Photovoltaic Module Reliability Wohlgemuth All Chapter

Photovoltaic module reliability
Wohlgemuth
Visit to download the full and correct content document:
https://ebookmass.com/product/photovoltaic-module-reliability-wohlgemuth/
Photovoltaic Module
Reliability
Photovoltaic Module Reliability
John H. Wohlgemuth
Virginia
USA
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 John H. Wohlgemuth 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.
Limit of Liability/Disclaimer of Warranty

In view of ongoing research, equipment modifications, changes in governmental regulations, and the
constant flow of information relating to the use of experimental reagents, equipment, and devices,
the reader is urged to review and evaluate the information provided in the package insert or instructions
for each chemical, piece of equipment, reagent, or device for, among other things, any changes in the
instructions or indication of usage and for added warnings and precautions. While the publisher and
authors have used their best efforts in preparing this work, they make no representations or warranties
with respect to the accuracy or completeness of the contents of this work and specifically disclaim
all warranties, including without limitation any implied warranties of merchantability or fitness for
a particular purpose. No warranty may be created or extended by sales representatives, written sales
materials or promotional statements for this work. The fact that an organization, website, or product is
referred to in this work as a citation and/or potential source of further information does not mean that
the publisher and authors endorse the information or services the organization, website, or product may
provide or recommendations it may make. This work is sold with the understanding that the publisher
is not engaged in rendering professional services. The advice and strategies contained herein may not
be suitable for your situation. You should consult with a specialist where appropriate. Further, readers
should be aware that websites listed in this work may have changed or disappeared between when this
work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of
profit or any other commercial damages, including but not limited to special, incidental, consequential,
or other damages.
Library of Congress Cataloging‐in‐Publication data applied for

HB ISBN: 9781119458999
Cover Design: Wiley

Cover Image: © foxbat/Shutterstock
Set in 9.5/12.5pt STIXTwoText by SPi Global, Pondicherry, India
10 9 8 7 6 5 4 3 2 1
v
Contents
Acknowledgments xi
1 Introduction 1
1.1 Brief History of PVs 2
1.2 Types of PV Cells 4
1.3 Module Packaging – Purpose and Types 8
1.4 What Does Reliability Mean for PV Modules? 12
1.5 Preview of the Book 13
References 15
2 Module Failure Modes 17

2.1 roken Interconnects 17
B
2.2 Broken/Cracked Cells and Snail Trails 21
2.3 Delamination 24
2.4 Corrosion of Cell Metallization 26
2.5 Encapsulant Discoloration 28
2.6 Failure of Electrical Bonds Particularly Solder Bonds 31
2.7 Glass Breakage 33
2.8 Junction Box Problems 35
2.9 Loss of Elastomeric Properties of Back Sheets 36
2.10 Reverse Bias Hot Spots 37
2.11 By-Pass Diodes 39
2.12 Structural Failures 41
2.13 Ground Faults and Open Circuits Leading to Arcing 43
2.14 Potential Induced Degradation 46
2.15 Thin-Film Specific Defects 48
2.15.1 Light-Induced Degradation 48
2.15.2 Inadequate Edge Deletion 49
2.15.3 Shunts at Laser Scribes and Impurities in Thin Film 49
2.15.4 Failure of Edge Seals 50
References 51
vi Contents
3 Development of Accelerated Stress Tests 55

3.1 Thermal Cycling or Change in Temperature 57
3.2 Damp Heat 58
3.3 Humidity Freeze 59
3.4 Ultraviolet (UV) Light Exposure 60
3.5 Static Mechanical Load 61
3.6 Cyclic (Dynamic) Mechanical Load 62
3.7 Reverse Bias Hot Spot Test 63
3.8 Bypass Diode Thermal Test 63
3.9 Hail Test 64
References 65
4 Qualification Testing 67
4.1 JPL Block Buy Program 68
4.2 Evolution of IEC 61215 Qualification Test Sequence 75
4.3 IEC 61215 Test Protocol 80
4.3.1 MQT 01 – Visual Inspection 82
4.3.2 MQT 02 – Maximum Power Determination 82
4.3.3 MQT 03 – Insulation Test 82
4.3.4 MQT 04 – Measurement of Temperature Coefficients 83
4.3.5 MQT 05 – Measurement of NMOT 83
4.3.6 MQT 06 – Performance at STC and NMOT 84
4.3.7 MQT 07 – Performance at Low Irradiance 84
4.3.8 MQT 08 – Outdoor Exposure Test 85
4.3.9 MQT 09 – Hot Spot Endurance Test 85
4.3.10 MQT 10 – UV Preconditioning Test 88
4.3.11 MQT 11 – Thermal Cycling Test 88
4.3.12 MQT 12 – Humidity-Freeze Test 89
4.3.13 MQT 13 – Damp-Heat Test 89
4.3.14 MQT 14 – Robustness of Termination 90
4.3.15 MQT 15 – Wet Leakage Current Test 91
4.3.16 MQT 16 – Static Mechanical Load Test 91
4.3.17 MQT 17 – Hail Test 92
4.3.18 MQT 18 – Bypass Diode Test 92
4.3.19 MQT 19 – Stabilization 94
4.4 How Qualification Tests have been Critical to Improving the Reliability
and Durability of PV Modules 95
4.5 Limitations of the Qualification Tests 97
4.6 PV Module Safety Certification 98
4.6.1 Construction Requirements: IEC 61730-1 99
4.6.1.1 Components 99
4.6.1.2 Mechanical and Electromechanical Connections 101
4.6.1.3 Materials 103
4.6.1.4 Protection Against Electric Shock 105
4.6.2 Requirements of Testing IEC 61730-2 110
Contents vii
4.6.2.1 MST 01 – Visual Inspection 113

4.6.2.2 MST 02 – Performance at STC 113
4.6.2.3 MST 03 – Maximum Power Determination 114
4.6.2.4 MST 04 – Insulation Thickness Test 114
4.6.2.5 MST 05 – Durability of Markings Test 114
4.6.2.6 MST 06 – Sharp Edge Test 114
4.6.2.7 MST 07 – Bypass Diode Functionality Test 114
4.6.2.8 MST 11 – Accessibility Test 114
4.6.2.9 MST 12 – Cut Susceptibility Test 115
4.6.2.10 MST 13 – Continuity Test of Equipotential Bonding 115
4.6.2.11 MST 14 – Impulse Voltage Test 115
4.6.2.12 MST 16 – Insulation Test 116
4.6.2.13 MST 17 – Wet Leakage Current Test 116
4.6.2.14 MST 21 – Temperature Test 116
4.6.2.15 MST 22 – Hot Spot Endurance Test 117
4.6.2.16 MST 24 – Ignitability Test 117
4.6.2.17 MST 25 – Bypass Diode Thermal Test 117
4.6.2.18 MST 26 – Reverse Current Overload Test 117
4.6.2.19 MST 32 – Mechanical Breakage Test 118
4.6.2.20 MST 33 – Screw Connections Test – Test for General Screw
Connections MST 33a 118
4.6.2.21 MST 33 – Screw Connections Test – Test for Locking Screws MST 33b 119
4.6.2.22 MST 34 – Static Mechanical Load Test 119
4.6.2.23 MST 35 – Peel Test 119
4.6.2.24 MST 36 – Lap Shear Strength Test 120
4.6.2.25 MST 37 – Materials Creep Test 121
4.6.2.26 MST 42 – Robustness of Termination Test 121
4.6.2.27 MST 51 – Thermal Cycling Test 121
4.6.2.28 MST 52 – Humidity Freeze Test 121
4.6.2.29 MST 53 – Damp Heat Test 121
4.6.2.30 MST 54 – UV Test 122
4.6.2.31 MST 55 – Cold Conditioning 122
4.6.2.32 MST 56 – Dry Heat Conditioning 122
4.6.2.33 Recommendations for Testing of PV Modules from Production 122
References 123
5 Failure Analysis Tools 127

5.1 PV Performance – Analysis of Light I–V Curves 127
5.2 Performance as a Function of Irradiance 132
5.3 Dark I–V Curves 136
5.4 Visual Inspection 137
5.5 Infrared (IR) Inspection 143
5.6 Electroluminescence (EL) 145
5.7 Adhesion of Layers, Boxes, Frames, etc. 149
References 149
viii Contents
6 Using Quality Management Systems to Manufacture PV Modules 151

6.1 Quality Management Systems 151
6.2 Using ISO 9000 and IEC 61215 153
6.3 Why just Using IEC 61215 and ISO 9000 is No Longer Considered
Adequate? 154
6.4 Customer Defined “Do It Yourself” Quality Management and Qualification
Systems (IEC 61215 on Steroids) 156
6.5 Problems with the “Do It Yourself” System 157
References 163
7 The PVQAT Effort 165

7.1 Task Group 1: PV QA Guidelines for Module Manufacturing 167
7.2 Task Group 2: Testing for Thermal and Mechanical Fatigue 169
7.3 Task Group 3: Testing for Humidity, Temperature and Voltage 175
7.3.1 Corrosion 176
7.3.2 Delamination 177
7.3.3 PID 179
7.3.4 Delamination Due to Voltage Stress 181
7.4 Task Group 4: Testing for Diodes, Shading and Reverse Bias 182
7.5 Task Group 5: Testing for UV, Temperature and Humidity 186
7.6 Task Group 6: Communications of Rating Information 189
7.7 Task Group 7: Testing for Snow and Wind Load 189
7.8 Task Group 8: Testing for Thin-Film Modules 190
7.9 Task Group 9: Testing for Concentrator Photovoltaic (CPV) 190
7.10 Task Group 10: Testing for Connectors 190
7.11 Task Group 11: QA for PV Systems 191
7.12 Task Group 12: Soiling and Dust 191
7.13 Task Group 13: Cells 192
References 192
8 Conformity Assessment and IECRE 195

8.1 Module Conformity Assessment – PowerMark, IECQ, PVGAP, and IECEE 195
8.1.1 PV-1: “Criteria for a Model Quality System for Laboratories Engaged in Testing
PV Modules” 196
8.1.2 PV-2: Model for a Third-Party Certification and Labeling Program for PV
Modules 197
8.1.3 PV-3: Testing Requirements for a Certification and Labeling Program for PV
Modules 197
8.1.4 PV-4: Operational Procedures Manual for the Certification Body of the PV
Module Certification Program 197
8.1.5 PV-5: Application and Certification Procedures for the PV Module Certification
Program 198
8.2 IECRE – Conformity Assessment for PV Systems 201
References 206
Contents ix
9 Predicting PV Module Service Life 209

9.1 Determining Acceleration Factors 210
9.1.1 Thermal Cycling 212
9.1.2 Discoloration of the Encapsulant 213
9.1.3 PET Hydrolysis 213
9.2 Impact of Design and Manufacturing on Failure or Degradation Rates
for PV Modules 215
9.3 Impact of Location and Type of Mounting on Failure or Degradation Rates
for PV Modules 216
9.4 Extended Stress Testing of PV Modules 221
9.5 Setting Up a True Service Life Prediction Program 226
References 227
10 What does the Future Hold for PV and a Brief Summary 229
10.1 Current Work on Updating Standards 229
10.1.1 Second Edition of IEC 61215 Series 229
10.1.2 Amendment 1 to Second Edition of IEC 61730-1 and IEC 61730-1 234
10.1.3 IEC TS 63126 – Guidelines for Qualifying PV Modules, Components
and Materials for Operation at High Temperatures 234
10.2 Looking to the Future 237
10.2.1 Degradation Rates 237
10.2.2 Module Lifetime 238
10.3 Brief Summary 239
10.3.1 Personal Reflections 240
References 240
Index 243
xi
Acknowledgments
In my 50‐year PV journey from graduate student through retirement from National Renewable
Energy Laboratory (NREL), I have been assisted and befriended by many colleagues along
the way. I would like to take this opportunity to acknowledge some of them.
The first one I would like to acknowledge is Hillard B. Huntington of RPI, my PhD thesis
advisor. Then, I would like to acknowledge my two Post‐Doctorate advisors, Donald Brodie
of the University of Waterloo who provided me with the opportunity to begin research
work in semiconductors and Martin Wolf of the University of Pennsylvania who gave me
the opportunity to begin my career in photovoltaics.
I certainly need to acknowledge the impact of Joseph Lindmayer and Peter Varadi on my
career. They offered me the opportunity to work for Solarex, one of the first PV‐module
manufacturers. Joseph provided me with my first education in PV while Peter provided the
lessons in business. From the early days of Solarex, I would like to acknowledge the assis-
tance and training I received from Chuck Wrigley, John Goldsmith, and Ramon Dominguez.
I would also like to acknowledge the efforts of those who actually made the solar cells and
modules for the technology group during this time including Don Warfield, Joe Creager,
Dan Whitehouse, George Kelly, and Tim Koval. It was John Corsi and Alain Ricaud who
brought me back to Solarex after a brief stint out of PV. Alain is also the one I must thank
for getting me involved in IEC standards where I spent 40 years as a member of Working
Group 2 (Modules) of IEC TC82 – the Technical Committee on PV. From that era at Solarex,
I would also like to acknowledge the support from, and collaborations with, Ray Peterson,
Steve Shea, Mohan Narayanan, Jim Emming, and Jean Posbic. I would also like to acknowl-
edge the contributions of Mark Conway and David Meakin who made most of the meas-
urements for technology.
Solarex then became BP Solarex and finally just BP Solar. It was Peter Bihuniak who
changed my focus from making more efficient and less expensive cells and modules to
module reliability, performance measurements and standards. From the BP Solar days,
I would like to recognize Steve Ransome, Danny Cunningham, David Carlson, Paul
Garvison, and Zhiyong Xia for their contributions to PV and for their collaborations
with me. I would also like to recognize the engineering staff from module technology,
including Andy Nguyen, Jay Shaner, Jay Miller, Dinesh Amin, and Neil Placer.
I would like to thank Sarah Kurtz for facilitating my move to NREL. Sarah gave me the
opportunity to transition my work in PV‐module reliability to NREL and beyond where it
could impact much of the PV world. While at NREL, I would like to acknowledge support
xii Acknowledgments
and collaboration with Sarah, Michael Kempe, Nick Bosco, and David Miller. I would also
like to recognize a number of additional members of the NREL staff for their valuable work
in PV including Dirk Jordan, Tim Silverman, Mike Deceglie, Ingrid Repins, and Peter
Hacke.
There are also a number of people from outside the organizations that I worked for
whom I would like to acknowledge for their collaborations with me and for their contribu-
tions to PV reliability including Bob Hammond of ASU, Govindasamy TamizhMani (Mani)
of ASU, Neelkanth Dhere of FSEC, and Jim Galica of STR and Stevens Urethane.
I would also like to recognize specific contributions to IEC PV standards. PV standards
would not be nearly as effective or as advanced as they are today if it weren’t for the efforts
of the secretaries to IEC TC82, particularly Jerry Anderson, Howard Barikmo, and George
Kelly. I would also like to thank Steve Chalmers and Alex Mikonowicz, long‐serving
technical advisors to the US TAG for IEC TC 82, for their efforts. I would particularly like
to acknowledge Tony Sample, who took over the convenorship of WG2 when I retired.
Chris Flueckiger and Kent Whitfield should also be recognized for the support they gave in
developing the US version of the IEC module safety standard.
I would also like to acknowledge those who worked with me to found PVQAT ‐ namely
Sarah Kurtz, Tony Sample, Michio Kondo, and Masaaki Yamamichi. I would like to thank
my wife, Beth for all of her support and encouragement during my four decade PV career
and particularly during the writing of this book.
Finally, PV in general and PV reliability in particular, has been a great technology field to
work in with many dedicated professionals working together to advance the technology,
and to solve the industry’s problems and eventually the world’s energy problems. For any
students using this book for their University course work, I applaud your choice of a
technology to study and hope that this book advances your knowledge in PV reliability and
maybe encourages you to consider it for your career choice.
1
Introduction
Photovoltaics (PVs) is the direct conversion of light into electricity. Typically, this means
generation of electricity from sunlight, a renewable energy process without release of pol‑
lution or greenhouse gases. PV is one of the renewable energy sources that offers the poten‑
tial to replace burning of fossil fuels and, therefore, to slow the growing effects of global
climate change.
When the author began working in PV at Solarex in 1976, the entire worldwide annual
production of PV modules was less than 100 kW. Numerous groups [1, 2] are predicting
that more than 100 GW of PV modules will be produced and shipped in 2019. That is a
growth in production volume of more than six orders of magnitude across a span of little
more than 40 years. PV has gone from a small niche business, providing electricity for
remote power applications to a mainstream, electric power producing industry. It has been
estimated that PV provided approximately 3% of the world’s electricity in 2018. PV and
wind have been the two fastest growing sectors in commercial electricity production in the
world for a number of years [3].
So why has PV been so successful? Certainly, the fact that PV is a clean, non‐polluting
source of electricity is very important. The fact that the prices for PV modules have fallen
dramatically from more than $40 per peak watt in the 1970s to around $0.40 per peak watt
today [4] has also certainly helped. Today’s price level makes PV one of the lowest‐cost
sources of electricity in the world. However, none of these would really matter if PV tech‑
nology, in general and PV modules in particular, were not very reliable and have long ser‑
vice lives. How many products do you know of that have to work outdoors in all kinds of
weather and yet are provided with a 25‐year warranty? Since PV is a solid‐state process,
there are no moving parts and very little to wear out so PV modules should be able to oper‑
ate for a long time. Longevity is critical to the value of PV since the investment to install a
PV system is made in the beginning and then the income accumulates over years as the
electricity is sold. Without reliable PV products, no one would be risking billions of dollars
to purchase and install the PV power systems that have made PV the success it is today.
This book is about Photovoltaic Module Reliability. Modules are usually the most reliable
component within a PV system, but it is important to continue to study their reliability.
Modules are the most expensive and by far the hardest to replace component in the PV
system if something goes wrong. Any large‐scale failure of modules could result in dra‑
Photovoltaic Module Reliability, First Edition. John H. Wohlgemuth.

© 2020 John Wiley & Sons Ltd. Published 2020 by John Wiley & Sons Ltd.
2 Photovoltaic Module Reliability
matic reductions in growth of the PV power industry. This book has been written from a
historical prospective to help the reader understand how PV modules were able to achieve
25‐year lifetimes and a well‐deserved reputation for reliability. Knowing how this level of
reliability was achieved, can help in understanding how to maintain similar reliability in
the future with new generations of products.
This introductory Chapter will provide some background for the in‐depth look into mod‑
ule reliability in subsequent chapters. The first section provides a brief history of PV. The
second section discusses some of the different types of materials and devices used for com‑
mercial solar cells. The third section covers module packaging, including their purpose; the
types of structures used for different modules, and a brief introduction to the types of mate‑
rials used in today’s commercial PV modules. The next section discusses what the author
means by the reliability of PV modules as well as introducing several other terms that will
be used throughout the book. The final section in this first chapter provides a brief over‑
view of what is contained in each of the subsequent chapters.
1.1 Brief History of PVs
To outsiders, it may seem like PV appeared quickly out of nowhere. In reality, the technol‑
ogy has been under development for a long time. Let’s take a brief look at the history of PV.
●● Edmond Becquerel discovered the PV effect in 1839. So, PV is certainly not a new tech‑
nology [5].
●● Albert Einstein published a paper explaining how the PV effect worked in 1905. In 1921,
he received the Nobel Prize in Physics for his discovery of how the PV effect works [6].
PV was a lot less controversial than relativity at that time.
●● In 1954, a group at Bell Laboratories developed the first practical silicon solar cells [7].
●● The Bell Laboratories development was just in time for PV to provide power for all of the
US satellites designed to perform in space for more than a few days. Vanguard 1 launched
in March 1958, was powered by PV and continued to transmit data back to earth for six
years [8], while purely battery‐powered satellites typically only provided data for a few
months. Most US satellites continue to use PV as their primary energy supply. This
means that your satellite weather forecasts, long‐distance communications and TV sig‑
nals have always been powered by PV. Most of us have been taking advantage of PV in
this way for decades.
●● In the 1970s, a terrestrial PV business was developed by two small companies (Solarex
Corporation and Sensor Technology) to provide power systems for remote applications.
In these remote site applications, PV was cost effective even at a $20–$40/Wp cost for
modules. These remote applications included telecommunications, weather stations,
navigational aids, water pumping, fence charging, remote vacation homes and cathodic
protection.
●● Partially because of the US energy crises of 1973 and 1979, the US government began an
expanded effort in renewable energies. The Flat‐Plate Solar Array (FSA) Project, funded
by the US Government and managed by the Jet Propulsion Laboratory, was formed in
1975 to develop the flat‐plate module and array technologies needed to attain widespread
Introduction 3
terrestrial use of PV [9]. While many important developments came out of this effort,
three had particularly important impacts on future PV efforts. The first was the initial
efforts to evaluate the potential for PV technologies to undergo significant cost reduc‑
tions and, therefore, eventually compete with traditional sources of electricity. The sec‑
ond was the proof that PV modules could be assembled into larger‐scale systems that
could then power real‐world applications. The third was the Jet Propulsion Laboratory
(JPL) Block Buy Program to be discussed in more detail in Chapter 4. This effort led to
the use of acceleration stress tests and the establishment of qualification tests for PV
modules. This went a long way toward improving the reliability and increasing the ser‑
vice life of PV modules.
●● The election of Ronald Reagan to the Presidency in 1980 resulted in a huge setback for
PV. The national budget for PV was slashed drastically and PV research in the US dropped
to a small fraction of what it had been under Jimmy Carter’s Administration. As a result,
PV progress slowed appreciably and the center of PV development shifted away from the
US to Europe and Asia.
●● In 1994, The Japanese Ministry of Economy, Trade and Industry (METI, formerly called
MITI) launched a subsidy program for residential PV systems with an overall goal of
installing 4.82 GW of PV by 2010. The program was launched with a subsidy of 50% of the
cost of the PV system. The program attracted homeowners not only because of the sub‑
sidy, but also because the residential electricity rates in Japan were about 24 Yen/kWh
equivalent to about $0.24/kWh at that time, among the highest residential rates in the
world. On the other hand, mortgage interest rates in Japan are low (1–2%) and were
extendable to cover the costs of residential PV systems. From 1994 until 2003, the
Japanese PV market grew steadily by about 30% a year even though the subsidy level was
reduced every year. By 2003, the Japanese market was the largest in the world, represent‑
ing >40% of the world’s PV demand.
●● Germany used a different approach to subsidizing PV systems. Rather than assisting with
the initial purchase as was done in Japan, the German program pays a rate or tariff‐based
incentive on the electricity actually produced. The PV system owner is paid a specified
rate for each kWh produced by the PV system. While such a program had been ongoing
in Germany since the late 1990s, a change in the structure of the incentive program that
began in 2004, resulted in explosive growth of the German PV market in 2004 and 2005.
The feed in tariff rates were established at €45.7 ¢/kWh for ground mounted systems
(with a 6.5% annual reduction), at €57.4 ¢/kWh for rooftop residential and at ~€54 ¢/
kWh for rooftop commercial (with a 5% annual reduction). In addition to the feed in
tariff, preferential loans were made available for PV. This whole program was focused on
PV’s C02 lowering potential. In Germany, the incentive for PV went from environmental
to economic with significant financial returns earned via investment in PV systems. In
2004, the German PV market more than doubled, overtaking Japan as the world’s largest
PV market. The dramatic market growth was actually constrained in late 2004 and 2005
by lack of module availability.
●● In the mid‐2000s, China began to realize that PV was going to be an important energy
industry and so Chinese companies began manufacturing PV modules. By 2010, an
appreciable share of PV modules in the world was being produced in China. By
2015/2016 a majority of all PV modules manufactured in the world were being made
in China. The volume has continued to grow since then. The combination of large‐
volume production and low costs for labor and infrastructure has led to dramatic
decreases in the selling price for PV modules. According to Paula Mints [4] average
worldwide module costs have gone from $3.00 to $4.00 per watt in 2007/2008 to less
than $0.5 per watt in 2018.
The combination of market need, driven by feed in tariffs and the building of large facto‑
ries in China led to an explosive growth in the PV industry. With the low cost of modules
today, PV is competing successfully with conventional forms of electricity generation.
Large (>100 MW) PV systems are being installed around the world with China leading the
way both as producer and consumer of PV modules. For a detailed account of the history
of how the Photovoltaic Industry began see Peter Varadi’s book entitled “Sun Above the
Horizon” [10]. For a detailed description of how the PV industry was able to grow so rap‑
idly see Peter’s book entitled “Sun Towards High Noon” [11].
1.2 Types of PV Cells
Solar cells are the devices that convert sunlight into electricity. Solar cells are made of semi‑
conductor materials and use a junction to separate the carriers of different charges, that is,
to separate the electrons from the holes to create a voltage. There are a number of good
articles on how solar cells work including one by the author [12] as well as others in print
[13] and on line [14]. This book will not go into great detail about how solar cells work, but
will discuss them in terms of how they impact the reliability of PV modules.
A variety of materials have been used commercially to make solar cells. The list below
discusses those that have been successfully utilized in terrestrial PV modules and are avail‑
able commercially today.
Crystalline silicon materials: Crystalline silicon materials were the first utilized for fabri‑
cation of solar cells by Bell Laboratories [7]. They were also the first used for space applica‑
tions and then the first commercialized for remote power terrestrial applications. Crystalline
silicon technology has been, and still is, the dominant commercial PV material, with more
than 90% of the terrestrial PV market in 2018 [4].
Crystalline Si is a wafer‐based technology. Today, all of the commercial wafers are cut
from ingots. There are two different types of crystalline silicon materials, single or mono‐
crystalline and multi‐crystalline. They differ in the way the silicon ingots are grown. The
single‐crystal ingots are grown by a crystal growth process called Czochralski (CZ), where
a single‐crystal seed is dipped into a bath of molten silicon. As the seed is withdrawn, the
silicon grows the same orientation as the seed. This is the same process that is used to make
the wafers for many semiconductor devices. Single‐crystal ingots are round, but round
wafers do not pack very well into rectangular modules so, in most cases, the round wafers
are cut into square or pseudo‐square wafers for PV. Multi‐crystalline Si is grown using
direction solidification in a crucible or mold. This process was developed specifically for PV
in the late 1970s [15]. According to Paula Mints [4], in 2018 multi‐crystalline Si technology
accounted for slightly more than 50% of the worldwide shipments of PV modules, while
mono‐crystalline Si accounted for approximately 45%.
Introduction 5
Metal grid lines AR coating
Solar cell p-n junction Back metallization
Figure 1.1 Cross-sectional drawing of screen-printed cry-Si solar cell.
For both mono‐ and multi‐crystalline Si, the ingots are cut into wafers using wire saws.
The resultant wafers are then processed into cells and the cells incorporated into modules
in very similar ways. There are a number of cell processes in use today. Let’s take a look at
screen‐print technology (one of the simpler methods) since most of the cell components
will be similar no matter which cell design is used. Figure 1.1 shows a diagram of the
important features of a screen‐printed cry‐Si solar cell.
The solar cell is fabricated on the Si wafer. A p‐n junction is grown in the front (sun side)
usually by diffusion. Metal grid lines are printed on the front to collect the current gener‑
ated by the cell. The back side is also metallized. The drawing shows continuous back
metallization though a grid can also be used on the back. Finally, because bare silicon is
quite reflective, an antireflective coating (AR) is usually applied to the front. For mono‐
crystalline Si the front surface is usually textured to increase optical absorption. This is
one of the reasons that mono‐crystalline cells are usually a few percent higher in effi‑
ciency than multi‐crystalline cells. It is important to remember the different components
in the cell when addressing reliability, because it is usually not the Si wafer or the p‐n
junction that degrade, but rather the contacts, the interconnections between cells or even
the AR coating.
Screen‐printed cry‐Si cells are typically fabricated on 15.6 × 15.6 cm wafers. Over the
years, the wafer thickness has been slowly reduced to lower the cost of the Si used in the
module. Today, Si wafers are typically less than 200 μm thick, which makes them more
susceptible to breakage than the thicker wafers that were used in the earlier days of PV.
Today most commercial screen‐printed cells have a conversion efficiency in the range of
15–19% with the multi‐crystalline Si cells at the low end and monocrystalline Si cells at
the high end.
A number of commercial cell manufacturers have utilized specialty structures in order to
increase cell performance. Several examples are:
●● SunPower has commercialized cells with all of the contacts on the back of the cell. This
cell structure eliminates front surface shadowing and can provide improved collection of
the high currents associated with large area cells. This cell structure requires silicon sub‑
strates with very high lifetimes and excellent front surface passivation. SunPower offers
their commercial cells with conversion efficiency up to 22.7% [16].
●● The HIT (Heterojunction with intrinsic thin layer) cell uses heterojunctions between a‐Si
and crystalline silicon to produce a much higher voltage than the standard p‐n junction.
So HIT cells have high efficiency with research cells reported at 25.6% [17] and commer‑
cial modules available from Panasonic with reported efficiencies of 19.7% [18].
●● One of the latest methods for increasing cell efficiency is the use of PERC Technology
(Passivated emitter rear cell). PERC cells use the same screen‐printed front surface as
standard screen‐print technology, but the rear is modified by replacing the full metal
coverage with a passivated dielectric layer with small area back contacts. PERC improves
the cells by reducing the back‐surface recombination of carriers and improving the
reflection of long wavelength light back into the cells [19]. Solar World has reported
PERC cell efficiencies of 22% [20]. There have been a number of forecasts that PERC will
gain significant market share in the next few years.
Each solar cell produces a voltage determined by the semiconductor junction. For crys‑
talline Si the typical p‐n junction cell has an open circuit voltage of 0.6–0.7 V and a peak
power voltage of around 0.5 V. Therefore, to reach useful voltages a number of cells are
connected together in series into a module. Each cell has metal contacts typically copper
ribbons attached to the front metal grid and then to the back metallization of the next cell
in the string. Ribbons from the front of the cell before it in the string are attached to its back
metallization. In this way cell voltages are combined to reach useful levels. Today, most
cry‐Si power modules have 60 or 72 cells in series. The next section will talk about the mod‑
ule packaging in more detail.
Thin Films Materials: In the thin‐film approach, the various layers of a PV device are
deposited directly onto a substrate or superstrate. The advantages of thin films are the
potential ability to:
●● Dramatically reduce the amount of material utilized. In some cases, the active semicon‑
ductor only needs to be a fraction of a micron thick in order to absorb most of the inci‑
dent sunlight.
●● Directly integrate into a higher voltage module, thereby eliminating much of the han‑
dling and labor necessary to produce cry‐Si modules.
For monolithically integrated thin film modules, the deposition processes and the inte‑
gration processes are described in Table 1.1 and the resultant structure shown in cross
section in Figure 1.2. Figure 1.2 is not drawn to scale as the thin film layers and the three
different scribe lines are exaggerated in size so you can see them. In thin film devices, the
semiconductors have considerably higher resistance than cry‐Si so metal grids do not
work well. Instead, a Transparent Conductive Oxide (TCO) is often used on the semicon‑
ductor surface to provide a conductive path as well as serving as the AR. Because the TCO
layers are not as conductive as a metal grid, thin film solar cells are usually long and
skinny, so that the collected current only flows a short distance across the cell to the next
cell. Monolithically integrated thin film modules often have a large number of skinny cells
connected in series, resulting in higher voltage, lower current and reduced series resist‑
ance losses.
Thin film layers of materials are more susceptible to corrosion than bulk materials.
When thin film layers are deposited directly in contact with glass, they may also be suscep‑
tible to ion flow in the glass [21]. This is why many thin film modules are protected in pack‑
ages that are designed to keep moisture out for the lifetime of the product.
Introduction 7
Table 1.1 Process for making monolithically integrated thin film modules.
Step # Step Name Description
1 Transparent Conductive Oxide TCO Deposit TCO layer onto the substrate
2 P1 Scribe Scribe to remove TCO from selected areas
3 Semiconductor Deposition Deposit the semiconductor layers that make the
solar cells
4 P2 Scribe Scribe to remove the Semiconductors from selected
areas
5 Metallization Deposit the back metallization
6 P3 Scribe Scribe to remove the Metallization from selected
areas
Semiconductor layers Metallization
TCO
Substrate
P1 P2 P3
Figure 1.2 Cross-sectional drawing of monolithically integrated thin film module.
According to Mints [4], thin films only represented about 5% of total worldwide module
shipments in 2018, a declining percentage from previous years. The thin‐film PV commer‑
cial market includes the following three materials.
1) Cadmium telluride (CdTe): CdTe is a well‐known semiconductor often used in high
performance infrared (IR) sensors. CdTe absorbs visible light very strongly, so very
thin films (1–2 μm) are sufficient to absorb most of the sunlight. Commercial CdTe
modules are typically fabricated on glass with a structure similar to the cross‐
sectional drawing in Figure 1.2. First Solar is by far the largest supplier of CdTe
modules. First Solar now offers large area, monolithically integrated CdTe modules
with efficiencies of up to 18% [22]. CdTe technology is especially attractive because of
low‐cost manufacturing. In 2018, First Solar, selling just CdTe modules shipped about
3% of the world’s PV modules [4].
2) Copper Indium Diselenide (CIS) and Copper Indium Gallium Diselenide (CIGS): Both
CuInSe2 (CIS) and Cu(InGa)Se2 (CIGS) are ideal PV‐absorber materials. The band gap
is near the optimum for absorbing the terrestrial spectrum. These materials have strong
optical absorption so very thin films (~1 μm) are sufficient to absorb most of the sun‑
light. Grain boundaries and surfaces in CIS and CIGS are electronically benign, so sim‑
ple polycrystalline films yield reasonably high efficiency PV devices without complex
grain boundary passivation. Typical CIS and CIGS cells use thin CdS layers to form p‐n
junctions, molybdenum for ohmic contacts to the CIS or CIGS, and transparent conduc‑
tors like zinc oxide and indium‑tin oxide for contact to the CdS and to serve as an AR
coating. In most cases, CIS and CIGS cells are deposited upside down, meaning that the
back metal is deposited first and the cells are deposited from back to front onto the
metal. So, CIS and CIGS modules often have the cells deposited onto the back glass of a
glass/glass package or onto a metallic substrate that is then diced up to produce indi‑
vidual cells that are packaged like you would wafer‐based cells. While CIS and CIGS are
attractive PV materials, it has taken a long time to move from the laboratory to commer‑
cialization. Two reasons for this appear to be the difficultly in scaling to larger sizes and
larger volumes, related to uniformity of the deposited films, and sensitivity to environ‑
mental stresses. Reliability issues relate to the humidity sensitivity of the contacts and
TCO necessitating hermetic sealing to achieve long‐term life. Solar Frontier from Japan
is the world’s largest manufacturer of CIS PV modules offering large area, monolithi‑
cally integrated CIS modules with conversion efficiencies up to 15% [23]. In 2018, CIS/
CIGS made up about 1% of the worldwide shipments of PV modules [4].
3) Amorphous Silicon (a‐Si): Alloys of a‐Si made of thin‐film hydrogenated silicon
(a‐Si:H) can be deposited on either glass superstrates or flexible metallic substrates.
Because of the low minority carrier lifetimes in doped a‐Si, p‐i‐n (p‐type, intrinsic,
n‐type) cell structures are utilized rather than normal p‐n junctions. Even in undoped
a‐Si, the minority carrier lifetimes are short and because of the Staebler‐Wronski effect,
in which light induces additional electrically active defects, the thickness of individual
layers must be minimized. Therefore, to achieve reasonable efficiencies, multi‐
junctions are usually utilized. Commercial a‐Si is typically fabricated with a structure
similar to the cross‐sectional drawing in Figure 1.2. The maximum reported efficiency
for a‐Si modules is about 12% [17], which is less than the efficiencies of commercially
available CdTe or CIGs modules. Commercial a‐Si modules have efficiencies in the
5–8% range after light stabilization. Most a‐Si products are used in consumer‐type prod‑
ucts (watches, calculators, etc.) not for power production.
Over the years, there have been a number of efforts to use optical concentration to
increase the output power that can be obtained from PV devices. Some of these efforts have
resulted in laboratory systems with high efficiencies and good performance. However, to
date, none have been able to establish a sustainable commercial market. For this reason,
reliability of concentrator PV will not be discussed in this book.
1.3 Module Packaging – Purpose and Types
While the solar cells actually produce the electricity, the module package is important for the
continued operation of the solar cells. Often the costs associated with the packaging exceed
the costs of the cells themselves. Historically, it is usually the package that fails first, ulti‑
mately leading to degradation of the cells, conductors, connectors and diodes resulting in
failed or degraded modules. The PV module package provides for the following functions:
●● Mechanical support – holding the cells in place pointing toward the sun.
●● Dielectric protection – keeping the high voltage away from people and keeping current
from flowing out of the array circuit (to ground or in a loop) where it has the potential to
cause a fire.
Introduction 9
Table 1.2 Typical commercial module constructions.
Glass Superstrate: Cry‐Si Cells

●● Glass/encapsulant/cry‐Si cells/encapsulant/backsheet
●● Glass/encapsulant/cry‐Si cells/encapsulant/glass
Glass Superstrate: Thin Film Cells
●● Glass/thin film cells on front glass/encapsulant/glass with edge seal
●● Glass/thin film cells on front glass/encapsulant/backsheet
●● Glass/encapsulant/thin film cells on back glass/substrate
Flexible substrates
●● Transparent frontsheet/encapsulant/thin film cells/flexible substrate
●● Transparent frontsheet/encapsulant/cry‐Si cells/encapsulant/flexible substrate
●● Protection of the cells, diodes and interconnects from the weather (UV, rain, humidity,
hail, etc.)
●● Coupling of the maximum amount of light energy possible into the solar cells (at all
angles at the wavelengths that the cells can utilize).
●● Cooling of the cells to minimize their temperature increase.
There are really just a few types of module constructions that make up the vast majority
of commercial PV modules. Table 1.2 provides a list of the types of typical commercial
module constructions. A vast majority of PV modules use glass as the front surface because
of its excellent optical properties and as we will see in Chapter 2 as the main structural sup‑
port because of the low thermal expansion coefficient of glass.
The first construction for cry‐Si modules (Glass/encapsulant/c‐Si cells/encapsulant/
backsheet) has certainly been used on more modules than any other and still remains the
most popular in the industry. Figure 1.3a shows a cross‐sectional drawing of a Glass/
encapsulant/cry‐Si cells/encapsulant/backsheet module construction. The second
construction for c‐Si modules substitutes a second glass layer for the standard backsheet
as shown in cross‐section in Figure 1.3b. This type of construction is becoming more
popular especially for use in bifacial designs (modules that produce electricity from light
that falls on both sides, not just the front). The third construction for c‐Si modules is a
flexible design. This is shown in Figure 1.3c. Flexible modules are usually designed as
portable power supplies to be carried and deployed when needed. They are not designed
for continuous outdoor exposure.
Thin film cells are deposited onto a foreign substrate. These substrates can be glass where
the cells are deposited right side up or upside down depending on the technology of the
Glass
Cells
Encapsulant
Backsheet
Figure 1.3a Cross-sectional drawing of glass/encapsulant/cry-Si cells/encapsulant/backsheet

module.
Glass
Cells
Encapsulant
Glass
Figure 1.3b Cross-sectional drawing of glass/encapsulant/cry-Si cells/encapsulant/glass module.
Transparent frontsheet
Encapsulant
C-Si cells
Flexible or rigid substrate
Figure 1.3c Cross-sectional drawing of flexible cry-Si module.
particular thin film material being used. Figures 1.4a and 1.4b show the cross section of
these two types of module constructions. In Figure 1.4a, the thin film is deposited on the
backside of the front glass. This is typical of how CdTe and a‐Si modules are fabricated.
Figure 1.4a has been drawn with edge seals as this is typically how CdTe modules are fab‑
ricated today. The edge seals are designed to keep moisture from reaching the active cell
area for the lifetime of the product (typically warrantied by the manufacturer for 25 years).
In Figure 1.4b, the thin film is deposited on the front side of the back glass. This is typical
of how CIS and CIGS modules are fabricated. Figure 1.4b has also been drawn with edge
seals, but edge seals are not as prevalent in these types of modules. In this case, the super‑
strate can also be made of glass though other materials are often used.
In some cases, thin films are deposited in large areas and then cut to cell size afterwards.
Basically, creating wafers out of thin films which then have to be electrically connected in
series just like cry‐Si cells. Figure 1.4c shows the typical construction used for such thin
film modules, although any of the constructions used for cry‐Si wafers could also be used
Edge seal Front glass

Encapsulant
Thin film PV Back glass
Figure 1.4a Cross-sectional drawing of front glass/thin film cells/encapsulant/back glass modules.
Glass
Edge seal Encapsulant
Thin film PV
Substrate
Figure 1.4b Cross-sectional drawing of front glass/encapsulant/thin film cells/substrate modules.

Introduction 11
Transparent frontsheet
Thin film PV Encapsulant

on substrates
Flexible or rigid substrate
Figure 1.4c Cross-sectional drawing of module structures for thin film wafer like cells.
to package these thin film “wafers.” Some CIGS modules have been made with these types
of cells. They are particularly of interest for fabrication of large‐area flexible modules.
There are only a handful of materials that appear in these drawing so let’s briefly take a
look at the properties required and those typically selected for use in PV modules.
Glass: When glass is used as the superstrate, one of the properties of interest is the optical
transmittance over the wavelength range, that solar cells can effectively use the photons,
from about 300 nm to 1100 nm for cry‐Si for example. To maximize performance without
significantly increasing the cost, most cry‐Si modules and some thin‐film modules are built
using low iron glass which has better transmittance than the standard soda lime (window)
glass. Some thin‐film PV modules do use regular soda lime glass to keep the cost down. In
addition, most cry‐Si modules use tempered or heat‐strengthened glass to provide added
strength to withstand wind and snow loads as well as hail impact. Some thin‐film modules
can’t use heat‐strengthened glass because the thin‐film deposition process occurs at such a
high temperature that the heat strengthening would be removed from the glass. In this
case, the modules are usually built with double glass (glass on front and back) to provide
the strength necessary to survive in the field.
Encapsulant: The encapsulant is the material that surrounds the cells and the “glue” that
holds the whole package together. The encapsulant should provide good adhesion to all of
the other components within the module so that everything in the package stays stuck
together for 25 or 30 years. This is usually assisted by addition of a primer into the encapsu‑
lant formulation itself. Of course, any of the encapsulant material that is used in front of
active solar cells must be optically transparent and resistant to UV exposure. Since the
encapsulant surrounds the solar cells, it helps to provide electrical isolation. So, materials
used as encapsulants must have low‐bulk conductivity (or high‐bulk resistivity) to minimize
flow of leakage currents. Some encapsulants are cross‐linked during module lamination to
provide stability at the high temperatures at which they operate. Others, however, do not
have to be cross‐linked since they are stable enough not to flow or creep at typical module
operating temperatures. Typical examples of materials used as encapsulants are listed below:
●● Ethylene vinyl acetate (EVA) has been used in more modules than any other encapsulant
as it is reasonably priced and readily available as a formulated film for PV with primers,
cross‐link agents and UV stabilizers incorporated into the film itself.
●● Silicones were used in the early days of PV and worked well in the field but were aban‑
doned due to their high costs and because liquid encapsulants were more difficult to use
in manufacturing.
●● Polyolefins are similar to EVA and are also available as formulated sheets. They have
become more popular as a replacement for EVA in recent years.
●● Ionomers were used in the past, especially by Mobil Solar and ASE Americas. They were
typically used in glass/glass constructions but the particular formulation used had prob‑
lems with delamination in the field probably due to less than ideal adhesion to the glass.
Backsheets: As the name implies, the backsheet is the outside material on the back or
non‐sun side of the PV module. The functions of the backsheet include:
●● Protecting the rest of the module from the weather – rain, snow, hail, etc.
●● Screening the materials inside it from UV.
●● Providing protection from the high voltage within the module, so backsheets must have
high resistance and high dielectric strength.
●● Providing protection from accidental exposure to the active components within the mod‑
ule. So, the backsheet must have high‐tensile strength and be scratch resistant.
●● The backsheet is the first line of defense in case the module catches fire so its properties
will impact the fire rating that a module can obtain.
●● The backsheet must provide for secure bonding of junction boxes, connectors, frames
and/or mounting rails.
Backsheets are usually comprised of multi‐layers of material as the different layers pro‑
vide different functions. Often one of the layers (usually the center layer) is a poly(ethylene
terephthalate) (PET) or polyester to provide the dielectric isolation and high resistance.
The outer layer has to provide UV resistance to the layers inside and is most often com‑
posed of a fluoropolymer. The inner most layer must bond well to the encapsulant. One
typical example of a multilayer backsheet is Tedlar/polyester/Tedlar.
Edge Seals: Modules that are constructed with impermeable (or extremely low permeabil‑
ity) front and backsheets designed to protect moisture‐sensitive PV materials, may suffer from
moisture ingress from the sides. Edge Seal materials are low‐diffusivity materials that are
placed around the edges of a module between the impermeable front and backsheets to pre‑
vent moisture ingress. Edge seals were borrowed from the insulated glass industry where they
are used to keep moisture from penetrating between the two panes of glass. In addition to
restricting moisture ingress, edge seal materials must have high electrical resistivity to provide
electrical insulation as frontsheets and backsheets do. To continue to perform these functions
for the lifetime of the module, edge seal materials must remain well adhered to the front and
back sheets of glass. Edge seal materials are usually made of Polyisobutylene and filled with
desiccants to keep moisture from penetrating throughout the useful life of the module.
Frontsheets: Frontsheets must meet all of the same requirements as backsheets, with the
additional requirement of having high optical transmittance over the wavelength range
that solar cells are effective, for example from about 300 nm to 1100 nm for cry‐Si. Glass is
used the most as a frontsheet, but some modules are made using fluoropolymer frontsheets,
particularly Tefzel or Ethylene tetrafluoroethylene (ETFE).
1.4 What Does Reliability Mean for PV Modules?
Since the book is entitled “Reliability of Photovoltaic Modules,” it is important to explain

what reliability means in this context. In the simplest sense, reliability of a product means
that it performs the promised functions for the expected lifetime. For a PV module,
Introduction 13
the promised function is continued production of the specified amount of electricity in a

safe manner. Typically, the product warranty will specify both the amount of power retained
(something like retention of 80% of initial power) and the time period over which it should
retain this power level (often 25 years). So, if a particular module fails to produce this
amount of power or if it becomes unsafe (maybe it has too high a leakage current or it falls
out of the frame), then it is considered a failure. For a particular type of module, the term
reliable would indicate that a certain percentage (maybe something like >95%) are still
operating within the specified power limit throughout the length of the warranty period.
Another word often used to describe PV modules is their durability. The dictionary
defines durability as “the ability to withstand wear, pressure, or damage.” PV modules must
be durable since they are exposed to the stresses of the outdoor environment when
deployed. For modules, durability has come to mean the continued ability to provide out‑
put power over its lifetime. Rather than focusing on failure as reliability does, durability
focuses on maintaining the output power level. As a module ages, it may slowly degrade in
output power. This is usually reported as an annual degradation rate although the rate is
often not linear, but that is the way it is reported for simplicity. So, what does it really mean
if one particular module type deployed at a specific geographic location has a reported
annual degradation rate of 1% after having been deployed at the site for five years? This
really means that the average module output power has dropped by 5% over the five‐year
exposure time. The annual degradation rate is an important factor used in evaluating
potential financial payback from the investment in the PV array. Investors want to know
what annual degradation rate they can expect from the module type they are purchasing.
The final term to discuss in this section is module lifetime. This is a measure of how long
the module will continue to produce power at the specified level in a safe manner. The war‑
ranty typically provides an estimated lifetime which should be the minimum time over
which the product will continue to operate and meet the warrantied power. Once again,
these values are typically 80% output power after 25 years. The real question is how the
customer defines the lifetime. When does the owner of the PV modules decide that they
have degraded too much and need to be retired and perhaps replaced? Experience indicates
that the 80% level may be a reasonable benchmark to use because once modules have
degraded below this power level, they often start having other problems like ground faults,
breakage, etc. So, while there is certainly no consensus on defining module lifetime, this
book will consider the useful life of a module over when the power has decreased to less
than 80% of the original specification. For an array of modules, this would typically mean
that the power of the array or the average power of the modules has dropped to 80% of the
initial specified output power.
1.5 Preview of the Book

This book has been written from a historical perspective to guide the reader through how
the PV industry learned what the failure and degradation modes were, how accelerated
tests were developed to cause the same failures and degradation in the laboratory and then
how these tests were used as tools to guide the design and fabrication of more reliable and
longer‐life modules. The following sections will provide a brief description of what will be
contained within each of the subsequent chapters.
Chapter 2 begins the story with a survey of PV module failure and degradation modes
that have been observed in the field. Pictures are provided for many of these to show what
the failures look like and to provide the reader with a guide for their own module inspec‑
tions. The book starts with failures not because PV modules are unreliable, but because
identifying likely failure modes is the first step in avoiding them. For some of the failure
modes, there is description of the relatively simple fixes that were used to solve some of the
early problems with PV modules and therefore led to more reliable products.
The subject of Chapter 3 is accelerated stress tests (ASTs), describing what they are and
how they have been used in PV to improve module reliability and lifetime. In developing
ASTs, we must cause degradation. The degradation occurring in the AST must be due to the
same failure mechanism we saw outdoors. The work to develop the appropriate AST for the
different module failure modes is based on more than 35 years of experience. The Chapter
concludes with a discussion of various ASTs that have been found to be useful for testing
PV modules.
Chapter 4 introduces the concept of Qualification Tests, which are a set of well‐defined
ASTs developed out of a reliability program. The purpose of qualification testing is to rap‑
idly detect the presence of known failure or degradation modes that may occur in the
intended operating environment. In these tests, the stress levels and durations are limited
so the tests can be completed within a reasonable amount of time and cost. The chapter
starts with a history of the development of Qualification Testing for PV. It then provides a
summary of the testing performed in IEC 61215; the main qualification test standard used
for PV modules. The chapter then discusses how Qualification Tests have been critical to
improving the reliability and durability of PV modules as well as some of the limitations of
the Qualification Tests. The chapter concludes with a discussion of module safety testing
including a summary of IEC 61730, the main safety test standard used for PV modules.
Chapter 5 discusses some of the tools used to better understand what has gone wrong
within a failed or degraded module. So, it presents characterization tools that look to define
what properties of the module (or cells) have degraded and what may have been the cause
of such degradation. Methods presented include, how to analyze the I‐V parameters, meas‑
urement of performance at different irradiances, visual inspection, Infrared (IR) Inspection,
Electroluminescence (EL) Inspection and evaluation of adhesion.
Chapter 6 is about the use of Quality Management Systems in the manufacture of PV
modules. The premise is that to continually build quality modules, the manufacturers
should be using a Quality Management System that have been developed specifically for
PV module manufacturing. The chapter provides the history behind how Quality
Management Systems evolved in PV, indicating how successful this has been but also iden‑
tifying some of the issues with a “do it yourself” system and the need for further improve‑
ments that led to the creation of the International PV Quality Assurance Task Force
(PVQAT) discussed in the Chapter 7.
Chapter 7 tells the story of PVQAT including its creation and the establishment of its
research goals. The three goals include development of improved accelerated stress testing
for PV modules, establishment of a Quality Management System for PV module manufac‑
turing and establishment of a conformity assessment system for PV power plants. The
chapter concludes with a summary of the objectives and activities of each of the PVQAT
Task groups.
Introduction 15
Chapter 8 introduces the concept of Conformity Assessment for PV. The first part of the
chapter discusses development of conformity assessment systems for PV products, mostly
PV modules though some PV products like PV lanterns have also been covered. The second
part discusses the extension of conformity assessment from products to PV systems
deployed in the field and how this required the creation of a whole new Conformity
Assessment organization at IEC called the IEC System for Certification to Standards
Relating to Equipment for Use in Renewable Energy Applications or IECRE for short.
Chapter 9 discusses how to predict service life of PV modules. The chapter starts out by
addressing how to determine the acceleration factors for the different ASTs that are typi‑
cally performed on PV modules. It then discusses the impact of module design and control
of the manufacturing process on module failure rates and how that impacts lifetime predic‑
tions. The third section explains the impact of the weather at the geographic location where
the module is deployed and the type of mounting system used on module degradation and
failure rates and how those impact lifetime predictions. The fourth section talks about the
efforts to get the PV community to agree on one set of extended ASTs that evaluate modules
for wear out. The final section discusses development of a methodology for how a PV mod‑
ule manufacturer could set up a system to predict the lifetime of one of their products.
Chapter 10 shifts the focus to the future. Since PV is a dynamic industry, the technology
and the testing standards are constantly evolving. The first section provides an update on
the changes already in progress for some of the more important module qualification and
safety standards. The second section takes a longer‐range view, discussing how PV module
reliability is likely to change in the future and what sort of accelerated stress testing will be
necessary to validate the quality of the huge volume of modules that will be produced. The
book will end with a brief summary of the status of PV module reliability today.
R
eferences
1 Beetz, B. (2018). https://www.pv‐magazine.com/2018/12/31/14‐pv‐trends‐for‐2019

(Accessed 28 August 2019).
2 McKerracher, C. (2019). https://www.bloomberg.com/professional/blog/transition‐energy‐
transport‐predictions‐2019 (Accessed 28 August 2019).
3 IEA (2019). https://www.iea.org/statistics/electricity (Accessed 28 August 2019).
4 Mints, P. (2019). Photovoltaic Manufacturer Capacity, Shipments, Price & Revenues
2018/2019. SPV Market Research, Report SPV‐Supply7.
5 Williams, R. (1960). Becquerel photovoltaic effect in binary compounds. The Journal of
Chemical Physics 32 (5): 1505–1514.
6 All Nobel Prizes in Physics. https://www.nobelprize.org/prizes/lists/all‐nobel‐prizes‐in‐
physics (Accessed 28 August 2019).
7 Chapin, D.M., Fuller, C.S., and Pearson, G.L. (1954). A new silicon p‐n junction photocell for
converting solar radiation into electrical power. Journal of Applied Physics 25: 676.
8 National Aeronautics and Space Administration. https://nssdc.gsfc.nasa.gov/nmc/
spacecraft/display.action?id=1958‐002B (Accessed 28 August 2019).
9 FSA. https://www2.jpl.nasa.gov/adv_tech/photovol/summary_overview.htm (Accessed 28
August 2019).
10 Varadi, P.F. (2014). Sun Above the Horizon. Pan Sanford Publishing.
11 Varadi, P.F. (2017). Sun Towards High Noon. Pan Sanford Publishing.
12 Wohlgemuth, J. Photovoltaic cells. Kirk‐Othmer Encyclopedia of Chemical Technology
https://doi.org/10.1002/0471238961.16081520070125.a01.pub3.
13 Green, M.A. (1995). Silicon Solar Cells. Bridge Printery Pty Ltd.
14 ChemMatters. https://www.acs.org/content/acs/en/education/resources/highschool/
chemmatters/past‐issues/archive‐2013‐2014/how‐a‐solar‐cell‐works.html (Accessed 28
August 2019).
15 Lindmayer, J. (1976). Semi‐crystalline silicon solar cells. 12th IEEE PVSC in Los Angeles,
Usa (15–18 November 1976).
16 Sunpower. https://us.sunpower.com/products/solar‐panels (Accessed 28 August 2019).
17 NREL. https://www.nrel.gov/pv/module‐efficiency.html (Accessed 28 August 2019).
18 Panasonic. https://na.panasonic.com/us/energy‐solutions/solar/solar‐panels/n330‐
photovoltaic‐module‐hitr‐40mm (Accessed 28 August 2019).
19 REC (2014). http://www.recgroup.com/sites/default/files/documents/whitepaper_perc.pdf
20 Fuhs, M. and Sieg, M. (2016). http://www.pv‐magazine.com/news/details/beitrag/
solarworld‐hits‐22‐perc‐efficiency_100022790/#axzz4Kzjs1p29 (Accessed 28 August 2019).
21 Carlson, D. (2002). Accelerated Corrosion Testing of Tin Oxide Coated Glass. Proceedings
of NREL Thin Film Module Reliability National Team Meeting, Colorado, USA (6–10 May
2002).
22 First Solar (2019). http://www.firstsolar.com/‐/media/First‐Solar/Technical‐Documents/
Series‐6‐Datasheets/Series‐6‐Datasheet.ashx (Accessed 28 August 2019).
23 Solar Frontier. http://www.solar‐frontier.com/eng/solutions/products/index.html
17
Module Failure Modes
Initial product reliability assessments are based on the environment in which the product
is going to be exposed, the outdoor terrestrial environment in the case of photovoltaic (PV)
modules. Such an analysis can provide some clues as to the level of stresses to be encoun‑
tered. The earliest module manufacturers understood that the modules would be required
to endure exposure to the weather (rain, hail, and snow), high temperatures, UV, humidity,
and thermal cycling. However, in most cases, the stresses were underestimated. The first
generation of terrestrial PV modules was not very reliable nor did the modules survive for
very long in the field. However, this first generation of product served an important func‑
tion in that they failed in the field (often very quickly) allowing for subsequent analysis and
development of accelerated stress tests to be described in Chapter 3. It wasn’t until the
product designs could be tested using the accelerated stress tests that reliability was signifi‑
cantly improved.
The study of PV module reliability starts with the identification of field failures. Over
the years, there have been many reports of module field failures. Table 2.1 provides a list
of field failures observed in crystalline‐Si PV modules. The vast majority of deployed PV
modules have been crystalline silicon so this technology has the most history. The follow‑
ing sections will discuss the details of each of these failure modes, with emphasis on what
causes the failure, what construction or material selections make it worse or better and
how the industry eventually modified their products to improve reliability and lifetime.
Each subsection will also mention whether this is a failure mode likely to be observed
for thin‐film modules. A final section in the chapter will address failure modes specific to
thin‐film modules.
2.1 Broken Interconnects
The interconnect ribbons that connect solar cells together can break due to stress caused by
thermal expansion and contraction or due to repeated mechanical stress. Many of the earli‑
est modules only had one interconnect ribbon per cell so they suffered open circuits when
an interconnect broke. Redundancy of interconnect ribbons was introduced to keep mod‑
ules from failing prematurely. Figure 2.1 shows an Electroluminescence (EL) picture of a
module with multiple broken interconnect ribbons. (EL is a useful tool for inspecting
Photovoltaic Module Reliability, First Edition. John H. Wohlgemuth.

© 2020 John Wiley & Sons Ltd. Published 2020 by John Wiley & Sons Ltd.
Table 2.1 Failure modes observed for c-Si modules.
Broken interconnects
Broken/cracked cells and snail trails
Corrosion of cells, metals and connectors
Delamination/loss of adhesion between layers
Loss of elastomeric properties of encapsulant or backsheet
Encapsulant discoloration
Solder bond failures
Broken glass
Glass corrosion
Reverse bias Hot Spots
Ground faults due to breakdown of insulation package
Junction box and module connection failures
Structural failures
Bypass Diode failures
Open circuiting leading to arcing
Potential Induced Degradation (PID)
Figure 2.1 Electroluminescence (EL) picture of module with multiple broken interconnect ribbons.
Module Failure Modes 19
Figure 2.2 Magnified picture of interconnect ribbon with a crack [1]. Source: reprinted from
author’s PVSC article.
PV modules and will be described in more detail in Chapter 5). The dark areas in the pic‑
ture are the regions where broken interconnect ribbons are impeding the current flow.
Since the interconnects in these regions are broken, the current is no longer efficiently col‑
lected from those regions. Those regions are not producing as much energy as they should
be. This module has lost approximately 10% of its initial STC power (that is peak power
measured at Standard Test Conditions defined as 1000 W/m2 irradiance with the reference
AM1.5 solar spectrum at a junction temperature of 25 °C.)
Figure 2.2 shows a picture of a broken interconnect ribbon. The module from which this
ribbon was taken also had multiple dark areas in its EL picture. This particular type of
module suffered from broken interconnect ribbons because of a design flaw (the ribbons
were soldered to the cell edge on both front and back so there was not enough free intercon‑
nect ribbon). Many failures of interconnect ribbons are due to poor design and usually take
a significant amount of time in the field (at least five years) to have a major impact on
module performance. There are a number of factors in module design that facilitate inter‑
connect breakage including:
●● Substrates (or superstrates) with high linear thermal expansion coefficients: If the
substrate or superstrate has a high linear thermal expansion coefficient, it will expand
and contract more during normal operation resulting in extra stress on the intercon‑
nect ribbons. Values for the linear thermal expansion coefficient of some materials
relevant to PV are given in Table 2.2 [2]. Substrates made of plastics, epoxy board and
even metals have high thermal expansion coefficients which stresses the interconnect
ribbons during outdoor exposure causing some of the ribbons to break. Many of the
earliest module designs including some of the original Jet Propulsion Laboratory
(JPL)Block I modules had fiber board substrates. As will be described in Chapter 4,
these modules failed open circuited in a few years, with those deployed in desert
Table 2.2 Linear thermal expansion coefficient of selected

materials relevant to PV [2].
Material m/mK (* 10−6)
Aluminum 21–24
Copper 16–16.7
Epoxy without fillers 45–65
Epoxy wit glass fillers 36
EVA 180
Germanium 18.4
Glass 9
Plastics 40–120
PET or polyster 59.4
Silicon 3–5
Solder (tin‑lead) 25
Stainless Steel 10–17
c limates failing the fastest. For almost four decades now, the vast majority of PV mod‑
ules have used the glass superstrate design because glass has one of the lowest ther‑
mal expansion coefficients.
●● Larger cells: Larger cells expand and contract more than smaller cells and therefore exert
more stress on the interconnect ribbons.
●● Thicker ribbons: To reduce the series resistance losses it is advantageous to use the thick‑
est interconnect ribbon practical. However, thicker ribbons are stiffer, resulting in the
ribbon itself breaking more easily or in it stressing the solder bonds resulting in their
failure (See Section 2.7).
●● Stiffer ribbon: If a ribbon is soft and deforms when stressed, it is much less likely to break
in a module. Over the years, the ability to make softer ribbons has dramatically improved
the reliability of PV modules in terms of ribbon breakage, solder bond failure and even
cell breakage.
●● Kinks in ribbon: In the early days of PV, it was considered good practice to put a stress
relief loop in the interconnect ribbon. However, experience showed that all these loops
did was concentrate the stress in the loop instead of along the whole length of the free
ribbon. The ribbons tended to fail at the loop.
●● Not enough free ribbon between solder bonds on adjacent cells: It is the free region
between bonds on the two cells that must take up the changes in spacing between cells
caused by the thermal cycling. Even though copper (or any other metal for that matter)
has a much higher thermal expansion coefficient than Si and glass, if there are only a
few millimeters of free ribbon between solder bonds on adjacent cells, there isn’t enough
material to provide the necessary expansion and contraction. The ribbon in Figure 2.2
came from a module where the ribbons were soldered to the ends of the bus bars on
both front and back of the cells. These types of modules suffered from multiple broken
ribbons and excess power loss.
The improvements made in module design and the improved properties of the materials
used for interconnect ribbons has dramatically reduced the instances of interconnect
breakage in modern PV modules.
2.2 Broken/Cracked Cells and Snail Trails
Crystalline‐Si cells can (and will) break due to mechanical and thermal stresses. Early
modules suffered open circuits due to broken cells since there was only one electrical
attachment point per cell for each polarity. This led to the development of the tab across
design, where tabs run the full length of the cell and are soldered in multiple places along
the cell length on both sides. This design has been used for most crystalline silicon modules
since the 1980s. With the tab across design, a small amount of cell breakage doesn’t often
lead to power loss, but a large amount of breakage can.
Electroluminescence (EL) is an excellent tool for observing cell breakage. A description
of how EL works will be provided in Chapter 5. Figure 2.3 shows an EL picture of a module
with extensive cell breakage. The dark lines in the picture indicate cracks in the Si. However,
most of these cracks just appear as thin gray lines. These do not impact the performance of
the solar cell. The dark areas in the EL picture (for example, the dark area in the fifth cell
from the left in the top row) indicate regions where there is no EL signal because those
parts of the cells are disconnected from the module circuit and are not providing power.
Such regions not only reduce the active area of the cell itself but they also lead to cell mis‑
match as a cell with a significant reduction in area cannot produce as much current as a full
cell. Such a cell will act as a resistive load to the current produced by the remainder of the
module. The module in Figure 2.3 was down 9% in peak power. We were not sure why this
module suffered from broken cells while the remainder of the modules in the system had
few, if any, broken cells and still provided 100% of rated power.
Figure 2.3 Electroluminescence (EL) picture of module with many broken cells [3]. Source: reprinted
from author’s PVSC article.
There are multiple reasons why the cells in modules crack. Some of them are related to
construction of the module while others are caused by handling or improper mounting.
Some of the factors to consider are:
●● Cell thickness: In the 1980s, most cry‐Si solar cells were 300 μm or more in thickness.
Today, most cry‐Si cells are less than 200 μm in thickness. The trend is for continued
reduction in cell thickness to reduce the amount of Si per kW‐hr of electricity produced.
Thinner Si wafers break more easily than thicker ones. So we expect today’s cry‐Si mod‑
ules to suffer from more broken cells than older modules did.
●● Cell size: In the late 1970s, the terrestrial PV industry transitioned from using 2.25 in. round
cells to 3 in. round cells. Today, the standard cry‐Si cell is 15.6 cm by 15.6 cm or 6 in. by 6 in.
(either fully‐square polycrystalline Si or pseudo‐square single crystal Si). The breakage
rate of larger cells is higher than for smaller cells everything else being equivalent.
●● Module size: In the late 1970s, modules were typically fabricated using 36 2.25 in. cells
with an area of 81 square inches or 0.05 square meters. Today, standard commercial mod‑
ules with 72 15.6 by 15.6 cells have an area of 3000 in.2 or 2 m2. Most such modules still
use 3.2 mm thick tempered glass for the superstrate. The new modules deflect more
under wind and snow load. Therefore, there is significantly more stress imparted to the
cells in today’s larger modules.
●● Crystallinity: Single‐crystal wafers have cleave planes along which the cells crack easily.
If the cleave plane is oriented along the bus bar the cell is more likely to crack.
●● Pre‐stressed or chipped cells: If the solar cells have built in stress say due to crystal
growth problems or microcracks or chips caused by the processing, they will be more
prone to breakage initiated at the site of the damage.
●● Poor packaging of modules during shipment: Shipping can be hard on the cells within
the modules. It is not unusual to observe a broken module or two at the bottom of a pallet
of modules delivered to a work site. I have had personal experience shipping perfectly
good modules to a test laboratory and have them arrive with multiple broken cells. Some
module manufacturers test their packaging methods, but use module breakage as the
pass criteria not cell breakage within the modules. IEC has now published a transporta‑
tion standard, IEC 62759‐1 [4] which not only stresses the modules within their transpor‑
tation packaging but also stresses the modules afterwards to determine whether cells
have been damaged during the transportation testing.
●● Poor handling of modules during installation: The cells within modules can also be broken
during installation. Everything from walking on modules to dropping modules to torque‑
ing them down too hard onto the support structure can break cells (or even glass) or make
the cells susceptible to future breakage. IEC has recently published a guideline for system
installation, IEC 63049 [5] in an effort to minimize damage during installation.
Particularly because of the latter two factors, many system installers now use EL to
screen the installed modules to determine whether there is significant cell breakage within
their array.
Snail trails are a relatively new phenomenon that occurs as a result of broken or cracked
cells. A typical example is shown in Figure 2.4. In the region around a crack in a cell,
the intrusion of moisture and/or air results in corrosion of the Ag in the screen‐printed
metallization [6]. This can occur quickly in the field, often during the first year of
Figure 2.4 Cells with snail trails.
eployment. Early Ethylene Vinyl Acetate (EVA) modules did not exhibit snail trails, but
d
with the introduction of large numbers of module manufacturers and material suppliers
from China in the 2000s, this phenomenon began to be observed.
Fraunhofer ISE has observed that snail trail products (the corrosion on the grid lines)
include silver phosphate (Ag3PO4), silver sulfide (Ag2S), silver carbonate (Ag2CO3) and sil‑
ver acetate (Ag2[CH3COO]2) [7]. In their work, the source of the impurities was evaluated
and subsequent accelerated testing was performed to determine how the corrosion might
progress. Some of their main findings were:
●● Silver acetate growth resulted from interaction with the EVA and was accelerated by UV
exposure. There is no limit on its formation during a PV module’s lifetime
●● Silver sulfide results from interaction with sulfurous‐containing antioxidants probably
from the backsheet and was accelerated by damp heat. It is not clear whether this could
be a major cause of snail trails, but it can be reduced by using a backsheet with lower
oxygen transmission rate (OTR).
●● Silver carbonate is formed via a reaction with metal oxides on the grid lines and it disap‑
pears after exposure to damp heat. Since there are limited metal oxides and it disappears
from humidity exposure its formation is limited.
●● Silver phosphate forms from a reaction with phosphorous‐containing antioxidants in the
EVA and does not grow under damp heat or UV exposure. Its formation is most likely
limited by the amount of phosphorous available.
The finding that silver acetate is the likely culprit is consistent with the finding that dif‑
ferent EVA’s have vastly different amounts of free acetate. It would appear that those with
a large amount of free acetate are much more likely to develop snail trails than EVA formu‑
lations with little free acetate.
One of the remaining open questions is whether snail trails are just cosmetic or if the
corrosion of the grid metallization will ultimately lead to power loss. While some module
manufacturers are telling customers it is only cosmetic, most are doing all they can to
reduce the occurrence of snail trails in their products via their selection of EVA and
backsheet materials.
2.3 Delamination
Delamination refers to separation of different layers within the module package. It can
occur between a variety of different layers within the package. The impact on module per‑
formance and lifetime is dependent on which surface delaminates. Let’s look at several
different types of delamination that have been observed in the field.
●● Delamination of the encapsulant from the glass: Encapsulant‐glass delamination has
been observed to occur in specific module types. National Renewable Energy Laboratory
(NREL) has reported on such delamination in Mobil Solar and later ASE America mod‑
ules at a number of sites with a variety of different climates (see Figure 2.5) [8]. These
modules were fabricated with a non‐EVA encapsulant which did not have as high an
initial adhesion to glass as other encapsulants like EVA. This material tended to delami‑
nate from the front glass either directly over the junction box or in the corners. This
problem was observed at numerous sites often with every module exhibiting delamina‑
tion directly over the junction box. However, observations at the Springerville, AZ site
indicated that modules fabricated before September, 2001 are delaminating while those
fabricated after September, 2001 are not delaminating. This suggests that the module
manufacturer likely made a material or process change in this timeframe. The area over
the junction box runs hotter than the rest of the module so it is likely that the higher
temperature plays a role in the delamination. It is also possible that the mechanical stress
of the junction box actually helps to pull the encapsulant off of the glass.
Delamination between the glass and encapsulant may not indicate as severe a problem
as delamination between other layers within the module. It all depends upon whether
the delamination results in exposure of the cells or electric circuit to the elements. If the
encapsulant stays well attached to the cells and the bus bars, it may continue to protect
them from moisture. As in the case of the module shown in Figure 2.5, there is a cos‑
metic defect, but the module has continued to operate adequately.
Figure 2.5 Delamination over the Junction Box of the encapsulant from front glass for ASE Module
[8]. Source: reprinted from author’s PVSC article with the permission of David Miller of NREL.
Most EVA‐encapsulated modules do not exhibit this type of encapsulant‐glass delami‑

nation. To help us understand why, Nick Bosco at NREL utilized a recently developed
cantilever beam method to measure the adhesion between EVA and glass in old Arco
Solar modules – comparing the results obtained from modules exposed in the field for
27 years with those obtained from a module continually stored in a shed for the same
amount of time [9]. For the sample that had been in a shed for the whole time, the weak‑
est interface was the EVA‐cell interface with a debond energy (adhesion) of between 800
and 1000 J/m2. For the sample that was exposed in the field to sunlight for 27 years, the
weakest interface was the glass‐EVA interface with an adhesion of between 200 and
300 J/m2. When measuring the adhesion of EVA to glass in new samples, Nick typically
obtains values of around 2000 J/m2. So the front surface adhesion has dropped by nearly
an order of magnitude. However, these modules show no evidence of delamination, indi‑
cating that 200 and 300 J/m2 is adequate for module survival for more than 25 years.
In a recent set of experiments undertaken by PVQAT Task Group 5, the attachment
strength between encapsulant and glass was evaluated as a function of UV exposure
using various temperatures and humidity levels [10]. While these accelerated exposures
did reduce the measured strength and adhesion of EVA to glass by more than 50%, they
remained at levels that are still adequate to prevent delamination. This result shows that
while EVA to glass adhesion will be reduced during the initial stages of field deployment,
it may not continue to degrade throughout the lifetime of the module. This experiment
will be described in more detail in Chapter 7.
●● Delamination of the encapsulant from the cells: For EVA‐based modules, the more com‑
monly observed delamination is between the EVA and the cell surface, selectively occur‑
ring around the interconnect ribbons as shown in Figure 2.6 for a Siemens Solar module.
This failure mode tends to occur in a large fraction of the modules in certain arrays.
Figure 2.6 Delamination between encapsulant and cell surface in Siemens Solar Module [11].
Source: reprinted from author’s PVSC article with the permission of David Miller of NREL.
The module in Figure 2.6 was deployed in Florida [11], where all of the modules in the
array exhibited a similar level of delamination. However, there are numerous Siemens
Solar arrays with little or no delamination. It is interesting to note that such delamination
appears in multiple installations from Arco Solar, Siemens Solar and even Shell Solar
indicating that the source of the delamination was part of the module design or manufac‑
turing process for many years. One possibility is related to the lamination process that was
used. Arco, Siemens and Shell all continued to use the standard cure EVA formulation
long after most other companies had switched to fast cure EVA. Their lamination process
consisted of a short cycle in the laminator followed by a much longer cure in an oven. This
process appeared to work well to cure the EVA without outgassing, but it may not have
been adequate to activate the silane prime resulting in occasional adhesion problems.
This type of delamination is more serious than some other types of delamination
because the grid metallization is exposed and can corrode. Therefore, we expect and do
see significant power loss due to increased series resistance from such modules.
This type of delamination is not observed during damp heat, humidity freeze or
thermal cycling from the standard qualification testing discussed in Chapter 4. There is
evidence that this phenomenon is not strictly a delamination but may be caused by
generation of gas at the metallization surface. This will be discussed in more detail in
Chapter 7 in conjunction with the PVQAT effort.
●● Delamination of the backsheet: Delamination has also been observed between the
encapsulant and the backsheet or between layers of the backsheet itself. It is often hard
to distinguish between these two failure modes without destructive failure analysis.
There are multiple examples where one or more layers of a backsheet have completely
delaminated from the inner layer that is attached to the encapsulant. This type of
backsheet delamination may lead to potential safety concerns as the backsheet ensures
electrical isolation between the system voltage within the module and the outside world.
However, backsheets do not play as important a role in terms of PV performance as
encapsulants. We have seen PV modules in the field with their backsheets completely
delaminated from the back encapsulant that are still producing power in accordance
with their warranty.
2.4 Corrosion of Cell Metallization
Moisture induced corrosion of cell metallization is usually associated with significant

power loss and premature failure. Corrosion observed in fielded modules is often associ‑
ated with delamination of the encapsulant. Then water can condense in the void caused by
the delamination. Liquid water then corrodes the cell metallization. One example is shown
in Figure 2.7 [10]. This is an extreme case where the adhesion between the encapsulant and
the cell surface has failed across larger areas of the module.
Most of the modules in this system showed similar levels of delamination and subse‑
quent corrosion of the metallization. Needless to say, these modules are no longer produc‑
ing any power. The primary cause of delamination and then corrosion appears to be the use
of EVA without primer. Only the glass was primed before lamination so the EVA did not
adhere well to the cells.
Figure 2.7 Corrosion of the front metallization on solar cells [11]. Source: reprinted from author’s
PVSC paper.
Figure 2.8 Corrosion on corner of glass–glass module [11]. Source: reprinted from author’s
PVSC paper.
A second (less destructive) type of corrosion is shown in Figure 2.8 [11]. This is a glass–
glass module where the non‐EVA encapsulant did not adhere very well to the glass. Almost
every night, water would condense in the delamination voids in the corner of this module.
Ultimately, that liquid water corrodes the cells metallization locally. This array was actually
deployed in Arizona so it shows that a humid environment is not required to corrosion to
occur. In this particular case, less than 10% of the modules exhibited any corrosion at all.
The main way to prevent corrosion of the cell metallization is to prevent encapsulant
delamination. Long‐term adhesion of the encapsulant to all of the other components of the
module is critical. The second most important design is the use of a moisture/corrosion
Figure 2.9 Corrosion of cells in Polyvinyl butyral (PVB) Module [12]. Source: reprinted from PVSC
with permission of Ronald Ross of the Jet Propulsion Laboratory.
resistant metallization system. The screen printed pastes used today are much more toler‑
ant of moisture than the materials that were available 20 years ago. When damp heat tests
were first introduced, modules had trouble meeting the pass/fail criteria of a 5% power loss.
Today, most commercial cry‐Si modules exhibit no power loss after 1000 hours of damp
heat testing.
The final design protection against metallization corrosion is the use of an encapsu‑
lant that retains low ionic conductivity when wet. In the early 1980s, module manufac‑
turers tried to use Polyvinyl butyral (PVB) as an encapsulant as it was (and still is) used
extensively in the laminated glass industry. The problem with PVB is that its ionic con‑
ductivity increases in higher humidity while that of EVA does not [12]. Cry‐Si modules
made with screen‐printed metallization and PVB encapsulant failed in the field as the
Ag from the screen‐printed metallization migrated into the encapsulation as shown in
Figure 2.9.
2.5 Encapsulant Discoloration
Discoloration of EVA encapsulant has been observed as a major cause of PV module power
degradation. A typical example of the observed discoloration is shown in Figure 2.10. This
module was fabricated using the original STR EVA formulation A9918 (standard cure EVA)
and deployed on an open rack mount system at the STAR facility in Tempe, AZ as part of
the control in an experiment to evaluate the susceptibility of different EVA formulations to
discolor [13].
The initial observation and worst reported case of EVA discoloration occurred in the
early 1990s at the Arco Solar Carrizo Plains site in California [14]. The Carrizo Plains sys‑
tem employed mirror enhancement so the modules were exposed to more UV light and
Figure 2.10 Example of discolored Ethylene Vinyl Acetate (EVA) formulation A9918 in the STR
experiment at the Tempe site [13]. Source: reprinted from author’s PVSC article with the permission
of David Miller of NREL.
higher temperatures than in standard open rack mounted terrestrial systems. The Carrizo
system suffered severe power loss, originally thought to be entirely due to EVA browning.
However, the author’s team at Solarex was able to restore much of the lost power to several
modules by repairing back solder bonds [15]. Much of the power loss in the system was due
to solder bond issues, but the discoloration in these modules did lead to power loss in the
order of 10–20%.
Additional observations confirmed both that the A9918 formulation of EVA did discolor
and that the discoloration was caused by UV exposure at higher temperatures. At NREL,
we had the opportunity to evaluate modules with A9918 EVA after 27 years of exposure at
SMUD (Sacramento Municipal Utility District). Current losses of 10–12% were measured.
This should be typical of the level of degradation expected for long‐term exposure of this
type of module. So it is not enough to invoke the warranty (usually based on 20% degrada‑
tion) but it is a significant reduction allowing for much smaller reductions from all degra‑
dation modes before reaching the warranty limit.
Figure 2.11 provided by Govindasamy Tamizhmani of ASU shows images of two mod‑
ules from the same module manufacturer where the module in images (Figure 2.11a) has
been stored indoors (not exposed to the terrestrial environment) and the module in images
(Figure 2.11b) has been exposed in the field for 10 years in Florida. Let’s compare the two
sets of pictures:
●● Visual Image: In the picture of the unexposed module, no visual defects are observed. In
the picture of the exposed module, there is no major discoloration, but with closer inspec‑
tion some discoloration is observable in the centers of the cells.
●● EL (Electroluminescence Image): In the EL picture of the unexposed module, a few
darker spots are visible across the fingers. In the EL picture of the exposed module, the
EL intensity is lower than for the unexposed module. Bright spots at cell interconnects
probably correspond to solder bond degradation which will be discussed in Section 2.6.
Visual image EL image IR image UVF image
Figure 2.11a Characterization of unexposed module. Source: images provided by Govindasamy

Tamizhmani of ASU.
Visual image EL image IR image UVF image
UVF (close up shot)

Cell center → browning
Cell peripherals → clear

due to photobleaching
mechanism
Figure 2.11b Characterization of module after 10 years of field exposure in Florida. Source: images
provided by Govindasamy Tamizhmani of ASU.
●● IR (Infrared) Images (See Section 5.5 for a discussion of using IR as a tool to evaluate
changes in PV modules.): The IR image of the unexposed module has a uniform
response with no hot spots. The IR image of the exposed module has hot spots in the
same places that the EL image had bright spots, namely where solder bond degrada‑
tion has occurred.
●● UVF (UV Florescence) Images (See Ref. [16] for a description of how to use UVF as a
tool to evaluate degradation in PV encapsulants.): The UVF image of the unexposed
module shows no signal from the encapsulant. The UVF image of the exposed module
shows a bright signal from the center of the cells with almost no signal from the edges
of the cell. The bright areas indicate degradation of EVA material in the center of the
cells, exactly where we see discoloration.
STR was able to replicate the discoloration and subsequent loss in short‐circuit current
via exposure to UV in a Weather‐Ometer [17]. The Springborn Testing & Research Inc.
(STR) work demonstrated that use of 15295 EVA formulation and cerium oxide in the glass
achieves a combined reduction in the discoloration rate by a factor of 12.5 from the A9918
with standard low‐iron glass [17, 18]. STR’s experiments with the different additives within
the A9918 package determined that the discoloration was not related to the EVA resin itself
but was related to the additives, most importantly, the antioxidant (Naugard P) and its
interaction with the cross‐linking agent (Lupersol 101).
In the 1990s, the immediate solution for the PV industry was to use fast‐cure formula‑
tions with cerium oxide glass for a superstrate. Over time, EVA manufacturers developed
better formulations that virtually eliminated the discoloration problem. However, as we
will see later in Chapter 6, loss of short‐circuit current is still the dominate mechanism for
power degradation in fielded cry‐Si modules [19].
2.6 Failure of Electrical Bonds Particularly Solder Bonds
One of the major failure modes observed in the field is failure of the electrical bonds par‑
ticularly solder bonds within the module. Electrical bonds can fail due to the stresses
induced by thermal cycling or mechanical vibrations. PV modules are particularly
susceptible to this because of the high currents flowing through the bonds during normal
operation. So the modules themselves are heated by the sun while the bonds have an
additional heat load from the current flow.
Early modules typically had only one solder bond at each end of the interconnect ribbon
and only one ribbon per cell so failure of one solder bond resulted in an open circuit failure
of the whole module. The tab across design used in most cry‐Si modules results in multiple
solder bonds on both the front and back of each cell providing redundancy for long‐term
survival in the field. Even today, non‐cell solder bonds often have little or no redundancy so
failure of one or two of these bonds can lead to the drop out of a cell string, a whole module
or even a whole string of modules. Such single‐point failures like that shown in Figure 2.12
are often workmanship related.
Most bus bar to cell solder bonds are made using automated equipment. In this case,
process control and maintenance of the equipment is critical to achieving quality bonds.
Since there is significant redundancy, failure of a few percent of those bonds will not affect
long term performance. However, there are occasionally examples of large‐scale failures
within a product as shown in Figure 2.13. In this case, new production equipment was
utilized, but it turned out the soldering equipment was not producing large enough solder
bonds. This example will be discussed further in Chapter 4 as this particular failure mode
required a change in the qualification testing method.
Figure 2.12 Example of single point solder bond failure on bus bar [1]. Source: reprinted from
author’s PVSC article.
Figure 2.13 Example of Multiple Solder Bond Failures on One Cell [20]. Source: reprinted from
author’s EU PVSEC paper.
Over the years, the PV industry has made great strides in its efforts to alleviate solder
bond failures including:
●● Utilize multiple solder bonds on each tabbing ribbon.
●● Utilize softer ribbon so there is less stress on the bonds themselves.
●● Perform preventive maintenance on the soldering equipment to ensure a controlled
process.
●● Perform periodic pull tests to ensure quality of solder bonds being made.
●● Perform thermal cycle tests well beyond the 200 cycles from the qualification tests.
●● Implement training and QA inspections to ensure that non‐cell solder bonds are being
fabricated correctly.
●● In critical areas (like termination wires) use both solder and mechanical connections.
2.7 Glass Breakage
We are all familiar with the fact that glass breaks if stressed enough. Heat strengthening or
tempering does make glass stronger and less susceptible to breakage. The glass in cry‐Si
modules is usually tempered or heat strengthened. Such glass is usually strong enough to
withstand the bending and flexing encountered during normal operation. However, there
are still instances where this stronger glass breaks, including:
●● High impact like a rock or a bullet will break it. Can almost always identify spot where
the object hit. (See Figure 2.14.)
●● Failure of or misuse of the support structure can lead to glass breakage. Pictures of this
type of breakage will be shown in Section 2.10 on Structural Failures.
●● High temperature (hot spot or arc) can also break glass. (See Figure 2.15 where
you can readily see that the discolored area at the bus bar solder joint is the source
of the break.)
●● Improper handling, shipping and installation can also result in broken glass.
It is clear from both Figures 2.14 and 2.15, that when tempered glass breaks, the whole
piece shatters into small fragments at the same time. Most of the time, a cry‐Si module
continues to work for a time after the glass breaks. However, over time the module suffers
stresses like thermal cycling, wind loading and penetration of water to cause corrosion.
Once the glass is broken the module lifetime is limited.
Annealed glass (similar in strength to the windows in your house) is not as strong as heat
strengthened or tempered glass and the break grows across the surface in a random pat‑
tern. Thin‐film modules often use annealed glass because the temperatures required for
Figure 2.14 Module with tempered glass broken by impact.

Figure 2.15 Module with

tempered glass broken due to
overheating at bad solder bond.
depositing the thin‐film semiconductors on the glass are high enough to anneal the heat
strengthening out of the glass. Annealed glass breaks from the same stresses listed above
for heat strengthened or tempered glass, but can also break due to:
●●Thermal gradients: This is particularly important for thin‐film PV because they are
designed to absorb most of the light (not let it through like most glass). So shading such
as provided by partial snow cover can result in thermal gradients that are large enough
to break the glass. Solarex provided several kilowatts of a‐Si Millenia modules to NREL
for deployment in an array on the roof of their Outdoor Test Facility. The array was
doing fine until there was a snow storm followed by a bright sunny day. The following
week, it was observed that every single one of the
Millenia modules had cracked glass. At Solarex,
we then performed some experiments and vali‑
dated that the annealed glass of the Millenia
modules would crack with a temperature gradient
in excess of 25 ° C from center to edge.
● Stress induced during processing (lamination):

Lamination can result in edge pinch where the
lamination process forces some of the encap‑
sulant out of the corners and deforms the cor‑
ners of the glass inward. Because this glass is
under stress it tends to break soon after
deployment.
● Stress from the mounting system: Since the
annealed glass is not as strong as the tempered
glass, it can break if installed under stress.
● Stress from handling: Any defects introduced into
the glass during handling such as edge chips will
ultimately result in breakage starting at the site of
Figure 2.16 Breakage of an annealed the defects.
glass thin-film module [20]. Source:
reprinted from author’s EU PVSEC paper Figure 2.16 shows a picture of a broken thin‐film
with permission of David Miller. module made with annealed glass.
Another random document with
no related content on Scribd:
ought to wear over it a short bed-gown reaching to the hips, and have
on a flannel petticoat to meet it, and then she should put on a
dressing-gown over all. If it be winter, the dressing-gown had better
either be composed of flannel or be lined with that material. The
stays must not be worn, as they would interfere with the progress of
the labor.
509. The valances of the bed, and the carpet, and the curtains at
the foot of the bed, had better all be removed; they are only in the
way, and may get soiled and spoiled.
510. “The guarding of the bed.”—This is done in the following way:
Cover the right side of the bed (as the patient will have to lie on her
left side), with a large piece—a yard and a half square—of waterproof
cloth, or bed-sheeting, as it is sometimes called, which is sold for the
purpose;[93] over this, folded sheets ought to be placed. If a
waterproof cloth cannot be procured, an oil-cloth table-cover will
answer the purpose. Either of the above plans will effectually protect
the bed from injury.
511. The lying-in room should be kept not hot, but comfortably
warm; if the temperature of the room be high, the patient will
become irritable, feverish, and restless.
512. Every now and then, in order to change the air, let the door of
the room be left ajar; and if, in the early periods of the labor, she
should retire for awhile to the drawing-room, let the lying-in room
window be thrown wide open, so as to thoroughly ventilate the
apartment, and to make it fresh and sweet on her return. If the
weather be very warm, the lower sash of the window may for a few
inches be opened. It is wonderful how refreshing to the spirits, and
how strengthening to the frame, a well-ventilated room is to a lying-
in patient.
513. Many attendants are not only unnecessary but injurious. They
excite and flurry the patient, they cause noise and confusion, and rob
the air of its purity. One lady friend besides the doctor and the
monthly nurse is all that is needed.
514. In making the selection of a friend, care should be taken that
she is the mother of a family, that she is kind-hearted and self-
possessed, and of a cheerful turn of mind. At these times all
“chatterers,” “croakers,” and “potterers” ought to be carefully
excluded from the lying-in room. No conversation of a depressing
character should for one moment be allowed. Nurses and friends
who are in the habit of telling of bad cases that have occurred in their
experience must be avoided as the plague. If nurses have had bad
cases, many of them have probably been of their own making; such
nurses, therefore, ought on every account to be shunned.
515. During the progress of the labor, boisterous and noisy
conversation ought never to be permitted; it only irritates and excites
the patient. Although boisterous merriment is bad, yet at such times
quiet, cheerful, and agreeable conversation is beneficial.
516. A mother on these occasions is often present; but of all
persons she is the most unsuitable, as, from her maternal anxiety,
she tends rather to depress than to cheer her daughter. Though the
mother ought not to be in the room, it is, if practicable, desirable that
she should be in the house. The patient, in the generality of cases,
derives comfort from the knowledge of her mother being so near at
hand.
517. Another preparation for labor is to soothe her mind by telling
her of the usual safety of confinements, and by assuring her that, in
the generality of instances, it is a natural process; and that all she has
to do is to keep up her spirits, to adhere strictly to the rules of her
doctor, and she will do well.
518. Tell her that “sweet is pleasure after pain;”[94] tell her, too, of
the exquisite happiness and joy she will feel as soon as the labor is
over, as perhaps the greatest thrill of delight a woman ever
experiences in this world is when her babe is first born. She, as if by
magic, forgets all the sorrow and suffering she has endured. “A
woman when she is in travail hath sorrow, because her hour is come;
but as soon as she is delivered of the child, she remembereth no
more the anguish, for joy that a man is born into the world.”[95]
Keble, in the Christian Year, well observes:
“Mysterious to all thought,
A mother’s prime of bliss,
When to her eager lips is brought
Her infant’s thrilling kiss.”
Rogers, too, in referring to this interesting event, sweetly sings:

“The hour arrives, the moment wished and feared;
The child is born, by many a pang endeared!
And now the mother’s ear has caught his cry—
Oh! grant the cherub to her asking eye!
He comes—she clasps him; to her bosom pressed,
He drinks the balm of life, and drops to rest.”
519. The doctor, too, will be able to administer comfort to her

when he has “tried a pain” or has “taken a pain,” as it is called, and
when he can assure her that it “is all right and straight”—that is to
say, that the child is presenting in the most favorable position, and
that everything is progressing satisfactorily. He will, moreover, be
able to inform her of the probable duration of the labor.
520. Let me in this place urge upon the patient the importance of
her allowing the medical man to inquire fully into her state. She may
depend upon it that this inquiry will be conducted in the most
delicate manner. If there be anything wrong in the labor, it is in the
early stage, and before the “waters have broken,” that the most good
can be done. If a proper examination be not allowed to the medical
man whenever he deems it right and proper (and a judicious doctor
will do it as seldom as he can), her life, and perhaps that of her child,
might pay the penalty of such false delicacy.
521. French brandy, in case it is wanted, ought always to be in the
house; but let me impress upon the minds of the attendants the
importance of withholding it, unless it be ordered by the doctor, from
a lying-in woman. Numbers have fallen victims to brandy being
indiscriminately given. I am of opinion that the great caution which
is now adopted in giving spirits to women in labor is one reason,
among others, of the great safety of the confinements of the present
day, compared with those of former times.
522. The best beverage for a patient during labor is either a cup of
warm tea, or of gruel, or of arrow-root. It is folly in the extreme,
during the progress of labor, to force her to eat: her stomach recoils
from it, as at these times there is generally a loathing of food, and if
we will, as we always ought to, take the appetite as our guide, we
shall never go far wrong.
523. A patient during labor ought frequently to make water; by
doing so she will add materially to her ease and comfort, and it will
give the adjacent parts more room, and will thus expedite the labor. I
wish to call attention to this point, as many ladies, especially with
their first children, have, from false delicacy, suffered severely from
not attending to it; one of the ill effects of which is inability after the
labor is over to make water without the assistance of the doctor, who
might in an extreme case deem it necessary to introduce a catheter
into the bladder, and thus to draw the water off.
524. I recommended, in a previous paragraph, that the doctor
ought to have either the drawing-room or a bedroom to retire to, in
order that the patient may, during the progress of the labor, be left
very much to herself, and that thus she may have full opportunities,
whenever she feels the slightest inclination to do so, of thoroughly
emptying either her bladder or her bowels. Now, this advice is of
very great importance, and if it were, more than it is, attended to,
would cause a great diminution of misery, of annoyance, and
suffering. I have given the subject great attention; as I have had large
experience in midwifery practice; I therefore speak “like one having
authority,” and if my advice in this particular be followed, this book
will not have been written in vain.
525. If the patient, twelve hours after the labor, and having tried
two or three times during that time, is unable to make water, the
medical man ought to be made acquainted with it, or serious
consequences might ensue.
CHLOROFORM IN HARD AND IN LINGERING LABOR.
526. Mothers and doctors are indebted to Dr.—now Sir James—

Simpson for the introduction of chloroform, one of the greatest and
most valuable discoveries ever conferred on suffering humanity.
527. Sulphuric ether was formerly used to cause insensibility to
pain; but it is far inferior to chloroform, and is now, in this country,
very seldom employed; while the inhalation of chloroform, especially
in cases of hard and of lingering labor, is every day becoming more
general, and will do so still more extensively as its value is better
understood, and when, in well-selected cases, its comparative
freedom from danger is sufficiently appreciated.
528. Chloroform, then, is a great boon in midwifery practice;
indeed, we may say with Dr. Kidd,[96] that in labor cases “it has
proved to be almost a greater boon than in the experimental and
gigantic operations of the surgeon.” It may be administered in labor
by a medical man with perfect safety. I have given it in numerous
instances, and have always been satisfied with the result.
529. The inhalation of chloroform causes either partial or complete
unconsciousness, and freedom from pain either for a longer or for a
shorter time, according to the will of the operator. In other words,
the effects might with perfect safety be continued either for a few
minutes, or from time to time for several hours; indeed, if given in
proper cases, and by a judicious medical man, with immense benefit.
530. Chloroform is more applicable and useful in a labor—more
especially in a first labor—when it is lingering, when the pains are
very severe, and when, notwithstanding the pain, it is making but
little progress,—then chloroform is a priceless boon.
531. Chloroform, too, is very beneficial when the patient is of a
nervous temperament, and when she looks forward with dread and
apprehension to each labor pain.
532. It might be asked,—Would you give chloroform in every case
of labor, be it ever so easy and quick? Certainly not: it is neither
advisable nor expedient, in an ordinary, easy, quick labor, to
administer it.
533. The cases in which it is desirable to give chloroform are all
lingering, hard, and severe ordinary labors. In such I would gladly
use it. But before administering it, I would, as a rule, wait for at least
six hours from the commencement of the labor.
534. Oh, the delightful and magical effects of it in the cases here
described! the lying-in room, from being in a state of gloom,
despondency, and misery, is instantly transformed, by its means,
into one of cheerfulness, hope, and happiness!
535. When once a lying-in woman has experienced the good effects
of chloroform in assuaging her agony, she importunately, at every
recurrence of “the pain,” urges her medical man to give her more! In
all her subsequent confinements, having once tasted the good effects
of chloroform, she does not dread them. I have frequently heard a
lady declare that, now (if her labor be either hard or lingering) she
can have chloroform, she looks forward to the period of childbirth
with confidence and hope.
536. It might be asked,—Does the inhalation of chloroform retard
the patient’s “getting about”? I emphatically declare that it does not
do so. Those who have had chloroform have always, in my practice,
had as good and as speedy recoveries as those who have not inhaled
it.
537. One important consideration in the giving of chloroform in
labor is, that a patient has seldom, if ever, while under the effects of
it, been known to die; which is more than can be said when it has
been administered in surgical operations, in the extraction of teeth,
etc. “I know there is not one well-attested death from chloroform in
midwifery in all our journals.”[97]
538. One reason why it may be so safe to give chloroform in labor
is, that in the practice of midwifery a medical man does not deem it
needful to put his patient under the extreme influence of it. He
administers just enough to ease her pain, but not sufficient to rob her
of total consciousness; while in a surgical operation the surgeon may
consider it necessary to put his patient under the full influence of
chloroform: hence the safety in the one, and the danger in the other
case. “It is quite possible to afford immense relief, to ‘render the
pains quite bearable,’ as a patient of mine observed, by a dose which
does not procure sleep or impair the mental condition of the patient,
and which all our experience would show is absolutely free from
danger.”[98]
539. There is another advantage in chloroform,—the child, when
he is born, is usually both lively and strong, and is not at all affected
by the mother having had chloroform administered to her. This is a
most important consideration.
540. The doctor, too, as I before remarked, is deeply indebted to
Sir J. Simpson for this great boon: formerly he dreaded a tedious
and hard labor; now he does not do so, as he is fully aware that
chloroform will rob such a lying-in of much of its terror and most of
its pain and suffering, and will in all probability materially shorten
the duration of the confinement.
541. Chloroform ought never to be administered, either to a labor
patient or to any one else, except by a medical man. This advice
admits of no exception. And chloroform should never be given unless
it be either in a lingering or in a hard labor. As I have before advised,
in a natural, easy, everyday labor, nature ought not to be interfered
with, but should be allowed to run its own course. Patience,
gentleness, and non-interference are the best and the chief requisites
required in the majority of labor cases.
HINTS TO ATTENDANTS IN CASE THE DOCTOR IS

UNAVOIDABLY ABSENT.
542. It frequently happens that after the first confinement the

labor is so rapid that the child is born before the doctor has time to
reach the patient.
543. It is consequently highly desirable—nay, imperatively
necessary—for the interest and for the well-doing both of the mother
and of the baby, that either the nurse or the lady friend should, in
such an emergency, know what to do and what NOT to do. I
therefore, in the few following paragraphs, purpose, in the simplest
and clearest language I can command, to enlighten them on the
subject.
544. In the first place, let the attendants be both calm and self-
possessed, and let there be no noise, no scuffling, no excitement, no
whispering, and no talking, and let the patient be made to thoroughly
understand that there is not the slightest danger; as the principal
danger will be in causing unnecessary fears both as to herself and
her child. Tens of thousands are annually delivered in England, and
everywhere else, without the slightest assistance from a doctor,[99]—
he not being at hand or not being in time; and yet both mother and
child almost invariably do well. Let her be informed of this fact—for
it is a fact—and it will be a comfort to her and will assuage her fears.
The medical man, as soon as he arrives, will soon make all right and
straight.
545. In the mean time let the following directions be followed:
Supposing a child to be born before the medical man arrives, the
nurse ought then to ascertain whether a coil of navel-string be
around the neck of the infant; if it be, it must be instantly liberated,
or he might be strangled. Care should be taken that he has sufficient
room to breathe, that there be not a “membrane” over his mouth;[100]
and that his face be not buried in the clothes. Any mucus about the
mouth of the babe ought, with a soft napkin, to be wiped away, or it
might impede the breathing.
546. Every infant, the moment he comes into the world, ought to
cry; if he does not naturally, he should be made to do so by smacking
his buttocks until he does cry. He will then be safe:
“We came crying hither.
Thou know’st, the first time we smell the air
We waul and cry.”[101]
547. If the doctor has not arrived, cheerfulness, quietness, and

presence of mind must be observed by all around; otherwise, the
patient may become excited and alarmed, and dangerous
consequences might ensue.
548. If the infant should be born apparently dead, a few smart
blows must be given on the buttocks and on the back; a smelling-
bottle ought to be applied to the nostrils, or rag should be singed
under the nose, taking care that the burning tinder does not touch
the skin; and cold water must be freely sprinkled on the face. The
navel-string, as long as there is pulsation in it, ought not to be tied.
549. The limbs, the back, and the chest of the child ought to be
well rubbed with the warm hand. The face should not be smothered
up in the clothes. If pulsation has ceased in the navel-string (the
above rules having been strictly followed, and having failed), let the
navel-string be tied and divided,[102] and then let the child be plunged
into warm water—98° Fahr. If the sudden plunge does not rouse
respiration into action, let him be taken out of the warm bath, as the
keeping him for any length of time in the water will be of no avail.
550. If these simple means should not quickly succeed, although
they generally will, Dr. Marshall Hall’s Ready Method ought in the
following manner to be tried: “Place the infant on his face; turn the
body gently but completely on the side and a little beyond, and then
on the face, alternately; repeating these measures deliberately,
efficiently, and perseveringly, fifteen times in the minute only.”
551. Another plan of restoring suspended animation is by artificial
respiration, which should be employed in the following manner: Let
the nurse (in the absence of the doctor) squeeze, with her left hand,
the child’s nose, to prevent any passage of air through the nostrils;
then let her apply her mouth to the child’s mouth, and breathe into
it, in order to inflate the lungs; as soon as they are inflated, the air
ought, with the right hand, to be pressed out again, so as to imitate
natural breathing. Again and again, for several minutes, and for
about fifteen times a minute, should the above process be repeated;
and the operator will frequently be rewarded by hearing a convulsive
sob, which will be the harbinger of renewed life.
552. Until animation be restored, the navel-string, provided there
be pulsation in it, ought not to be tied. If it be tied before the child
has breathed, and before he has cried, he will have but a slight
chance of recovery. While the navel-string is left entire, provided
there be still pulsation in it, the infant has the advantage of the
mother’s circulation and support.
553. If Dr. Marshall Hall’s Ready Method and if artificial
respiration should not succeed, he must be immersed up to his neck
in a warm bath of 98° Fahrenheit. A plentiful supply of warm water
ought always to be in readiness, more especially if the labor be either
hard or lingering.
554. Should the child have been born for some time before the
doctor has arrived, it may be necessary to tie and to divide the navel-
string. The manner of performing it is as follows: A ligature,
composed of four or five whity-brown threads, nearly a foot in length
and with a knot at each end, ought, by a double knot, to be tightly
tied, at about two inches from the body of the child, around the
navel-string. A second ligature must, in a similar manner, be applied
about three inches from the first, and the navel-string should be
carefully divided midway between the two ligatures. Of course, if the
medical man should shortly be expected, any interference would not
be advisable, as such matters ought always to be left entirely to him.
555. The after birth must never be brought away by the nurse: if
the doctor has not yet arrived, it should be allowed to come away
(which, if left alone in the generality of cases, it generally will) of its
own accord. The only treatment that the nurse ought in such a case
to adopt is, that she apply, by means of her right hand, firm pressure
over the region of the womb: this will have the effect of encouraging
the contraction of the womb, of throwing off the after birth, and of
preventing violent flooding.
556. If the after birth does not soon come away, say in an hour, or
if there be flooding, another medical man ought to be sent for; but on
no account should the nurse be allowed to interfere with it further
than by applying firm pressure over the region of the womb, and not
touching the navel-string at all; as I have known dangerous, and in
some cases even fatal, consequences to ensue from such meddling.
REST AFTER DELIVERY.
557. A lady ought never to be disturbed for at least an hour after

the delivery; if she be, violent flooding might be produced. The
doctor, of course, will make her comfortable by removing the soiled
napkins, and by applying clean ones in their place.
558. Her head ought to be made easy; she must still lie on her side;
indeed, for the first hour let her remain nearly in the same position
as that in which she was confined—with this only difference, that if
her feet have been pressing against the bedpost, they should be
removed from that position.
CLOTHING AFTER LABOR.
559. She ought, after the lapse of an hour or two, to be moved from
one side of the bed to the other. It ought to be done in the most
tender and cautious manner. She must not, on any account
whatever, be allowed to sit erect in the bed. While being moved, she
herself should be passive—that is to say, she ought to use no
exertion, no effort, but should, by two attendants, be removed from
side to side; one must take hold of her shoulders, the other of her
hips.
560. A patient, after delivery, usually feels shivering and starved;
it will therefore be necessary to throw additional clothing, such as a
blanket or two, over her, which ought to envelop the body, and
should be well tucked around her; but the nurse ought to be careful
not to overload her with clothes, or it might produce flooding,
fainting, etc.; as soon, therefore, as she is warmer, let the extra
clothing be gradually removed. If the feet be cold, let them be
wrapped in a warm flannel petticoat, over which a pillow should be
placed.
561. A frequent change of linen after confinement is desirable.
Nothing is more conducive to health than cleanliness. Great care
should be taken to have the sheets and linen well aired.
REFRESHMENT.
562. A cup of cool, black tea, directly after a patient is confined,

ought to be given. I say cool, not cold, as cold tea might chill her. Hot
tea would be improper, as it might induce flooding.
563. As soon as she is settled in bed, there is nothing better than a
small basin of warm gruel.
564. Brandy ought never, unless ordered by the medical man, to be
given after a confinement. Warm beer is also objectionable; indeed,
stimulants of all kinds must, unless advised by the doctor, be
carefully avoided, as they would only produce fever, and probably
inflammation. Caudle is now seldom given; but still, some old-
fashioned people are fond of recommending it after a labor. Caudle
ought to be banished the lying-in room; it caused in former times the
death of thousands.
BANDAGE AFTER A CONFINEMENT.
565. (1) This consists of thick linen, similar to sheeting, about a

yard and a half long, and sufficiently broad to comfortably support
the belly. It ought to be put on moderately tight; and should be
retightened every night and morning, or oftener, if it become slack.
(2) Salmon’s Obstetric Binder is admirably adapted to give support
after a confinement, and may be obtained of any respectable
surgical-instrument-maker.
566. If there be not either a proper bandage or binder at hand—(3)
a yard and a half of unbleached calico, folded double, will answer the
purpose. The best pins to fasten the bandages are the patent safety
nursery-pins. The binder requires no pins.
567. A support to the belly after labor is important: in the first
place, it is a great comfort; in the second, it induces the belly to
return to its original size; and lastly, it prevents flooding. Those
ladies, more especially if they have had large families, who have
neglected proper bandaging after their confinements, frequently
suffer from enlarged and pendulous bellies, which give them an
unwieldy and ungainly appearance.
POSITION.
568. The way of placing the patient in bed.—She ought not,

immediately after a labor, under any pretext or pretense whatever, to
be allowed to raise herself in bed. If she be dressed, as recommended
at paragraph 508, her soiled linen may readily be removed; and she
may be drawn up by two assistants—one being at the shoulders and
the other at the legs—to the proper place, as she herself must not be
allowed to use the slightest exertion.
569. Inattention to the above recommendation has caused violent
flooding, fainting, bearing down of the womb, etc., and in some cases
even fatal consequences.
THE LYING-IN ROOM.
570. The room to be kept cool and well ventilated.—A nurse is too
apt, after the confinement is over, to keep a large fire. Nothing is
more injurious than to have the temperature of a lying-in room high.
A little fire, provided the weather be cold, to dress the baby by, and
to encourage a circulation of the air, is desirable. A fire-guard ought
to be attached to the grate of the lying-in room. The door must
occasionally be left ajar, in order to change the air of the apartment;
a lying-in woman requires pure air as much as any other person; but
how frequently does the nurse fancy that it is dangerous for her to
breathe it!
571. After the affair is over, the blinds ought to be put down, and
the window curtains should be drawn, in order to induce the patient
to have a sleep, and thus to rest herself after her hard work. Perfect
stillness must reign both in the room and in the house.
572. It is really surprising, in this present enlightened age, how
much misconception and prejudice there still is among the
attendants of a lying-in room; they fancy labor to be a disease,
instead of being what it really is—a natural process; and that old-
fashioned notions, and not common sense, ought to guide them.
573. The patient should, after the labor, be strictly prohibited from
talking; and noisy conversation ought not to be allowed; indeed, she
cannot be kept too quiet, as she may then be induced to fall into a
sweet sleep, which would recruit her wasted strength. As soon as the
baby is washed and dressed, and the mother is made comfortable in
bed, the nurse ought alone to remain; let every one else be banished
the lying-in room. Visitors should on no account, until the medical
man gives permission, be allowed to see the patient.
THE BLADDER.
574. Ought a patient to go to sleep before she has made water?—

There is not the least danger in her doing so (although some old-
fashioned person might tell her that there is); nevertheless, before
she goes to sleep, if she have the slightest inclination she should
respond to it, as it would make her feel more comfortable and sleep
more sweetly.
575. Let me urge the importance of the patient, immediately after
childbirth, making water while she is in a lying position. I have
known violent flooding to arise from a lying-in woman being
allowed, soon after delivery, to sit up while passing her water.
576. The “female slipper”[103] (previously warmed by dipping it in
very hot water and then quickly drying it) ought, at these times, and
for some days after a confinement, to be used. It is admirably
adapted for the purpose, as it takes up but little room and is
conveniently shaped, and readily slips under the patient, and enables
her to make water comfortably, she being perfectly passive the while.
It should be passed under her in front, and not at the side of the
body.
577. If there be any difficulty in her making water, the medical
man must, through the nurse, be immediately informed of it. False
delicacy ought never to stand in the way of this advice. It should be
borne in mind that, after either a very lingering or a severe labor,
there is frequently retention of urine,—that is to say, that although
the bladder may be full of water, the patient is, without assistance,
unable to make it.
578. After the patient, while lying down, trying several times to
pass her water, and after, allowing twelve or fifteen hours to elapse,
and not being able to do so, it will be well for her to try the following
method: Let the pot de chambre be well warmed, let the rim be
covered with flannel, let her, supported the while by the nurse, kneel
on the bed, her shoulders being covered with a warm shawl; then let
her, with the pot de chambre properly placed between her knees on
the bed, try to make water, and the chances are that she will now
succeed.
579. If she does not, twenty-four hours having elapsed, the doctor
must be informed of the fact; and it will then be necessary, absolutely
necessary, for him, by means of a catheter, to draw off the water. It
might be well to state that the passing of a catheter is unattended
with either the slightest danger or pain; and that it is done without
exposing her, and thus without shocking her modesty.
THE BOWELS.
580. The bowels are usually costive after a confinement. This

confined state of the bowels after labor is doubtless a wise provision
of nature, in order to give repose to the surrounding parts—
especially to the womb; it is well, therefore, not to interfere with
them, but to let them have perfect rest for three days. Sometimes
before the expiration of the third day the bowels are relieved, either
without medicine or merely by the taking of a cupful of warm coffee.
If such be the case, all well and good; as it is much better that the
bowels should be relieved without medicine than by medicine; but if,
having taken the coffee, at the end of the third day they are not
opened, then early on the following—the fourth—morning, a dose of
castor oil should be given in the manner recommended at paragraph
281. Either a teaspoonful or a dessertspoonful, according to the
constitution of the patient, will be a proper dose. If, in the course of
twelve hours, it should not have the desired effect, it must be
repeated. The old-fashioned custom was to give castor oil on the
morning after the confinement; this, as I have before proved, was a
mistaken plan.
581. After a lying-in, and when the bowels are not opened either
naturally or by the taking of a cupful of warm coffee, if medicine be
given by the mouth, castor oil is the best medicine, as it does not
irritate either the patient’s bowels, or, through the mother’s milk,
gripe the infant. Aperient pills, as they most of them contain either
colocynth or aloes, or both, frequently give great pain to the babe,
and purge him much more than they do the mother herself; aperient
pills, therefore, after a confinement ought never to be taken.
582. If the patient object to the taking of castor oil, let the nurse,
by means of an enema apparatus, administer an enema. This is an
excellent, indeed the best, method of opening the bowels, as it
neither interferes with the appetite nor with the digestion; it does
away with the nauseousness of castor oil, and does not, in the
administration, give the slightest pain. If the first enema should not
have the desired effect, let one be given every quarter of an hour until
relief be obtained. One of the best for the purpose is the following:
Take of—Olive oil, two tablespoonfuls;
Table salt, two tablespoonfuls;
Warm oatmeal gruel, one pint;
To make a clyster.
Another capital enema for the purpose is one made of Castile soap
dissolved in warm water.
583. If the patient object both to the taking of the castor oil and to
the administration of an enema, then the following draught will be
found useful; it will act kindly, and will neither gripe the mother nor
the child:
Take of—Concentrated Essence of Senna, half an ounce;
Syrup of Ginger, one drachm;
Distilled Water, seven drachms:
To make a draught. To be taken early in the morning.

If in twelve hours the above draught should not have the desired
effect (although, if the essence of senna be good, it usually does long
before that time), let the draught be repeated. If the bowels be easily
moved, half of the above draught is generally sufficient; if it be not so
in twelve hours, the remainder should be taken. But let every lying-in
woman bear in mind that as soon as her bowels will act, either
naturally or by the taking of a cupful of warm coffee, without an
aperient, not a particle of opening medicine should be taken.
584. But, after all that can be said on the subject, there is no better
method in the world for opening a lying-in patient’s bowels, when
costive, than (if the cup of coffee be not sufficiently powerful) by
giving her an enema, as advised in a previous paragraph. An enema
is safe, speedy, painless, and effectual, and does not induce
costiveness afterward, which castor oil, and all other aperients, if
repeatedly taken, most assuredly will.
585. An enema, then, is an admirable method of opening costive
bowels, both during suckling and during pregnancy, and deserves to
be more universally followed than it now is; fortunately, the plan just
recommended is making rapid progress, and shortly will, with ladies
at such times, entirely supersede the necessity of administering
aperients by the mouth.
586. Aperients, after a confinement, were in olden times, as a
matter of course, repeatedly given both to the mother and to the
babe, to their utter disgust and to their serious detriment! This was
only one of the numerous mistakes and follies that formerly
prevailed in the lying-in room. Unfortunately, in those days a
confinement was looked upon as a disease, and to be physicked
accordingly. A better state of things is happily now dawning.
587. When the patient’s bowels, for the first few days after the
confinement, require to be opened, she ought to use either the
French bed-pan or the bed-pan of the Liverpool Northern Hospital.
Either the one or the other of these pans is a great improvement on
the old-fashioned bed-pans, as they will readily slip under the
patient, and will enable her, while lying down and while she is
perfectly passive in bed, to have her bowels relieved, which at these
times is very desirable. The French bed-pan or the bed-pan of the
Liverpool Northern Hospital is admirably adapted for a lying-in
room; indeed, no lying-in room ought to be without either the one or
the other of these useful inventions. “A flannel cap for the toe-part,
held on by strings round the heel, will afford considerable comfort to
the patient.”[104]
“CLEANSINGS”—ABLUTIONS.
588. The “Cleansings.”—This watery discharge occurs directly after

a lying-in, and lasts either a week or a fortnight, and sometimes even
longer. It is, at first, of a reddish color; this gradually changes to a
brownish hue, and afterward to a greenish shade; hence the name of
“green waters.” It has in some cases a disagreeable odor. A moderate
discharge is necessary; but when it is profuse, it weakens the patient.
589. Some ignorant nurses object to have the parts bathed after
delivery; they have the impression that such a proceeding would give
cold. Now, warm fomentation twice a day, and even oftener, either if
the discharge or if the state of the parts requires it, is absolutely
indispensable to health, to cleanliness, and comfort. Ablutions,
indeed, at this time are far more necessary than at any other period
of a woman’s existence.
590. There is nothing better for the purpose than a soft sponge and
warm water, unless the parts be very sore; if they be, a warm
fomentation, two or three times a day, of marshmallows and
chamomile,[105] will afford great relief, or the parts may be bathed
with warm oatmeal gruel, of course without salt. In these cases, too, I
have found warm barm (yeast) and water a great comfort, and which
will soon take away the soreness. The parts ought, after each
fomentation, to be well but quickly dried with warm, dry, soft towels.
591. If the internal parts be very sore, it may be necessary, two or
three times a day, to syringe them out, by means of an india-rubber
vaginal syringe,[106] with either of the above remedies. Hence the
importance of having a good monthly nurse, of having one who
thoroughly understands her business.
592. Let the above rules be strictly followed. Let no prejudices and
no old-fashioned notions, either of the nurse or of any female friend,
stand in the way of the above advice. Ablution of the parts, then,
after a confinement, and that frequently, is absolutely required, or
evil results will, as a matter of course, ensue.
REST AND QUIETUDE.

593. A horizontal—a level—position for either ten days or a
fortnight after a labor is important. A lady frequently fancies that if
she supports her legs, it is all that is necessary. Now, this is absurd; it
is the womb and not the legs that requires rest; and the only way to
obtain it is by lying flat either on a bed or on a sofa: for the first five
or six days, day and night, on a bed, and then for the next five or six
days she ought to be removed for a short period of the day either to
another bed or to a sofa; which other bed or sofa should be wheeled
to the side of the bed, and she must be placed on it by two assistants,
one taking hold of her shoulders and the other of her hips, and thus
lifting her on the bed or sofa, she herself being perfectly passive, and
not being allowed to sit erect the while. She ought, during the time
she is on the sofa, to maintain the level position.
594. She ought, after the first nine days, to sit up for an hour; she
should gradually prolong the time of the sitting; but still she must,
for the first fortnight, lie down a great part of every day. She should,
after the first week, lie either on a sofa or on a horse-hair mattress.
595. The above plan may appear irksome, but my experience tells
me that it is necessary, absolutely necessary. The benefit the patient
will ultimately reap from it will amply repay the temporary
annoyance of so much rest. Where the above rules have not been
adopted, I have known flooding, bearing down of the womb, and
even “falling” of the womb, frequent miscarriages, and ultimately
ruin of the constitution, to ensue.
596. “Falling of the womb” is a disagreeable complaint, and the
misfortune of it is, that every additional child increases the infirmity.
Now, all this might, in the majority of cases, have been prevented, if
the recumbent posture for ten days or a fortnight after delivery had
been strictly adopted.
597. If a patient labor under a “falling of the womb,” she ought to
apply to a medical man, who will provide her with a proper support,
called a pessary, which will prevent the womb from “falling down,”
and will effectually keep it in its proper place.
598. It is only a medical man, accustomed to these matters, who
can select a pessary suitable for each individual case. A proper kind
of, and duly-adjusted, pessary is a great comfort to a patient, and will
enable her both to take her proper exercise and to follow her
ordinary employments; indeed, if a suitable pessary be used, it is so
comfortable that the patient often forgets that she is wearing one at
all. Those pessaries ought only to be employed that can be removed
every night, as there is not the least necessity for a patient to sleep in
one, as the womb does not usually come down when the patient is
lying down. Moreover, a pessary ought to be kept perfectly clean, and
unless it be daily removed it is utterly impossible to keep it so. It is a
great comfort and advantage to a patient to be able both to introduce
and to remove the pessary herself, which, if a proper kind be
employed, she can, when once taught, readily do.
599. If “falling of the womb” be early and properly treated, there is
a good chance of a patient being perfectly cured, and thus of being
able to dispense with a pessary altogether.
DIETARY.
600. For the first day the diet should consist of nicely made and
well-boiled gruel, arrow-root and milk, bread and milk, tea, dry toast
and butter, or bread and butter; taking care not to overload the
stomach with too much fluid. Therefore, either a cupful of gruel, or of
arrow-root, or of tea, at a time, should not be exceeded, otherwise
the patient will feel oppressed; she will be liable to violent
perspiration, and there will be a too abundant secretion of milk.
601. For the next—the second day.—Breakfast,—either dry toast
and butter, or bread and butter, and black tea. Luncheon,—either a
breakfast-cupful of strong beef-tea,[107] or of bread and milk, or of
arrow-root made with good fresh milk. Dinner,—either chicken or
game, mashed potatoes, and bread. Tea, the same as for breakfast.
Supper,—a breakfast-cupful of well-boiled and well-made gruel,
made either with water or with fresh milk, or with water with a
tablespoonful of cream added to it.
602. If beef-tea and arrow-root and milk be distasteful to the
patient, or if they do not agree, then for luncheon let her have either
a light egg pudding or a little rice pudding instead of either the beef-
tea or the arrow-root.
603. On the third and fourth days.—Similar diet to the second
day, with this difference, that for her dinner the patient should have
mutton—either a mutton-chop or a cut out of a joint of mutton,
instead of the chicken or game. The diet ought gradually to be
improved, so that at the end of four days she should return to her
usual diet, provided it be plain, wholesome, and nourishing.
604. The above, for the generality of cases, is the scale of dietary;
but of course every lying-in woman ought not to be treated alike. If
she be weak and delicate, she may from the beginning require good
nourishment, and instead of giving her gruel, it may, from the very
commencement, be necessary to prescribe good strong beef-tea, veal-
and-milk broth,[108] chicken-broth, mutton-chops, grilled chicken,
game, the yelk and the white of an egg beaten up together in half a
teacupful of good fresh milk, etc. Common sense ought to guide us in
the treatment of a lying-in as of every other patient. We cannot treat
people by rule and compass; we must be guided by circumstances;
we can only lay down general rules. There is no universal guide, then,
to be followed in the dietary of a lying-in woman; each case may and
will demand separate treatment; a delicate woman, as I have just
remarked, may, from the very first day, require generous living;
while, on the other hand, a strong, robust, inflammatory patient
may, for the first few days, require only simple bland nourishment,
without a particle of stimulants. “And hence the true secret of
success rests in the use of common sense and discretion—common
sense to read nature aright, and discretion in making a right use of
what the dictates of nature prescribe.”[109]
BEVERAGE.
605. For the first week, either toast and water, or barley-water and
milk,[110] with the chill taken off, is the best beverage. Wine, spirits,
and beer, during this time, unless the patient be weak and exhausted,
or unless ordered by the medical man, ought not to be given. All
liquids given during this period should be administered by means of
a feeding-cup; this plan I strongly recommend, as it is both a comfort
and a benefit to the patient; it prevents her from sitting up in bed
every time she has to take fluids, and it keeps her perfectly still and
quiet, which, for the first week after confinement, is very desirable.
606. When she is weak, and faint, and low, it may, as early as the
first or second day, be necessary to give a stimulant, such as either a

Photovoltaic Module Reliability Wohlgemuth All Chapter

Uploaded by

Copyright:

Available Formats

You might also like

Photovoltaic Module Reliability Wohlgemuth All Chapter

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Photovoltaic Module Reliability Wohlgemuth All Chapter

Uploaded by

Copyright:

Available Formats

Photovoltaic module reliability

Limit of Liability/Disclaimer of Warranty

Library of Congress Cataloging‐in‐Publication data applied for

Cover Design: Wiley

Set in 9.5/12.5pt STIXTwoText by SPi Global, Pondicherry, India

2 Module Failure Modes 17

3 Development of Accelerated Stress Tests 55

4.6.2.1 MST 01 – Visual Inspection 113

5 Failure Analysis Tools 127

6 Using Quality Management Systems to Manufacture PV Modules 151

7 The PVQAT Effort 165

8 Conformity Assessment and IECRE 195

9 Predicting PV Module Service Life 209

Photovoltaic Module Reliability, First Edition. John H. Wohlgemuth.

1.1 ­Brief History of PVs

1.2 ­Types of PV Cells

Metal grid lines AR coating

Solar cell p-n junction Back metallization

Figure 1.1 Cross-sectional drawing of screen-printed cry-Si solar cell.

Step # Step Name Description

Semiconductor layers Metallization

Figure 1.2 Cross-sectional drawing of monolithically integrated thin film module.

1.3 ­Module Packaging – Purpose and Types

Table 1.2 Typical commercial module constructions.

Glass Superstrate: Cry‐Si Cells

Figure 1.3a Cross-sectional drawing of glass/encapsulant/cry-Si cells/encapsulant/backsheet

Figure 1.3b Cross-sectional drawing of glass/encapsulant/cry-Si cells/encapsulant/glass module.

Flexible or rigid substrate

Figure 1.3c Cross-sectional drawing of flexible cry-Si module.

Edge seal Front glass

Thin film PV Back glass

Figure 1.4b Cross-sectional drawing of front glass/encapsulant/thin film cells/substrate modules.

Thin film PV Encapsulant

Flexible or rigid substrate

1.4 ­What Does Reliability Mean for PV Modules?

Since the book is entitled “Reliability of Photovoltaic Modules,” it is important to explain

the promised function is continued production of the specified amount of electricity in a

1.5 ­Preview of the Book

1 Beetz, B. (2018). https://www.pv‐magazine.com/2018/12/31/14‐pv‐trends‐for‐2019

Module Failure Modes

Photovoltaic Module Reliability, First Edition. John H. Wohlgemuth.

Table 2.1 Failure modes observed for c-Si modules.

Table 2.2 Linear thermal expansion coefficient of selected

Material m/mK (* 10−6)

2.2 ­Broken/Cracked Cells and Snail Trails

Figure 2.4 Cells with snail trails.

Most EVA‐encapsulated modules do not exhibit this type of encapsulant‐glass delami‑

2.4 ­Corrosion of Cell Metallization

Moisture induced corrosion of cell metallization is usually associated with significant

Visual image EL image IR image UVF image

Figure 2.11a Characterization of unexposed module. Source: images provided by Govindasamy

Visual image EL image IR image UVF image

UVF (close up shot)

Cell peripherals → clear

2.6 ­Failure of Electrical Bonds Particularly Solder Bonds

Figure 2.14 Module with tempered glass broken by impact.

Figure 2.15 Module with

Rogers, too, in referring to this interesting event, sweetly sings:

1.1 Brief History of PVs

1.2 Types of PV Cells

1.3 Module Packaging – Purpose and Types

1.4 What Does Reliability Mean for PV Modules?

1.5 Preview of the Book

2.2 Broken/Cracked Cells and Snail Trails

2.4 Corrosion of Cell Metallization

2.6 Failure of Electrical Bonds Particularly Solder Bonds