Series on Innovation in Structures and Construction -Vol.

Earthquake-Resistant Design
of Masonry Buildings

Miha Tomazevie
Slovenian National Building and Civil Engineering Institute

Imperial College Press

\
Published by
Imperial College Press
57 Shelton Street
Covent Garden
London WC2H 9HE

Distributed by
World Scientific Publishing Co. Re. Ltd.
5 Toh Tuck Link, Singapore 596224
USA ofice: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601
UK ofice: 57 Shelton Street, Covent Garden, London WC2H 9HE

Library of Congress Cataloging-in-PublicationData

TomaZeviE, Miha.
Earthquake-resistant design of masonry buildings / Miha Toma'ievi'c.
p. cm. -- (Series in innovation in structures and
construction; vol. 1)
Includes bibliographical references and index.
ISBN 1-86094-066-8
1. Masonry. 2. Buildings. 3. Earthquake resistant design.
I. Title. 11. Series.
TH5321.T66 1999
693.8'52--d~21 99- 18293
CIP

British Library Cataloguing-in-PublicationData

A catalogue record for this book is available from the British Library.

First published 1999

Reprinted 2000,2006

Copyright Q 1999 by Imperial College Press

All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any m e w ,
electronic or mechanical, including photocopying, recording or any information storage and retrieval
system now known or to be invented, without written permission from the Pubtisher.

For photocopying of material in this volume, please pay a copying fee through the Copyright
Clearance Center, Inc., 222 Rosewood Drive, Danvers, M A 01923, USA.In this case permission to
photocopy is not required from the publisher.

Printed by FuIsland Offset Printing (S) Pte Ltd, Singapore

 PREFACE

From many points of view, writing this book Earthquake-Resistant Design

of Masonry BuiEdings was not an easy task. In most countries, masonry is still
used as a traditional, non-engineered construction material: in this regard, the
engineering approach to masonry construction had to be encouraged. Several,
although not many, books on this topic have been published in the last decade,
therefore, a touch of originality had to be given to this book. And most
importantly, taking into account the complexity of masonry construction world-
wide, the request that the book should be useful for a broad readership of
structural engineers, interested in earthquake resistant design of masonry
buildings, has to be respected.
In the recent books on masonry structures, the idea that reinforced grouted
masonry is the earthquake-resistant type of masonry construction, is usually
followed. However, buildings built by traditional masonry construction systems,
such as plain and confined, as well as reinforced masonry, where normal or
specially shaped hollow or perforated masonry units are used to accommodate
the reinforcement, embedded in mortar or concrete, can also resist earthquakes.
As in most countries these buildings outnumber reinforced grouted masonry
ones. The discussion in this book is limited to traditional, unreinforced and
reinforced masonry construction systems. However, general aspects of
earthquake-resistant masonry construction are preferred to aetails specific for
different masonry construction systems.
In order to assist the implementation of the recent European structural
design codes, Eurocodes, where a clear concept for limit states verification has
also been introduced for masonry structures, the concepts of Eurocodes’
structural verification philosophy, which has general validity, are followed.
However, although detailed information is given on the requirements of
Eurocodes, it was not the aim of the book to collect a set of design and
construction rules, accompanied with relevant equations for structural
V
vi Earthquake-Resistant Design of Masonry Buildings

verification, and produce a state-of-the-art manual for structural engineers who

wish to become familiar with earthquake-resistant design of masonry buildings.
There have been excellent books on masonry published recently which more
than fulfii this task.
Being involved in experimental research in seismic behaviour of masonry
structures for most of my career, now 30 years long, I have ambitiously decided
to make the readers familiar with masonry “behind the scenes”. Namely, an
attempt has been made in this book to provide information regarding the actual
behaviour of masonry structures when subjected to earthquakes, as well as to
present experimental background for the development of mathematical models
for seismic resistance verification. It is hoped that the readers will better
understand the design philosophy and the meaning of parameters in the design
equations. Therefore, experimental evidence to support the criteria and
mathematical models is provided, wherever possible. Given the opportunity, the
methods developed on the basis of experimental research, carried out in
Ljubljana, are mainly presented and discussed.
Although the book is not a review of state-of-the-art earthquake-resistant
design of masonry structures, an atte-mpt has been made to balance the
discussion on recent code requirements, the state-of-the-art methods of
earthquake-resistant design, and my research work in order to make the book
useful for a broader application in the design practice. I have also attempted to
present, in a condensed but easy to understand way, all the information needed
for earthquake-resistant design of masonry buildings constructed in traditional
masonry construction systems.
The book first deals with earthquakes and seismic performance of masonry
buildings. To be compatible with Eurocodes, the new European macroseismic
scale is briefly presented as a measure of earthquake intensity. Then, the basic
aspects regarding the selection of masonry materials and construction systems
are discussed, including the testing methods for the determination of mechanical
properties of masonry materials and evaluation of test results. Next, the
architectural and structural aspects of earthquake-resistant masonry construction
are discussed. The requirements regarding the shape and dimensions of
buildings are given. Simple buildings, in which the verification of seismic
resistance by calculation is not mandatory, are described.
In the chapters to follow, the basic concepts of limit states verification are
presented and equations for seismic resistance verification of masonry walls of
all types of masonry construction - unreinforced, confined, and reinforced -
Preface vii

as well as masonry infilled reinforced concrete frames, are explained. A method

for seismic resistance verification, compatible with recent code requirements, is
discussed. In all cases, experimental results are used to explain the proposed
methods and equations.
Finally, an important part of the book discusses the problems of seismic
repair, retrofit and rehabilitation of existing masonry buildings, including
historical houses in urban nuclei. Methods of strengthening the masonry walls,
as well as improving the structural integrity of existing buildings, are described
in detail. Wherever possible, experimental evidence regarding the effectiveness
of the proposed strengthening methods is given.
A list of references is given at the end of each chapter. Considering the
concept of writing the book, the list of references has no intention of providing
information regarding the state of the art of the topic discussed. It is only aimed
at orienting the reader to the source where he/she could obtain additional
background information.
Although an attempt has been made to cover as many problems related with
the earthquake-resistant design of masonry buildings as possible, not everything
could have been mentioned in the book. At least two important issues in the
field of modern masonry construction have been completely omitted.
Consequently, the interested reader should search, on his own, for information
regarding the large panel prefabricated construction, which is used in the case
of residential masonry buildings, and seismic isolation, which can be applied to
new and existing masonry buildings.
It is my pleasure to acknowledge Imperial College Press, who have given me
opportunity to write this book. My gratitude also goes to former and recent
colleagues who have assisted me in carrying out my research in the past years:
most of their names can be found on the list of references in individual chapters.
Without them, there would not be much to write about in this book. Although
most of the technical work is now done by powerful computers, the hands of Ms
Zvonka Rekete were needed to prepare the drawings.
Last but not least, my deepest thanks go to NataSa and Nejc - who have
not seen much of their husband and father recently - for their encouragement
and support during the days when the end of this adventure had seemed
unattainable.

Miha Toma2eviC
 CONTENTS

PREFACE V

1. INTRODUCTION 1
2. EARTHQUAKES AND SEISMIC PERFORMANCE
OF MASONRY BUILDINGS 5
2.1 Introduction 5
2.2 Earthquakes and Seismic Ground Motion 6
2.2.1 Causes of earthquakes 6
2.2.2 Seismic waves and earthquake ground motion 9
2.2.3 Magnitude and intensity 11
2.2.4 Occurrence of earthquakes 17
2.2.5 Earthquake ground motion and
effects on buildings 18
2.3 Seismic Performance of Masonry Buildings 25
2.3.1 Structural typology 25
2.3.2 Seismic performance and classification of damage 28
2.4 References 32
3. MASONRY MATERIALS AND
CONSTRUCTION SYSTEMS 35
3.1 Introduction 35
3.2 Masonry Materials 36
3.2. I Masonry units 36
3.2.2 Mortar 41
3.2.3 Concrete infill 42
3.2.4 Reinforcing steel 43
3.2.5 Masonry 45

ix
x Earthquake-Resistant Design of Masonry Buildings

3.3 Construction Systems 54

3.3. I Unreinforced masonry 55
3.3.2 Confined masonry 58
3.3.3 Reinforced masonry 62
3.4 References 69
4. ARCHITECTURAL AND STRUCTURAL
CONCEPTS OF EARTHQUAKE-RESISTANT
BUILDING CONFIGURATION 71
4.1 Introduction 71
4.2 Building Configuration 72
4.3 Dimensions, Building Height and Number of Stories 76
4.4 Distribution of Structural Walls 77
4.5 Wall Openings 78
4.6 Simple Buildings 79
4.7 Non-structural Elements 80
4.8 References 82
5. FLOORS AND ROOFS 85
5.1 Introduction 85
5.2 Floors 85
5.3 Bond-beams 88
5.4 Lintels, Balconies anc Overuangs 90
5.5 Roofs 92
5.6 References 92
6. BASIC CONCEPTS OF LIMIT STATES VERIFICATION
OF SEISMIC RESISTANCE OF MASONRY BUILDINGS 95
6.1 Fundamentals 95
6.2 Safety Verification and Partial Safety Factors
for Materials 96
6.3 Design Seismic Action 99
6.3.1 Seismic action and design response spectrum 100
6.3.2 Design base shear 102
6.3.3 Distribution of design base shear 103
6.3.4 Behaviour factor 105
6.4 References 107
Contents xi

7. SEISMIC RESISTANCE VERIFICATION OF

STRUCTURAL WALLS 109

7.1 Introduction 109

7.2 Experimental Simulation of Seismic
Behaviour of Masonry Walls 112
7.3 Idealisation of Experimental Results 115
7.4 Shear Resistance 124
7.4.1 Unreinforced masonry 124
7.4.2 Reinforced masonry 127
7.4.3 Confined masonry 137
7.5 Flexural Resistance 141
7.5.1 Unreinforced masonry 141
7.5.2 Reinforced masonry 146
7.5.3 Confined masonry 149
7.6 Sliding Shear 150
7.7 Flanged Sections 151
7.8 Out-of-plane Beh aviour 154
7.9 Non - structural Elernents 158
7.10 References 159
8. MASONRY INFILLED REINFORCED
CONCRETE FRAMES 163
8.1 Introduction 163
8.2 Seismic Behaviour and Mechanisms 164
8.3 Seismic Resistance Verification 169
8.3.I Design seismic loads 169
8.3.2 Lateral resistance 171
8.3.3 Stiflness 174
8.3.4 Seismic resistance verification 176
8.4 References 177
9. SEISMIC RESISTANCE VERIFICATION OF
MASONRY BUILDINGS 179
9.1 Introduction 179
9.2 Calculation Procedures 181
9.3 Structural Models 183
9.4 Storey Resistance Envelope 188
xii Earthquake-Resistant Design of Masonry Buildings

9.4.I Assumptions 190

9.4.2 Procedure for calculation 191
9.4.3 Experimental verification 196
9.5 Seismic Resistance Verification 197
9.6 References 200
10.REPAIR AND STRENGTmNXNG OF
MASONRY BUILDINGS 203
10.1 Introduction 203
10.2 Criteria for Repair and Strengthening 207
10.3 Verification of Seismic Resistance of
Existing Masonry Buildings 210
10.3.1 Dynamic characteristics and behaviour factor 210
10.3.2 Material properties 212
10.3.3 Seismic vulnerability evaluation 214
10.4 Methods of Strengthening of Masonry Walls 216
10.4.1 Repair of cracks 217
10.4.2 Repointing 220
10.4.3 Reinforced-cement coating 221
10.4.4 Grouting 225
10.4.5 Prestressing 23 1
10.4.6 Reconstruction 232
10.5 Methods of Improving Structural Integrity 234
10.5.1 Tying of walls with steel ties 235
10.5.2 Interventions in floor structures and roofs 240
10.5.3 Repair of corners and wall intersection zones 244
10.5.4 Strengthening of walls by conflnement 247
10.6 Foundations 249
10.7 Non - Str uctur a1 Elements 25 1
10.8 References 252
SYMBOLS 257
SUBJECT INDEX 263
 CHAPTER 1
INTRODUCTION

Besides wood, masonry is the most important construction material in the history
of mankind. Masonry has been used, in a wide variety of forms, as a basic
construction material for public and residential buildings in the past several thousand
years: from the Tower of Babylon, which should have reached the sky if it had been
completed, to the Great Wall of China, which is the only man-made structure visible
from the Moon.A great number of well-preserved old masonry buildings still exist,
proving that masonry can successfully resist loads and environmental impacts,
therefore providing shelter for people and their goods for a long period of time, if
adequately conceived and constructed. In recognition of their importance and value,
many of those buildings have been ranked among the assets of highest category of
mankind’s historical and cultural heritage.
Many urban settlements are located in seismic zones. This is a consequence of
the fact that most convenient geographical locations for building a city are valleys
and cross-roads, which frequently follow the locations and intersections of active
seismic faults. As proven by the historical data, many ancient and medieval towns
and cities have already been destroyed by earthquakes. Some of them have been
rebuilt at the same site, while others have been relocated to avoid future possible
seismic impacts.
Although some specific features have been invented during the course of time
to improve the seismic behaviour of masonry buildings, such as connecting
stones, strengthening the comers and wall intersections zones, as well as tying of
the walls, even today, masonry construction represents the most vulnerable part
of existing building stock. This is not only the case in undeveloped or developing
countries, but it is also the case in the most developed regions of Europe and
USA. As many recent earthquakes in those regions have proven, most earthquake

1
2 Earthwake-Resistant Design of Masonry Buildings

damage and loss of human lives have occurred in historical urban nuclei due to
inadequate seismic resistance of old masonry buildings.
When subjected to earthquake ground motion, inertia forces, proportional to
the masses of building components and induced accelerations, cause the vibration
of the structural system. As a result of vibration, additional bending and shear
stresses develop, which often exceed the strength of materials and cause damage
to structural elements. Since masonry, which can be stressed relatively high in
compression, is not a suitable material for carrying the bending and shear, the
resulting damage is severe and often causes the collapse of a building.
Consequently, for a long time, masonry has not been considered to be a suitable
material in earthquake-resistant construction. Other materials which developed in
modern times, such as steel and reinforced concrete, have replaced the traditional
clay brick in the case of multi-storey building construction in the developed part
of the world.
However, in the last few decades, considerable research on the behaviour of
masonry walls and buildings subjected to seismic actions has been carried out in
many countries. The behaviour of masonry buildings during earthquakes has been
analysed, and experiments to determine the basic parameters of the seismic
resistance of masonry walls and buildings have been carried out.
As a result of these investigations, masonry survived, and because of good
physical properties which make living in masonry buildings comfortable, the use
of masonry seems to be on the rise. However, clay brick is no more the
elementary element. In order to save raw materials and improve the thermal
insulation properties of masonry walls, perforated bricks and hollow blocks have
been developed and new technologies are being used for the production of
ceramic, and concrete blocks.
The quality of masonry products has also been improved. Using cement
mortars, relatively high-strength masonry can be produced, which makes it
possible to increase the floor spans even in the case of multi-storey buildings,
Reinforcement has been added in order to improve the resistance of masonry to
bending and shear. Although grouted reinforced masonry, which is closer to
reinforced concrete than to traditional masonry, is the main type of masonry
construction in seismic-prone areas of the USA, traditional type of masonry
construction with or without mortar bed joint reinforcement still prevails in
Europe, Asia and Latin America. New structural systems have been developed,
such as large panel systems, where prefabricated elements, cast in the factory Or
assembled on site, can be used.
Introduction 3

By means of experimental research, new data on the strength and stifkess

degradation and deterioration, ductility and energy dissipation capacity of
different types of masonry have been obtained. The results of investigations made
possible the improvement in masonry construction, and have resulted in the
development of analytical models and mathematical tools for earthquake-
resistance verification and design for seismic loads. The investigations have also
resulted into the development of new codes for masonry construction and design,
for example, Eurocode 6: Design of masonry structures and Eurocode 8 : Design
provisions for earthquake resistance of structures. In structural Eurocodes, a
clear concept for limit states verification has also been introduced, for the first
time, for masonry structures.
Considering that old masonry buildings of all types, including historical
monuments, represent an important part of the existing building stock, which is
highly vulnerable to earthquakes, considerable experimental research has been
also carried out in the last few decades to investigate the causes of damage and
develop technologies suitable for seismic retrofit and rehabilitation of existing
masonry buildings. As a result of earthquake damage observation and
experimental research, various technical solutions for the improvement of
structural and material defficiences of existing masonry buildings, are now
available. As the experiments and earthquakes prove, by applying these methods
to old buildings, the same level of seismic resistance can be obtained as in the
case of new buildings designed by seismic codes.
 CHAPTER 2

EARTHQUAKES AND SEISMIC

PERFORMANCE OF MASONRY BUILDINGS

2.1 Introduction

Seismic actions are accidental actions which, depending on the seismicity of

the location, rarely occur in the building's life-time. However, because of the
destructive power of earthquakes, the stability and safety of buildings located in
earthquake-prone areas should be verified for seismic loads. Such verifkation is
based on the results of geological and seismological studies, which provide data
on the seismic activity of the location and recommend the values of parameters
to be used in the assessment of the expected seismic actions. However,
verification is also based on the analysis of earthquake damage and mechanisms
of collapse, as well as subsequent experimental investigations in the seismic
behaviour, which provide the basis for the development of methods for structural
verification of newly designed buildings. In addition, on the basis of such
analysis and investigations, the deficiences of existing structural systems can
also be identified and measures for their future improvement developed.
To better understand the basic concepts of seismic resistance analysis and
seismic resistant design and construction of masonry structures, the causes of
earthquakes and the characteristics of seismic ground motion should first be
known. On the basis of the expected intensity and dynamic characteristics of the
ground motion, the forces generated by earthquakes can be assessed. On the
basis of the probability of occurrence of maximum expected intensity earthquake
at the location, and the importance and type of building under consideration, the
decision regarding the magnitude of design seismic actions can be made.
5
6 Earthquake-Resistant Design of Masonry Buildings

However, to understand the significance of the properties of materials and

structural characteristics specific for masonry structures, which influence the
response of masonry structures to seismic ground motion, the observed behaviour
of masonry buildings, as well as the causes of any damage that occurred to
buildings during earthquakes, should be also carefully analysed.

2.2 Earthquakes and Seismic Ground Motion

2.2.I Causes of earthquakes

Ground motion, which is generated by sudden displacements within the

earth's crust, is called an earthquake. Earthquakes are caused by various natural
phenomena, such as tectonic processes, volcanic eruptions, sudden failure of
parts of the ground in karstic terrain, as well as by human activities, such as large
excavations in mines, explosions, and large water reservoirs.

Tectonic plate L
1
Mid-ocea

Con1tinent
Subduction
Earthquakes
Asthenosphere
Subduction "\\
Earthquakes
~

3
\
Conth
nent

Figure 2.1. Mechanism of motion of tectonic plates at their boundaries.

According to the generally-accepted plate tectonics theory [ 1, 21, the crust

and upper part of the mantle of the earth, called the lithosphere, which is about
50 km thick under the deepest ocean and 150 km thick under the highest
mountains, is subdivided into tectonic plates, which move as rigid bodies on a
relatively soft asthenosphere (Fig. 2.1). At midoceanic ridges, the plates separate
and the hot magmatic materials fiom the asthenosphere flow up. As they cool
down at the bottom of the ocean, they form a plate. Because of the separation of
tectonic plates at midoceanic ridges, subduction or collision takes place at
midocean trenches or mountain ridges, respectively. In the case of subduction,
the plates are absorbed back into the asthenosphere. In the case of collision,
Earthquakes and Seismic Performance of Masonry Buildings 7

mountains are lifted up. At some boundaries, however, the plates simply slide
along one another at transform faults.
Because of the relative displacements between two adjoining plates, high
stresses are induced in the bedrock materials within the affected zones. In the
case where the stresses exceed the materials’ strength, the accumulated strain
energy is released in a form of an earthquake. The failure plane, where the
relative slip of adjacent rock formations has occurred, is called a tectonic fault.
In most cases, the faults are formed deep within the lithosphere and are not
visible on the surface. Such is the case of faults where earthquakes occur several
hundreds kilometres deep in the subduction zones. However, quite often large
relative displacements in the vertical and horizontal directions are visible also on
the surface of the earth. The most famous example of an active earthquake fault is
the longer than 1300 km San Andreas fault in California, which was the cause of
many earthquakes in that region. A 300-km long slip of 6.4 m caused the San
Francisco earthquake of 1906, with a Richter magnitude of M = 8.3 [3].
Most earthquakes are the result of relative movement of tectonic blocks along
active faults. Rarely, however, earthquakes occur because of overstressing of
rock formations within tectonic blocks, such as earthquakes in the middle of
Indian peninsula or eastern United States and Canada [4]. New faults rarely
develop because of earthquakes. Therefore, in the recent history of the earth,
faults can be considered to be the causes rather than the results of recent
earthquakes.
From a scientific point of view, active faults are faults that have exhibited
displacements within the last several hundred thousand years and will continue-to
do so in the fbture. From a practical engineering point of view, however, the
definition of an active fault depends also on the type and importance of the
building to be constructed. In the case of ordinary buildings, where earthquakes
with a return period of 475 years are considered in the design, the fault, which
exhibited no deformation within the last 1 1 000 years period, may be considered
as “not active”. In the case of nuclear power plants, however, the criteria are
more severe: a fault, where indications of motion have been found for at least
once within the last 350 000 years, or more than once within the last 500 000
years, should be considered as an active fault [5].
The slipping of the bedrock formation along a fault can be either in the
vertical or horizontal direction, or a combination of both (Fig. 2.2). Depending
on the direction of motion, the various types of motion are usually classified, as
seen from the position of observation, into: (a) normal fault, where the upper
8 Earthq uake-Res is tant Design of Masonry Buildings

block moves downwards, (b) reverse fault, where the upper block moves
upwards, (c) right lateral fault, where the other block moves to the right, and (d)
left lateral fault, where the other block moves to the left [3].

Normal fault

. .._* . .

Left oblique fault

Figure 2.2. Main types of fault motion.

Figure 2.3. Tangshan, China, 1976: broken pipe indicates slippage at ground surface.
Earthquakes and Seismic Performance of Masonry Buildings 9

The elastic rebound theory is used to explain the mechanism of a typical

earthquake. According to this theory, strain that has accumulated in rock
materials along the fault due to relative deformation of tectonic blocks in a long
period of time between two earthquakes, has attained the ultimate limit. As the
rock materials at the fault have been broken and crushed in the previous
earthquakes, the strength of that zone is reduced. Slippage, which occurs at the
fault zone causes rebound, which, in turn, generates seismic waves. An example
of slippage, visible at the ground surface, is shown in Fig. 2.3.
The surface along which the slippage has occurred, and which has generated
seismic waves, is called the focus or hypocentre. The projection of the
hypocentre onto the earth’s surface is called the epicentre. Although the
hypocentre and epicentre of an earthquake are idealised as two points, it should
be borne in mind that they are actually represented by areas where the slippage
between two tectonic blocks along the fault has occurred and where the impact of
the earthquake on the surface was most severe, respectively. The area of
epicentral zone depends on the intensity of the earthquake, size of the slippage
zone, and depth of the focus. It can be several tens of kilometres long and several
tens of kilometres wide. The earthquakes are considered to be shallow if the
depth of the focus does not exceed 70 km. In the case of deep earthquakes, the
depth of the focus goes down to 700 km. However, if not extremely strong by
magnitude, the effect of such earthquakes on the ground is not too severe.

2.2.2 Seismic waves and earthquake ground motion

Two types of seismic waves are generated at the slippage zone. Since they
propagate within the rock of the earth crust, these two types of seismic waves are
also called body waves, namely longitudinal or compressive waves, which
propagate in the same direction as the vibration, and transverse or shear waves,
which propagate in the direction perpendicular to the vibration. Since the
propagation velocity of longitudinal waves is faster than the propagation of
transverse waves, longitudinal waves are sometimes called primary waves (P-
waves) and transverse waves are called secondary waves (S-waves). Some typical
values of wave propagation velocities are indicated in Table 2.1.
Surface waves are the result of reflection and refraction of P- and S-waves
during propagation in stratified formations of the earth’s crust. They propagate
on the earth’s surface in two principal forms: as L (Love) waves, which vibrate in
a plane parallel to the earth’s surface and perpendicular to the direction of wave
10 Earthquake-Resistant Design of Masonry Buildings

propagation, and R (Raleigh) waves, which vibrate in an elliptical form in a plane

perpendicular to the earth's surface.

Table 2.1. Wave propagation velocities in various types of soil.

Sand 300-900 100-500

Clay 400-2000 100-600
I Sandstone I 2400-4300 1 900-2100 I
I Limestone I 3500-6500 I 1800-3800 1
I Granite I 4600-7000 I 2500-4000 I
I Basalt I 5400-6400 I 2900-3200 I
The seismic waves, generated in the focus, propagate through different layers
of rock and soil. On their way to the surface, they reflect and refiact, but also
change their amplitude and fiequency of oscillation. In other words, the waves
are filtered and amplified (or attenuated) when passing through various layers of
rock and soil with different mechanical characteristics. When the seismic waves
finally reach the surface and induce vibration to buildings, they reflect not only
the characteristic of earthquake source and mechanism, but also the
characteristics of the bedrock and soil on their way of propagation to the site of
construction (Fig. 2.4).

Angle of incidence

Figure 2.4. Propagation of seismic waves from the rock to the surface.
Earthquakes and Seismic Performance of Masonry Buildings 11

2.2.3 Magnitude and intensity

The impact of an earthquake on built environment is closely related to the

amount of energy released in the focus. The measure of released energy is
magnitude M, which was first defined by Richter in 1935, and is therefore called
“Richter’s magnitude”. According to Richter, the magnitude of an earthquake M
is given by a logarithm of a maximum displacement amplitude A (in pm),
recorded by a standardised instrument, located at exactly 100 km from the
epicentre :

Since it is never the case that the standardised instrument is located at a pre-
defined distance from the epicentre, the magnitude is calculated by taking into
account the actual distance from the epicentre, as well as the characteristics of
the propagation of the seismic waves.
Richter’s magnitude is also called “local magnitude” M - . Namely, the
characteristics of seismic waves change with increased distance fkom the focus.
In an earthquake record, arrival times of body waves and surface waves can be
clearly distinguished. Therefore, new measures of magnitude have been
introduced on the basis of the measurements of amplitudes of body waves
(magnitude Mb) and surface waves (magnitude Ms). Unfortunately, the measures
do not yield the same numerical values. For example, magnitude Mscorrelates
with magnitude ML in the range of magnitudes 6-6.5. In the case of stronger
earthquakes, however, magnitude Ms is greater than ML. In order to avoid
confusion, Richter’s (local) magnitude ML is usually used for moderate
earthquakes with magnitude M = 6.5 or less, whereas magnitude Msis used in the
case of stronger earthquakes.
The size of an earthquake, measured by magnitude, is directly related to the
amount of released energy. Fortunately, a great part of the energy is dissipated in
the process of crushing and warming rock formations during an earthquake, or
transformed into potential energy which will generate future displacements within
the fault zone. Only a minor part of the energy is used to generate seismic waves.
Many formulae have been developed to assess the relationship between the
energy of seismic waves E in joules and the magnitude M. An example is given
below [6J :
log E = 4.8 + 1.5 M.
12 Earthquake-Resistant Design of Masonry Buildings

As can be seen, the energy is increased by 32 times if the magnitude is

increased by 1, and by 1000 times if M is increased by 2.
There are various ways to measure earthquakes and their effects on people,
built environment, and nature. The effects of earthquakes, which can be observed
on the surface of the earth’s crust, are measured by means of various intensity
scales. A 12-grade Mercalli-Cancani-Sieberg (MCS) scale has been used in
th
Europe from the beginning of the 20 century, a 12-grade modified Mercalli
(MM) scale and 8-grade Japan Meteorological Agency (JMA) scale are still used
in the US and Japan, respectively. A 12-grade Medvedev-Sponheuer-Karnik
(MSK-64), originally proposed in 1964 and subsequently modified (MSK-76,
MSK-78), has been suggested as an example of harmonisation of seismic
intensity scales. In MSK scale, the types of buildings are well defined, and
severity of damage, as well as the amount of damage are quantified to make the
assessment of intensity consistent.
Seismologists tried to propose a relationship between the magnitude M and
epicentral intensity lo of an earthquake in dependence on the depth of the focus h.
The relationship is expressed by an empirical equation of the form:
A4 = a lo + b log h + c, (2.3)
where a, b, and c are constants depending on the zone, and should be determined
on the basis of analysis of historical and instrumental data.
Intensity scales, also called macroseismic scales, are based on observations
of damage to traditionally built buildings, impact of earthquakes on the
environment, and human feelings. Since the typology of buildings affected by
earthquakes varies from country to country, the correlation between intensities
assessed by different scales is not always clear. The correlation is not clear even
in cases where the same intensity scale has been applied to earthquake-damaged
buildings in different countries, where there is a substantial difference in building
typology and quality of construction. For example, a stone-masonry house in
China or India is not the same as a stone-masonry house in an European historical
urban centre.
Taking this into consideration, and realising that the number of buildings
designed by seismic codes, which should not have suffered substantial damage
during strong earthquakes, already prevails in many earthquake-prone areas, a
new 12-grade European macroseismic scale (EMS), which is a modification of
the MSK scale, has been recently developed [7].
Earthquakes and Seismic Performance of Masonry Buildings 13

I Classification of damage to masonry buildings

Grade 1: Negligible to sliglit damage

(no structural damage)
Hair-line cracks in very few walls;
fall of small pieces of plaster only.
Fall of loose stones from upper
parts of buildings in very few
cases only.

Grade 2: Moderate damage (slight

structural damage, moderate
non-structural damage)
Cracks in many walls; fall of
fairly large pieces of plaster;
parts of chimneys fall down.

3 a d e 3: Substantial to heavy damage

(moderate structural damage,
heavy non-structural damage)
Large and extensive cracks in
most walls; pantiles or slates slip
off. Chimneys are broken at the
roof line; failure of individual
non-structural elements.

Grade 4: Very heavy damage (heavy

structural damage, very heavy
non-structural damage)
Serious failure of walls; partial
structural failure.

Grade 5 : Destruction (very heavy

structural damage)
Total or near total collapse.

Figure 2.5. EMS scale - classification of damage to masonry buildings (after [S]).
14 Ear thq uake-Resis tant Design of Masonry Buildings

Vulnerability class
Type of
masonry
A B
--
Rubble stone; 0
field stone
Adobe (earth brick) 0 -1
Simple stone t- 0
Massive stone
Unreinforced brick/
t
concrete block
-_
t- 0.
Umeinforced brick
with r.c. floors t
Reinforced brick
(confined masonry)

Figure 2.6. EMS scale - differentiation of masonry structures into vulnerability classes (&er [8]).

In the new EMS scale, the definitions are based on:

(a) Effects on humans.
(b) Effects on objects and nature (excluding damage to buildings, effects on
ground and ground failure).
(c) Damage to buildings.
In this scale, damage to masonry and reinforced concrete buildings is
classified in a very detailed way (Fig. 2.5). To account for seismic-resistant
design, however, the buildings at risk are classified according to their expected
seismic vulnerability (Fig. 2.6). Definitions of quantity are also specified: few =
10 %, many = 20-50 %, and most = 60-100 %. There is a 10 % overlapping
between the categories.
The following classification of earthquake intensities is proposed in the new
EMS scale:
I. Not felt:
(a) Not felt even under the most favourable circumstances.
(b) No effect.
Earthquakes and Seismic Performance of Masonry Buildings 15

No damage.
11. Scarcely felt:
The tremor is felt only by a very few (less than 1 %) individuals at rest and in
a specially receptive position indoors.
No effect.
No damage.
111. Weak:
The earthquake is felt indoors by a few. People at rest feel a swaying or light
trembling.
Hanging objects swing slightly.
No damage.
IV. Largely observed:
The earthquake is felt indoors by many and outdoors only by very few
people. A few people are awakened. The level of vibration is not frightening.
The vibration is moderate. Observers feel a slight trembling or swaying of the
buildings, room or bed, chair, etc.
China, glasses, windows and doors rattle. Hanging objects swing. Light
furniture shake visibly in a few cases. Woodworks creak in a few cases.
No damage.
V. Strong;:
-

The earthquake is felt indoors by most, outdoors by few people. A few

people are frightened and run outdoors. Many sleeping people are awakened.
Observers feel a strong shaking or rocking of the whole building, room or
furniture .
Hanging objects swing considerably. China and glasses clatter together.
Small, top-heavy andor precariously supported objects may be shifted or fall
down. Doors and windows swing open and shut. In a few cases window panes
break. liquids oscillate and may spill from well-filled containers. Animals
indoors may become uneasy.
Damage of grade 1 to a few buildings.
VI. Sligh&diamaging
Felt by most people indoors and by many outdoors. A few people lose their
balance. Many people are frightened and run outdoors.
Small objects of ordinary stability may fall and furniture may be shifted. In
few instances dishes and glassware may break. Farm animals (even outdoors)
mav be frightened.
16 Earthquake-Res istant Design of Masonry Bu ildings

(c) Damage of grade 1 is sustained by many buildings; a few suffer from grade 2
damages.
VII. Damaging:
(a) Most people are frightened and try to run outdoors. Many find it difficult to
stand, especially on the upper floors.
(b) Furniture is shifted and top-heavy furniture may be overturned. Objects fall
from shelves in large numbers. Water splashes from containers, tanks and
pools.
(c) Many buildings of vulnerability class B and a few of class C suffer damage
of grade 2. Many buildings of class A and a few of class B suffer damage of
grade 3; a few buildings of class A suffer damage of grade 4. Damage is
particularly noticeable in the upper parts of buildings.
VIII. Heavilv damaging:
(a) Many people find it difficult to stand, even outdoors.
(b) Furniture may be overturned. Objects like TV sets, typewriters etc. fall to the
ground. Tombstones may occasionally be displaced, twisted or overturned.
Waves may be seen on very soft ground.
(c) Many buildings of vulnerability class C suffer damage of grade 2. Many
buildings of class B and a few of class C suffer damage of grade 3. Many
buildings of class A and a few of class B suffer damage of grade 4; a few
buildings of class A suffer damage of grade 5 . A few buildings of class D
suffer damage of grade 2.
IX. Destructive:
(a) General panic. People may be forcibly thrown to the ground.
(b) Many monuments and columns fall or are twisted. Waves are seen on soft
ground.
(c) Many buildings of vulnerability class C suffer damage of grade 3. Many
buildings of class B and a few of class C suffer damage of grade 4. Many
buildings of class A and a few of class B suffer damage of grade 5 .
(d) Many buildings of class D suffer damage of grade 2; a few suffer grade 3
damages. A few buildings of class E suffer damage of grade 2.
X . Very destructive:
(c) Many buildings of vulnerability class C suffer damage of grade 4. Many
buildings of class B and a few of class C suffer damage of grade 5 , as do
most buildings of class A.
Earthquakes and Seismic Performance of Masonry Buildings 17

Many buildings of class D suffer damage of grade 3; a few suffer grade 4

damages. Many buildings of class E suffer damage of grade 2; a few suffer
grade 3 damages. A few buildings of class F suffer damage of grade 2.
XI. Devastating:
(c) Most buildings of vulnerability class C suffer damage of grade 4. Most
buildings of class B and many of class C suffer damage of grade 5 .
Many buildings of class D suffer damage of grade 4; a few suffer grade 5
damages. Many buildings of class E sufer damage of grade 3; a few suffer
grade 4 damages. Many buildings of class F suffer damage of grade 2, a few
suffer grade 3 damages.
XII. Completely devastating:
(c) Practically all structures above and below ground are destroyed.
After being used for a three-year test period, the ESM scale will be
introduced as an international standard by the European Seismological
Commission.

2.2.4 Occurrence of earthquakes

Earthquakes do not occur regularly in space and time. Whereas possible

locations of earthquake sources can be reasonably well defined on the basis of
geological studies and the plate tectonics theory, the prediction of occurrence of
earthquakes at a given location in time needs further correlation studies and
instrumental observations.
In the case of sufficient number of historical data about earthquakes in the
zone, correlation between the magnitude M and number n of earthquakes with
that, or a greater, magnitude can be found in the form
log n =a - b M, (2.4)
where a and b are constants, depending on the zone and time interval [ 6 ] .On the
basis of this correlation and assuming the Poisson’s model of distribution, the
probability of occurrence of an earthquake can be estimated. There are usually
two ways of determining the probability:
By assessing the return period R of an earthquake, and
By assessing the probability of the occurrence of the event PT in a given time
period .
The return period R of occurrence of an earthquake is determined by an
average time interval between two events of a given (or greater) magnitude. For
18 Earthquake-Resistant Design of Masonry Buildings

example, in the case where the return period of an earthquake is 100 years, an
earthquake of a given magnitude is expected to occur 10 times in the period of
1000 years [81.
Consequently, the probability of occurrence of an earthquake with return
period R in one year is P1 = 1/R (0.01 for an earthquake with 100 years return
period) , and the probability that it will not occur in the same year is 1 - Pi (0.99
for an earthquake with a return period of 100 years). These relationships can be
used to assess the probability of occurrence in a longer period of time. Where the
period of time is T years, the probability that the earthquake with a 100-year
T
return period will not occur is (1 - Pi) . It can be seen that probability that the
earthquake of a given or greater magnitude will occur in a given time interval T,
is given by:
PT= 1 - ( I -PI)T. (2.5)
For example, the probability that a 100-year return period earthquake will
occur within the time interval equal to the return period, i.e. 100 years, is
Ploo = 1 - (1 - l / l O O ) 100 = 0.63.
The probability that the very same earthquake will not occur is 1 - PT= 0.37.
It is interesting to see that the probability of occurrence of an earthquake with a
given return period R within the time interval equal to the return period R, is
always 0.63.
Since the probability of occurrence of a 475-year return period earthquake
within the time interval of 100 years is still relatively high,
I00
1 - (1 - 11475) = 0.19,

this return period is considered when assessing the design seismic loads for
seismic resistance verification of buildings of normal importance.

2.2.5 Earthquake ground motion and eflects on buildings

Observed intensity reflects the effects of earthquakes on buildings. However,

as a measure of earthquakes, it can only be used to assess the damage to similar
type of buildings when subjected to earthquakes of similar intensity. The
intensity of the earthquake cannot be used as a design parameter, unless it is
correlated with physical quantities used in the design (accelerations, forces).
Peak (or maximum) g r o u d acceleration ag is often considered as an obvious
Earthquakes and Seismic Performance of Masonry Buildings 19

physical parameter to determine the intensity of earthquakes. Forces acting on the

building during an earthquake are induced by ground motion and depend on the
intensity of the motion.
Therefore, a correlation between the seismic intensity and expected values of
peak ground acceleration ag that could be used in earthquake-resistant design of
buildings, has been proposed already in the MSK-64 intensity scale. The
correlation has been improved in fhther developments of the scale. A formula
that correlates the values of peak ground acceleration ag and intensity I has also
been developed for MM intensity scale [9]:

In order to obtain an idea about their magnitude, the values of ground

accelerations are often given in terms of acceleration of gravity, g = 9.81 ms-2.
Some typical proposed values are given in Table 2.2.

Table 2.2. Average values of peak ground acceleration ag(in % of gravity) in dependence on intensity.

1
[Intensity scale Grade VIIIGradeVIII 11 -
I MSK-64 I 5-10 I 10-20 I 20-40 I
MSK-76,78 10 20 40
Eq. (2.6) 10 18 32

However, forces induced in the structure depend not only on the intensity of
the seismic input, but also on the response of the structure to seismic ground
motion. For example, in the case of the Friuli earthquake of 1976, where peak
ground accelerations of more than 0.5 g have been recorded, substantial damage
occurred to rigid masonry houses, whereas flexible r.c. structures resisted the
earthquake with only minor damages incurred. The situation in Mexico City was
different: although peak ground acceleration did not exceed 0.17 g, many flexible
high-rise buildings collapsed, whereas rigid masonry building remained
undamaged.
Generally, seismic ground motion is complex and tri-dimensional, so special
seismological instruments are needed to record the components of the motion.
Usually, accelerations in the North-South, East-West and vertical directions are
recorded, and velocities and displacement are calculated on the basis of these
records. Typical acceleration records of only one component of the ground
20 Earthquake-Resistant Design of Masonry Buildings

motion of characteristic short- and long-period earthquakes are shown in Fig. 2.7
[lo, 111.
From these records, the influence of foundation soil on the characteristics of
the seismic ground motion at the site can be clearly seen. Within the last decades,
comprehensive studies have been carried out to investigate the propagation of
seismic waves from the bedrock to the surface of the ground. As shown in Fig.
2.4, when passing the stratified layers of soil, seismic waves refract and reflect,
but also amplify or attenuate. Stochastic characteristics of seismic waves,
generated in the focus, are filtered on the way to the surface: depending on the
characteristics of the soil, significantly different records of the same earthquake
can be obtained. If the waves have passed through soft soils, their frequency
characteristics become narrow, with longer periods, amplified amplitudes of
vibration, and longer duration of strong motion. If however, they travel along
bedrock, their stochastic characteristics are retained.

I I I I

5 10 15 20

0 30 60 90 120 150 180

6)

Figure 2.7. Acceleration records of (a) N-S component of Petrovac record of the Montenegro Earthquake
of 1979, and (b) N90W component of SCT record of the Mexico City Earthquake of 1985.

Many parameters can be calculated on the basis of these records in order to

analyse the ground motion. As the damage to buildings is correlated with the
amount of energy, which is a velocity-dependant quantity, peak ground velocity
as well as different measures of energy-based intensity parameters, such as
Housner spectrum intensity and Arias intensity, are calculated [12, 131. The
Earthquakes and Seismic Performance of Masonry Buildings 21

duration of the strong phase of ground motion and predominant periods of

vibration represent the parameters, which are also important in the process of
structural verification and design.

Damping: 0,2,5,10 and 20 ?4of critical

0
0 0.5 1.0 1.5 2.0 2.5
Period (s)

Figure 2.8. Acceleration response spectra of N-S component of Petrovac record

of Montenegro Earthquake of 1979 (after [101).

40

Damping: 0,2,5, 10 and 20 % of critical

0
0 1 2 3 4 5
Period (s)

Figure 2.9. Acceleration response spectra of N90W component of SCT record

of the Mexico City Earthquake of 1985 (after [1 11).
22 Earthquake-ResistantDesign of Masonry Buildings

However, most data needed in earthquake resistance verification and design

can be obtained by calculating the response spectra [14]. The response s p e c t m
is defined as a relationship between the maximum response of a single-degee-of-
freedom (SDOF) system to a given ground motion and the system’s natural
period of vibration and damping. The dependence of acceleration response
amplitudes on the type of ground motion can be clearly seen in Figs. 2.8 and 2.9,
which show the response spectra derived from ground acceleration records s h o w
in Fig. 2.7.
Masonry buildings represent typical rigid structures with short periods of
vibration. In such a case, most data needed for structural verification and design
can be obtained directly from ground acceleration records, which in many cases
simplifies the assessment of seismic loads and seismic resistance verification. As
the analyses of seismic behaviour of masonry buildings show, the values of
effective peak ground accelerations are in good correlation with the induced
seismic forces and observed damage to buildings. For earthquake resistance ana-
lysis and design, the actual response spectra, obtained on the basis of a great
number of ground acceleration records, are statistically evaluated, averaged and
smoothed. Also, to take into account the non-linear behaviour of structures,
elastic response spectra are further reduced on the basis of either the ductility or
energy dissipation capacity of the structure under consideration (see Chapter
3.2).
In order to separate the effects of soil and intensity of earthquakes, response
spectra are normalised with regard to maximum ground acceleration. In such a
case, response spectra represent amplification of the ground motion to be taken
into consideration for structures with different dynamic characteristics, whereas
design ground acceleration values represent the expected intensity of earthquakes
with a specified return period of occurrence. The shape of the elastic response
spectrum, proposed to determine the design seismic action by Eurocode 8 : Design
provisions for earthquake resistance of structures [ 151, is shown in Fig. 2.10.
The values &(T) are determined by a set of expressions, depending on the
first natural period of vibration of the structure under consideration:

if 0 < T < TB: (2.6a)

(2.6b)
Earthquakes and Seismic Performance of Masonry Buildings 23

(2.6~)

if TD < T: (2.6d)

where:
Se( T ) = the ordinate of the elastic response spectrum,
T = the vibration period of a linear SDOF system,
ag = the design ground acceleration for the reference return period (475 years
for buildings with importance factor 1.O),
S = the soil parameter with reference value 1 .O for subsoil class A,
q = the damping correction coefficient with reference value 1.0 for 5 %
viscous damping,
Po = the maximum normalised spectral value assumed constant between TB
and Tc (Po= 2.5), and
k l = 1, k2 = 2 = the exponents influencing the shape of the elastic response
spectrum.

I B C

Figure 2.10. Shape of the EC 8 elastic response spectrum.

The values of parameters describing the elastic response spectrum are given
in Table 2.3. As can be seen, the shape of the spectra depends on the subsoil
characteristics. Three classes of subsoil are defined in EC 8:
- Subsoil class A is represented by rock or other geological formation,
characterised by a shear wave velocity of at least 800 m/s, including at most
5 m of weaker material at the surface. Stiff deposits of sand, gravel or over-
24 Earthquake-Resistant Design of Masonry Buildings

consolidated clay, up to several tens of metres thick, characterised by gradual

increase of mechanical properties with depth, and by shear wave velocity at
10 m depth of at least 400 m/s, are also classified as class A.
- SubsoiZ class B is represented by deep deposits of medium dense sand, gravel
or medium stiff clays with thickness from several tens to many hundreds of
metres, characterised by minimum value of shear wave velocity increasing
from 200 m / s at a depth of 10 m, to 350 m / s at a depth of 50 m.
Subsoil class C is represented by loose cohesionless soil deposits with or without
soft cohesive layers, characterised by shear wave velocity values below 200 d s
in the uppermost 20 m. Deposits with predominant soft-to-medium stiff cohesive
soils, characterised by shear wave velocities below 200 m / s in the uppermost 20
m, are also categorised as subsoil class C.

Table 2.3. Values of parameters describing elastic response spectrum.

Subsoil class S PO kl k2 TB TC TD
A 1.o 2.5 1.o 2.0 0.10 s 0.40 s 3.0 s
B 1.o 2.5 1.o 2.0 0.15 s 0.60 s 3.0 s
C 0.9 2.5 1.o 2.0 0.20 s 0.80 s 3.0 s

EC 8 recommends that the design ground acceleration values correspond to

earthquakes with a reference return period of occurrence of 475 years. However,
no recommendation is given in the document with regard to the values that would
be adequate for the zones of low, moderate, and high seismic intensity,
respectively. The values will be determined by each country in her National
Application Document to EC 8. Therefore, the values of effective ground
accelerations, proposed in the MSK seismic intensity scale (Table 2.2) are given
as indicatives to be taken into account for the following seismic resistance
verification:
High intensity, seismic intensity zone IX: ag = 0.4 g,
Moderate intensity, seismic intensity zone VIII: ag = 0.2 g,
Low intensity, seismic intensity zone VII: ag = 0.1 g.
Since masonry buildings are rigid structures, the amplification defined by the
flat part of the elastic response spectrum (Eq. 2.6b) will be most often taken into
account.
Earthquakes and Seismic Performance of Masonry Buildings 25

2.3 Seismic Performance of Masonry Buildings

2.3.I Structural typology

Masonry has been used as a handicracft material of old. With the exception
of monumental buildings, which have sometimes been designed on the basis of
experiments and the simple theory of structures, masonry buildings have been
built on the basis of tradition and experience. Structural walls, in cases where
large openings have been needed, replaced by arches, have been provided to
carry the vertical loads, resulting from floors. Not so often, measures to improve
the earthquake resistance, developed on the basis of experience, have been
applied.
According to available materials, climatic and functional requirements,
technical knowledge and traditional practices specific to different countries, a
variety of masonry typologies can be found. Masonry buildings can be classified
according to materials used for construction (adobe, stone, brick, block) and
structural system (plain, confined, reinforced masonry), place of construction
(rural, urban), period of construction (historical, prior to World War I, between
both World Wars, post-war period, after the adoption of seismic codes), and use
of buildings (residential, public).
In many zones of high seismic risk in Asia and Latin America, mud- or sun-
dried brick (adobe) and stone is still used as the main structural material for
residential construction. Timber fiame, infilled with adobe or brick, represents an
improved structural system. Floors and roofs are usually wooden, with freely
supported joists. Roofs are often covered with heavy earthen topping.
In Mediterranean and Central European countries, the buildings in the
historical urban nuclei are also traditionally built stone- or brick-masonry
buildings. The buildings of this type are 3-4 storeys high and are built in clusters
(Fig. 2.11). They usually have a regular structural layout, with thick walls
uniformly distributed in both directions. However, the floors are wooden and the
wall ties are often omitted. Typically, floor structures above the ground floor, the
staircases and corridors are brick vaults.
Historical brick-masonry buildings have similar structural characteristics as
th
stone-masonry buildings. At the turn of the 19 century, the structural layout,
especially that of public buildings, is often irregular, with many offsets and
setbacks. Poor quality lime mortar and wooden floors, sometimes replaced by
brick vaults supported by steel beams, were the reasons for poor behaviour of
such buildings during earthquakes at the turn of the century. At that time,
26 Earthquake-Resistant Design of Masonry Buildings

however, specific regulations have been issued to improve the construction of

brick-masonry buildings to be earthquake resistant. Requirements for the
minimum thickness and distribution of structural walls in plan, as well as the
guide-lines for anchoring the floors and tying of the walls with steel strip ties and
anchors have been introduced.

Figure 2.1 1 . Typical buildings in the historical Figure 2.12. Typical brick masonry condominium,
centre of the city of Ljubljana. built in the thirties (Ljubljana).

After World War I, r.c. tie-beams, sometimes accompanied by monolithic r.c.

or prefabricated masonry floor slabs have been implemented. The number of
stories increased to 6-7, and storey height to 3.5-4 m (Fig. 2.12). Mixed
structural systems are often found, using inner r.c. columns as load-bearing
elements. In that case, the number of storeys is less, but the storey height often
exceeds 4 m.
During the post-World War I1 period of reconstruction of cities and t o w s in
many earthquake-prone regions, apartment buildings up to six stories high0 have
ofetn been constructed with load-bearing walls in the transverse direction only.
Longitudinal walls have not been considered as load-resisting elements, because
Earthquakes and Seismic Performance of Masonry Buildings 27

they have been made weak by many window and door openings. Also, not much
attention has been paid to the quality of materials and construction.

Figure 2.13. Typical pre- 1964 seismic code

URM tower block (rear, Ljubljana).

Block-type masonry units have been introduced for load-bearing and

structural walls. Just before the introduction of modern seismic codes in the
sixties, many block-masonry residential tower-blocks more than 10-stories high
have been constructed in unreinforced masonry (Fig. 2.13). However, in the last
several decades, the allowable number of stories for unreinforced masonry
3
buildings has been drastically decreased, and improved earthquake-resistant
systems have been developed and required by seismic codes, such as confined
and reinforced masonry.
28 Earth¶ uake-Resistant Design of Masonry b ii rl dings

2.3.2 Seismic performance and classifjcation of damage

Of the great number of masonry buildings subjected to strong earthquakes,

many were severely damaged and collapsed. Consequently, masonry has long
been considered as an unsuitable material for the construction of buildings in
seismic zones. However, there were cases where some buildings survived the
earthquake only slightly damaged or even undamaged, although they have been
built at the same location as the damaged buildings.
Considering the response to earthquakes, adobe and stone-masonry buildings
suffered severe damage. Especially in the case of stone-masonry, poor quality
mud mortar resulted in the disintegration of masonry and loss of support to
floors. Heavy earthen roof topping, which buries the inside of the building, is
often the main reason for severe consequences of earthquakes. Although
continuous attempts are being made to improve the new construction or
strengthen the earthquake-damaged buildings [ 16, 171, a high toll is usually paid
after strong earthquakes (Fig.2.14).

Figure 2.14. Killari, India, 1993: destroyed village.

With regard to historical stone- andor brick-masonry buildings in urban and rural
nuclei, cracks at the comers and at wall intersections, which occur as a result of
insufficient connections and a lack of connection between the walls and floors,
represent a characteristic damage pattern (Fig 2.15). Sometimes, separation of walls
Earthquakes and Seismic Performance of Masonry Buildings 29

and even out-of-plane collapse occurred (Fig. 2.16). Also, many times, despite the
favourable structural layout of those buildings in plan and good connection of walls,
the quality of masonry materials was not good enough to spare the walls from
diagonal cracking, disintegration, and ultimate collapse.

Figure 2.15. PosoCje, Slovenia, 1976: Figure 2.16. Budva, Montenegro, 1979:
separation of walls at the corners. out-of-plane collapse of perimetral walls.

Irregular structural layout in plan, large openings and poor quality of

masonry materials, especially mortar, often cause severe damage or collapse of
contemporary masonry buildings. In many cases, characteristic diagonal cracks in
load-bearing window piers have been observed as a result of inadequate shear
resistance (Fig. 2.17).
In the case of contemporary masonry buildings, adequate structural layout
turned out to be an extremely important issue. The buildings with structural walls
in one, usually the transverse, direction only, were not able to resist earthquakes
with predominant ground motion in the weak direction of the building (Fig. 2.18).
The behaviour of mixed structural systems was also very poor, because, most
often, formation of cracks in perimetral load-bearing walls resulted in severe
30 Earthquake-Res istant Design 0f Mas onry Buildings

stiffness and strength degradation of the entire structural system. Inner columns,
designed to carry the vertical loads, were not able to withstand the ultimate
lateral loads redistributed from the walls at ultimate state.

Figure 2.17. Budva, Montenegro, 1979: typical shear cracks

in window piers of a brick-masonry building.

Figure 2.18. Neftegorsk, Russia, 1995: large-block-masonry buildings with transverse load-bearing
walls did not resist the earthquake in the longitudinal direction (photo by courtesy of G.Koff).
Earthquakes and Seismic Performance of Masonry Buildings 31

Masonry buildings, designed and constructed according to requirements of

modern seismic codes, behaved adequately. Cases of collapse were rare and were
limited to buildings where the requirements of codes, especially those related to
the quality of construction, were only partly met.
Although the structural typology of masonry buildings varies in different
regions, their damage resulting fiom earthquakes can be classified in an uniform
way. The following typical types of damage can be identified by the analysis of
the observed earthquake damage patterns:
Cracks between walls and floors,
Cracks at the comers and at wall intersections,
Out-of-plane collapse of perimetral walls,
Cracks in spandrel beams and/or parapets,
Diagonal cracks in structural walls,
Partial disintegration or collapse of structural walls,
Partial or complete collapse of the building.

Figure 2.19.Deformation of the building and typical damage to structural wall.

The analysis of damage patterns can clearly identify the weak and good
points of different structural systems. On the basis of damage analysis, the failure
mechanisms of individual structural walls and the entire structural system can be
defined. On the basis of the observed mechanism, forces that develop in the
structural system during earthquakes can be defined. However, any quantitative
data regarding seismic action or structural resistance cannot be assessed unless
additional experimental investigations that simulate the observed behaviour and
damage patterns have been carried out.
32 Eart hq uake-lies istan t Design of Mas onry Buildings

Deformation and the typical type of damage to structural walls of a simple

masonry building, subjected to seismic loads, is shown in Fig. 2.19. As can be
seen, structural walls which are perpendicular to seismic motion are subjected to
out-of-plane bending, causing vertical cracks at the corners and in the middle of
the walls. In the plane of the walls, however, bending and shear caused horizontal
and diagonal cracks in the walls, respectively.
Distribution of forces and stresses, induced in a typical pier of a simple
masonry building, subjected to earthquake ground motion, is shown in Fig. 2.20.

Figure 2.20. Distribution of forces and stresses in window piers.

2.4 References

[l J B.A. Bolt. Earthquakes. A Primer (W.H. Freeman and Co., San Francisco,
1978).
[2] J.M. Gere and H.C.Shah. Terra Non Firma (W.H. Freeman and Co., New
York, 1984).
[31 M. Wakabayashi. Design of Earthquake Resistant Buildings (McGraw Hill,
New York, 1986).
Earthquakes and Seismic Performance of Masonry Buildings 33

L. Seeber, S.K. Jain, C.V.R. Murty and N. Chandak. Surface rupture and
damage patterns in the Ms = 6.4, September 29, 1993 Killari (Latur)
earthquake in central India. NCEER Bulletin, 7 (4) (National Center for
Earthquake Engineering Research, Buffalo, 1993), p. 12.
G.W. Housner and P.C. Jennings. Earthquake Design Criteria (Earthquake
Engineering Research Institute, Berkeley, 1982).
B. Gutenberg and C.F. Richter. Seismicity of the Earth and Associated
Phenomena (Princeton University Press, Princeton, 1954).
G. Griinthal, Ed. New European Macroseismic Scale 1992, (European
Seismological Commission, Luxembourg, 1993).
D. AniEiC, P. Fajfar, B. PetroviC, A. Szavits-Nossan and M. TomaZeviC.
Earthquake Engineering. Buildings (Gradevinska knjiga, Beograd, 1990 -
in Serbo-Croatian).
J.R. Murphy and J.L. O’Brien. The correlation of peak ground acceleration
amplitude with seismic intensity and other physical parameters. Bulletin of
the Seismological Society of America, 67 (3) (1977), pp. 877-91 5.
Bulletin of the Strong Motion Accelerograms, No. 1 (Institute of Earthquake
Engineering and Engineering Seismology, Skopje, 1 984).
Accelerogram, recorded at SCOP Centre of SCT. Earthquake of September
19, 1985. Report IPS - IOB (Instrumentacion Sismica, Instituto de
Ingenieria, UNAM, Ciudad de Mexico, 1985 - in Spanish).
G.W. Housner. Strong ground motion. Earthquake Engineering, R.L.
Wiegel, Ed. (Prentice Hall, Englewood Cliffs, New Jersey, 1970).
A. Arias. A measure of earthquake intensity. Seismic Design of Nuclear
Power Plants. R. Hanson, Ed. (MIT Press., Cambridge, 1970).
N.M. Newmark and W.J. Hall. Earthquake Spectra and Design (Earthquake
Engineering Research Institute, Berkeley, 1 982).
Eurocode 8. Design provisions for earthquake resistance of structures.
Part I-1: General rules for buildings - Seismic actions and general
requirements for structures. ENV 1998-1- 1 : 1994 (CEN, Brussels, 1994).
Guidelines for earthquake resistant non-engineered construction (The
International Association for Earthquake Engineering, Tokyo, 1986).
Improving Earthquake Resistance of Low Strength Masonry Buildings -
Guidelines, Indian Standard IS 13828: 1993 (Indian Standard Institute, New
Delhi, 1993).
 CHAPTER 3

MASONRY MATERIALS AND

CONSTRUCTION SYSTEMS

3.1 Introduction

Masonry is a typical composite construction material which consists of:

Masonry units,
Mortar,
Concrete infill and/or concrete, and
Reinforcing steel.
A wide variety of raw materials, both natural and arWicia1, is used for the
production of traditionally and industrially made masonry units. Mortar is
composed of lime, cement, and sand in different proportions, and mixed with
water, with or without additives. Deformed and smooth steel reinforcing bars of
different shapes and qualities, embedded in mortar or placed in the holes and
grouted, are used to reinforce the masonry. Although each component of a
masonry wall has its own specific mechanical characteristics, they are all
expected to act together as a homogeneous structural material in case they are
subjected to permanent and temporary actions. Of course, not all materials are
always encountered together. Depending on which materials and/or how they
are composed together in a structure, masonry construction systems are further
subdivided in to:
0 Unreinforced (plain) masonry, consisting of mortar and masonry units.
Confined masonry, consisting of masonry units, mortar, reinforcing steel
and concrete, and

35
 Reinforced masonry, consisting of masonry units, mortar, reinforcing steel
and concrete infill.
Because of specific characteristics of each constituent masonry material,
especially masonry units, it is not easy to predict the mechanical characteristics
of a specific masonry construction type by knowing only the characteristics of its
constituent materials, mortar and masonry units. It is therefore of relevant
importance that, for each type of masonry, experiments to correlate the strength
characteristics of constituent materials with the characteristics of masonry are
carried out.
Because of its complexity, masonry and its constituent materials should
comply with specific requirements of standards and codes, especially when they
are used for the construction of engineered structures, where the resistance of
elements and the entire structure to gravity and seismic loads is verified by
calculation. In cases where the mechanical characteristics of the constituent
materials and masonry as structural material do not comply with the assumptions
of numerical verification, unreliable conclusions may be obtained.
The basic requirements to be fulfilled for masonry materials and types of
construction are specified in Eurocode 6: Design of masonry structures [ 11.
Additional requirements for masonry materials and construction systems to be
considered in seismic zones are specified in Eurocode 8 : Design provisions for
earthquake resistance of structures [21. A set of European standards, which
determine the basic properties of masonry materials and the way of testing,
accompanies the Eurocodes.

3.2 Masonry Materials

3.2.I Mmonry units

Apart from the load-bearing capacity, the following aspects should also be
considered when selecting the most suitable type of masonry unit:
- adequate thermal and sound insulation capacity of masonry, especially in the
case of external walls,
- reduction of the weight of the building in order to reduce the seismic loads,
and
- economy of construction.
Adobe and natural stone are still used in some countries and regions for the
construction of plain, low-cost masonry houses. Adobe units (usual dimensions
Masonry Materials and Construction Systems 37

are 300/400/120mm) are made of clayey earth, consisting of 30-40 % clay and
70-60 % earth, which is mixed with water. Straw is added to the mix of clay and
earth (7-10 kg per m3 of earth) in order to prevent cracking. Sometimes, cement
or gypsum is added as a stabiliser in order to improve the compressive strength of
the units. The units are dried under the sun [3].
Besides masonry units manufactured from traditional and modem materials,
different kinds of natural stone (lime-stone, slate etc.) are still used for the
construction of masonry walls. According to EC 6 and EC 8, only the use of
dimensioned stone units, i.e. square dressed units with parallel horizontal faces,
is allowed for the construction of masonry buildings in seismic zones. Traditional
stone-masonry construction with two outer layers of uncoursed stone with an
inner infill of rubble stones is not considered to be earthquake resistant. In any
case, natural stone units are considered as units, where Category I1 of
manufacturing control values of partial safety factors for material properties are
taken into account.
EC 6 gives specifications regarding the use of:
Fired clay units, including lightweight clay units.
Calcium silicate units.
Concrete units, made with dense or lightweight aggregates.
Autoclaved aerated concrete units.
Manufactured stone units.
Dimensioned natural stone units.
In all cases, the quality of masonry units should comply with the
requirements of relevant European standards (EN 77 1-1-6, respectively).
Requirements regarding materials and manufacture, dimensions and tolerances,
density, mechanical strength, water absorption, frost resistance, soluble salts
content, safety in case of fire, etc., are specified for each type of masonry units.
Solid or perforated, hollow, cellular and horizontally perforated masonry
units with different dimensions and shapes, made of different materials and suited
to different construction systems, are produced. Solid or equivalent solid
masonry units are either units without recesses or units with recesses that are
filled with mortar during construction, or units with up to 25 % by volume of
vertical holes. Perforated masonry units are units with more than 25 % but less
than 50 % by volume of formed vertical holes, which may pass through the unit.
Hollow masonry units are units with more than 50 % by volume of vertical holes.
Horizontally perforated units are units, which have no more than 50 % by cross-
38 Earthquake-Resistant Design of Masonry Buildings

sectional area of formed horizontal holes, which pass through the units. Typical
shapes of masonry units are shown in Fig. 3.1.
In the case of hollow block masonry units, the disposition and size of holes
should be limited in order to avoid premature cracks in webs and shells, either
during manufacture and handling, or use. With regard to the total volume of
holes, volume of any hole, area of any hole, as well as combined thickness of
webs and shells, EC 6 classifies the units into four groups (Table 3.1).

Perforated unit
BHollow unit

Horizontally perforated unit

Cellular unit

Figure 3.1. Types of masonry units (after [4)).

This classification is used to select the appropriate value of correction factors

K in cases where the characteristic compressive fi( and shear strength of the
masonry fvk are assessed by calculation on the basis of empirically obtained
correlation between the normalised compressive strength of masonry units Jb and
mortarf, (Eqs. 3.2 and 3.4).
In addition to that and in order to avoid local brittle failure of hollow units
under a combination of vertical and lateral loads, EC 8 requires that masonry
units to be used for earthquake-resistant construction of masonry buildings shall
also meet the following requirements:
- The units have not more than 50 % of holes (in % of the gross volume).
- The minimum thickness of shells is 15 mm.
- The vertical webs in hollow and cellular units extend over the entire
horizontal length of the unit.
Masonry Materials and Construction Systems 39

Table 3.1. EC 6 requirements for the grouping of masonry units.

Group of masonry units

1 2a I 2b 3
Volume of > 25-45 for > 45-55 for
holes (% of clay units, clay units,
I25 > 25-50 for > 50-60 for I70
the gross
volume)' concrete concrete
aggr. units aggr. units
Volume of 5 12.5 for S 12.5 for
any hole(% clay units, clay units, Limited by
5 12.5 I 25 for I25 for are
of the gross
volume) concrete concrete cL (see below)
aggr. units aggr. units
2 2 800 mm2
except units
~

Area Limited by Limited by Limited by

of any hole volume volume volume with a single
(see above) (see above) (see above) hole when the
hole should be
_____ ~ ~ ~ _ _ _ _ _ _ ~
I 1 8 000 mm2
Combined
thickness (YO 2 37.5 2 30 2 20 No
of the overall requirement
width)3
Notes:
1. Holes may consist of formed vertical holes through the units or frogs or
recesses.
2. If there is national experience, based on tests, that confirms that the
safety of the masonry is not reduced unacceptably when a higher
proportion of holes is incorporated, the limit of 55 % for clay units and
60 % for concrete aggregate units may be increased for masonry units
that are used in the country with the national experience.
3 . The combined thickness is the thickness of the webs and shells,
measured horizontally across the unit at right angles to the face of the
wall.

Relatively low minimum mean values of compressive strength of masonry

units to be used for the construction of structural walls are specified in relevant
standards (EN 77 1- 1-6):
- Clay units: minfb = 2.5 MPa.
- Calcium silicate units: minfb = 5.0 MPa (normalised value).
40 Earthquake-Resistant Design of Masonry Buildings

- Concrete aggregate units: minfb = 1.8 MPa.

- Autoclaved aerated concrete units: minfb = 1.8 MPa.
- Manufactured stone units: minfb = 15 MPa.
According to EC 8, the minimum normalised compressive strength of a
masonry unit, normal to the bed face, is = 2.5 MPa. Nevertheless, it is
recommended, that in the case of hollow clay and concrete aggregate units, the
mean values of the units’ compressive strength should not be kept below 7.5
MPa, especially where the units are used for the construction of reinforced
masonry walls. The reason for such recommendation is explained in Section
3.3.2.
According to EC 6, the so called “normalised compressive strength of
masonry units”fb should be used in the design. This is the mean value of a
reference strength determined by testing of at least ten (six in the case where the
coefficient of variation is not greater than 15 %) equivalent, air dried, 100-mm
wide and 100-mmhigh specimens cut from the unit (for example, this is the way
to test calcium silicate and autoclaved aerated concrete units). If the strength is
obtained by testing full-sized units, the mean value of strength is multiplied by
the shape factor 6, which takes into account the actual dimensions of the unit. In
case the compressive strength of masonry units is specified as characteristic
strength, this should be first converted to the mean equivalent using a conversion
factor based on the coefficient of variation, and then multiplied by the shape
factor 6. The values are given in Table 3.2.

Table 3.2. Shape factor 6 for conversion of mean value of unit’s strength to normalised value (EC 6),

Height Least horizontal dimension (mm)

(mm) 50 I 100 I 150 I 200 I >250
50 I 0.85 I 0.75 I 0.70 I - ~~~
I -
65 0.95 0.85 0.75 0.70 0.65
100 1.15 1 .oo 0.90 0.80 0.75
150 1.30 1.20 1.10 1 .oo 0.95
200 1.45 1.35 1.25 1.15 1.10
>250 1.55 1.45 1.35 1.25 1.15

Interpolation of values of the shape factor 6 is permitted where the

dimensions of the unit are different than those specified in Table 3.2.
Masonry Materials and Construction Systems 41

3.2.2 Mortar

Mortar is a mixture of inorganic binders (lime and/or cement), aggregates and

water, which binds together masonry units. Sometimes, additives are added to
mortar to improve its workability, or for other reasons.
According to the classification used in EC 6, different types of mortar are
used in masonry construction:
General purpose mortar, which is the traditional type of mortar used in joints
with a thickness greater than 3 mm and in which only dense aggregate is
used.
Thin layer mortar, which is intended for use in masonry with a nominal
thickness of joints 1-3 mm. Thin layer mortars are typical mortars, designed
to fulfill the specific requirement of masonry construction.
Lightweight mortar is made using perlite, pumice, expanded clay, expanded
shale, or other lightweight materials. Lightweight mortars are also designed
to fulfill specific requirements of masonry construction, and have a dry,
hardened density lower than 1500 kg/m3.
In order to attain the specified compressive strengthf,, mortar mixes can be
either prescribed on the basis of experience or designed. In Table 3.3, typical
compositions of prescribed general purpose mortar mixes and expected mean
values of compressive strength are indicated [4].

Table 3.3. Typical prescribed composition and strength of general purpose mortars [4].

Mean Approximate composition in parts of volume

Mortar compressive
tYPe strength Cement Hydrated lime Sand
M2 2.5 MPa 1 1.25 - 2.50
M5 5 MPa 1 0.50 - 1.25 2.25-3-times
M10 10 MPa 1 0.25 - 0.50 cement and lime
M20 20 MPa 1 0 - 0.25

According to EC 8, for the construction of unreinforced and confined

masonry, only mortars of type M5 (compressive strength& = 5 MPa) or stronger
are allowed. If weaker mortars are used, disintegration of masonry may take place
when subjected to vibration during a strong earthquake. In the case of reinforced
masonry, however, the minimum allowable mortar type is M10 (compressive
42 Earthauake-Resistant Design of Masonry Buildings

strengthf, = 10 MPa) to ensure the assumed transfer of internal forces from steel
reinforcement to masonry.
In the case where reinforcement is embedded in mortar, the recommended
values of characteristic bond strength fbok for typical classes of mortars, which
can be assumed in the design, are specified in Table 3.4.

Table 3.4. Characteristic anchorage bond strength of reinforcement in mortar (EC 6).

I Mortar I M5-M9 1 M10-Ml4 I M15-Ml9 I M20

I fbok for plain
bars (MPa)
I 0.7 1 1.2 I 1.4 I 1.5
fbok for high- 1 .o 1.5 2.0 2.5
bond bars (MPa)

Mechanical properties of mortar are determined by testing mortar prisms

40/40/160 mm (EN 1015-1 1). The prisms are first tested for bending, and then
for compressive strength. The compressive strength of the mortar is the mean
strength of six specimens.
Generally, the thickness of bed and perpend (head) joints in the case of
general purpose and lightweight mortars should be between 8 mm and 15 mm. If
such is the case, Eqs. (3.2) and (3.4) can be used to assess the characteristic
compressive strengthh and shear strength of the masonryf&. Except in zones of
low seismicity, all perpend joints should be fully filled with mortar!

3.2.3 Concrete infill

A concrete mix of suitable consistency and aggregate size, which is used to

fill the holes where reinforcing steel bars are placed in the case of reinforced
masonry, is called concrete infill or grout. According to specifications given in
EC 6, the maximum aggregate size should not exceed 10 mm where the least
dimension of the void is 50 mm and the cover to reinforcement is between 15 mm
and 25 mm, and should not exceed 20 mm where the dimensions are not less than
100 mm or 25 mm, respectively.
In order to ensure workability and pourability of the infill, sufficient water
should be added to the mix. However, in order to reduce the risk of cracking in
the infill due to shrinkage resulting from loss of water into the masonry,
expanding agents may be used.
Masonry Materials and Construction Systems 43

The characteristic compressive cylinder or cube strength of the concrete infill

f c k should not be less than 12 MPa or 15 MPa, respectively. The values of
characteristic compressive cubekylinder strength of concrete infill fck, as well as
the values of characteristic shear strength of concrete infillfcvk to be used in the
design and seismic resistance verification, are specified in Table 3.5 for the
relevant concrete design classes.

Table 3.5. Characteristic compressivef,k and shear strength&k of concrete infill (EC 6).

Strength class of 12/1 5 C 16/20 C20/25 C25/30or

concrete stronger I

fck ( m a ) 12 16 20 25
fcvk (ma) 0.27 0.33 0.39 0.45

3.2.4 Reinforcing steel

Smooth or high bond (deformed) reinforcing steel with distinct yield plateau,
which is used for reinforced concrete structures, is also used for reinforced and
confined masonry construction. Especially shaped prefabricated ladder-type or
truss-type reinforcement is often placed in mortar bed joints (Fig. 3.2).

Figure 3.2. Typical ladder- and truss-type prefabricated bed joint reinforcement.

According to EC 6, reinforcing steel may be assumed to possess adequate

elongation ductility, if the following requirements are satisfied:
- For high ductility class: E U >~ 5 % and (ft lfy)k > 1.08.
- For normal ductility class: Euk > 2.5 % and (ft lfy)k > 1.05,
44 Earthquake-Resistant Design of Masonry Buildings

where:
EUk = the characteristic value of the unit elongation at maximum tensile
stress,
ft = the tensile strength of the reinforcing steel,
fy= the yield strength of the reinforcing steel, and
lfy)k = the characteristic value of ft lfy.

a(MPa)
800
t fy = 500 MPa

600

400

200

0 ,
0 0.05 0.10 0.15 0.20 0.25 E

Figure 3.3. Typical stress-strain relationship of a plain reinforcing steel at tension.

“ t

I E, = 200 kN/mm2

Figure 3.4. Idealised stress-strain relationship of a reinforcing steel

at tension and compression (EC6).
Masonry Materials and Construction Systems 45

In the case where high bond reinforcing steel with diameter less than 6 mm,
including prefabricated ladder-type and truss-type bed joint reinforcement, is
used, it should not be considered as having high ductility.
A typical stress-strain diagram, obtained by tension test of a plain reinforcing steel
reinforcing bar, is shown in Fig. 3.3. An idealised diagram, used in the design of
reinforced masonry members, is shown in Fig. 3.4.

3.2.5 Masonry

When verifjhg the load-bearing capacity of masonry walls and structures to

vertical and lateral loads, the values of mechanical properties of masonry,
considered as an assemblage of masonry units and mortar, and not mechanical
properties of constituent materials, such as masonry units and mortar, are used. In
EC 6, the following intrinsic mechanical properties of the masonry, which should
be obtained by standard test methods, are specified:
The compressive strength, f,
The shear strength,&,
The flexural strength, fx,
The stress-strain relationship, 0--E.
EC 6 makes a distinction between the masonry itself - considered as an
assemblage of masonry units and mortar, which has intrinsic mechanical
properties - and the structural masonry element, a wall, the mechanical properties
of which depend on the intrinsic mechanical properties of the masonry, the
geometry of the element, and the interaction of adjacent parts.
In addition to mechanical characteristics specified by EC 6, the following
mechanical properties of masonry and masonry elements are also needed in
numerical verification:
The tensile strength,ft, as an equivalent to shear strength, f v ,
The modulus of elasticity, E,
The shear modulus, G, and
The ductility factor (indicator), p.
The ductility factor is a typical mechanical property of a masonry wall. It
cannot be attributed to the masonry alone.
As a rule, mechanical characteristics of masonry are determined by testing
standard specimens of masonry wallets and walls according to a set of standards
EN 1052. Compressive strength is determined by testing either small wallets of at
46 Earthquake-Resistant Design of Masonry Buildings

least 1.5 units length and 3 units height, or walls of 1.0-1.8 m long and 2.4-2.7
m high (Fig. 3.5). Specimens are placed in a testing machine, and the vertical
load is increased at a uniform rate so that the failure occurs after 15-30 minutes
after the beginning of testing. In cases where the wall is slender (heightkhickness
ratio greater than 20), lateral displacements at the mid-height of the wall are
measured in order to take into account the slenderness. If 6 is the displacement
just before the attainment of maximum vertical load, and t is the thickness of the
wall, the test value can be increased by a factor of
t
t-6’
provided that the increase is not greater than 15 %.

DT
VDT
40 4

Figure 3.5. Masonry specimens for compression tests (EN 1052-1).

According to EN 1052- 1, three identical specimens are tested and the results
evaluated. The mean compressive strength f of masonry is adjusted if the
compressive strength of masonry units and mortar deviate fiom the design mean
values within f 25 % of the specified strength. The characteristic compressive
strength of masonry is determined as the smaller value of either Ji( = f /1.2
or fk = fmin.
Masonry Materials and Construction Systems 47

In case that no test data are available, the characteristic compressive strength
of unreinforced masonry made with general purpose mortar may be calculated on
the basis of the normalised compressive strength of masonry units and
compressive strength of mortar fm using the equation:
Jk = KJb0.65fm 0.25 (inMPa), (34
provided that fm is not greater than 20 MPa nor greater than 2 Jb, whichever is the
0.10
smaller. The value of constant K (in MPa ) depends on the classification of
masonry units into groups according to Table 3.1. EC 6 recommends that the
constant K may be taken as:
- 0.60 for Group 1 masonry units in a wall without longitudinal mortar joint,
- 0.55 for Group 2a masonry units in a wall without longitudinal mortar joint,
- 0.50 for Group 2b masonry units in a wall without longitudinal mortar joint
and for Group 1 masonry units in a wall with longitudinal mortar joint.
- 0.45 for Group 2a masonry units in a wall with longitudinal mortar joint,
- 0.40 for Group 2b masonry units in a wall with longitudinal mortar joint and
for Group 3 masonry units.
The correlation between typical, experimentally-obtained values of
characteristic compressive strength of masonry fi(, obtained by testing wall
specimens [9, lo], and expected values, calculated by Eq. (3.1), is given in Table
3.6.
Table 3.6. Correlation between experimentally-obtained and predicted values
of characteristic compressive strength of masonry.

Unit Group Mortar

@Pa)

I 15 I 1 -clay I 2.5 I 2.5 I 4.4 I

I 7.5 I 2 a - clay I 2 I 5.0 I 2.4~~
I
15 2a - clay 2.5 2.5 4.0
15 2a - clay 5 3.0 4.8
I 7.5 12a- other I 5 I 3.5 I 3.0 I
I 7.5 I 2 a - other I 5 I 4.0 I 3.0 I
I 7.5 I 2b - clay I 2.5 I 4.5 I 2.3 I
48 Earthquake-Resistant Design of Masonry Buildings

Good correlation between experimental and assessed values can be obtained

in some cases. In others, however, the correlation is far from good. It is therefore
recommended that tests to determine the material characteristics of masonry
should be carried out whenever a new shape of masonry unit, or a new
construction technology, is introduced.
Shear strength, which is the strength of masonry subjected to shear forces, is
defined as a combination of initial shear strength under zero compressive strength
and increment in strength due to compressive stress perpendicular to shear in the
member at the level under consideration. Initial shear strength at zero
compressive stress fvko is determined by testing so called triplet specimens
according to EN 1052-3 (Fig. 3.6). The way of supporting the specimens and
applying the load should ensure that only shear stresses develop in the mortar to
masonry unit contact planes.

Figure 3.6. Determination of initial shear strength (EN 1052-3).

A minimum of five triplets are tested. The minimum acceptable value of

shear strength at zero compressive stress is 0.03 MPa. The characteristic shear
strength of unreinforced masonry is then calculated by using the equation:

fvk =fvko + 0.4 Od,

where Od is the design compressive stress perpendicular to shear. However, this
value should be not less than 0.065 (which is not less than fvko), or less than a
limiting value specified in EC 6 and depending on masonry unit’s group and
mortar quality. In Table 3.7, typical values for masonry, constructed with units
and mortar quality recommended for the construction in seismic zones, are
indicated.
Masonry Materials and Construction Systems 49

A different approach is also used, which defines the shear strength of

unreinforced masonry as the maximum value of principal tensile stress,
developed in a masonry wall of a specified geometry, idealised as an elastic,
homogeneous and isotropic panel at the attained shear resistance of the wall [ 6 ] .
The value of principal tensile stress, developed in the wall under those
assumptions at shear failure, is called tensile strengthft. Eq. (7.16), which is used
to evaluate the value of tensile strength from experimental results, and the
assumptions of this theory are explained in details in Chapter 7.4.1.
There is yet no standard testing procedure proposed to determine the shear
strength fv or tensile strength of masonryft by direct testing. Either cyclic and
monotonic racking or simple diagonal compression tests can be used to determine
the values of bothf, andft. As has been found by correlating the results obtained
from different types of testing, the differences are not significant [73. Whether the
results are evaluated forf" orft, the effect of compressive stresses in the masonry
wall panel is taken into account. Typical experimentally-obtained values of
characteristic tensile strength of masonry fk are correlated with initial shear
strength valuesfvko at zero compressive stress, recommended by EC 6, in Table
3.8.

Table 3.7. Typical values of initial shear strength at zero compression&,

and limiting values of characteristic shear strengthfvk for general purpose mortar (EC6).

1 M10- M20 0.3 1.7

clay M2.5-M9 0.2 1.5
I I

1 M10- M20 0.2 1.7

other M2.5-M9 0.15 1.5
[ M10- M20 I 0.3 I 1.4 I
2a other M10- M20 0.2 1.4
2b clay M2.5-M9 0.15 1.2

If Eqs. (3.2) and (7.16), which define the shear and tensile strength of
masonry, respectively, are considered as statistical regression equations used to
present the same experimental results in two different ways, it can be seen that no
50 Earthquake-Resistant Design of Masonry Buildings

statistically significant difference is obtained if the results are presented in one

way or another (Fig. 3.7).

Table 3.8. Correlation between experimental values of tensile strengthf& and

initial shear strengthfvko of masonry.

Unit Group Mortar I Strength (MPa) I

WPa)
0.5 1 0.04 1 0.10 I
I 15 I 1 -clay 2.5 I 0.18 I 0.20 I
I 7.5 I 2 a - clay 2 I 0.30 I 0.10 I
2a - clay 2.5 I 0.12 I 0.20 I
2a - clay 5 I 0.18 I 0.20 I
r7.5 I2a - other 5 I 0.27 I 0.15 I
I 7.5 12a- other 5 1 0.27 1 0.15 I
I 7.5 1 2b -clay 3 I 0.10 I 0.20 I

-
fU
ft
2.5

2.0

1.5

1.0

0.5

0.0
0 1 2 3 4 5 6 7 8 9 10 11 0,
ft

Figure 3.7. Correlation between experiments and shear resistance theory (after [S]).
Masonry Materials and Construction Systems 51

As has been found by correlating a large number of test results, the ratio
between the tensile and compressive strength of any type of masonry varies
between (compare Tables 3.6 and 3.8)

which makes it possible to assess the values of characteristic tensile strength if

only the value of characteristic compressive strength for a particular type of
masonry is available.
If the resistance of a masonry wall is verified for the out-of-plane loads,
flexural strength, i.e. the strength of masonry in pure bending, is the governing
parameter. However, as specified in EC 6, the value offxi should be taken as
zero, i.e. no contribution of unreinforced masonry should be taken into account in
the case of design for seismic resistance. Depending on the direction of failure
plane, flexural strength parallel to bed joints,fxl, or flexural strength perpendi-
cular to bed joints, fd,are distinguished (Fig. 3.8). The values of flexural
strengths fxl and fd should be determined by testing specially-made wallets
according to EN 1052-2.

Figure 3.8. Definition of failure planes to determine flexural strengthsfxi andfd (EC6).

On the basis of compression tests, the modulus of elasticity of masonry can

also be evaluated. Modulus of elasticity E is defined as a secant modulus at
service load conditions, i.e. at 113 of the maximum vertical load. It is calculated
on the basis of the mean strain p measured at that level of vertical load (Fig. 3.9).
On the basis of experimental values, EC 6 proposes idealised stress-strain
relationship for the design of masonry in bending and compression (Fig. 3.10). In
52 Eartha uake-Res istant Des inn of Masonry Buildings

the absence of a value of E determined tests, the value to be taken into

consideration in the structural analysis can be assessed to be equal to
E = lOOOfi(. (3 -4)

(r (MPa)
7

0
0 2 4 6 8

Figure 3.9. Typical experimental stress-strain relationships of masonry at compression [ 5 ] .

d
Idealised
A -------
/=-r ow.--
7
0’ I I
/ I I
1 I I
/ I I
i I
Design
I
fk
f d =-
Ym

I I
-
0.002 0.0035 E

Figure 3.10, Stress-strain relationship for the design of masonry in bending and compression (EC 6).
Masonry Materials and Construction Systems 53

As was the case of compressive strength, the values of modulus E, assessed

by Eq. (3.4), are sometimes far from reality. As can be seen by comparing the
values in Tables 3.6 and 3.9, the actual values vary between
2 0 0 h 5 E I 2000h. (3.4a)

As specified in EC 6 , it may be assumed that the shear modulus, G, is 40 %

of the elastic modulus, E. However, as experimental results indicate, the actual
values of shear modulus are much lower than assumed by EC 6 . Typical
correlation between experimental and predicted values is given in Table 3.9.

Table 3.9. Correlation between experimental and predicted values of moduli E and G.

15 -
2a clay 5 5000 3000 300 1200
7.5 2a - other 5 5000 3500 500 1400
7.5 2 a - other 5 6000 4000 600 2400
7.5 2b - clay 3 1800 4500 270 1800

It can be seen that the values of shear modulus G vary fkom 6 % to 25 % of

the elastic modulus E. In no case have values as high as 40 % of E, as specified
in EC 6 , been obtained. If shear modulus G is expressed in terms of tensile
Strengthftk, the following correlation is obtained:

with most results falling close to G = 2000ftk.

Taking into consideration the wide range of variation of the possible values
of strength and deformability characteristics of masonry, which do not depend on
the relevant characteristics of constituent materials in a uniform way, the testing
of masonry becomes one of the basic aspects of seismic resistance verification of
54 Earthquake-ResistantDesign of Masonry Buildings

masonry structures. Using the data obtained by testing each specific type of
masonry, and not relying upon the code recommended relationships, will make
the results of the seismic resistance verification accurate.

3.3 Construction Systems

Masonry buildings are box-type structural systems composed of vertical and

horizontal structural elements, walls and floors, connected in every direction.
Horizontal connecting elements, steel ties or, more often, reinforced-concrete
bond-beams (tie-beams) are provided at floor levels to connect the walls. During
earthquakes, floors should act as rigid horizontal diaphragm, which distribute the
seismic inertia forces among structural walls in proportion to their stiffhesses.
Any type of floors may be used, provided that general requirements of continuity
and effective diaphragm action are satisfied.
According to EC 6, the main types of structural walls are distinguished as
follows:
Single-leaf wall, which is a wall without a cavity or continuous vertical joint
in its plane.
Double-leaf wall, which is a wall consisting of two parallel leaves with a
longitudinal joint between them, not exceeding 25 nun, and solidly filled with
mortar. The leaves are tied together with wall ties to achieve common action
under vertical and lateral loading.
Cavity wall, which is a wall consisting of two parallel single-leaf walls, tied
together with wall ties or bed joint reinforcement, with either one or both
leaves supporting vertical loads. The space between the leaves is left as a
continuous cavity or filled, or partially-filled, with non-load bearing thermal
insulating material.
Grouted cavity wall, which is a cavity wall with two parallel leaves, spaced at
least 50 mm apart and tied securely together with wall ties and bed joint
reinforcement, and with a cavity filled with concrete to achieve common
action under vertical and lateral loading.
Typical examples of wall types are shown in Fig. 3.1 1.
Although no restriction regarding the use of any of the structural walls’ type
in seismic zones is specified in EC 8, it is obvious that single-leaf walls should be
preferred to double-leaf walls, and reinforced grouted cavity walls to cavity
walls, since they ensure monolithic behaviour of the wall under seismic
conditions.
Masonry Materials and Construction Systems 55

Figure 3.1 1. Cross-section of a (a) singleleaf, (b) double-leaf and (c) cavity wall (EC6).

Various types of masonry construction systems, both traditional and

engineered, are used in various countries. Because of the characteristics of
masonry, the behaviour of different construction systems subjected to seismic
loads is different: whereas unreinforced, plain masonry represents basically a
non-ductile structural material, confined and especially reinforced masonry
represent structural systems of improved strength and ductility.

3.3.I Unreinforced masonry

Although it is not considered to be earthquake-resistant by EC 6 and EC 8,

traditional stone-masonry construction, constructed as two-layered walls with an
inner infill, can be made earthquake-resistant by providing connecting stones at
least one per m2 of the vertical area of the wall. Mortar type M2 should be used
for the construction, and care should be taken that all the voids between the
stones are filled with mortar, especially the inner infill part of the wall.
Dimensioned stones must be used at the comers and wall intersections to
provide adequate connection in these, critical, zones. Construction course must
be levelled at each 1.0 m height in the zones of high, and at each 2.0 m height in
the zones of moderate and low seismicity. R.c. bond-beams should be provided,
connecting the structural walls all around the building (Fig. 3.12).
Brick and block masonry walls should be constructed according to the rules
of good mason workmanship, such as:
- When necessary, masonry units should be soaked in water before the
construction in order to prevent burning of the mortar, especially cement.
56 Earthquake-Resistant Design of Masonry Buildings

stone

U
0.5 m

Figure 3.12. Construction of earthquake-resistanttraditional stonemasonry wall.

Figure 3.13. Bonding arrangements using Group 1 masonry units (EC 6).

- Masonry units should be overlapped on alternate courses so that the wall acts
as a single structural element. To ensure adequate bonding, masonry units
should overlap by a length equal to at least 0.4 times the height of the unit or
Masonry Materials and Construction Systems 57

4 0 mm, whichever is the greater (Figs. 3.13 and 3.14). At the comers and wall
intersections, the overlap should not be less than the thickness of the units.
Units should be cut in order to achieve the specified overlap.
- Vertical stacked bond in not allowed in seismic regions.
- If general purpose mortar is used, the thickness of mortar joints should be not
less than 8 mm and not more than 15 mm. Perpend joints should be fully filled
with mortar.

Figure 3.14. Bonding arrangements using Group 2a and 2b masonry units (EC 6).

Although the recommended minimum thickness of load-bearing walls by EC

6 is only 100 mm, EC 8 specifies that, in seismic zones, the thickness of
unreinforced masonry shear walls should not be less than 400 mm in the case of
natural stone, 300 mm in the case of manufactured stone, and 240 mm in the case
of confined and reinforced masonry. This value may also be recommended for
masonry units made of materials other than stone. In many countries, however,
190-mm thickness is considered as the minimum for masonry construction in
seismic zones.
To take into account the stability of structural walls, the ratio of the effective
height to thickness of the wall should not be more than 9 in the case of natural
stone, 12 in the case of manufactured stone, and 15 in all other cases. Also, the
58 Earthquake-Resistant Design of Masonry Buildings

length of a structural wall should be at least 1/2 of the greater clear height of the
openings adjacent to the wall in the case of natural and manufactured stone, and
1/3 in the case of confined masonry.
As a rule, the same type of masonry units and mortar should be used for
structural walls in the same storey. Bracing walls should be constructed
simultaneously with load-bearing walls. It is also recommended that the thickness
of individual walls be kept constant over the entire height of the building.

3.3.2 Confined masonry

Confined masonry is a construction system, where masonry structural walls

are confined on all four sides with reinforced concrete or reinforced masonry
vertical and horizontal confining elements, which are not intended to carry either
vertical or horizontal loads, and are consequently not designed to perform as a
moment-resisting frames (Fig. 3.15).

Figure 3.15. Masonry confined within (a) reinforced masonry

and (b) reinforced concrete bond-beams and column.

In the case of masonry-infilled frames, the r.c. frame structure, which is

designed to resist vertical and seismic loads without infill, is constructed first.
Masonry filler walls are very often constructed as non-structural elements after
the completion of the main r.c. structure. In the case of confined masonry,
however, masonry walls are intended to carry all vertical and seismic loading,
The structural walls, which support the floors, are constructed first. Then, the
Masonry Materials and Construction Systems 59

floors with horizontal bond-beams elements are put in place, and finally, r.c.
vertical confining elements are constructed, well connected with horizontal
confining elements.
As the experimental investigations and the experiences obtained after
earthquakes have shown, confining the masonry walls with bond-beams and tie-
columns results in:
- Improvement in the connection between structural walls.
- Improvement in the stability of slender structural walls.
- Improvement in strength and ductility of masonry panels.
- Reduction in the risk of disintegration of masonry panels damaged by the
earthquake.
In order to ensure structural integrity, vertical confining elements should be
located at all corners and recesses of the building, and at all joints and wall
intersections. In addition, they should be placed at both sides of any wall opening
2
which, according to EC 8, has an area of more than 1.5 m . This is a severe
requirement which, as experiences indicate, can be relaxed to 2.5 m . Vertical
2

confining elements should also be placed at all free ends of structural walls. As is
the case of horizontal bond-beams, the distance between the vertical confining
elements should not exceed 4.0 m (Fig. 3.16).

H
Figure 3.16. Typical distribution of vertical confining elements in the plan of a building.

One of the dimensions of the cross-section of r.c. vertical tie-columns is

usually equal to the thickness of the wall, but not less than 150 mm. The
dimension may be less than the thickness of the wall in order to accommodate an
outer thermal insulation layer. The minimum dimension in the direction of the
60 Earthquake-Res istant Design of Masonry Buildings

wall should be 150 mm. Concrete of at least grade C15 should be used.
According to EC 6, the minimum cross-section area of steel reinforcement of tie-
2
columns is 200 mm , corresponding to 4 plain steel (yield stress 240 m a ) , 8-mm
diameter reinforcement bars. Although the minimum cross-section of the tie-
2
columns’ reinforcement is increased to 240 mm in seismic zones (EC 8), it is
recommended that, at comers and wall intersections, tie-columns are reinforced
with at least 4 plain steel, 10-mm diameter reinforcement bars, with a total cross-
2
sectional area of 314 mm . Stirrups of 6-mm diameter plain bars at 200 mm
intervals are used to confine the longitudinal reinforcement.
According to EC 6, the resistance of the r.c. or reinforced masonry confining
elements should not be taken into account in the design of confined masonry
structures in seismic situations. This is a mere consequence of fact that very little
experimental information regarding the mechanism of action and distribution of
lateral seismic loads among confining elements during earthquakes is available.
Therefore, the amount of reinforcement in confining elements is determined on
empirical basis [lo]. In Table 3.10, a proposal is given to choose the adequate
dimensions and number of plain steel (yield stress 240 MPa) reinforcing bars of
vertical confining elements on the basis of the number of storeys of the building
under consideration and the seismicity of the location.

Table 3.10. Typical reinforcement of vertical confining elements (after [lo]).

No. of Low: Moderate: High:

storeys a g < 0.1 g 0.1 g == ag c0.2 g 0.2 g c ag c0.4 g
2 1-2 448 4410 4412
4 1-2 448 4410 4412
4 2-4 448 4410 4412
6 1-2 4410 4412 4414
6 3-4 448 4410 4412
6 5-6 448 4410 4012

In order to provide the integrity of the walls’ connecting system, care should
be taken such that the reinforcement of the bond-beams and tie-columns is
adequately spliced and anchored at the corners and wall intersections. Sixty
diameters overlaps are required by EC 8. In addition, tie-columns should be
connected with masonry walls with reinforcing bars of not less than 6 ~lltn
Masonry Materials and Construction Systems 61

diameter and spaced no more than 60 cm apart, adequately anchored into the
mortar joints.
Fig. 3.17 shows the damage to a confined masonry wall pier at the end of a
“shaking-table” test of a model confined masonry building [ 111. Although the
confinement significantly improved the seismic performance of an unreinforced
masonry wall, it did not prevent the disintegration of the middle part of the
masonry panel at ultimate state. It can be seen that, in order to fully utilise the
resistance and energy dissipation capacity of masonry, a masonry wall panel
should be reinforced with horizontal bed joint reinforcement.

Figure 3.17. Disintegration of a confined masonry wall pier without bed joint reinforcement
at the end of “shaking-table” test of a confined masonry building model.

To reinforce the masonry, the use of specially shaped ladder-type or truss-

type reinforcement (Fig. 3.2), embedded in mortar with a vertical spacing of no
more than 600 mm, and adequately anchored into the tie-columns at the ends, is
recommended. In case where the masonry bed joint reinforcement is anchored
into the tie-columns, the connectors, required by EC 8 to connect the masonry
and confining elements, are not necessary.
62 Earthquake-Resistant Design of Masonry Buildings

3.3.3 Reinforced masonry

Reinforced masonry is a construction system, where steel reinforcement in

the form of reinforcing bars or mesh is embedded in the mortar or placed in the
holes and filled with concrete or grout. By reinforcing the masonry with steel
reinforcement, the resistance to seismic loads and energy dissipation capacity
may be improved significantly. To achieve this, the reinforcement should be
integrated with masonry so that all materials of the reinforced masonry system act
monolithically when resisting gravity and seismic loading.
There are various ways in which steel reinforcement can be used in a
reinforced masonry structural system. Basically, however, reinforced masonry
systems can be classified into:
Reinforced hollow unit masonry,
Reinforced grouted cavity masonry, and
Reinforced pocket type walls.

Concrete Reinforcement Mortar or

infill laid inmortar I / cocrete infiill

Figure 3.18. Reinforced hollow unit masonry (EC 6).

Reinforced hollow unit masonry represents the basic form of reinforced

masonry construction (Fig. 3.18). Specially shaped units with vertical holes
where vertical reinforcement is placed and filled with infill concrete or grout,
with or without grooves to accommodate horizontal, bed joint reinforcement, are
used for the construction of masonry walls. Before laying the masonry units,
vertical reinforcement is placed in position. Then, the first course of units is laid
in the mortar and horizontal bars or bed joint reinforcement are placed in the
grooves or in the mortar joints. The holes containing vertical bars are filled with
either concrete or grout, and the grooves containing the horizontal steel are filled
with either grout or mortar, as the construction of the wall progresses. In order to
Masonry Materials and Construction Systems 63

improve the resistance and depending on the shape of the units, all holes in the
hollow blocks are often grouted or filled with concrete infill.
Reinforced cavity masonry, however, is different by technology of
construction, and consequently, by structural characteristics and behaviour. As
can be seen in Fig. 3.19, it consists of two leaves (wythes) of masonry units,
separated by a cavity into which the vertical and horizontal reinforcement is
placed and grouted with either concrete infill or grout. The two leaves of a cavity
wall are tied together with wall ties or connectors, which should be designed to
carry lateral loads, induced by earthquakes. It is recommended that, in seismic
regions, at least one 6-mm diameter stirrup, or an equivalent tie, should be
2
provided at every 0.25 m of the wall area.

Reinforcement
Concrete laid in mortar
infill

Figure 3.19. Reinforced grouted cavity masonry construction systems (EC 6).

The wythes are usually 100 mm (the thickness of masonry units) thick and the
cavity 60-1 00 mm wide. The masonry units should be laid in running or stretcher
bond: vertical stacked bond is not allowed in seismic zones. The grout can be
poured either as the work progresses or after the masonry units in the whole
storey have been laid. In the first case, vertical reinforcing bars are first placed
into position. Then, the horizontal bars and wall ties (connectors) are placed and
grouted as the laying of courses of masonry progresses. In the second case,
however, the mesh of vertical and horizontal reinforcement is first placed into
position. Then, masonry units are laid on each side of the mesh, connected
together with wall ties. The ties should be laid in the bed joints along the same
vertical line in order to facilitate the vibrating of the grout pours. After the
masonry is built to a full storey height, the cavity is filled with grout. Before
64 Earthq uake-Resistant Design of Masonry Buildings

grouting, all mortar droppings should be removed fiom the foundations or other
bearing surfaces and reinforcement. Cleanout openings should be provided to
allow flushing away of mortar droppings and debris at the bottom of each pour.
Sometimes, vertical reinforcement is placed in the pockets formed in the wall
by special bonding arrangement (Fig. 3.20). As in the case of reinforced hollow
unit masonry, vertical reinforcing bars are placed into position before the laying
of masonry units. Depending on the units used, horizontal bed joint reinforcement
is placed in the mortar joints at vertical spacing not exceeding 600 mm. The
pockets containing vertical bars are filled with either concrete or grout, as the
construction of the wall progresses.

Mortar or
conlcrete infil

Figure 3.20. Reinforced pocket type walls (EC 6).

Reinforcing steel in masonry should be corrosion resistant or protected

adequately against corrosion due to environmental conditions. If ordinary carbon
steel is placed in the holes and embedded in concrete infill, it should be protected
by a concrete cover at least 20 mm thick in the case of a dry environment, 25 mm
thick in a humid environment, and 40 mm thick in the case of an aggressive
chemical environment. In the case of the mortar bed joint reinforcement,
however, the minimum depth of mortar cover from the reinforcing steel to the
face of the masonry should be 15 mm.
As specified in EC 8, horizontal reinforcement should be placed in the bed
joints or suitable grooves in the units with a vertical spacing not exceeding 600
mm, and the minimum percentage, referred to as the gross area of the section,
should be not less than 0.05 %. However, high percentages of horizontal
reinforcement leading to compressive failure of the units prior to yielding of the
steel should be avoided. Bars of not less than 4-mm diameter should be bent
around the vertical steel at the edges of the wall.
Masonry Materials and Construction Systems 65

Accordin to EC 8, vertical reinforcement with a cross-sectional area not less

4
than 400 mm should be placed at both free edges of every wall element, at every
wall intersection, or at at least 4 m intervals within the wall.
The reinforcing bars should be provided with a sufficient anchorage length so
that internal forces can be transmitted from reinforcement bars to the mortar and
concrete infill, and, subsequently, to masonry units. Anchorage may be achieved
by straight anchorage, hooks, bends or loops (Fig. 3.21). Straight anchorage or
bents are not allowed for plain reinforcing steel of more than 8 mm diameter.
Hooks, bends or loops are not allowed to anchor reinforcing steel in compression.

1
1
1
1 &-L 0.7 Ib
Ib
Straight anchorage Hook

Figure 3.21, Typical details of anchorage of reinforcing steel (EC 6).

EC 6 specifies that the straight anchorage length Zb depends on the diameter

of a reinforcement bar. It should be determined by calculation assuming that the
bond stress along the bar is constant:

Q = the effective diameter of reinforcing steel,

fyk = the characteristic strength of reinforcing steel,
fbok = the characteristic anchorage bond strength
ys, YM = the partial safety factors.
For hooks, bends and loops, the anchorage length for the bars in tension can
be reduced by 30 %. In the case where a larger area of reinforcement is provided
66 Earthquake-Resistant Design of Masonry Buildings

than required by design, the anchorage length can be reduced proportionally,

provided that for a reinforcing bar in tension the anchorage length is not less than
the greater value of 0.32b, 10 bar diameters, or 100 mm, and for a bar in
compression not less than the greater value of 0.62b, 10 bar diameters, or 100
mm.
The lap length is also determined on the basis of Eq. (3.6). In the calculation,
however, the smaller of the two bars lapped is considered. The lap length
provided between two reinforcing bars should be equal to
- 2b for bars in compression and for bars in tension where less than 30 % of the
bars in the section are lapped, and where the clear distance between the
lapped bars in a transverse direction is not less than 10 bar diameters and the
concrete or mortar cover is not less than 5 bar diameters.
- 1.4Zb for reinforcing bars in tension where either 30 % or more of the bars at
the section are lapped, or if the clear distance between the lapped bars in a
transverse direction is less than 10 bar diameters, or the concrete or mortar
cover is less than 5 bar diameters.
- 2Zb for reinforcing bars in tension where both 30 % or more of the bars at the
section are lapped, and the clear distance between the lapped bars in a
transverse direction is less than 10 bar diameters or the concrete or mortar
cover is less than 5 bar diameters.
However, code requirements regarding the amount and distribution of
reinforcement in reinforced hollow unit masonry need cautious consideration. As
has been shown by experiments, the effectiveness of horizontal and vertical
reinforcement strongly depends on the type and quality of masonry units and
.mortar, as well as on the way the reinforcing bars are anchored at the ends.
Obviously, the minimum and maximum amount of reinforcement in hollow unit
masonry is closely correlated with the type and quality of masonry.
In a simplified way, the behaviour of reinforced hollow unit masonry
subjected to seismic loading can be modelled by a truss-beam mechanism (Fig.
3.22). Under seismic situation, the bending moment and shear acting on the wall
are resisted by tension and compression forces that develop in the tension and
compression chord, as well as by tension and compression that develop in
verticals and diagonals of an equivalent truss, respectively. The tension chord is
represented by vertical reinforcing bars at the tensioned side, the compression
chord, however, by compressed masonry and vertical reinforcement at the
compressed side of the wall. As can be seen in Fig. 3.22, “verticals” are
Masonry Materials and Construction Systems 67

represented by horizontal bars in mortar bed joints, and compressed diagonals are
represented by masonry.

Figure 3.22. Simplified mechanism of reinforcement action.

Figure 3.2-3.Failure of reinforced hollow block masonry wall

due to crushing of shells of masonry units.
68 Earthquake-Res istant Design of Masonry Buildings

The mechanism works if there is enough vertical and horizontal steel

adequately embedded and anchored to resist the induced tension forces, and if
masonry units are strong enough to resist the induced compression and shear
(Fig. 3.23). Therefore, the recommendation of EC 8 that “high percentages of
horizontal reinforcement leading to compressive failure of the units prior to
yielding of the steel, should be avoided” should be modified to take into account
the balance between the load-bearing capacity of vertical and horizontal steel and
masonry. In this regard, experimental investigations are inevitable, particularly in
the case where new reinforced masonry construction systems are introduced.
When subjected to cyclic loads, such as seismic loads, the bond between
mortar and steel deteriorates. Consequently, the tension capacity of steel cannot
be fully utilised, although the bars are properly embedded and bent around the
vertical steel at the edges of the wall. Additional limitations should be given to
improve the behaviour.
Flexural type of failure is preferred to shear type because it represents ductile
type of seismic behaviour. However, crushing of face shells of hollow units under
cyclic lateral loads may lead to unexpected failure mechanism, although the
compressive strength of the units might be adequate. As a result, the resistance
capacity of such walls, calculated on the basis of the materials’ strength
characteristics, may be overestimated. Fig. 3.24 shows a typical example.

100

80

60

v
% I I/ Unreinforced
40

20

or- I I I I I

0 5 10 15 20 25 30
n (mm)

Figure 3.24. Vertical reinforcement at the edges did not improve the
the resistance of the walls because of weak hollow masonry units [9].
Masonry Materials and Construction Systems 69

Although the walls have been designed and a sufficient amount of horizontal
reinforcement to resist the shear has been provided, the resistance envelopes,
obtained by testing the walls under simulated seismic loads, clearly show that the
increased amount of vertical steel at the borders of the walls had no effect on
resistance because of the crushing of face shells of hollow masonry units.
Whereas reinforced grouted masonry behaves more or less as reinforced
concrete, the seismic behaviour of reinforced masonry with hollow or perforated
units and reinforcing bars in the vertical holes and bed joints depends on the type
and quality of masonry units. For optimum behaviour, balance should be always
found between the compression and shear capacity of masonry units, tension
capacity of vertical and horizontal steel, and available bond between mortar and
steel. In this regard, the minimum percentage of steel, either vertical or
horizontal, depends on the quality of the basic masonry wall. The maximum
amount of reinforcement should also be limited in correlation with the quality of
masonry units, mortar and grout, and not only with regard to the spacing between
the bars and depth of cover for protection against corrosion.

3.4 References

Eurocode 6: Design of masonry structures, Part 1-1: General rules for

buildings. Rules for reinforced and unreinforced masonry. ENV 1996-1-1;
1995 (CEN, Brussels, 1995).
Eurocode 8: Design provisions for earthquake resistance of structures,
Part I-3: General rules - Specific rules for various materials and elements.
ENV 1998-1-3 : 1995 (CEN, Brussels, 1995).
N. Bayiilke. The state of the masonry buildings in Turkey. EAEE WG6
Input Report (Ankara, 1989).
International Recommendations for Design and Erection of Unreinforced
and Reinforced Masonry Structures. CIB Recommendations, Publication 94
(CIB, Rotterdam, 1987).
M.TomaieviE and R. 2arniC. The behaviour of horizontally reinforced
masonry walls subjected to cyclic lateral in-plane load reversals. Proc. 8th
European Con$ on Earthquake Engrg., Vol. 4 (Laboratorio Nacional de
Engenharia Civil, Lisbon, 1984), pp. 7.6/1-8.
V. TurnSek and F. eaEoviE. Some experimental results on the strength of
brick masonry walls. Proc. 2nd Int. Brick-Masonry Con$ (British Ceramic
Society, Stoke-on-Trent, 197 1), pp. 149-1 56.
70 Earthquake-Res istant Design of Masonry Buildings

A. Bernardini, C. Modena, V. TurnSek and U. Vescovi. A comparison of

three laboratory test methods used to determine the shear resistance of
masonry walls. Proc. 7th World Conf on Earthquake Engrg., Vo1.7 (Int.
Association for Earthquake Engrg., Istanbul, Turkey, 1980), pp. 181-1 84.
V. Turngek and P. Sheppard. The shear and flexural resistance of masonry
walls. Proc. of Int. Research Conf on Earthquake Engineering (IZIIS,
Skopje, 1981), pp. 517-573.
M. TomaieviC, M. Lutman and L. PetkoviC. Seismic resistance tests of
hollow block masonry walls. Report ZAG/PI-96/04 (Slovenian National
Building and Civil Engineering Institute, Ljubljana, 1996).
Construction Under Seismic Condition in the Balkan Region. Vol. 3:
Design and Construction of Stone and Brick-masonry Buildings
(UNIDOLJNDP, Vienna, 1984).
M. TomaieviC and I. Klemenc. Seismic behaviour of confined masonry
buildings. Part two: Shaking-table tests of model buildings M1 and M2 -
analysis of test results. Report ZAG/PI-95/06 (Slovenian National Building
and Civil Engineering Institute, Ljubljana, 1996).
 CHAPTER 4

ARCHITECTURAL AND STRUCTURAL

CONCEPTS OF EARTHQUAKE-RESISTANT
BUILDING CONFIGURATION

4.1 Introduction

Many lessons have been learned from earthquakes regarding the influence of
structural layout on the seismic behaviour of masonry buildings. The
observations of earthquake damage and subsequent analysis of the causes of
damage indisputably show that besides the quality of structural materials,
building configuration is of relevant importance. The buildings with regular
structural layout, with the walls properly connected together at floor levels, have
often performed well, even when they are not designed to resist earthquakes.
Adequate seismic behaviour of those buildings proved that it is possible to
improve the seismic resistance by considering simple principles of architectural
and structural planning, and meeting the requirements for the quality of
materials and construction at the same time.
The following basic principles should always be considered when conceiving
a seismically-resistantmasonry structure:
Structural simplicity and regularity, i,e. uniformity and symmetry,
0 Redundancy (robustness),
Rigid floor diaphragm action, and
0 Adequate foundation.
In the case where the structure is regular and simple, gravity and seismic
loads are transmitted in a clear and undisturbed way from element to element.
71
72 Earthquake-Resistant Design of Masonry Buildings

Under seismic conditions, the induced seismic energy will dissipate uniformly
over the entire structure. If structural elements are not distributed unifomly in
the plan and elevation of the structural system, however, concentration of stresses
might occur in the zones of nonuniformity, resulting in heavy damage and
collapse of the structure.
Earthquake ground motion is a tri-dimensional phenomenon. It is, however,
not known which will be the main direction of ground motion during an expected
seismic event. Therefore, the resisting elements of each structure in a seismic
zone should be designed to resist the seismic excitation in both principal
directions of the building. The consequences of neglecting the necessity of
having an adequate number of sufficiently resistant structural walls in both
principal directions of the building are severe (Fig. 2.18).
Symmetric distribution of resisting elements in the plan of the building will
prevent possible torsional vibration, which often causes unexpected behaviour of
the structure when subjected to strong seismic ground motion. For the same
reason, the dimensions of setbacks and recesses should be limited.
In the case of masonry structures, the importance of good connection of walls
and rigid horizontal diaphragm action of floors has already been emphasised.
Last but not least, an adequate foundation system should be provided to transmit
the ultimate seismic loads, developed in the upper structure, into the foundation
soil. If the foundation soil fails before the ultimate resistance of the structure is
attained, the building is exposed to a severe risk of overturning.

4.2 Building Configuration

Masonry is a specific construction material, which, because of its mechanical

properties, requires specific structural configuration even when it is intended to
carry only vertical loads. The basic rules for the construction of masonry
structures are based on tradition and experience. As a result, traditional structural
systems are generally simple and regular, consisting of load-bearing walls and
cross walls which do not change their position and shape along the height of the
building, and are evenly distributed in both directions of the building.
The simplicity and regularity of a structure not only improve the expected
structural behaviour, but also make possible simplification of methods and ways
of seismic resistance verification. Experience and subsequent parametric analysis
of response of buildings to earthquakes have indicated that the following general
criteria for structural regularity in plan and elevation should be considered [ 1-5 3:
Architectural and Structural Concepts of Earthquake-Resistant Building Configuration 73

- The building structure is approximately symmetrical along each principal

axis in plan, regarding both lateral stiffness and mass distribution. A
sufficient number of structural walls, with approximately the same cross-
sectional area and stiffness, should be provided in each direction of the
building (Fig. 4.1).

U
+
n
Adequate Not adequate

Figure 4.1, Distribution of structural walls in plan.

- The plan configuration should be simple. Simple square or rectangular

buildings perform better when subjected to earthquake actions than those
with many projections. In this regard, the total dimension of projections, re-
entrant comers or recesses in one direction should not exceed 25 % of the
overall dimension of the building in the corresponding direction (Fig. 4.2).

- In order to avoid torsional effects resulting from differences in ground

motion in the case of long rectangular buildings, it is desirable to limit the
length of a single part of the building to four times its width. If longer
74 Earthquake-Resistant Design of Masonry Buildings

buildings are required, the building should be divided into separate parts with
adequate separation.
- Separation of large buildings with composite shape in plan (L, T, U or +
shaped plan) into several parts may be required in order to obtain symmetry
and rectangularity of each individual part (Fig. 4.3). To prevent hammering
effects between the adjacent parts, a sufficient separation between individual
parts should be provided. It is recommended that the width of separation
should not be less than 30 mm, and 10 mm should be added for each storey
(or 3.0 m) when the building height exceeds 9.0 m.

1-L 4B -t

Separation Separation

Figure 4.3. Irregular masonry buildings should be separated in regular sections.

- The building should also be regular in elevation. The distribution of resisting

elements, stiffnesses and masses along the height of the building should be as
uniform as possible. Concentration of masses at upper storeys should be
avoided. Sudden changes in stifhess due to changes in dimensions in plan,
distribution and type of structural elements in the adjacent stories along the
height of the building result in severe concentration of stresses, energy
dissipation demand and possible damage to those zones (Fig. 4.4).
- Mixed structural systems, such as a combination of masonry structural walls
in one storey and r.c. frame structural system in the adjacent storey, are not
allowed (Fig. 4.5). Sometimes, a combination of r.c. columns and masonry
shear walls within the storey is required because of architectural reasons. If
this is the case, structural masonry walls should be reinforced with vertical
Architectural and Structural Concepts of Earthquake-Resistant Building Configuration 75

and horizontal reinforcement to improve their ductility and energy dissipation

capacity. R.c. columns should be part of a frame system, connected with
masonry shear walls with rigid horizontal diaphragms, to ensure a uniform
distribution of seismic loads among the reinforced masonry shear walls and
r.c. columns. A detailed structural analysis should be carried out to verify the
transfer of seismic forces from one type of structural element to another. In
the design, special attention should be paid to structural details to ensure that
the structure is able to resist the assumed effects of seismic actions.

Adequate Not adequate Bad

Figure 4.4. Principles for building configuration in elevation.

Figure 4.5. Mixed structural system should be avoided.

Although the criteria for structural regularity and simplicity are fulfilled, the
potential seismic resistance of the structure is not fully utilised, unless the
requirements for rigid horizontal floor diaphragm action and good connection of
the walls are fulfilled at the same time. Namely, if the walls are not connected
together at floor levels, out-of-plane vibration will cause their separation along
vertical joints at the corners and wall intersections. Uncoupled vibration of
separated walls during earthquakes will significantly reduce the resistance of the
16 Earthquake-Resistant Design of Masonry Buildings

building to lateral loads and will, consequently, lead to the partial or total
collapse of a building.
In order to ensure rigid horizontal diaphragm action of floors, the following
requirements should be taken into consideration:
- Each floor should be situated in a single plane. Sharp dislevelements should
be avoided.
- The rigid behaviour of horizontal diaphragm should not be altered by the
presence of discontinuity, such as stairways. Large opening zones should be
strengthened with special reinforcement or bond-beams.
- Two-way slabs are preferred to one-way slabs, as they distribute vertical
gravity loads more uniformly onto the structural walls.

4.3 Dimensions, Building Height and Number of Stories

On the basis of experience and tradition, on the extent of earthquake damage

due to the type of materials used and structural systems employed, and
considering the current level of technical knowledge and construction
technology, limitations regarding the dimensions and height of masonry buildings
have been set in most existing seismic codes.
However, based on the results of recent experimental and theoretical
investigations, and on improvements in technology and methods of design,
limitations regarding the dimensions, building height and number of stories have
been relaxed. Except for unreinforced masonry located in seismic zones with
ag 2 0.3 g, which is not allowed for earthquake-resistant shear walls in buildings
higher than two storeys, no limitations regarding the size and height of masonry
buildings are specified in EC 6 and 8. In the case of confined and reinforced
masonry buildings, which fulfill the basic requirements for the quality of
materials and structural configuration, the size and height of the building depend
on the load-bearing capacity of masonry materials. However, the structure should
undergo subsequent verification of serviceability and ultimate limit states.
In order to reduce typical effects of temperature differences, shrinkage of r.c.
floors, and differential soil settlements common to long buildings, which may all
result into cracking of masonry structural walls even before the earthquake, and
to avoid the unfavourable effects, such as torsion due to differences in ground
motion along the length of the building in the case of earthquakes, it is
recommended that the length of masonry buildings of all masonry structural
Architectural and Structural Concepts of Earthquake-Resistant Building Configuration 77

systems or their separated parts should be limited to 40 m in the zones of high (ug
2 0.3 g), and to 50 m in the zones of moderate and low seismic intensity (ag< 0.3 g).
In the case of unfavourable soil conditions, further limitations regarding the
maximum dimensions of masonry buildings should be observed, depending on the
specific soil characteristics.
Although the seismic resistance of all masonry buildings should be verified
by calculation, it is recommended that, having in mind the available quality of
masonry materials and technology, the height and number of stories of masonry
buildings constructed in one of the construction systems may not exceed the
recommended values specified in Table 4.1. The limitations are specified for the
traditional brick or hollow unit masonry construction. Reinforced grouted cavity
masonry is exempted from these limitations.

Table 4.1. Recommended maximum building height Hand number of stories n.

4.4 Distribution of Structural Walls

In order to obtain a good performance when subjected to an earthquake,

structural walls in a masonry buildings should be uniformly distributed in two
orthogonal directions. Their number and strength should be sufficient to resist the
induced seismic loads. The walls should be firmly connected to the floors, which,
acting as rigid horizontal diaphragms, distribute horizontal seismic forces to the
walls in proportion to their stiffness.
From the viewpoint of the structural system, which resists the seismic loads,
the walls of a masonry building can be classified as:
- Structural walls, which resist the horizontal loads acting on the building, and
78 Earthquake-Resistant Design of Masonry Buildings

- Non-structural walls, having exclusively the function of partitioning the

building space.
With regard to the carrying of vertical loads, the structural walls can be
further subdivided into two categories, namely:
- Load-bearing walls, which carry their own weight and vertical loads from the
floors, and
- Bracing walls, which carry their own weight, but do not support the floors.
Considering the significance of structural walls, these walls should have a
minimum thickness of 400 mm in the case of stone and 300 mm in the case of
unreinforced masonry with artificial units, whereas 240 mm is the minimum
thickness in the case of confined and reinforced masonry (see Section 3.3.1).
However, there is no specification in EC 6 and 8 regarding the minimum distance
between them.
In order to obtain adequate performance for different masonry structural
systems, it is recommended that the distance between the structural walls should
be limited depending on the structural system and seismicity of the zone as
indicated in Table 4.2.

Table 4.2. Recommended maximum distance between structural walls [3].

Design ground < 0.2 g 0.2-0.3 g 2 0.3 g

acceleration ag
I Unreinforcedmasonry 1 10 m 1
~~~

8m r
~~

6m 1
Confined masonry 15 m 12 m 8m
Reinforced masonry 15 m 12 m 8m

Although the above recommended values are used when conceiving the
structural layout of the building, its structural stability should be verified by
calculation. Limiting factors may be the vertical load-bearing capacity and the
out-of-plane bending capacity of these walls.

4.5. Wall Openings

The size and position of wall openings, such as windows and doors, have a
strong effect on the in-plane-resistanceof a masonry shear wall. When subjected
to seismic loads, stress concentration takes place in the opening zones, which
Architectural and Structural Concepts of Earthquake-Resistant Building Configuration 79

may result in unexpected cracking of masonry and the subsequent deterioration of

resistance of masonry elements to in-plane lateral loads. In order to improve the
behaviour of masonry buildings when subjected to earthquakes, the following
recommendations should be observed regarding the location and size of the wall
openings:
Openings should be located in those walls which are subjected to smaller
intensity of vertical gravity loads.
Openings should be located outside the zones of direct influence of
cocentrated loads at beam supports.
On each storey, openings should be located in the same position along the
vertical line.
In order to provide an uniform distribution of resistance and stiffness in two
orthogonal directions, openings should be located symmetrically in the plan
of the building.
The top of the openings in the storey should be at the same horizontal level.
Openings should not interrupt r.c. bond-beams at the top of structural walls.
In addition to the above, the total length of openings in a shear wall should
not exceed half of the wall’s length. It is also recommended that, in the case of
brick and hollow unit masonry construction in the zones of high expected seismic
intensity, the total cross-sectional area of structural walls in each of two
orthogonal directions should not be less than 3 % of the gross floor area.

4.6 Simple Buildings

For masonry buildings which comply with the provisions regarding the
quality of masonry materials and construction rules (see Chapter 3), and with
additional structural limitations specified in EC 8, an explicit safety verification
is not mandatory. Such buildings are called “simple buildings”.
Simple buildings are regular buildings with an approximately rectangular
plan, where the ratio between the length of the long and short side is not more
than 4, and the projections or recesses from the rectangular shape are not greater
than 15 % of the length of the side parallel to the direction of the projection.
The number of stories above ground for different construction systems is
limited depending on seismic zones to the values given in Table 4.3.
The resisting walls (shear walls, structural walls) should be arranged almost
symmetrically in plan in two orthogonal direction. A minimum of two parallel
walls should be placed in each orthogonal direction, the length of each wall being
80 Earthquake -Res istant Design of Masonry Buildings

greater than 30 % of the length of the building in the same direction, and the
distance between these walls not being greater than 75 % of the length of the
building in the other direction. In the case of unreinforced masonry buildings,
walls in one direction should be connected with walls in the orthogonal direction
at a maximum spacing of 7.0 m.

Table 4.3. Number of stories above ground, allowed for simple buildings @C 8).

Design ground < 0.2 g 0.2-0.3 g 2 0.3 g

acceleration ag
I

Unreinforced masonry 3 2 1
Confined masonry 4 3 2
Reinforced masonry 5 4 3

At every floor, the cross-sectional area of structural walls in two orthogonal

directions, given as a percentage of the total floor area above the level
considered, should be not less than the values given in Table 4.4.

Table 4.4. Minimum horizontal shear wall cross-section,

given as % of the total floor area above the level considered.

Design ground < 0.2 g 0.2-0.3 g 2 0.3 g

acceleration ag
Unreinforced masonry 3 5 6
Confined masonry 2 4 5
Reinforced masonry 2 4 5

In addition to that, at least 75 % of the vertical load should be supported by

the structural walls, and the difference in the mass and in horizontal cross-section
of structural walls between the adjacent stories in two orthogonal directions
should not be greater than 20 %.

4.7 Non-structural Elements

Failures or fall-downs of non-structural elements, such as partition walls,

chimneys, masonry veneer, ornamentations, etc., might cause casualties and
structural damage during strong earthquakes. The falling-down of non-structural
Architectural and Structural Concepts of Earthquake-Resistant Building Configuration 81

elements might also obstruct passages and emergency exits, hence preventing
emergency interventions after the earthquakes. In this regard, when designing
masonry buildings to resist seismic loads, attention should also be paid to
adequate structural detailing of non-structural elements.
Partition walls are made of Group 3 masonry units (see Table 3.1) and are
usually about 100 mm thick, or less. Depending on their dimensions and the
seismic zone, partition walls may be either unreinforced or reinforced with bed
joint reinforcement to prevent their out-of-plane instability. If reinforced, 4-6
mm diameter bars are usually placed in the bed joints with a vertical spacing of
400-600 mm. Partition walls are fixed between the floor slabs by means of
cement mortar joints, whereas their connection with structural walls or tie-
columns along the vertical borders is achieved either by bond or by steel
anchors. The out-of-plane stability of partition walls should be verified by
calculation (see Sections 7.8 and 7.9).
It is recommended that masonry gable end walls and attics higher than 0.5 m
are anchored to the uppermost floor bond-beams. In order to connect those walls,
r.c. bond-beams should also be provided on top of those walls. In the case where
the height of those walls exceeds 4 m, intermediate bond-beams should be added
at intervals not exceeding 2 m. In addition to that, r.c. tie-columns, as specified in
Section 3.3.2, should be provided at distances not exceeding 4 m, and should be
well connected together with r.c. bond-beams (Fig. 4.6).

Figure 4.6. Tying of gable end walls and attics with r.c.tie-beams and columns.
82 Earthquake-Resistant Design of Masonry Buildings

Masonry veneer represents an architectural feature used to improve the

outlook of the faces of a building. Masonry veneer is supported by the main
structural system and can either be adhered or anchored to the backing structure.
If veneer is not a part of a structural wall, and is made as a free standing wall of
special veneer units, it should be adequately bonded to the backing structure,
even though it does not contribute to its strength.
In the case where masonry veneer wall is attached to a masonry wall, many
problems related to differential movements between veneer and support due to
shrinkage, short and long term deflections, temperature differences, and the like,
occurring in the case of masonry veneer attached to r.c. or steel structures,
disappear. However, in order to prevent its falling-out during earthquakes, even
adhered masonry veneer should be adequately anchored to structural walls with
steel anchors or connectors. Although no specific requirements regarding
masonry veneer are provided in EC 6 and EC 8, it is obvious that similar rules as
in the case of non-structural walls should be considered to verify the stability of
masonry veneer (see Sections 7.8 and 7.9).
Free standing chimneys and ventilation stacks should be constructed using
cement mortars. Adequate anchoring into the top floor and reinforcement above
the top floor level should be provided. Ornamentations, such as cornices, vertical
or horizontal cantilever projections, etc., should be reinforced with steel
reinforcement and adequately anchored into the main structural system of the
building. The adequacy of the anchoring should be verified by calculation as
specified in Sections 7.8 and 7.9.

4.8 References

Eurocode 6: Design of masonry structures, Part 1-1: General rules for

buildings. Rules for reinforced and unreinforced masonry. ENV 1996-1-1 :
1995 (CEN, Brussels, 1995).
Eurocode 8: Design provisions for earthquake resistance of structures,
Part 1-2: General rules and rules for buildings - General rules for
buildings. ENV 1998-1-2: 1995 (CEN, Brussels, 1995).
Eurocode 8: Design provisions for earthquake resistance of structures,
Part 1-3: General rules - Specijic rules for various materials and elements.
ENV 1998-1 -3 : 1995 (CEN, Brussels, 1995).
Architectural and Structural Concepts of Earthquake-Resistant Building Configuration 83

[4] Construction Under Seismic Condition in the Balkan Region. Vol. 3 :

Design and Construction of Stone and Brick-masonry Buildings
(UNIDO/UNDP, Vienna, 1984).
[5] International Recommendations for Design and Erection of Unreinforced
and Reinforced Masonry Structures. CIB Recommendations, Publication 94
(CIB, Rotterdam, 1987).
 CHAPTER 5
FLOORS AND ROOFS

5.1 Introduction

Masonry buildings represent box-type structural system composed of vertical

structural elements - walls, and horizontal structural elements - floors and
roofs. Vertical gravity loads are transferred from the floors and roof, which act
as horizontal flexural elements, to the bearing walls, which support the floors
and act as vertical compression members. Finally, the loads are transferred fkom
the bearing walls to the foundation system and into the ground.
In the case of earthquakes, however, floors and roof act as horizontal
diaphragms which transfer the seismic forces, developed at floor levels, into the
walls. In addition to this, floors and roofs connect the structural walls together
and distribute the horizontal seismic forces, developed in a masonry building,
among the structural walls in proportion to their lateral stiffness. Bond-beams
are provided along each structural wall at floor levels to assist the floors in
connecting the structural walls.

5.2 Floors

According to EC 6 and EC 8 [l, 21, a floor or roof structure can be made of

reinforced or precast concrete or timber joists incorporating boarding, provided
the floor or roof structure is capable of developing horizontal diaphragm action.
The connection between the floors and walls should be provided by steel ties or
r.c. bond-beams. The design lateral loads should be transferred between the
walls and interconnecting elements either by means of anchors (straps) or by the
frictional resistance between the walls and the floors or roofs.
85
86 Earthquake-Resistant Design of Masonry Buildings

Different types of floors can be used in the earthquake resistant construction

of masonry buildings. Monolithic r.c. slabs, which are cast simultaneously with
r.c. bond-beams (Fig. 5. l), represent the most simple solution. Sufficient bearing
length, being not less than 65 mm in normal cases, should provide the required
bearing capacity and transfer of shear forces.

- -- Rc. slab

-
Figure 5.1. TypicaI example of monolithic cast-in-place r.c. slabs with bond-beams (after [3]).

prefabricated element

st-in-place concrete

earn
U

Figure 5.2, Typical example of prefabricated slabs with r.c topping and bond-beams (after [3]).
Floors and Roofs 87

In cases where the floors are made of prefabricated, reinforced-concrete or

reinforced-masonry elements, precast elements should be well anchored into the
bond-beams along the wall. R.c. topping having a minimum thickness of 40 mm,
made of at least grade C 20 concrete and reinforced with at least 6-mm diameter
bars at 200 mm intervals in both orthogonal directions, placed at the mid-depth of
the topping, should be cast simultaneously with r.c. bond-beams (Fig. 5.2).
In cases where the floors are made of large prefabricated elements without
r.c. topping, steel connectors should be provided along the connections between
two elements to transfer the shear and tension forces developed in the horizontal
diaphragm during an earthquake fiom one element to another. Steel connectors
should be strong enough to ensure monolithic rigid diaphragm action of the floor
during the strongest expected earthquake. Adequate anchors should also be
provided at supports to ensure good connection of structural walls and rigid
horizontal diaphragm action of such type of floors (Fig. 5.3).

icntcd elcmcnl

elem

m-
I

Figure 5.3. Typical example of a slab made of large prefabricated elements (after [3]).

Wooden floors represent flexible horizontal diaphragms. As some recent

experimental and analytical studies indicate, the flexibility of horizontal wooden
floors, adequately anchored to structural walls, may improve the seismic
behaviour of masonry buildings in the specific case of long span structural walls
141
88 Earthquake-Resistant Design of Masonry Buildings

In the case of wooden floors, rigid horizontal diaphragm action is provided

by nailing plywood to timber joists at the bottom and top, or by nailing boards or
planks to timber joists in both diagonal directions: in one direction at the bottom
and in the other one at the top (Fig. 5.4). Sometimes, wooden floors are stiffened
by nailing boards or planks in both orthogonal directions only at the top of the
joists.
In any case, however, r.c. bond-beams or steel ties should be provided along
the walls, and timber joists should be anchored into the bond-beams or walls with
steel anchors.

Figure 5.4. Stiffening of wooden floors.

5.3 Bond-beams

Horizontal r.c. bond-beams (tie-beams) should be constructed at the top of all

structural walls at every floor level, or with a distance not exceeding 4 m between
them. Bond-beams represent a horizontal framing system which
- Transfers the horizontal shear induced by the earthquakes fiom the floors to
the structural walls.
- Connects the structural walls.
- Improves the in-plane rigidity of horizontal floor diaphragms.
- In combination with vertical tie-columns, improves the strength and energy
dissipation capacity of masonry walls.
In order to achieve the assumed behaviour, a number of structural details and
recommendations has been developed on the basis of experience and
experimental research. Concrete of at least class C 15 should be used. According
to specifications given in EC 8, the cross-section of horizontal bond-beams
should not be less than 150 by 150 mm. Usually, the vertical dimension of a r.c.
section of a bond-beam is equal to the thickness of the floor structure, whereas its
Floors and Roofs 89

horizontal dimension may be less than the thickness of the wall in order to
accommodate an outer thermal insulation layer, but not less than 150 mm.
Following the experiments and observations, at least four plain steel (yield stress
2
240 m a ) , 10-mm diameter reinforcing bars, with a total of 314 mm cross-
sectional area, should be used to reinforce the bond-beam, although a cross-
2
section of 240 mm is specified as the minimum by EC 8. These requirements
need to be considered in correlation with the limitation of the height of
unreinforced masonry buildings in seismic zones (see Section 4.3).
In order to provide the integrity of the walls’ connecting system, care should
be taken that the reinforcement of the bond-beams is adequately spliced and
anchored at the corners and at wall intersections. Stirrups of 6-mm diameter plain
bars at 200 mm intervals are used to confine the longitudinal reinforcement. Sixty
bar diameters overlaps are required by EC 8 for the longitudinal reinforcement of
the bond-beams.
According to EC 6 and EC 8, the resistance of the r.c. bond-beams should not
be taken into account in the design, and the bond-beams do not need to be
designed for seismic loads. As was the case of vertical confining elements, this is
a mere consequence of fact that limited experimental information regarding the
mechanism of action and distribution of lateral seismic loads onto bond-beams
and structural walls during earthquakes is available. The amount of reinforcement
in the bond-beams is determined on empirical basis [3]. In Table 5.1, a proposal
is given to choose the adequate dimensions and number of plain steel (yield stress
240 MPa) reinforcing bars of horizontal confining elements on the basis of the
number of storeys of the building under consideration and the seismicity of the
location.

Table 5.1. Typical reinforcement of horizontal r.c. bond-beams (after [3]).

No. of Position Low: Moderate : High:

storeys (storey) ag c 0.2 g 0.2 g S ag c 0.3 g 0.3 g 5 ag
2 1-2 4 bars, 4 8 mm 4 bars, 4 10 mm 4 bars, 4 12 mm
-- ~ ~~ ~ ~

4 1-2 4 bars, 4 1 0 mm 4 bars, 4 12 mm 4 bars, 4 14 mm

4 2-4 4 bars, 4 8 mm 4 bars, 4 10 mm 4 bars, 4 12 mm
-
6 1-2 4 bars, 4 12 mm 4 bars, 4 14 mm 4 bars, 4 16 mm
6 3-4 4 bars, 4 10 mm 4 bars, 4 12 mm 4 bars, 4 14 mm
6 5-6 I
4 bars, 4 8 m m 4 bars, 4 1Omm I I 4 bars, 4 12mm
90 Earthquake-Resistant Design of Masonry Buildings

However, as indicated by experiments (see Section 10.5.1.), two mechanisms

govern the behaviour of bond-beams in the case of flexible floors: bending due to
out-of-plane vibration of the walls and tension due to seismic shear which
developed in the walls in the direction of seismic action.
However, in the case of monolithic floors and rigid horizontal floor
diaphragm action, the bond-beams represent a constituent part of the diaphragm.
In this particular case, their reinforcement is carrying the tension developed in
the tensioned chord of the diaphragm, subjected to seismic loads.

5.4 Lintels, Balconies and Overhangs

For vertical loads lintels function as beams, which support the weight of the
wall and floor above the opening. Lintels can be made of either cast-in-place
concrete, or prefabricated reinforced concrete, or reinforced masonry elements.
Depending on the distance from the top of the opening to the top of the adjoining
floor, cast-in-place lintels can be cast either independently or monolithically with
the bond-beam and floor slab. The latter solution provides improved behaviour in
the case of an earthquake (Fig. 5.5).

Bond-beam Monolithic bond-beam and lintel

Lintel

Figure 5.5. Typical configuration of lintels in seismic zones (after [3]).

In seismic zones, the lintels and parapets should be regularly bonded to the
masonry of the adjoining walls and connected to them with horizontal
reinforcement. In the case where r.c. tie-columns are used to confine the walls
along the openings, the reinforcement of the lintels should be anchored into r.c.
tie-columns. In order to prevent local collapse due to crushing of supports in
earthquakes, sufficient bearing length should be provided at the end of the lintels.
Floors and Roofi 91

It is recommended that a minimum of 250 mm bearing length should be provided

at both ends.
As a rule, the lintel width should be equal to the thickness of the wall. In the
case of external walls, however, the lintel width may be reduced to accommodate
the outer thermal insulation layer. As is the case of bond-beams, the lintel width
should not be less than 150 mm.

Figure 5.6. Typical disposition of cantilever slabs and overhangs (after [3]).

Balconies and overhangs are typical cantilever structural elements, where

non-desirable vertical vibration can be induced during earthquakes. In order to
reduce the vertical oscillation, it is recommended that the span of balconies,
overhangs and other cantilever elements should be limited to (Fig. 5.6):
- 1.20 m for cantilever slabs cast continuously with the floor slabs, and
- 0.50 m for cantilever slabs anchored into the bond-beams without the
continuity with the floor slab.
The stability of cantilever balconies and overhangs, exceeding the
recommended span, should be verified by taking into account the vertical
component of seismic loads. According to EC 8 [ 6 ] ,the analysis for determining
the effects of the vertical component of the seismic action can be carried out on
the basis of a partial model of the structure, which includes the elements under
consideration and takes into account the stiffhess of the adjacent elements.
According to EC 8 [7], the effects of vertical components of seismic motion
are modelled by the response spectrum as defined in Section 2.2.5, but with the
ordinates reduced by a factor of 0.7 for vibration periods T shorter than 0.15 s,
by a factor of 0.50 for vibration periods longer than 0.50 s, and linearly
interpolated value of factor for a vibration period between 0.15 s and 0.50 s.
92 Earthquake-Res istant Design of Masonry Buildings

5.5 Roofs

In order to transfer inertia forces developed at the roof level into the
supporting walls, the roof system should be adequately braced in both orthogonal
directions, and should be adequately anchored into the r.c. bond-beam,
constructed at the top of the load-bearing and structural walls (Fig. 5.7).
Structural systems which exert lateral forces on the attic masonry walls and
gable walls should be avoided. If such situations cannot be avoided, the attic
walls should be anchored to the uppermost floor by means of adequately spaced
r.c. tie-columns.

Figure 5.7. Typical timber roof structure.

In order to reduce seismic loads, light roof structural systems and roof cover
(tiles) are preferred to massive structures. Where prefabricated elements are used,
r.c. cast-in-place topping with a minimum thickness of 40 mm should be
provided. In such a case the ends of the prefabricated elements should be
embedded into the r.c. bond-beam along the complete perimeter of the roof.

5.6 References

[l] Eurocode 6: Design of masonry structures, Part 1-1: General rules for
buildings. Rules for reinforced and unreinforced masonry. ENV 1996-1- 1 :
1995 (CEN, Brussels, 1995).
[2] Eurocode 8: Design provisions for earthquake resistance of structures,
Part 1-3: General rules - Specific rules for various materials and elements.
ENV 1998-1-3:1995 (CEN,Brussels, 1995).
Floors and Roof’s 93

[3] Construction Under Seismic Condition in the Balkan Region. Vol. 3:

Design and Construction of Stone and Brick-masonry Buildings
(UNIDO/UNDP, Vienna, 1984).
[4] J.C. Kariotis, A.M. El-Mustapha and R.D. Ewing. Dynamic reponse of
building systems with reinforced masonryy shear walls. Proc., 8th Int.
BricMBlock Masonry Con$, Vol. 2 (Elsevier, London, 1988), pp.740-75 1 .
[S] International Recommendations for Design and Erection of Unreinforced
and Reinforced Masonry Structures. CIB Recommendations, Publication 94
(CIB, Rotterdam, 1987).
[6] Eurocode 8: Design provisions for earthquake resistance of structures,
Part 1-2: General rules and rules for buildings - General rules for
buildings. ENV 1998-1-2: 1995 (CEN, Brussels, 1995).
[7] Eurocode 8: Design provisions for earthquake resistance of structures,
Part 1-1: General rules and rules for buildings - Seismic actions and
general requirements for structures. ENV 1998-1- 1 : 1995 (CEN, Brussels,
1995).
 CHAPTER 6

BASIC CONCEPTS OF LIMIT STATES

VERIFICATION OF SEISMIC RESISTANCE
OF MASONRY BUILDINGS

6.1 Fundamentals

After many centuries of traditional use and decades of allowable stresses

methods, limit states verification of seismic resistance of masonry structures has
been recently introduced to seismic codes. The philosophy of Eurocode 6:
Design of masonry structures [I] and Eurocode 8: Design provisions for
earthquake resistance of structures [2], which regulate the design and
construction of masonry structures, is based on the fundamental requirement
that a structure should be designed so that, with acceptable probability, it will
remain in use within the expected life period and under expected maintenance
conditions. This means that the structure should withstand all actions and
influences likely to occur in its life time without substantial damage, but will
also not be damaged disproportionally in cases where accidental events such as
explosions, impacts, earthquakes or human errors might occur.
In seismic regions, two basic requirements are considered in the design:
0 No collapse requirement, and
Damage limitation requirement.
The structure should be designed and constructed to withstand the design
seismic action without local or general collapse. It should also retain structural
integrity and load-bearing capacity after being subjected to an earthquake with
expected intensity (design earthquake). If the structure is subjected to seismic
95
96 Earthquake-Resistant Design of Masonry Buildings

actions having higher probability of occurrence than the design earthquake, but
of a lesser intensity, no damage to structural or non-structural elements should
occur that might limit the use of the building, or the costs of which would be
disproportionally high.
Therefore, two basic limit states, corresponding to the above criteria, need to
be verified in the design of a structure to resist seismic loads:
Ultimate limit state which is associated with collapse or other forms of
structural failure which may endanger the safety of people, and
Serviceability limit state which is associated with the occurrence of damage,
deformations or deflections, beyond which the specified service requirements
of the building are no longer met.
Because of specific characteristics of masonry structures and masonry
materials, there is usually no need to check the serviceability limit states.
Generally, masonry buildings are rigid structures in the case of which even
ultimate deformations and displacements are relatively small. In most cases, if a
masonry structure is verified for ultimate state, the requirements for serviceability
limit will be automatically fulfilled.

6.2 safety Verification and Partial Safety Factors for Materials

The safety of a structure against earthquakes is a probabilistic function which

depends on the expected seismic action and ability of the structural system to
resist the earthquake. According to EC 6 and EC 8, the following general
relationship shall be satisfied for all structural elements:

where Ed is the design value of the actions' effects, and Rd is the design
resistance capacity of a structural member under consideration. When
considering a limit state of transformation of the structure into a mechanism, it
should be verified that a mechanism does not occur unless the actions exceed
their design values.
According to EC 8, the design value Ed of the actions' effects, i.e. the design
value of bending moments, axial and shear forces in the seismic design situation
is determined by combining the characteristic values of the relevant actions
defined in EC 8 and Eurocode 1: Basis of design and action on structures [3]:
Basic Concepts of Limit States Verification of Seismic Resistance of Masonry Buildings 97

where
Gly = the characteristic value of permanent action j, i.e. self-weight of the
structure (dead load ), fixed equipment, etc.,
AEd = the design value of seismic action for the reference return period,
Pk = the characteristic value of prestressing action, if any,
&i = the characteristic value of variable action i. Snow, wind and fire are
not taken into account in the case of an earthquake,
y~= the importance factor,
~ 2 ,=i the combination coefficient for quasi permanent value of variable
action i (live load). In the case of residential and office buildings, ~2 = 0.3,
in the case of congregation areas and shopping, however, ~2 = 06.
Importance categories and importance factors for buildings y1 are given in
Table 6.1. The importance factor = 1.O is associated with the design earthquake
having a reference return period of 475 years.

Table 6.1. Importance categories and importance factors for buildings (EC 8).

Importance Buildings
I
I
Importance
factors, y~
~-

Buildings whose integrity during

I ~ earthquakes is of vital importance for I 1.4
~ civil protection, e.g. hospitals, fire
stations, power plants, etc.
Buildings whose seismic resistance is of
I1 importance in view of the consequences ~ 1.2
associated with a collapse, e.g. schools,
assembly halls, cultural institutions, etc.
Buildings of intermediate size and
I11 normal use, e.g. apartment houses, I 1.o
office buildings, etc.
~~ ~

Buildings of minor importance for public

safety, e.g. agricultural buildings, etc.
I 0.8

According to design philosophy of the Eurocodes, characteristic values of

material mechanical properties should be taken into account in the calculations of
design resistance capacity Rd of masonry structural members. However, to be in
compliance with the design actions’ effects in the seismic situation, the
characteristic values of material mechanical properties are further reduced by
partial safety factors for materials YM in the case where the ultimate limit state is
98 Earthquake-Resistant Design of Masonry Buildings

verified in accordance with Eq. (6.1). The values of partial safety factors of masonry
in a nonnal situation are given in Table 6.2.

Table 6.2. Partial safety factors for material properties "(M (EC 6).

YM Category of execution
A B C
Masonry Category of I 1.7 2.2 2.7
manufacturing
(see note) control of
masonry units I1 2.0 2.5 3.0
Anchorage and tensile and compressive 2.5 2.5 2.5
resistance of wall ties and straps
Anchorage bond of reinforcing steel 1.7 2.2 -
Steel (Ys) 1.15 1.15 -
Note: The value of y~ for concrete infill should be taken as that
appropriate to the category of manufacturing control of the
i
masonry units in the location where the infill is being used

In a seismic situation, however, the values of partial safety factors for

masonry strength ym are reduced. The recommended values are specified in Table
6.3. The partial safety factor for reinforcing steel should be taken as ys= 1.O.

Table 6.3. Partial safety factors for masonry strength'ym in seismic situations (EC8).

ICategory of execution I
I A I B I C I
Category of
manufacturing
control of
masonry units

It is clear that, once the safety level of the structure against earthquakes have
been decided upon, the determination of both the design seismic action and
design resistance capacity becomes an interdependent procedure. The philosophy
of determination of partial and global safety factors used in the calculation of R d
and structural behaviour factors used in the calculation of Ed should be
compatible, otherwise the verification according to Eq.(6.1) would lead to the
Basic Concepts of Limit States Verification of Seismic Resistance of Masonry Buildings 99

wrong conclusions. In order to estimate accurate values of parameters appearing

on both sides of Eq. (6.1), realistic data about the seismic behaviour of the
structure under consideration and the expected seismic ground motion should be
known. Since limit states of structural behaviour are considered, mathematical
models should be developed on the basis of the experimentally verified
theoretical hypotheses that take into account the characteristics of the actual
seismic behaviour of structures.

6.3 Design Seismic Action

To determine the design value of seismic action, the presence of all gravity
loads (in other words: masses, inducing seismic forces) should be taken into
account in the following combination:

where:

VEi = (p V2i,
and the values of (p are given in Table 6.4.

Table 6.4. Values of <p for calculating mi(EC8).

I Type of variable
action
I Occupation 1 Position I
Residential, office, Storeys Top storey: 1 .o
congregation independently Other storeys: 0.5
occu-;
I I LO-]
ylGu LSA
~ ~-

Residential, office, Some Iitoreys TOP storey:

,,I
having ccll ,c,l
IGlalGU
Correlated
congregation occupancies occupancies: 0.8
Other storeys: 0.5
Shopping, 1 .o
storage, garage
100 Earthquake-Resis rant Design of Masonry Buildings

6.3.I Seismic action and design response spectrum

Seismic action can be represented in various forms, such as ground

acceleration or velocity time-history (recorded or artificial), power spectrum, and
response spectrum. The form of seismic action to be used in seismic resistance
verification depends on the importance and complexity of the structure under
consideration. In some cases, even the tri-dimensional character of seismic
ground motion should be taken into account. However, taking into account the
regularity of masonry buildings, the response spectra representation, as described
in Section 2 . 2 5 , will give adequate results for most cases of masonry structures.
Ground acceleration or velocity time-history represent the direct form of
representation of seismic action, which is used to calculate the structural
response, and hence, action effects. Response spectra, however, already imply the
calculation of structural response. In the case where the design seismic loads are
determined on the basis of response spectra, only the calculation of action effects
is needed.
Earthquake damage observations and subsequent experimental and analytical
simulation of seismic behaviour show that most structural systems also resist
seismic loads in the non-linear range. Therefore, there is no need to design the
structure for elastic forces. However, during a strong earthquake the period of
vibration T of the structure is no longer a constant. As indicated by experiments,
once the structure is damaged and enters the non-linear range, the stiffness as
well as energy dissipation of the structure significantly change during the
vibration [4-7J. Typical changes, represented as a function of the duration of
seismic excitation, are shown in Fig. 6.1.

20 4

E
15
n
6
3 Input energy A 3rd storey
E z
m0 2 2nd storey

Y I st storey
1
ipated hysteretic energy
0 0 1 I I

0.5 1.5 2.5 3.5 0.0 2.0 4.0 6.0 8.0

t (9 t (s)

Figure 6.1. Changes in (a) lateral stiffness and (b) energy dissipation
during response of masonry building model to strong seismic ground motion.
Basic Concepts of Limit States Verificationof Seismic Resistance of Masonry Buildings 101

As a result of these changes, the structural response is also changed. Typical

results of shaking table tests of masonry building models are presented in Table
6.5, where the values of the dynamic amplification factor, defined as a ratio of
maximum measured model response acceleration a m , m a to maximum ground
acceleration ag,ma, are given at characteristic limit states. As can be seen,
dynamic amplification factor decreases with increased intensity of motion, thus
resulting in structural damage.

Table 6.5. Typical changes in dynamic amplification with increased intensity of shaking [5,7].

Model M1 Model M2 Model M3

arn,max/ag,max am,rna/ag,max am,max/ag,max
Elastic limit 4.72 1.53 2.53
I Maximum resistance I 0.92 I 1.07 I 1.99 I
I Ultimate state I 0.92 I 0.74 I 0.64 I
Note: M1 = plain-, M 2 = reinforced-, M3 = confined-masonry model.

Direct non-linear dynamic analysis, which takes into account the non-linear
characteristics of masonry structural elements, is the only way to obtain accurate
information regarding the actual behaviow of a masonry structure subjected to
seismic loads. However, to avoid the sophisticated direct non-linear dynamic
analysis, the non-linear behaviour and energy dissipation capacity of the structure
is taken into account by performing simple linear elastic analysis, but considering
a reduced response spectrum, called a design spectrum, obtained by introducing
the behaviour factor, i.e. force reduction factor q (see Section 6.3.5).
The design spectrum &J (T), the ordinates of which are normalised by the
acceleration of gravity g = 9.81 ms-2, is defined similarly as the elastic response
spectrum (see Section 2.5.5):

(6.5a)

(6.5b)

(6.5~)
102 Earthquake-Resistant Design of Masonry Buildings

if TD < T: (6.5d)

where:
S d ( T ) = the ordinate of the design response spectrum, normalised by g,
T = the vibration period of a linear SDOF system,
a = the ratio of the design ground acceleration ag to the acceleration of
gravity g = 9.81 msW2(a= ag/g),
q = the behaviour factor (see Section 6.2.3),
S = the soil parameter with reference value 1.O for subsoil class A,
q = the damping correction coefficient with reference value 1.0 for 5 %
viscous damping, and
Po = the maximum normalised spectral value assumed constant between TB
and Tc (Po= 2.5),
kdl = 213, kd2 = 513 = the exponents influencing the shape of the design
response spectrum.
If the period of vibration T of the structure is longer than Tc, then the design
spectrum value S d ( T ) should be not less than 0.20 a.

6.3.2 Design base shear

The following equation determines the design base shear force Fbd:

where:
s d (T) = the ordinate of the design response spectrum, determined by Eqs.
(6.51,
W = the weight of the building above ground level, determined in accordance
with Eq. (6.3).
Very often the value of the design base shear is represented in a
nondimensional form of the “design base shear coefficient”, which is a ratio of
the base shear to the weight of the building above the critical section. In this
particular case, the design base shear coefficient is represented simply by the
ordinate of the normalised design response spectrum:
Basic Concepts of Limit States Verification of Seismic Resistance of Masonry Buildings 103

Taking into account the recommended values of behaviour factors and

assuming the soil parameter S = 1.0 (normal soil conditions), the damping
correction coefficient q = 1.0 (at 5 % viscous damping) and the maximum
spectral value assumed constant Po = 2.5 (natural period of vibration between 0.1
s and 0.4 s), which is the normal case of a masonry building, the values of the
design base shear coefficient BSCd for different systems of masonry construction
to be constructed in different seismic zones can be calculated. They are given in
Table 6.6.

Table 6.6. Design values of base shear coefficient Bscd (EC 8).

I Seismic intensity (MSK) I VII I VIII I IX I

I Design ground acceleration I 0.1 g I 0.2 g I 0.4 g 1
Plain masonry: q = 1.5 0.17 0.33 0.67
Confined masonry: q = 2.0 0.125 0.25 0.50
I Reinforcedmasonry: q = 2.5 I 0.10 I 0.20 I 0.40 I
In EC 8, the values of behaviour factors, as well as parameters defining the
shape of the normalised elastic response spectra are given as indicative.
However, no indication is given regarding the values of design ground
accelerations. The design ground acceleration value ag, which is a measure of
earthquake’s intensity, corresponds to a reference earthquake with a return period
of 475 years. The probability of occurrence of such an earthquake during the 100
years life-time of an average building, with the assigned value of importance
factor y1 = 1.0, is 19 %. If the values proposed in seismic intensity scales are
used, the resulting values of BSCd for masonry structures are relatively high. In
fact, they are up to 100 % greater than they used to be for the verification of
ultimate limit state according to many existing seismic codes.

6.3.3 Distribution of design base shear

Design base shear force, i.e. the maximum value of the resultant seismic
forces that develop along the height of the building when subjected to
earthquakes of expected intensity, is calculated by Eq. (6.6). However, in order
to calculate the design seismic actions AEd, i.e. internal forces such as bending
moments, axial and shear forces in individual structural elements, the design base
shear should be distributed along the height of the building. Considering the
104 Earthquake-Resistant Design of Masonry Buildings

regularity of typical masonry structures, it can be assumed that horizontal forces

that develop in the building during an earthquake are distributed along the height
of the building in proportion to the shape of the first mode of vibration (Fig.
6.2a).

where:
Fid = the design horizontal force acting at i-th storey,
the displacements of masses mi, mj in the first mode shape,
Si, Sj=
Wi, Wj = the weight of masses mi, mj, calculated in accordance with Eq. (6.3).

S"

Figure 6.2: Distribution of base shear along the building's height.

The calculation can be further simplified if inverse triangular distribution of

lateral displacements along the height of the building is assumed (Fig. 6.2b):

where:
zi, z j = the heights of masses mi, mj above the level of application of seismic
loads.
Basic Concepts of Limit States Verijication of Seismic Resistance of Masonry Buildings 105

6.3.4 Behaviourfactor

Although masonry is considered to be a brittle structural material, the

experiments and analyses of earthquake damage show that even plain masonry
buildings possess a relatively high energy dissipation capacity, which makes
possible the reduction of elastic seismic forces. In a qualitative and simplified
way, the well-known definition of behaviour factor q (force reduction factor) is
explained in Fig. 6.3, where the seismic response envelope curve of an actual
structure, idealised as a linear elastic - perfectly plastic envelope, is compared
with the response of a perfectly elastic structure having the same initial elastic
stiffhess characteristics. As a result of the energy dissipation capacity of the
actual structure, which is expressed by the global ductility factor pu = du/de,there
is usually no need for the structure to be designed for strength, i.e. for the
expected elastic seismic load He.The structure should be designed for the
ultimate design load Hdu. The ratio between the two is called the behaviour factor
q = H$Hdu. The reserve strength ratio, called overstrength yo = &a/Hdu, where
Hma is the actual maximum resistance of the structure, results in an increased
value of behaviour factor q' = yo q.

Figure 6.3. Definition of structural behaviour factor q.

The following values of behaviour factors q are recommended for use in the
seismic resistance verification of masonry buildings:
0 For unreinforced masonry: q = 1.5,
106 Earthquake-Resistant Design of Masonry Buildings

For confined masonry: q = 2.0, and

For reinforced masonry: q = 2.5.
In order to attain the assumed energy dissipation capacity, specific
requirements for the basic structural configuration and details such as the quality
of materials, distribution of structural walls, bonding elements and reinforcement,
should be taken into account when designing the structure. Specific rules for
masonry buildings, given in EC 8, specify the minimum requirements needed to
attain the above specified values of behaviour factors in the case of masonry
buildings.

Figure 6.4. Shaking-table tests are used to

assess the values of structural behaviour factor q.

Not much experimental evidence is available to verify the proposed values of

behaviour factors. Taking advantage of the experimental data, obtained by testing
two three-storey, plain (M1) and reinforced-masonry building (M2) models of
identical structural configuration [4] and a model of a three-storey confined
masonry building (M3) on a shaking-table [ 5 ] , an attempt has been made to verify
the values of q-factors proposed by EC 8. Since the analysis of the dynamic
behaviour of the tested models indicated the predominant first mode of vibration
Basic Concepts of Limit States Verification of Seismic Resistance of Masonry Buildings 107

and storey mechanism which defines the failure mode, the global behaviour of the
models has been used to evaluate q-factors according to definition explained in
Fig. 6.3.
To compare the elastic and non-linear behaviour of the tested models, the
responses of hypothetical models with elastic stiffness characteristics to
maximum shaking-table accelerations, which the actual models had resisted, have
been calculated. As a result of the comparison, the values of H$Hmaxratio
H $ H m a = 3.74, H$Hmax= 2.69, and H$Hmax = 2.84 have been obtained for the
reinforced, confined, and plain-masonry building models, respectively.
Experimental values confirm the general validity of EC 8 proposed values of 4
factors, indicating a possible reserve at the same time.
If the requirements of EC 8 were taken into account in the design of
prototype buildings (which, in fact, have been designed by an elastic design
method), the design resistance capacity Hdu of the tested model buildings would
be lower than experimentally obtained. Consequently, a substantial reserve in
strength (overstrength) would have been observed, which would further increase
the observed q-factor values.
As indicated by these experiments, the proposed values adequately reflect the
assumed differences in the behaviour of different types of masonry construction.
However, it seems that masonry structures of all types of construction possess
more energy dissipation capacity than attributed by the proposed values of the
behaviour factor 4. This might be one of the main reasons why masonry
buildings, designed for low values of Bscd by former seismic codes, resist
earthquakes. However, further experimental and analytical research is needed to
confirm these observations.

6.4 References

11 I Eurocode 6: Design of masonry structures, Part 1-1: General rules for

buildings. Rules for reinforced and unreinforced masonry. ENV 1996-1-1 :
1995 (CEN, Brussels, 1995).
121 Eurocode 8: Design provisions for earthquake resistance of structures,
Part 1-1: General rules and rules for buildings - Seismic actions and
general requirements for structures. ENV 1998-1-1 : 1995 (CEN, Brussels,
1995).
131 Eurocode I: Basis of design and actions on structures, Part 1: Basis of
design. ENV 1991-1: 1994 (CEN, Brussels, 1994).
108 Earthquake-Resistant Design of Masonry Buildings

D. Benedetti and G.M. Benzoni. A numerical model for seismic analysis of

masonry buildings. Earthquake Engrg. Struct. Dynamics, 12 (12) (John
Wiley & Sons, Chichester, 1984), pp. 81 7-83 1.
M. Toma2eviC, P. Weiss, T. Velechovsky, C . Modena. Seismic behaviour
of masonry buildings. Shaking-table study of masonry building models with
different structural configuration - summary report. Report ZRMxPI-90/04
(Institute for Testing and Research in Materials and Structures, Ljubljana,
1990).
M. TomaieviC and P. Weiss. Seismic behavior of plain- and reinforced-
masonry buildings. J. Struct. Engrg., 120 (2) (ASCE, New York, 1994), pp.
323-338.
M. TomaieviC and I. Klemenc. Verification of seismic resistance of
confined masonry buildings. Earthquake Engrg. Struct. Dynamics, 26 (1 0 )
(John Wiley & Sons, Chichester, 1997), pp. 1073-1 088.
 CHAPTER 7

SEISMIC RESISTANCE VERIFICATION

OF STRUCTURAL WALLS

7.1 Introduction

If no earthquakes are expected in the region, the stability of masonry

structures needs to be verified only for gravity loads. In case of an earthquake,
however, the structure will be subjected to a series of cyclic horizontal actions,
which will often cause high additional bending and shear stresses in structural
walls, exceeding the elastic range of the behaviour of masonry materials.
Structural walls, which are the basic resisting element to seismic loads, will be
damaged, and, if they had not been properly designed and detailed to withstand
inelastic deformation and to dissipate energy, the induced inertia forces might
cause heavy damage or even collapse of the building.
Since the ground motion is tridirectional, both vertical and horizontal
inertia forces are induced, changing in time, and resulting in tridimensional
vibration of the structure. In addition, due to the distributed mass of masonry
walls, inertia forces perpendicular to the planes of the walls are also induced,
resulting in the out-of-plane vibration of structural and non-structural walls.
Because of typical structural configuration and reserve in strength of masonry
materials with regard to carrying vertical gravity loads, there is generally no
need to verify the load-carrying capacity of masonry walls and floors for vertical
seismic action. Also, because of uniform distribution of walls in both orthogonal
dixections, geometric requirements for shear walls (effective height, size and
position of openings) and connection between walls and floors, out-of-phne
resistance to seismic action is usually not critical Therefore, the verification of
109
110 Earth4 uake-Resistant Design of Masonry Buildings

seismic resistance for lateral out-of-plane loads is necessary only in the case of
extreme spans between structural walls, which exceed the code recommended
values.
According to the results of earthquake damage analysis and subsequent
experiments, three types of mechanism and failure modes define the seismic
behaviour of structural masonry walls when subjected to in-plane seismic loads.
The mechanisms depend on the geometry of the wall (height/width ratio) and
quality of materials, but also on boundary restraints and loads acting on the wall
(Fig. 7.1).

Figure 7.1, Typical failure modes of masonry walls, subjected to in-plane seismic load.

In the case of low vertical load and poor quality mortar, seismic loads
frequently cause shearing of the wall in two parts and sliding of the upper part of
the wall on one of the horizontal mortar joints. The mechanism is called sliding
shear failure. If the level of vertical load and axial compression stresses in the
wall are within the usual limits, the wall may fail either in shear or in bending.
Shear failure, which is a typical mode of failure of masonry walls subjected to
seismic loads, takes place where the principal tensile stresses, developed in the
wall under a combination of vertical and horizontal loads, exceed the tensile
strength of masonry materials. Characteristic diagonal cracks develop in the wall
just before the attainment of lateral resistance. The cracks can either follow the
mortar joints or pass through the masonry units, or both. In the case of improved
shear resistance and high momentlshear ratio, however, crushing of compressed
zones at the ends of the wall usually takes place, indicating the flexural mode of
failure.
It is not simple to model the non-elastic, non-homogeneous and unisotropic
character of masonry by calculation. The prediction of lateral load-bearing
capacity and deformability of masonry walls is usually based on the analogy with
Seismic Resistance Verification of Structural Walls 111

reinforced concrete structural elements. Since the behaviour under seismic loads
is quite similar, only minor adjustments are needed in the case of reinforced
grouted masonry. Good correlation with experimental results can be obtained for
hlly grouted masonry walls if general computer programs for predicting the
inelastic cyclic behaviour of r.c. structures are used [l]. However, in the case of
traditional plain or reinforced brickblock masonry construction, mathematical
models developed for r.c. elements should be modified in order to take into
account specific mechanical properties of masonry materials.
Lateral load-displacement relationships have been modelled on the basis of
different physical models. Combined arch and truss mechanisms [2] or a
combination of dowel, pullout and friction mechanisms [3] have been considered
in order to predict lateral load-displacement skeleton curve in the case of shear
failure of reinforced masonry walls, as well as cyclic hysteretic behaviour. A
global implicit dimensionless analytical hysteretic model has been developed on
the basis of experimental results from cyclic tests of reinforced masonry walls
[4]). Hysteretic behaviour of plain masonry walls failing in shear has also been
modelled by parameter functions, shear modulus and its viscous counterpart, also
determined by experiments [ 5 ] .
By finite element approach (FEM), where masonry blocks and mortar joints
are modelled as discrete or continuum elements with links between them, lateral
load-displacement relations, failure mode and crack patterns can be predicted
relatively well [6-81. Recently, good results have been obtained by applying a
dilatant interface constitutive model, in addition to the mortar joints modelled
with interface elements and masonry units modelled with smeared crack elements
191.
In the case of either FEM or physical mechanism models, the basic skeleton
curve of hysteresis loops can be accurately predicted if experimentally obtained
input data on the mechanical properties of constituent masonry materials are
properly used in the calculations. However, data on the hysteretic behaviour of
the walls, such as strength and stiffness degradation and deterioration and energy
dissipation capacity, can be obtained only by experimental simulation of seismic
behaviour of masonry walls as a whole.
Although these models yield good results for the specific cases, simulating
the conditions under which they have been developed, they are too complex to be
used for practical evaluation of seismic resistance of masonry walls. For practical
use, simple procedures and equations should be developed based on the familiar
laws of the theory of elasticity, but reflecting the specific characteristics of
112 Earthquake-Res istant Design of Masonry Buildings

seismic behaviour of masonry. In order to develop such procedures, experimental

simulation of seismic behaviour of masonry walls and buildings is inevitable.

7.2 Experimental Simulation of Seismic Behaviour of Masonry Walls

In order to determine the parameters of seismic resistance of masonry walls,

tests should be carried out by subjecting the specimens to similar loading
conditions as they are subjected to in a building. In order to simulate the
observed failure mechanism, the specimens should be of similar geometry and
should be restrained in a similar way as in the building's structural system.

Figure 7.2.Shear failure of typical brick-masonry window piers. Budva, Montenegro, 1979.

A considerable amount of experimental research in the behaviour of plain and

reinforced-masonry walls subjected to simulated seismic in-plane loading has
been carried out in the last few decades. Full-scale or reduced-size masonry wall
specimens made of different materials have been tested at different boundary
conditions. Different testing procedures, static and dynamic, cyclic and
monotonic have been used in order to simulate the effects of seismic loads. As a
result of these tests, mechanisms of behaviour of masonry walls subjected to
seismic loads have been defined and mathematical models for the prediction of
seismic resistance developed [lo-221. By the same tests, data on the values of
parameters of seismic resistance of different types of masonry have also been
obtained. Whereas reliable values of mechanical properties of masonry can be
Seismic Resistance Verjfication of Structural Walls 113

obtained by relatively simple tests, parameters of stiffness degradation and

deterioration and energy dissipation capacity can be determined only by
simulating the cyclic character of seismic loads.
Typically, the behaviour of a window pier about 1.0 m long and 1.5 m high,
as shown in Fig. 7.2, is studied in the laboratory. Since the typical pier can be
considered to be symmetrically fixed at the bottom and top to the structural
system, special laboratory testing facilities, which simulate such restraints, have
been developed to experimentally study the seismic behaviour of that kind of
structural walls (Fig. 7.3).

Figure 7.3. Typical facility designed for seismic resistance testing of masonry walls.

In the real structure, vertical compression in a wall changes during

earthquake because of stresses, that develop due to restraints that prevent the
rotation of the wall at large lateral displacements. However, since it would be
difficult to simulate the actual restraints, the walls are tested at a controlled level
of vertical load, as well as at controlled conditions of boundary supports.
Usually, the level of vertical load is kept constant during seismic resistance test,
within the allowable limits of vertical loading in a real building.
114 Eart hq uake-Res istan t Design of Masonry Buildings

Various cyclic lateral displacement patterns, applied statically or

dynamically, are used to simulate the seismic loads (Fig. 7.4). In the case of most
simple test, monotonically increased displacements are applied statically. In a
more sophisticated testing procedure, however, a real seismic displacement
response pattern can be applied dynamically. Hysteretic relationships between the
imposed displacements and resistance of the wall represent the basic parameters
in the analysis of test results (Fig. 7.5).

Monotonic Sinusoidal cyclic

1 (s)
5 41 I Earthquake response
S i 11iisoi da I cy cl ic

Figure 7.4. Typical imposed lateral displacement time histories used for simulation of seismic loads [23].

A recent study that investigated the experimental simulation of seismic

behaviour by applying different imposed displacement patterns [231, indicated
that significant differences in the results may be obtained if different
displacement patterns are used to test equal masonry wall specimens (Fig.7.6).
Since the differences in test results exceed the normally tolerated scattering of
quality of masonry materials, it is of relevant importance that testing procedures,
on the basis of which the mathematical models for seismic resistance verification
of masonry walls have been developed, be harmonised with the procedures used
for the determination of the input parameters. In order to develop mathematical
tools for reliable verification, however, the testing procedures should also reflect
the actual seismic behaviour of masonry walls.
Seismic Resistance Verification of Structural Walls 115

42

28

14
z
W

2
0

-14

-28

-42 '
-1 5
I 1

-10
I

-5
I I
0
I I

5
1 I

10
1
15
d (mm)

Figure 7.5. Typical lateral displacements-lateral resistance hysteresis loops,

obtained during seismic resistance test of a masonry wall.

75

I
50

25

0 I
I I I I I I I I I I

0 5 10 15 20 25
d (m)

Figure 7.6. Hysteresis envelopes, obtained by testing the same type

of wall with different displacement patterns [23]

7.3 Idealisation of Experimental Results

In order to simplifi the analysis and design, the values of sectional forces,
stresses and strains are usually determined based on the gross cross-sectional
geometrical characteristics of the walls, and assuming the elastic, homogeneous
116 Earthquake-Resistant Design of Masonry Buildings

and isotropic global properties of masonry as the structural material. These

assumptions permit that equations, based on the simple theory of elasticity, can
be used for seismic resistance verification, modified by taking into account the
basic characteristics of the observed seismic behaviour of masonry walls.
As was the case of more refined mathematical models, practical equations for
the calculation of lateral resistance and deformability of masonry walls have been
developed on the basis of extensive series of tests of plain, confined and
reinforced masonry walls subjected to simulated earthquake loads.
Although simplified, those equations reflect the actual failure mechanisms.
Therefore, the values of mechanical quantities, which determine the load-bearing
capacity and deformability of masonry walls in those equations, such as:
0 The compressive strength of masonry, f,
0 The shear&, or tensile strength of masonry,ft,
0 The modulus of elasticity (secant modulus), E,
0 The shear modulus, G, and.
0 The ductility factor, p,
sh uld also be determined experimentally by using t sting procedures lhich a
compatible with the experiments on the basis of which the equations for
calculations have been developed. Otherwise, errors can be made in predicting
the seismic resistance of the building.
In order to make the calculations simple, the actual hysteretic behaviour of a
masonry wall, subjected to a combination of constant vertical load and a
sequence of lateral load reversals (Fig. 7.5) is represented by an idealised bi- or
trilinear resistance envelope (Figs. 7.7 and 7.8, respectively). To idealise the
experimental envelope, three limit states in the observed behaviour of the tested
wall are first defined:
Crack limit, determined by displacement dcr and resistance Hcr at the
formation of the first significant cracks in the wall, which change the slope of
the envelope.
Maximum resistance, determined by maximum resistance Hma, attained
during test, and corresponding displacement d ~ ~ ~ .
Ultimate state, determined by maximum displacement attained during test
dma and corresponding resistance Hdm,.
Obviously, the initial slope of the idealised envelope is best defined with a
secant stiffness at the formation of cracks, which is called effective stifkess of
Seismic Resistance Verification of Structural Walls 117

the wall Ke. It is calculated as the ratio between the resistance and displacement
of the wall at crack limit:

t #.I
.I I
I
I
I
I
I \I
I

Figure 7.7. Idealisation of experimental resistance envelope with bilinear relationship (after [24-251).

If the resistance envelope is idealised with a bilinear relationship, the

ultimate resistance of the idealised envelope Hu is evaluated by taking into
account the equal energy dissipation capacity of an actual and idealised wall: the
areas below the actual and idealised curves should be equal (Fig. 7.7). When
idealising the experimental curve, and knowing the initial stifthess Ke, the
ultimate resistance Hucan be easily calculated from 124-251:

where:
Aenv = the area below the experimental resistance envelope.
It should be emphasised, at this point, that ultimate resistance Hu does not
represent the design, but the idealised maximum experimental value. As has been
found by evaluating the results of tests of more than 60 walls, the average value
ratio is 0.9 118,261. Consequently, in the case of bilinear idealisation
118 Earl hq uake -Res istant Design of Mas onry Buildings

of resistance envelope, the calculated values of maximum resistance should be

multiplied by 0.9:
Hu = 0.9 H m a . (7.3)
Arbitrarily, the ultimate idealised displacement du is defined as the
displacement value where the idealised line intersects the descending branch of
the experimental one. Consequently, ultimate ductility factor (indicator) pu is
defined as a ratio

(7.4)

and the displacement at the idealised elastic limit de is evaluated from

d e = -H"
.
(7.5)
Ke
In seismic resistance verification, however, the value of ultimate ductility
factor pu is limited to avoid excessive damage to structural walls. Usually,
allowable values for individual structural walls, which are taken into account in
the calculations of the idealised resistance envelopes of structural walls, are
greater than the values of behaviour factors recommended for each particular
type of masonry construction. However, although experimental results might
indicate that larger values could be acceptable, it is recommended that the values
of ultimate ductility factor pu of individual walls are limited to:
pu = 2.0-3.0 for the case of plain masonry walls,
pu = 3.0-4.0 for the case of confined masonry walls,
pu = 4.0-5.0 for the case of reinforced masonry walls.
Where the idealisation is trilinear (Fig.7.8), the slope of the initial branch of
the idealised curve is defined by the same value of effective stiffness Ke as in the
case of bilinear idealisation (Eq. 7.1). In the case of trilinear idealisation,
however, the calculated value of maximum resistance Hmm is used without any
reduction.
As regards the theoretical value of resistance at elastic (crack) limit, the
value is either calculated by using adequate mathematical models, or maximum
resistance is reduced by considering an experimentally obtained value of a
reduction factor Ccr, which takes account for the observed ratio between the
resistance of the wall at crack limit and at maximum resistance:
Seismic Resistance Verijication of Structural Walls 119

Figure 7.8. Idealisation of experimental resistance envelope with trilinear relationship [27].

As has been shown by most experiments, HCr/’Hmax ratio varies from 0.6-0.8.
It is therefore recommended that the resistance of the wall, built in any
construction system, at the formation of cracks be calculated by multiplying the
maximum resistance value Hmaxby Ccr = 0.7.
For the theoretical value of lateral resistance at ultimate displacement Hd,,
there is no simple mathematical model available to assess the strength
degradation of a masonry wall as a function of lateral displacements. Therefore,
experimentally evaluated strength degradation factors are used to evaluate the
resistance at ultimate displacement:

where:
Csd = the strength degradation factor.

As the experiments indicate, strength degradation In the non- inear range of

behaviour of masonry walls before collapse is relatively large (the values of
strength degradation factor vary from 0.4-0.8). However, because of severe
deterioration of the wall before collapse it is recommended that no more than 20
% strength degradation (Csd= 0.8) be tolerated in practical calculations.
120 Earthq uake-Resistant Design of Masonry Buildings

Lateral stiffness of structural walls is an important parameter: period of

vibration of the building under consideration depends on the stiffness of the
walls, and seismic shear is distributed among the walls according to their
individual stiffnesses.
By definition, the stifhess of a structural element is defined by the action
effect, shear and/or bending moment, which causes an unit displacement and/or
rotation of the element under consideration (Fig. 7.9). The element's stiffness
depends on the mechanical properties of constituent materials, the geometry and
boundary restraints.

H .dl

1 h

Figure 7.9. Deformation of a fixed-ended wall subjected to lateral loading.

When subjected to lateral load H , a fixed-ended masonry wall exhibits lateral

deformation d, which is partly due to bending and partly to shear:

Hh3 KHh
d= +
12EI, G A ,
9

where:
d = the displacement of the wall,
h = the height of the wall,
t z3
IW' - = the moment of inertia of the wall's cross-section,
12
Seismic Resistance Verification of Structural Walls 121

Aw =tI = the area of the wall’s horizontal cross-section

K = 1.2 = the shear coefficient for rectangular cross-section.

If corresponding expressions for Iw and A,,,,, and the value of K are substituted
and Eq. (7.8) is rearranged, the following general equation for effective stiffness
of a masonry wall can be obtained:

where a’ = coefficient determining the position of the bending moment’s

inflection point along the height of the wall. a’ = 0.83 in the case of a fixed-
ended and 01’ = 3.33 in the case of a cantilever wall.

3
G
9
2

Figure 7.10. Stiffness degradation of reinforced masonry walls depending

on nonnalised lateral displacements (after [23]).

In the non-linear range, however, the stifhess of a structural wall is defined

as a secant stiffness, which follows the displacements on the idealised curve. In
the case of bilinear idealisation, for example, stifhess K is defined as a fhction
of effective stiffness Ke and displacement d

(7.10)
122 Earthquake-Resistant Design of Masonry Buildings

As the analyses of experimental results indicate, the shape of stiffhess

degradation as a function of lateral displacements in’a non-dimensional form is
quite similar for all types of masonry walls, including plain, confined and
reinforced masonry (Fig. 7.10).
Two examples of simple equations of the stiffhess degradation h c t i o n , i.e.
the relationship between the stiffness ratio WKe and lateral displacements of the
walls, are presented. In the first case, the stiffhess K is calculated directly as a
h c t i o n of normalised lateral displacements d/dmax:

(7.1 1)

where a and p are the stiffness degradation parameters, which depend on lateral
load history and compression stresses due to vertical loading, acting on the wall
[23].If equation (Eq. 7.1 1) needs to be used, the values of parameters should be
determined experimentally. If there is a lack of experimental data, however, the
values of a = 0.3 and p = -0.85 can be taken into account for the case of normal
compression stresses, not exceeding 20% of masonry’s compressive strength, and
cyclic lateral loads.
In the second case, however, secant stiffness K is calculated as a function of
effective stiffness Ke and damage occurring to the walls, defined with damage
index Id. As has been found by statistical analysis and correlation between
damage to the walls and lateral stiffness values, the following two-parameter
equation best defines the degradation of stiffness [27]:

(7.12)

Correlation between the damage index I d and the extent of damage in the case
of a shear failure of a reinforced masonry wall is as follows:
I d = 0.25:the formation of the first significant crack. Crack limit.
I d = 0.50: a network of diagonally oriented cracks. Usually, the attainment of
maximum resistance.
I d = 0.75: increased width of cracks. Crushing of masonry units in the middle
of the wall, splitting of units and grout at the compressed zones of the wall.
Id = 1.00: heavy damage (beyond repair) or collapse of the wall. As has been
found by the analysis of experimental results, just before, or at I d = 1.00,
secant stiffness amounts to only 2-5 % of the wall’s effective stiffness Ke.
Seismic Resistance Verifjcationof Structural Walls 123

As was the case of parameters a and p, the stiffness degradation parameters a

and b depend on the lateral load history and compression stresses due to vertical
load, acting on the wall. As a rule, they should be determined experimentally.
The basic parameters, that need to be calculated in the case of a bi- or
trilinear idealisation of actual resistance envelopes, are indicated in Table 7.1.
Equations to calculate those parameters, as well as the physical and mathematical
models, on the basis of which the equations have been developed, will be
explained in the sections to come for each type of masonry structural walls.

Table 7.1. Parameters defining the idealised resistance envelopes.

Bilinear idealisation Trilinear idealisation

K d H K d H
Elastic limit Ke de 0.9 Hmax Ke de Ccr Hmax
Max. resistance - - 0.9 Hmax KHm, dmax &ax
Ultimate state - ~ d e 0.9Hmax KH” ~ d e CsrHmax

By representing the seismic behaviour of masonry walls with idealised bi- or

trilinear resistance envelopes, the redistribution of seismic loads from one wall to
another may be carried out by assuming that the walls as a whole, and not their
critical sections, behave as ductile structural elements.
In a way, the plastification of the wall’s section or the formation of a plastic
hinge takes place in the case of the flexural failure of reinforced masonry section.
In that case, the ductile behaviour of the wall and energy dissipation are due to
the rotation of the wall’s section at the position of a plastic hinge. However, the
idea of plastification of the section makes no sense in the case of a plain masonry
wall which rotates as a rigid body, due to cracks that developed at the wall’s
ends. It also makes little sense in the case of the shear failure of plain and
reinforced walls, which is by far the most characteristic mode of failure caused
by seismic loads. Similarly, the ductility of a masonry wall should be defined as
the capacity of the wall to dissipate energy and to carry the vertical loads at large
displacement amplitudes, despite substantial damage that has been caused to the
wall at repeated lateral load reversals.
124 Earthauake-Resistant Design of Masonry Buildings

7.4 Shear Resistance

In the case where cracks develop in the wall, oriented in one or both diagonal
directions, the shear resistance of the wall is not sufficient to resist the induced
seismic loads. Because of typical mechanical properties of masonry materials and
the geometry of structural walls, shear failure is the most common type of failure
of a masonry wall subjected to seismic loads, in all types of masonry
construction.

7.4.I . Unreinforced masonry

It has been generally accepted that shear resistance depends on vertical

compression stresses in the walls. However, two basically different hypotheses,
which lead to virtually same results, have been developed in order to physically
model the shear failure mechanism of a plain masonry wall.
In the first case, which has been accepted for shear resistance evaluation by
EC 6: Design of masonry structures [28], the shear strength of masonry fv is
defined according to fiiction theory by the expression:
(7.13)
where:
fvo = the
shear strength under zero compression stress,
p c = the constant defining the contribution of compression stresses,
od = the design compression stress, perpendicular to shear.
Values of fvo and pc should be determined by tests. This theory can be
accepted in the case of brick and/or block masonry wall built of strong masonry
units in poor quality mortar, where diagonally oriented cracks, typical for shear
failure, pass through the mortar joints. In such a case, friction between mortar
and masonry units, moving along the bed-joints during the vibration of the wall,
determines the resistance of the wall to lateral seismic loads. Shear strength
under zero compressive stress, however, should not be taken into consideration
once the wall is diagonally cracked and moves laterally during an earthquake.
In order to evaluate the lateral resistance of a structural wall, failing in shear,
the characteristic value of shear strength is multiplied by the resisting cross-
sectional area of the wall. However, for seismic resistance verification according
to EC 6, where design values of resistance and action are correlated,
characteristic value of the shear strength is reduced by material partial safety
Seismic Resistance Verijication of Structural Walls 125

factor YM for masonry, so that the design shear resistance &d,w of a plain masonry
wall is given by
- fvkt
Hsd,w Y
(7.14)
YM
where:
t = the thickness of the wall,
Zc = the length of the compressed part of the wall.
According to an alternative theory, it is assumed that diagonal cracks at shear
failure are caused by the principal tensile stresses which develop in the wall when
subjected to a combination of vertical and lateral load (7.10). By this assumption,
formation of diagonally oriented cracks, passing through masonry units in the
case of a brick-masonry wall, or passing through stones and mortar in a
homogeneous stone-masonry wall without regular mortar joints, can be easily
exp1ained.

Figure 7.1 1 . Brittle shear failure of plain masonry wall.

126 Earthquake-Resistant Design of Masonry Buildings

By considering the masonry wall as an elastic, homogeneous and isotropic

structural element, the basic equation for the evaluation of the shear resistance of
plain masonry walls can be derived by taking into account the assumptions of the
elementary theory of elasticity. Under the combination of vertical, IV, and lateral
load, H, the following principal compression:

0, =,/(+) L
+(b2)2 +%,
2 (7.15a)

and principal tensile stresses develop in the middle section of the wall:

q =I(?) 2
+(b-c)2 --.0 0
2
(7.15b)

As can be seen, the principal plane orientation

+, = & =0.5arctg-,22 (7.152)

00

is in the diagonal direction of the wall (Fig.7.11).

The meanings of symbols in Eq. (7.15) are:
oo = N/Aw = the average compression stress due to vertical load N ,
z = H/Aw = the average shear stress due to lateral load H,
Aw = the horizontal cross-section area of the wall,
b = the shear stress distribution factor, depending on the geometry of the wall
and N/Hmax ratio. In the case of the wall with geometrical aspect ratio WZ =
1.5, b = 1.5.
Assuming that the wall behaves as an elastic, homogeneous and isotropic
panel up until the attainment of its maximum resistance Hmax,referential
principal tensile stress that would develop at that instant is called the tensile
strength of masonry

(7.16)

where:
fi = the tensile strength of masonry,
ZH,,,~= the average shear stress in the wall at the attained maximum
resistance Hmm.
Seismic Resistance Verification of Structural Walls 127

Originally proposed equation has been modified to take into account the
influence of geometry of the wall and distribution of actions (ratio between
vertical and lateral load) at maximum resistance [29]. The lateral resistance Hsyw
of a plain masonry wall panel, failing in shear is evaluated by:

ft Go
Hsyw=Aw-J-+l, (7.17)
b ft
In the case where the resistance envelope is idealised as a bilinear
relationship, Eq. (7.17) is multiplied by 0.9. If, however, the design value of the
shear resistance Hsdywshould be correlated with the design seismic action,
Eq. (7.17) is modified by considering the characteristic value of tensile strength
and by taking into account the material partial safety factor yM:

(7.17a)

Evidently, Z H may ~ ~be considered as equivalent to the shear strength

obtained by Eq.(7.13):

(7.18)

7.4.2Reinforced masonry

When failing in shear, plain masonry walls behave: as brittle structural

elements with limited energy dissipation capacity, especiarly when subjected to
high compression stresses (Fig.7.11). In order to improve lateral resistance and
ductility, masonry walls are reinforced with steel reinforcement, which is placed
either in the joints between the units and embedded in the mortar, or in specially
provided holes and channels within the units and grouted with concrete or grout
(see Chapter 3 for the description of masonry construction systems).'
If a masonry wall is reinforced horizontally, the reinforcemen; prevents the
separation of the wall's cracked part at shear failure, hence improving the
resistance and energy dissipation capacity of the wall when subjected to repeated
lateral load reversals (Fig. 7.12). In the case of unreinforced masonry walls, a
single diagonal crack causes severe deterioration in strength and subsequent
brittle collapse. However, if the walls are reinforced horizontally, many cracks,
evenly distributed over the entire surface of the walls, develop (Fig. 7.13). At
128 Earthquake-Resistant Design of Masonry Buildings
~

ultimate state, crushing of masonry units due to a combination of bending and

shear is often observed, indicating that the load-bearing capacity of masonry
units is fully utilised.

1 1 I 1

-48 -32 -24 -16 -8 0 8 16 24 32 48 -18 -32 -24 -16 -8 0 8 16 24 33, 48

d (mm) / n (mm)

Figure 7.12. Comparison of experimentally obtained lateral resistancedisplacement hysteresis

loops of a (a) plain and (b) horizontally reinforced masonry wall.

I
Figure 7.13. Ductile shear failure of a horizontally reinforced masonry wall.
Seismic Resistance Verificationof Structural Walls 129

For case when the wall is reinforced with distributed vertical and horizontal
reinforcement, the basic mechanism of reinforcement action at shear failure of
the wall is shown in Fig. 7.14. Generally, the shear resistance of reinforced
masonry walls depends on different mechanisms, such as tension of horizontal
steel, dowel action of vertical steel, the combination of truss and arch-beam
action of vertical and horizontal reinforcement and masonry, as well as
interlocking between the parts of the walls separated by diagonal cracks. Few
attempts have been made to theoretically model this mechanism [2, 3, 301.
However, because of the complexity of mechanisms, not all the proposed
equations are suitable for practical design, and the validity of theoretical
solutions is limited to specific cases.

H n” H 8”

Figure 7.14. Mechanism of action of vertical and horizontal reinforcement

of a masonry wall failing in shear (after [121).

For example, the combination of arch-beam and truss mechanism of vertical

and horizontal steel reinforcement and masonry [2] assumes that compression is
carried by an arch, formed within the wall and consisting of a part of vertical
steel and masonry, whereas the tension is carried by a truss, consisting of the
remaining part of vertical steel and masonry, and horizontal steel (Fig. 7.15).
The contribution of each part to the resistance of the wall is determined by
taking into account the compatibility and equilibrium conditions. In the case of
the arch mechanism, the following conditions should be considered [ 181:
130 Earthquake-Resistant Design of Masonry Buildings

(7.19a)

(7.19b)

(7.19~)

Pa=Pkwa
Qt a Pt=Pkw

1 3

i
Arch mechanism Truss mechanism

Figure 7.15. Arch-beam and truss mechanism of a reinforced masonry wall failing in shear (after [2]).

In the case of the truss mechanism, however, the conditions are:

R = 2 &- &,ht,
~ (7.2Oa)
o wt AW +
OrhtArhdrv = 0,
(7.20b)
fi drh

(7.20~)

Symbols are explained in Fig. 7 . 1 5 . The indexes, however, are self-

explanatory: w is for wall, rv and rh is for vertical and horizontal reinforcement,
respectively, a is for arch-beam, and t is for truss mechanism.
Seismic Resistance Verifjcation of Structural Walls 131

On the basis of compatibility and equilibrium conditions, which link the arch-
beam and truss part of the model, the measured results obtained on a series of
walls, reinforced with vertical reinforcement at the ends and distributed bed joint
reinforcement, have been correlated with theoretical calculations [ 181. As can be
seen in Fig. 7.16, where the experimental and calculated values of lateral
resistance of the wall H and contributing force, developed in horizontal
reinforcement Hrh, are plotted against the rotation of the wall, a good agreement
between the measured and theoretical results can be obtained by this model.

H
40

3 30
W

Q
20

10 -Experimental
----- Calculated
I
0
0 2 4 6 8 10 12 14 16 18 20
R (%o)

Figure 7.16. Correlation between theoretical and experimentally obtained resistance

and contribution of horizontal reinforcement at shear failure [181.

In practical calculations, however, the shear resistance of reinforced masonry

walls is calculated in a much simpler way. According to recommendations in EC
6, for example, the design shear resistance of a horizontally reinforced masonry
wall Hsd r is evaluated as a sum of contributions of masonry and reinforcement.
The following equation is given in EC 6:

(7.21)

where &d,w iS given by Eq.(7.14) and Hsd rh is calculated by:

132 Earthq uake-Res is tant Design of Masonry Buildings

Hsd,rh = 0.9 d --,

fvk (7.22)
Ys

where:
d = the effective depth of the wall,
Arh = the area of shear (horizontal) reinforcement,
s = the spacing of shear reinforcement,
fyk = the characteristic yield strength of steel,
ys = the partial safety factor for steel.
In the case where masonry walls are reinforced with vertical steel, part of the
shear capacity of the wall can also be attributed to vertical reinforcement acting
in bending (dowel action - Figs. 7.14 and 7.17).

contact
R, 13

stresses
bending
moments
T 1

Figure 7.17. Dowel mechanism of vertical reinforcement at shear failure

of a reinforced masonry wall (after [ 121).

Following the mechanism shown in Fig. 7.17, the amount of shear that can be
carried by a dowel action of a single vertical bar Hd,m, embedded in the mortar or
grout, can be calculated by:

(7.23)

where:
drv = the diameter of vertical reinforcing bar,
fm = the compressive strength of embedding mortar or grout.
fy = the yield strength of reinforcing steel.
Seismic Resistance Verification of Structural Walls 133

It can be easily calculated that in the normal case, where the strength of
mortar or grout does not exceed 20 MPa (the value can be increased due to
triaxial stress state in the embedment, confined by a masonry unit) and the yield
stress of reinforcing steel does not go beyond 400 MPa, the contribution of a
single vertical bar does not exceed

Hd,rv = 0.25 d k fy , (7.2 3a)

which is 3.1 times less than the tension capacity of the same reinforcing bar
placed horizontally.
However, as the post-earthquake damage observations and experiments
indicate, vertical steel alone is not capable of contributing to the shear resistance
of masonry. Walls reinforced only with vertical reinforcement will fail in shear,
despite their predicted flexural behaviour (Fig.7.18).

Figure 7.18. Santiago, Chile, 1985: shear failure of a vertically reinforced

masonry wall (photo courtesy of M.Astroza).
134 Eartha uake -Res istan t Design of Mas onry Buildings

Also, as the test results indicate, in the case of masonry walls with mortar
bed-j oint reinforcement, the tension capacity of horizontal steel, as assumed in
EC 6, cannot be fully utilised because of the bond failure between mortar and
steel [IS, 261. Although the results of experiments clearly show that horizontal
steel acts in tension when subjected to lateral loads, as expected theoretically
(Fig. 7.19), yielding of reinforcement has not often been observed.

40
30
20
10
0
-10
-20
-30
-40
I - . I .
-50
-10 -6 -2 2 6 10 14 18 22 26 30
E(E-4)

Figure 7.19. Typical relationship between lateral load and strain

in horizontal reinforcement at shear failure.

Masonry walls with various amounts of horizontal, mortar bed-joint

reinforcement, and withlor without vertical edge reinforcement (Table 7.2) have
been tested in the laboratory [18, 261.

Table 7.2. Configuration and reinforcement of tested walls.

Seismic Resistance Verification of Structural Walls 135

Accumulation of permanent strain in horizontal reinforcement after the

unloading has been observed before yielding. This indicates that horizontal steel
keeps together the cracked parts of the wall, which tend to separate due to
vertical load, by horizontal prestressing. As a result, many uniformly distributed
diagonal cracks develop in a horizontally reinforced wall at shear failure.
The comparison of tension forces that develop in bed-j oint reinforcement
with the measured resistance of the wall shows that, before the formation of
diagonal cracks, the contribution of horizontal reinforcement to the resistance of
the wall is small. The contribution of reinforcement increases significantly after
the formation of cracks. At maximum resistance, the masonry itself still carries
part of the imposed lateral load. However, when subjected to lateral load
reversals in the non-linear range, the major part of lateral load is gradually
resisted by the reinforcing steel. Ultimately, all the resistance of the wall can be
attributed to tension in the bed-joint reinforcement, although the steel is still far
from yielding (Fig.7.20).

Figure 7.20. (a) Contribution to lateral resistance and (b) effectiveness of

horizontal bed-joint reinforcement of the wall when subjected to lateral loads.

This can be clearly seen in Table 7.3, where the results of tests are
summarised in terms of effectiveness of reinforcement CeR, expressed as a ratio
of maximum tension, measured in the horizontal reinforcing bars Hrh, to yielding
capacity of reinforcement Hrhy

(7.24)

as well as horizontal reinforcement capacity reduction factor Crh:

 Next Page

136 Earthquake-Resistant Design of Masonry Buildings

(7.25)

where Hm,, is the experimental maximum resistance of a horizontally

reinforced wall, and H m a is the experimentally obtained maximum resistance of
a referential plain masonry wall.

Table 7.3. Effectiveness of horizontal steel Ceff and horizontal reinforcement

capacity reduction factor crh depending on reinforcement ratio ph.

A trend of decreasing the effectiveness of horizontal bed joint reinforcement

with increased reinforcement ratio has been observed in all series of tests. As can
be seen, if the reinforcement has been only embedded in mortar (walls series A
and B), it has been best utilised in the case of a small reinforcement ratio.
However, in the case of stirrups embedded in mortar and bent around the vertical
steel at the ends (walls of series C and D), the tendency of decreasing of
effectiveness with increasing reinforcement ratio is not so important. The loss of
bond between mortar and steel in the bed joiqts, as well as crushing of blocks
prevented the yielding of horizontal steel in most cases.
The experiments confirmed the idea that the shear resistance of reinforced
masonry walls can be most easily assessed as a sum of contributions of the
masonry wall panel and reinforcement. However, if the wall is reinforced with
mortar bed joint reinforcement, a modification of Eq. (7.21) is proposed, where,
in the presence of the shear reinforcement, the contribution due to dowel
mechanism of vertical steel may also be taken into account. In such a case, the
design shear resistance of a reinforced masonry wall is evaluated by:

(7.26)

where:
 CHAPTER 8

MASONRY INFILLED REINFORCED

CONCRETE F U M E S

8.1 Introduction

Very often, the open space in a reinforced concrete or steel frame system is
separated with masonry filler walls, constructed between the columns and beams of
the main structure. In this sense, filler walls are partitions which determine the
architectural layout of the building in plan according to the specific needs of the
users. Since filler walls are installed after the completion of the main structural
system, masonry is an ideal material to use. Masonry walls are strong enough to
resist the out-of-plane impacts and provide adequate sound insulation. Materials can
be prepared on, or brought to the site, so that there is no need for assembling the
foms needed at the location in case concrete is used instead of masonry.
With regard to carrying vertical gravity loads, masonry filler walls represent
typical secondary, non-structural elements which add additional weight to the
basic structural system, but do not contribute to the load-bearing capacity.
However, the situation is different in the case where the building is subjected to
seismic loads, where the role and influence of masonry infill on the behaviour of
the main structure depends on the connection between the infdl and the frame.
Although considered as non-structural elements, filler walls are often rigidly
connected with the main r.c. or steel frame and impede the deformation of the
structure. In such a case, masonry filler walls represent a constituent part of the
vibrating structural system. As a result of connection, interaction forces develop
in the contact zone between the frames and filler walls, influencing the
behaviour of the infilled frame elements as well as the entire structure.
163
164 Earthquake-Res istant Design of Masonry Buildings

Bearing in mind the influence of masonry infill on the seismic behaviour of

r.c. frame buildings, the construction of masonry infill should be carefully
controlled by design. Basically, there are two approaches to control the seismic
behaviour of a r.c. frame structure with the masonry infill:
- Masonry infill is constructed as a secondary, non-structural part of the main
structural system. In that case, the infill is separated from the structure by
special details, so that it will not impede the vibration of the main structure
during earthquakes.
- Masonry infill is constructed as a resisting part of the structural system. In
that case, the beneficial influence of masonry filler walls during earthquakes
is taken into account in the design. Adequate detailing is provided to achieve
ductile behaviour of the infill and good connection between the masonry and
concrete elements.
However, since they are constructed only after the completion of the main
r.c. structure, masonry filler walls are never designed for vertical loads.
Since the contribution of masonry filler walls to the resistance of the building
is substantial and their construction is simple and economical, masonry infill is
often used to improve the behaviour of seismically inadequate flexible r.c. frame
structures. In the case of seismic retrofit or rehabilitation of r.c. frame structures
by filling the fiames with masonry filler walls, care should be taken such that the
filler walls are adequately distributed in the building’s plan and properly
connected and anchored to the basic structural system.

8.2 Seismic Behaviour and Mechanisms

The influence of masonry infill on the basic frame structural system is

manifold. In the case where a flexible r.c. frame structure is connected with rigid
filler walls, the dynamic properties of the building change. Short periods of
vibration in many cases result in increased seismic actions. In the elastic range
and at small amplitudes of vibration, those actions will be mainly carried by the
rigid infill, since the contribution of the flexible main system to the lateral
resistance of the building at small displacements is not significant. If the masonry
filler walls had not been designed to resist lateral forces at increased lateral
deformations, the masonry will be damaged. At that instant, the redistribution of
loads from filler walls to the main structural system will take place. If the failure
of the infill is brittle and the seismic actions are substantially increased just
Masonry Infilled Reinforced Concrete Frames 165

before the local collapse of the infill, severe damage to the main structural
system, which had not been designed to resist the increased lateral loads, occurs.
However, often the infill has a beneficial effect on the behaviour of buildings
subjected to earthquakes. If masonry filler walls are damaged before the
development of high shear forces, which might possibly damage the main
structural system, they dissipate seismic energy and prevent large deformations of
r.c. frames as well as damage that would occur to other non-structural elements as
a result of excessive deformations (Fig. 8.1).

Figure 8.1. Ciudad de Mexico, 1985: masonry infill

prevented collapse of r.c. frame building.

In some cases, however, interaction forces, which develop at the contacts

between the masonry infill and r.c. elements of the main system, cause
unexpected behaviour of the structure, and result in severe damage to individual
structural elements, or even partial or complete collapse of the building.
Such consequences of interaction between masonry infill and r.c. frame are
observed where the height or length of masonry filler walls is smaller than the
clear height or length of columns and beams. In such a case, masonry infill only
166 Earthquake-Resistant Design of Masonry Buildings

capacity but reduced flexibility of r.c. sections, high shear forces develop in the
free parts of columns and beams. If the critical sections of columns and beams
had not been properly reinforced, severe damage or even shear failure of
elements may occur (Figs. 8.2 and 8.3).

Figure 8.2. Budva, Montenegro, 1979: impeded deformation due to masonry

partitions at the bottom and top caused shear failure of r.c. column.

At low level of lateral load and at small lateral deformations, the masonry
infilled r.c. frame acts as a monolithic composite structural element. Because of
the initial masonry infill to r.c. frame stiffness ratio, the contribution of the
flexible frame to the lateral resistance is small and the major part of lateral
seismic load is carried by the rigid masonry infill. However, as the lateral
deformations increase, the relatively weak masonry infill is no longer capable of
carrying the increased lateral load. As the cracks develop in the masonry and the
filler wall separates into two or more parts, the r.c. frame deforms, depending on
the type of separation of the filler wall and the length of the remaining contact
zone between the masonry filler wall and frame members. In the case where the
resistance capacity of the infilled frame section, acting monolithically, exceeds
Masonry InBlled Reinforced Concrete Frames 167

the combined resistance capacity of the masonry infill and the frame, acting
separately, resistance degradation takes place (Fig. 8.4). In any case, the ultimate
resistance of the infilled frame depends on the resistance of the masonry infill
and the ultimate resistance mechanism of the r.c. frame.

w
Figure 8.3.Budva, Montenegro, 1979:impeded deformation due to
masonry infill caused shear failure of r.c. beam.

240

160

80
n

SO
-80

-160

- .

-25 -20 -15 -10 -5 0 5 10 15 20 25

d (mm)

Figure 8.4. Lateral load - displacement hysteresis loops, showing strength and
stifiess degradation after the separation of masonry infill (after [4]).
168 Earthquake-Resistant Design of Masonry Buildings

Observations of earthquake damage and comprehensive experimental

investigations in the last decades indicated various possible failure mechanisms
of masonry infilled frame systems subjected to lateral loads. Basically, the failure
mechanism is of shear type and depends on the masonry filler wall to r.c. frame
rigidity ratio, the quality of materials, and contact between the filler wall and r.c.
frame [l-81. Typical mechanisms, shown in Fig. 8.5, are characterised by:
- Sliding shear failure of masonry infill along horizontal mortar joints and
separation into several parts. Separated parts of masonry infill permit free
deformations of columns, ultimately resulting in plastic hinging of columns at
the joints between columns and beams (Fig. 8.5a). In the case of squat
masonry infills, horizontal slippage is often accompanied by diagonal cracks,
propagating from the mid-height of the filler wall to the joints between
columns and beams at the top and bottom of the infill. Energy dissipation is a
result of plastic hinging of r.c. members at the joints and friction between the
separated parts of the filler wall.

Figure 8.5. Typical failure mechanisms of masonry infilled frames.

Sliding shear failure of masonry infill and separation in two parts along
mortar joints at the mid-height. As a result of slippage of the two separated
parts of the infill, shear failure of the free parts of columns due to short
column effect may take place, with plastic hinging at the bottom and top of
the free parts of the columns (Fig. 8.5b).
Diagonal tensile cracking of the filler wall usually occurs if the masonry is
strong and the contact between the masonry and frame is good. A windward
column, supported by the infill, fails in shear, whereas plastic hinging occurs
at the bottom and top of the free to deform leeward column (Fig. 8.5~).
Although practically all types of masonry infilled r.c. frames can be classified
into the category of rigid frames with flexible infill [ 9 ] , this is not a guarantee
Masonry Inflled Reinforced Concrete Frames 169

that only one of the various possible failure mechanisms is likely to occur in the
particular case studied. There are other parameters, such as strength of the
masonry, quality of workmanship, and the amount and distribution of
reinforcement in the frame members, i.e. resistance and ductility capacity of the
frame, which also influence the ultimate mechanism. Therefore, several possible
failure mechanisms of the infilled frame should be assumed in the analysis. The
most unfavourable mechanism should be taken into account in the seismic
resistance verification of the building.

8.3 Seismic Resistance Verification

8.3.I Design seismic loads

When assessing the design seismic action effects, the fact that filler walls are
not responsible for the stability of the frame structure under gravity loads, but
change its dynamic characteristics, is taken into consideration. Therefore, the
design seismic action effects, obtained on the basis of the response of the bare
reinforced concrete frame structure, are modified by taking into account the
reduced natural period of vibration of the structure where filler walls are added to
the frames. In the determination of the ordinate s d of the design response
spectrum (see Section 6.6, Eqs. (6.5)),the average value of the first vibration mode
period T; of the infilled fiame structure, determined by:

is taken into account, where:

T l b = the first vibration mode period of the bare structure without infills,
Tli = the first vibration mode period of the structure with masonry infill.
Unless more accurate dynamic analysis to determine the first vibration mode
period had been carried out by modelling the influence of masonry infill, the first
vibration mode period of a r.c. Erame structure with masonry filler walls may be
estimated in a simplified way. According to EC 8, the following simple
expression can be used:
m
170 Earthauake-Resistant Design of Masonry Buildings

where:
A, = the average horizontal cross-sectional area of filler walls per storey in
the relevant direction,
G = the shear modulus of masonry,
g = the acceleration of gravity (g = 9.81 mi2),
H = the height of the building,
W = the weight of the building, determined by Eq. 6.3.
As an even simpler alternative, the minimum value of:
Tli = 0.065 n, (8.3a)

(8.3b)

TI i = 0.075 H3’4, (8.3~)

can be considered in the calculation of the average value of the first vibration
mode period of the infilled frame structure, where:
n = the number of storeys,
H = the height of the building (in m),
B = the width of the building in the direction considered (in m).
In the case of severe irregularities resulting from nonsymmetric arrangement
of filler walls, spatial models should be used for the analysis of the structure. In
the case of irregularities in elevation, a local increase of the seismic action
effects in the respective storey should be considered. According to EC 8, the
calculated action effects should be increased by a multiplication factor a,equal
to:

where:
AVR= ~ the total reduction of the resistance of masonry walls in the storey
concerned, compared to the more infilled adjoining storey,
ZvSd = the sum of the design seismic shear forces acting on all structural
vertical elements of the storey concerned.
In case that a < 1.1, there is no need for the modification of action effects.
Masonry Infilled Reinforced Concrete Frames 171

8.3.2 Lateral resistance

No specific guidance regarding the modelling of seismic behaviour of

masonry infilled r.c. frames is specified in EC 8 (8.10). Some general
requirements and criteria regarding the consequences of irregularities in plan and
elevation of the building, as well as local effects due to frame-infill-interaction
are only given.
On the basis of comprehensive experimental research, various models have
been proposed and various methods for the assessment of lateral resistance and
deformability of masonry infilled r.c. frames subjected to seismic loads have
been developed [2, 3, 8, 9, 11-15]. As indicated by experiments, classical finite
element models, based on the theory of elasticity, can be used for the prediction
of the linear behaviour of masonry infilled frame systems. In the non-linear
range, however, the assumptions of the theory of elasticity are no longer valid. In
order to predict the ultimate behaviour of masonry infilled frames, mathematical
models should be developed on the basis of the observed failure mechanisms.

II h hfr

ri"t
H H

ii"
Figure 8.6. Modelling of seismic behaviour of a typical infilled frame
before and after separation of masonry infill (after [3]).
172 Earthauake-Resistant Design of Masonry Buildings

As an example, an analytical model for the assessment of resistance envelope

of an infilled frame element, which behaves according to failure mechanism
shown in Fig. 8.5c, has been developed on the basis of the following
considerations [3J :
- At small deformation level and before the separation of the filler wall, the
infilled frame behaves as a monolithic structural element (stage 1 - Fig.
8.6a),
- After the separation of the infill, the infilled frame behaves as a combination
of a confined masonry wall panel and a diagonally-braced r.c. frame (stage 2
- Fig. 8.6b).
The contribution of the r.c. fiame to the resistance of the infilled panel can be
neglected at a low level of lateral deformations (stage 1). If the development of
interaction forces is ensured by good contact between the masonry and r.c. frame,
compression stresses, which increase the resistance of the masonry filler wall, are
induced in the masonry panel due to interaction forces. Eqation (7.30), explained
in Section 7.4.3 for the case of confined masonry wall panels, is modified to take
account of the stress state in a masonry filler wall, where all vertical loads are
carried by the frame and, consequently, compression stresses are the result of
interaction forces. The resistance of a single-storey, single bay masonry infilled
frame at diagonal cracking of the infill (at the end of stage 1) is therefore given
by:

where:
Aj = the area of horizontal cross-section of the infill,
ft = the tensile strength of the infill masonry,
001 =the compression stress induced in the filler wall due to interaction
forces, determined by Eq. (7.29). Since the infill is built after the
construction of the main structure, there is no contribution due to gravity
loads.
CI = 2ab/h = interaction coefficient, which takes into account the assumed
shape of distribution of interaction forces, as well as shear stresses along the
horizontal cross-section of the masonry filler wall. In the particular case
explained, the values of a = 7/8 and b = 1.1 have been determined by
experiments.
Masonry Infilled Reinforced Concrete Frames 173

Once cracked, the infill separates from the frame. As the frame deforms more
or less freely, the masonry filler wall and the frame no longer act as one single
monolithic structural element. After the separation, at stage 2, the ultimate
resistance of the masonry infilled frame is a s u m of contributions from both
resisting elements:

where:
CR= the masonry infill shear resistance capacity reduction factor, and
Hu,fr= the ultimate resistance of the frame.
The value of shear resistance capacity reduction factor should be determined
by experiments. As has been found, the value of CR varies between 0.5-0.7. In
the cases of good contact between the masonry and infill, the value of CR = 0.7
will be adequate. The ultimate resistance capacity of the frame is determined
according to the observed frame failure mode. In the case of the mechanism
shown in Fig. 8Sc, the windward column is supported by the lower part of the
infill, whereas the leeward column is free to deform. Consequently, the windward
column becomes a short column and fails in shear, whereas the leeward column
fails in bending after plastic hinges develop at the bottom and top of the column.
The following equation determines the ultimate resistance of the fiame Hu,fi in
this specific case:

where:

2MY
Hi&=hf,-

Hs,c= the shear resistance of the windward column,

= the relevant area of transverse column’s reinforcement,
Hf,c = the flexural resistance of the leeward column, assuming plastic hinges
at the bottom and top,
My= the flexural capacity of the column’s section
hfi = the height of the frame.
In the case of different frame failure mechanisms, the resistance of the frame
should be calculated by adequately modelling the relevant mechanism. In the case
174 Earthquake-Resistant Design of Masonry Buildings

of mechanism shown in Fig. 8Sa, for example, the ultimate resistance of the
frame H,,fi should be calculated as

(8.10)

As can be seen in Fig. 8.5a in the particular case discussed, plastic hinging of
both windward and leeward columns at the bottom and top has occurred at
ultimate state.
In the case where the design values should be calculated, partial safety
factors for constituent materials YM should be taken into account.

8.3.3Stiffness

At stage 1 before the separation of the infill, the effective stifhess of a

single-storey, single bay infilled frame panel is calculated by taking into account
both the shear and flexural deformations of the equivalent cross-sectional area of
the panel. Assuming fixed-ended conditions, the following equation is obtained:

(8.1 1)

where:
EC %+), 2
I , =Ii +2-I, +2 (8.12)
E E

(8.13)

and:
Ie = the equivalent moment of inertia of the composite horizontal cross-
section of the infilled frame,
Ae = the equivalent area of the composite horizontal cross-section of the
infilled frame.
Ii, Ic = the moment of inertia of the masonry infill and r.c. column,
respectively,
Ac = the area of horizontal cross-section of the r.c. column,
E, Ec = the modulus of elasticity of masonry and concrete, respectively,
Masonry InJilled Reinforced Concrete Frames 175

G, Gc= the shear modulus of masonry and concrete, respectively,

h, hfi = the height of the infill and frame, respectively,
Zfi = the span of the frame.
Eq. (8.11) should be modified to take into account the actual restraints and
boundary conditions of the masonry infilled r.c. frame element under
consideration. However, after the separation of the infill at stage 2, the stiffness
of the infilled panel is calculated by considering the structure to be an equivalent
diagonally-braced kame. The characteristics of the equivalent diagonal
compression strut are calculated on the basis of the observed failure mechanism
of the infill (Fig. 8.7).

Figure 8.7. Determination of equivalent diagonal strut (after [1 1 and 31).

In the case of situation shown in Fig. 8.7a, the effect of masonry infill is
modelled by diagonal compression strut, connecting the frame corners Ill]. The
effective height of the strut hs depends on the stiffness of the frame to masonry
filler wall ratio and ultimate lateral loading conditions. In most practical cases,
however, the height of the strut hs should be taken as 25 % of the strut’s length Is:
hs = 0.25 1s (8.14a)
For the situation shown in Fig. 8.7b, however, the effect of masonry infill is
modelled by a compression strut, which supports the windward column at 2/3 of
the filler wall’s height [13]. By making equal the stifhesses of triangular
supporting part of masonry filler wall and equivalent strut, the cross-sectional
176 Earthquake-Resistant Design of Masonry Buildings

area and the height of the strut have been evaluated. In this particular case, the
height of the strut hs should be taken as 13 % of the strut’s length Is:
hs= 0.13 Is. (8.14b)

8.3.4 Seismic resistance verifzcation

The influence of masonry filler walls on the dynamic characteristics of the

r.c. frame structure and distribution of action effects among structural members
should be assessed by calculation. Although linear static analysis is usually used
to assess the seismic action effects, the elastic stage 1 situation of monolithic
behaviour of the masonry infilled r.c. frame is not considered in the calculation.
The structure is generally verified for a stage 2 situation of separated filler walls,
where the effect of masonry filler walls is modelled by equivalent diagonal
compression struts connected to the main frame structure by means of pins (Fig.
8.8).

.c. frame

strut

Figure 8.8. Equivalent struts to model the effects of masonry infill.

The position of connection and dimensions of equivalent compression

diagonals are determined depending on the expected failure mechanism. Relevant
mechanical characteristics of masonry materials should be considered as the input
data for calculation. The main r.c. frame structure is then verified for seismic
Masonry Infilled Reinforced Concrete Frames 177

action effects resulting from the modified, and not from the original, structural
system.

8.4 References

RC frames under earthquake loading - state of the art report. CEB

Bulletin 23 1 (Thomas Telford, London, 1996).
T.P. Tassios. Behaviour of walls including infilled frames under cyclic
loading. Proc. CIB Symposium on Wall Structures (International Council
for Building and Documentation, Warsaw, 1984).
R. ZarniC and M. TomaieviC. Study of the behaviour of masonry infilled
reinforced concrete frames subjected to seismic loading. Proc. 7th Int.
Brick-Masonry Con$, Vol. 2 (Brick Development Research Institute,
Melbourne, 1985), pp. 1315-1325.
R. ZarniC, M. TomaieviC and T. Velechovsky. Experimental study of
methods for repair and strengthening of masonry infilled reinforced
concrete fiames. Proc. 8th European Con$ on Earthquake Engrg, Vol. 5
(Laboratorio Nacional de Engenharia Civil, Lisbon, 1986), pp. 11.1/41-48.
T. Schmidt. Experiments of the nonlinear behaviour of masonry infilled
reinforced concrete frames. Darmstadt Concrete, 4 (Technisce Hochscule
Darmstadt, Darmstadt, 1989), pp. 185-1 94.
T. Valiasis and K. Stylianidis. Masonry infilled WC frames under
horizontal loading. Experimental results. European Earthquake Engrg., 3
(3) (Patron Editore, Bologna, 1989), pp. 10-20.
F.M.Guerreiro Pires. Influence of masonry infill on the behaviour of r.c
frames subjected to horizontal actions. Specialist 's Thesis, Report Proc.
082/I U9346 (Laboratorio Nacional de Engenharia Civil, Lisbon, 1990 - in
Portuguese).
T. Paulay and M.J.N.Priestley. Seismic Design of Reinforced Concrete and
Masonry Buildings (John Wiley & Sons, New York, 1992).
J.R.Riddington and S.B. Smith. Analysis of infilled frames subjected to
racking loading with design recommendations. The Structural Engineer, 55
(6) (1977), pp. 263-268.
Eurocode 8: Design provisions for earthquake resistance of structures,
Part 1-3: General rules - Specific rules for various materials and elements,
ENV 1998-1-3: 1995 (CEN, Brussels, 1995).
178 Earihg uake-Res istant Design of Masonry Buildings

B. Stafford-Smith and C. Carter. A method of analysis for infilled frames.

Proc. of Institution of Civil Engineers, 44 (1 9691, pp. 3 1-49.
E. Vintzileou, T.P. Tassios. Seismic behaviour and design of infilled R.C.
frames. European Earthquake Engrg., 3 ( 2 ) (Patron Editore, Bologna,
1989), pp. 22-28.
T.N. Valiasis, K.C. Stylianidis and G.G. Penelis. Hysteresis model for weak
brick masonry infills in R/C frames under lateral reversals. European
Earthquake Engrg., 6 (1) (Patron Editore, Bologna, 1993), pp. 3-9.
R. 2arniC. Inelastic model of r/c kame with masonry infill - analytical
approach. Int. J. for Engrg. Modelling, 7 (1-2) (Split, 1994), pp. 47-54.
P.G. Papadopoulos. Simplified nonlinear seismic analysis of infilled plane
RC frames. European Earthquake Engrg., 9 ( 3 ) (Patron Editore, Bologna,
1995), pp. 39-50.
 CHAPTER 9

SEISMIC RESISTANCE VERIFICATION

OF MASONRY BUILDINGS

9.1 Introduction

Masonry buildings are typical shear-wall structures. Masonry shear-walls in

two orthogonal directions of the building, which are linked together with floors,
represent the basic resisting elements for both vertical gravity loads and
horizontal seismic loads. Consequently, basic principles, hypotheses and
mathematical models used for seismic resistant design of reinforced concrete
shear walls and shear-wall structures can also be applied to masonry buildings.
However, mathematical models developed for seismic resistance verification of
r.c. structures should be modified to take into account the specific mechanical
characteristics of masonry and constituent materials, as well as specific
structural characteristics of various types of masonry construction.
Masonry has long been considered as a traditional, non-engineered building
material. In many countries, allowable stresses method has been used for the
structural verification and design of masonry buildings. Until recently, only a
few seismic codes provided specifications and guidelines regarding the
verification of limit states of masonry structures. In most seismic codes, only the
requirements regarding the quality of masonry materials and structural
configuration and details have been given.
If allowable stresses method is used and the seismic loads are distributed
onto structural walls according to the elastic theory of structures, it can usually
be seen that it is practically impossible that all structural elements attain the
allowable stress state simultaneously. For technological reasons, which do not
179
180 Earthmake-Resistant Design of Masonry Buildings

permit the dimensions of masonry walls and quality of masonry materials to vary
in a single storey, there will be only a few sections of masonry walls in the entire
building which will be fully stressed at the design loading conditions.
Consequently, the actual load-bearing capacity of a masonry structure will be
greater than the design load, multiplied by a global structural safety factor
(Fig.9.1).

Allow. Limit
str ss states Mechanism
rne&od method

Figure 9.1. Typical correlation between actual and design resistance of structure.

In the case of the limit states verification, the sectional capacity of structural
walls, determined by taking into account the partial material safety factors, is
compared with the sectional forces (action effects), calculated on the basis of the
elastic theory of structures. Since specific possible failure modes of individual
structural walls are taken into consideration in the analysis, the actual lateral
load-bearing capacity of the structure is better estimated than in the previous
case. However, if no redistribution of loading is accounted for in the calculation,
the structure still possesses a substantial reserve in strength, which will result in
an underestimated value of the design seismic resistance.
It is the ductility of structural walls and their sections which makes it
possible for the structural systems to resist earthquakes. Because of ductility, the
vertical load-bearing capacity of structural walls is maintained, despite the fact
that the walls are damaged, deformed, and displaced when subjected to seismic
loads. The ductility of walls makes possible the redistribution of seismic loads
from the most damaged walls to other less damaged or undamaged walls, and
makes possible the dissipation of energy during the seismic response of the
building. The structure or part of a structure resist the seismic loads as long as
Seismic Resistance Verijkation of Masonry Buildings 181

they are not changed into a mechanism. In a hypothetical case of structural

elements with ideal elastic-perfectly plastic non-linear sectional characteristics,
this will occur when the structure attains the maximum resistance, but is no
longer stable for lateral loads. Depending on the distribution and number of
elements or sections of elements with severe non-linearity, either partial or
complete mechanisms will develop, and the structure will collapse either locally
or globally. It is obvious that in the case of the mechanism, the internal forces,
i.e. action effects, are no longer distributed among the walls according to the
rules of the theory of elasticity. It is also obvious that in the case where the
resistance of a masonry structure is determined on the basis of the assumed
ultimate mechanism, the difference between the calculated design and actual
resistance of the structure is reduced to a minimum.

9.2 Calculation Procedures

In the case of masonry structures, there is usually no need for sophisticated

non-linear dynamic analyses to be carried out for seismic resistance verification.
Because of the regularity and simplicity of a typical masonry structure, an
equivalent static analysis will provide reliable information regarding the safety of
the building under consideration against expected seismic loads.
It is also not necessary that, because of the tri-dimensional nature of seismic
ground motion, complex spatial mathematical models are used to model the
response of masonry buildings to earthquakes. In normal cases, the verification of
the seismic resistance of the structure in two orthogonal directions will be
appropriate.
No specific procedures or algorithms for seismic resistance analysis and
verification are required by EC 8 111. However, the calculation procedure
depends on whether linear or non-linear methods are used for assessing the
seismic action effects. The following general sequence of calculation is usually
carried out for linear analysis and seismic resistance verification:
- The weight of the building, concentrated at floor levels, is determined by
taking into account the combination of gravity loads according to Eq. (6.3).
- Using appropriate mathematical models (Eq. 7.9),the stiffness of individual
walls in the storey under consideration is calculated. The stiffness matrix of
the entire structure is evaluated.
- The period of vibration is calculated when necessary and the ordinate of
response spectrum determined according to Eqs. (6.5).
182 Earthauake-Resistant Design of Masonry Buildings

- The design base shear is determined according to Eq. (6.6).

- The base shear is distributed along the height of the building according to
Eqs. (6.8) or (6.9).
- The storey shear is distributed among the walls according to their stiffness
and the design values of action effects are calculated according to Eq. (6.2).
- Finally, the design resistance of wall sections is calculated and compared to
the design action effects.
As specified in EC 8, in the case where linear analysis is used for the
distribution of the total base shear among structural walls, the redistribution of
shear to the walls is possible, provided that global equilibrium is satisfied and the
shear in any wall is neither reduced more than 30 % nor increased more than 50 %.
In the case of non-linear static analysis, the first steps of the procedure,
which are needed to determine the design seismic loads, are identical to the
sequence for linear analysis. However, the seismic resistance of the structure is
calculated on the basis of the assumed ultimate resistance mechanism, which
includes the redistribution of action effects to individual walls according to the
attributed ductility capacity. In the final verification, the calculated design
resistance of the storey under consideration is compared with the design shear. In
that case, there is no need to compare the design resistance of individuals walls to
the design action effects. The results of the analysis themselves will indicate the
potential weak points in the structural system, and will make it possible for
improvement in the strength and ductility of critical walls by redesign.
The idea that the propagation of damage and development of plastic hinges in
a r.c. frame structure can be followed by a step-wise increase of the lateral loads,
acting on the structure, is several decades old. At each step of the analysis, i.e. at
each increment of lateral loads, the distribution of loads among structural
members is carried out on the basis of their stiffness ratio, assessed in the
previous step of calculation. If a plastic hinge is formed in any of structural
members, the structural system is accordingly modified and the calculation is
continued by analysing the modified structural system. The procedure is
terminated when a sufficient number of plastic hinges is formed so that the
structure becomes a mechanism. By definition of a mechanism, the lateral load
can no longer increase and the structure becomes unstable. Today, the calculation
procedure of this kind is known as the “push-over method”.
When applied to masonry buildings, the idea of increasing the lateral load
has been modified, following the principle of laboratory testing of masonry walls
and assemblages: displacements, and not lateral loads are imposed on the
Seismic Resistance Verification of Masonry Buildings 183

structure to be tested. Therefore, in the calculation, also, the structure is

displaced and the resistance of structural elements to the imposed displacement is
calculated [2, 31. This way, the behaviour of the structure can be followed by
calculation all the way to ultimate collapse.

9.3 Structural Models

In order to obtain reliable information on the seismic resistance of the

building, the structural model used for the seismic resistance verification should
be capable of representing the basic features of the behaviour of both the entire
system and individual structural elements in the linear and non-linear range.
According to EC 8, the stiffness of structural walls should be evaluated by
taking account of both their flexural and shear deformability, but also the axial
defonnability, if relevant. Floor diaphragms may be considered as rigid in their
own plane, if their openings do not significantly affect the overall in-plane
rigidity of the floors. Frame analysis may be used for the determination of the
action effects, provided that masonry spandrels, taken into account as coupling
beams, be regularly bonded to the adjoining walls and connected to the bond-
beams and lintels.

Deformation and crack pattern Shears and bending moments

Figure 9.2. Cantilever walls linked by flexible floor slabs.

Regarding the structural configuration, shear walls, solid or pierced by

window and door openings in each storey, represent the basic structural elements
184 Earthquake-Resistant Design of Masonry Buildings

of a masonry structure resisting the seismic loads. Considering 'their

configuration, the type of construction, and resulting seismic behaviour and
failure mechanism, masonry shear walls are classified into three main categories
141
Cantilever walls (Fig. 9.2), where two or more cantilever walls are connected
together with floor slabs. In the case of an earthquake, seismic forces are carried
by the walls. Floor slabs, rigid in their plane but flexible in the orthogonal
direction to the plane, distribute the lateral loads onto the walls in proportion to
their stiffness. However, they do not transfer any moments resulting from the
bending of the walls. Consequently, large bending moments that develop in the
walls require strong reinforcement and careful detailing at the lower-most
sections of the cantilevers in order to provide ductile behaviour of the structure.

Deformation and crack pattern Shears and bending moments

Figure 9.3. Coupled shear wall with weak piers.

Very often, masonry shear walls are pierced by window and door openings.
Above and below the opening, spandrels connect the walls and transfer the
seismic forces. Depending on the proportion of openings, either piers are
relatively weaker than spandrels (Fig. 9.3) or spandrels are relatively weaker
than piers (Fig. 9.4).
If the piers are weaker than spandrels, which is most often the case in
traditional unreinforced masonry construction, damage will first occur to the
piers (Fig. 2.17). The spandrels are rigid so that the piers may be considered as
fixed above and below them. Depending on the geometry and quality of masonry
Seismic Resistance Verijlcation of Masonry Buildings 185

materials, the piers will either fail in shear due to diagonal compression or rock
until crushing of masonry occurs at the compressed zones. Shear failure of
masonry piers is the most frequent failure mechanism of unreinforced masonry
buildings. However, with regard to the ductility and energy dissipation capacity,
this mechanism is not favourable. Although the pier action of unreinforced
masonry walls is not so fatal for a masonry structure compared to hinging of
columns in the case of a r.c. frame, the non-ductile behaviour of weak piers
should be improved by means of adequately distributed bed joint reinforcement.

Deformation and crack pattern Shears and bending moments

Figure 9.4. Coupled shear wall with weak spandrels.

In the case of reinforced masonry structures, spandrels, including bond-

beams and effective parts of floor slabs, are sometimes weaker than piers. If such
is the case, spandrels behave as coupling beams, which connect the walls together
and transfer bending moments. If adequately reinforced, the piers and spandrels
will be subjected to predominant flexural behaviour that will significantly
improve the lateral resistance and energy dissipation capacity of the structure.
Damage will occur not only to piers, but also to spandrels. As a result of the
carrying over of bending moments, the amount of reinforcement in the walls is
reduced and energy dissipation capacity is distributed over the entire height of
the shear wall. Therefore, the mechanism of coupled walls with spandrel hinging
represents the most desirable mechanism among the three mechanisms presented.
In practical calculations, shear walls are most often modelled as frames (Fig.
9.5). As can be seen, shear forces and bending moments are induced in the piers
186 Earthquake-Res istant Des inn of Masonry Buildings

and spandrels of the idealised frame structure. As a result of overturning

moments, additional axial forces develop in the piers, which, depending on the
direction of seismic loading, either increase or decrease the stresses in the piers
due to gravity loads. Considering the fact that the masonry structure behaves as a
monolithic box, the level of additional stresses induced in the piers due to
overturning moments is relatively small. Rarely, however, foundations should be
verified for overturning at the attainment of the ultimate resistance of the upper
structure.

Wall 1 Wall 2

n-th storey

i-th storey

1st storey

Figure 9.5. Distribution of action effects in a shear wall: bending moments (M),
shear forces (Q), and axial forces (N).

In the case of reinforced masonry, where reinforcement in the walls and

spandrels provides the carrying of bending moments, the resultant seismic forces
in each storey are distributed among individual piers and spandrel beams
according to their stiffness. The shear forces carried by individual piers are
determined by

where
F = ZCFi = the seismic shear in the storey considered, and
Kei = the effective stiffness of the i-th wall in the storey considered.
Seismic Resistance Verification of Masonry Buildings 187

The effective stiffness of the pier is calculated by taking into account the
actual restraints. Knowing the shear force acting on individual walls, the bending
moments at the bottom n/ii,b:
(9.2a)

Mi,t = (1 -a)hi Qi (9.2b)

of the i-th pier can be calculated, where:

a = the coefficient determining the position of the moment inflection point.
a = 0.5 for the case of a fixed-ended, and a = 1.0 for the case of a cantilever
wall.
hi = the height of the 6th pier.
Equilibrium of moments at the pier - spandrel joints is the starting point for
the calculation of moments at the ends of spandrel beams. Additional axial forces
induced in the piers due to seismic loads are calculated on the basis of the simple
theory of structures.

Wall I Wall i Wall n

n-th storey

1st storey

Figure 9.6. Distribution of action effects in a shear wall with rigid spandrels and
unreinforced masonry piers: bending moments (M), and shear forces (Q).

Frame analogy may be simplified in the case of unreinforced masonry

buildings, where spandrels are rigid and strong, and the flexural capacity of the
walls’ sections is low SO that predominant shear behaviour of piers is expected. In
188 Earths uake-Resistant Design of Masonry Buildings

this case, the piers may be considered to be symmetrically-fixed at the top and
bottom to the spandrel beam (Fig. 9.6). Under such an assumption, the piers are
fixed-ended and the bending moments caused by the shear forces at the bottom
and top of the piers are calculated simply by
Qi hi
(9.3)
Mi =2 9

i.e. a , the coefficient determining the moment inflection point, is equal to 0.5.

9.4 Storey Resistance Envelope

As has been shown by shaking-table tests of models of masonry buildings, the

first mode of vibration prevails in the response of masonry structures to
earthquakes [5-81, Although the models of buildings with different masonry
construction types have been tested, such as plain, confined and reinforced
masonry, the shape of the first vibration mode was typically shear, especially at
the ultimate state of the seismic behaviour (Fig. 9.7).

1.oo

Figure 9.7. Typical first vibration mode shape of (a) unreinforced, (b) confined,
and (c) reinforced masonry building at ultimate state (after [5,6]).

In the case of masonry buildings, which fulfill the basic requirements for
rigid horizontal diaphragm floor action, a concentration of damage in the lower-
most storey has been observed by both post-earthquake observations and
experiments. This is a result of structural concept, where constant dimensions of
the sections of structural walls along the height of the building do not follow the
intensity of action effects, namely, because the predominant first vibration mode
shape, maximum shear, and bending effects develop in the first storey. However,
Seismic Resistance Verifjcation of Masonry Buildings 189

although the increased level of vertical loads improves the resistance of the walls
in the first storey, the increased effects of lateral loads prevail.
Fig. 9.8 shows an example of ultimate state of a three-storey confined
masonry building model tested on a shaking table. It can be clearly seen that,
although the damage is distributed over the entire height of the structure, the
ultimate damage, causing collapse, was concentrated in the lower-most storey.

Figure 9.8. Damage to a three-storey confined masonry

building model at ultimate state.

Observations and experiments conclude that many parameters regarding the

seismic resistance of a masonry building can be determined if the relationship
between the resistance of the critical storey and corresponding storey drift can be
assessed. In other words, if the so-called storey resistance envelope, representing
the full-line curve in Fig. 6.3 where the definition of structural behaviour factor q
is explained, can be calculated, the ultimate and serviceability limit states of
masonry structures can be easily verified.
In the following section, the idea of subjecting the structure to imposed
lateral displacements and to calculate the resisting forces, or pushing the
190 Earthquake-Resistant Design of Masonry Buildings

structure and observing the damage and redistribution of action effects, will be
discussed.

9.4.I Assumptions

The storey resistance envelope, which determines the relationship between

the resistance and relative storey drift, is obtained as a superposition of resistance
envelopes of all walls in the storey under consideration. Assuming the boundary
restraints and relevant mechanism (cantilever walls, fixed-ended piers, piers
coupled with spandrels), the resistance envelope of each wall in the storey can be
evaluated using relevant equations explained in Section 7 for the shear and
flexural resistance, as well as the initial effective stiffness of the wall. Either
bilinear or tri-linear idealisation, as explained in Section 7.3, may be used to
represent the resistance envelope of each contributing wall.
In the calculation, the following assumptions are taken into account:
Rigid horizontal floor diaphragm action. Structural walls are connected
together with rigid floors and bond-beams, so that the displacements and
action effects are distributed to the walls in proportion to their stiffness.
Differential displacements and action effects due to torsional rotation are also
distributed to the walls in proportion to their stiffness.
Predominant first vibration mode shape. When pushing the structure, lateral
displacements in the form of the first vibration mode shape are imposed. In
order to simplifL the calculation, inverse triangular distribution of
displacements over the height of the building is sometimes assumed.
The contribution of an individual wall to the lateral resistance of the storey
depends on the lateral displacement attributed to that wall and the shape of
the wall’s resistance envelope. The walls resist the imposed displacements up
to the attainment of their ductility capacity. Although they fail for lateral
loads, they are still assumed to carry the vertical loads.
Depending on the masonry construction system, the walls of composite cross-
sections, such as L, T, and + shaped walls, may be considered as separate
along the vertical joints between their parts.
Seismic Resistance Verifwation of Masonry Buildings 191

9.4.2 Procedure for calculation

Different shapes of the distribution of imposed displacements along the

height of the building can be used in the analysis. Usually, the first vibration
mode shape is taken into account. Sometimes, inverse triangular distribution of
displacements is used to simplify the calculation. In the most simple case of a
pier mechanism of walls and the prevailing rocking of the walls at ultimate state,

7
only the critical storey of the building needs to be displaced (Fig. 9.9).

m
1 .oo 1.oo

1 .oo

1.33 1 .oo

7 (b)

Figure 9.9. Typical distribution of imposed displacements along a building’s height for push-over analysis:
(a) first vibration mode shape, (b) inverse triangle, and (c) pier mechanism.

In order to assess the first vibration mode shape and stress-state in structural
walls due to gravity loads, the masses, concentrated at floor levels, and
stiffnesses of individual walls are determined. On the basis of distribution of
masses and stiffhesses along the height of the building, the first vibration mode
shape can be calculated. On the basis of compressive stresses that develop in the
walls due to gravity loads, lateral resistance of each individual wall can be
assessed. Knowing the lateral resistance and initial (effective) stiffness, the
resistance envelope of each individual wall in the storey under consideration can
be calculated. Depending on the shape of the idealised resistance envelope, the
parameters, which define the characteristic points of the envelope, need to be
calculated as specified in Table 7.1.
To take account of the possible torsional effects, the positions of the storey
mass and stiffness centres are calculated and eccentricities of the resultant
seismic loads acting on the building in the two orthogonal directions of the
building are determined. The position of the storey mass-centre does not always
192 Earthquake-Res is tant Design of Masonry Buildings

coincide with the position of the storey stiffhess centre. As a result of

eccentricity, rotation of the storey around the storey stiffness centre causes
additional stresses to structural elements.

Ys

Figure 9.10. Eccentricity of storey mass and stiffness centre.

In the case of non-symmetry, the actual eccentricity of the mass and stifhess
centre should be determined by calculation. For example, the eccentricity ex in
the direction perpendicular to the direction of seismic load Hy, is calculated by
(see Fig. 9.10):

e,=x,-x,, (9.4)
where:

are co-ordinates of the storey mass and stiffness centre, respectively,

perpendicular to the direction of seismic load Hy, and
Seismic Resistance Verification of Masonry Buildings 193

CToi = the average compressive stress in the horizontal section of the i-th wall,
A W i = the cross-section area of the i-th wall,
Xi = co-ordinate of the area centre of the i-th wall, perpendicular to the direction
of seismic load Hy,
Kyi = the stifhess of the i-th wall in the direction of seismic load Hy

Eccentricity in the perpendicular direction is evaluated in a similar way.

According to EC 8 [9],to cover the uncertainties in the location of masses and in
the spatial variation of the seismic motion, the calculated position of the mass-
centre at each floor should be considered displaced from its nominal location in
each direction by an accidental eccentricity of at least:
e=+0.05L, (9.7)
where:
e = the accidental eccentricity of the storey mass from its nominal location,
L = the floor dimension perpendicular to the direction of the seismic action.
Additional accidental eccentricity should be taken into account when
evaluating the torsional effects.

Elastic limit Maximum resistance Ultimate state

d"

de n

Figure 9.11. Characteristic steps of push-over analysis of a masonry structure.

In the calculation of the storey resistance envelope, the structure is displaced

by a small value, assuming the chosen shape of distribution of displacements
along the height of the building. The shear walls are deformed according to the
194 Earthauake-Resistant Design of Masonry Buildings

assumed structural model and the resisting forces in structural members are
calculated. If torsional rotation takes place by displacing the mass centres of
individual storeys, the displacements of individual piers are accordingly
modified.
The calculation is repeated step-by-step by increasing the imposed
displacements (Fig. 9.1 1). Once the walls enter into the non-linear range, the
structural system of the building and stiffness matrices are modified. The
stiffness and resistance of individual walls in each step of calculation are
determined considering the calculated storey displacements and idealised
resistance envelopes of structural walls in each storey.
As a result of calculation, the relationship between the resistance of the
critical story and interstorey drift, i.e. the resistance envelope, is obtained. At the
given lateral displacement of the i-th wall d,, the resisting storey shear Htot is
determined as a sum of resistances of structural walls Hi in the storey under
consideration (Fig. 9.12):

H t

Figure 9.12. Construction of storey resistance envelope on the basis

of bilinear resistance envelopes of structural walls.

where the resistance of the i-th wall H , depends on the deformation of the wall d;.
In case the resistance envelope of structural walls is idealised with a bilinear
Seismic Resistance Verification of Masonry Buildings 195

relationship, the following simple conditions determine the contributing stiffness

and resistance of individual structural walls:
(9.9a)

(9.9b)

(9.9c)

where:
di = the displacement of the i-th wall,
dei = the displacement of the i-th wall at the attained elastic limit,
dui = the ultimate displacement of the i-th wall,
Hui = the resistance capacity of the i-th wall,
Ki = the stiffness if the i-th wall,
Kei = the effective stiffness of the i-th wall,
n = the number of walls.
Depending on the masonry construction system, different ductility capacities
are attributed to different masonry systems (see Section 7.3). Taking this into
consideration, the ultimate displacement of each wall type in the critical storey is
determined by taking into account the recommended value of ultimate ductility
factor pui:
dui = Pui dei- (9.10)
Beyond the ultimate displacement dui, the i-th wall no longer resists the
lateral load. However, it is assumed that it still carries the gravity loads.
Three characteristic limit states are defined on the resistance envelope of the
critical storey considered in the calculation:
Elastic limit, where the displacement of the first wall in the storey exceeds
the elastic limit,
Maximum resistance, and
Ultimate state, where the resistance of the storey deteriorates beyond the
acceptable limit. In practical cases, degradation below 70 % of maximum
resistance may be defined as the collapse of the building.
196 Earthq uake-Res istant Design of Masonry Buildings

Needless to mention, the resistance envelope of the critical storey is

calculated and the seismic resistance is verified for the two orthogonal directions
of the building under consideration.
In cases where the seismic resistance verification of the building is carried
out according to the requirements of EC 8, characteristic values of mechanical
properties of masonry materials and partial material safety factors YM should be
considered in the calculation of idealised resistance envelopes of structural walls.
Namely, by EC 8, the design situation should be taken into account when
verifying the seismic resistance of masonry buildings. Therefore, the design
resistance envelope, and not the actual envelope of the critical storey should be
calculated.
By assessing the storey resistance envelope, the situation of each wall in the
storey considered can be followed. Depending on the layout of the computer
programme used for the calculation of the envelope, the state of individual walls
at characteristic limit states of the storey resistance envelope is indicated, so that
. I

the critical walls in the storey can be identified and redesigned, whenever
necessary.

9.4.3 Experimental verification

Experimental results obtained from shaking-table tests of models of

unreinforced, confined and reinforced masonry buildings have been used to
verify the validity of the proposed method [ 5 , 6, 81. In all cases, good correlation
between experimental and theoretical results has been obtained, confirming the
general usability of this relatively simple static non-linear method for the
determination of the storey resistance envelope.
As an example, experimentally-obtained and calculated first storey resistance
- first storey drift envelopes, obtained for the shaking-table test of the confined-
masonry building model, shown in Fig. 9.7,are correlated in Fig. 9.13.A similar
degree of correlation between experimental and calculated results has been
obtained in other cases.
In the particular case shown in Fig. 9.13, trilinear idealisation of resistance
envelopes of confined masonry walls has been used in the calculation of storey
resistance envelope. It should be noted that average values of mechanical
properties of masonry materials, without taking into account the partial material
safety factors YM have been considered in the calculation. Therefore, the
Seismic Resistance Verification of Masonry Buildings 197

calculated resistance envelope represents the actual ultimate, and not the design
ultimate resistance envelope.

35 -.R100
30
25
20
4
.el
Q IS
I0
5
0
0 10 20 30 40
d (-1

Figure 9.13. Correlation between experimental and calculated storey

resistance envelopes in the case of confined masonry building model (after [6]).

9.5 Seismic Resistance Verification

According to the requirements of EC 8 [l], the structural system should be

verified for the required strength and ductility properties. The strength and
ductility assigned to the structure, are related to the capacity of the structure to
dissipate the seismic energy and to withstand non-linear deformations without
losing the stability. In this regard, the extent of permissible damage occurring to
the structure in the non-linear range, should also be carefully evaluated and,
consequently, the serviceability limit of the structure should be verified. Due to
the structural configuration, however, there is usually no need for masonry
buildings to be verified also for the serviceability limit state.
Basically, such a balance between the required strength and ductility is
specified by the value of the behaviour factor q, which varies according to
different masonry construction systems (see Section 6.3.5). If the structure is not
able to dissipate any energy, it should be designed for strength. In such a case, no
account is taken of any hysteretic energy dissipation, and the behaviour factor is
equal to q = 1.0.
Besides strength and ductility, the structure as a whole should also be
verified for overturning and sliding stability. It should be verified that the
198 Earthquake -Res istant Design of Masonry Buildings

foundation elements and foundation soil are capable of resisting the ultimate
action effects resulting from the response of the superstructure, without
substant ial permanent deformations.
Taking into consideration the basic requirements for strength and ductility,
which should be filfilled in the case of a seismically resistant structure, the
verification of the seismic resistance of masonry buildings on the basis of the
calculated resistance envelope has two major advantages:
- design resistance capacity of the structure under consideration can be easily
correlated with the design seismic action, and
- design ductility capacity of the structure can be assessed and compared with
code requirements for the specific type of the structure considered.
The procedure is explained in Fig. 9.14, where the calculated resistance
envelope of an actual masonry building is presented in terms of non-dimensional
quantities, such as seismic resistance coefficient SRC, which is the ratio of design
resistance to the weight of the building, and storey drift rotation angle 0,which
is the ratio of storey drift to storey height.

Figure 9.14. Seismic resistance verification on the basis

of resistance envelope of an actual masonry building.

In order make the verification easier, the calculated envelope is smoothed

and presented as a bilinear relationship. The same principle of equality of areas
below the calculated and idealised envelopes is taken into account, and a similar
Seismic Resistance Verification of Masonry Buildings 199

equation as in the case of bilinear idealisation of lateral resistance curve of

masonry walls (Eq. 7.2) is used to determine the idealised storey resistance
envelope. In the idealisation, the initial stifhess of both envelopes is assumed to
be equal.
Once the design resistance curve is idealised, the global ultimate ductility
factor of the structure under consideration can be assessed as a ratio between the
ultimate rotation and rotation at the attained elastic limit:

(9.1 1)

where:
if>,, = the ultimate storey drift rotation angle,
if>e = the storey drift rotation angle at elastic limit.
In the verification of the seismic resistance of an actual building structure,
the design resistance in terms of ultimated design seismic resistance coefficient
SRCd, is compared with the design base shear coefficient BSCd. In case that the
basic condition of seismic resistance:
SRCd, 2 BSCd, (9.12)
is fulfilled, the global ductility requirement should be also verified. On the basis
of the simple definition of structural behaviour factor, shown in Fig. 6.3, the
following relationship between the structural behaviour factor q, and global
ductility factor pu can be derived:

(9.13)

This gives an estimate for global ductility demand

- q2 + 1
PU - (9.14)
2

Taking relationship (9.14) into consideration, and considering the values of q

factors proposed in EC 8 for various existing types of masonry construction, the
following global ductility demand requirements should be fulfilled for the case
Of:
0 Unreinforced masonry buildings: q = 1.5, pu = 1.6,
Confined masonry buildings: q = 2.0, p,, = 2.5, and
200 Eartha uake-Res istant Design of Masonry Buildings

Reinforced masonry buildings: q == 2.5, pu = 3.6.

In case the calculated values of global ultimate ductility factors pu according
to Eq. (9.1 1 ) exceed the above values, and the basic condition for seismic
resistance given in Eq. (9.12) is fulfilled, the seismic resistance of the building
under consideration fulfills the requirements of EC 8 for earthquake-resistant
masonry buildings.
However, if the resistance envelope meets the requirements of Eq. (9.12), but
does not hlfill the requirements for global ductility demand (Eq. 9.1 l), the
ductility should be verified at a lower resistance level. If, at the level which
conforms with the required global ductility, the resistance level is still in
accordance with Eq. (9.12), the structure may be considered to be adequate. If
not, the structure should be redesigned.

9.6 References

Eurocode 8: Design provisions for earthquake resistance of structures,

Part 1-1: General rules and rules for buildings - Seismic actions and
general requirements for structures. ENV 1998-1-1 : 1995 (CEN, Brussels,
1995).
M. TomaieviC, V. TurnSek and S. Tertelj. Computation of the shear
resistance of masonry buildings. Report ZRMK-ZK (Institute for Testing and
Research in Materials and Structures, Ljubljana, 1978).
M. TomaieviC and V. TurnSek. Verification of the seismic resistance of
masonry buildings. Proceedings of the British Ceramic Society No. 30,
(Shelton House, Stoke-on-Trent, 1982), pp. 360-369.
T. Paulay and M.J.N. Priestley. Seismic Design of Reinforced Concrete and
Masonry Buildings (John Wiley & Sons, New York, 1992).
M. TomaieviC and P. Weiss. Seismic behavior of plain- and reinforced-
masonry buildings. J. Struct. Engrg., 120 (2) (ASCE, New York, 1994), pp.
323-338.
M. Toma2eviiJ and I. Klemenc. Verification of seismic resistance of
confined masonry buildings. Earthquake Engrg. Struet. Dynamics, 26 (1 0 )
(John Wiley & Sons, Chichester, 1997), pp. 1073-1088.
M.P. Limongelli and I). Pezzolli. Analysis of the seismic response of
masonry buildings excited by a shaking table. European Earthquake
Engrg., 7 (2) (Patron Editore, Bologna, 1994), pp. 18-30.
Seismic Resistance Verification of Masonry Buildings 201

[ 8J A.C.Costley and D.P.Abrams. Dynamic response of unreinforced masonry

buildings with flexible diaphragms. Technical Report NCEER-96-0001
(National Center for Earthquake Engineering Research, Buffalo, 1996).
[91 Eurocode 8: Design provisions for earthquake resistance of structures,
Part 1-2: General rules - General rules for buildings. ENV 1998-1-2: 1995
(CEN, Brussels, 1995).
 CHAPTER 10

REPAIR AND STRENGTHENING

OF MASONRY BUILDINGS

10.1 Introduction

By definition, “repair” refers to the post-earthquake repair of damage,

caused by seismic ground motion, that does not increase the seismic resistance
of a structure beyond its pre-earthquake state. “Strengthening”, “seismic
strengthening”, or “seismic upgrading”, however, comprises technical
interventions in the structural system of a building that improve its seismic
resistance by increasing the strength and ductility. According to the proposed
terminology 113, strengthening a building before an earthquake is called
“rehabilitation”, whereas strengthening after the earthquake is called “retrofit”.
Therefore, the decision to repair a building after an earthquake or
strengthen it before an earthquake occurs depends on the building’s seismic
resistance. If the resistance is adequate and damage which occurred in the
structure after an earthquake is a consequence of energy dissipation processes
predicted by the design, the structure will be repaired in order to return the
building to its original seismic resistance.
If, however, the degree of damage in the building during earthquake is
beyond the predicted level, then the seismic resistance of a building was
obviously not adequate. When the decision is made to retain an earthquake-
damaged building, its structural system should be adequately strengthened in
order to attain the desired level of seismic resistance. In this case, however,
strengthening of the structural system also includes repair of damaged structural
and non- structural elements.
203
2 04 Earrhq uake -Res istant Design ofA4asonry Bu iidings

As has been learned from all recent earthquakes, modern-including masonry

buildings-designed according to requirements of state-of-the-art earthquake
engineering, successfully resist strong ground motion, whereas many existing,
non-engineered as well as engineered buildings, “old” by the standards of
earthquake-resistant design, and not merely by the age of construction, collapse
or suffer severe damage.
In Asia, Latin America, and also in Central Europe, the Mediterranean
region, as well as in the USA, old masonry buildings of all types of construction,
including historical monuments, represent an important part of the existing
building stock. Since old masonry buildings are typical representatives of
traditional non-engineered construction, their seismic vulnerability is,in general,
relatively high. Indeed, most earthquake damage and loss of life in these regions
is caused by inadequate seismic behaviour of existing masonry buildings, in most
cases residential houses in urban and rural areas which are of traditional type of
construction (Fig. 10.1). On the other hand, there have been cases where
adequately constructed masonry buildings survived even the most devastating
earthquakes (Fig. 10.2).

Figure 10.1.Podbela, Slovenia, 1976: Historic village destroyed by the Friuli earthquake.

Insufficient attention has been paid to the post-earthquake repair of masonry

buildings. Many buildings have been repaired on the basis of the designer’s
intuition, without knowing the cause of damage. Sometimes, buildings have been
repaired cosmetically by covering cracks with plaster and paint, believing that
Repair and Strengthening of Masonry Buildings 205

another strong earthquake will not occur again in the building’s life-time. The
situation was worse in building reconstruction and renewal, where fatal errors
have been made by removing structural walls, adding new levels, etc., without
adequate strengthening of the existing structure.

Figure 10.2. Tangshan, China, 1976: A masonry building which survived the earthquake.

As a significant part of EC 8: Design provisions for earhquake resistance of

structures, Part 1-4: General rules - Strengthening and repair of buildings, has
recently been approved by CEN as prestandard [2]. Although this is not the first
attempt that basic guidelines for seismic repair and strengthening of masonry
buildings have been given in specific codes [3, 41 or manuals and
recommendations[5-7 1, Part 1-4 of EC 8, which was “specifically developed
because:
- For most of the old structures seismic design was not considered originally,
whereas the ordinary actions were considered, at least by means of traditional
construction rules.
- Seismic hazard evaluations in accordance with the present knowledge may
indicate the need for strengthening campaigns.
206 Earthquake-Resistant Design of Masonry Buildings

- The occurrence of earthquakes may create the need for important repairs,”
includes basic considerations for evaluation and seismic redesign of existing
buildings, based on the same philosophy as in the case of a new construction, and
provides technical recommendations for seismic strengthening of all structural
types, including masonry structures.
Old masonry buildings in historical urban and rural nuclei require specific
treatment. Whereas monumental buildings, classified into the category of cultural
monuments, are architectural masterpieces as individual buildings-which, in
most cases, have retained their original form for centuries-historical urban
nuclei are monuments represented by clusters of buildings, originally built as
individual units, but reconstructed and rebuilt in the course of time to meet the
needs of their residents, finally obtaining the form of today.
Why do old buildings need strengthening? In most cases, structural systems need
repair and strengthening either because of deterioration of structural materials as a
consequence of age and inadequate maintenance, air pollution, etc., or because of
damage to structural elements caused by accidents, wars, natural disasters, etc. In the
case of old residential buildings in urban nuclei, however, interventions in structural
system are often required because of rearrangement of living space during the
buildings’ renewal. In order to modemise buildings, windows and doors may be
enlarged, new openings may be made, staircases reconstructed, etc.
Structural strengthening of historical monuments is part of the complex,
multidisciplinary procedure of conservation and restoration of cultural monuments.
Whenever an intervention in the structural system of an historical building is required,
the principles of preservation and restoration of cultural monuments should be
followed, which severely limit the application of many possible technical
interventions. These principles usually do not permit the reconstruction or
replacement of structural elements, typical for the erection period of the building
under consideration, with elements made of modem construction materials.
Modernisation, however, requires interventions in those buildings, which, if the
rules of conservation and restoration were respected rigourously, would not have been
permitted. On the other hand, policy of protection of lives of residents, as well as
economic considerations regarding investments of large h d s needed for the renewal
of old urban nuclei, also require structural strengthening of historical buildings. It
would not be acceptable that people living in recently renewed, but unstrengthened
buildings, are exposed to a higher risk regarding loss of their property and lives, than
people who live in modem buildings designed by contemporary building codes.
Repair and Strengthening of Masonry Buildings 207

Therefore, in the specific case of strengthening of buildings in historical urban

and rural nuclei, a compromise between the requirements of structural engineering and
conservation of historic monuments should be achieved and special “building-
friendly” technologies should be developed based on the use of traditional building
materials.

10.2 Criteria for Repair and Strengthening

There are many criteria which should be considered when deciding upon the
repair and strengthening (retrofit or rehabilitation) of a building. Besides the
technical aspects, which include tests and structural analysis to assess the
structural resistance and to define efficient technical measures needed to improve
structural deficiencies, some general criteria, related with costs of intervention
and the importance of the building, availability of adequate technology and
skilled workmanship to implement the proposed measures, duration of the works
and problems of occupancy, etc., need to be taken into serious consideration
before the final decision is made. In the case of historical buildings, however, the
principles of preservation of the historical value of architectural heritage should
also be taken into account.
In seismic areas, the basic criterion for repair and strengthening is based on
the correlation of the expected seismic loads with the resistance of structural
system, i.e. on seismic resistance verification. If the assessed resistance of the
structure is not sufficient to resist the earthquake of expected intensity within
acceptable limits of damage, the structure needs seismic strengthening. Seismic
resistance analysis will define the causes of potential damage and indicate parts
of the structure that need to be strengthened.
As in the case of a new construction, the criteria for the determination of the
acceptable level of seismic risk for strengthening the existing buildings depend
on the economic potential of each country. Although the level of design seismic
loads, which should be taken into account for the redesign of existing buildings,
is usually the same as in the case of a new construction, Part 1-4 of EC 8
specifies that, if the code required values of design ground acceleration
immensely increase the anticipated total costs, or lead to unacceptable
architectural alterations (e.g. in the case of historic monuments), the values of
effective peak ground acceleration defined for a general case may be reduced for
redesign purposes.
208 Earthquake-Res istant Design of Masonry Buildings

With regard to the seismic resistance, the mechanisms of seismic behaviour

and causes of damage can be defined on the basis of earthquake-damage
observations and damage analysis. However, additional investigations to
establish the structural characteristics and material properties need to be carried out
in order to evaluate the seismic resistance of existing buildings. Input parameters
that are used in the calculations should be based on information about the quality
of existing materials, type and extent of the previous and present structural
damage, as well as earlier repair measures, if any. Mathematical models and
methods, however, should be calibrated taking into account the observed seismic
behaviour and experimental data.
When strengthening the building, elements of the existing structural system
are systematically strengthened. Sometimes, additional elements are constructed
uniformly distributed in plan and elevation of the building. These elements
improve the lateral resistance and ductility of the system. When strengthening
only individual elements, attention should be paid to prevent worsening the
distribution of seismic loads onto the structural elements. The torsional effects or
irregularities in vibration modes, which take place in the case of concentration of
stiffness and resistance at one part of the building in plan or elevation, may result
in the concentration of damage and subsequent poor seismic behaviour of the
“strengthened” building.
The repair and strengthening of a structure against earthquakes represent a
complex procedure: it is not enough to strengthen individual structural elements
or structural assemblages as, for example, in the case of strengthening the
structure for static gravity loads. In the case of seismic strengthening, however,
good performance of the whole structural system must be ensured. Therefore,
both the lateral resistance of the structure and its ductility and energy dissipation
capacity characteristics need to be verified.
Despite different structural materials and type of construction of existing
masonry buildings, their earthquake-damage can be classified in a uniform way.
Generally, damage is a result of inadequate structural layout and insufficient
load-bearing capacity of the walls (Fig. 10.1) and/or inadequate connection
between the walls (Fig. 10.3).
Accordingly, the technical measures for repair and strengthening of masonry
buildings are classified into (a) measures for strengthening of masonry walls and
(b) measures for tying the walls, and for anchoring and stiffening of floors. The
following technical criteria should be considered for successful strengthening:
Repair and Strengthening of Masonry Buildings 209

- The structural walls should be uniformly distributed in both orthogonal

directions of the building. They should be sufficient in number and strength
to resist the expected seismic loads.
- The walls should be adequately tied and connected. Rigid floor diaphragm
action should be provided, and the floors should be well anchored into the
walls to prevent out-of-plane vibration of the walls.

Figure 10.3.kllari, India, 1993: Separation of walls.

- The foundation system must be capable of transfering the increased ultimate

loads from the strengthened upper structure into the soil.
When implementing these measures to buildings, it is of relevant importance
that:
- They provide the required degree of seismic resistance. In this regard,
experimental verification of the effectiveness of the proposed technical
solution is necessary.
- The chosen technical solution is simple and economical to carry out.
- In the case of historical monuments, the proposed methods fulfil the basic
requirements of restoration and conservation of cultural monuments.
210 Earthauake-Resistant Design of Masonry Buildings

10.3 Verification of Seismic Resistance of Existing Masonry Buildings

The same philosophy as in the case of modem buildings is used for seismic
resistance verification of existing masonry buildings. However, mathematical
models which have been developed on the basis of structural characteristics of
modem masonry structures, should be modified to take into account the
irregularities of structural layout and lack of connection between walls and floors
in the case of non-engineered masonry construction. In the case of historic urban
buildings, specific models have also been developed based on the observed
earthquake damage and kinematic mechanism which caused that damage [S]. In
all cases, however, experiments are needed to verify the proposed mathematical
models for assessing the lateral resistance. Experiments are also needed to
determine the values of basic parameters, which define the lateral resistance and
design seismic loads: strength and deformability characteristics of existing
masonry, as well as dynamic characteristics and energy dissipation capacity of
masonry structural systems.
According to Annex K to Part 1-4 of EC 8 [2], the methods involving a non-
linear time domain dynamic analysis are not recommended because of the lack of
reliable models of the hysteretic in-plane and out-of-plane behaviour of walls. EC
8 recommends that the method, described in detail in Chapter 9, is applied for the
evaluation of existing masonry buildings. According to Part 1-4 of EC 8, seismic
vulnerability evaluation may also be used for undamaged low-rise masonry
buildings, as a preliminary approach to evaluate the need for seismic retrofitting.
In this case, however, the use of vulnerability index is recommended only where
the procedure has been calibrated to the observed damage to buildings in past
earthquakes.

f 0.3.I Dynamic Characteristics and behaviour factor

Because of the structural complexity and unknown mechanical characteristics

of masonry materials, it is not easy to evaluate the dynamic characteristics of
existing masonry buildings by mere calculation. In order to assess the values
needed to estimate the design seismic loads, ambient vibrations of typical
buildings have been measured and analysed [9-11]. Some characteristic results of
these studies are summarised in Table 10.1,
As can be seen, the tested buildings, although different in size, type, masonry
materials and period of construction, are all rigid structures. The values obtained
in the case of two historical stone-masonry buildings are similar-close to 5 Hz-
Repair and Strengthening of Masonry Buildings 21 1

although the buildings were quite different in plan and elevation. The values
obtained in the case of much higher, up to 13-storey pre-1963 residential tower-
blocks are also similar-close to 2 Hz-although the buildings have been built
with0 different materials. In all cases, however, the measured values of the first
natural frequency are within the limits of the flat part of the design response
spectra proposed by EC 8, a fact that simplifies the procedure: in the case of
normal structural configuration of masonry structures there is usually no need to
carry out modal analysis to assess the response spectra amplification values.

Table 10.1. First natural frequency of vibration of existing masonry buildings.

1 Description No. of Dimensions Height 1 st natural frequency

of building storeys in plan E-w N-S Torsion
(m)x (m) (m) 1 (I-IZ) (I-IZ) WZ)
Stone,
historical 2 29.3 x 13.0 7.1 3.4 4.2 5.8
Stone, 4 18.6 x 25.3 4.6 4.6
historical 13.0 3.8

Hollow 11 22.4 x 16.1 29.7 2.2 2.2 2.8

block, pre-
1963
Concrete- I
filled block, 13 18.6 x 17.9 37.7 1.8 2.2 2.6
pre- 1963

In order to assess the design seismic loads for seismic resistance verification,
elastic design loads are reduced by the behaviour factor, q, which depends on the
structural system under consideration. In the elastic range, at small amplitudes of
vibration, even existing masonry buildings vibrate as monolithic box. At large
amplitudes, however, when subjected to strong seismic motion, the shape and
mechanism of vibration depend on the interconnection of walls and type of
floors. Where timber joists are not anchored and the walls are not tied, the walls
separate along their vertical joints or intersections, the vibrations of the walls
become uncoupled and perimetral walls collapse due to out-of-plane forces.
The influence of tying the walls on the seismic behaviour of simple two-
storey stone- and brick-masonry model houses has been investigated
experimentally [12, 131. AS can be seen in Fig. 10.4, the potential seismic
212 Earthquake-Resistant Design of Masonry Buildings

resistance of a masonry house, expressed in terms of cumulative seismic input

energy, which the model buildings had resisted, in dependence on damage
occurred to the walls, is significantly reduced if the walls are not adequately tied
at floor levels.
As can be seen, twice the amount of input energy was needed to cause the
collapse of adequately tied buildings than of buildings without ties. It is therefore
clear that, when assessing design seismic loads for seismic resistance verification,
distinction should be made between existing buildings with adequately tied walls
and rigid floors and buildings with flexible wooden floors without ties. Whereas
behaviour factor q = 1.5, as proposed in EC 8 for plain masonry structures, may
be used in the first case, it is proposed that no reduction of elastic seismic loads,
i.e. q = 1.0 be taken into account in the other.

2.5

0
0.25 0.5 0.75 1
Damage index Id

Figure 10.4. Damage to models in dependence on cumulative input energy. As , Ab - stone- and brick-
masonry house without ties; 8,,Bb - stone- and brick-masonry house with ties.

10.3.2 Material properties

With regard to masonry materials and structure of masonry walls, a great

variety of different kinds of traditional masonry construction can be found.
Brick- and stone-masonry, sometimes replaced by mixed stone-and-brick-
masonry, have been used as the main construction materials for centuries. Stone-
masonry buildings, which are sometimes four to five storeys high, are generally
made of irregularly sized, two-layered, rubble stone masonry with an infill
Repair and Strengthening of Masonry Buildings 213

consisting of smaller pieces of stone bound together with lime mortar. Cut stone,
or partly cut stone is used in the case of public and monumental buildings.
Locally available material, such as limestone and slate, can be found. The load-
bearing layers of stone-masonry walls are sometimes connected at regular
intervals with connecting stones. Regular cut stones are used for better
connections of walls at the comers and wall intersections. Also, the use of iron
ties to connect the walls horizontally at floor levels is evident.
Clay brick laid in mortar of poor to good quality prevails in many
earthquake-prone areas of Central Europe and the Mediterranean region. Old,
Austrian format bricks (30 x 15 x 6.5 cm) and normal format bricks (25 x 12 x 6.5 cm)
have been used before and after World War I, respectively. Perforated bricks and
hollow concrete blocks have been used after World War 11. Before the adoption
of seismic codes, the height of many post-World War I1 plain masonry buildings
exceeded ten storeys.

Figure 10.5. In-situ lateral resistance Figure 10.6.Laboratory lateral resistance test of
test of stone-masonry wall [20]. brick-masonry wall, cut from the building 1191.
214 Earthauake-Resistant Design of Masonry Buildings

Since masonry is a non-elastic, non-homogeneous and unisotropic material, it

is-as a rule-not possible to determine the mechanical properties of historic
masonry walls by testing their constituent materials in the laboratory. It is also
difficult to reproduce the existing masonry walls in the laboratory, eventhough
very thorough chemical and mechanical tests of the mortar, brick and/or stone
may have been carried out. The only reliable method of determining the Ioad-
carrying capacity of existing old masonry walls involves the carrying out of tests
in situ (Fig. 10.5), or cutting out specimens from these walls and testing them in
the laboratory (Fig. 10.6). Sometimes, flat-jacks are used in situ, either to
determine the compressive strength of the masonry or simply to evaluate the
stress state in the wall.
Some indicative data regarding mechanical properties of existing masonry
materials are given in Table 10.2 [ 14-2 13.

Table 10.2. Mechanical properties of existing masonry (characteristic values).

I Stone-masonry I Brick-masonry I
Compressive strength,Ji( (ma) 0.3 - 0.9 1.5 - 10.0
Tensile strength,f~ ( m a ) 0.08 - 0.21 0.10 - 0.70
I Modulus of elasticity, E (MPa) I -
200 1000 I -
1500 3800 I
I Shear modulus, G ( m a ) I 70 - 90 I 60-165 I

10.3.3 Seismic vulnerability evaluation

If the seismic resistance of a large group of existing masonry buildings needs to

be evaluated, e.g. for planning purposes, a seismic vulnerability evaluation approach
represents an adequate method, especially when the method has already been
calibrated to the observed damage of the same type of buildings during past
earthquakes.
Different methodologies are used to assess the seismic vulnerability of existing
masonry buildings (for example, see [22-281). A simple method which might be
applied to typical Central European urban masonry construction [26], where
earthquakes of degree VIII-IX by MSK seismic intensity scale can be expected, takes
into account the following basic parameters:
Type and quality of masonry (weighing factor 1.O),
Repair and Strengthening of Masonry Buildings 215

Structural wall area (weighing factor LO),

Distribution of structural walls (weighmg factor 1.O),
Connection of walls (weighing factor 1S), and
Other (weighing factor 0.5).
Parameters are rated between one and five: the closer the parameter reflects the
code requirements, the lower the rating (one means that the parameter is in full
agreement with code requirements). Consequently, the sum of weighted values of all
parameters for individual buildings varies between Ps= 5 and Ps= 25 (Ps= 5 means
that structural characteristics of the buildings under consideration are in full agreement
with code requirement).

sRc t
o Residential buildings
x Public buildings

0.16

0.12

0.08

0.04

0
15 17 19 2'1 23 V,

Figure 10.7. Correlation between the results of seismic vulnerability analysis

and seismic resistance verification (after [27]).

The seismic resistance of masonry structures depends on the number of storeys, n,

and storey height, H. The following values of parameter Ph are taken into account for:
216 Earthq uake-Resis rant Design of Masonry Buildings

n C 2, H < 3.5 M:P h = o ,

n = 3, H > 3.5 m or n = 4, H < 3.5 m: Ph = 2,
n = 4, H > 3.5 m or n = 5, H < 3.5 m: Ph = 4,
n = 5 , H > 3.5 m or n = 6, H < 3.5 m: Ph = 6.
The seismicity of micro-location of the building, expressed in terms of design
ground acceleration ag is defined by the parameter Pi:
ag=O.l5: Pi=O,
ag = 0.20: Pi = 3,
0 ~g = 0.25: Pi = 9,
ag= 0.30: Pi = 15.

The seismic vulnerability index of individual buildings is then calculated by:

(10.1)

As can be seen, the frnal index is corrected by factor Fs, which takes into account
the structural system. According to the proposed methodology, Fs = 1.0 for plain and
Fs= 0.75 for confined masonry buildings. The method has not yet been calibrated to
earthquake-damaged buildings. However, as can be seen in Fig. 10.7, where the
assessed values are correlated with the values obtained by detailed seismic resistance
verification of a group of existing masonry buildings in the City of Ljubljana [27],
reliable estimates can be obtained by this, although simple, method.
Recently, expert systems, based on large databases of the values of many
parameters influencing the seismic resistance of buildings, have been developed [283.
In most cases, non-parametric models and neural networks are used to model the
seismic vulnerability.

10.4 Methods of Strengthening of Masonry Walls

Different technical measures have been developed for repairing and

strengthening masonry buildings. Some of them are based only on the analysis of
earthquake damage and engineering judgement, and have never been actually
verified. Others, however, are based on both earthquake damage observation and
experimental investigations, and have been verified either in the laboratory or by
a real earthquake. Different methodologies are available for different types of
masonry walls. The type and quality of masonry materials and a building's
structural layout are the main criteria to be considered when choosing the
Repair and Strengthening of Masonry Buildings 217

adequate method of strengthening. However, the choice of the most suitable

technical solution depends also on the required degree of improving the
resistance of the wall. This is indicated by seismic resistance verification and
redesign of the existing structure.
While efficient interventions in stone-masonry are more or less limited to
injecting the grout into the void parts of the walls, various possibilities are
available for the repair and strengthening of brick- and block-masonry walls. The
procedures can be classified into the following main groups:
Repair of cracks.
Repointing the joints with cement mortar.
Application of reinforced cement coating on one or both sides of the walls.
Grouting with cement, modified cement, or epoxy grout.
Prestressing the walls in vertical or horizontal direction.
Reconstruction of the most damaged parts of the walls.

10.4.I Repair of cracks

According to recommendations given in Annex K to Part 1-4 of EC 8, cracks

may be sealed with mortar if the crack width is large (about 10 mm), and the
thickness of the wall is relatively small. Otherwise, the cracks should be injected
with cement grout which contains admixtures against shrinkage. If the cracks are
fine, the use of epoxy is recommended, but if the cracks are larger than 10 m,
the damaged area should be reconstructed using stitching bricks.
In the case of brick and/or block masonry, the cracks can be injected either
with epoxy, if their width does not exceed 1.0 mm, or with cement grout, if their
width varies between 0.3-3.0 mm. Taking into account the effectiveness of
intervention and the cost of materials, the use of cement grouts has been
preferred to epoxy injections for the repair of earthquake-damaged masonry
walls. However, experience has shown that the price of epoxy grout can be
reduced and application range increased by adding inert filler materials, such as
fine-grained quartz sand, to epoxy resins.
In order to inject the cracks with cement grout, the damaged part of plaster is
removed from the wall, and the injection nozzles are built into the holes, drilled
into the wall along the crack at 0.3-0.6 m intervals. The surface of the wall is
then cleaned, and the cracks are sealed and the nozzles fixed to the wall with fast
binding mortar. Before injecting, cracks are washed with water. Usually, the mix
consisting of 90% of Portland cement and 10% of Pozzolana, is used. In the case
21 8 Earthquake-Resistant Design of Masonry Buildings

of grouting of 5.0-10.0 mm wide cracks, fine sand can be added to the mix for
better results. When injecting the grout, the pressure in the container is slowly
increased up to 3 bar and is kept constant until the wall absorbs the grout. After
that, the pressure is increased to 4 bar and kept constant for 5-10 minutes in
order to densify the mix and to drain excess water.
The effect of grouting the cracks on the lateral resistance of brick- and block-
masonry walls has been tested experimentally [29]. It has been found that by
injecting the cracks with cement or epoxy grout, the original load-bearing
capacity of the walls is recovered (sometimes even improved, as in the case of
poor quality masonry), but the rigidity in most cases is not. Therefore, the
grouting of cracks in brick- or block-masonry walls represents a typical method
of structural repair. Typical results are given in Table 10.3.

Table 10.3. Effect of injecting the cracks on tensile strength and shear modulus (af€er1291).

Original I Grouted cracks

Masonry unit

Brick (B 20) 0.5 0.07 - 0.1 1 -

Brick (B 20) 3 .O 0.20 - 0.25 -
Ceramic block (B 20) 4.8 0.15 360 0.26 250
Ceramic block (B20) 6.1 0.19 240 0.18 380
I
~ Light concrete block (B 7.5) 2.9 0.19 380 0.28 380
Fly-ash block (B 15) 1.3 0.14 370 0.14 230
Fly-ash block (B 15) 1.3 0.16 480 0.22 490

In the case of excessively wide cracks in brick and/or block masonry walls,
the damaged area is reconstructed as recommended in Part 1-4 of EC 8, or cracks
are sealed with cement mortar or grouted, but in addition, the area around the
crack is coated with reinforced-cement coating (Fig. 10.8), as described in
subsequent sections. In such a way, good connection between the separated parts
of the wall is provided and the integrity of resisting elements ensured.
Repair and Strengthening of Masonry Buildings 219

0.3 m 0.10-0.15 m 0.3 m 0.3 m 0.10-0.15 m 0.3 m

m t
= 0.

. . . , , . ,,,,
~

Figure 10.8.Repairing of heavily cracked brick-masonry wall with r.c. coating (after [S]).

Heavily damaged part Rebuilt with original material

Figure 10.9. Reconstruction of the central part of a heavily cracked stone-masonry wall (after [S I).

In the case of stone-masonry walls, however, reconstruction of parts or

complete walls is usually necessary where severe cracks occur in the middle part
of the wall, or if that part of the wall has been disintegrated during the
earthquake. In the case of reconstruction, original materials, mortar and stone,
which are compatible with the materials of the remaining part of the wall, should
be used (Fig. 10.9). If, for example, the part of the masonry that is removed is
220 Earthquake-Resistant Design of Masonry Buildings

replaced by concrete, the rigid concrete element represents a potential cause of

serious damage to neighbouring masonry during the future seismic events.
However, care should be taken to improve the integrity of the masonry wall by
using connecting (stitching) stones, as well as to provide good connection
between the existing and new masonry.

10.4.2 Repointing

Where the bed-joints are relatively level, the mortar is poor and the units are
good, the resistance of a wall to vertical and lateral loading can be considerably
improved by replacing a part of the existing mortar with mortar of significantly
better quality, e.g. cement mortar.

ti3 2i3 t tt3 ti3 tf3 ti3 ti3 ti3

w -i1 L 1
1 tt-t-t

Figure 10.10. Repointing of a brick-masonry wall.

For that purpose, existing mortar up to 1/3 of the wall’s thickness is removed
from the joints on one or both sides of the wall using clamps or electric tools. In
order to prevent vertical instability, the wall is first repointed on one and then the
other side. Care should be taken and numerical verification of stability (buckling)
is needed to determine the allowable depth of removal of mortar. Once the
existing mortar has been removed, the surface of the wall is cleaned and
moistened with water jet. Steel reinforcement is sometimes placed in the bed-
joints to improve the ductility and energy dissipation capacity of the structure. In
Repair and Strengthening of Masonry Buildings 22 1

the case of normal bed-joint thickness (1.0-1.5 cm), 6-mm bars are placed at
0.3-0.5 m intervals along the height of the wall, adequately anchored at the ends
of the wall, e.g. into the vertical tie-columns. The joints are then repointed with
cement mortar. After sufficient strength of the mortar is attained, the procedure
of repointing is repeated on the other side of the wall (Fig. 10.10).

10.4.3Reinforced-cement coating

In the case of seriously damaged brick- and block-masonry walls, or where

there is a need to strengthen the existing structure, the application of reinforced-
cement coating (jacket) on one or both sides of the wall represents a logical way
of improving the lateral resistance and energy dissipation capacity of the system.
Since the method is easy to apply and very efficient, it has been widely used for
strengthening existing masonry walls all over the world. The possibilities of
using ferro-cement and, most recently, carbon fibre coating instead of ordinary
reinforcing steel, have been experimentally studied.
If reinforcing steel is used, plaster is first removed from the wall. Mortar is
removed from the joints between the bricks or blocks, 10-15 mm deep, and the
cracks in the wall are grouted. The wall surface is cleaned, moistened with water
and spattered with cement milk. The first layer of cement coating, i.e. 10-15 mm
thick cement mortar layer (compressive strength 20-30 MPa), is applied. The
reinforcing mesh is then placed on both sides of the wall (4-6-mm diameter bars
at 100-150 mm intervals in the vertical and horizontal directions), connected
together with steel anchors (6-mm diameter bars, placed in the pre-drilled holes
and cemented or epoxied, 4-6 pieces per m2 of the wall's surface). After the
reinforcing mesh is connected to the anchors, the second layer of cement coating
is applied, so that the total thickness of coating does not exceed 30 mm
(Fig. 10.1 1).
In a similar way, feno-cement, and carbon- or polyester-fibre coating can be
applied, Similar results as in the case of conventional r.c. coating have been
reported with regard to the improvement in lateral resistance of brick masonry
walls, in the case where adequate connection between coating and masonry has
been achieved [3 1, 321. For better results, reinforcement can be placed diagonally
in order to follow the possible direction of cracks. Although both technologies
have several advantages in comparison with conventional methods, they have not
yet been widely applied because of their relatively high costs.
222 Earthquake-Resistant Design of Masonry Buildings

The effect of strengthening the brick- and block-masonry walls by coating

has been experimentally investigated in the laboratory and in situ [18, 29, 31,
321. The experiments, which have all shown that r.c. coating improves the lateral

-
resistance, have also indicated the importance of adequate anchoring of coating
reinforcement to the existing masonry. If the connection was not adequate to
prevent splitting, the coating separated from the wall at the occurrence of cracks
in the masonry wall and buckled.

z 0.5 m 10-15 mm

Figure 10.11. Application of r.c. coating to brick-masonry walls (after [5]).

As shown by experiments, by applying r.c. coating, the shear resistance of the

tested walls was significantly improved (Table 10.4). Consequently, in the case
of walls with geometry aspect (heightlwidth) ratio greater than 1.0, the failure
mechanism changed from shear to bending. As indicated by experiments, the
improvement in the lateral resistance is inversely proportional to the quality of
the original walls. It is significant in the case of poor-quality masonry, but not so
in the case of good-quality walls.
Other systems of application of cement coating can also be used (e.g. shot-
creting), otherwise the coating is simply applied by concreting. In the latter case,
however, the thickness of coating is increased to 80-100 nun. If concrete is
poured into forms, reinforcing mesh consisting of 8-10 mm diameter bars at 250
mm intervals, is used. Concrete is used for stone-masonry walls, in the case of
Repair and Strengthening of Masonry Buildings 223

which 30 mm thick cement-plaster layers would not be enough to strengthen a

two-leaf wall which is more than 50 cm thick. Typical examples of strengthening
stone-masonry walls with reinforced-concrete coating are shown in Figs. 10.12
and 10.13.

Table 10.4. Effect of coating on the lateral resistance of a wall.

Type of Resistance I
Grade
Brick I3 20 I M 0.4
Grade cement
I Steel 34 1 118
II Multiplier

3.5
Brick B 10 I M 0.3 I Steel 47 1 167 I 3.6
C. blockB 7.5 I M 5 I Steel 128 I 167 1 1.3
Brick B 20 I M 7.2 I Ferro-cem. 276 I 693 I 2.5"
Brick B 15 I - I Carbon 299 I 426 I 1.4

Note: * Diagonal compression tests [31].

Existing Cage reinforcement detail

wall

10-20 cm = 5 cm

Cage
7

R.c.

Figure 10.12.Connecting the r.c. coating to stone-masonry wall with shear connectors (after [5]).

In the case of stone-masonry, plaster and loose stones are first removed from
the wall and any cracks are ealed or injected with cement mortar or grout,
 Next Page

224 Earthquake-Resistant Design of Masonry Buildings

respectively. Then, holes for steel anchors-6-8-mm diameter reinforcing steel

bars placed on either side of the wall-that will connect the coating
reinforcement-are drilled through the wall at 50-60 cm intervals. Sometimes,
stones are removed fiom the wall at regular intervals and a reinforcement cage is
placed in the chase or the void created by removing the stone. By filling the
chase with concrete, a shear connector is formed which efficiently transfers the
forces from the new coating to the wall. This system is used where only one-side
coating is provided and where heavy and thick stone-masonry makes the drilling
of holes for the anchors difficult.

Existing

Figure 10.13. Application of r.c. coating on one side of the wall with new r.c. slab (after [S]).

Before concreting, dust is removed and the surface of the wall is washed with
water jet. Depending on the thickness of the coating, the concrete is either
shotcreted (80 mm) or poured in the form (at least 100 mm). In the case of
shotcrete, the reinforcing mesh (8-10 mm diameter bars at 250 mm intervals) is
placed on the finished first layer of shotcrete, the anchors are tied to the bars or
welded mesh fabric, and the second layer of the shotcrete is applied. If the
concrete is poured in, the reinforcing mesh is fixed to the position before
concreting.
As the test results and subsequent analyses indicate, it is not possible to
estimate the lateral resistance of a coated wall panel by simple calculation. As a
rule, the calculated values, obtained on the basis of simple combination of the
theoretical lateral resistance of the existing wall and reinforced-cement coating,
do not correlate with experimentally obtained results. Therefore, practical
 SYMBOLS

= area of horizontal cross-section of r.c. column

= equivalent area of composite section of infilled frame
= design value of seismic action
= area of horizontal cross-section of infill
= area of horizontal reinforcement
= area of vertical reinforcement
= area of horizontal cross-section of wall
= stiffness degradation parameter
= depth of stress block
= peak ground acceleration
= width of a building
= design base shear coefficient
= stiffness degradation parameter;
shear stress distribution factor
Ccff = coefficient of effectiveness of horizontal reinforcement
CI = interaction coefficient
Ccr = Hcr/Hmax ratio
CR = masonry infill capacity reduction factor
crh = horizontal reinforcement capacity reduction factor
crv = vertical reinforcement capacity reduction factor
Csd = strength degradation factor
Dmin = minimum tie diameter
d = displacement; effective depth of wall
dcr = displacement at crack limit
de = displacement at elastic limit
dmax = maximum displacement
257
258 Earthquake-Resistant Design of Masonry Buildings

dHlnax = displacement at maximum resistance

drv = diameter of vertical reinforcing bar
E = energy
E = modulus of elasticity;
modulus of elasticity of masonry
Ec = modulus of elasticity of concrete
Ed = design value of action effect
e = eccentricity
eU = ultimate eccentricity
e, = eccentricity in x-direction
eY = eccentricity in y-direction
F = seismic force, acting in a storey
Fi = seismic force, acting on the i-th wall in a storey
Fa = seismic force, acting on element
Fbd = design base shear force
Fid = design seismic force, acting at the i-th storey
Fr = compression force in compressed reinforcement
Fw = resultant of compressive stresses
f = compressive strength of masonry
fb = normalised compressive strength of masonry units
fbok = characteristic anchorage bond strength
fck = characteristic compressive strength of concrete
fcvk = characteristic shear strength of concrete
Ji( = characteristic compressive strength of masonry
fm = compressive strength of mortar
fv = shear strength of masonry
fvk = characteristic shear strength of masonry
fv0 = shear strength of masonry at zero compressive stress
fvko = initial characteristic shear strength of masonry at zero
compressive stress
h = tensile strength of masonry;
tensile strength of steel
hk = characteristic tensile strength of masonry;
characteristic tensile strength of steel
fx = flexural strength of masonry
fx 1 = flexural strength of masonry parallel to bed joints
fx2 = flexural strength of masonry perpendicular to bed joints
Symbols 259

= yield strength of steel

= shear modulus;
shear modulus of masonry
= shear modulus of concrete
= characteristic value of permanent actionj
= acceleration of gravity (9.8 1 ms-2>
= height of building;
lateral load
= resistance at crack limit
= resistance at maximum displacement
= resistance of vertical steel due to dowel action
= design resistance of vertical steel due to dowel action
= elastic load
= flexural resistance of r.c. column
= flexural resistance of reinforced masonry wall
= design flexural resistance of reinforced masonry wall
= flexural resistance of plain masonry wall
= design flexural resistance of plain masonry wall
= maximum resistance
= maximum resistance of horizontally reinforced wall
= tension in horizontal steel
= tension capacity of horizontal steel
= shear resistance of confined masonry wall;
shear resistance of r.c. column
= design shear resistance of r.c. coated masonry wall
= design shear resistance of prestressed masonry wall
= design shear resistance of a reinforced masonry wall
= contribution of horizontal steel
= resistance of wall to sliding
= shear resistance of plain masonry wall
= design shear resistance of plain masonry wall
= resisting storey shear
= ultimate lateral load
= resistance of confined masonry panel before the separation of infill
= ultimate resistance of confined masonry panel
= ultimate design load of plain masonry wall
= ultimate resistance of frame
260 Earthquake- Res istant Design of Masonry Buildings

= ultimate resistance of critical segment of a building

= height of wall
= height of frame
= height of compressed strut
= intensity of earthquake
= moment of inertia of r.c. column’s section
= damage index
= equivalent moment of inertia of composite section
= moment of inertia of masonry infill’s section
= moment of inertia of masonry wall’s section
= strength correction factor
= stiffness of wall
= effective stiffness of wall
= effective stiffness of equivalent r.c. coated wall
= effective stiffness of masonry wall
= effective stiffness of r.c. coating
= effective stiffness of the i-th wall
= stiffness of wall at maximum resistance
= stiffness of wall at ultimate resistance
= stiffnesses of wall in x- and y-direction
= exponents influencing the shape of elastic response spectrum
= exponents influencing the shape of design response spectrum
= length of building
= anchorage length
= span of frame
= length of compressed strut
= magnitude of earthquake
= bending moment of fixed-ended i-th wall
= bending moment at the bottom of the i-th wall
= bending moment at the top of the i-th wall
= design flexural capacity of wall’s section
= ultimate flexural capacity of wall’s section
= flexural capacity of r.c. column’s section
= prestressing effect multiplier
= multiplier for horizontal prestressing
= multiplier for vertical prestressing
= mass at the i-th storey level
Symbols 261

= vertical load
= number of stories;
number of events
= probability of occurrence
= characteristic value of prestressing action
= shear force
= shear force, carried by the i-th wall
= characteristic value of variable action i
= behaviour factor
= behaviour factor associated with structural element
= return period
= design resistance capacity
= soil parameter
= seismic coefficient associated with structural element
= ordinate of design response spectrum
= ordinate of elastic response spectrum
= seismic resistance coefficient
= design seismic resistance coefficient
= spacing of horizontal reinforcement
= displacements of masses mi and mj in the first mode shape
= period of vibration;
time
= period of vibration of structural element
= first vibration mode period of bare r.c. frame without infills
= first vibration mode period of structure with infills
= tension force in reinforcing steel
= thi c h e s s
= weight of building (gravity loads)
= weight of structural element
= weights of masses mi and mj
= co-ordinates of area centre of the i-th wall
= co-ordinates of storey mass centre
= co-ordinates of storey stiffness centre
= section modulus; height of centre of structural element above the
base
= heights of masses mi and mj above the level of seismic loads
262 Ear fhquake-Resistant Design of Masonry Buildings

= design ground acceleratiodacceleration of gravity ratio;

stiffness degradation parameter
= coefficient defining position of moment’s inflection point
= stiffness degradation parameter
= maximum normalised spectral value
= importance factor, associated with structural element
= importance factor
= partial safety factor for material properties
= partial safety factor for masonry
= global safety factor
= partial safety factor for steel
= shape factor; displacement
= characteristic value of unit elongation at maximum stress
= damping correction coefficient
= shear coefficient
= ultimate ductility factor
= horizontal reinforcement ratio
= principal compression stress
= design compression stress
= average compression stress
= compression stress due to horizontal prestressing
= compression stress due to vertical prestressing
= principal tension stress
= average shear stress
= average shear stress at maximum resistance
= diameter of reinforcing steel
= principal compression stress angle
= storey drift rotation angle at elastic limit
= principal tension stress angle
= ultimate storey drift rotation angle
= storey load participation factor
= seismic action combination coefficient
= action combination coefficient
SUBJECT INDEX

Index Terms Links

Acceleration response spectrum:

design 101–102
elastic 22–24
Acceptable seismic risk 207
Allowable stress method 179
Arch-beam mechanism 130

Balconies 91
Base shear coefficient 103 199
Behaviour factor:
masonry structures 105–106 212
non-structural elements 158
Bond-beams:
function 88
dimensions 88–89
installation of new 243–244
reinforcement 89
Bond strength 42
Brick (see Masonry units)
Bulged stone-masonry wall 233

Cavity wall 54
Compressive strength:
concrete infill 43
masonry 46–47
Compression test of masonry 46
Concrete infill 42–43
Strengthening of walls by
confinement:
installation of tie-columns 247–248
installation of vertical ties 248–249

This page has been reformatted by Knovel to provide easier navigation.

Index Terms Links

Damage index 122

Design base shear:
force 102
coefficient 102 199
distribution 104
Design ductility capacity 198–200
Design seismic action 99
Design resistance capacity 96–97 198
Design value of action effect 96
Distribution of action effects in
shear walls 186–188
Dowel action of vertical steel 132–133
Ductility factor 45
Ductility demand 199–200
Ductility of masonry 124
Dynamic amplification factor 101
Dynamic characteristics of existing
masonry buildings 210–211

Earthquakes:
causes of 6–9
focus 9
intensity 12–17
magnitude 11–12
probability of occurrence 17–18
return period 18
Eccentricity 192–193
Effect of grouting:
on lateral resistance 228–231
on shear modulus 228
Effective flange width 152
Effective depth of wall 132
Effectiveness of horizontal bed joint
reinforcement 135–136
Epoxy grout 217
Equivalent compressive stress
block 144
Equivalent diagonal strut of
masonry infill 175–176

This page has been reformatted by Knovel to provide easier navigation.

Index Terms Links

Eurocode 6:
masonry materials 36–54
construction systems 54–66
Eurocode 8:
elastic response spectrum 22–23
design response spectrum 101–102
design seismic action 99–108
repair and strengthening 205–209
safety verification 96–99
simple buildings,
subsoil classes 23–24
Experimental simulation of seismic
behaviour:
masonry walls 112–115 127 133–134 138
142 146 213
masonry buildings 189 196 238

Failure:
brittle 125 127
ductile 127
Failure mode 110
Flanged sections: 151–154
Flexural resistance:
unreinforced masonry 141–145
reinforced masonry 146–149
confined masonry 149–150
Flexural strength of masonry 51
Floors:
monolithic r.c. slab 86
prefabricated slab 87
r.c. topping 87
wooden floor 87–88
Foundation soil:
failure 249
influence on ground motion 20

Ground acceleration:
effective 24
design 103
peak 19

This page has been reformatted by Knovel to provide easier navigation.

Index Terms Links

Ground acceleration: (Cont.)

time history 20 100
Grouting:
composition of grout 217 229
masonry-friendly 231
procedure 226–227
strength of grout mix 229–230
Grouted cavity wall 54

Hollow clay block (see Masonry units)

Hollow concrete block (see Masonry units)
Horizontal reinforcement 62–65 127–131
Horizontal reinforcement capacity
reduction factor 137
Hysteretic behaviour 111 115

Idealisation of hysteresis envelope:

bilinear 117 123
trilinear 118–119 123
Idealisation of seismic behaviour:
assumptions 115–116
Importance factor 97
Initial shear strength test of
masonry 48
Injecting the cracks with cement
grout:
procedure 217–218
effect 218
In-situ tests 213
Installation of new r.c. slab:
bearing 241
anchoring 241
Interaction between masonry and
confinement 140–141

Lateral restraint 113

Lateral stiffness 120
Lightweight unit (see Masonry units)

This page has been reformatted by Knovel to provide easier navigation.

Index Terms Links

Limit states:
cracks limit 116 195
maximum resistance 116 195
serviceability limit state 96
ultimate state 96 116 195
Linear elastic analysis 101
Lintel 90
Lateral loading history 114–115

Masonry bond 56–57

Masonry construction systems:
unreinforced masonry 35–36 55–58
confined masonry 35–36 58–61
reinforced masonry 35–36 62–69
Masonry building:
configuration 72–76
distribution of walls 78
earthquake damage 31
number of stories 76–77
seismic performance 28–32
simple building 79–80
structural typology 25–27
Masonry infill:
influence on vibration period
and design seismic loads 169–170
influence of irregularities on
design seismic load 110
Masonry infilled r.c. frames
calculation of seismic resistance 171–176
concept of construction 164
seismic behaviour 164–168
failure mechanism 168
Masonry units:
aerated concrete unit 37–39
adobe 37
calcium silicate unit 37–39
clay unit 37–39
concrete unit 37–39
minimum compressive strength 39–39
minimum thickness of shells 38

This page has been reformatted by Knovel to provide easier navigation.

Index Terms Links

Masonry units: (Cont.)

natural stone unit 37
shape factor 40
Masonry veneer 82
Mechanism of action of
reinforcement 66–69
Modelling of mechanism of
reinforcement action:
at shear failure 129–137
at flexural failure 146–149
Modelling of seismic behaviour of
masonry walls 110–112
Modelling of seismic behaviour of
masonry buildings 179 182
Modulus of elasticity of masonry 53
Mortar:
composition 41
general purpose mortar 41
lightweight mortar 41
thin layer mortar 41
Mortar joint 42
Multiplier for r.c. coating 225
Multiplier for prestressing 232

Non-linear dynamic analysis 101

Non-linear static analysis 182
Non-structural element 80–82 158
Normalised compressive strength of
masonry unit 40

Out-of-plane behaviour:
seismic action 155–156
flexural resistance capacity 156
Overstrength ratio 105

Partial safety factor:

material properties 98
masonry in seismic situation 98

This page has been reformatted by Knovel to provide easier navigation.

Index Terms Links

Partition wall 81
Percentage of reinforcement:
horizontal 64 134–136
vertical 65
Plastic hinge 123
Position of moment inflection point 121 187
Prestressing 231–232
Principal tensile stress 125–126

Reconstruction of walls:
brick-masonry 232–233
stone-masonry 219 233
Reinforced-cement coating:
brick masonry 219 221–222
effect on resistance 223
stone-masonry 222–224
corners and wall intersections 246–247
Reinforcing steel:
anchorage 65
anchorage length 65–66
elongation ductility 43
tensile strength 43–44
yield strength 43–44
Repair:
definition of 203 208
of cracks 217–220
Repointing 220–221
Resistance of r.c. coated masonry
wall 225
Rocking 143
Roof 92

Sectional forces at flexural failure:

plain masonry wall 144–145
reinforced masonry wall 147–148
confined masonry wall 149–150
Seismic action 5 100
Seismic fault 7–8

This page has been reformatted by Knovel to provide easier navigation.

Index Terms Links

Seismic intensity scales:

EMS scale 12–17
JMA scale 12
MCS scale 12
MM scale 12 14
MSK scale 12 14
Seismic resistance verification:
masonry wall 96 116
masonry structure 96 179–181 197–200
masonry infilled r.c. frames 176
Seismic vulnerability:
evaluation 214–216
index 216
parameters 214–215
Seismic waves:
types of 9–10
propagation 10
Separation of masonry infill 172
Shear modulus of masonry 53
Shear resistance:
unreinforced masonry 124–127
reinforced masonry 127–137
confined masonry 137–141
Shear strength:
concrete infill 43
masonry 48–51
initial shear strength of masonry 48–51
Simple building 79–80
Single-leaf wall 54
Sliding shear failure 150–151
Splicing of reinforcing steel 66
Steel ties:
structural details 235–237
installation and position 237
design 239–240
Siffening of existing floors:
bracing with metal truss 242
diagonal ties 241
Stiffness of masonry walls:
definition of 120–121
effective elastic 121
in non-linear range 121–123
Stiffness of masonry infilled frame 174–175
This page has been reformatted by Knovel to provide easier navigation.
Index Terms Links

Stiffness degradation parameter 122

Stiffness of r.c. coated masonry
wall 225
Storey mass centre 192
Storey resistance envelope:
definition of 189–190
assumptions of calculation 190
calculation 193–196
design resistance envelope 196
Storey stiffness centre 192
Strength degradation factor 119
Strengthening:
corner zones 244–247
foundations 249–251
definition of 203 208
masonry walls 216–217 247–249
non-structural elements 251
Stress-strain diagram:
masonry 52
reinforcing steel 44
Structural models of masonry
shear-walls:
cantilever walls 183–184
coupled shear wall with weak
piers 184
coupled shear wall with weak
spandrels 185

Tensile strength:
masonry 49–51
reinforcing steel 43–44
Tension capacity of horizontal steel 134
Tie-column:
dimensions 60
distribution 59
reinforcement 60
Truss mechanism 130
Tying of walls:
mechanism 234
method 235–238
effect 212 238

This page has been reformatted by Knovel to provide easier navigation.

Index Terms Links

Ultimate design seismic resistance

coefficient 199
Ultimate ductility factor 118
Ultimate limit state 96 195

Vertical reinforcement 62–65

Wall opening 78–79

Yield strength of reinforcing

steel 43–44

This page has been reformatted by Knovel to provide easier navigation.