the chair and the user. This may be easier in warm climates where fewer clothes
are worn because more skin may be exposed on the body where it will contact the chair. In
warmer temperatures, a greater level of body moisture helps make contact through clothing
fabric. In colder climates where several layers of clothes may be worn and low-humidity
conditions may arise, it can be difficult to establish reliable contact between the body and
the chair.
The second difficulty is to ensure that the chair provides reliable grounding to the ESD
control floor surface. A chair foot or wheel may contact a very small area of the floor and
is prone to buildup of dirt between the foot/wheel and the floor. This is particularly so in
areas where dust and contamination may be a concern.
As current standards do not use seats for grounding personnel, compliance would require
a tailoring statement giving the technical basis and rationale, backed by data, establishing
that the practice is acceptable. Qualification and compliance verification test methods and
pass criteria for suitable seating would need to be established as these are not provided by
standards in this case. For these reasons, a study would need to be done to establish the
reliability of grounding personnel through seating and establishing suitable tests and pass
criteria. The variation of resistance from the user’s body to ground and body voltage should
be studied over a range of temperature and humidity conditions. Cold and dry conditions
may provide a “worst case.” This and any other anticipated worst-case conditions should
be carefully studied. The results of the study should be documented in a report that can
be summarized and referenced in the ESD program documents and held on file for later
reference. An explanation of the practice and reference to the test report should be included
in the ESD program documents.
If seats are used for grounding personnel, then the body voltage must be maintained
less than the maximum required in the ESD program. This should be established using
body voltage measurements under simulated worst-case operating conditions including
low-humidity air conditions. One problem is that unless outer clothing is specified (e.g.
by use of ESD coats), the charge generation properties of outer clothing against the seat
cover material can be highly variable with clothing material changes.
If several types of seat are used, all these should be tested and qualified, and each seat type
to be used should be type tested to demonstrate that body voltage is adequately controlled
and establish the range for acceptable resistance to ground. Seats purchased for use should
then be restricted to the types tested and qualified.
Suitable test methods for compliance verification and pass criteria should be considered.
A simple resistance to ground test of the seat would not establish grounding of the user.
A resistance test from the body of the person sitting on the seat to ground would be needed to
verify adequate connection (Section 11.8.3). A body voltage measurement based on Section
11.8.9, made with the subject simulating normal working activity, would also be advisable.
7.10 Ionizers 217
7.10 Ionizers
7.10.1 What Does an Ionizer Do?
Ionizers are commonly used for reducing charge levels, voltages, and electrostatic fields
on essential insulators in use near ESDS devices (see Section 4.6.3). The primary means of
controlling voltages on conductors is to ensure that they are reliably grounded, but where
grounding cannot be used, it may be possible to use an ionizer to reduce voltages on conduc-
tors to an acceptable level. Reduction of electrostatic fields and voltages can be important for
reducing ESD risk and for reducing electrostatic attraction (ESA) of contaminant particles
to items in clean areas (Section 4.7.9).
An ionizer works by “spraying” positive and negative charges into the air as ions (see
Section 2.8). Opposite polarity ions drift under electrostatic fields to the surfaces that are the
sources of the fields. Arriving at the surface, they neutralize excess charges of the opposite
polarity or reduce the voltage by accumulating there.
The ions are produced from the air present in the environment. Producing air ions
in an ionizer does not produce any contamination and so ionization is suitable for
clean room use, although the erosion of emitter points in electrical ionizers should be
considered.
Use of ionization to neutralize charge and voltages suffers two main disadvantages. First,
it takes time to neutralize the charge or voltages, limited by the rate of arrival of ions at the
surface being neutralized. This can take many seconds. If charges are being generated at a
greater rate than can be neutralized by the arrival of air ions, the charge neutralization is
unsuccessful.
Second, depending on the type of ionizer used, the balance of positive and negative ions
produced by most types of ionizer is often not exactly balanced. This results in charging of
all ungrounded items or materials within the ionized region to a small offset voltage. The
user must establish that this offset voltage is insignificant for their purpose. The ionizer
imbalance typically changes with time and age.
Opposite polarity air ions attract each other due to their opposite charge. This means that
they tend to drift toward each other and recombine to form neutral molecules or particles.
The rate of arrival of ions, and hence charge or voltage reduction, can be highly dependent
on distance between the ionizer and the surface, ionizer orientation, local air flows, and
other factors. Effective neutralization occurs within a limited region around the output of
the ionizer and typically takes longer for greater distance from the ionizer. If an item is
within the ionized region for insufficient time, the surface voltages will not be completely
neutralized and may remain to cause ESD risk. In specifying an ionization system, it is
important to understand the size and position of the effective neutralization region, the
neutralization time, and the offset voltage characteristics.
to produce a well-balanced air ion source. X-ray ionizers use “soft” X-rays for the same
purpose.
Nuclear ionizers commonly use a polonium 210 nuclear source. This is packaged for
nuclear safety to prevent escape of the nuclear material but allows emitted alpha particles
to be released. These alpha particles (helium nuclei) collide with and split gas molecules
in the air to produce equal numbers of positive and negative ions and a well-balanced ion
source. Nuclear ionizers do not contain high voltages and so do not produce electrostatic
fields. They are unusual in that they produce a completely balanced ion stream and so have
zero offset voltage. One disadvantage is that the rate of ion generation is limited by the
rate of nuclear disintegration. The ionizer may be effective only for a short range although
airflow can be used to help distribution of the ions produced.
AC ionizers use AC high voltages applied to corona needle electrodes to produce positive
and negative ions alternately from the same electrodes. The high voltage AC is usually at
mains power frequency. As the ions of both polarities are produced at the same electrodes,
they can quickly recombine. A fan or other air movement system is often used to transport
the ions to the intended effective neutralization region.
DC systems have positive and negative DC high-voltage sources applied to separate
corona discharge needles. These needles are placed at some distance apart. Loss of ions
due to recombination is typically less than for AC systems. Lower air flow may then be
used to transport the ions through the intended region of effective neutralization. If the
ionization electrodes are far apart, the ion balance close to the electrodes may be poor,
resulting in high offset voltages close to the ionizer.
Pulsed DC ionizers used the same electrodes with both polarities of high voltage alter-
nately to generate both polarities of ions. The voltage is applied with a relatively low alter-
nating pulse rate, usually below 10 Hz. This results in a lower recombination rate than AC
systems, but the offset voltage may cycle negative and positive especially close to the ionizer
emitter electrodes. One advantage is that at the high voltage pulse rate, delivery of ions can
be adjusted to improve performance over the intended effective neutralization region.
Soft X-ray ionizers use X-ray sources to ionize the air, producing a well-balanced stream
of positive and negative ions. The ions are produced along the beam of the X-ray, which
can extend a meter or so from the ionizer. Personnel must be shielded from exposure to the
X-ray radiation. Advantages of the system are that well-balanced source of ions is produced
in a defined region without electric fields or air flow.
electrostatic charging that can cause ESD and contamination (particle attraction) within
these workstations. AC, DC, pulsed DC, or nuclear ion sources may be used.
Work surface ionizers are used to control electrostatic charging in specific work surface
regions. AC, DC, pulsed DC, or nuclear ion sources may be used. Common types include
fans to take ions into the controlled region.
Where electrostatic control is needed in a small specific region within a process or
machine, point-of-use ionizers may be used. Compressed air nozzles may incorporate
ionizers to reduce electrostatic charge during blow-off particle removal. They may include
AC, DC, pulsed DC, X-ray, or nuclear ion sources.
Some systems use feedback to adjust ion balance and improve performance. These rely on
sensors to sense ion balance in the region around them. The sensor can assure balance only
in a limited region around it. Where balance is required over a large area, it may be better to
use several systems, each having their own sensor controlling smaller regions, rather than
a single system controlling a large area. Use of a feedback-controlled system can reduce
maintenance by automatically adjusting for changing emitter efficiency.
of an ionizer can be considerably reduced. Some types of ionizers include fans to assist
air flow. The work area should be arranged to remove any obstruction to air flow to the
neutralization target.
Electrical ionizers use high voltages and sharp points. If insufficiently guarded, these can
give electrical shock or accidental injury risks to personnel. Electrical ionizers generate a
small amount of ozone that must not be allowed to become a threat to the health of per-
sonnel or to processes. Ionizers must meet local and national electrical, X-ray or nuclear
safety, ozone generation, or other regulations. Electrical ionizers also generate EMI that
may interfere with sensitive electronic equipment.
In summary, when selecting an ionizer, it may be necessary to consider the following:
● The application and task of the ionizer
● ESDS device withstand voltage
● Required ionizer offset voltage
● The size and position of the region to be controlled
● Whether the process can be modified to make the ionizer’s task easier or more reliable
● High voltages and electrostatic fields generated, and the regions and levels of ion imbal-
ance, especially near ionizer electrodes
● Environmental aspects such as airflow and blocks to ionized airflow
● Use of airflow to enhance neutralization
● Safety issues and regulations
● Maintenance requirements
● Installation, operation, and maintenance costs as well as equipment cost
● Need for associated facilities such as air supply
● Cleanliness and contamination, especially where air supplies and clean areas are involved
● Ozone and EMI effects (for electrical ionizers)
● For large systems, power distribution requirements
Nuclear ionizers should not be used in a chemical environment without consulting the
manufacturer.
Tests should also be performed with the ionizer working in the process and environment
in which it will be used, testing performance at positions that represent the range of work-
ing positions of the ESDS devices or other items to be neutralized. Specific and nonstandard
tests should be devised to demonstrate that the ionizer reliably fulfills the task intended.
Tests might include, for example, investigation of the charge decay time and residual volt-
ages on an insulating item, an ESDS device, or an isolated conductor within the process
under conditions as close as possible to operating conditions. During in-situ qualification,
it may be useful to determine a suitable test arrangement or CPM location for compliance
verification testing. In some circumstances (e.g. within operating machines), it may be dif-
ficult or not viable to make measurements under working conditions.
It may be necessary to use nonstandard CPM instruments as well as other instruments
such as electrostatic voltmeters or field meters to quantify operation under real conditions.
As well as electrostatic tests, it may be necessary to test other aspects of ionizer performance
such as ozone buildup, particulate contamination, or EMI.
Where the ionizer is protecting against low-probability ESD events that may happen in
running production circumstances, the final proof of effectiveness may be demonstration
that it solves the identified problem and does not introduce another.
Figure 7.11 Typical ceiling mounted pulsed DC ionizer charging of an object at tabletop height.
Source: D E Swenson.
used in a situation where the performance is less critical or where verification history has
shown reliable and consistent performance.
Table 7.1 Types and characteristics of ESD control garments according to IEC 61340-4-9:2016.
Grounded via a single wire Groundable static Resistance point to point <109
continuous monitoring control garment Resistance to groundable point
equipment system (garment in
combination with a Integrated wrist strap in <3.5 × 107
person) accordance with IEC 61340-4-6
Disposable garments, as the name implies, are intended to be used a small number of
times. They are typically made from nonwoven materials. ESD control properties are often
provided by finishes or treatments that have a limited life. Their properties may be depen-
dent on atmospheric humidity and be impaired under low-humidity conditions. They may
be useful in processes where contamination gives a short garment life and a need for fre-
quent replacement.
Reusable topically treated garments may require retreatment after each cleaning.
Treatments typically depend on atmospheric humidity and become inadequate under
low-humidity conditions. Some materials have conductive fibers but also require topical
treatment. For these materials, the properties may be less dependent on humidity.
Garments with permanent ESD control properties should maintain their ESD control
properties for their intended life. They usually have a grid or stripes of conductive fibers
woven into the fabric (Figures 7.12 and 7.13). If not grounded, exposed conductive fibers
could represent an ESD source.
Static control garments can be made in various materials. Some are disposable, whereas
others are reusable. They may be made from homogenous untreated fabrics or coated fab-
rics or may have conductive threads aligned in one direction only or in two directions as
a grid.
Groundable static control garments include a means of grounding the conductive
fibers in the fabric through one or more groundable points. The panels of these garments
should be electrically connected. Fabrics that have a grid pattern may produce the most
reliable connection. Connection between panels may be improved by including additional
non-insulative (electrically conducting) material in the seams. The garment can be
grounded via the wearer’s body through conductive wrist cuffs or other areas of direct
Conductive threads
Textile Textile
Figure 7.12 Stripe (left) and grid (right) conductive fiber patterns (Paasi et al. 2005a,b).
Figure 7.13 Examples of ESD control garment fiber types (Paasi et al. 2005a,b).
7.11 ESD Control Garments 225
contact with the wearer’s skin, providing the wearer is grounded through a wrist strap or
footwear-flooring system. Alternatively, grounding may be achieved via a ground cord.
Fabrics that have a stripe pattern rather than grid may not have a good connection
between the conductive fibers. These may need topical treatments or conductive material
in seams to improve connection between the fibers.
Fabrics that depend on topical treatments can degrade unevenly and leave areas of fabric
isolated from the ground connection.
Some experts have been dissatisfied with measurement of garment materials only on
the basis of material resistance. Buried conducting fiber materials are usually rejected by
resistance measurements as the measurement electrodes do not make contact with the con-
ducting fiber. These simple tests do not characterize many of the properties of garment
materials (Paasi et al. 2004, Baumgartner 2000). Potentially relevant factors that are ignored
by the resistance test method include
In the early 2000s the ESTAT garments European Project researched the importance of
these factors and produced recommendations for use and test of ESD-protective garments,
considering surface and core conducting fiber and stainless steel thread materials. Their
recommendations have not been widely adopted in ESD standardization, but their final
report makes interesting reading (Paasi et al. 2005b). Their results broadly confirmed the
value of simple resistance measurements in evaluating groundable garment materials.
They found that the ESD-protective fabric and garment characteristics were very depen-
dent on the fabric and garment design, atmospheric humidity, and whether the garment
and the wearer were grounded. The main functions of an ESD control garment are
Charging of ESDS devices by accidental rubbing against the garment fabric could not be
easily averted by choice of fabric but could be reduced by minimizing the risk of rubbing, e.g.
by avoiding loosely fitting sleeves. Loose garments have lower protective value than close
fitting garments. ESD could occur from garment materials with ungrounded conducting
threads or sufficiently large insulating areas (over 20 × 20 mm). Most effective ESD control
was achieved for materials having surface resistance in the range 100 kΩ–100 GΩ.
Paasi et al. (2005b) classified garments as Class A (using electrically continuous, low
charging, static dissipative, or conductive materials and grounded) or Class B (low charging
but need not have measurable electrical conductivity, and grounding not required). These
226 7 Selection, Use, Care, and Maintenance of Equipment and Materials for ESD Control
approximately correspond with the “groundable static control garment” and “static control
garment” classifications of IEC 61340-4-9 (Table 7.1).
Grounding the conducting parts of the groundable garment (e.g. to the grounded wearer)
improves the performance of the garment both in terms of external electrostatic fields and
as a potential source of ESD.
Table 7.2 Factors that might be considered in decision to issue ESD control garments.
Factors that might indicate ESD control Factors that might indicate ESD control
garments are desirable garments are unnecessary
Low CDM ESD withstand voltage Components handled have moderate or high
components. ESD withstand voltages and low or moderate
High cost and low withstand voltage ESDS. cost.
Consequence of ESD failure is highly Consequence of occasional ESD failures is
undesirable. acceptable to some extent.
Clean area where ESA is undesirable. ESA is not a significant issue.
Likelihood that high charge generating Likelihood that single layer low-charging
clothing that is not close fitting may be worn clothing will be worn, e.g. warm moist climate.
e.g. cool dry climate.
High reliability product or market where low Consumer market where moderate failure
ppm failure rate is required. rates are accepted.
Use of ESD control garments may help
establish ESD control culture or help
demonstrate best practice and care (e.g. to
visitors).
7.11 ESD Control Garments 227
The main ESD control risk from ordinary garments is that they might charge highly and
give an electrostatic field near ESDS devices. This electrostatic field could induce high volt-
ages on devices, leading to charged device or charged board ESD. ESA can also be a concern
in clean areas, where electrostatic fields can encourage attraction of contamination to the
items handled. In most cases, the areas of the body most likely to come close to ESDS are
the front of the torso and arms. It is therefore particularly important to shield ESDS from
electrostatic fields from clothing in these areas. The clothing of the lower body is often less
important as it is less likely to come near to unprotected ESDS devices.
Woven materials can also release fibers that can give unacceptable contamination in
clean areas. Coveralls used in clean areas can be made entirely from ESD control materials
designed for low fiber release to reduce ESA of particulate contaminants.
Once the decision to use ESD garments is made, a suitable type must be selected. In some
cases (e.g. clean room), this will be a relatively simple matter of specifying an ESD control
version of a garment that is required for a specific purpose (e.g. contamination control or
personal protection against hazards). Ordinary personal protective clothing may be made
from insulative materials that can be a strong source of electrostatic fields.
An ESD control garment should cover the underlying garments completely over the key
areas of the body that need to be covered. As a guide, where the clothes are likely to come
within 30 cm of ESDS devices, they should be covered by an ESD control garment. The
material should be designed to reduce the electrostatic fields generated by underlying gar-
ments. It should not itself be a source of electrostatic fields or ESD. Materials used should
not contain low resistance fibers or metal parts that could charge and become a source of
ESD or could be an electrical safety concern.
According to Paasi et al. (2005b), a dense grid of conductive fibers (e.g. 5 × 5 mm) gives
improved electrostatic field control. For materials with conductive stripes, the stripes
should be less than 10 mm apart. For a grid fabric, the grid squares should be 20 × 20 mm
or less. They found that core conductive fiber fabrics cannot give good grounding of the
fibers under dry air conditions, but this does not necessarily preclude their value in ESD
protective garments. They did not recommend use of a garment as the ground path of the
wearer.
In some countries (e.g. in Europe), garments that have a PPE function are subject to
regulations that require they comply with specific safety-related regulations. Individual
organizations may also have safety rules that must be followed when selecting garments
for use within that organization.
Many garment materials are affected to some extend by atmospheric humidity. Cotton is
inherently hygroscopic, but man-made fibers such as PET and nylon are usually less so. Top-
ical treatments are often highly dependent on humidity for their performance. So, resistance
measurements should be made under controlled low-humidity conditions. Testing over
the full range of temperature and humidity conditions likely in practice may be advisable.
Garments that use fabrics that have core conductive fibers cannot be successfully mea-
sured using surface resistance measurements, as the test electrodes do not make good con-
tact with the buried core conductive fibers. No standard test methods are currently defined
for measurement of these materials for use in electronics environment (Swenson 2011). The
user would need to define suitable test methods and pass criteria for qualification of these
materials.
Paasi et al. recommended that as well as surface resistance tests for qualification of gar-
ments, EN1149-3 Method 1 could be used to evaluate tribocharging, and EN1149-3 Method
2 evaluate induction charging of all types of garment materials (British Standards Institute
2004). Holdstock et al. (2003) have developed a “capacitance loading” test method that can
be used to evaluate charging of garment materials.
Garments made from fabrics that claim permanent ESD control properties should have
tests that compare measurements on new fabrics with measurements on fabrics after
repeated cleaning simulating long-term use. IEC 61340-5-2 recommends typically 50
cleaning cycles should be used.
be tested more frequently that those that include a conductive fiber matrix. For garment
materials that have measurable surface resistance, standard point-to-point resistance test
methods can be used (see Section 11.8.2). For groundable garments, the resistance to the
groundable point should also be tested.
Resistance test methods are not suitable for verification of garment materials that use
buried conductor fibers. If these materials are selected for use, suitable verification test
methods and pass criteria will need to be defined.
The frequency of compliance verification tests should reflect the expected permanence
of the garment fabric technology.
Hand tools are normally made ESD safe by ensuring that metal parts have a grounding
path through to the user’s hand. This usually means that insulative parts of the tool have
been replaced by noninsulative parts.
In some cases, when handling ESDS devices that have low CDM ESD withstand, it may
be necessary to ensure that metal parts (even though grounded) do not contact the ESDS
devices. The materials that contact the ESDS devices may need to be resistive and a min-
imum resistance be specified to reduce the strength of any discharge that may occur on
contact with the ESDS devices.
be observed. Guidance on selection, care, use, and maintenance of PPE in Europe can be
found in CEN/TR 16832, and the gloves may be subject to requirements given in EN16350
(European Committee for Standardisation [CEN] 2014).
Selection of gloves and finger cots should first identify and address the main reasons
for needing them, e.g. personal protection or avoidance of product contamination. These
may well dictate the basic form of glove to be selected. Other considerations may then
include
● Disposable or multiuse
● Physical and safety protection considerations
● Need for grounding of handheld items
● Possible charging of items handled
● Cost and sensitivity of ESDS handled
● Ease and convenience of use
● Suitability for use in clean areas
● Possible consequences of contamination due to antistats
Some types of glove, e.g. thin latex and vinyl gloves can have surprisingly good properties
for ESD control use even though they were not intended for this purpose. If tests confirm
this is the case, qualified types may be usable in the ESD Control Program.
References
Baumgartner G. (2000) ESD TR2.0-01-00. ESD Association Technical Report – Consideration For
Developing ESD Garment Specifications. Rome, NY, EOS/ESD Association Inc.
British Standards Institute. (2004) BS EN 1149-3:2004. Protective clothing – Electrostatic
properties – Part 3: Test methods for measurement of charge decay. ISBN 0 580 43736 1
EOS/ESD Association Inc. (1999) ESD TR13.0-01-99. ESD Association Technical Report EOS
Safe Soldering Irons requirements. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2013) ANSI/ESD S1.1-2013. Standard for protection of Electrostatic
Discharge Susceptible Items - Wrist Straps. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2014a) ANSI/ESD STM9.1-2014. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Footwear - Resistive Characterization.
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2014b) ANSI/ESD S20.20-2014. ESD Association Standard for the
Development of an Electrostatic Discharge Control Program for – Protection of Electrical and
Electronic Parts, Assemblies and Equipment (excluding Electrically Initiated Explosive Devices).
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015) ANSI/ESD STM97.1-2015. ESD Association Standard Test
Method for the Protection of Electrostatic Discharge Susceptible Items – Floor Materials and
Footwear – Resistance Measurement in Combination with a Person. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2016) ANSI/ESD STM97.2-2016. Floor Materials and
Footwear – Voltage Measurement in Combination with a Person. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2017) ANSI/ESD S8.1-2017 Draft Standard for the Protection of
Electrostatic Discharge Susceptible Items – Symbols – ESD Awareness. Rome, NY, EOS/ESD
Association Inc.
European Committee for Standardisation (CEN) (2014) EN16350-2014 Protective gloves -
Electrostatic properties. Brussels, CEN.
European Committee for Standardisation (CEN) (2015) PD CEN/TR 16832:2015. Selection, use,
care and maintenance of personal protective equipment for preventing electrostatic risks in
hazardous areas (explosion risks) Brussels, CEN.
European Union (2016) Regulation (EU) 2016/425 of the European Parliament and of the
Council of 9 March 2016 on personal protective equipment and repealing Council Directive
89/686/EEC. Available from: https://eur-lex.europa.eu/legal-content/EN/TXT/?qid=
1484921753526&uri=CELEX:32016R0425 [Accessed 17th Aug. 2019]
Holdstock, P., Dyer, M.J.D., and Chubb, J.N. (2003). Test procedures for predicting surface
voltages on inhabited garments. In: Proc. of the EOS/ESD Symp. EOS-25, 300–305. Rome, NY:
EOS/ESD Association Inc.
International Electrotechnical Commission. (1998) IEC 61340-5-1:1998. Electrostatics - Part 5-1:
Protection of electronic devices from electrostatic phenomena - General requirements. Geneva,
IEC.
International Electrotechnical Commission. (2001) IEC 61340-4-3:2001. Electrostatics - Part 4-3:
Standard test methods for specific applications – Footwear. Geneva, IEC.
236 7 Selection, Use, Care, and Maintenance of Equipment and Materials for ESD Control
Further Reading
EOS/ESD Association Inc. (2016) ESD TR20.20-2016. ESD Association Technical Report -
Handbook for the Development of an Electrostatic Discharge Control Program for the Protection
of Electronic Parts, Assemblies and Equipment. Rome, NY, EOS/ESD Association Inc.
Paasi J, Kalliohaka T, Luoma T, Soininen M, Salmela H, Nurmi S, Coletti G, Guastavino F, Fast
L, Nilsson A, Lemaire P, Laperre J, Vogel C, Haase J, Peltoniemi T, Viheriäkoski T, Reina G,
Smallwood J, and Börjesson A. (2004) Evaluation of existing test methods for ESD garments
VTT Research Report BTUO45-041224 Available from: http://estat.vtt.fi/publications/vtt_
btuo45-041224.pdf [Accessed 11th Oct 2017]
239
Ordinary packaging materials are made from a wide variety of materials (Figure 8.1).
Many of these are made of insulating materials such as plastics that charge up easily giving
electrostatic fields and electrostatic discharge (ESD) risks. Others such as papers and
cardboard may have characteristics that vary widely with type and moisture content. Those
that are not insulative at moderate humidity often become insulative at low humidity
below about 30% rh.
Ordinary (non-ESD control) papers and cardboard have highly variable properties that
depend on moisture content and atmospheric humidity as well as the grade of material.
Figure 8.2 shows the variation of surface resistance of some ordinary papers found in the
author’s office, with atmospheric humidity. Many types of cardboard or papers have resis-
tance that varies by several orders of magnitude and may become insulative under dry air
conditions. Some types may be insulative even under moderate humidity conditions. The
unpredictability of the characteristics of these materials makes them undesirable in an area
where static electricity is to be controlled. For this reason, ordinary papers and cardboard
are usually excluded from areas where electrostatic control is required. The moisture con-
tent of papers and cardboard can also be affected by processes. For example, papers passed
through a photocopier may emerge in a dried and charged condition.
Cardboard can be formulated for ESD protective packaging use and is often used to make
boxes. Papers formulated for use in ESD protected areas (EPAs) are also available.
ESD protective packaging is required for two ESD control purposes. First, there is a need
for materials that have known and defined characteristics and can be relied on to provide
little in the way of ESD risk when used in an EPA. Second, materials and packaging are
required that can protect ESD–sensitive (ESDS) devices from ESD risks when outside the
EPA during storage and transport.
ESDS devices, printed circuit boards (PCBs), and assemblies are protected within an
EPA by reducing the ESD threats and sources to a level where ESD damage is insignificant.
When the ESDS devices are to be taken outside the EPA, protective packaging is used to
provide protection from these ESD threats and sources. The ESD protection function may
be combined with other functions such as organizing the ESDS devices into batches or
quantities, types or other groupings, and protection against physical damage or humidity.
The ESD Control Program Handbook, First Edition. Jeremy M Smallwood.
© 2020 John Wiley & Sons Ltd. Published 2020 by John Wiley & Sons Ltd.
240 8 ESD Control Packaging
1.E+13
Office
1.E+12
News
Surface resistance (Ω)
Computer
1.E+11
Viking outer
Viking inner
1.E+10
Mylar
Jiffy bag
1.E+09
Cardboard
1.E+08
1.E+07
20 30 40 50 60 70 80
Relative humidity % rh
ESD protective packaging is now available in an extraordinary variety of types and forms
(Figure 8.3) to suit different types of ESDS devices.
ESD packaging is so important that the International Electrotechnical Commission
(IEC) and ESD Association (ESDA) standards systems have specific standards dealing
with ESD packaging. In the IEC system, ESD packaging requirements are given by IEC
61340-5-3 (International Electrotechnical Commission 2015) and American National
8.2 Packaging Functions 241
Figure 8.3 Some examples of the extraordinary variety of packaging types in use having ESD
control functions.
Standards (ANSI)/ESD S541 in the ANSI/ESDA system. Nevertheless, the topic of ESD
packaging remains one of the most poorly understood areas of ESD control.
● Physical protection
● Holding product ready for use in processes such as automated handling and assembly
● Bearing identification markings or other information
When selecting packaging, all the properties required at all stages of use must be considered.
In ESD control packaging, various terms have been defined by standards or are in common
usage. Unfortunately, they are not always consistent even where standard definitions exist.
used in exposed surfaces of ESD packaging as it may be prone to charge generation and
retention. The exact resistance levels used for these classifications may vary with different
ESD standards. The definitions used in IEC 61340-5-1 (International Electrotechnical Com-
mission 2016b) and ESD S20.20 standards systems (EOS/ESD Association Inc. (2014, 2016))
at the time of writing are given in Section 8.8.
A material may be classified as “electrostatic (or electric) field shielding” if it attenu-
ates electrostatic fields. A material that acts as a barrier to ESD may be classified as “ESD
shielding.” Different ESD packaging standards may vary in the detailed wording of these
definitions and terminology.
The term antistatic has unfortunately been widely used to mean many different things.
Many people refer to ESD packaging as “antistatic” packaging. In some ESD packaging
standards, the term is not used at all because of the risk of confusion. In others, it may be
used with specific meanings.
The property of low electrostatic charge generation is desirable in ESD control packaging
materials. Materials that have lower charge generation properties than ordinary materials
may be classified as “low-charging” materials. They may also be called antistatic materials
by some.
These terms may have specific definitions in ESD packaging standards (Sec. 8.8 and
Tables 8.). The resistance properties of the materials are defined in terms of the surface
resistance Rs or volume resistance Rv measured using defined test methods. Electrostatic
(electric) field shielding is also defined in these standards in terms of the material resis-
tance. In contrast, ESD shielding is defined in two different ways. For bags, it is defined
as the energy measured appearing within the package in response to a standard ESD
waveform applied to the outside of the package in a standard test method. For packaging
other than bags, ESD shielding is defined as a combination of criteria.
materials that contact it and the circumstances. Materials can be classified or ranked accord-
ing to their relative charging polarity in the triboelectric series (see Section 2.2). Triboelec-
trification is, however, a highly variable effect and notorious for its poor reproducibility.
Materials may be ranked differently by different experimenters in their triboelectric series
(Cross 1987).
Some materials have been designed to be “low charging” or “antistatic” in that their
propensity to generate electrostatic charge is less than other materials. The definitions used
in IEC 61340-5-1 and ESD S20.20 standards systems at the time of writing are given in
Section 8.8. It has been found difficult to standardize tests of this property due to the irre-
producibility of results obtained. No agreement has been possible on pass criteria for a
low-charging property, and it remains a comparative parameter.
Triboelectric charging is often confused with the ability of a material to hold an electro-
static charge and surface or volume resistance properties. There is no correlation between
resistance and triboelectric charge generation properties. In tribocharging, two materials
are involved in contact and charging. If one of these is an insulator, it will become charged
even if the other contacting material is not an insulator. It is quite possible for triboelectric
charging to be stronger when a material contacts a static dissipative or conductive material
than it is when it contacts an insulating material.
electrically isolated from each other. On the other hand, it may be desirable to have a high
resistance or even insulating barrier layer that prevents significant conduction of direct ESD
current between the inner and outer surfaces (see Sec. 8.4.5).
(a) 6.0
Applied field (kV/m)
5.0
4.0
Electrostatic field (kV/m)
(b) 6.0
Applied field (kV/m)
5.0
4.0
Electrostatic field (kV/m)
3.0
2.0
1.0
0.0
–0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
–1.0 Time (seconds)
Figure 8.5 (a) Transient electrostatic field arising within an insulating package (polythene bag)
due to external varying field. (b) Transient electrostatic field arising within an intermediate
resistance package (pink polythene at 50% rh) due to external varying field. (c) Transient
electrostatic field arising within a metalized ESD shielding bag at 50% rh due to external varying
field.
8.5 Use of ESD Protective Packaging 247
field shielding effect. Under dry conditions, pink polythene can show little electrostatic field
shielding. Under humid conditions, the same material can show considerable electrostatic
field shielding although transient fields may arise within the package (Figure 8.5b). The
moisture adsorbed by surface antistats allows slow charge migration to reduce the inter-
nal field. When the external field is removed, an internal transient field is produced as the
charges again migrate to a new equilibrium.
In the case of an ESD shielding bag, there is a lower resistance metallized shielding layer
that allows rapid movement of charge in response to the external field. So, the internal field
transient is small (Figure 8.5c).
Until recently, ESD packaging standards, electrostatic field shielding materials have been
classified according by their surface or volume resistance. The definitions and requirements
given in IEC 61340-5-1 and ESD S20.20 standards systems at the time of writing are given in
Section 8.8.
insulating surfaces as well as conductive parts. These parts charge up by contact with the
intimate packaging surface. Charging of the ESDS is dependent on the mutual charging
properties of the ESDS materials and the intimate packaging material and is not related
to the resistance of the intimate packaging material. Charge removal from the conducting
parts of the ESDS device depends on the ESD packaging material resistance, but charge is
not removed from the insulating parts of the ESDS device in this way. In some cases, it can
be most important to minimize electrostatic charging of the ESDS device in this way, and
that is not necessarily done by selecting a non-insulating material.
Where a charged device contacts a low-resistance material, the discharge that occurs can
give charged device ESD risk. This can be minimized by avoiding contact between the ESDS
device and low resistance materials. It is usually desirable for ESD control packaging to have
surface resistance greater than a minimum value (at least 10 kΩ) for this reason.
For very small components, charge retention on the component or on the ESD control
packaging can also result in uncontrolled ejection of the components from the package by
electrostatic attraction or repulsion forces.
EPA is that it should not introduce ESD risk to ESDS handled there. For this reason, pack-
aging used in the EPA must not have exposed insulating materials that could charge up
and cause ESD risks. Exposed packaging surfaces must be low-charging, static dissipative
of conductive. This applies equally to intimate and proximity packaging materials.
Clearly, secondary (non-ESD-protective) packaging usually contains exposed insulating
materials or at least materials that have not been characterized for ESD purposes. These
must be kept well away from ESDS devices due to the possibility they could cause ESD risk.
It is therefore normal practice to keep these materials away from workstations where ESDS
may be handled. Many ESD programs find it preferable to keep these materials out of the
EPA altogether.
Occasionally ESD protective packaging could be required within the EPA to protect ESDS
devices against specific threats that are not well controlled by standard ESD control mea-
sures. It is more usual that additional ESD protective properties will be required where ESD
packaging will be used to protect ESDS devices taken out of the EPA.
Some types of ESDS may place additional requirements on ESD packaging materials that
they may contact or be stored within. For example, for a PCB containing a battery, a mini-
mum surface resistance may be specified for intimate packaging due to a concern that the
battery charge may be drained by contact with the material.
Pink polythene is polythene that has been loaded or surface coated with a chemical
antistat. The pink color is a colorant that was originally added to differentiate static
control materials from ordinary secondary packaging. Nevertheless, a pink color does not
guarantee that a material has low-charging characteristics. Users of these materials should
be careful to ensure they know their characteristics before use.
In some types, a form of amine was used that caused oxidation of some metals and stress
cracking in some plastics. For this reason, some manufacturers switched to amide-based
antistats. An effect on device solderability has also been a concern.
The materials are made by compounding the antistat additive with the polymer base
material, usually low-density polyethylene. Other polymers may be used to change proper-
ties such as stiffness and sealability. A material such as crushed silica or calcium carbonate
is added to counter the tendency for the finished film to stick to itself. The finished mate-
rial may have a greasy feel if the proportions of additive, polymer, and antiblock are out of
balance. The additive may contaminate any surface that it contacts. The final material can
be made into film, sheets, or foams or can be injection molded into trays or other forms.
For effective performance, first the antistat must diffuse to the material surface. Second,
atmospheric moisture must be adsorbed to the surface. The amount of antistat at the sur-
face depends on the material formulation but also can vary and reduce with age of the
material, due to evaporation of the antistat. The amount of antistat present at the surface
depends on the relative rates of evaporation and diffusion to the surface from the bulk mate-
rial. The amount of atmospheric moisture available to be adsorbed to the antistat varies
with atmospheric relative humidity. These factors mean that the material properties can
change considerably with atmospheric humidity and age. At low humidity and extended
age, the material may become ineffective. This means that many pink polythene materials
have short shelf life.
Pink polythene materials have also been implicated in solderability concerns and crack-
ing of polycarbonate materials. The latter is due to the additive dissolving into the polycar-
bonate material. Properties such as label adhesion and printability may be reduced.
Figure 8.6 Typical variation of polymer resistance with conductive particle loading. Source: Blythe
and Bloor (2008).
paths through the material. When the additive loading increases beyond a level, an abrupt
reduction in resistance occurs (Figure 8.6). This level is known as the critical loading or
percolation threshold. The level is influenced by the nature of the resin substrate, size dis-
tribution, and shape of additive particles and the quality of contact between the particles.
Different fillers are added at different loading to produce a given resistance range. The per-
colation threshold occurs around 10–20% loading for spherical conductive particles but at
lower concentrations for fibrous conductive materials.
The material resistance drops quickly with increasing conductive particle loading above
a threshold loading level. At the same time, particle loading variation will inevitably be
present due to variability of mixing. This means that there can be considerable variation of
material resistance over short distances. Processes producing final packaging products can
also affect the local additive loading and material conductivity and result in high resistance
or insulating regions in the material.
The aspect ratio of the particles is the ratio of particle length to diameter. High-aspect
ratio particles improve contact between particles and hence higher conductivity
(lower resistance).
Adding the conductive particulate material influences the physical properties of the
material, such as flexural modulus, tensile strength, hardness, viscosity, and heat distortion
temperatures. Sloughing or particle shedding can be a problem and can prevent use in
clean areas.
These materials are commonly used to make packaging products such as tote boxes, bags,
tubes, bins, and trays by vacuum forming, injection molding, or film extrusion.
8.6 Materials and Processes Used in ESD Protective Packaging 253
Carbon black is the most common additive used to produce conductive or static dis-
sipative materials as they are cost effective. Carbon-loaded materials are normally black
in color. Carbon black in low loading is used in materials not intended for ESD control,
giving black color and other properties. At these low loadings, the materials remain highly
insulating. So, it is important to understand that a black color does not necessarily imply
a material has static dissipative or conductive properties or is suitable for ESD control use!
8.6.9 Embossing
Embossing can be used to change the surface profile of a material. This can in some cases
be used to modify charge generation properties by reducing surface contact area.
8.6.12 Lamination
Several layers of materials with different properties may often be laminated together to form
a package with the desired properties. Examples are ESD shielding bags and moisture bar-
rier bags. Some laminated materials may be vacuum formed, but the forming process may
alter the properties of the laminated layers.
8.7.1 Bags
There are various types of ESD packaging bags on the market that have different com-
binations of physical and ESD control properties. They may be heat sealable or have a
zip or self-adhesive seal. They are available in low-charging, static dissipative, or conduc-
tive or metallized ESD shielding materials. Low-charging bubble wrap bags are also avail-
able. Some materials are transparent, whereas others are opaque. Some examples of bags
designed for ESD control or protection are shown in Figure 8.7. The bag properties specified
in ANSI/ESD S11.4 (EOS/ESD Association Inc. 2012a) and MIL-PRF-81705-D standards are
given in Table 8.1.
8.7 Types and Forms of ESD Protective Packaging 255
Figure 8.7 ESD packaging bags include pink polythene, black polythene, and metalized ESD
shielding bags among other types.
Table 8.1 ESD control bags standard classification and properties specified.
EMI
Standard Type Electrostatic property shielding
Surface Electrostatic
Low resistance Charge field ESD Moisture
charging or resistivity decay shielding shielding barrier
Figure 8.10 Comparison of ESD shielding capability of 2 pink polythene, 3 black polythene, and
11 ESD shielding bag specimens. Source: Smallwood and Robertson (1998).
8.7.3 Foam
Foams are available in insulative, low-charging, dissipative, and conductive materials. They
may be used to cushion items within boxes or as a means of holding leaded package devices
securely (Figure 8.12). A high resistance foam insert within a box or other package system
can help give a barrier to ESD for ESD shielding.
Figure 8.12 Foam is often used as intimate packaging or cushioning in packaging systems (left
and mid) black carbon loaded static dissipative foam (right) pink low charging polythene foam.
specifically to suit the product they are to contain. They may be designed for use with
components or assemblies of finished products. They may be made from injection-molded
materials such as polypropylene or vacuum formed from materials such as high-density
polyethylene.
Shipping and storage boxes may also be made from cardboard, typically faced with con-
ductive or static dissipative coatings and containing a foam insert.
Packaging systems based on boxes can include conductive layers or coatings to give elec-
trostatic field shielding and air gaps, high-resistance materials, or even buried insulating
layers as a barrier to ESD for ESD shielding.
Joint Electron Device Engineering Council (JEDEC) and waffle trays (Figure 8.14) are
stackable trays containing multiple pockets in a matrix, designed to hold component pack-
age size and form made to dimensions defined by JEDEC standards. Trays may also be made
to custom sizes.
260 8 ESD Control Packaging
adhesive. The tape is then wound onto the reel for transport, storage, and dispensing of
components.
Tape and reel packaging is available in paper or polymer insulative materials for passive
component handling, as well as in dissipative or conductive ESD control materials for use
with ESDS. The reel itself can be made of insulative, dissipative, or conductive material.
When used to contain ESDS devices, the tapes and reel should be made from static dissipa-
tive of conductive materials. The intimate surfaces of the tapes should be static dissipative
or conductive and low charging.
The small feature sizes of the tapes make them impossible to test with current standard
test methods that use large electrodes. The tape forming method can affect the resistivity of
the material and can result in isolation of conductive or dissipative regions within the tape
pockets.
Stick magazines or tubes are usually constructed of extruded rigid, clear, or translucent
PVC. Metal sticks have also been used, and carbon-loaded static dissipative or conductive
plastic tubes are available. The extrusion shape is tailored to component package types or
shapes. The extruded magazines are surface treated, usually by dipping the tubes in a liquid
solution. The pins and plugs are also treated.
Definition
Classification IEC61340-5-3:2015 ESD S541-2019
Low charging Not defined User-defined level that will ensure that ESDS items will
not be charged excessively (produces unacceptable risk
of discharge)
8.8 Packaging Standards 263
Table 8.3 Classification of packaging materials in IEC 61340-5-3:2015 and ANSI/ESD S541-2019
based on their surface and volume resistance range.
Definition
Classification or term IEC61340-5-3:2015 ESD 541-2019
Dissipative Rs ≥ 104 Ω and < 1011 Ω Provides an electrical path for charge to
Rv ≥ 104 Ω and < 1011 Ω dissipate from the package
Rs ≥ 104 Ω and < 1011 Ω
Rv ≥ 104 Ω and < 1011 Ω
Surface conductive Rs < 104 Ω Provides an electrical path for charge to
dissipate from the package
Rs < 104 Ω
Volume conductive Rv < 104 Ω Provides an electrical path for charge to
dissipate from the package
Rv < 104 Ω
11
Insulative Rs ≥ 10 Ω Rs ≥ 1011 Ω
11
Rv ≥ 10 Ω Rv ≥ 1011 Ω
Table 8.4 Classification of packaging materials as electrostatic field or ESD shielding in IEC
61340-5-3:2015 and ANSI/ESD S541-2019.
Definition
Classification or term IEC 61340-5-3:2015 ESD 541-2019
These ESD packaging standards also give examples of markings that are recommended
for use to identify packaging used in ESD control and protection. These are used to identify
the packaging material or warn the user that ESDS devices may be contained within the
package. This is discussed further in Section 8.10.
8.8.2.2 MIL-PRF-81705
MIL-PRF-81705D 2004 gave specifications for heat-sealable electrostatic protective flexible
barrier materials used for military packaging of ESDS for use by the US Department of
Figure 8.17 Example of moisture sensitive device packaging label specified in MIL-PRF-18705D
and ESD S11.4.
8.8 Packaging Standards 265
Table 8.5 Applications of static control moisture barrier material classifications defined by
MIL-PRF-81705D.
I Water vapor proof, electrostatic protective, Water vapor proof, electrostatic, and
electrostatic, and electromagnetic shielding electromagnetic protection of microcircuits,
and semiconductor devices, such as diodes,
field effect transistors, and sensitive resistors
II Transparent, waterproof, electrostatic Use where transparency and static
protective, static dissipative dissipation is required and contact with oil or
grease is not contemplated
III Transparent, waterproof, electrostatic Use where a transparent, waterproof,
protective, electrostatic shielding electrostatic-protective, and electrostatic
field protective barrier is required
Defense. It classified materials as Type I, Type II, or Type III (Tables 8.5 and 8.6). Each
type has two subclasses: Class 1 for unlimited use, and Class 2 for use on automated
bag making machines only. The standard specifies physical characteristics such as seam
strength, water vapor transmission rate, thickness, and transparency as well as electrostatic
properties.
266 8 ESD Control Packaging
Property
Low Electrostatic ESD
Level Application Structure charging field shielding shielding Surface resistance (𝛀) Moisture barrier
1 Device packaging: Multilayer Yes Yes Yes Yes. Inner surface may be different Yes
Items subject to metalized plastic <20 nJ from outer surface Inner: ≤0.002 g/100 in.2 /d
reflow soldering 104 ≤ Rs < 1011
Outer:
Rs < 1011
2 General packaging: Multilayer plastic Yes Yes Yes Yes. Inner surface may be different Yes
Items not subject to <20 nJ from outer surface Inner: ≤0.02 g 100 in.2 /d
reflow soldering 104 ≤ Rs < 1011
Outer:
Rs < 1011
3 “Static shielding” to Multilayer plastic Yes Yes Yes Yes. Inner surface may be different No
protect ESDS <20 nJ from outer surface Inner:
104 ≤ Rs < 1011
Outer:
Rs < 1011
4 Conductive to Conductive Yes Yes No Yes. No
protect ESDS extruded plastic Rs < 1011
5 Static dissipative to Extruded plastic Yes No No Yes. Inner and outer surface No
protect ESDS with antistat typically similar Inner:
104 ≤ Rs < 1011
Outer:
104 ≤ Rs < 1011
Level 5 bags are static dissipative bags for protection of ESDS items. They have low-charging
properties intended to avoid charge accumulation on or in the bag that could be damag-
ing to ESDS items. Their inner and outer surface resistance is typically similar and is
intended to allow charge to dissipate when the surface is grounded. They are typically
made from extruded plastic containing or coated with an antistat.
Surface resistivity is defined as a material surface parameter that is corrected for the mea-
surement electrode form (see Section 1.7.2). In contrast, surface resistance is the result
of a measurement made on the surface, with no correction for electrode form. The elec-
trodes specified in IEC 61340-2-3 (International Electrotechnical Commission 2016a) and
ANSI/ESD S11.11 give a factor of 10 reduction in the resistance measured, compared to the
surface resistivity of the material. So, a material with surface resistivity of 1012 Ω according
to MIL-PRF-18705 would also be at the surface resistance limit of 1011 Ω as specified and
measured according to IEC 61340-5-3 or ANSI/ESD S541.
8.9.1 Introduction
In selecting suitable packaging for an ESDS, many factors may need to be considered. These
include the type of ESDS and likely ESD threats, the environment through which it will
pass, physical and environmental protection requirements, and customer requirements.
The overall cost of the packaging is likely to be a significant consideration.
to do damage. ESD can be prevented by ensuring that contact between the ESDS device
and other conductors cannot happen, by ensuring that significant voltage differences do
not exist between the ESDS device and conductors that it may make contact, or that when
ESD occurs, its strength (peak current) is limited and discharge energy is absorbed by high
resistance of the contacting material.
There are two types of ESD threats that should be considered. The main one is protection
of the ESDS devices against direct ESD that may occur from an external ESD source or from
a charged ESDS on contact with an external conductor.
Electrostatic fields are a second threat. While most ESDS devices do not suffer damage
directly from electrostatic fields, they can set up the conditions under which ESD can occur
by inducing high voltages on conductors within the ESDS device (see Section 2.4.4).
Where the ESD threats to the ESDS are not well understood, then it is best to protect the
ESDS device against direct ESD to or from the ESDS device. Protection against electrostatic
fields should also be considered. It may often be most cost effective to take this approach
rather than analyze and evaluate the specific ESD threats to the ESDS device.
If the specific ESD threats to the ESDS device are well defined, it may be possible to choose
ESD packaging that directly prevents those threats. An example is an ESDS electronic mod-
ule built into a metal shielding housing, in which the only threat is the possibility of ESD
to connector pins. In this case, it may be sufficient to protect the connectors against direct
ESD, for example by use of static dissipative or conductive connector cover caps. In some
cases, even insulative caps may be acceptable if these do not present ESD risk within EPAs.
There is a risk of ESD wherever a conductive part of an ESDS device can contact another
conductor (person, equipment, or object) in the environment. The ESD threat, of course,
depends on the susceptibility of the exposed ESDS part to the ESD source with which it
makes contact. Possible sources of ESD include
● ESD from charged personnel
● ESD from charged packaging, objects, or equipment
● The possibility that the ESDS may become charged and discharge to an item that they
touch
The number of ways in which ESD can enter an ESDS device can vary tremendously with
the type and form of ESDS device. A PCB or multipin component can have many ESD entry
points that could touch an external item or person resulting in ESD. Some types of modules
may have few contact points limited to exposed flying leads or connector pins. In some
cases, connector pins may be recessed in a connector housing to the extent that contact
with them is highly unlikely. In other cases, a connector may extend in a way that it is
extremely likely that it will be the first point of contact with another item.
The ESD susceptibility of an ESDS device, and ways in which ESD can enter the ESDS
device, is likely to change with the build state. Ultimately, many ESDS products are pro-
tected against ESD by being built into a housing or potted, preventing ESD from occurring
in all but a small number of well-defined ways. The final product may be considered insen-
sitive to ESD. In some markets, e.g. Europe, the product may need to pass ESD susceptibility
tests before it can be placed on the market. Some customers may have similar requirements.
A change in build state can also introduce a new ESD risk. Potting or building a PCB into
a plastic housing can prevent ESD into all parts except connector pins. At the same time, it
8.9 How to Select an Appropriate Packaging System 271
introduces the risk that the housing can become highly charged and induce high voltages
on the internal PCB, giving a risk of damaging “charged module” ESD on connection to a
wiring loom or other external item.
In many cases, the packaging will need to protect the ESDS device for transport and stor-
age outside the EPA. In this case, it must protect against the anticipated ESD threats as well
as provide no ESD risks to ESDS devices while within the EPA.
Within an EPA, the package may be handled only by trained personnel, which may reduce
the requirement, for example for physical protection. Outside the EPA the package may
be handled by untrained personnel and must withstand the physical threats likely to be
encountered when handled in the same way as any other package.
Within an EPA, it is often necessary to use ESD packaging for non-ESDS items. Secondary
packaging must be kept well away from ESDS devices. Many ESD programs keep secondary
packaging out of the EPA completely. Any suitable ESD packaging can be used to package
non-ESDS within, and for entry into an EPA. For example, a conductive or static dissipative
tote box can equally be used to contain non-ESDS items. Conductive or static dissipative
bags can be used for non-ESDS items instead of ordinary polythene bags. Low-charging
(pink polythene) can also be used for non-ESDS device inside or outside the EPA, providing
the atmospheric humidity within the EPA is not expected to go below about 30% rh. Pink
polythene should also be checked regularly to make sure it has not lost its properties due to
loss of antistat.
If the package is required to enter a clean area, it may be necessary to select materials that
will not give a risk of contamination. This may prevent use of some common ESD control
materials such as carbon-loaded materials or pink polythene. Intrinsically dissipative or
conductive polymer materials may be more acceptable in this case.
The physical and environmental threats are likely to be more severe where packages go
out of the factory for transport between sites or to the customer.
Use of third-party courier or postage services may require a greater level of protection
than use of company transport services. Packages may be subject to transport by air, ship,
rail, or road during a journey.
Where the package crosses national borders, there may be a need to inspect the contents
by personnel who are untrained in ESD control procedures.
The temperature and humidity likely to be encountered during transport and storage may
be subject to considerable variation. The packaging may need to protect against wide tem-
perature and humidity extremes during vehicle or air transport. Even within an EPA, the
humidity may vary dramatically with season, weather, and the presence of heat sources
such as ovens.
Packaging used within an EPA should always be selected to maintain its properties to
the lowest humidity level that may be encountered in practice. Modern packaging stan-
dards usually require materials to be characterized at low humidity (e.g. 12% rh). Some
materials (e.g. pink polythene) rely on antistats and atmospheric moisture for their prop-
erties. These materials be used with caution where the humidity is likely to drop below
about 30% rh. Below this humidity, their properties may become little better than ordinary
secondary packaging materials.
If the package is to take the items to a customer, then customer requirements may be
a strong consideration. Some customers may only accept items packaged as specified in
contracts or standards.
The way in which the package is used, stored, or handled at the customer’s site may give
specific requirements for the package and package marking. For example, a retail customer
8.9 How to Select an Appropriate Packaging System 273
may require packaging that looks attractive for retail display of the product. The package
may need to carry information on the contents to help the consumer select a product for
purchase.
In some cases, it may be sufficient to protect the package contents from ingress of rain,
for example during postage and transport. In other cases, the need for moisture protec-
tion may include maintaining low-humidity conditions within the package to avoid process
problems at a later stage. High humidity can cause problems to electronic parts including
corrosion and difficulties in soldering. It may be necessary to enclose parts susceptible to
such problems within moisture barrier packaging.
The expected temperature range during transit and storage could be an important con-
sideration. In a sealed package, the relative humidity level doubles with a 10 ∘ C fall in
temperature. If the temperature falls below the dew point (see Section 1.11) condensation
may occur on surfaces within the package. A desiccant may be used to reduce the air humid-
ity to a minimum, or the air can be evacuated during sealing to prevent this. The package
may need to have a moisture barrier to prevent vapor ingress.
Removing the water in the internal space of the package with a desiccant can result in
the “worst-case” low-humidity conditions where electrostatic charging may be a concern.
Even without the presence of desiccant, low-humidity conditions can arise due to temper-
ature changes. When ambient temperature increases, the internal relative humidity drops,
halving with a 10 ∘ C rise in temperature.
8.9.7 Selecting the ESD Packaging Type and ESD Protective Functions
8.9.7.1 Intimate Packaging
Intimate packaging is the material in contact with the ESDS device within the package.
It must not charge up appreciably to avoid electrostatic risk to the ESDS device and other
items in an EPA.
The material must not cause significant charging of the ESDS device that is in contact
with it. At first sight, a packaging material that does not retain charge might be expected to
also minimize charging of the ESDS device. However, charge accumulation on any material
is the product of two characteristics – charge generation and the dissipation of charge from
the material. Typically, the ESDS device has some insulating surfaces as well as conducting
surfaces. Any charge generated by contact of these insulating surfaces with the packaging
can be retained on the ESDS device. In contrast, the packaging material typically should be
a noninsulative material and dissipate the charge produced on this material. So, the charge
retained on the packaging is not correlated to the charge retained on the ESDS device. For
low charging of the ESDS device, a material that will give low charge generation character-
istics with the ESDS device should be selected.
It is impossible to avoid some charge generation by contact of the ESDS device with
the intimate packaging material. For dissipation of charge generated by contact with the
ESDS device, the intimate packaging material should be static dissipative or conductive
(see Section 8.3.1 and Table 8.3). Any making or breaking of contact with the ESDS device
will result in ESD at some level. If a high-resistance intimate packaging material is used,
the ESD that occurs has its peak current limited by the high resistance of the material. The
274 8 ESD Control Packaging
energy in the discharge is mainly absorbed in the intimate packaging material rather than
the ESDS device.
It may also be necessary to set a lower limit to the surface resistance of the intimate pack-
aging. For example, a PCB with on-board battery may require an intimate packaging of
sufficiently high resistance to avoid discharge of the battery by contact with the packaging.
The physical design of the intimate packaging can also give benefits. For example, a pur-
pose made rigid material can be shaped to hold the ESDS device firmly while introducing an
air gap to minimize contact with the surfaces of the ESDS device (and hence charging) and
at the same time act as a barrier to direct ESD. As another example, foam inserts within
a box can be used to hold ESDS devices of a variety of size or form firmly with minimal
movement and can virtually eliminate making and breaking of contact with ESDS leads.
Figure 8.18 Example of packaging marking to identify ESD control packaging or materials
according to IEC 61340-5-3, ANSI/ESD S541 and ANSI/ESD S11.4. Source: Reproduced by
permission of the EOS/ESD Association Inc.
The ESD packaging standards IEC 61340-5-3 and ANSI/ESD S541 give examples of ESD
packaging marking that can be used with ESD control and protective packaging. The outer
surface of an ESD protective package should be marked with a symbol and/or wording
identifying it as a package that is expected to contain ESDS devices (Figure 8.18). This is
so that personnel handling the package can easily recognize that the package may contain
ESDS devices and should not be opened further outside an EPA. Where a packaging system
has several materials in layers, each material layer should be identified as an ESD control
material. The “hand” symbol should also normally have a letter code beneath, indicating
the main packaging function codes as follows: S for electrostatic discharge shielding,
F for electrostatic field shielding, C for electrostatic conductive, and D for electrostatic
dissipative.
Unfortunately, in practice packaging manufacturers are rather variable in how they
mark ESD control packaging materials and products. Markings from older and superseded
standards are often still seen. The letter codes showing the packaging function are often
missing.
References
ASTM International. (2013) ASTM F1249 13 Standard Test Method for Water Vapor
Transmission Rate Through Plastic Film and Sheeting Using a Modulated Infrared Sensor.
West Conshohocken, PA, ASTM.
ASTM International. (2014) ASTM D257-14 Standard Test Methods for DC Resistance or
Conductance of Insulating Materials. West Conshohocken, PA, ASTM.
Bellmore, D. (2001). Anodized aluminium alloys – insulators or not? In: Proc EOS/ESD Symp.
EOS-23, 141–148. Rome, NY: EOS/ESD Association Inc.
Blythe, T. and Bloor, D. (2008). Electrical Properties of Polymers, 2e. Cambridge University
Press. ISBN: 978-0-521-55838-9.
Cross, J.A. (1987). Electrostatics Principles, Problems and Applications. Adam Hilger. ISBN:
0-85274-589-3.
Department of Defense (2002) MIL-STD-3010). Test Method Standard. Testing Procedures for
Packaging Materials. Test Method 4046 Electrostatic Properties, 33–41. Washington, D.C.:
DoD.
Department of Defense (2004) MIL-PRF-81705D:2004). Military Specification. Barrier
Materials, Flexible, Electrostatic Protective, Heat Sealable. Washington, D.C.: DoD.
References 277
Smallwood, J.M. and Millar, S. (2010) Paper 3B4). Comparison of methods of evaluation of
charge dissipation from AHE soak boats. In: Proc. EOS/ESD Symp, 233–238. Rome, NY:
EOS/ESD Association Inc.
Smallwood J M, Robertson C J. (1998) Evaluation of Shielding Packaging for Prevention of
Electrostatic Damage to Sensitive Electronic Components. ERA Report 97-1079R, ERA
Technology Ltd., Cleeve Rd, Leatherhead, Surrey, KT22 7SA
Further Reading
EOS/ESD Association Inc. (1995). ESD ADV11.2-1995. Advisory for the Protection of
Electrostatic Discharge Susceptible Items - Triboelectric Charge Accumulation Testing. Rome,
NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2017) ANSI/ESD S8.1-2017. Draft Standard for the Protection of
Electrostatic Discharge Susceptible Items – Symbols – ESD Awareness Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2017). ESD ADV1.0-2017. ESD Association Advisory for Electrostatic
Discharge Terminology – Glossary. Rome, NY, EOS/ESD Association Inc.
Fowler S. (2000) ESD protective packaging. Available from: http://www.esdjournal.com/
techpapr/twenty1/intro.doc [Accessed 10th Nov. 2017]
Gale S F. (2006) Zero tolerance for ESD. Solid State Technology http://electroiq.com/blog/
2006/09/zero-tolerance-for-esd [Accessed 29th Nov. 2017]
Huntsman J. R., Yenni D. M., Mueller G. E. (1980) Fundamental requirements for static
protective containers. Nepcon West VI pp. 624–635
Huntsman, J.R. and Yenni, D.M. (1982). Test methods for static control products. In: Proc. of
EOS/ESD Symp. EOS-4, 94–109. Rome, NY: EOS/ESD Association Inc.
Huntsman, J.R. (1984). Triboelectric charge: its ESD ability and a measurement method for its
propensity on packaging materials. In: Proc. EOS/ESD Symp. EOS-6, 64–77. Rome, NY:
EOS/ESD Association Inc.
International Electrotechnical Commission. (2018) IEC TR 61340-5-2:2018. Electrostatics – Part
5-2: Protection of electronic devices from electrostatic phenomena - User guide. Geneva, IEC.
Koyler, J.M. and Anderson, W.E. (1981). Selection of packaging materials for electrostatic
discharge (ESDS) items. In: Proc. EOS/ESD Symp. EOS-3, 75–84. Rome, NY: EOS/ESD
Association Inc.
Matisoff, B. (1997). Handbook of Electronics Manufacturing Engineering, 3e. Springer.
Swenson, D.E. and Lieske, N.P. (1987). Triboelectric charge-discharge damage susceptibility of
large scale IC’s. In: Proc. EOS/ESD Symp. EOS-9, 274–279. Rome, NY: EOS/ESD Association
Inc.
Swenson, D.E. and Gibson, R. (1992). Triboelectric testing of packaging materials: practical
considerations – what is important? What does it mean? In: Proc. EOS/ESD Symp. EOS-14,
209–217. Rome, NY: EOS/ESD Association Inc.
Texas Instruments. (2002) Electrostatic Discharge (ESD) Protective Semiconductor Packing
Materials and Configurations. Application Report SZZA027A - April 2002
Further Reading 279
9.1 Introduction
There are always several ways of looking at the “fitness for purpose” of an electrostatic
discharge (ESD) program. As this phrase implies, it is necessary first to understand the
various objectives that the ESD control program may be addressing. The primary purpose
is presumably to control ESD risks and reduce ESD damage to an acceptable level.
A second purpose is often to help satisfy customers that the organization takes sufficient
care in the manufacture or handling of their product commensurate with the market and
reliability requirements. In this way, the ESD control program may form a positive contri-
bution to the marketing of the product handled. Customers may audit the facility from time
to time as part of their supplier quality assurance procedures. Some customers may require
that their suppliers comply with an ESD control standard, and some of these may insist on
compliance with their own preferred ESD control standard.
Another important aspect of evaluation is of the cost effectiveness of the ESD control pro-
gram. It is usually desirable to attempt to get the maximum benefit from the least investment
in resources (time and money). Evaluating the likely benefit from the ESD control program
may be difficult but can be helpful in selecting the objectives of the program as well as
deciding the level of resources to invest in it.
Identification and evaluation of these is discussed further in Chapter 9. The usual sources
of ESD risk include
● Charged personnel
● Charged metal or conductive objects or materials
● Charged devices
Other ESD risks may also occur, such as
● Charged board ESD
● Charged cable ESD
● Charged modules or assemblies
Understanding the ESD risks is an important step toward determining effective ESD con-
trol measures.
A final product that is not susceptible to ESD damage often relies on a housing or covers
for protection of the ESDS internal parts. If the housing or covers are removed and the ESDS
parts exposed, it may be necessary to again consider the item to be susceptible to ESD.
An assembly containing ESDS devices should be considered also to be ESDS devices
unless there is good reason or evidence that it is adequately protected against ESD at that
build level. The following possible ESD sources should be considered:
● Direct ESD from charged personnel to the ESDS parts of the product, especially via hand-
held metal tools
● ESD occurring on contact with charged metal parts, chassis, or machinery
● The possibility that the internal parts could reach high voltage by induction or tribocharg-
ing leading to ESD on contact with another conductive item
● ESD occurring due to exposed terminations of a charged product contacting external con-
ductive items
● Connection of charged cables or wiring looms to the assembly
Where only specific ESD threats occur in a process step, it may be sufficient to use spe-
cific ESD control measures rather than general ESD control measures. General ESD control
measures may not adequately protect against specific ESD risks.
The responses to risks are usually specified in accordance with IEC 61340-5-1 and ESD
S20.20 or the guidance of ESD SP10.1. For example, ESD SP10.1 recommends that in
a region 15 cm around the ESDS critical path, all conductive machine parts should be
grounded and insulative parts made static safe. S20.20 and 61340-5-1 require that the
electrostatic field at the position of the ESDS device is <5 kVm−1 , and items having surface
voltage greater than 2 kV must be kept >30 cm from the ESDS device. The responses to
risks may also need special measures not prescribed by these documents. These standards
also require that
● Personnel handling unprotected ESDS devices must be grounded according to the stan-
dards.
● Metal objects contacting the ESDS devices must be grounded. If they cannot be grounded,
the voltage difference between the ESDS devices and metal object must be reduced to
below a hazard threshold level (±35 V in 61340-5-1:2016 and 20.20-2014).
● To reduce charged device ESD risks, low-resistance conductors that contact the ESDS
device may be replaced with intermediate- or high-resistance conductors.
● To address field-induced charged device ESD risk, nonessential insulators that might
become charged are removed from the vicinity of the ESDS devices. Essential insulators
are treated according to ESD risk.
● To address charged device ESD risks caused by contact charging or rubbing of ESDS
devices, process steps that cause tribocharging of the ESDS devices are minimized.
Notice that all the ESD risks listed involve contact between the ESDS devices and a person
or material. For most ESDS devices, where there is no contact, there is no ESD risk. The
contact that causes ESD may, however, occur during a following step rather than the step
at which charging occurs.
Electrostatic fields usually provide an ESD risk only where there is contact with the ESDS
device in the presence of the field. There are, however, a small number of ESDS device types
(e.g. reticles) that may be damaged by electrostatic field alone due to induced voltage differ-
ences causing breakdown of internal low-voltage insulating layers or air gaps. Smallwood
(2019) has shown that damage could be caused to some voltage-sensitive devices such as
metal-oxide-semiconductor field-effect transistors (MOSFETs) with high-impedance termi-
nals, by charge injection from fast-changing electrostatic fields.
When evaluating processes, it’s likely that the ESD specialist will need assistance from
process specialists to ensure that ESDS movements are correctly evaluated, risk points
identified, relevant measurements made, and suitable solutions found. Examples are given
by Gaertner (2007), Halperin et al. (2008), Lin et al. (2014), and ESD TR17.1-01-15. The
TR17.1-01 document draws on several earlier works.
TR17.1-01 following Jacob et al. identifies an evaluation procedure steps as follows:
● Assess potential risk
● Identify measurement points
● Make measurements
● Assess measurement results
● Define and implement corrective actions
9.3 Evaluating Process Capability Based on HBM, MM, and CDM Data 285
The first step involves demonstration of the process under real conditions, with the goal of
identifying potential risks and measurement points. Making measurement and evaluating
the results presents its own set of challenges. Some of these are discussed by Lin et al. (2014).
parameters such as inductance, resistance, and source capacitance. Real-world sources are
unlikely to have the same capacitance, resistance, and inductance as the HBM, MM, and
CDM models. This means that HBM, MM, and CDM discharges do not happen in the real
world – they happen only in the component susceptibility test equipment. Real-world ESDs
are simply discharges from the human body, a metal part, a device, or another source. Each
source has very different ESD waveform characteristics from the HBM, MM, and CDM mod-
els and would yield a different withstand voltage (TR17.0-01, Tamminen and Viheriäkoski
2007; Gaertner and Stadler 2012). So, ESDS devices might be more or less susceptible to
damage from real sources than it is to HBM, MM, or CDM. A simple action such as changing
the position of an ESDS devices can change a relevant parameter such as its capacitance as
well as the voltage and hence the ESD damage susceptibility. Tamminen and Viheriäkoski
(2007) found that the real charged device ESD risk in a process can be significantly less than
suggested by the CDM withstand voltage. As components approach a flat grounded surface,
the device voltage could in some cases reduce by 95 percent compared to the initial value
away from the surface. This drop varied with component package, being least for tall com-
ponents such as dual in-line (DIL) packages. Similar results were found with a component
approaching a floating PCB. They were able in some cases to get a better estimate of ESD
risk by taking account of the initial voltage and charge on the device, especially where the
geometry of the device and environment were taken into account. They commented that
in placing of a component, the resistive characteristics of solder paste may affect discharge
characteristics and ESD risk.
Even in the well-controlled world of the device ESD susceptibility tester, there can be
significant differences in parameters due to parasitic components in the tester. These can
lead to variation in withstand voltages measured for the same device with different testers.
Changing a device package type can also lead to changes in measured withstand voltage.
Some ESDS devices such as PCBs or assemblies are not normally tested for susceptibil-
ity using standard HBM, MM, or CDM ESD. Even if they were, the susceptibility would be
likely to change for every possible contact or ESD entry point on the ESDS devices. A com-
plex ESDS device such as a PCB thus could have a large number of possible contact points.
Susceptibility to ESD could vary with each new component added to an assembly and with
each ESDS position in an assembly process.
Some workers have suggested that it would make more sense to specify parameters such
as peak current (Bellmore 2004; Smallwood and Paasi 2003), charge, power, or energy
instead of the source voltage as withstand data parameters. This added complexity would
at least allow use of relevant parameters in evaluation of ESD risk in real-world situations.
Even this would not necessarily give the complete answer. It can be difficult to measure real
ESD parameters in a process, especially in automated equipment. Inserting measurement
equipment and sensors usually modifies the parameters that govern the ESD waveforms
that are being measured. In fast-moving automated equipment, it can be difficult to add
measurement equipment safely and without modification of the process.
A further complication is that it is not always easy to obtain ESD withstand data for all
components. In practice, this may not matter too much as it is likely that the process eval-
uation will focus on the most sensitive components. As long as these are identified and the
ESD control designed for handling them, then the less-sensitive components are probably
adequately protected.
9.3 Evaluating Process Capability Based on HBM, MM, and CDM Data 287
This analysis depends on the ESDS device being susceptible to the energy in the discharge.
If it is susceptible to the voltage or charged transferred in the discharge, then a different
analysis would need to be applied (Paasi et al. 2003). For simplicity, it is tempting to specify
that the maximum voltage between any external metal ESD source and the ESDS device
should not exceed the MM ESD withstand voltage. This would, however, greatly overesti-
mate the ESD risk, and it could be difficult to maintain these low voltages. In practice, the
MM ESD susceptibility test is being discontinued by IC manufacturers.
As with charged device ESD, if the contacting material has significant resistance, this
absorbs some of the discharge energy and to some extent protects an energy-susceptible
ESDS device. If the ESDS device is voltage or charge susceptible, resistance in the contacting
material may not have any protective effect.
Identify conductor
No
No
No
Evaluate and
ameliorate risk
Document
evaluation and
ESD control
measures
ESDS device and the conductor must be less than ±35 V. Measuring this is easier said than
done. It requires two measurements.
● The voltage range and polarity on the ESDS devices must be measured under working
conditions.
● The voltage range and polarity on the conductor must be measured under working con-
ditions.
● The range in difference between the two voltages must be calculated.
Measurement of voltage on the ESDS device and conductor under working conditions
may be far from easy in practice, especially under working conditions in a fast-moving
automated process. If the ESDS device or the conductor is small, contact or noncontact
electrostatic voltmeters capable of measurement on small objects must be used (Steinman
2010, 2012).
One way to bring the voltage difference between a conductor and an ESDS device to a
low level is to place them in the ion stream from an effective ionizer. Given sufficient time
and low charge generation processes, each will come to the same voltage - the offset voltage
of the ionizer. This may not work well, however, in a fast-moving process where sufficient
time may not be available or if charge generation processes are greater than the rate of
neutralization that the ionizer can achieve in the installation.
290 9 How to Evaluate an ESD Control Program
If the ESD concern is restricted to charged device ESD (i.e. an otherwise unconnected
ESDS device makes contact with the conductor) and if the conductor has surface resistance
>10 kΩ, the ESD risk is likely to be negligible (see Section 9.3.4).
Device
Capacitance
between
device and
ESD
surface
Minimal
Metal surface surface
resistance
Resistance in
ground path
(a)
Device
Capacitance
between device
ESD and surface
Surface
Resistive resistance
surface
(b)
Figure 9.2 Charged device ESD to metal (above) or resistive (below) surface. Adding resistance to
a ground path from a contacting conductor does not limit the charged device ESD current as using
a resistive surface does.
● For a charged insulator, it may be possible to replace the insulator with a grounded con-
ductor.
● Increasing the separation distance between the voltage source and the device path.
● Adding an electrostatic field shield between the voltage source and the device.
The parts of an ESDS device most at risk from this are likely to be high-impedance lines
that have low capacitance and low-voltage breakdown. This result in sensitivity to damage
from electrostatic fields even where no contact is made with the ESDS device.
Paasi et al. (2006) showed that for a MOSFET, the ESD risk threshold was adequately
described by a charge threshold, whereas the HBM voltage value was inadequate as an indi-
cator of ESD risk. The damage risk occurred when the induced charge could elevate the volt-
age to the gate breakdown value. When the MOSFET was mounted on a PCB, the gate break-
down voltage could replace the charge threshold as a risk indicator. The real ESD risk is of
course modified by the circumstances and the probability of making contact to the MOSFET
in a way that could cause ESD damage. ESD risk did not arise directly from the electrostatic
field strength but rather from charge induced on the MOSFET or PCB. The field threshold
9.3 Evaluating Process Capability Based on HBM, MM, and CDM Data 293
for damage to a device on a PCB was lower than for an unmounted component because the
PCB had greater area exposed to the field and greater induced charge for the same field.
Measure electrostatic
charging under worst-
case conditions
Is charging
negligible under Yes
worst-case
conditions?
No
No
any spark and the impedance of the contacting material at the point of contact as well as
voltage difference between the ESDS device and conductor. The impedance of this circuit
typically changes with material characteristics and ESDS device position and orientation.
Standards may also give requirements that can be used to evaluate fields from charged
insulators. For example, the IEC 61340-5-1:2016 standard gives requirements that the elec-
trostatic field at the position of the ESDS device must be <5 kVm−1 . In IEC 61340-5-1:2016
and ESD S20.20-2014, insulators charged to >125 V must be kept at least 2.5 cm from the
ESDS device, and if charged to >2 kV must be kept >30 cm from the ESDS device. If these
conditions are fulfilled, the electrostatic fields and voltages can be considered negligible for
the purposes of these standards. Nevertheless, these limits may not be sufficiently low or
well specified for safe handling of some types of ESDS devices (Stadler et al. 2018). This is
a subject of research and discussion among experts at the time of writing.
2012). He used a 14-pin DIL package positive metal-oxide semiconductor (PMOS) device
(Siliconix SM110CJ with 200 V HBM and 125 V CDM ESD withstand voltage) and a
discrete MOSFET (Motorola 3 N157 with 200 V HBM and 150 V CDM ESD withstand
voltage) as his test devices. He mounted these in an electrostatic field created by charged
plates of various sizes charged by momentary contact with a preset high voltage. Devices
were placed on a glass plate or a grounded metal plate. A grounded metal-wire probe
was touched to a device pin while within the field. The discharge current waveform
was recorded. Swenson then produced charts showing the damage rates experienced by
devices at different applied voltage (electrostatic field) and separation distance from the
ESDS devices.
With the SM110C device, Swenson found that the field level required for damage was
greater when the device was placed on a ground plane compared to when it was on a glass
plate. With the device on the glass plate, it is likely that the field coupled to the device is
intensified. With the ground plane present, some of the field is coupled to the plane, the field
at the device is reduced, and voltage required for damage is increased. Differences were also
observed with the 3N157 device. Swenson noted that damage actually occurred when the
devices were touched by the ground lead, creating a field-induced charged device ESD.
Swenson went on to perform experiments with a uniformly charged insulating plastic
plate. He found that with the SM110C device on a ground plane, direct contact between
the ESDS device and the charged plastic plate did not damage the device until the charged
insulator voltage was 18 kV. When the device pin was grounded in the presence of the field,
damage occurred at 10 kV. (In practice, the surface voltage and surrounding field will be
altered by the presence of conductors such as the ground plane and ESDS device. The insu-
lator voltage was measured using a field meter at 2.5 cm distance. (The measured insulator
surface voltage would not have been the same as at the time of damage occurring.)
In experiments with the 3N157, again direct contact between the ESDS device and the
charged plastic plate did not damage the device until the charged insulator voltage was
18 kV. Touching a device pin with a grounded probe resulted in damage with the insulator
charged to only 2 kV.
Swenson’s experiments showed that an ESDS device, if grounded via a metal contact in
the presence of an electrostatic field, experiences a potentially damaging ESD. The device
becomes charged in the process and, if the field is removed, can subsequently experience
another potentially damaging ESD. The risk to the ESDS device depends on the size of the
electrostatic field source and its distance from the ESDS device, as well as the charged sur-
face voltage. Low voltages close to the device can be of concern where very sensitive devices
are handled. It is the act of grounding, rather than the field, that caused the damaging ESD.
Direct contact with an insulator did not result in damage up to high charge levels. Swen-
son’s results show that proximity or contact between a charged insulator and ESDS device is
not a significant ESD risk in most circumstances, providing the ESDS device does not make
contact with a conductor in the presence of the field. Small charged objects provide lower
risk than large objects – Swenson found damage occurred to both devices when they were
grounded in the presence of plates >4 cm × 4 cm size charged to 500–1000 V. The presence
of a ground plane nearby increased the field required for damage.
Swenson’s results supported the strategy of removing nonessential insulators from the
vicinity of the ESDS device to reduce ESD risks. They also confirm that field strength is an
296 9 How to Evaluate an ESD Control Program
indicator of these risks where contact between the ESDS device and a grounded conductor
occurs in the presence of the field.
It is arguably the CDM ESD withstand voltage of the devices that is of most relevance to
the ESD risk in Swenson’s experiments. The voltages required for damage in the experiment
(>500 V) were considerably greater than the CDM ESD withstand voltage of the devices (125
and 150 V).
9.3.7 Troubleshooting
Where failures have been experienced and ESD damage is suspected, the first action should
be to check that normal ESD control processes are in place and are operating correctly. Rec-
tify any deviations from the ESD control program and evaluate whether they could cause
the damage.
Check that personnel handle the ESDS device using the required procedures and personal
grounding equipment. Human nature sometimes causes personnel to invent more quick
and convenient procedures that may not be so effective for ESD control, especially when
under pressure to work fast.
If possible, obtain ESD withstand data for the damaged components to help identify pos-
sible sensitivity to likely ESD sources. If possible, failure analyze the damaged components
to establish, or rule out, ESD as a possible cause. Semiconductor manufacturers will some-
times failure analyze components on behalf of their customers.
If atmospheric humidity data is available, check whether failure rates correlate with low
atmospheric humidity. If this correlation exists, it can be a strong indicator that ESD is a
likely cause of damage and ESD control is insufficient. Low humidity can occur in areas
where strong heat sources are present, such as near ovens or near the cooling fan exhausts
of equipment or computers. Local low humidity conditions can occur near heat sources
even in air-conditioned rooms where humidity is controlled.
If there is no obvious deviation from the ESD program that could be responsible for the
failures, more detailed evaluation of the process may be required, including following the
path of the ESDS device through the process. Try to narrow down the problem to part of the
process. For example, if failed products pass tests at one stage and then fail at a later stage,
it is likely that the problem has occurred in the process steps between the two test points.
Critically examine the process for stages at which the ESDS device contacts other con-
ductors. In a manual process, this should include handling by personnel. These contact
points are points at which ESD is likely to occur. Processes in which ESDS devices may
contact metal or highly conductive items should be examined carefully for ESD risks. This
is discussed further in Section 9.4.
One area that is often overlooked is where an ESDS device is placed on a test jig that
makes contact prior to test. If the ESDS device becomes charged before contact or contact is
made in the presence of an electrostatic field, ESD can occur. Test jigs often contain essential
insulating parts that could charge and give rise to an electrostatic field during use. I have
also often found jig covers made from high-charging materials – these could induce voltages
on ESDS devices immediately prior to contact with test jig contacts. The possibility that the
test jig contacts and wiring might be charged or energized should also be checked.
Another often overlooked area is connection of cables to an ESDS device. ESD can occur
if the cable is charged, if the ESDS device is charged, or if the connection is made in the
9.4 Evaluating ESD Protection Needs 297
presence of an electrostatic field. Cables can be highly charged as a result of being supplied
packaged within a polythene bag or taken off a reel. The presence of charged packaging can
also induce high voltage on cables, modules, or assemblies nearby.
It is often assumed that a potted module is immune to ESD. A module can have its exterior
highly charged by contact with, or storage within, insulating packaging. The surface charge
on the module can then induce high voltage on internal conductors that can lead to ESD
when connected to cables or other conductors.
One area that can be problematic is handling or testing ESDS devices where high voltages
are present, e.g. test process. Safety considerations often dictate that personal grounding or
other standard ESD control measures cannot be used. Personal protective equipment such
as high-voltage protection gloves or footwear may be demanded.
● Is the conductor grounded? If not, what would be the ESD risk if it became charged?
● Could the ESDS device become charged before touching the conductor? Could the voltage
difference between the conductor and the ESDS device result in ESD risk?
● What is the resistance of the conductor? Is it high enough to prevent ESD risk?
ESDS devices or isolated conductors can attain high voltage by two means. First, they
can become charged by contact with other materials (triboelectrification). Second, they can
attain high voltage under the influence of an electrostatic field.
Insulating materials are often the main source of electrostatic fields when they become
charged by contact with other materials. Nonessential insulators are therefore removed
from the vicinity where unprotected ESDS devices are handled, or even from the EPA. Many
insulators are, however, an essential part of the product or process. These essential insula-
tors cannot be removed from the vicinity of the unprotected ESDS devices. The ESD risk
associated with the presence of the insulator must be evaluated.
If an ESDS devices is exposed to an electrostatic field but does not contact another con-
ductor, there is unlikely to be a risk of ESD damage (Gaertner 2007). Few ESDS devices
are susceptible to damage from electrostatic fields alone. An exception may be where ESDS
devices contain voltage-sensitive devices such as MOSFETs with high-impedance circuit
nodes (Smallwood 2019). So, the presence of an electrostatic field from a charged insula-
tor does not necessarily amount to an ESD risk. An ESDS device can enter a field and exit
without damage. If, however, contact is made with another conductor while within the
298 9 How to Evaluate an ESD Control Program
electrostatic field, ESD will occur. An ESDS device that enters the field uncharged will exit
the field uncharged unless charge has been lost or gained by ESD, by conduction through
contact with other materials or via unbalanced ionization.
Triboelectric charging of ESDS devices can happen whenever the ESDS device contacts
other materials. The amount of charge generated depend on the material that contacts the
ESDS device and many other factors and conditions (e.g. humidity and rubbing action). It is
often thought that triboelectrification occurs with contact only by insulating materials. This
is not correct as any material contacts give triboelectrification, although if a conductor is
grounded, the separated charge escapes from the conductor without any voltage developing.
Insulating parts of an ESDS device such as PCB substrate, coatings, or potting compound
can be charged by contact with conductors, and the charge developed on the insulator is not
removed by contact with the conductor. In some cases, a greater level of charging has been
found for contact with static dissipative materials than with insulators or lower-resistance
conductors (Viheriäkoski et al. 2012). The voltage developed on the charged item depends
on the amount of charge and its capacitance. Capacitance is in turn dependent on the prox-
imity of other materials and conductors.
The risk of damage occurring when an ESDS device at high voltage contacts a conductor
depends on the part of the ESDS device that is subjected to ESD, the magnitude (energy,
current, or other parameter related to possible damage) of the ESD, and the sensitivity of
the ESDS device to the form of ESD that occurs. Unfortunately, it is difficult to predict the
susceptibility of ESDS devices to a real ESD event even if HBM, MM, or CDM ESD withstand
voltage data is available.
Common ways in which an ESDS devices could attain high voltage include
● Triboelectrification during handling by operators wearing gloves
● Triboelectrification during transport by suction grip
● Triboelectrification during contact with a support wheel or conveyor
● Induced voltages due to proximity to charged insulators, machine parts, computer
screens, clothing, gloves, or another electrostatic field source
● Triboelectrification during removal of a label, masking material, or adhesive item
● Triboelectrification during brushing to remove debris
● Induced voltages due to assembly into a charged insulative housing
● Triboelectrification during rubbing or removal adhesive tape or packaging from a plastic
housing or potted component
Common ways in which conductors often contact ESDS devices include
● Touching a metal tool or machine part
● Placing of a component on a PCB track
● A PCB track coming against an end stop in a machine
● A PCB track contacting “bed of nails” pogo pin or support
● Contact of a cable with a terminal on the ESDS device
If an ESDS device is found to attain high voltage and contact another conductor while in
this state, an ESD risk is clear. Similarly, a risk is clear if an isolated (ungrounded) conductor
is found to attain high voltage and contacts an ESDS device while in this state. To avoid these
risks, a means must be found of reducing voltage difference between the conductor and the
9.4 Evaluating ESD Protection Needs 299
ESDS device before contact. A possible alternative is to replace the conductor with a static
dissipative material to reduce the ESD current occurring on contact.
It is usual, if possible, to ground any conductor that contacts the ESDS device. This elim-
inates the risk that the conductor will attain high voltage due to triboelectric or induction
charging. It does not, however, eliminate the possibility that the ESDS device may attain
high voltage due to triboelectrification or induction. If this occurs, charged device ESD can
occur on contact.
Understanding the process and evaluating the associated risks can be particularly difficult
for automated processes. A working automated process is usually fast and inaccessible. See-
ing the steps can be difficult and measuring ESD-related parameters during live operations
is often impossible. Some common practices used in automated equipment are discussed
in Chapter 5.
Rgmax C = 𝜏
An example might be the resistance through the handle of a handheld tool. If the capac-
itance from bit to hand through the handle is measured to be 20 pF and a decay time of
one second is deemed to be adequate, the maximum acceptable resistance through the han-
dle is 1/10−11 = 1011 Ω (100 GΩ).
In practice, it might be convenient to set a lower resistance that is easier and more con-
venient to measure (see Section 9.4.5).
● Protection of personnel against electric shocks in a fault condition with high voltages
present (see Section 4.7.6)
● Incorporation of inherent resistance in a material that contacts an ESDS device, for reduc-
tion of charged device ESD risk (see Section 4.7.5)
In practice, it might be convenient to set a lower resistance that is easier and more conve-
nient to measure. For example, setting an upper limit of 20 MΩ might make the tool more
easily verifiable using a lower specification resistance meter or even a wrist strap checker.
PCB may be contained within a subassembly isolated from the main assembly. In each of
these cases, the ESDS PCB can achieve high voltage by triboelectrification or induction due
to nearby electrostatic fields. ESD can occur when the ESDS PCB is connected to another
system component. Damage can occur if the connection and ESD occurs to a susceptible
part of the circuit.
It can often be difficult to prevent this type of problem unless the system components are
designed to avoid it. The risk of damage can often be reduced by connecting first to a 0 V,
ground, or power connection before I/O connections are made.
When packaging for transport modules that have flying leads, the ESD protective packag-
ing should be specified to prevent accidental contact between the flying leads and external
items.
The costs of an inadequate ESD control program can include (Smallwood et al. 2014)
● Repair and replacement of ESDS devices product failures
● Dealing with product unreliability or drift in characteristics
● Failure analysis
● Purchase of materials that have incorrect properties for ESD control
● Production delays
● Need to overstock commonly failing components
● Expenditure on unnecessary or ineffective control materials or equipment
● Dealing with customers failures occurring in the field
● Dealing with disputes with customers about their perception of adequacy of ESD control
in the facility
● Effects on product and company reputation, and sales
One way to estimate component losses is by throughput evaluation (Halperin 1986). This
includes the following steps:
1. Identify the ESDS devices and determine the discrepancy between the volume purchased
and the volume used in production.
2. Analyze ESDS usage, including average inventory levels and locations, requisitioning
departments, purchase volume, and unit cost.
3. Define burden costs associate with ESDS devices and assemblies.
4. Evaluate the overall impact of ESD damage.
One approach is to identify the finished products produced during a particular period that
contain ESDS components or assemblies, especially those that contain very sensitive (low
ESD withstand voltage) devices. The number of product produced is usually easily found
from production statistics, and the number of devices or assemblies used in the product
can be calculated. Sometimes a difference between planned and realized product numbers
is evident, which can itself indicate that something was going on that might be worth inves-
tigation, such as excessive rework, shortage of parts, or field problems.
If the ESD withstand data for the ESDS devices can be obtained, then the lowest ESD
withstand devices are the ones most likely to be worth further investigation. Unfortunately,
this data is often difficult to obtain as it is not always published on device data sheets.
For the devices selected for further investigation, inventory and purchasing records can
be obtained. The actual number of ESDS items purchased in the period can be obtained,
including the starting and final inventory and the cost of the ESDS devices. It can also be
useful to find details of the locations or departments at which the ESDS devices are used.
Analysis of these data and the production figures may reveal a discrepancy between the
number of devices leaving in the product and the number purchased and remaining in stor-
age. An excessive difference may indicate, for example, rework or field service consumption
due to failures. Of course, this would not in itself confirm ESD as the failure cause in all
cases, but identification of any failure mode is usually a useful outcome. The analysis may
give useful indication of where to focus further resources.
Knowledge of the number of failed devices and their cost allows calculation of failed
component cost, but this represents only a part of the cost of the failures. If the number
of reworked items or field or other failures can be identified, the associated costs can
304 9 How to Evaluate an ESD Control Program
be estimated. These may include labor, facility, power, and other expenditure. Average
costs may be easier to estimate than real costs per individual failed item. Halperin (1986)
described preparation of the data in the form of a spreadsheet table, listing the sensitive
parts, and for each part the ESD withstand voltage, difference between numbers sourced
and those used in final in product, cost of each item, estimated “burden” of associated
costs, and total lost cost of the components and associated burden. The magnitude of total
cost per item give a means of ranking for further work, and the overall total cost for all
items gives a view of the possible cost of failures.
The cost of the failure is likely to vary with the production stage at which the failure is
found, including field failures. The cost of failure typically increases as the product advances
through the production process. Field failures are usually the costliest of all. With some
markets and product types (e.g. satellites and aerospace), the cost of a field failure, in eco-
nomic and other terms, is extremely high and could include equipment downtime, loss of
the product, and even threat to life or property. Some product may be irreplaceable or unser-
viceable. This consideration in itself may justify considerable care in ESD control during
product manufacture, storage, transport, and handling.
Halperin (1986) provided an example of this type of analysis. An interesting case study
was also later provided by Helling (1996) (see Section 9.5.4).
While it may be desirable to confirm ESD damage with failure analysis of the failed com-
ponents, few organizations do this in practice. One reason may be that failure analysis of
components can be a time-consuming process requiring several person-days specialist effort
and equipment. The outcome is not always conclusive in identification of ESD failures. In
particular, ESD damage and electrical overstress damage often show similarities. The situ-
ation can be worsened if adequate ESD control is not maintained at all stages of handling
including between discovery of failure and into failure analysis, as ESD could still damage
an already failed component!
Dangelmayer (1999) reported several interesting case studies that formed a convincing
demonstration of the economic benefit of ESD control. These used several different
approaches to establish the source or presence of ESD failures and benefit of ESD controls,
including the following:
● Use of failure analysis to identify ESD failures
● Correlation of ESD control deviations and ESD losses
● Comparison of failure rates of batches of product assembled with, and without, ESD con-
trol measures
● Reproduction of failure of components using simulated ESD
● Testing of devices before and after operations to determine the source of ESD damage
One common reason for my clients to first contact me for help has been to evaluate their
ESD control after a customer audit has declared it inadequate. The customer is not always
right in terms of their perception of control of real ESD risks. They do, however, sometimes
opt to take their business elsewhere due to perceived inadequate of ESD control in the orga-
nization’s facility. A dispute with a customer over this costs time and resources and is not
good for customer relations, even if their evaluation is of ESD control is incorrect.
Implementation of a good ESD control program, with good documentation of ESD con-
trol processes and compliance with a standard, can do much to convince a customer that
adequate care is taken and ESD control is effective. For some customers, compliance with
an ESD control standard can be a prerequisite to doing business with a supplier organiza-
tion. In this situation, the organization’s positive attitude to ESD control can even become
a useful marketing benefit.
control measure. Gumkowski and Levit (2013) examined the use of air ionization and
compared different types of ionizers in a semiconductor manufacturing process.
Helling (1996) found that internal studies had shown that a failure to observe an ESD con-
trol measure in their facility typically resulted in about 1% of ESDS devices being stressed
by ESD. About 10% of stressed devices resulted in a defect or failure. He therefore assumed
an ESD failure rate of 0.1% per ESD control fault. He calculated on this basis failure rates
for five production lines. He added the cost of repair of ESD failures. He found that 60% of
failures were found at PCB test, and 30% at system test. Ten percent of failures were found
at the customers site. His calculations were compared over two business years. The repair
costs at each stage was calculated as follows:
Stage repair cost = number of product × %failures × ESD failure rate (0.1%)
× repair cost per item
An example summary of some of Helling’s results for a manual process handling 80 000
products per year is given in Table 9.1. Two interesting things can be seen from this example.
First, the cost of failures increases as the failure is discovered at later stages in production.
Second, the most expensive cost is that of failures at the customers site. These characteristics
are likely to be typical of most cost of product failure profiles.
Helling estimated the repair costs for other processes in the same way and was then able
to add these to estimate the cost of ESD damage related repairs to the business. He then
estimated the cost of ESD control measures required for an example facility in terms of
packaging, ESD control equipment, training, audit, and other items for each business year
under consideration. The ROI could then be calculated as the cost/benefit ratio of expendi-
ture on ESD control compared to anticipated saving on ESD failures. For the two business
years considered he found ROI values of 3:1 and 11:1, respectively. He commented that ESD
protection faults led to the product being more expensive than necessary, and this alone,
aside from the loss of customer reputation that results from defects and failures, cost the
business more than ESD protection measures.
Other workers have reported high return on investment figures. Danglemayer (1999)
reported in some of his case studies ROI as high as 185% and even 950%. In another case
study, an investment of $1000 resulted in a saving of $6 000 000. He concluded that using
ESD controls can lower operating costs with a ROI of up to 1000% while simultaneously
enhancing product quality and reliability.
Table 9.1 Costs of repair of ESD damage in a manual process (Helling 1996).
In contrast, a simple standardized ESD control program implementing the same con-
trol measures in different areas may lead to reduced training and compliance verification
needs and costs and help prevent operator confusion. It may also reduce the risk and cost
of conflict with customers or auditors.
Against this, equipping EPAs with the same equipment, whether it is necessary for ESD
risk control or not, is likely to increase equipment and compliance verification costs. Any
additional equipment must of course be regularly tested and verified.
When optimizing an ESD control program, it is essential to analyze and understand the
ESD risks that are present in the processes and facilities. Without this, ESD controls may
be implemented that are not necessary or risks may be present that are not addressed. ESD
risks are often not obvious or well understood. Adequate skills and experience are required
for successful analysis of these. Implementation of a standard will address the standard
well-known ESD risks but may not cover more unusual ESD risks (Gaertner 2007).
With a little knowledge and understanding, the cost of ESD control can be high and
the effectiveness low (Smallwood et al. 2014). This is perhaps most likely to occur when
the low ESD withstand voltage devices are handled. With a high level of knowledge and
understanding, the cost of ESD control can be reduced and the effectiveness maximized.
Investment in high-level training in the principles and practice of ESD control for the ESD
coordinator and other personnel working on development of the ESD control program can
be worthwhile if not essential. Strategies for optimizing the ESD control program are further
discussed in Section 10.7.
The IEC 61340-5-1 states under the heading “ESD Coordinator” that “A person shall be
assigned by the organization with the responsibility for implementing the requirements of
this standard including establishing, documenting, maintaining, and verifying the compli-
ance of the program.” This translates into the checklist of Table 9.2.
Table 9.3 gives a checklist of the broad requirements of the ESD Control Program Plan.
The detail of compliance of the Compliance Verification Plan is covered in Table 9.7. As an
example, the Compliance Verification Plan might be counted as “defined” if some elements
of compliance verification are covered in some way, e.g. testing of wrist straps. That does not
mean it is necessarily considered adequate – further requirements are tested in Table 9.7.
Table 9.4 covers the requirements for “tailoring” given by the standard. This does not mean
that some tailoring must be specified – if none is needed, none needs be documented.
Compliance
Requirement evaluation Notes
Compliance
Requirement evaluation Notes
Compliance
Requirement evaluation Notes
Compliance
Requirement evaluation Notes
Table 9.5 covers the required content of the ESD Training Plan in detail, and Table 9.6
covers the detailed content of the ESD Control Product Qualification Plan. Table 9.7 covers
in more detail the required content of the Compliance Verification Plan.
Table 9.8 concerns the definition of grounding and bonding systems. Suitable grounding
methods must be defined, but not all the methods in this table must be used in all facili-
ties. Usually only one will be sufficient. For example, many facilities will use the electrical
protective earth as ESD earth (ground).
9.6 Evaluation of Compliance of an ESD Control Program with a Standard 311
Compliance
Requirement evaluation Notes
Compliance
Requirement evaluation Notes
Compliance
Requirement evaluation Notes
Compliance
Requirement evaluation Notes
Table 9.9 concerns personal grounding. The detailed pass criteria given in current stan-
dards 61340-5-1:2016 and ESD S20.20-2014 are given in Chapter 6, in this case in Table 6.5.
Table 9.10 gives general requirements for EPAs. Requirements for EPA equipment such as
floors, bench mats, and chairs are given in in Chapter 6, in this case in Table 6.6.
Table 9.11 covers the requirements for ESD protective packaging. The classification of
packaging materials and requirements of IEC 61340-5-3:2015 are given in in Chapter 6, in
this case in Tables 6.10 and 6.11. Table 9.12 covers the requirements for marking for ESD
control purposes.
Compliance
Requirement evaluation Notes
Compliance
Requirement evaluation Notes
Compliance
Requirement evaluation Notes
be tested in the Compliance Verification Plan and hence in evaluation of compliance of the
facility with the plan.
References
EOS/ESD Association Inc. (2014). ANSI/ESD S20.20–2014. ESD Association Standard for the
Development of an Electrostatic Discharge Control Program for – Protection of Electrical
and Electronic Parts, Assemblies and Equipment (excluding Electrically Initiated Explosive
Devices). Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015). ESD TR17.0–01-15. Technical Report for ESD Process
Assessment Methodologies in Electronic Production Lines – Best Practices used in Industry.
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016a) ESD SP10.1–2016. Standard practice for protection of
Electrostatic Discharge Susceptible Items–Automated handling Equipment (AHE). Rome, NY,
EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016b) ESD Association Electrostatic Discharge (ESD) Technology
roadmap – revised 2016. Available from: https://www.esda.org/assets/Uploads/docs/
2016ESDATechnologyRoadmap.pdf [Accessed: 10th May 2017]
Gaertner R. (2007) Do We Expect ESD-failures in an EPA Designed According to International
Standards? The Need for a Process Related Risk Analysis. Proc. EOS/ESD Symposium
EOS-29 Paper 3B.1 pp. 192–197
Gaertner R., Stadler W. (2012) Paper 3B.5. Is there a Correlation Between ESD Qualification
Values and the Voltages Measured in the Field? In: Proc. EOS/ESD Symposium EOS-34.
Rome, NY, EOS/ESD Association Inc.
Gumkowski G., Levit L. (2013) EOS-35 Paper 7B4. A New Look at the Financial Impact of Air
Ionization. In: Proc. EOS/ESD Symposium. Rome, NY, EOS/ESD Association Inc.
Halperin, S.A. (1986). Estimating ESD losses in the complex organisation. In: Proc. EOS/ESD
Symp. EOS-8, 12–18. Rome, NY: EOS/ESD Association Inc.
Halperin S. A., Gibson R, Kinnear J. (2008) EOS-30 2B-2. Process Capability & Transitional
Analysis. In: Proc. EOS/ESD Symp. Rome, NY, EOS/ESD Association Inc.
Helling, K. (1996). ESD protection measures – Return on investment calculation and case
study. In: Proc. EOS/ESD Symp. EOS-18, 130–144. Rome, NY: EOS/ESD Association Inc.
Industry Council on ESD Target Levels (2010) White paper 2: A case for lowering component
level CDM ESD specifications and requirements. Rev. 2.0. http://www.esdindustrycouncil
.org/ic/en/documents/6-white-paper-2-a-case-for-lowering-component-level-cdm-esd-
specifications-and-requirements [Accessed: 10th May 2017]
Industry Council on ESD Target Levels (2011) White paper 1: A case for lowering component
level HBM/MM ESD specifications and requirements. Rev. 3.0. Available from: http://www
.esdindustrycouncil.org/ic/en/documents/37-white-paper-1-a-case-for-lowering-
component-level-hbm-mm-esd-specifications-and-requirements-pdf [Accessed: 10th May
2017]
International Electrotechnical Commission. (2015) IEC 61340–5-3:2015. Electrostatics.
Protection of electronic devices from electrostatic phenomena. Properties and requirements
classifications for packaging intended for electrostatic discharge sensitive devices. Geneva, IEC.
International Electrotechnical Commission (2016) IEC 61340–5-1: 2016. Electrostatics – Part
5–1: Protection of electronic devices from electrostatic phenomena - General requirements.
Geneva, IEC.
International Rectifier. (2004). ESD Testing of MOS Gated Power Transistors. AN-986. http://
www.infineon.com/dgdl/an-986.pdf?fileId=5546d462533600a40153559f9f3a1243 [Accessed:
10th May 2017]
316 9 How to Evaluate an ESD Control Program
Jacob P., Gärtner R., Gieser H., Helling K., Pfeifle R., Thiemann U., Wulfert F., Rothkirch W.
(2012) EOS-34. Paper 3B.8. ESD risk evaluation of automated semiconductor process
equipment – A new guideline of the German ESD Forum e.V. In: Proc. EOS/ESD Symp.
Rome, NY, EOS/ESD Association Inc.
Lin N., Liang Y., Wang P. (2014) DOI: 10.1109/ICEPT.2014.6922951. Evolution of ESD process
capability in future electronics industry. In: Proc. 15th Int. Conf. Electronic Packaging Tech.
Bristol, England, IOP Publishing.
Paasi, J., Smallwood, J., and Salmela, H. (2003). EOS-25 Paper 2B4. New Methods for the
Assessment of ESD Threats to Electronic Components. In: Proc. EOS/ESD Symp, 151–160.
Rome, NY: EOS/ESD Association Inc.
Paasi, J., Salmela, H., and Smallwood, J.M. (2006). Electrostatic field limits and charge
threshold for field-induced damage to voltage susceptible devices. J. Electrostat. 64: 128–136.
Smallwood J M. (2019) Can ElectroStatic Discharge Sensitive electronic devices be damaged by
electrostatic fields? In: Proc. Electrostatics 2019. J. Phys. Conf. Se. Vol. 1322 01 2015 Available
from: https://iopscience.iop.org/article/10.1088/1742-6596/1322/1/012015/pdf [Accessed
Oct. 2019]
Smallwood J., Paasi J., (2003) Assessment of ESD threats to electronic devices, VTT Research
Report No BTUO45-031160
Smallwood J., Taminnen P., Viheriaekoski T. (2014) EOS-36. Paper 1B.1. Optimizing
investment in ESD Control. In: Proc. EOS/ESD Symp. Rome, NY, EOS/ESD Association Inc.
Stadler W., Niemesheim J., Seidl S., Gaertner R., Viheriaekoski T. (2018) EOS-40 Paper
1B.4.The Risks of Electric Fields for ESD Sensitive Devices. In: Proc. EOS/ESD Symp. Rome,
NY, EOS/ESD Association Inc.
Steinman A. (2010) EOS-32 Paper 3B3. Measurements to Establish Process ESD Compatibility.
In: Proc. EOS/ESD Symp. Rome, NY, EOS/ESD Association Inc.
Steinman A. (2012) EOS-34 Paper 2B.4. Process ESD Capability Measurements. In: Proc.
EOS/ESD Symp. Rome, NY, EOS/ESD Association Inc.
Swenson D. E. (2012) EOS-34 paper 3B.6. Electrical fields: What to worry about? In: Proc.
EOS/ESD Symp. Rome, NY, EOS/ESD Association Inc.
Tamminen, P. and Viheriäkoski, T. (2007). EOS-29 Paper 3B3. Characterization of ESD Risks in
an Assembly Process by Using Component-Level CDM Withstand Voltage. In: Proc.
EOS/ESD Symp, 202–211. Rome, NY: EOS/ESD Association Inc.
Viheriäkoski T, Ristikangas P, Hillberg J, Svanström H, Peltoniemi T. (2012). Paper 3B.2.
Triboelectrification of static dissipative materials. In: Proc. EOS/ESD Symp. Rome, NY,
EOS/ESD Association Inc.
Wallash, A. (2007). EOS-29 Paper 2B8. A Study of “Soft Grounding” of Tools for ESD/EOS/EMI
Control. In: Proc. EOS/ESD Symp, 152–157. Rome, NY: EOS/ESD Association Inc.
317
10
3.
A full-time ESD coordinator to serve as a consultant and oversee the plan
4.
An active committee to help implement the ESD program
5.
Realistic requirements
6.
Training for measurable goals
7.
Auditing using scientific measures
8.
ESD test facilities to qualify and test ESD control equipment
9.
A communication program to keep people aware of the ESD issue and demonstrate
progress
10. Systematic planning
11. Human factor engineering to take care of employee needs and render human error
unlikely
12. Continuous improvement
To specify and use appropriate and effective ESD controls, it is important to understand
ESD control technology and the role of different items of ESD control equipment. Speci-
fication of equipment without this understanding can lead to an overly expensive and yet
poorly effective ESD control program (Smallwood et al. 2014). The equipment specified may
be unnecessary or may fail to function as a system in the absence of other items.
present or that contain batteries that could give electric shock risk to personnel. Another
area is where processes require the operator to use personal protective equipment against
the effects of chemicals, high temperatures, or other risks.
Often, by careful evaluation of the risks and process steps, ways of working can be found
that both maximize safety and minimize ESD risks. As an example, handling an unpro-
tected ESDS device may be required while setting up for live testing with high voltages
present. Using personal grounding with live voltages present could give electric shock risks.
If it can be arranged that the handling of the ESDS device can be done before the high
voltages are present, the safety risk is removed. Not touching an ESDS device when live
voltages are present at the same time prevents ESD risk from charged personnel and pre-
vents electric shock to the personnel! If it is necessary to work on the ESDS device in
a live state, precautions such as use of insulating rubber gloves may be required to pro-
tect the person against electric shock risks. The same gloves may give some protection
to the ESDS device against human body ESD at body voltages up to the breakdown volt-
age of the gloves. (If the person’s electrostatic body voltage could exceed the breakdown
voltage of the glove, an ESD risk could result, but more importantly the protection pro-
vided by the glove against shock risk is compromised.) When using rubber gloves in this
way, an ESD risk could arise from electrostatic fields due to charging of the glove mate-
rial. This is usually a lower risk but should be evaluated and if necessary countered in
some way.
In practice, the effect of electric shock depends on the electrical current flowing through
the body, rather than the source voltage present. The effects were summarized by Dalziel
(1972) and become more hazardous as current flow level and duration of the shock
increases. The sensitivity of the body to shock also depends on the body parts that are
subjected to the current flow, and whether the source is direct current (DC) or alternating
current (AC), and the frequency of AC current. Women are typically susceptible at lower
current levels than men, and individuals vary greatly in their sensitivity. Other conditions
such as presence of moisture can be influential. As current is increased, the effects start
with perception of the current (tingling or warmth) and can include at higher levels painful
shocks, stopping of breathing, inability to let go, burns, ventricular fibrillation (cessation
of heart action), and immediate or delayed death.
Where safety guidelines or regulations do not cover a situation, the guidelines in
(Table 10.1) are often given in electrical textbooks, e.g. Nave and Nave 1985. Safety can be
evaluated by comparing possible electrical currents with these thresholds.
1 Threshold of perception
5 Maximum harmless current
10–20 Sustained muscular contraction “let go current”
50 Pain, possible fainting, exhaustion, heart, or respiratory effects
>100 Ventricular fibrillation, death possible
320 10 How to Develop an ESD Control Program
device to a low-resistance conductor often carries a charged device ESD risk. This risk can
be minimized by contacting the ESDS device through a resistive material (see Section 10.3)
rather than a low-resistance material such as a metal. Where ESDS devices must contact
conductive items (including other devices, tools, or PCBs), the ESD risks should be evalu-
ated, and ESD prevention measures may be required e.g. to reduce the voltages possible on
the ESDS device and conductors or increase the resistivity of the conductor.
The ESD control measures required within the EPA will be selected to address the ESD
risks found. In a typical EPA with manual processes, these will include
● Grounding of personnel handling ESDS devices either by wrist strap or by ESD control
footwear and flooring. (Occasionally other means e.g. personal grounding garments may
be used.) Suitable ground points for wrist straps must be provided.
● Where personnel are to be seated, ESD control seating is used, and personal grounding
via wrist straps is used.
● Specification of grounded ESD control materials for surfaces on which unprotected ESDS
devices could be placed, e.g. workstations and carts.
● Insulating materials are evaluated to decide whether they are essential to the process
or not.
⚬ Nonessential insulators should be kept well away from unprotected ESDS devices.
The ESD program may specify that they are excluded from the EPA.
⚬ Insulators that are essential to the process must be evaluated to determine whether
any ESD risk arises in their use. If an ESD risk is found, some means of ameliorating
it must be devised. Often this may be done by using an ionizer.
● Ionizers may be specified for control of voltages on essential insulators or nongroundable
isolated conductors
● ESD control packaging may be selected to protect ESDS devices when outside the EPA or
to prevent ESD risks arising from packaging materials used inside the EPA.
● ESD control garments may be specified to reduce the risk of electrostatic fields arising
from clothing worn by operators.
All the equipment and processes in the EPA should be designed with these require-
ments in mind. Well-designed common ESD control equipment normally fulfills these
requirements.
Other control items such as ESD control tools may also be specified. These control mea-
sures are further discussed in Section 10.5.13.4.
Sometimes ESD risks are identified that are not addressed by the usual ESD control mea-
sures and equipment. In this case, special ESD control measures and precautions must be
devised.
Operator safety should always be considered when developing an ESD program
(see Section 10.1.3).
● The ESD Control Program Plan, specifying the equipment and other ESD control mea-
sures to be used.
● An ESD Training Plan, specifying necessary ESD-related training requirements.
● An ESD Control Product Qualification Plan, specifying the criteria and requirements by
which ESD control equipment and products will be selected for use in the ESD control
program.
● A Compliance Verification Plan, specifying the checks and test program to be used.
When writing the ESD control program, bear in mind that there are usually at least two
audiences who must clearly understand the documents. First, the personnel who must
work with the documents, using them to maintain and implement the ESD control pro-
gram, must understand them.
Second, auditors (first, second, or third party) may need to audit the ESD control program
and check that what is happening in practice reflects the documented ESD control require-
ments. These auditors may need also to check that the documents are compliant with the
selected standard. It can save a lot of time and effort for auditors if the documentation is
laid out in a way that quickly shows that these requirements are met. If parts of the ESD
control program are documented in separate documents or procedures or are provided in a
different form (e.g. paper or intranet-based documents), suitable cross-referencing can be
used to make this clear.
324 10 How to Develop an ESD Control Program
Introduction
Scope
Terms and definitions
Personal safety
ESD control program
● ESD control program requirements
● ESD coordinator
● Personal grounding
References
Once the heading structure is in place, the text under each heading can be added accord-
ing to the requirements for ESD control in the facility and processes, in compliance with
the requirements of the standard. In some cases, it can be appropriate to include text based
on that of the standard being followed to ensure compliance with the standard. The follow-
ing sections give an example of how the basis of an ESD program based on IEC 61340-5-1:
2016 can be drafted from the heading structure. An outline of the content of each heading is
given. The requirements of 61340-5-1: 2016 are discussed in more depth in Chapter 6. Once
the document content is drafted, the headings can, if necessary, be reorganized to better
reflect the content and for clarity.
The ESD program documents should be written to conform to the organization’s internal
quality system procedures.
10.5.4 Scope
The section summarizes the scope of the ESD program. This should specify the areas and
activities in which ESDS devices are handled and will be covered by the ESD program. It
should cover all applicable parts of the organization’s work. The standard lists “activities
that: manufacture, process, assemble, install, package, label, service, test, inspect, transport,
or otherwise handle” ESDS devices.
The scope should also specify the range of ESD susceptibility of the components covered
by the ESD program. The default range from the standard gives “electrical or electronic
parts, assemblies, and equipment with withstand voltages greater than or equal to 100 V
HBM, 200 V CDM, and 35 V for isolated conductors.” If the devices handled are limited
to an ESD withstand voltage range higher than this default, the default can normally be
accepted. If devices of lower ESD withstand voltage are handled, a withstand voltage range
should be specified that includes the lowest withstand voltage device to be handled. For
example, if 60 V HBM or 180 V CDM devices are handled, the ESD withstand voltage range
should be specified to include these, e.g. greater than or equal to 50 V HBM and 150 V CDM.
Table 10.3 An example of specification of requirements for personal grounding equipment with
test method and test frequency.
Wrist strap, while worn, TM1 <35 MΩ and On each entry to EPA
resistance from body to >750 kΩ indicated by
groundable point green light on tester
ESD control footwear, while TM2 <10 MΩ and On each entry to EPA
worn, hand to metal foot plate >750 kΩ indicated by
green light on tester
ESD control footwear worn while TM3 <1 GΩ 6 monthly
standing on ESD control floor,
hand to ground.
Body voltage generated while TM4 100 V peak body 6 monthly
wearing ESD control footwear voltage
and walking on ESD control floor
Table 10.4 An example of EPA equipment pass requirements, test methods and test frequency
summarized as a table.
recognized by the user. All ESD control items will be connected to a common ground point
using this method. Grounding should comply with requirements of the National Electrical
Code. Any requirements of this that apply to the ESD ground can be specified or referenced
under this heading, if appropriate.
It is worth specifying that all conducting items and materials that contact ESDS devices
must be connected to this common ground point if possible. Any conductor that contacts
ESDS devices but for some reason cannot be electrically connected to ground must be
treated as an isolated conductor.
It is not good practice to have two different ground points in an EPA that are not connected
together as they could be at different voltages. It follows that if mains electricity protective
earth is present in the EPA, this should be used as, or connected to, ESD ground if possible.
consideration of the standard requirements and appropriateness of the facility and process.
In this example, the resistance of footwear, while worn by personnel, measured between
the hand and a metal plate under the foot, is specified to be <10 MΩ, which is much less
than the requirements of the standard. As it is within the requirements of the standard, no
tailoring explanation or justification is required.
As a key ESD control measure, it is wise to test personal grounding (wrist straps or
footwear) before use on each day of use. Some organizations require personnel to test these
at the beginning of each work shift or on each entry into an EPA. It is good practice to keep
records, such as a simple log (e.g. signed list) of this test. A failed wrist strap should always
be removed from use and repaired or replaced.
Where wrist straps are used, wrist strap grounding points should be regularly checked
for connection to ground. Where these are of a temporary nature (e.g. via connection to a
ground plug in a mains socket), tests should be more frequent due to the risk that accidental
disconnection of the ground path may occur. In the standard, testing of wrist strap bonding
points is addressed under “EPA Equipment.”
The boundaries of the EPA must be clearly identified and noticeable to personnel entering
or leaving the EPA. The way that the organization does this should be defined and described
here. Examples of typical signage should be shown.
The 61340-5-1: 2016 requires that “Access to the EPA shall be limited to personnel who
have completed ESD training. Untrained individuals shall be escorted by trained personnel
while in an EPA.” A clear statement like this should be included here, or at some other
appropriate point in the ESD Control Program Plan.
10.5.13.2 Insulators
This section should be used to explain and define identification and treatment of essential
and nonessential insulators used in the EPA. A similar approach can be taken to dealing
with other known electrostatic field sources.
Essential insulators are those that are required to be present as part of the product
or process, without which the process cannot be completed. Nonessential insulators
are those without which the process can be completed. The most common source of
high electrostatics fields in the workplace include insulators used in plastic (non-ESD
protective) packaging, documentation (file dividers and covers), equipment housings and
exposed construction, and other components used in the product. While electrostatic fields
do not in general cause direct damage to ESDS devices, they do lead to induced voltages on
ungrounded conductors and lead to the conditions in which ESD can arise if the conductors
are, or touch, ESDS devices. Contact between the ESDS device and a metal item while
within an electrostatic field can give a charged device ESD. It may be worth explaining this
risk in this section or somewhere else in the ESD control program such as the introduction.
There are essentially three ways of dealing with nonessential insulators in an EPA. The
organization should decide which approach they are going to take in each case and write
the ESD control program accordingly.
The first approach is to replace the insulator with a noninsulating (conducting) material
and ground it.
The second approach is to remove or eliminate the nonessential insulator from the EPA.
This approach is simple and clear and often easy to manage and train personnel to do. In
many cases, ESD control materials (static dissipative or conductive) can replace insulators
for use in the EPA. In some facilities and processes, however, it could give great inconve-
nience for various reasons. Inconvenience usually leads to difficulty in compliance.
A third approach is to keep nonessential insulators a minimum distance from worksta-
tions where ESDS devices are handled. This approach can sometimes be used in convenient
practices. (An example might be to have sign-off papers or computer equipment at a work-
station next to, but sufficiently far from, the workstation area at which work is done on the
ESDS device.) The disadvantage can be that effective procedures can be less clear to EPA
personnel and may require greater training for reliable implementation. Insulators or ESDS
can easily be transported from a position where they are allowed to a position where they
might cause ESD risk by personnel who are unwary or lack sufficient understanding of the
ESD control procedures.
The requirements for electrostatic field or voltage limits should be stated in the ESD
Control Program Plan. These limits can be taken from the standard, or other limits can be
defined. If the electrostatic field and voltage limits defined in the ESD Control Program Plan
332 10 How to Develop an ESD Control Program
are outside the limits given in the standard, this requires justification and documentation
as tailoring. This section should also define the actions to be taken if these limits are found
to be exceeded.
Papers and cardboard are highly variable materials that can vary by orders of magni-
tude in their electrical characteristics between grades and with humidity changes (and the
weather!). So, it is considered good practice to keep these materials away from ESDS devices.
61340-5-1: 2016 requires that “All non-essential insulators and items (plastics and paper),
such as coffee cups, food wrappers, and personal items shall be removed from the worksta-
tion or any operation where unprotected ESDS are handled.” A clear statement such as this
should be included in this section of the ESD Control Program Plan.
Pink polythene material used for ESD control relies on the presence of humidity for its
low-charging properties. It is often used within an EPA as a replacement for polythene
packaging e.g. for enclosing non-ESDS device items or documents. This material becomes
ineffective and can act like an insulator at low ambient humidity (<30% rh), charging highly.
There are other common items that could occur in practice and must be controlled
because they generate electrostatic fields, such as cathode ray tube (CRT) displays. These
should be identified, and how they will be treated should be defined.
Essential insulators cannot, by definition, be removed from the process in which they are
used. A strategy for dealing with these is shown in Figure 10.1. The first task is to determine
whether the insulator provides a significant ESD risk to the ESDS device in the process. This
can be evaluated by determining whether it can charge up and give a significant electrostatic
field at the possible position of the ESDS device. If not, it is necessary only to document the
evaluation.
No
Could it give electrostatic fields Document in the ESD
at the site of the ESDS device? Program Plan
Controlled Risk
No
Yes Can it be replaced by a
grounded conductor?
No
Yes Can it be moved away from the
ESDS device?
Use an ioniser
or other means of risk
Replace with reduction
grounded conductor Yes
Controlled Risk
Move it away
Document in the ESD
Program Plan
Controlled Risk
If the essential insulator can provide significant electrostatic fields at the position of the
ESDS device, it remains to decide what to do with it. Sometimes they can be replaced with
static dissipative materials and grounded. An example might be part of a test or assembly
jig that does not need to be made of insulating material. Sometimes, although the insulator
must remain nearby, it can be moved sufficiently far away from the position where ESDS
devices are handled to reduce the field at the position of the ESDS devices to a low level.
If neither of these techniques can be used, an ionizer or some other means of reducing the
electrostatic field must be found. Whichever technique is used should be documented in
the ESD Control Program Plan.
Can it be grounded?
No
Ground it
Controlled risk
Yes
Use an ioniser
If the isolated conductor could touch an ESDS device, one way to reduce the voltage
on it is to use an ionizer. If an ESDS device and isolated conductor spend sufficient time
close together in the ion stream from an ionizer, they are both likely to reach a similar
voltage near the offset voltage of the ionizer. The time taken to achieve this will depend
on the ionizer and process circumstances. To show that this has been successful, it would
be necessary to measure the voltages on the ESDS device and the conductor with an elec-
trostatic voltmeter and compare them. Measurement of these voltages would require an
electrostatic voltmeter capable of measuring voltage on small items. This is not an easy
task, especially in a fast-moving automated process.
Requirements for other items not specified in the standard may also need to be specified,
e.g. gloves, finger cots, tools, or soldering irons. Test methods and pass criteria will also need
to be specified for these items.
Sometimes it is desirable to specify a limit that is different from that given in the standard.
If this limit is within the range given by the standard, it does not represent tailoring of the
ESD program. For example, the organization may have selected a floor material that when
installed has a resistance to ground below 10 MΩ. In this case, 10 MΩ could be specified as
the upper limit for compliance verification to pick up changes in material performance or
due to surface contamination.
In contrast, if the proposed limit is outside the range given in the standard, this should be
considered a tailored requirement. As an example, many organizations that established an
ESD control program under earlier versions of 61340-5-1 would have accepted seats with
resistance to ground up to the previously specified limit of 10 GΩ. In the 2016 version, the
limit for seats was dropped to 1 GΩ. For compliance with 61340-5-1:2016, the organiza-
tion would have the choice of replacing all the seats according to the new requirements or
including a tailored limit of 10 GΩ in their ESD control program. If they choose the tailored
option, the decision should be supported by some tests demonstrating that no additional
risk is provided by this specification.
Don’t use words like conductive, antistatic, or dissipative to specify equipment unless they
are defined for that product in a standard that is referenced as part of the specification.
Contrary to popular usage, these words do not have generally standardized meanings for
many products. Even worse, they can mean different things for the same product in different
industry areas and for different products. So, unsuitable products may be purchased if using
these words as the specification.
Instead, all products are best specified using verifiable parameters (e.g. resistance), mea-
sured using standard test methods. It is important to specify the test method used, because
different test methods may give different measurement results. Common measurable
parameters include resistance to ground or to a groundable point for many types of ESD
control equipment, resistance point to point on a surface, material or garment, surface or
volume resistance for packaging, and decay time and offset voltage for ionizers.
An example of ESD control equipment summarized with test method, pass criteria, and
test frequency is given in Table 10.4. The pass criteria and test frequency should be specified
after consideration of the standard requirements and the needs of the facility and processes
present. The test methods TM5–TM9 must be written up and referenced. These should be
based on those given in IEC 61340-5-4 (International Electrotechnical Commission 2019)
or ESD TR53-01-15 (EOS/ESD Association Inc 2015; see Chapter 11).
Bench Mats and Other Surfaces on Which ESDS Devices Are Placed
Any surface on which unprotected ESDS device could be placed, for example storage racks
and cart shelves, should be subject to the same requirements as work surfaces. If ESD con-
trol packaging materials are used in this way, they should also be specified as per work
surfaces.
Compliance verification of these surfaces is normally specified and measured as resis-
tance between the surface and ESD ground (Rg ) (see Section 11.8.1).
336 10 How to Develop an ESD Control Program
Seating
If seating is needed at a workstation where ESDS devices are handled, then it must be
specified for ESD control according to the standard requirements. Seats are normally most
conveniently grounded through an EPA floor, and so an ESD control floor must usually also
be present. In the absence of an ESD control floor, it is sometimes possible to ground the
seat through a ground cord, but this can be inconvenient and may easily become discon-
nected. The requirement for seating should be thoroughly investigated before adding to the
ESD Control Program Plan.
Compliance verification of seats is normally specified and measured as resistance
between the seat surface and ESD ground (Rg ) (see Section 11.8.1).
Tools
When handling very sensitive ESDS devices, any tools that could contact ESDS devices
(e.g. cutters or pliers) could provide a risk of ESD damage. This can be addressed by spec-
ifying ESD tools in which the parts that contact ESDS devices are grounded through the
user’s body.
Compliance verification of tools can be specified and measured in various ways. There
is no standard test specified in current standards. One method is to specify the resistance
between the tool contact tip and ESD ground (Rg ) when the tool is held by the operator
(see Section 11.9.5).
Another way is to specify the maximum charge decay time measured using a charged
plate monitor (CPM) (see Section 11.9.8.1).
10.5 Documentation of ESD Procedures 337
Ionizers
Ionizers may be needed if there are essential insulators or isolated conductors present that
evaluation has shown can charge up and cause ESD risk (see Section 4.6).
Ionizers are specified for compliance verification in terms of charge decay time and offset
voltage, measured using a CPM (see Section 11.8.8).
10.5.16 References
It is normally useful to list the references used in preparing the document, including the
standards used and their user guides, and any in-house procedures, reports, test procedures,
or other documents specified.
protection. It is possible to use nonstandard or tailored ESD control methods in EPAs where
ESDS devices are handled by less common processes. The choice of ESD control mea-
sures used governs the cost of ESD control measures used in the EPAs and ESD protective
packaging used outside those areas. Documentation, training, and compliance verification
programs are also essential parts of ESD control. Trade-offs can be made between these
costs in order to optimize the ESD control program, although a high level of expertise may
be needed to do this successfully (Smallwood et al. 2014).
The first and most obvious benefit of an ESD control program is to protect ESDS devices
during processes and maintain the level of ESD damage at an acceptably low level. There
are also other benefits that could be part of the rationale for ESD control. Maintenance of an
ESD control program to a recognized international standard could be an important part of
the organization’s quality program. It may even be essential to prequalification of the orga-
nization as a provider of product to some customers. If a customer audits the organization,
a favorable impression of ESD control (whether the customer has a good understanding of
the topic or not) can be a vital step in the path to obtaining an order. An adverse impres-
sion and disagreements over the adequacy of ESD control can provide a significant block
to customer acceptance.
The main cost of inadequate ESD control is often thought to be damage to ESDS devices
handled in the organization’s processes. This cost, for many organizations, is far from clear
or quantifiable as failure analysis often is not done to the level that ESD failures are detected.
Other costs can also be significant but are often not considered. These can include prod-
uct failures, unreliability or drift in characteristics, additional test and rework expenditure,
delays to shipments, or need to overstock frequently failing items. Some of the most expen-
sive costs can be those related to customer service – cost of reaction to customer complaints
and failures at the customer’s site, impaired product or company reputation, and a possible
result in reduction of sales.
The real cost/benefit balance of an ESD control program depends on the time and
resources spent on all the aspects of ESD control. Optimizing the ESD control program
should consider all the desired benefits and areas of cost and determine an appropriate
balance between them. This means that trade-offs between investment in the different
aspects are possible and probably necessary.
So, it is worth carefully considering where the EPA boundary should be placed in order to
enclose all necessary activities while minimizing the EPA size and excluding unnecessary
and incompatible activities.
If inspection or test of ESDS devices is required, an EPA should be provided for the
purpose. It is preferable not to use this for any non-EPA activity, as this can be a means of
introducing noncompliant materials and lax EPA procedures. An EPA bench used for other
purposes should be checked for compliance and recommissioned before use as an EPA.
Goods In and Stores areas do not necessarily need an EPA. They will need one only if
ESD protective packaging must be opened and the ESDS device handled (e.g. for counting
or inspection) in an unprotected state. If possible, it can be preferable to avoid having this
sort of activity and therefore avoid the necessity of having an EPA. If necessary, the EPA can
be as simple as a single carefully specified workstation.
There is a considerable risk that EPA workstations in Goods In and Stores areas become
contaminated with secondary packaging and other insulating materials. To minimize this
risk, EPA workstations should not be used for non-EPA activities. Personnel using these
facilities may need to be carefully trained in avoiding non-compliances and ESD risks.
Siting of EPA workstations should be carefully chosen to be well away from areas where
high-level static generators such as secondary packaging are present. As an example, an
EPA workstation that backs onto a storage rack in an uncontrolled area could be subject to
considerable ESD risk from the static generators stored on the rack.
10.8.3 Kitting
Kitting activities may or may not require use of an EPA for handling unprotected ESDS
devices. Nevertheless, any kit destined for use in an EPA will be a source of noncompli-
ant packaging or other materials and items unless carefully specified to avoid this. Tote
boxes holding kits for transport into an EPA must be specified as compliant with EPA entry.
Items included within the kit, including mechanical, consumable, or non-ESDS device
components, must be supplied in suitable ESD protective packaging materials. Insulating
materials and items must be minimized or eliminated. If not eliminated, how they are used
and where they occur in the EPA must be carefully controlled to avoid ESD risks to ESDS
devices. If noncompliant items are included in a kit, they will cause noncompliance in the
EPA to which they are delivered.
10.8.4 Dispatch
Dispatch is another area that may or may not need EPAs as well as unprotected areas.
Dispatch areas commonly have large amounts of secondary packaging materials including
insulating packing tapes and other high-level static generators. Many of the considerations
for these areas are like Stores and Goods In areas.
Another consideration is that high levels of paper or cardboard dust, fibers, and larger
particles can be generated during handling of packaging materials. These can cause con-
tamination problems for some types of products handled in the vicinity.
10.8.5 Test
Test is probably the area most likely to require some tailoring of ESD control measures
due to safety considerations if high voltages and manual handling are present. This
often means that personal grounding requirements must be carefully thought through to
maximize safety and minimize ESD risk.
References 343
References
EOS/ESD Association Inc. (2015). ESD TR53-01-15. Technical Report for the Protection of
Electrostatic Discharge Susceptible Items – Compliance Verification of ESD Protective
Equipment and Materials. Rome, NY, EOS/ESD Association Inc.
International Electrotechnical Commission (2016) IEC 61340-5-1: 2016.
Electrostatics – Part 5-1: Protection of electronic devices from electrostatic phenomena - General
requirements. Geneva, IEC.
International Electrotechnical Commission. (2019) IEC TR 61340-5-4:2019. Electrostatics -
Part 5-4: Protection of electronic devices from electrostatic phenomena – Compliance
Verification. Geneva, IEC.
Nave, C.R. and Nave, B.C. (1985). Physics for the Health Sciences, 3e. W B Saunders.
ISBN: 0 7216 1309 8.
Smallwood, J., Taminnen, P., and Viheriaekoski, T. (2014). Paper 1B.1. Optimizing investment
in ESD control. In: Proc. EOS/ESD Symp EOS-36. Rome, NY: EOS/ESD Association Inc.
Snow, L. and Dangelmeyer, G.T. (1994). A successful ESD training program. In: Proc. EOS/ESD
Symp. EOS-16. Rome, NY: EOS/ESD Association Inc. pp. 94-–94-12.
345
11
ESD Measurements
11.1 Introduction
All materials and equipment used for electrostatic discharge (ESD) control are designed
and made to have certain properties that enable them to function. These commonly are
based on the principles of replacing exposed insulating materials with noninsulating mate-
rials and establishing a ground path for any charge generated on the equipment or material
to dissipate to electrostatic discharge protected area (EPA) ground. In addition, charge on
essential insulators may need to be neutralized using an ionizer or ESD risks controlled by
other means.
Many of the facilities equipment and materials needed for use in ESD prevention are
specified in various standards worldwide. These standards establish performance criteria
for compliance, and test methods for measuring this performance. The two main standards
discussed in this book are IEC 61340-5-1 and ESD Association ANSI/ESD S20.20 (EOS/ESD
Association Inc. (2014a)). IEC 61340-5-1 has also been adopted as a national standard in
many countries, in Europe becoming the European Norm EN61340-5-1. Individual coun-
tries may have their own versions, and, in the United Kingdom, this is BS EN 61340-5-1
(British Standards Institute. (2016)).
This chapter explains how to make basic measurements for use in the assessment of
equipment for compliance with these standards. It is a guide to the main measurements
that ESD coordinators will wish to make in their facilities, giving some practical guidance
and “work-arounds” where necessary, and giving some information on the more unusual
tests and nonstandard tests that can be used where standard tests are not specified or are
unsuitable.
For clarity, the 61340-5 series terminology is used, although the Electrostatic Discharge
Association (ESDA) series test methods are in many cases nearly identical and are cross-.
In some cases, differences with the ESDA standards are noted.
61340-5-1 and 20.20 use a variety of basic test methods to assess the performance of ESD
prevention equipment and materials. These include
● Point-to-point resistance
● Resistance to ground or groundable point
● Surface and volume resistance of ESD protective packaging
● End-to-end resistance of a ground cord
● Personal grounding resistance tests
● Electrostatic fields and voltages
● Walk test of footwear and flooring
● ESD shielding test of shielding bags
A “groundable point” can be, for example, a stud on a bench or floor mat that is intended
to be connected to ESD ground in the final installation.
Table 11.1 Summary of test methods and their application in 61340-5-1, and corresponding ESD
Association standards.
Table 11.2 Other IEC test methods that can be used in ESD protective
equipment, materials, and packaging evaluation.
Corresponding
IEC standard IEC title ESDA standard
The test methods will be described and demonstrated in the following sections. Simple
and clear test procedures for each measurement are given, which could be used to form the
basis of a company test procedure.
The test method standards available from IEC and ESDA at the time of writing are listed
in Tables 11.1 and 11.2. Many of the test methods in the IEC system have corresponding test
methods in the ESDA system, and vice versa, and these are indicated. It should be noted
that “correspondence” does not mean “equivalence” – there may be differences between
these documents. In some cases, the correspondence is with only part of the corresponding
document, or there may be other differences. As always, when aiming to comply with a
standard, it is necessary to refer to that standard for full and up-to-date details.
consideration in developing compliance verification tests is that these will need to be done
on a regular basis and test many items during an audit. A quick and easy but effective test
is required to minimize the time and effort required. Compliance verification tests are done
under the ambient atmospheric conditions in the EPA. It is good practice to make a note of
these during the test.
The 61340-5-1 and S20.20 standards have related documents IEC 61340-5-4 (Inter-
national Electrotechnical Commission 2019) and ESD TR53-01 (EOS/ESD Association
Inc. (2018a)) that specify tests adapted from other standards specifically for compliance
verification testing.
the 20.20 system. Some gaps in the standard test methods are also shown. Table 11.2 gives
some additional test method standards that can be useful in the evaluation of ESD protective
equipment, materials, and packaging.
Table 11.3 Test voltages used in resistance measurements required in 61340-5-1 and S20.20
standards.
Standards
Usage Test voltage Resistance range referenced
a) Where measurements involve the human body, the resistance meter should be current limited to a
maximum of 0.5 mA (See Section 10.1.3).
11.6 Measurement Equipment 351
Where high voltages are used for measurement, always assess the safety issues and take
appropriate safety precautions. Most handheld instruments are unlikely to source danger-
ous electrical currents, but check your instrument to evaluate any risks.
ESDA STM2.1
ESDA STM4.1
ESDA STM7.1
Characteristic 61340-4-1 61340-2-3 EN100015-1 ESDA STM12.1
Mass (kg) 2.5 ± 0.25 or 2.5 ± 0.25 2.5 ± 0.5 2.27 ± 0.06
5.0 ± 0.25a)
Diameter (mm) 65 ± 5 63.5 ± 1 75 63.5 ± 0.25
b)
Electrode surface 60 ± 10 50–70 50–70
hardness (Shore A)
Resistance (electrode <1 kΩb) <1 kΩb) <1 kΩc)
placed on metal sheet)
Notes:
a) According to 61340-4-1:2003, the 2.5 kg electrode is to be used for measurements on hard nonconformable
surfaces. The 5 kg to be used for measurements on all other surfaces. The first is applicable to most appli-
cations under 61340-5-1.
b) Conductive rubber pad need not be used with conformable (e.g. textile) surfaces.
c) Measured at 10 V between two electrodes placed on metallic surface.
352 11 ESD Measurements
Figure 11.1 An example of 2.5 kg electrodes according to IEC 61340-2-3 and ESD STM11.11.
Organizations in Europe that have been involved in ESD control for many years often
have EN100015 (CENELEC 1992) electrodes. The EN100015 electrodes may give different
results due to their different diameter and lack of conductive rubber face material but may
give results that correlate with the other electrodes. The lack of rubber face material means
that they might not make good contact with hard surface materials. These electrodes are
not recommended for making measurements according to 61340-5-1.
Table 11.5 Comparison of concentric ring surface resistance measurement electrodes from
various standards.
d2 = 57 mm d1 = 30 mm
Figure 11.2 Concentric ring electrode contact area according to IEC 61340-2-3 and ESD STM11.11
and ESD STM11.12.
Figure 11.4 Two-point probe electrode according to IEC 61340-2-3 and ESD STM11.13.
Figure 11.6 Example of standard handheld electrodes (right) and a nonstandard electrode (left).
leads to test equipment (Figure 11.6). In most cases, the handheld electrode size and shape
will have little effect on the measurement results, providing there is sufficient contact area
with the skin of the hand.
Figure 11.7 A simple tool test electrode (left) and underside (right) showing insulating feet.
11.6 Measurement Equipment 357
oxidize. Aluminum is not suitable due to its formation of an alumina surface layer that can
affect measurements.
Figure 11.9 Electrostatic field meters. (left) Induction type and (center and right) two field mill
type instruments.
The apparent voltage reading increases if the meter is brought closer to the surface. A
correction can be made if the distance is known, but it is normally most convenient to make
readings with the field meter/voltmeter held at the calibrated distance. Some meters have
a means of gauging the correct distance, e.g. converging lights or pillars that are placed in
contact with the surface.
These instruments operate by measuring the charge induced on a capacitive sensor plate
by an electrostatic field. The instrument must be grounded correctly when measurements
are made.
Some types of simple low-cost induction type instrument sense the field by charge
induced on a simple metal plate (Figure 11.9, left). These tend to suffer from drift. It is
important to zero these instruments before each measurement, with the sensitive aperture
shielded by a grounded surface or in a zero-field region, before each measurement is made.
“Field mill” type instruments have a rotating mechanical shield interrupting the electro-
static field to the sensor (Figure 11.9, center and right). These instruments are self-zeroing
and do not suffer from drift as much as the simple instruments do.
Field meters used to measure surface voltage of insulators for evaluation of risks in
61340-5-1 and 20.20 should be calibrated to read voltage at a distance of 2.5 cm (1 in.).
Meters calibrated at other distances will give different voltage readings.
Figure 11.10 A handheld noncontact electrostatic voltmeter. The sensitive tip contains a vibrating
reed sensor.
Figure 11.11 A contact voltmeter. Source: D E Swenson. The sensing tip has extremely high
impedance.
High-voltage source
Many CPMs on the market are not built according to these standards and have a smaller
plate size and capacitance. They are often smaller and lighter than a standard CPM and are
used for functional and comparative measurements with ionizers. These may often give
different results to the standard CPM and should be correlated to a standard CPM for com-
pliance verification measurement against standard ionizer requirements.
In practice, most organizations will use a nonstandard CPM for comparative and func-
tionality measurements on ionizers in their working positions in a workstation. In some
cases, a smaller sized nonstandard CPM plate can better represent the smaller area of a
typical ESDS device.
11.7 Common Problems with Measurements 361
Figure 11.13 Example of a CPM built according to the IEC 61340-4-7 and ESD STM3.1 standard.
Note the large 15 × 15 cm plate. Handheld electrode and voltmeter attachment accessories are also
shown. Source: D E Swenson.
11.7.1 Humidity
It is good practice to measure the ambient relative humidity when making electrostatic
measurements. High ambient humidity can cause insulating materials to show lowered
surface resistance <1011 Ω (100 GΩ). The instrument may be measuring the resistance of
the water layer adsorbed on the material. If in doubt, repeat the measurement under dry
(<30% rh) conditions.
hands may affect the measurement. This is usually only a problem with high-resistance
measurements above about 1010 Ω (10 GΩ) but can be worse under high-humidity condi-
tions. Sometimes test leads can be further isolated for use in high-resistance measurements
by threading them through a clean polyethylene tube.
Sometimes it is necessary, e.g. for resistance to ground measurement, to use very long
leads at least on one side of the meter. Connect the long lead on the earthy side of the meter
to reduce leakage problems.
Equipment required:
● One 2.5 kg resistance measurement electrode.
● Connector to ESD ground (0 Ω resistance).
● 10/100 V high-resistance meter with test leads.
● Work surface or floor to be tested.
High-resistance meter
10V / 100V
2.5 kg resistance
measurement
electrode
Surface of item
measured
ESD ground
Procedure:
Common Problems
ESD ground is often, but not always, mains electrical safety earth. Make sure you have
identified the correct ESD ground before making connection. The ESD ground should be
specified in the ESD Control Program Plan document.
If the ESD earth is mains electrical safety earth, it is usually convenient to measure the
resistance from the electrode to an ESD earth connector. Many of these have an internal
1 MΩ resistor. If present, the value of this must be subtracted from the measurement result.
If the measured item has resistance very much greater than 1 MΩ, this added resistance
may be neglected. To avoid confusion, it is often better to reserve ground cords and earth
connectors that do not have these built-in resistors for use in ESD measurements.
If you check resistance to ground back to a convenient mains electrical earth point, don’t
forget to check that the earth point is really connected back to safety earth. An extension
lead can be found to be unplugged or the electrical earth had somehow disconnected!
A higher than expected resistance result can be due to dust and dirt or other contamina-
tion accumulated on the surface. This is an indication that the cleaning regime may need
to be improved.
364 11 ESD Measurements
Resistance meter
10V / 100V
ESD ground
If the surface is dusty, it can be wiped with a dry cloth or paper towel before repeating
the measurement. If the surface is sufficiently dirty to require cleaning with a liquid clean-
ing material, it must be allowed to dry before measurement; otherwise, the measurement
results may be affected. Even a small amount of moisture on the surface will lower the
measured resistance results.
Equipment required:
● One 2.5 kg resistance measurement electrode.
● ESD earth connector.
● 10/100 V high-resistance meter with test leads.
● Sample chair for testing.
Procedure:
● Place electrode on chair seat.
● Connect the resistance meter between electrode and ESD earth.
● Measure result at 10 V.
● If R > 1 MΩ, measure result again at 100 V.
Resistance meter
10V / 100V
Separately test
resistance to ground
point and each
conductive wheel
Groundable point
(Figure 11.17). Resistance to several wheels or feet may be tested to see how many provide
groundable points.
To isolate the chair from any conduction paths through the floor, the chair must be placed
on an insulating surface that has a resistance considerably higher than the resistance
expected of the chair. This is a test that is best done in a test workshop and not in the EPA.
Equipment required:
● One 2.5 kg resistance measurement electrode.
● 10/100 V high-resistance meter with test leads.
● Insulating support >1010 Ω.
● Metal plate electrode (for under wheel).
● Sample chair for testing.
Procedure:
● Set the chair on the insulating support. Place metal plate electrode under a grounding
wheel or foot.
● Place the resistance measurement electrode on chair seat.
● Connect the high-resistance meter.
● Measure result at 10 V.
● If R > 1 MΩ, measure result again at 100 V.
Common Problems
It is usually wished to test parts of the chair such as the seat back or arms, which may not
be horizontal or have large flat areas on which an electrode can be easily balanced. Some
ingenuity is required to get around these problems. If the chair has an easily connected
groundable point, then the matter is simplified. The resistance from seat back to groundable
point may be measured with the chair laid on its back on an insulating support plate.
366 11 ESD Measurements
If the chair has an exposed metal part of the chassis that is confirmed to be connected
reliably to the groundable point, then the resistance of arms and back can be measured to
this metal part in subsequent measurements.
Resistance meter
10V / 100V
2.5 kg
Conductive
rubber
300 mm
Equipment required:
● Two 2.5 kg resistance measurement electrodes.
● 10/100 V high-resistance meter.
● Sample work surface.
Procedure:
● Place the electrodes on work surface at least 250 mm apart.
● Connect the resistance meter.
● Measure result at 10 V.
● If R > 1 MΩ, measure result again at 100 V.
● Note the result.
The measurement can be repeated in different locations and orientations to detect
whether the material has significant variation with position or direction.
Figure 11.20 Measurement of resistance point to point of a garment across panels. The garment
has been placed on a slab of insulating support material.
(Figure 11.20). Cuff-to-cuff measurements verify the connection of the garment materials
via the materials and seams between the cuffs.
Equipment required:
● 2 off 2.5 kg resistance measurement electrodes.
● 10/100 V high-resistance meter with test leads.
● Garment specimen to be tested.
● Insulating support.
Procedure:
● Place the insulating support on a working surface. Lay out the garment on the insulating
support, if possible, with a single layer of fabric.
● Place the electrodes on separate panels of the garment a convenient distance apart.
● Connect the high-resistance meter.
● Measure resistance result at 10 V.
● If R > 1 MΩ, measure result again at 100 V.
● Note the result.
Making several measurements in different areas of the material and orientations of elec-
trodes will indicate the variability of the surface resistance, and whether there is any direc-
tionality.
11.8 Standard Measurements Specified by IEC 61340-5-1 and ANSI/ESD S20.20 369
Common Problems
Many garment materials have high point-to-point resistance. Measurements above 1010 Ω
(10 GΩ) tend to be unreliable, and results are dependent strongly on humidity.
If the insulating mat is not sufficiently insulating, the measurement may be affected, and
the resistance result reduced. Check the point-to-point resistance of the insulating support
before the test to make sure it’s at least 10 times the expected resistance of the garment.
Equipment required:
● 2 off 2.5 kg resistance measurement electrodes.
● 10/100 V high-resistance meter with test leads.
● Garment specimen to be tested.
● Insulating support.
● Insulating cuff inserts. (These can be made from sheet insulating plastic material.)
Procedure:
● Place the insulating support on a working surface. Lay out the garment on the insulating
support.
● Place the electrodes on separate panels of the garment.
● Connect the high-resistance meter.
Figure 11.21 Measurement of cuff-to-cuff resistance of a garment, (above) with garment hanging
or (below) with garment resting on an insulating support. Source: D E Swenson.
370 11 ESD Measurements
Figure 11.22 Contacting (left) the inside or (right) the outside of a garment cuff with a 2.5 kg
electrode. The garment is placed on an insulating support. An insulating separator is used when
contacting the outside of the garment sleeve.
Equipment required:
● Two clamp electrodes suspended from an insulating frame.
● 10/100 V high-resistance meter with test leads.
● Garment specimen to be tested.
Procedure:
● Connect the clamp electrodes to the garment cuffs and suspend the garment.
● Connect the high-resistance meter to the clamp electrodes.
● Measure resistance result at 10 V.
● If R > 1 MΩ, measure result again at 100.
● Note the result.
Figure 11.23 Measurement of cuff to cuff resistance of a garment hung from clamps. Source: D E
Swenson.
and the resistance should also be measured to these. The resistance should be measured
between a 2.5 kg electrode placed on the material panels and the inner surface of the cuffs
(Figures 11.21 and 11.22).
Equipment required:
● 10/100 V resistance meter with appropriate leads and connectors
● Ground cord to be tested
372 11 ESD Measurements
Procedure:
● The wrist strap cord to be tested is connected to the resistance meter using appropriate
connectors.
● Note the result.
Common Problems
Do not be tempted to hold the cord ends in contact with test leads with the fingers. If you
do this, you risk connecting your body in parallel with the cord. This would then give an
incorrect result. Also, make sure any connectors used do not contact a conductor such as
an ESD control bench surface.
Equipment required:
● Handheld electrode
● 10/100 V resistance meter with test leads
● Wrist strap and cord to be tested
Procedure:
● The subject to be tested should wear the wrist strap in contact with their skin.
● The wrist strap cord to be tested is connected to one terminal of the resistance meter.
● The second terminal of the resistance meter is connected to the handheld electrode.
● The subject holds the electrode. Note the result.
11.8 Standard Measurements Specified by IEC 61340-5-1 and ANSI/ESD S20.20 373
Figure 11.25 Simple proprietary wrist strap checkers make verification easy. The pass/fail limits
of these checkers must correspond with the requirements of the ESD control program.
Figure 11.26 A handheld electrode used with suitable resistance meter allows checking of wrist
strap as worn.
374 11 ESD Measurements
Common Problems
If a wrist strap checker is used, the pass/fail limit of the checker must be selected to corre-
spond with the limits specified in the ESD control program.
Many wrist strap checkers also give a “low fail” if the resistance is less than a minimum
value. This “low fail” limit must correspond to the lower limit specified in the ESD control
program.
Note that current versions of 61340-5-1 and 20.20 do not specify a low limit for wrist strap
resistance, (see Section 7.5.3). Nevertheless, many organizations will often wish to specify
minimum resistance from body to ground for safety in the presence of high voltages or other
reasons.
Equipment required:
● Handheld electrode
● 10/100 V resistance meter with test leads
● ESD footwear
● Foot plate electrode, big enough to accommodate an entire foot
● Insulating support mat
Procedure:
● The foot plate is placed on the insulating support mat and connected to the resistance
meter. The second terminal of the resistance meter is connected to the hand touch elec-
trode.
● The subject should wear the footwear to be tested.
● The subject stands with one foot on the foot plate and one foot on the insulating mat.
● The subject holds the handheld electrode.
● Note the result.
● repeat test for the other foot.
Common Problems
If a proprietary footwear checker is used, make sure that the upper and lower pass/fail limits
correspond with those specified in the ESD control program.
Figure 11.28 Measurement of resistance to ground when grounded via a groundable garment.
Source: D E Swenson.
376 11 ESD Measurements
Figure 11.29 Checking the resistance from person to ground using a portable wrist strap checker.
earth bonding point or EPA floor. The test can also be done with a portable wrist strap
checker (Figure 11.29) or footwear checker, providing the pass/fail thresholds correspond
to the ESD control program requirements.
Equipment required:
● Handheld electrode
● 10/100 V resistance meter with test leads
● ESD ground connector (0 Ω resistance)
Procedure:
● The person to be tested should wear their personal grounding equipment (wrist strap
or ESD control footwear). The wrist strap should be connected to ground in its normal
working position. For footwear-flooring grounding, the person should stand on the ESD
control floor to be tested.
● Connect one side of the resistance meter to ESD ground via the ESD ground connector.
● Connect the other side of the resistance meter to the handheld electrode.
● The subject holds the electrode. Note the result.
Common Problems
The test gives an easy way of checking personal grounding as a system in operation. Don’t
forget to check that the ESD earth point connected to the resistance meter is also connected
back to ESD earth!
The test does not distinguish between individual grounding systems if more than one is
in operation (e.g. simultaneous grounding through wrist strap and footwear-flooring).
11.8 Standard Measurements Specified by IEC 61340-5-1 and ANSI/ESD S20.20 377
Sometimes, personnel who have dry skin do not easily make good connection with wrist
straps. This effect may be most evident just after the wrist strap is first put on and before a
sweat moisture layer has formed between the wrist and the wrist strap band. Sometimes it
can be necessary to augment the moisture layer, for example by using a skin lotion.
Contamination of the footwear sole or floor surface can lead to high-resistance results
for footwear-flooring grounding. If necessary, clean these and retest. If liquid cleaners are
used, make sure they are fully dried before retesting.
Equipment required:
Procedure:
● Connect one side of the resistance meter to ground via a 0 Ω ground plug.
● Connect the other side of the meter to the earth bonding point under test.
● Record the result.
Common Problems
This measurement is usually straightforward.
These three electrode systems often do not give the same results. ESD packaging materials
can have highly variable surface resistance at different positions, and the electrodes respond
differently to these variations. The concentric ring electrodes tend to give the lower range
of resistance of the material around the circumference of the electrode (Smallwood 2017).
It is not directional in response.
The two-pin point-to-point electrode tends to give a result about a factor of four greater
than the concentric ring electrode, for a homogenous material of the same surface resis-
tivity. It also is highly influenced by any small areas of high-resistance material under the
point electrodes and is directional in response. So, this electrode gives greater variation in
results if the material is variable in resistance. It is more likely to indicate the upper range
of resistance of the material under the electrode points.
378 11 ESD Measurements
Figure 11.30 Measuring the surface resistance of an ESD protective packaging material. The
sample is placed on an insulating base material.
A point-to-point measurement made with closely spaced 2.5 kg electrodes can give similar
results to the concentric ring electrodes but has a directional response.
Equipment required:
● 10/100 V resistance meter with test leads
● Concentric ring electrode
● ESD protective packaging material to be tested
● Insulating support
Procedure:
● Place the sample packaging upon the insulating support.
● Place the concentric ring electrode on the packaging surface.
● Connect the resistance meter to the electrode.
● Measure the resistance at 10 V.
● If R > 1 MΩ, measure result again at 100 V.
● Note the result.
Common Problems
In practice, with high resistance, materials readings do not fully stabilize in a short time.
In this circumstance, the reading should be taken after an appropriate electrification time,
e.g. 15 seconds after application of the test voltage.
11.8 Standard Measurements Specified by IEC 61340-5-1 and ANSI/ESD S20.20 379
Figure 11.31 Measurement of surface resistance of small or curved ESD packaging using a
two-pin probe.
The electrode needs a flat surface larger than the electrode. For profiled or curved surfaces
and small items, the surface resistance is not correctly measured as the electrode contact
area is reduced.
Equipment required:
● Two-pin probe electrode with test leads
● 10/100 V resistance meter
● Insulating support
● ESD protective packaging to be tested
Procedure:
● Connect resistance meter to the electrode.
● Place the sample packaging upon the insulating support.
● Place the electrode in contact with the packaging surface.
● Measure result at 10 V.
● If R > 1 MΩ, measure the result again at 100 V.
380 11 ESD Measurements
Common Problems
ESD protective packaging materials can have considerable variation in their surface resis-
tance at different positions. This measurement is not directly equivalent to the concentric
ring electrode or point-to-point 2.5 kg resistance measurement electrode methods and will
give different results. This two-point electrode tends to give results showing the resistance
of highest resistance of the material under the pins (Smallwood 2017, 2018). As such, the
results may be highly variable compared to results found with the concentric ring electrode.
This may be an advantage or a disadvantage depending on the purpose of the test.
For a homogenous material, the two-point probe may give results about 2.5–5 times
higher than concentric ring electrodes.
Figure 11.32 Measurement of surface resistance ESD packaging using two 2.5 kg solid resistance
measurement electrodes.
11.8 Standard Measurements Specified by IEC 61340-5-1 and ANSI/ESD S20.20 381
The electrodes should be placed close together without risking touching each other to
give results similar to the concentric ring electrode.
Equipment required:
Procedure:
Common Problems
This measurement is not directly equivalent to the concentric ring electrode or two-point
probe methods and will give different results. The results vary with the distance between
the electrodes. This arrangement is most likely to give results similar to the concentric ring
electrodes when the two electrodes are separated by only a few millimeters.
Equipment required:
Procedure:
Figure 11.33 Volume resistance measurement using a concentric ring electrode (left). The
electrodes are shown on the right. This concentric ring electrode requires a weight to give the
correct total mass.
● Connect the resistance meter to the electrodes. The applied voltage is normally connected
to the metal plate electrode, and the current sense terminal is connected to the inner ring
electrode. The outer ring electrode is optionally connected to the resistance meter guard
terminal, if available.
● Measure the resistance at 10 V.
● If R > 1 MΩ, measure result again at 100 V.
● Note the result.
Common Problems
In practice, with high resistance materials, the readings do not fully stabilize in a short time.
In this circumstance, the reading should be taken after an appropriate electrification time,
e.g. 15 seconds, after application of the test voltage.
ESD generator
1kV HBM
Electrode
500 Ω
Electrode
Oscilloscope
standard HBM ESD generator. The applied waveform is similar in specification to that used
to test ESD withstand voltage of components.
When an HBM ESD is applied to the upper electrode, a small impulse is detected by the
capacitive electrode. A transient current flow through the resistor and current sensor and is
recorded by the digitizing oscilloscope. The energy, W, in the detected transient is calculated
from the digitized current samples I over n samples covering the duration the waveform.
n
∑
W = 500 I 2 dt
0
Start
Yes
Is inner surface
insulative?
No
Not acceptable as
No
ESD protective packaging
Yes
Is there a high resistance
barrier material or air gap
between outer and inner
surfaces?
Acceptable as ESD
shielding packaging No
Not acceptable as
ESD shielding packaging
Figure 11.35 Flowchart procedure for evaluation of ESD shielding properties of packaging.
Ionizer
CPM measurement unit
CPM plate
Figure 11.36 Measuring the decay time of an ionizer. The CPM plate is positioned simulating the
normal position of items requiring neutralization.
11.8 Standard Measurements Specified by IEC 61340-5-1 and ANSI/ESD S20.20 385
100 1200
0 1100
–5 0 5 10 15 20 25 1000
–100
900
CPM voltage (V)
Standard tests require an initial CPM plate voltage of over 1000 V and measures the decay
time from 1000 V to a final voltage of 100 V (Figure 11.37). The offset voltage is measured
after the plate has reached a reasonably constant voltage sometime after the decay curve.
Some fluctuation of the offset voltage is normal.
Equipment required:
● Charge plate monitor
● Ionizer under test
Procedure:
● Set up the ionizer in its intended working position.
● Set up the charge plate monitor in a position typical of items requiring neutralization.
● Charge the plate and observe the decay of voltage.
● Measure the decay time between the initial voltage and the final voltage required by the
test.
● Note the residual “offset” voltage after it has stabilized.
Equipment required:
● CPM or body voltage measurement instrument
● Handheld electrode and long wire
● Computer or recording equipment (optional)
Procedure:
● Set up the CPM in a convenient position.
● Connect the long wire to the hand-held electrode and CPM plate.
● Ask the subject to hold the electrode and walk around. (Note: ESD STM97.1 requires a
specific step pattern to be used.)
● Monitor the voltage readings and record the “peaks” of positive or negative polarity.
● Calculate the average of the five highest peaks (if required).
Figure 11.38 Measurement of body voltage of personnel (left) using electrostatic voltmeter
instrument and (right) using CPM.
11.9 Useful Measurements Not Specified by IEC 61340-5-1 and ESD S20.20 387
Equipment:
● Electrostatic field meter
● Grounding wire (if required)
● Workstation position to be evaluated
Figure 11.39 Measurement of electrostatic fields in the region where an ESDS device may be
present. The field meter is held at different positions and orientations to look for electrostatic field
sources and measure the strength of fields.
388 11 ESD Measurements
Procedure:
● Ground the electrostatic field meter. (Many field meters can be grounded simply by being
held in the hand of a grounded person.)
● Move the field meter around the region in which the ESDS may be situated and monitor
the readings.
● Note any high field readings and the positions of their sources.
When high field readings are noted, the field meter can be used to home in on the source
for further evaluation.
Calibration
distance Voltage (V) = electrostatic field (E) x distance (d)
Electrostatic
field meter
Charged
surface
voltage V
Figure 11.40 Measuring electrostatic voltages on a planar target surface using an electrostatic
field meter calibrated as a voltmeter.
Figure 11.41 Using a field meter calibrated as a voltmeter to measure electrostatic surface
voltages on a charged object.
390 11 ESD Measurements
Equipment:
● Electrostatic voltmeter
● Grounding wire (if required)
● Objects or material to be tested
Procedure:
● Ground the electrostatic voltmeter. (Many meters can be grounded simply by being held
in the hand of a grounded person.)
● Hold the voltmeter the correct calibration distance from the surface to be measured.
● Take a reading.
Common Problems
The electrostatic voltmeter must be grounded, or it will not read the target voltage correctly.
Any voltage on the meter is added to the voltage read on the target surface. The reading will
also change if not held at the correct distance from the surface being measured. If too close,
the reading will be too high.
Do not hold a sample in your hand for measurements – if it conducts electricity, any
charge would be conducted to your body. If your body is not grounded, any voltage on your
body could appear also on the sample.
The voltage on small objects, curved objects, insulators, or small isolated conductors is
not correctly measured. The voltage on insulators and small isolated conductors changes
with increased capacitance due to the presence of the meter.
Equipment:
● Noncontact electrostatic voltmeter
● Grounding wire (if required)
● Objects or material to be tested
Procedure:
● Ground the electrostatic voltmeter. (Many meters can be grounded simply by being held
in the hand of a grounded person.)
● Hold the voltmeter within the correct calibration distance range from the surface to be
measured.
● Note the reading.
11.9 Useful Measurements Not Specified by IEC 61340-5-1 and ESD S20.20 391
Figure 11.42 Measurement of voltage of a small item using a noncontact electrostatic voltmeter.
Figure 11.43 Measurement of voltage of a small item using a contact electrostatic voltmeter.
Source: DE Swenson.
392 11 ESD Measurements
Equipment:
● Contact electrostatic voltmeter
● Grounding wire (if required)
● Objects or material to be tested
Procedure:
● Ground the electrostatic voltmeter. (Many meters can be grounded simply by being held
in the hand of a grounded person.)
● Touch the voltmeter tip to the surface to be measured.
● Note the reading.
Equipment required:
● 10/100 V high-resistance meter
● Test leads and clips
● Tool to be tested
Procedure:
● Connect the test leads and clips to the resistance meter.
● Connect one test lead to the tool bit.
● Connect the second test lead to the tool handle.
● Measure result at 10 V.
● If the resistance is >1 MΩ, measure result again at 100 V.
Common Problems
One problem with this test is that it can be difficult to make good electrical contact with
a handle made from hard high-resistance material. Contact can be improved by wrapping
a conductive self-adhesive metal tape around the handle to provide a larger area contact.
Measuring the tool as held in the hand gets around this problem.
Equipment required:
● 10/100 V resistance meter
● Wrist strap or footwear-flooring grounding for the test person
● Tool test electrode
Procedure:
● Connect one terminal of the resistance meter to ground.
● Connect the second terminal of the resistance meter to the tool test electrode.
● Wear the grounded wrist strap and hold the tool.
394 11 ESD Measurements
Figure 11.45 Test of resistance to ground of hand-held tool. The person holding the tool can be
grounded in their usual manner.
Common Problems
If the user’s hand touches the metal tool bit, the resistance of the handle is bypassed by the
resistance of the user’s hand, and a false low resistance reading is obtained.
Equipment required:
● Low-resistance meter
● Test leads and clips
Procedure:
● Connect one terminal of the resistance meter to the soldering iron groundable point.
● Connect the second terminal of the resistance meter to the soldering iron tip.
● Measure the resistance result.
11.9 Useful Measurements Not Specified by IEC 61340-5-1 and ESD S20.20 395
Figure 11.46 Measurement of the resistance to groundable point of a soldering iron bit.
Common Problems
Corrosion of the bit surface contact with the iron can give variable results in this test. But,
detection of this is one of the reasons for doing the test!
Equipment required:
● Low-resistance meter
● Test leads and 0 Ω earthing connector
● Tool test electrode
Procedure:
Common Problems
Corrosion of the bit contact with the iron can give variable results in this test – but then
detecting this part of the reason for doing the test!
396 11 ESD Measurements
Figure 11.47 Measurement of resistance to ground of a soldering iron tip while in use.
Common Problems
Typical wrist strap testers have a fixed upper “fail” limit, often 35 MΩ. This upper limit
must correspond to the upper limit specified in the ESD control program for the resistance
11.9 Useful Measurements Not Specified by IEC 61340-5-1 and ESD S20.20 397
Figure 11.48 Handheld (left) and CAFE (right) electrodes. Source: D E Swenson.
Figure 11.49 Measurement of resistance through glove to ground using CAFE electrode. Source: D
E Swenson.
398 11 ESD Measurements
Figure 11.51 Testing of system resistance through a glove and wrist strap using a wrist strap
tester. Source: D E Swenson.
11.9 Useful Measurements Not Specified by IEC 61340-5-1 and ESD S20.20 399
Figure 11.52 Measurement of resistance to ground through a glove using a handheld electrode.
of glove when worn. If the upper limit accepted by the ESD program for the glove is higher
than the “fail” threshold of the tester, gloves having resistance above the fail threshold are
“failed” incorrectly. If the upper limit accepted by the ESD program for the gloves is lower
than the “fail” threshold of the tester, gloves having resistance above the acceptable limit of
the ESD program, but below the “fail” threshold of the tester, are “passed” incorrectly.
Initial voltage Vi
Voltage
curve
Decay time
Decayed
voltage Vf
0 Time t
zero. This typically gives a quasi-exponential voltage curve (Figure 11.53). The time taken
for the voltage to reach a defined level V f , often chosen to be 100 V, is measured.
Slow decay curves can sometimes be monitored manually, but faster decay curves may
need a digitizing oscilloscope or chart recorder to aid measurement. The time taken for
the voltage to decay from the chosen initial voltage (V i in Figure 11.53) to the chosen final
voltage V f , is taken as the decay time.
The user should select the final voltage as appropriate for their purpose. Often a fraction
of the initial voltage (e.g. 0.1 Vi ) or a “hazard threshold voltage” (e.g. 100 V) is specified.
Typically, the charge decay time result will depend on the initial test voltage, perhaps
varying in some cases by orders of magnitude. Longer decay times are usually found for
lower test voltages.
A commercial CPM can be conveniently used for charge decay tests. This is often limited
in test voltage to the 1000 V and final voltage of 100 V designed for use with ionizer testing.
Many charge decay test methods are in effect indirect ways of comparing the resistance
to ground of the item measured. The resistance to ground of the item bleeds the charge
from the plate capacitance Cp . When touched by a handheld tool, the resistance to ground
of the tool Rt is connected in parallel with the plate capacitance Cp . The voltage decays with
a time constant Rt Cp . The decay time measured from 1000 to 100 V is approximately 2 Rt
Cp . Where the resistance is very high, this technique can be more reproducible than using
a direct resistance measurement method.
Figure 11.54 Measurement of charge decay time of tools. The tool is touched to the CPM plate
and the voltage decay time is measured. Both the operator and CPM are grounded.
Equipment required:
● Charge plate monitor
● Wrist strap or footwear – flooring grounding of operator
● Sample tools for test
Procedure:
● The operator must be grounded.
● Charge the CPM to 1000 V.
● The operator holds the tool in their hand but does not touch the tool blade. The tool blade
is then touched to the CPM.
● Observe and time the decay to 100 V.
It is helpful if the decay curve is recorded, as this can reveal the true charge decay char-
acteristics of the tool and be used to document the test for product qualification records.
Some examples of typical waveforms are given in Figures 11.55–11.57. A tool that has a low
resistance will give a fast drop in voltage to zero (Figure 11.55). A tool with high-resistance
handle will show a longer, slower decay time (Figure 11.56). An initial fast drop may be
seen in this case due to increasing capacitance and charge sharing as the tool bit contacts
the CPM plate.
402 11 ESD Measurements
1200
Tool touches CPM plate
1000
CPM voltage (V)
800
600
400
200
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (seconds)
Figure 11.55 Decay curve from tool that has low-resistance handle.
1200
Tool touches CPM plate
1000
CPM voltage (V)
800
600
400
200
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (seconds)
Figure 11.56 Decay curve from tool that has intermediate-resistance handle.
1200
Tool touches CPM plate
1000
CPM voltage (V)
800
200
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (seconds)
Figure 11.57 Decay curve from tool that has insulative handle and shows capacitive voltage
reduction.
11.9 Useful Measurements Not Specified by IEC 61340-5-1 and ESD S20.20 403
If the tool handle is insulative, the plate voltage does not reduce to zero in an acceptable
time. Again, there may be a step in voltage as the tool bit contacts the plate as the charge
stored on the plate is shared with the tool capacitance (Figure 11.57).
Common Problems
The metal tool blade should not be touched with the fingers, as the resistance of the handle
will be bypassed.
Typical digital displays on CPM instruments may update only every half-second or so.
Short decay times of a second or less are difficult to measure without monitoring the plate
voltage on an oscilloscope type display.
Equipment required:
Procedure:
Figure 11.58 Measuring the charge decay of a glove when worn. The gloved finger is brought into
contact with the charged plate and the decay observed.
404 11 ESD Measurements
Figure 11.59 Charge decay system test of tool held in gloved hand.
Common Problems
Typical digital displays on CPM instruments may update only every half-second or so. Short
decay times of a second or less are difficult to measure without monitoring the plate voltage
on an oscilloscope type display.
When the gloved finger is removed from the CPM plate, the plate may become charged.
This may indicate unacceptable charging of the plate by the glove material.
Common Problems
Make sure the fingers do not accidentally contact the tool bit, short-circuiting the handle.
Figure 11.60 Unshielded Faraday pail on CPM plate. This pail was purchased in the kitchen
department of a local store!
Simple unshielded Faraday pails are prone to errors from induced charge due to fields
from nearby charged insulators, personnel, or voltage sources. This can be reduced by con-
taining the measurement pail isolated within an earthed metal screen (Figure 11.61).
Where a Faraday pail is used with a coulombmeter, the charge is measured directly by
the coulombmeter.
Where a Faraday pail is used with a CPM or electrostatic voltmeter, the charge gives a
voltage rise on the pail that is measured using the voltmeter or CPM. For comparative mea-
surements (e.g. comparing the charging of an item when handled by two different glove
types), it may be sufficient to compare the voltages produced on the pail. If the actual charge
values Q is required, it can be calculated, knowing the capacitance Cp of the pail in its mea-
surement arrangement and the voltage result V, from the relation.
Q = Cp V
When used with a voltmeter or CPM, a Faraday pail must usually be zeroed by momentary
connection to ground to bring the pail to zero volts. When used with a coulombmeter, the
coulombmeter must be zeroed before each measurement.
The principle of these tests is that a grounded person, wearing the glove or finger cots
under test, handles the item and then places them into a Faraday pail (see Section 11.9.9.1)
or on to a CPM (see Section 11.9.9.3).
If an unscreened Faraday pail is used, electrostatic fields from charged clothing, gloves,
or other insulating items in the vicinity can strongly affect the measurement result. In this
case, make sure that all potentially charged items are moved well away from the measure-
ment area and that the operator moves away before the measurement result is noted. If a
screened Faraday pail is used, this is much less affected by nearby electrostatic fields.
Equipment required:
● Faraday pail connected to a voltmeter or charge measurement instrument
● Wrist strap grounder
● Sample glove or finger cot for test
Procedure:
● Wear the wrist strap, and ground it.
● The subject holds the product in their hand and handles it in a representative way.
● If necessary, zero the Faraday pail and measurement instrument.
● Place the product in the Faraday pail.
● Move the hand well away from the Faraday pail.
● Note the charge reading.
11.9 Useful Measurements Not Specified by IEC 61340-5-1 and ESD S20.20 407
Common Problems
Touching the Faraday pail can cause errors due to charging or discharging of the pail by
contact.
An unshielded Faraday pail is prone to errors due to induced charge from nearby electro-
static field sources such as charged clothing, the user’s body, or insulators. Proximity to the
user’s hand, body, or other conductors can increase capacitance and reduce the readings
obtained.
Placing a charged ESDS in a metallic Faraday pail or on a metal CPM plate can cause
charged device ESD that could damage the ESDS. This can be prevented by lining the Fara-
day pail or CPM plate with static dissipative material having surface resistance >10 kΩ.
product on it. The voltage on the CPM after placing the product on it is indicative of the
charge on the product. Different gloves can be compared using repeated tests handling the
same item.
The charge induced on the CPM plate is an unknown fraction K of the charge on the item
measured. If the CPM plate capacitance Cp is known, the charge Q on the item measured
can be estimated from the CPM plate voltage V from the relation.
Q = KCp V
The unknown factor K must usually be assumed to be K = 1 unless it can be somehow
evaluated.
Common Problems
Touching the CPM plate can cause errors due to charging or discharging of the pail by
contact.
The CPM plate is prone to errors due to induced charge from nearby electrostatic field
sources such as charged insulators. Proximity to the user’s hand, body, or other conductors
can increase capacitance and reduce the readings obtained.
Placing a charged ESDS device on a metal CPM plate can cause charged device ESD that
could damage the ESDS. This can be prevented by lining the CPM plate with static dissipa-
tive material.
Equipment required:
● ESD event detector
Procedure:
● Switch on the ESD event detector and place it near the operation being investigated.
● Watch the response of the event detector during the process. Look for coincidence of an
ESD event with a contact made between a conductor and the ESDS.
Common Problems
ESD event detectors typically detect ESD from almost any source in the vicinity. These could
be from contactors or switches operating in equipment, room lighting being switched, or
other sources. Most of these sources are irrelevant to potential damage to the ESDS. It can
be difficult to distinguish between potentially damaging ESD events and irrelevant ESD
events. Where the ESDS can be observed through the process, look for coincidence between
an ESD event and a contact between the ESDS device and a conductor. If the ESDS cannot
References 409
be seen during the part of the process where ESD was noted, it may be necessary to review
the process steps while not in operation.
References
EOS/ESD Association Inc. (2014b) ANSI/ESD STM9.1-2014. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Footwear – Resistive Characterization.
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015a) ANSI/ESD STM3.1-2015. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Ionization. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2015b) ANSI/ESD STM11.11-2015. ESD Association Standard for
Protection of Electrostatic Discharge Susceptible Items – Surface Resistance Measurement of
Static Dissipative Planar Materials. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015c) ANSI/ESD STM11.12-2015. ESD Association Standard for
Protection of Electrostatic Discharge Susceptible Items. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015d) ANSI/ESD STM11.13-2015. ESD Association Standard Test
Method for the Protection of Electrostatic Discharge Susceptible Items – Two-Point Resistance
Measurement. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015e) ANSI/ESD S13.1-2015. Provides electrical soldering/
desoldering hand tool test methods for measuring current leakage, tip to ground reference point
resistance, and tip voltage. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2015f). ANSI/ESD STM97.1-2015. ESD Association Standard Test
Method for the Protection of Electrostatic Discharge Susceptible Items – Floor Materials and
Footwear – Resistance Measurement in Combination with a Person. Rome, NY, EOS/ESD
Association Inc.
EOS/ESD Association Inc. (2016a) ANSI/ESD SP3.3-2016. Standard Practice for the Protection
of Electrostatic Discharge Susceptible Items – Periodic Verification of Air Ionizers. Rome, NY,
EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016b) ANSI/ESD SP3.4-2016. Standard Practice for the Protection
of Electrostatic Discharge Susceptible Items – Periodic Verification of Air Ionizer Performance
Using a Small Test Fixture. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016c) ANSI/ESD STM97.2-2016. Standard Test Method for the
Protection of Electrostatic Discharge Susceptible Items – Floor Materials and Footwear – Voltage
Measurement in Combination with a Person. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2017) ANSI/ESD STM4.1-2017. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Worksurfaces – Resistance
Measurements. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2018a) ESD TR53-01-18. Technical Report for the Protection of
Electrostatic Discharge Susceptible Items – Compliance Verification of ESD Protective
Equipment and Materials. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2018b) ANSI/ESD STM11.31-2018. ESD Association Standard Test
Method for Evaluating the Performance of Electrostatic Discharge Shielding Materials – Bags.
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2018c) ANSI/ESD S541-2018. Packaging Materials for ESD Sensitive
Items. Rome, NY, EOS/ESD Association Inc.
European Committee for Electrotechnical Standardisation (CENELEC). (1992) EN 100015-1.
Basic specification. Protection of electrostatic sensitive devices. Harmonized system of quality
assessment for electronic components. Basic specification: protection of electrostatic sensitive
devices. General requirements. Brussels, CENELEC.
References 411
Further Reading
EOS/ESD Association Inc. (2012) ANSI/ESD STM4.2-2012. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – ESD Protective Worksurfaces – Charge
Dissipation Characteristics. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016) ESD TR20.20-2016. ESD Association Technical Report -
Handbook for the Development of an Electrostatic Discharge Control Program for the Protection
of Electronic Parts, Assemblies and Equipment. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (1999) ESD TR15.0-01-99. Standard Technical Report for the
Protection of Electrostatic Discharge Susceptible Items-ESD Glove and Finger Cots. Rome, NY,
EOS/ESD Association Inc.
EOS/ESD Association Inc. (2019) ANSI/ESD SP15.1-2019. ESD Association Standard Practice
for the Protection of Electrostatic Discharge Susceptible Items – In-Use Resistance Measurement
of Gloves and Finger Cots. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2019) ANSI/ESD STM9.2-2019. ESD Association Standard for the
Protection of Electrostatic Discharge Susceptible Items – Footwear – Foot Grounders Resistive
Characterization. Rome, NY, EOS/ESD Association Inc.
International Electrotechnical Commission. (2018) IEC TR 61340-5-2:2018. Electrostatics – Part
5-2: Protection of electronic devices from electrostatic phenomena - User guide. Geneva, IEC.
Smallwood, J. (2005). Standardisation of electrostatic test methods and electrostatic discharge
prevention measures for the world market. J. Electrostat. 63 (6–10): 501–508.
Vermillion R. (2016) Testing methods for ESD control packaging products. Controlled
Environments. Available from: https://www.cemag.us/article/2016/02/testing-methods-esd-
control-packaging-products [Accessed 6th June 2019].
413
12
ESD Training
has sufficient strength and is from a potentially damaging source. Only a small proportion
of the ESDS device handled will fail, and the failures may be identified during test at a later
stage.
Personnel rarely then get the firsthand feedback that loss of ESD control causes ESD fail-
ures. Static electricity and ESD are rarely noticeable by sight or sound, and neither can they
be felt unless at very high levels. Furthermore, static electricity is not always present in sig-
nificant levels – it appears and disappears as materials are handled and moved, and even
in response to the weather! There are many ways in which static charge is harmlessly dis-
sipated before it can cause damage – ESD may occur only when the many factors involved
conspire to prevent static charge dissipation.
If ESD damage occurs, it is usually detected at a test stage long after the damage was
caused. Even if it is recognized as an ESD failure, the action or control failure that caused
it is usually not obvious.
These factors tend to promote skepticism that ESD damage is a real issue and that ESD
control measures are necessary and can make a real difference.
ESD training often seems to be trying to convince trainees of a scarcely believable sce-
nario, and overselling this can be self-defeating (McAteer 1980). Nevertheless, the challenge
is to give personnel the understanding they need to know when, where, and how to use ESD
control equipment and procedures effectively. They understand why they are doing this and
believe that it is important. An ungrounded person is probably the greatest ESD source in
manual handling processes. A trained person correctly using ESD control equipment and
procedures and preventing noncompliances is arguably a most effective first line of defense
against ESD risks.
Personnel who have different roles often need different training levels or content to fulfill
those roles. The Training Plan should document the following:
Managers who need to enter EPAs may also need brief instructions on the use of foot
straps, garments, or other ESD control equipment they will use in the EPA, as well as on
any activities or actions they must avoid (e.g. touching ESDS). If they accompany visitors
into the EPA, they may need instruction on supervision of these visitors.
Trainers who need to develop and present effective ESD training may themselves need
training on the specific ESD control practices current in their organization. They may also
need sufficient understanding of ESD control principles and practice to answer trainee
questions during training sessions. They may need training on how to present effective
demonstrations relevant to electrostatics and the ESD control program.
Personnel who purchase ESD-related items may need an overview of ESD control
practices and their impact on product specification and procurement. If the ESD control
program complies with a standard, they may need a working knowledge of the standard
requirements for sourcing ESD control equipment.
Subcontractors who go into an EPA may need either specific training according to their
activities or instructions on any areas and activities they must avoid.
Visitors who go into an EPA should normally be accompanied by trained personnel to
ensure they do not do anything that might compromise ESD control. Nevertheless, they may
need brief instructions on the use of foot straps, garments, or other ESD control equipment
they will use in the EPA, as well as on any activities or actions they must avoid (e.g. touching
ESDS).
Skeptical engineers who do not believe in ESD control can be among the most resistant
and unreliable in using ESD control measures. They can also have a negative effect on
others around them, reducing their confidence in the need for ESD control. Conversely,
engineers who have good understanding of ESD issues and control measures can be a
great asset in implementing an ESD control program, helping to develop an effective
ESD control program and helping others understand the importance and how to use ESD
control equipment.
One way of summarizing the roles of personnel that need training, and the training they
will need, is to present them as a matrix (Table 12.1). This can help plan training courses
and their content, as well as who will receive them.
Training type
ESD Use of EPA Testing EPA Principles and practice ESD control
Personnel role awareness equipment equipment of ESD control for managers
√ √
Operators
√ √ √
Supervisors
√
Managers
√ √ √
Audit and test
personnel
√ √ √ √ √
ESD coordinators
12.4 Training Form and Content 417
● ESD awareness for manager, including the possible financial cost/benefit and other
impact of ESD damage and ESD control
● ESD measurement test procedures, for personnel who check ESD control equipment and
EPAs
● ESD control, safe working, and special procedures for personnel who work in
high-voltage or other areas with special safety issues
● Audit techniques and audit of ESD control procedures
● ESD coordinator training and knowledge development and standards update
● Cleaning regime cleaning materials and practice for personnel who clean within
the EPA
● Do’s and don’ts for visitors, and guidance for personnel who accompany visitors
● Instructions for contractors working in the EPA
● Instructions for facility maintenance activities in the EPA
teach. If the necessary expertise does not exist in-house, it should be sourced outside the
organization. This might include using an external trainer with appropriate expertise to run
a suitable course.
A commercial course prepared and presented by a third-party instructor can be a good
way of providing broad-based awareness training or general expertise such as ESD measure-
ments or ESD coordinator level training. A disadvantage of generic commercial training is
that it may not align well to the organization’s specific ESD control program, processes, and
facilities.
An in-house course prepared and presented by an expert instructor (in-house or external)
can be an excellent way of providing training that is closely aligned to the organization’s
ESD program and facilities and include details specific to them.
High-level training, e.g. for the ESD coordinator, may be available only as external
courses. Attendance of seminars, conferences, or symposia should be considered as a
means of updating knowledge of current standards, trends and ESD control techniques,
available equipment, know-how, and expertise.
The location and availability of this information and resources should be publicized so
that personnel can refer to it.
● Preparation
● Delivery
● Instructor-led demonstrations
● Hands-on learning
● Follow-up assessment and training
12.4 Training Form and Content 421
Many books have been written on the subject of creating and delivering successful presen-
tations. The best presentations use a mixture of verbal and visual elements. Points should
be presented in a logical sequence. Each point should lead to following points or build on
previous points. I have found that short videos, demonstrations, or audience participation
(discussions or activities) are all useful ways of keeping interest in a presentation high.
anything that might be a distraction or a barrier to communication. Try to make the presen-
tation as interesting and enjoyable as possible for the audience and yourself (Figure 12.1).
Eye contact, or an illusion of eye contact, is important, but maintaining unnatural eye
contact can make the recipient uncomfortable. In a small group, it may be possible and
important to make occasional eye contact with each attendee in a relaxed and random order.
In a larger group, there are usually a few attendees around the room who are obviously more
receptive and engaged than others. Keeping eye contact with some of these can give the
illusion of keeping eye contact with the whole audience. Keep eye contact moving around
the room to avoid making anyone feel uncomfortable! If the presenter feels self-conscious
about eye contact, looking at different points behind the audience can help give the illusion
of it.
The presenter should face the audience whenever possible and try to avoid turning their
back. This can be difficult when the display screen is behind the presenter, and it may be
necessary to point to parts of it. If possible, the presenter should avoid blocking the view of
the screen or demonstrations. Achieving this may require some thought into the arrange-
ment of the room and presenting area when setting up for the presentation. Before starting,
walk around the room and look at the presenter’s area from the point of view of the entire
audience, sitting at selected points, to look from attendee eye height.
Classes where learners can participate, discuss, and engage are more likely to be found
interesting and remembered by the audience. Demonstrations can be very valuable, espe-
cially if the audience participates in them. As learners best remember what they practice
frequently, it is valuable to have attendees practice the skills they will need to use. It is
best to immediately follow up this learning by practice in the workplace. Surprisingly, even
demonstrations that go wrong can entertain and make the course more memorable. As elec-
trostatic demonstrations can be fickle, I explain early on that at least one will probably go
12.4 Training Form and Content 423
wrong, but I don’t know which one. This creates a sense of interested anticipation! When
one does go wrong, its failure can often be used to reinforce some other learning point!
Learning is often facilitated by working in a small group and encouraging questions and
discussion. I find that attendee questions often bring up very interesting points specifically
relevant to the questioner. Questions indicate that the attendee is thinking about the course
material presented and trying, in their mind, to apply it in their work.
Unless a course is very short or the audience is large, I prefer to start by briefly introducing
myself and then asking each attendee to introduce themselves and their job role in a couple
of sentences. This helps “break the ice” and starts the attendees speaking and contribut-
ing. It also helps me understand my audience for later formulating answers to questions in
appropriate terms and level.
Early on in ESD awareness courses I ask the question, “Who here has experienced elec-
trostatic shocks in everyday life?” I then go on to ask typical circumstances they have felt
shocks and to explain that the shock is a form of electrostatic discharge. I encountered very
few people who claim not to have experienced an electrostatic shock. Getting the audience
to respond to this serves several useful purposes.
● Normally, everyone has experienced a static shock at some time, and everyone can
respond – this again helps establish the habit of audience participation.
● It generates interest.
● It shows they have themselves experienced ESD, and it is a matter of everyday experience.
● It gives me the opportunity to explain that to feel a shock the body voltage must be over
2000 V for most people, but many components can be damaged by ESD at lower body
voltages we do not feel shocks (this is later substantiated by demonstrating body voltage).
Specific points can often be illustrated by relevant anecdotes from real life. If an attendee
has an illustrative anecdote from their experience, this can be even better, bringing the
point into relief and immediate relevance to other attendees. Occasionally, attendees have
brought to the discussion some evidence of ESD damage in their experience – this can be
particularly valuable.
When preparing electronic slide presentations, the visual style, type face, color selection,
etc. should be selected for clarity. Transition effects should be used sparingly as they can
be distracting. Establishing a consistent slide style helps give a professional presentation
image. Be careful to make sure that text and information low down on a slide can be seen
be all the audience.
12.4.10 Conferences
Many of the organizations listed in Tables 12.2 and 12.3 organize conferences worldwide
that may include some ESD-related papers. Their current activities may be found via their
web sites.
The organizations listed in Table 12.2 are most likely to arrange specialist ESD
control–related conferences and publish papers in their proceedings.
Table 12.4 Some magazines, journals, and online resources that publish ESD-related articles.
Wang (2002), and Voldman (2004). There are several books on ESD control in the man-
ufacturing environment, e.g. Dangelmeyer (1999), McAteer (1990), Welker et al. (2006).
Unfortunately, few of these are up to date although their content may still be highly
relevant. Because of their age, they are usually not well aligned with current ESD control
standards. Welker et al. (2006) is notable in that it focuses on ESD control in the clean room
environment.
There are several online magazines that occasionally publish good-quality articles on
the subject (Table 12.4). Some academic journals that from time to time publish papers
on ESD-related topics are also listed here.
428 12 ESD Training
Voltage V = IR
CV = Q Water flow in
dQ/dt = I
C
Charge
Capacitance
generation R Stored water level
(current source) Resistance
Plug
Basin
Drain
Figure 12.2 Water analogy of a simple model of electrostatic charging that is easily understood
by many nontechnical people.
(analogous to an insulator blocking charge dissipation), even a small water flow in (such as
a dripping tap) produces a significant water level that can remain for a long time.
Finally, the height of water in a basin for a given charge depends on the shape of the basin.
A basin that has a smaller footprint will give a higher water level, and a basin with a larger
footprint will give a lower voltage level for a given amount of water. This is analogous to
the same charge giving higher voltage for smaller storage capacitance.
● The cost of failures tends to increase as the product goes through the production process
stages.
12.7 Electrostatic Demonstrations 431
● The most expensive failures are usually those that occur at the customer’s site.
● Cost/benefit indications of over 1:10 can be found for an effective ESD control program.
In a course including relevant attendees such as managers or QA personnel, I often ask
the hypothetical question – what is the likely approximate cost of a failure or unreliability at
the customer’s site for your product? This usually provokes some interesting discussion. In
many cases, the answer to this question alone makes it worth taking ESD control seriously.
Many people will not be convinced by accounts of the cost of ESD damage found in other
factories, especially if the sources are old. There is no doubt that if it is available or can be
determined, real current data on the cost of ESD damage at the attendee’s own facility will
be far more convincing. In the absence of this data, it can be as valuable to demonstrate cases
where component failure levels, although not analyzed and proven due to ESD, nevertheless
fell when ESD controls were improved.
attendees that it’s likely that at least one demonstration will not work, and I don’t know up
front which it will be. This helps to create a sense of interested attention and anticipation!
The most convincing and relevant demonstrations are often those made using materials
that are commonly found in a poorly controlled workplace. For this reason, I generally
avoid unusual props like balloons (I once did see balloons in an EPA workstation; I was
told it was the operator’s birthday!) or Van der Graaf generators. The following examples
give some demonstrations that I have used in my electrostatics training.
Figure 12.3 Demonstrating a charged insulator, in this case a slab of expanded polystyrene foam.
12.7 Electrostatic Demonstrations 433
Materials may need to be replaced from time to time as their charging and charge-holding
properties tend to diminish with handling. This is due to contamination with oils, salts, and
perspiration from the hands.
A selection of ESD control packaging can be useful. Low-charging materials such as pink
polythene should be kept in a package separately from the other demonstration materials.
This is because they tend to contaminate other materials with their antistat and can make
demonstrations of electrostatic charging rather unreliable.
It can be useful to have an electric hairdryer or hot air gun in the demonstration kit to dry
out materials under high-humidity conditions. This can then be used as a teaching point,
demonstrating how humidity affects electrostatic charging and activity.
An ESD detector can be useful for demonstrating that ESD occurs during experiments.
This should be selected to make a loud enough sound when ESD occurs, to be easily heard
by all trainees.
● The field meter is calibrated to read surface voltage but correctly does so only when taking
readings from a large flat surface a defined distance away.
● Electrostatic field drops rapidly with increasing distance from the source.
● The closer a field source is to the ESDS, the more concerned we will be about it.
● If an insulator is found to have greater voltage level than a defined risk level, the risk can
be reduced by keeping it a sufficient distance away from any ESDS. This is the basis of
the field and voltage limits given in the standards.
434 12 ESD Training
Figure 12.4 Using a clear plastic covered booklet to show how voltages appear and disappear
when the cover is opened and closed.
12.7 Electrostatic Demonstrations 435
Figure 12.5 Using a document holder to show how voltage appears when the document is pulled
out and disappears when it is returned.
I usually neglect two other factors that are present in these experiments, for the sake of
reducing possible confusion. First, presence of a conducting material nearby can reduce
the apparent voltage produced by a charge. This is a second reason the voltage disappears
when the paper is reinserted into the holder. Second, the paper is normally sufficiently
conducting that any charge on it can move to and from the body of the demonstrator while
they are performing the demonstration.
I have occasionally done these experiments while wearing highly insulating rubber
gloves. If this is done, it can often be shown that the paper is positively charged on
separation while the plastic is negatively charged.
12.7.7 Tribocharging
Tribocharging can be demonstrated with conductors or insulating materials. As many peo-
ple, even those more familiar with electrostatics, believe that conductors cannot generate
charge, it can be useful to demonstrate tribocharging of a conductor. One way of doing this
is to use a metal plate mounted on a highly insulating handle (Figure 12.6). The field meter
is positioned to show the voltage on the plate and display it.
A suitable charged plate monitor can also be used to do these experiments. The metal
plate, however, can be easier for trainees to relate to metal items and printed circuit board
(PCB) used in the workplace. The instructor can make the point that what happens to the
plate can be expected to happen to the conductors on a PCB isolated from ground.
In this arrangement, the plate can be tribocharged by rubbing with various materials or
other actions. I have typically used the following:
I have also used this arrangement to demonstrate the charging effect of processes, for
example use of a cooling spray can.
Under humid air conditions, the moisture layer condensing on the insulating handle can
be sufficient to allow slow discharge of the plate. In these conditions, a hair dryer or hot air
gun can be used to temporarily dry the insulator and prevent discharge. This can itself be a
436 12 ESD Training
useful teaching point, showing that the discharge of the plate across the insulator is due to
moisture on the surface that is removed by drying.
The conductor used to touch the plate and generate ESD does not have to be a ground wire
and does not have to be grounded. ESD can also be demonstrated by touching with a tool,
or even a grounded ESD tool held in the hand of a grounded instructor. Even experienced
trainees can sometimes be surprised that ESD can occur when a correctly grounded person
handling or working on ESDS using ESD tools. The explanation is of course that the ESD
can occur if the ESDS is itself charged. ESD can occur when a person touches an ESDS if
either the person is charged or the ESDS is charged, or both.
produced when the demonstrator touches the plate with the tool. The explanation is that
the demonstrator and the plate are at the same voltage and so ESD does not occur.
This experiment supports the following teaching points:
● ESD occurs when two conductors touch and there is a voltage difference between them.
● If there is no voltage difference between the conductors, no ESD occurs when they touch.
● Once an equipotential bonding connection is made, no ESD can occur between the con-
nected conductors. However, at the point that the connection is made, ESD is likely to
occur as the conductors will probably be at different voltages.
This experiment can lead into a discussion explaining grounding as a form of equipo-
tential bonding where all conductors are bonded to earth. It can also be explained that
in situations where grounding is not possible equipotential bonding can be used to con-
trol ESD sources. Current ESD control standards often consider equipotential bonding and
grounding to earth both to be acceptable forms of “grounding.”
Figure 12.9 Induction charging demonstrated using a charged plate monitor (left) voltage
induced on plate by nearby charged insulator, ready for ESD (middle). ESD occurs when charge
moves from plate on grounding; (right) plate is left charged to opposite polarity when insulator is
removed, ready for another discharge. Source: D. E. Swenson.
12.7 Electrostatic Demonstrations 439
● The charged insulator does not need to touch the conductor for the voltage to be induced.
● The induced voltage is negligible if the insulator is kept sufficiently far away.
● ESD occurs if the conductor is touched by another conductor or ground wire when at a
different voltage. After touching with the ground wire, the plate is now charged, although
at this point it has no voltage, until the field source is moved away.
● After the ground wire is removed and the insulator moved away, the plate can be seen to
be charged as it achieves a high opposite polarity voltage.
Figure 12.10 The “perpetual ESD generator.” The metal plate rises in voltage due to electrostatic
field changes. A conductor contacting the plate initiates ESD.
440 12 ESD Training
The effects shown in Sections 12.7.10 and 12.7.11 are often the main reason for control of
electrostatic fields and insulators in an EPA. Electrostatic fields set up the conditions under
which ESD is more likely to occur when the ESDS touches another conductor. The stronger
the field, the greater the concern and risk of ESD damage.
Figure 12.11 Demonstrating body voltage while walking, using a charged plate monitor. The
computer displays the body voltage waveform.
while the trainee is moving around. The body voltage is seen to reappear when the wrist
strap is disconnected.
This experiment supports the following teaching points:
● Body voltage of hundreds of volts are usually generated by ordinary actions such as walk-
ing in daily life in unprotected areas. The person has no idea these voltages are present
unless they are high enough to cause electrostatic shocks. Different footwear and flooring
combinations give different body voltage results.
● The body voltage is reduced to less than 100 V by wearing a wrist strap.
● If the wrist strap connection fails, the body voltage reappears. A continuous connection
is required to control body voltage.
442 12 ESD Training
If a similar experiment is done using ESD footwear and flooring, the body voltage is often
greater than when grounded by wrist strap. Using different combinations of footwear and
flooring, it can be demonstrated that body voltage typically is different for each combina-
tion. In a higher-level course, this can lead to a discussion of qualification of footwear and
flooring combinations with regard to body voltage generation, and the necessity of a walk
test. It can be demonstrated that body voltage is not usually well controlled if ESD control
footwear is used without ESD control flooring or the flooring used without ESD control
footwear.
Figure 12.12 Placing a field meter inside an ESD bag gives a means of showing differences in
electrostatic charge generation and electrostatic field shielding.
12.7 Electrostatic Demonstrations 443
A highly charged insulator can then be brought nearby. If the electrostatic field shielding
provided by the bag type is poor, then strong indication of field is shown by the field meter.
If the electrostatic field shielding is good, then little indication of field is shown.
It is more convincing if the field meter is then taken out of the bag, and the charged
insulator brought nearby to show that it is indeed charged and gives a strong electrostatic
field.
When the demonstration is done with pink polythene bags under dry (<30% rh) air con-
ditions, the bag often shows strong electrostatic charging and little electrostatic field shield-
ing. It can be explained to the trainees that this material relies heavily on atmospheric
moisture for its low-charging properties. Under dry conditions, its performance can be little
different from an ordinary polythene bag. Aging and loss of antistat can also give this loss
of performance.
Under humid air conditions (>50% rh) pink polythene gives little static charging and can
appear to give good electrostatic field shielding. After demonstrating this, the bag can be
dried using a hair drier or hot air gun. After drying, the experiment is repeated, and the bag
will often show strong electrostatic charging and little electrostatic field shielding.
Black polythene bags give little electrostatic charging and show good electrostatic field
shielding in this test and can be shown to be independent of humidity.
Metallized ESD shielding bags usually show little electrostatic charging in this demon-
stration. They often show a small electrostatic field appearing within the bag in the elec-
trostatic field shielding test. Heavily crumpled bags may show reduced electrostatic field
shielding.
This experiment supports the following teaching points:
● Different types of ESD packaging have different electrostatic charging and electrostatic
field shielding properties.
● Ordinary polythene bags generate strong charges and are transparent to electrostatic
fields.
● Pink polythene bags have variable properties. With sufficient humidity, they give low
electrostatic charging and may also give good field shielding. Under low humidity, or
when old, they may appear little different from ordinary polythene.
● Black polythene bags show low electrostatic charging and good electrostatic field shield-
ing.
● Metallized ESD shielding bags show low electrostatic charging and good electrostatic
field shielding, although a small residual electrostatic field often appears within the bag.
● Charge does not move around quickly on insulating materials and cannot move to the
ground wire.
● Insulators cannot be successfully grounded.
Figure 12.14 Demonstrating charge decay and offset voltage of an ionizer using an insulating
material.
can be seen to rise again to the offset voltage. It can then be explained any item within the
region around the ionizer becomes charged to the offset voltage. This could, if too high, cre-
ate ESD risks. The offset voltage may vary with age and condition of the ionizer, and regular
test and maintenance is necessary to be sure this is kept within acceptable levels.
In a second experiment a charged insulator can be moved around near the metal plate
(Figure 12.15). The metal plate can be seen to vary in induced voltage in response to this
Figure 12.15 Demonstrating inability of an ionizer to keep a metal plate at low voltage in the
presence of fast-changing electrostatic fields or charging.
446 12 ESD Training
insulator. It can be shown that an ionizer does not necessarily keep the voltage on an iso-
lated conductor to an acceptable level if the electrostatic field or charging conditions create
voltage faster than the ionizer can neutralize them.
This experiment supports the following teaching points:
● Ionizers take time to neutralize charge and voltage. It may be necessary to wait for volt-
ages to be reduced to an acceptable level.
● For many types of ionizer, the neutralized voltage is near zero but does not reach zero
due to the offset voltage.
● Ionizer offset charges items to the offset voltage.
● Neutralization does not necessarily maintain the voltage on an isolated conductor at a
low level in fast-changing conditions.
12.8 Evaluation
12.8.1 The Need for Evaluation
It is important to make sure that trainees have understood and remember key training
points. It is also important to ensure they can use key equipment correctly and, in some
cases such as personal grounding, test them correctly. This means that some form of test
(paper or practical) will be needed to cover key points. The IEC 61340-5-1 and ANSI/ESD
S20.20 standards require this as part of their ESD Training Plan (International Electrotech-
nical Commission 2016; EOS/ESD Association Inc 2014).
References
Amerasekera, A. and Duvvury, C. (2002). ESD in Silicon Integrated Circuits, 2e. Wiley. ISBN: 0
470 49871 8.
Dangelmeyer, T. (1999). ESD Program Management, 2e. Clewer. ISBN: 0-412-13671-6.
EOS/ESD Association Inc. (2014) ANSI/ESD S20.20-2014. ESD Association Standard for the
Development of an Electrostatic Discharge Control Program for – Protection of Electrical and
Electronic Parts, Assemblies and Equipment (excluding Electrically Initiated Explosive Devices).
Rome, NY, EOS/ESD Association Inc.
Helling, K. (1996). ESD protection measures – return on investment calculation and case study.
In: Proc. EOS/ESD Symp. EOS-18, 130–144. Rome, NY: EOS/ESD Association Inc.
INARTE (2017) Electrostatic Discharge Control Certification. Available from: https://inarte
.org/certifications/inarte-electrostatic-discharge-control-esd-certification [Accessed 7th
Dec. 2017]
International Electrotechnical Commission (2016) IEC 61340-5-1: 2016. Electrostatics – Part
5-1: Protection of electronic devices from electrostatic phenomena - General requirements.
Geneva, IEC.
IPC (2017) IPC Announces New ESD Control Certification Courses on IPC EDGE Courses
provide professional credentials for Certified ESD Trainers (CETs) and ESD Certified
Operators (ECOs). Available from: http://www.ipc.org/ContentPage.aspx?Pageid=IPC-
Announces-New-ESD-Control-Certification-Courses-on-EDGE [Accessed 7th Dec. 2017]
McAteer, O.J. (1980). An Effective ESD awareness training program. In: Proc. EOS/ESD Symp,
189–191. Rome, NY: EOS/ESD Association Inc.
McAteer, O. (1990). Electrostatic Discharge Control. San Francisco, CA: McGraw-Hill. ISBN:
0-07-044838-8.
Newberg C (2017) Certification. Available from: https://www.esda.org/certification [Accessed
5th Dec. 2017].
Snow, L. and Dangelmeyer, G.T. (1994). EOS-16 94-1 – 94-12. A successful ESD training
program. In: Proc. EOS/ESD Symp. Rome, NY: EOS/ESD Association Inc.
STAHA Association. Available from: http://www.staha.fi/index_files/STAHA_leaflet_2013.pdf
[Accessed 5th Dec. 2017].
Voldman, S.H. (2004). ESD Physics and Devices. Wiley. ISBN: 0-470-84753-0.
Wang, A.Z.H. (2002). On-Chip ESD Protection for Integrated Circuits. Klewer Academic Press.
Welker, R.W., Nagarajan, R., and Newberg, C. (2006). Contamination and ESD Control in
High-Technology Manufacturing. Wiley-Interscience, IEEE Press. ISBN-10: 0 471 41452 2,
ISBN-13: 978 0 471 41452 0.
448 12 ESD Training
Further Reading
13
The Future
The future, particularly in a fast-moving world like the electronics industry, has even more
capacity to surprise than the workings of electrostatics and electrostatic discharge (ESD).
Attempting to predict it might seem to be doomed to failure. To the engineers who first dis-
covered ESD damage, it must have been unexpected. Many of those who followed showed
disbelief, and skepticism remains rife in the industry to this day. Although the future can
never be accurately predicted, some general short- and medium-term trends seem likely
to continue. New technologies can of course disrupt the picture at any time. Nevertheless,
it’s likely that the future reader of this chapter may be surprised and even amused by the
difference between my predictions and the intervening reality!
Electronic components will continue to become smaller in their internal feature dimen-
sions, and this will tend to make them more susceptible to ESD damage. New device tech-
nologies will continue to be developed. Some of these will be more susceptible, and some
less susceptible to ESD.
New device packages and circuit assembly technologies will become available and more
widely used. Some may even replace current technologies, as through-hole components
have been widely replaced by surface-mounted technology today.
ESD control programs will continue to evolve, as will the standards that support them. As
low ESD withstand (very sensitive) components become more widely handled, ESD control
programs will need to develop to handle them, both in manual and automated processes.
I hope that ESD withstand data will become more widely published on component data
sheets or available to the ESD control professional on demand. Unfortunately, many com-
ponent manufacturers as yet seem unwilling to make this data easily available.
Knowledge and understanding will be more important – leading to greater need for ESD
training at high level and audit, as well as specific aspects such as ESD-related measure-
ments.
Some of these topics are discussed more fully in the following paragraphs.
4 kV
De facto
Target Level
Current Target Projected Target
Level Level
2 kV
1 kV
500 V
0V
1975 1980 1985 1990 1995 2000 2005 2010 2015 2020
Copyright © 2016 ESD Association
Reproduced Figure 1, from the Electrostatic Discharge (ESD) Technology Roadmap, May 2016, by
permission of The EOS/ESD Association, Inc.
1250 V
De facto
Target Level
1000 V
De facto
Target Levels
250 V
125 V
0V
1975 1980 1985 1990 1995 2000 2005 2010 2015 2020
From the mid-1990s, withstand voltages fell again as technology pressures made achiev-
ing high ESD withstand voltage increasingly difficult.
In the 2000s a group of semiconductor manufacturers joined together to recommend new
ESD target levels (Industry Council 2010a,b, 2011). The motivation was to set ESD protec-
tion requirements for the safe handling of ICs within electrostatic discharge protected areas
(EPAs), while responding to the increasing difficulty in achieving high ESD withstand volt-
age in the face of technology scaling and other technology changes in IC design. The result
was to recommend reducing target HBM ESD withstand voltage to 1000 V and machine
model (MM) ESD withstand voltage to 30 V (Industry Council 2011). This was followed by
a recommendation to reduce CDM ESD withstand voltage target level to 250 V (Industry
Council 2010a). They concluded that “basic” ESD controls were adequate for protection
of devices with ESD withstand voltage down to 500 V HBM. “Detailed” ESD controls are
required for handling devices with ESD withstand voltages less than 500 V HBM.
The Industry Council for ESD Target Levels recommended HBM ESD withstand target
levels are reduced to 1 kV. This is probably only a first step – it seems likely that the HBM
ESD withstand target level will be further reduced to 500 V, with a corresponding reduction
in CDM ESD withstand target level. CDM target levels have already been reduced to 250 V,
and the roadmap suggests that a target of 125 V CDM will be needed in the future. Some
high-performance devices already have ESD withstand voltages as low as 100 V HBM.
The 2016 roadmap also gives an expanded view of HBM and CDM ESD withstand trends
from 2010 until beyond 2020 (Figures 13.3–13.6). These show that as time goes on, a greater
proportion of ESD-sensitive (ESDS) devices will have ESD withstand voltages between
500 and 100 V HBM, and <200 V CDM (Figures 13.5 and 13.6). In Figures 13.3 and 13.4,
452 13 The Future
Current Target
Level
Projected Target
Level
≥2 kV
Basic Handling
(500 V)
500 V
S20.20 (100 V)
0V
Custom Controls
2010 2015 2020 Region (<100 V)
Reproduced Figure 2, from the Electrostatic Discharge (ESD) Technology Roadmap, May 2016, by
permission of The EOS/ESD Association, Inc.
Figure 13.3 Technology Roadmap expanded 2010 and beyond HBM ESD withstand projections.
Source: ESDA (2016c).
Current Target
Level Projected Target
Level
500 V
Process Specific
125 V Measures * (100 V)
Charge / Discharge
Measurements at
0V each process step**
(50 V)
2010 2015 2020
* - Include process specific measures to avoid charging or discharge
** - Include process specific measures to avoid charging and discharge
Figure 13.4 Technology Roadmap expanded 2010 and beyond CDM ESD withstand projections.
Source: ESDA (2016c).
13.2 ESD Withstand Voltage Trends 453
> 1000 V
500 – 1000 V
500 – 1000 V
< 500 V
< 500 V
2010–2015 By 2020
Reproduced Figure 3, from the Electrostatic Discharge (ESD) Technology Roadmap, May 2016, by
permission of The EOS/ESD Association, Inc.
250 – 500 V
125 – 250 V
125 – 250 V
< 125 V
< 125 V
2010–2015 By 2020
Reproduced Figure 7, from the Electrostatic Discharge (ESD) Technology Roadmap, May 2016, by
permission of The EOS/ESD Association, Inc.
the control levels achieved by an ESD control program compliant with ANSI/ESD S20.20
are also shown. Below these levels (100 V HBM and 200 V CDM), special “custom” and
process-specific ESD control measures are usually required.
Looking further than 2020, the roadmap suggests that the range of HBM ESD withstand
voltages may not change so much, but the distribution of devices within the range may
change, with a greater proportion falling in the 500–1000 V and 100–500 V withstand volt-
age ranges. Similarly, a greater proportion is likely to fall within the 250–500 V, 125–250 V,
and <125 V CDM withstand voltage ranges.
The drivers for these reductions in ESD withstand are rooted in technology changes. The
internal device dimensions are reducing into the 22 and 18 nm range, and this tends to
lead to inherently more sensitive internal circuitry. Increasing numbers of devices include
high-speed input/output (I/O) pins that will need to operate into the 10–30 Gbit range of
data rate. Radio frequency circuits are becoming ever more widespread. These device pins
often tolerate only minute capacitance added by ESD protection networks, before perfor-
mance is impaired. Increased device performance often comes at the expense of reducing
ESD withstand voltage.
The MM ESD withstand test is already being phased out, and future ESD withstand char-
acterizations will be normally in terms of HBM and CDM withstand voltages.
The CDM ESD susceptibility is also reduced with increasing package size. The roadmap
states that current packages of 3000 pins in land grid array (LGA) or ball grid array (BGA)
with high-speed I/O at 22 nm barely meet a CDM target of 125 V.
454 13 The Future
● A desire to optimize the ESD control program for best effectiveness and return on invest-
ment
● A desire to achieve very low risk of ESD damage (e.g. for high value or high reliability
product)
Even in the absence of these motivations, the continuing trends in reduction of device
ESD withstand voltages confirm a need for continuing improvement of ESD control proce-
dures and compliance in processes handling these devices.
The Industry Council (2011) commented that ESD Control Programs range across three
levels.
● Little or no ESD control
● Basic ESD control programs
● Detailed ESD control programs
It is likely that organizations will need to move toward better ESD controls as
time goes on, and more organizations will move into the “detailed ESD controls”
category.
Organizations that have little or no ESD control are likely to experience ESD losses even
with devices of 2 kV HBM or greater ESD withstand voltage. Without adequate personal
grounding, body voltage can easily exceed 2000 V. Even if simple ESD control measures may
be present, ESD training and compliance verification are often minimal or absent. It is no
coincidence that the key requirements of modern ESD control standards include documen-
tation of the ESD program, an ESD Control Product Qualification Plan, an ESD Training
Plan, and a Compliance Verification Plan. Omission of any of these means that the ESD
program is likely to be ineffective.
● Lack of an ESD control program document means that ESD control measures will be
poorly reproduced and key measures may become omitted.
● Lack of an ESD Control Product Qualification Plan can lead to purchase of equipment
that does not effectively fulfill its purpose in the long term.
● Lack of a Compliance Verification Plan means that ESD control equipment failures are
unlikely to be discovered and remedied.
● Lack of an ESD Training Plan means that personnel are likely to remain insufficiently
aware of the procedures and equipment required to be used in ESD control.
With the increasing proportion of components that have low ESD withstand voltage,
organizations that currently pay minimal attention to ESD control are likely to find that
increasing losses will be experienced. This will demonstrate the need for establishment of
an effective ESD control program having at least good basic control measures.
The techniques used to control ESD risks within the EPA are that
● All conductors especially personnel who might contact ESDS devices are equipotential
bonded, preferably to earth.
● Unnecessary insulators are removed from the vicinity of ESDS devices.
● Insulators that are necessary to the process or product are evaluated for ESD risk. Unac-
ceptable risks are reduced by some means to an acceptable level. Risks are normally
evaluated in terms of electrostatic fields and voltages.
For long-term success, it must include documentation and implementation of the ESD
program, an ESD Control Product Qualification Plan, an ESD Training Plan, and a Compli-
ance Verification Plan.
The current ESD control program standards IEC 61340-5-1:2016 and ANSI/ESD
S20.20-2014 are designed to protect devices down to 100 V HBM and 200 V CDM. ESD
from charged conductors is limited to 35 V.
A basic ESD control program well implemented can be very effective, providing any fail-
ures of ESD control equipment are quickly detected and remedied.
At the time of writing, many users have yet to adequately adopt adequate footwear and
flooring compliance verification procedures as per the standards. Use of walk tests is partic-
ularly important when handling low HBM ESD withstand voltage devices, as the objective
is to keep body voltage below a low level related to the HBM withstand voltage of the
devices handled. Studies have shown that body voltage generated is, for some combina-
tions of footwear and flooring, badly predicted by the resistance of the footwear and flooring
individually and even in combination.
Working with low ESD withstand devices may require some users adopt lower body volt-
age limits to reflect the ESD withstand voltage of the devices they handle. For example, an
upper limit for body voltage of 50 V might be selected when working with 50 V HBM ESD
withstand devices.
● Conductors that make contact with ESDS devices must be grounded if possible.
● Conductors that make contact with ESDS devices but cannot be grounded must have the
voltage difference between the conductor and ESDS limited to ±35 V.
458 13 The Future
These requirements address both one-pin and two-pin ESD threats. The charged device
ESD threat is also addressed by limitation of electrostatic field in the vicinity of the ESDS
device. Both types of ESD damage can also be prevented by preventing contact between the
ESDS devices and low-resistivity conductive items.
In practice, it is difficult to measure and verify the voltage difference between the ESDS
device and an ungrounded conductor. To do so normally requires measurement of the volt-
age on the ESDS, measurement of the voltage on the conductor, and then calculating the dif-
ference. These can be difficult measurements to make, especially in an operating machine
process, as the ESDS device and conductor may be small low-capacitance items. Special
electrostatic voltmeters may be required to make these measurements. So, it remains to be
seen how practical these requirements are in practice. The standards also do not yet define
in a measurable way what is meant by an ungrounded conductor.
13.3.7 “One-Pin” ESD Between the ESDS and Another Conductive Part
Over the last decade or so, there has been considerable interest in developing a methodol-
ogy for relating process capability to device CDM ESD withstand data. This is particularly of
interest in automated processes but also in understanding ESD risks from contact between
devices and metal objects (whether grounded or not). Some researchers have used existing
CDM withstand voltage as the parameter (e.g. Steinman 2010, 2012), but other researchers
have proposed that it is necessary to record peak current or other CDM waveform data,
rather than the source voltage, for comparison with real-world conditions (e.g. Tamminen
and Viheriäkoski 2007; Tamminen et al. 2017a,b). At the same time, it has proven difficult
to make in-process measurements for comparison with CDM waveform parameters. It’s
likely that this area will continue to be researched over the forthcoming decade especially
as device CDM withstand voltages are reduced. New measurement techniques or methodol-
ogy may be needed before success is achieved in this area (Tamminen et al. 2017a,b). Device
ESD withstand measurement practices may need to be modified to record the required data.
13.4 Standards 459
13.3.9 Optimization
Optimization of an ESD program in terms of return on investment has long been an aim for
some ESD control practitioners. Nevertheless, compliance with standards has often been
an overriding goal, particularly in organizations where understanding of ESD controls has
been at a lower level. As greater expertise is applied in successful ESD control, that expertise
will be also applicable to optimization. More organizations will be able to achieve improved
optimization of ESD control alongside compliance with the standards.
13.4 Standards
The core ESD control standards IEC 61340-5-1 and ANSI/ESD S20.20 will continue to be
refined and developed for the foreseeable future. At their current stage of maturity, the
changes are expected to be incremental rather than major. But, experience with using the
standards, the increasing use of low ESD withstand voltage devices, and developing ESD
control approaches will be reflected in rewording and changes to the detailed requirements
of the standards. As an example, the 61340-5-1:2007 standard set a limit of electrostatic
460 13 The Future
fields at 10000 V m−1 , and this was reduced to 5000 V m−1 in the 61340-5-1:2016 update. The
latter also added some requirements for dealing with ungrounded conductors that make
contact with ESDS devices. The ANSI/ESD S20.20 often leads the way in this development,
with a new version often appearing a year or so before the 61340-5-1. New versions of the
User Guides IEC 61340-5-2 and ESD TR20.20 usually accompany the new standard versions
(International Electrotechnical Commission 2018a; EOS/ESD Association Inc 2016b).
Alongside this, some of the related test methods and other standards undergo develop-
ment and change with experience of their usage. New test methods may emerge for specific
applications. Methodologies for process evaluation are likely to be improved and developed
and may emerge as standard practice or full standards.
● The 61340-5-1 and S20.20 documents could be developed to directly address automated
systems. Guidance for application to automated systems could be incorporated in the
61340-5-2 and ESD TR20.20 User Guide documents.
13.6 ESD-Related Measurements 461
References
EOS/ESD Association Inc (2010). ESD Association Electrostatic Discharge (ESD) Technology
roadmap. Revised March 2010 Available from: https://www.esda.org/assets/Uploads/docs/
2013ElectrostaticDischargeRoadmap.pdf (Accessed: 15th May 2017).
References 463
EOS/ESD Association Inc. (2013). ESD Association Electrostatic Discharge (ESD) Technology
roadmap. Revised March 2013 Available from: https://www.esda.org/assets/Uploads/docs/
2013ElectrostaticDischargeRoadmap.pdf (Accessed: 10th May 2017).
EOS/ESD Association Inc. (2014). ANSI/ESD S20.20-2014. ESD Association Standard for the
Development of an Electrostatic Discharge Control Program for – Protection of Electrical and
Electronic Parts, Assemblies and Equipment (excluding Electrically Initiated Explosive Devices).
Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016a) ANSI/ESD SP10.1-2016. Standard practice for protection of
Electrostatic Discharge Susceptible Items – Automated handling Equipment (AHE). Rome, NY,
EOS/ESD Association Inc.
EOS/ESD Association Inc. (2016b). ESD TR20.20-2016. ESD Association Technical Report -
Handbook for the Development of an Electrostatic Discharge Control Program for the Protection
of Electronic Parts, Assemblies and Equipment. Rome, NY, EOS/ESD Association Inc.
EOS/ESD Association Inc (2016c) ESD Association Electrostatic Discharge (ESD) Technology
roadmap. revised 2016. Available from: https://www.esda.org/assets/Uploads/docs/
2016ESDATechnologyRoadmap.pdf (Accessed: 10th May 2017).
Industry Council on ESD Target Levels (2010a) White paper 2: A case for lowering component
level CDM ESD specifications and requirements. Rev. 2.0. http://www.esdindustrycouncil
.org/ic/en/documents/6-white-paper-2-a-case-for-lowering-component-level-cdm-esd-
specifications-and-requirements (Accessed: 10th May 2017)
Industry Council on ESD Target Levels (2010b) White paper 3: System Level ESD Part I:
Common Misconceptions and Recommended Basic Approaches. Rev. 1.0 http://www
.esdindustrycouncil.org/ic/en/documents/7-white-paper-3-system-level-esd-part-i-
common-misconceptions-and-recommended-basic-approaches (Accessed: 10th May 2017).
Industry Council on ESD Target Levels (2011) White paper 1: A case for lowering component
level HBM/MM ESD specifications and requirements. Rev. 3.0. Available from: http://www
.esdindustrycouncil.org/ic/en/documents/37-white-paper-1-a-case-for-lowering-
component-level-hbm-mm-esd-specifications-and-requirements-pdf (Accessed: 10th May
2017).
Industry Council on ESD Target Levels (2012) White paper 3: System Level ESD Part II:
Implementation of Effective ESD Robust Designs. Rev. 1.0 http://www.esdindustrycouncil
.org/ic/en/documents/36-white-paper-3-system-level-esd-part-ii-effective-esd-robust-
designs (Accessed: 10th May 2017).
International Electrotechnical Commission (2016) IEC 61340-5-1: 2016. Electrostatics – Part
5–1: Protection of electronic devices from electrostatic phenomena - General requirements.
Geneva, IEC.
International Electrotechnical Commission. (2018a) IEC TR 61340-5-2:2018.
Electrostatics – Part 5-2: Protection of electronic devices from electrostatic phenomena - User
guide. ISBN 978-2-8322-5445-5 Geneva, IEC.
International Electrotechnical Commission. (2018b) IEC TR 61340-5-5:2018. Electrostatics -
Part 5-5: Protection of electronic devices from electrostatic phenomena – Packaging systems
used in electronic manufacturing. Geneva, IEC.
Jacob, P., Gärtner, R., Gieser, H. et al. (2012). Paper 3B.8. ESD risk evaluation of automated
semiconductor process equipment – A new guideline of the German ESD Forum e.V. In:
Proc. EOS/ESD Symp. EOS-34. Rome, NY: EOS/ESD Association Inc.
Marathe, S., Wei, P., Ze, S. et al. (2017). Paper 3A.4. Scenarios of ESD Discharges to USB
Connectors. In: Proc. EOS/ESD Symp. EOS-39. Rome, NY: EOS/ESD Association Inc.
464 13 The Future
Olney, A., Gifford, B., Guravage, J., and Righter, A. (2003). Real-world charged board model
(CBM) failures. In: Proc. EOS/ESD Symp. EOS-25, 34–43. Rome, NY: EOS/ESD Association
Inc.
Paasi, J., Salmela, H., and Smallwood, J.M. (2003). New methods of assessment of ESD threats
to electronic components. In: Proc EOS/ESD Symp. EOS-25, 151–160. Rome, NY: EOS/ESD
Association Inc.
Smallwood, J. and Paasi, J. (2003). Assessment of ESD threats to electronic components and
ESD control requirements. In: Proc. Electrostatics 2003. Inst. Phys. Conf. Ser. No. 178 Section 6,
247–252.
Stadler, W., Niemesheim, J., and Stadler, A. (2017). Paper 3A.1. Risk assessment of cable
discharge events. In: Proc. EOS/ESD Symp. EOS-39. Rome, NY: EOS/ESD Association Inc.
Steinman, A. (2010). Paper 3B3. Measurements to Establish Process ESD Compatibility. In:
Proc. EOS/ESD Symp. EOS-32. Rome, NY: EOS/ESD Association Inc.
Steinman, A. (2012). Paper 2B.4. Process ESD Capability Measurements. In: Proc. EOS/ESD
Symp. EOS-34. Rome, NY: EOS/ESD Association Inc.
Tamminen, P. and Viheriäkoski, T. (2007). Paper 3B.3. Characterization of ESD risks in an
assembly process by using component-level CDM withstand voltage. In: Proc. EOS/ESD
Symp. EOS-29, 202–211. Rome, NY: EOS/ESD Association Inc.
Tamminen, P., Smallwood, J., and Stadler, W. (2017a). Paper 1B.4. Charged device discharge
measurement methods in electronics manufacturing. In: Proc. EOS/ESD Symp. EOS-39.
Rome, NY: EOS/ESD Association Inc.
Tamminen, P., Smallwood, J., and Stadler, W. (2017b). Paper 4B.2. The main parameters
affecting charged device discharge waveforms in a CDM qualification and manufacturing.
In: Proc. EOS/ESD Symp. EOS-39. Rome, NY: EOS/ESD Association Inc.
Viheriäkoski, T., Kärjä, E., Gärtner, R., and Tamminen, T. (2017). Paper 4B.3. Electrostatic
discharge characteristics of conductive polymers. In: Proc. EOS/ESD Symp. EOS-39. Rome,
NY: EOS/ESD Association Inc.
Further Reading
Fung, R., Wong, R., Tsan, J., and Batra, J. (2017). Paper 1B.3. An ESD case study with high
speed interface in electronics manufacturing and its future challenge. In: Proc. EOS/ESD
Symp. EOS-39. Rome, NY: EOS/ESD Association Inc.
Koh, L.H., Goh, Y.H., and Wong, W.F. (2017). Paper 1B.2. ESD Risk Assessment Considerations
for Automated Handling Equipment. In: Proc. EOS/ESD Symp. EOS-39. Rome, NY:
EOS/ESD Association Inc.
465
In some cases, operators will enter the EPA and work for a short time handling unpro-
tected ESDS devices while standing. In this case, personal grounding will be done most
conveniently by wearing ESD control footwear and standing on an ESD control floor. In
other cases, operators will sit at the workstation for some time handling unprotected ESDS
devices, and a wrist strap will be required. ESD control seats will also be grounded using an
ESD control floor.
A typical EPA in the imagined facility includes
● An ESD workstation with a surface on which unprotected ESDS devices may be placed
● An ESD control floor mat for grounding standing personnel (through footwear) and ESD
control chair
● An ESD control chair
● Wrist straps for grounding seated personnel, and earth bonding points of connection of
straps
● Containers such as tote boxes
● ESD control garments (coats)
The ESD control Program Plan contains an ESD Control Product Qualification Plan and a
Compliance Verification Plan. These require several qualification procedures (QPs) and test
procedures (TPs). These are not given here but would be defined and documented including
the test equipment to be used, details of the test method, and a simple test procedure.
A.4.1 Introduction
This document provides an ESD control program for XXX Ltd. in compliance with IEC
61340-5-1:2016 and ANSI/ESD S20.20-2014.
A.4.2 Scope
This ESD program covers all activities that manufacture, process, assemble, install, pack-
age, label, service, test, inspect, transport, or otherwise handle ESDS devices at XXX Ltd.
Specifically, these activities include receipt of ESD-susceptible (ESDS) PCBs and assemblies
at Goods In, removal from their ESD protective packaging for inspection, and individual
repacking in bags for dispatch.
The ESD withstand voltages of the individual types of ESDS devices are not known. The
ESD program is designed to be compliant with the current standards and capable of han-
dling electrical or electronic parts, assemblies, and equipment with ESD withstand voltages
greater than or equal to 100 V human body model (HBM), 200 V charged device model
(CDM) and ±35 V for isolated conductors.
A.5 Personal Safety 467
CDM Charged device model. ESD stress model that approximates the
discharge event that occurs when a charged component is quickly
discharged to another object at a different electrostatic potential.
EPA ESD protected area. Area in which an ESDS can be handled with
accepted risk of damage as a result of electrostatic discharge or
fields.
ESDS ESD-susceptible device. A sensitive device, integrated circuit, or
assembly that may be damaged by electrostatic fields or
electrostatic discharge.
HBM Human body model. ESD stress model that approximates the
discharge from the fingertip of a typical human being onto a pin of
a device with another pin grounded.
Shall “Shall” indicates a requirement of the ESD control program. If it is
stated that some aspect shall be done, and if it is not, this
constitutes a noncompliance.
Tailoring Modification of the requirements of the standard after evaluation
of the applicability of each requirement. Requirements may be
added, modified, or deleted. Tailoring decisions, including
rationale, and technical justification, shall be documented.
Static dissipative A packaging material or item that has surface resistance <100 GΩ
and >10 kΩ as per IEC 61340-5-3.
Conductive A packaging material or item that has surface resistance <10 kΩ as
per IEC 61340-5-3.
Insulator A materials or item that has surface resistance >100 GΩ as per IEC
61340-5-3.
Conductor Item or material that has surface resistance <100 GΩ.
ESD shielding packaging ESD protective packaging compliant with IEC 61340-5-3 that has
static dissipative inner and outer surfaces and a barrier layer or air
gap to prevent ESD attenuate ESD current passing from the outside
to the inside of the package.
Compliance with local safety laws, regulations, and requirements shall take precedence
over ESD control requirements.
Where use of personal protective equipment (PPE) is required, this shall be specified
in line with safety requirements and, where possible, also in compliance with ESD con-
trol requirements. PPE specified for use within this ESD control program shall be specified
within this ESD Control Program Plan.
To avoid electric shock risk, personnel who might come into accidental contact with high
voltages shall be protected by control measures that limit the current flow under fault con-
ditions to less than 0.5 mA. To limit current and provide protection against electric shock
468 Appendix A: An Example Draft ESD Control Program
in case of accidental contact of the wearer with 250 VAC electric power systems, the min-
imum resistance from the wearer’s body tested to groundable point (wrist strap) or metal
plate (footwear as worn) shall be 750 kΩ (see Section A.7.2 of this plan).
Wrist strap and cord TP1 >750 kΩ <35 MΩ Each day before first use
when worn, wearer’s
body to cord end
groundable point
Footwear when worn, TP2 >750 kΩ <100 MΩ Each day before first use
tested between wearer’s
body and one foot placed
on metal plate
Temporary wrist strap Visual inspection Correctly connected Each day before first use
bonding point
Wrist strap bonding point TP3 <1 MΩ Monthly
Personnel handling ESDS when seated shall be grounded using a wrist strap. Wrist straps
shall make good direct contact with the wearer’s skin.
The personal grounding equipment requirements and test criteria are listed in Table A.1.
A.7.3.2 Insulators
The following items are considered likely to be made from insulators unless tested and
approved for use within an EPA:
All items made of insulators shall be excluded from EPAs unless they have been desig-
nated essential to the process or operations.
Principles and
ESD ESD audit and practice of
Personnel awareness measurements EPA cleaning ESD control
● Cleaning materials and techniques for the EPA floor and bench mats
Trainees shall be tested by questionnaire and observation of practice.
Unless otherwise specified, all ESD control materials, equipment used in EPAs, and ESD
protective packaging selected shall as a minimum have a data sheet or other document
showing compliance with IEC 61340-5-1 or ANSI/ESD S20.20 and requirements of this ESD
Program Plan.
Table A.4 ESD control product qualification tests and pass criteria.
ESD control footwear, QP1. Resistance from person to Resistance from the wearer’s body to
while worn ground while standing on ESD ground >100 kΩ and <1 GΩ
floor
QP2. Body voltage walk test of No peaks in body voltage greater
footwear and flooring system than 100 V
Wrist strap TP1 >750 kΩ
<35 MΩ
Floor mat QP3. Resistance to groundable <1 GΩ
point of work surfaces or mats
Chair or seat TP4. Measurement of resistance <1 GΩ
to ground from surface
Work surface on which QP3. Resistance to groundable <1 GΩ
ESDS may be placed point of work surfaces or mats
QP4. Point-to-point resistance of <1 GΩ
work surface
ESD control garment TP5. Point-to-point resistance of < 1 GΩ
ESD garments
ESD Protective TP8 Surface resistance < 100 GΩ
packaging
References 475
The ESD coordinator shall maintain a list of items and equipment approved for use in
EPAs. Listed items shall be identified by manufacturer, type, or other specific information.
Further product qualification requirements and procedures for specific ESD control items
are given in Table A.4.
Proprietary wrist strap and footwear checkers used for daily checking of wrist straps and
footwear shall be verified to have pass criteria as specified in Table A.1.
References
EOS/ESD Association Inc. (2014) ANSI/ESD S20.20-2014. ESD Association Standard for the
Development of an Electrostatic Discharge Control Program for – Protection of Electrical and
Electronic Parts, Assemblies and Equipment (excluding Electrically Initiated Explosive Devices).
Rome, NY, EOS/ESD Association Inc.
International Electrotechnical Commission. (2015) IEC 61340-5-3:2015. Electrostatics.
Protection of electronic devices from electrostatic phenomena. Properties and requirements
classifications for packaging intended for electrostatic discharge sensitive devices. Geneva, IEC.
International Electrotechnical Commission (2016) IEC 61340-5-1: 2016. Electrostatics – Part
5-1: Protection of electronic devices from electrostatic phenomena - General requirements.
Geneva, IEC.
International Electrotechnical Commission. (2019) IEC TR 61340-5-4:2019. Electrostatics - Part
5-4: Protection of electronic devices from electrostatic phenomena – Compliance Verification.
Geneva, IEC.
477
Index
gloves and finger cots, see gloves and garment 170, 203, 227–229, 329,
finger cots 366–367, 474
how to determine 321–323 gloves and finger cots 233–234
how to specify 334–338 ionizer 171, 220–221, 383
insulators, see insulators measurements, see measurements
ionizer, see ionizer packaging 262
personal grounding, see personal pass criteria, see pass criteria
grounding seat 215–216, 364–365, 474
rack, see ESD control, rack soldering irons 172, 231
return on investment 299, 305–306 storage racks and shelves 170, 209,
seating, see ESD control seating 366
tools, see ESD control tools test methods 127–128, 167–172
trends 461 tools 230, 401
ways of working 322–323, 341–343 wrist strap 169, 195, 327, 474–475
work surface, see work surface work surface 170, 194, 205–206,
ESD control product care and 327, 348, 474
maintenance 179–234 ESD control product role and purpose
carts and mobile equipment carts and mobile equipment 210
211–212 earth bonding point 189
floors 187–188 floor 183–184, 187
gloves and finger cots 234 gloves and finger cots 232
ionizer 221 ionizer 217
storage racks and shelves 208 packaging 239–241
work surface see work surface personal grounding 190
wrist strap 195 storage racks and shelves 207–208
ESD control product qualification tools 229
127, 130, 137–138, 165–172, work surface 203–204
179–180, 185–187, 194–195, wrist strap 192–193
200–201, 203–206, 209, 211–212, ESD control product selection
215–216, 220–221, 227–231, 233, 179–180, 184–185, 190, 194–195,
307, 309–311, 232–324, 326–330, 200, 204–205, 208, 214–215,
347–349, 364–367, 383, 386, 401, 219–220, 226–227, 232–233, 307
466, 468, 474–475 carts and mobile equipment
automated equipment 137–138 212–213
carts and mobile equipment 170, earth (ground) 181–182, 328–329
186–187, 211–213, 366 earth bonding points 190, 334
earth (ground) 183, 328–329 floors 184–185
earth bonding points 183, 190, 328 footwear 200
environmental conditions 180, garment 226–227
185, 349 gloves and finger cots 232–233
floors 169–170, 185–187, 474 ionizer 219–220
footwear 169, 171, 200–201, 466, packaging 269–275
468, 474–475 seat 214–215
footwear and flooring 169, 171, storage racks and shelves 208
201, 330, 348, 386, 474 tool 230
484 Index
ESD protection 317–321, 338–339, conductive 134, 139, 173, 242, 244,
341 247–254, 256, 258–259, 261–263,
basis of ESD protection, see ESD 266–267, 270, 272–276, 335
control program, basis of ESD customer requirements 269–270,
protection 272, 275, 313–314, 337–338
ESD protection measures, see ESD electrostatic field shielding 105,
control measures 123, 172, 245–248, 253, 255–257,
evaluation of needs 283–289, 262–263, 265–267, 270, 274–276,
293–294, 296–302, 319, 338 442–443
ESD protective packaging ESD shielding 105, 123, 172,
269–275, 317–318, 338–339, 341 242–243, 246, 248–249, 253–259,
limits of standard precautions 262–263, 265–268, 271, 274–276,
296–297 338, 346, 382–384
evaluation of risks 281–289, foam 242, 251, 258–259, 274–275
294–297, 319, 332–334 functions 123, 241–242, 273–275,
on chip protection 64–65, 68–71, 276
75, 79 humidity (moisture), effect of
on chip protection targets 68–70 239–240, 243, 247, 250–251, 256,
ESD protective packaging 12, 91–95, 433, 443
intimate 137, 242, 248–249, 256,
98, 100–101, 108–109, 122–124,
259, 261, 263, 273–275, 290
126, 130, 134, 137, 143, 147–149,
labels 262, 264, 276
173–174, 213, 239–276, 290,
low charging 173, 243–244,
297–298, 301–302, 306–307, 309,
249–251, 254–259, 261–262,
312- 313, 317–318, 322, 324, 326,
266–268, 272, 332, 433, 443
330, 332, 335- 339, 341–342,
marking 242, 262, 264, 272,
346–347, 350, 352–354, 356–357,
275–276, 314, 337–338
377–384, 413, 417–418, 425, 429, materials 250–254, 332, 342
433, 442–443, 455, 459, 461, 467, measurements 134, 149, 168, 244,
468–471 268–269, 346–347, 350, 352–354,
anodized aluminium 133, 140–141, 356–357, 377–383
144–146, 148, 253 moisture barrier 241, 253–255,
antistatic 173, 247, 250–251, 256, 257–258, 264–268
258, 266–268, 272, 335 opening 123–124, 256, 341–342
in automated systems 134, 137, operational environment 269–273
143, 147–149 pink polythene 245–247, 250–251,
bags 243, 245–247, 252–258, 261, 255–259, 272, 332, 433, 442–443
263–269, 271–272, 274–275, 297, properties 173, 239, 241–250, 332,
338, 341, 442–443 442–443
boxes and trays 251–252, 258–259, proximity 242, 249, 274–275
271, 342 resistance and resistivity 47, 100,
bubble wrap 254, 258 108–109, 123, 130, 134, 149, 168,
charge dissipation 122, 244, 247, 173, 239–249, 251–253, 255–256,
256, 273 258–259, 261–263, 265–271,
charging 123, 241, 243–244, 249, 273–275, 265, 268, 354–355, 338,
273–275, 433, 442 377–382
Index 487
care and maintenance 234 via floors 116–117, 181, 186, 191,
common problems 234 195, 198, 211, 214–215, 322–323,
compliance verification 234, 335, 329–330, 336, 348–349, 364,
337 376–377
grounding 231–234, 337, 319, functional earth 8, 110–111, 168,
399 171, 181, 183, 312, 328
insulating 231 gloves and finger cots, see ESD
qualification 233 control, gloves and finger cots
role and purpose 230, 232–233 hard 152
selection 232–233, 337 insulators 104, 443–444
standard requirements 335, 337 personnel, see grounding personnel
standard test methods 335, 337 qualification 181, 183, 190,
types 232, 319, 337 211–212, 215, 229, 233, 329–330
use 230–233, 297–298, 300–301, resistance to ground 13, 140–142,
319, 418 145, 147–149, 164, 166, 168, 174,
ground (grounding) 3–4, 7–8, 11–13, 181, 186–191, 193, 196–198,
98, 101, 104–105, 107–122, 200–201, 203–206, 209, 212–213,
127–130, 139–143, 145, 147, 215–216, 223, 227, 229, 231, 234,
167–171, 181–217, 223–234, 253, 287–288, 290–291, 299–300,
266, 268, 284, 288–292, 295–296, 327–330, 333, 335–337, 346,
310, 312, 322–324, 326–337, 342, 348–349, 351, 356–357, 363–364,
345–346, 348–349, 355–357, 374–377, 393–397, 399–400
362–366, 368, 370–371, 374–377, resistance to groundable point 180,
393–399, 404, 437–440, 443–444, 186–187, 194, 196, 203–204, 209,
446 212–213, 223, 227, 229, 327–328,
carts (trolleys), see ESD control, cart 335, 337, 346, 348, 365–366, 368,
(trolley) 370, 394–395
common problems 183, 363 role and function 110, 189
compliance verification 183, 190, safety, see safety, ground
195–197, 201, 203, 206, 209, 213, seats, see ESD control, seating
215–216, 228–229, 231, 310, selection 181–182, 190
328–329, 335–337, 348–349, 364, soft 152
415, 446 system 110–113, 115–116, 127, 167,
conductors 11, 139, 141–143, 147, 171, 326–327, 349, 374, 376, 393,
181, 217, 289–290, 292–293, 396, 399, 404
331–334, 436–440 tools, see ESD control, tools
elimination of ESD 110, 181, 217 types of ground 110–111, 171, 181,
equipotential bonding, see 328–329, 437–438
equipotential bonding work surfaces 113–115, 170, 180,
ESD control equipment 110–130, 182, 184, 204, 206–207, 209,
180–216, 230–234, 323–324, 327, 327–328, 335–336, 348, 363
336–337 wrist strap 139–140, 169, 191–197,
ESD control garments, see ESD 322–323, 328–330, 334, 355,
control, garment 376–377
490 Index
test 164, 168, 171, 181, 220–222, 376–378, 380–382, 385–386, 388,
328, 335, 337, 346, 359–360, 390, 393–396, 403–404, 407–408
383–385 compliance verification 167–172,
types 218 181, 188, 190, 195–196, 201, 203,
use 146, 172, 217, 220–221, 228, 206, 209, 213, 215, 221–222,
322, 333–334 228–229, 231, 234, 268, 311, 327,
X ray 42 336, 345–349, 364, 367, 383
ions 3, 32, 42–45, 106–107, 217–219 decay time 142, 145, 151, 220–221,
ion sources 217–219 230, 233, 268, 335–336, 346, 359,
isolated conductor 27–29, 44, 47, 75, 383–385, 399–404
101, 114, 119, 122, 141–142, 146, of gloves and finger cots 337,
149, 172, 217, 219, 221–222, 231, 403–404
287–291, 297–298, 300, 321–322, of packaging 268
324–325, 329, 333–334, 337, 393, of tools 230–233, 336, 346, 359,
435, 438–440, 446, 456, 466, 470 399–403
resistance to ground 149, 288, 300, earth 168
333–334 earth bonding point 190, 327, 377
standard requirements 172, 325, electrostatic field 168, 172, 346,
333 387–388
latent failures 52, 64, 66–68, 76–77 end to end resistance of ground cord
195, 327, 346, 371–372
m EMI detection, see ESD event
Machine Model (MM) 7, 54–55, ESD current 151
58–60, 62–64, 133, 135, 283–288 ESD event 408–409
marking 95–97, 167, 173–174, 215, ESD protective packaging 163,
231, 234, 242, 245, 262, 264, 272, 167–168, 244, 268–269, 346–347,
275–276, 309, 312, 314, 320–321, 352–354, 356–357, 377–384
326, 337–338, 451, 453–454, 458, ESD shielding 268, 346, 382–384
465, 468–469, 471 Faraday pail 404–407
EPA boundary 95–97, 320, 469 floor 158, 163, 170, 181, 186, 188,
measurements 163, 180, 181, 186, 336, 346, 348–350, 362–364
188, 190, 194–196, 200–201, footwear 163, 186, 169, 201–202,
203–206, 209, 212–213, 215–216, 346, 374
220–221, 225, 227–233, 244, footwear and flooring in combination
268–269, 284–286, 289, 291–293, 164, 169, 180, 186, 201, 200, 346,
299, 311, 327, 334–336, 345–409, 348–350, 374–377, 386
458, 461–462 garment 164, 170, 203, 225,
in automated equipment 147–151 227–229, 346, 350, 366–371
body voltage 169, 180, 186, 188, gloves and finger cots 231,
190, 194–195, 201, 203, 216, 348, 233–234, 346, 355, 359, 396–399,
386 403–404
carts (trolleys) 213 humidity 347–349, 361
charge 147, 151, 404–408 ionizer 147, 151, 164, 168, 171,
common problems 188, 205–206, 221–222, 334–335, 346, 359–360,
361–363, 365, 369, 372, 374, 383–385
492 Index