Reading PDF

SEISMIC RETROFITTING TECHNIQUES FOR
EXISTING UNREINFORCED MASONRY

STRUCTURES
By
SHIH-WEI CHUANG
M.E., Cornell University, U.S.A, 1995
Ph.D. Thesis
A thesis submitted to the School of Geosciences, Minerals
and Civil Engineering, University of South Australia in
partial fulfilment of the requirements for the degree of
Doctor of Philosophy
OF
MAY 2004 iA
I •)
LiBRARY
Preliminaries
STATEMENT OF ORIGINAL AUTHORSHIP
This thesis presents theoretical, numerical, and experimental work performed by the
author. All references to, and use, works by other researchers are fully acknowledged
throughout the text. The remaining work described herein, to the best of my knowledge
and belief, is original.
The work contained in this thesis has not been previously submitted, either in part or in
whole, for a degree at this or any other university.
Signed:
Date:
U
Preliminaries
ACKNOWLEDGEMENT
The work presented in this thesis would not have been possible without contribution
of many people. I would like to express my appreciation and thanks to my supervisor
Dr. Yan Zhuge for her guidance, patience, and enthusiastic support during the course
of this study. I would also like to express my appreciation to my associate supervisor
Mr. Peter McBean for his invaluable advice and continuous encouragement
throughout this study.
Financial supports for this study were provided under Department for Administrative
and Information Services at South Australia Government and University of South
Australia presidential scholarship. The supports are gratefully acknowledged.
The testing programme was time consuming and presented many technical
difficulties. Special thanks extended to Mr Ray Yerbury, Mr. Colin Sharp and Mr
Gerald Lewis from University of South Australia's team of technical staff. Their
advice and help during the setting up and testing was vital to the success of the
programme.
I would also like to thank my friends and all my colleagues at the School of
Geosciences, Minerals and Civil Engineering, for their friendship and support.
ill
Preliminaries
ABSTRACT
In the last decade, Australian civil engineers have not paid enough attention to
earthquake resistant design of structures. It was commonly believed that the
earthquake risk was insignificant thus earthquake resistant design was not considered
for most buildings. Earthquake resistant design of structures has become an important
issue in Australia following the Newcastle earthquake in 1989.
Masonry is one of the most commonly used materials throughout Australia for the
construction of low rise buildings. Even though the history of past earthquakes has
shown that masonry buildings suffered the maximum damage and accounted for the
maximum loss of life, they continue to be popular. It was evident from the Newcastle
earthquake that most unreinforced masonry structures were seriously damaged.
Therefore it is important to retrofit and strengthen existing masonry structures to resist
the potential earthquake damages.
For the last twenty years, several seismic retrofitting techniques for masonry
structures have been developed and practiced, but rarely validated with experiments
and numerical modelling. Further more, the research has been carried out mainly in
America and Japan where the risk of major earthquake is high. In Australia, although
unreinforced masonry is one of the most popular types of construction, research into
seismic retrofitting of masonry structures is rare.
The purpose of this research is to develop a new, cheaper and high strength seismic
retrofitting technique for masonry structures. An innovative retrofitting technique is
presented for improving the seismic resistance of unreinforced masonry walls using
cable systems. In this thesis, the experimental results from four unreinforced masonry
walls and eight unreinforced masonry walls retrofitted with cable systems are
presented. Seven unreinforced masonry walls retrofitted with FRP were also
conducted in this research for comparison purpose. All walls were tested under
combined constant gravity load and incrementally increasing in-plane lateral
displacement reversals. The results showed that both the strength and ductility of
iv
Preliminaries
tested specimens were significantly enhanced with this new technique. Seismic
retrofitting of unreinforced masonry walls with cables proved to be an effective and
reliable strengthening alternative.
In this thesis, a basic mechanical model has been introduced first using the simple
truss model. Then, two nonlinear finite element models based on this basic
mechanical model have been developed to validate the experimental results. One
model is developed for unreinforced masonry walls retrofitted by cable system and
the other model is developed for unreinforced masonry walls retrofitted by FRP. All
of the models have been found between the analytical and experimental results are
reasonably good agreement. The model takes into account the material nonlinearities
as well as damage due to progressive cracking. Behaviour of the masonry is modelled
using the theory of plasticity and cracking is modelled using smear cracking
approach. The model is generated using ABAQUS Finite Element program. The
validity of the model is established by comparison with the experimental results. It is
shown that both of the numerical models are capable of predicting not only the load
carrying capacity, but also the failure mode and, ductility of the retrofitted masonry
walls.
This thesis reviews the literature on all seismic retrofitting techniques and compares
their advantages and disadvantages to identify the most effective and economic
retrofitting method for unreinforced masonry buildings. It also presents the basic
concepts of seismic retrofitting and summarizes the findings from recent experimental
and analytical research activities on the seismic retrofitting of unreinforced masonry
buildings and provides some retrofit strategies on the most common failure modes. It
is expected that this thesis will provide some guidelines to assist Australian engineers
to retrofit unreinforced masonry buildings.
v
Preliminaries
LIST OF PUBLICATIONS
This thesis is the result of three years and three months of research work since March
2001 at the School of Geosciences, Minerals and Civil Engineering, University of South
Australia, Australia. During this period the following papers and reports have been
published or submitted for publication.
Chuang, S.W. and Zhuge, Y. (2002). "State-of-the-art report on seismic retrofitting of

unreinforced masonry buildings", Proceedings of 17th Australasian Conference on the
Mechanics of Structures and Materials, Gold Coast, Australia, pp. 627-632.
Chuang, S.W. and Zhuge, Y. (2002). " The development of seismic retrofitting
techniques for existing masonry and concrete structures ", Progress Report No.1,
Department for Administrative and information Services Building Management, South
Australia, Australia.
Chuang, S.W. and Zhuge, Y. (2003). "Seismic retrofitting of unreinforced masonry walls
by FRP strips", Proceedings of 2003 Pacific Conference on Earthquake Engineering,
Christchurch, New Zealand.
Chuang, S.W. and Zhuge, Y. (2003). "Seismic Retrofitting of Unreinforced Masonry

Buildings — A Literature Review ", Australia Journal of structural Engineering,
(submitted).
Chuang, S.W. (2003)."The development of seismic retrofitting techniques for existing

masonry structures". Final Report, Mawson Lakes Fellowship committee, Australia
Chuang, S.W. and Zhuge, Y. (2003). " The development of seismic retrofitting
techniques for existing masonry and concrete structures ", Progress Report No.2,
Department for Administrative and information Services Building Management, South
Australia, Australia.
vi
Preliminaries
Chuang, S.W. , Y. and McBean P.C. (2004). "Seismic Retrofitting of

Zhuge,
Unreinforced Masonry Walls by Cable System", The World Conference on
Earthquake Engineers, Vancouver, Canada, (accepted for publication).
Chuang, S.W., Zhuge, Y. and McBean P.C. (2004). "Seismic Retrofitting of

Unreinforced Masonry Walls by Cable System", The 7th Australasian Masonry
Conference, Newcastle, Australia, (accepted for publication).
Chuang, S.W., Zhuge, Y. and McBean P.C. (2004). "Seismic Retrofitting of

Unreinforced Masonry Walls by Cable System and FRP strips", Proceedings of 18th
Australasian Conference on the Mechanics of Structures and Materials, Perth, Australia,
(submitted).
Chuang, S.W., Zhuge, Y. and McBean P.C. (2004). "Numerical analysis of Unreinforced
Masonry Walls retrofitted by Cable System ", 13th International Brick/Block Masonry
Conference, Amsterdam, Holland, (accepted for publication).
vii
Preliminaries
TABLE OF CONTENTS
TITLE PAGE (i)

DECLARATION (ii)
ACKNOWLEDGEMENTS (iii)
ABSTRACT (iv)
LIST OF PUBLICATIONS (vi)
TABLE OF CONTENTS (viii)
LIST OF FIGURES (xv)
LIST OF TABLES (xxi)
NOTATION (xxii)
INTRODUCTION
1.1 General 1-1
1.2 Statement of the research problems 1-2
1.3 Research objectives 1-2
1.4 Thesis organization 1-4
2. REVIEW OF THE RELATED LITERATURE

2.1 Introduction 2-1
2.2 Basic concept and general procedures for seismic retrofitting 2-2
2.3 The seismic retrofitting methods for unreinforced masonry 2-4
2.3.1 Reduce the earthquake forces 2-4
2.3.2 Change structural system 2-7
2.3.3 Upgrade the individual element strength 2-7
2.3.3.1 Confinement/jacketing techniques 2-7
2.3.3.2 Add additional thin surface treatments
(plaster with wire mesh and cement mortar,
e.g. shotcrete, coatings, grouted steel bars) 2-10
viii
Preliminaries
2.3.3.3 Repointing 2-10

2.3.3.4 Internal reinforcement 2-12
2.3.3.5 Tying 2-14
2.3.3.6 Ground injection 2-14
2.3.3.7 Fifing in door or window openings 2-16
2.3.3.8 Enlarged openings 2-16
2.3.4 Upgrade the individual element strength —
adhesive techniques 2-16
2.3.4.1 Seismic retrofitting of URM walls byFRP 2-16
2.3.4.2 Sprayed FRP for seismic retrofitting 2-18
2.3.4.3 Seismic retrofitting of I.JRtVI walls by steel strips 2-18
2.3.4.4 Shape memory alloys for seismic retrofitting
of URM 2-19
2.3.4.5 A comparison of adhesive techniques 2-21
2.3.5 Shear wall hold-down anchors 2-21
2.4 Conclusions 2-27
3. BEHAVIOUR OF UNIREINFORCED MASONRY WALLS

AND CABLE SYSTEM
3.1 General 3-1
3.2 Unreinforced masonry walls 3-1
3.3 Parameters affecting the strength of unreinforced masonry walls 3-2
3.3.1 Wall aspect ratio 3-2
3.3.2 Type of loading 3-3
3.3.3 Axialloading 3-3
3.3.4 Construction joints 3-3
3.4 Failure model of URIVI under in-plane loads 3-3
3.5 Basic behaviour of the cable 3-5
3.6 Ductility and energy dissipation in unreinforced walls and cables 3-8
3.7 Overview of proposed retrofitting techniques 3-9
ix
Preliminaries
4. EXPERIMENTAL PROGRRAM
4.2 Test specimens 4-1
4.3 Wall dimension 4-2
4.4 Footing and top beam 4-3
4.5 Fabrication of wall specimens 4-4
4.5.1 Footing forms 4-4
4.5.2 Reinforcement cages 4-5
4.5.3 Beam construction 4-5
4.5.4 Wall construction 4-6
4.6 Retrofitting details 4-7
4.6.1 The detailed design for the cable system 4-7
4.6.2 The detail design of FRP 4-10
4.7 Material properties 4-11
4.8 Test setup 4-12
4.9 Loading history 4-15
4.10 Instrumentation and data acquisition system 4-16
4.11 Summary 4-20
5. OBSERVED BEHAVIOUR AND TEST RESULTS

5.1 General 5-1
5.2 Behaviour of unretrofitted walls 5-2
5.3 Behaviour of retrofitted walls 5-7
5.3.1 Retrofitted walls with cable system 5-7
5.3.2 Retrofitted walls with FRP 5-9
5.4 Comparison of hysteretic behaviour 5-11
5.4.1 Retrofitted walls with cable system 5-11
5.4.2 Retrofitted walls with FRP 5-13
5.5 Behaviour of retrofitting materials 5-16
5.5.1 Cable system 5-16
x
Preliminaries
5.5.2 FRP 5-18

5.6 Crack pattern of wall specimens 5-20
5.6.1 Wall 1 (Un-retrofitted wall) 5-20
5.6.2 Wall 2 (Retrofitted wall with cable system) 5-23
5.6.5 Wall 6 (Retrofitted wall with FRP) 5-32
5.6.6 Wall 7 (Retrofitted wall with FRP) 5-36
5.7 Comparison between two retrofitting techniques 5-39
5.7 Discussion and Conclusions 5-40
6. NUMERICAL MODELLING STRATEGIES FOR UNREINFORECED

MASONRY WALLS
6.2 Modelling techniques for unreinforced masonry wall 6-6
6.3 Summary of the proposed numerical model 6-9
6.4 Constitutive law of masonry 6-10
6.4.1 Properties of masonry 6-12
6.4.1.1 Uniaxial compressive behaviour of masonry 6-12
6.4.1.2 Uniaxial tensile behaviour of masonry 6-14
6.4.1.3 Biaxial behaviour 6-16
6.4.2 The constitutive law adopted by the current model 6-17
6.4.2.1 Strain rate decomposition 6-18
6.4.2.2 Plastic strain rate 6-18
6.4.3 Tension softening 6-20
6.4.4 Damaged elasticity 6-22
6.5 Failure criterion for unreinforced masonry 6-25
6.5.1 Introduction 6-25
6.5.2 Model of compression yield surface 6-29
6.5.3 Model of tensile failure 6-32
xi
Preliminaries
6.6 Masonry cracking models 6-34

6.6.1 Smeared-cracking model 6-35
6.6.2 Discrete-cracking model 6-39
6.6.3 Fracture-mechanics model 6-40
6.6.4 Crack detection surface 6-41
6.6.5 Crack direction assumptions 6-42
6.6.6 Crack shear retention 6-43
6.7 Conclusion 6-44
7. A BASIC MECHANICAL MODEL FOR THE RETROFITTED WALLS

7.1 General 7-1
7.2 Simple truss model 7-1
7.3 Validation of proposed Techniques 7-8
7.4 Effects of vertical and diagonal strut widths on the accuracy
of simple truss model 7-12
8. ANALYTICAL STUDY FOR RETROFITTED WALLS

8.2 Macro-modelling for the retrofitted wall 8-1
8.3 The finite element model for the retrofitted wall by cable system 8-2
8.3.1 Selection of element type 8-3
8.3.1.1 Masonry 8-3
8.3.1.2 Steel plate 8-3
8.3.1.3 Cable 8-4
8.3.1.4 Connection between the steel plate and
masonry wall 8-5
8.3.1.5 Connection between the steel plate and cable 8-5
8.3.2 The properties of the materials 8-6
8.3.2.1 The stress-strain relationship of masonry 8-6
xii
Preliminaries
8.3.2.2 Failure criteria for masonry 8-7

8.2.2.3 Steel plate 8-8
8.2.2.4 Cable 8-8
8.3.3 Geometry 8-8
8.3.4 Loading and boundary condition 8-11
8.3.5 Finite element discretization 8-12
8.4 The finite element model for the retrofitted wall by FRP 8-13
8.4.1 Selection of element type 8-14
8.4.1.1 Masonry 8-14
8.4.1.2FRP 8-14
8.4.1.3 Connection between FRP and masonry wall 8-14
8.4.2 The properties of the materials 8-15
8.4.2.1 Masonry 8-15
8.4.2.2FRP 8-15
8.4.3 Geometry 8-16
8.4.4 Loading and boundary condition 8-17
8.4.5 Finite element discretization 8-17
8.5 Material nonlinear analysis 8-18
8.6 Solution techniques for nonlinear analysis 8-19
8.7 Validation of the numerical modeffing results with the test results 8-21
8.7.1 URM walls retrofitted by cable system 8-21
8.7.1.1 Load-deflection plots 8-21
8.7.1.2 First cracking loads 8-24
8.7.1.3 Ultimate loads 8-25
8.7.1.4 Force carried by cable 8-25
8.7.2 URIVI walls retrofitted by FRP 8-26
8.7.2.1 Load-deflection plots 8-26
8.7.2.2 First cracking loads 8-28
8.7.2.3 Ultimate loads 8-28
8.7.2.4 Force carried by cable 8-29
xlii
Preliminaries
8.8 Comparison between the retrofitted walls by cable system

and FRP 8-30
8.8.1 Load-deflection plot 8-30
8.8.2 First cracking loads 8-31
8.8.3 Ultimate loads 8-31
8.8.4 Force carried by FRP 8-32
8.9 Conclusion 8-32
9. PERFORMANCE-BASED DESIGN FOR RETROFFITED WALLS

9.2 Concept of Performance-Based Design 9-5
9.3 Performance-based Seismic Rehabilitation Procedure 9-9
9.4 A performance-based design approach for retrofitting URM
low-rise walls 9-12
9.5 A practical design guideline for engineer 9-12
9.6 Sample design of steel strip system 9-14
9.6.1 Using Performance-based design according to
FEMA 274 9-16
9.6.2 Using Australian Standard according to
AS 1170.4—1993 9-19
9.6.3 The difference between Performance-Based Design
and Australian Standard - 9-22
9.7 Difficulty to implement PBD 9-23
10. SUMMARY AND CONCLUSIONS

10.1 Summary 10-1
10.3 Recommendations 10-6
REFERENCES
APPENDIX
xiv
Preliminaries
LIST OF FIGURES
2.1 Conventional structure 2-4

2.2 Base isolation structure 2-4
2.3 Friction damper 2-6
2.4 Active damper 2-6
2.5 Confinement of masonry wall with steel section 2-9
2.6 Jacketing with reinforced concrete 2-9
2.7 Structural repointing technique and combining with FRP laminates 2-11
2.8 Centre coring technique 2-12
2.9 Retro-reinforcement technique 2-13
2.10 Tension tie for wall to floor retrofit 2-14
2.11 Diagram of supplementary injection anchor in laboratory 2-15
2.12 Retrofitting masonry walls with FRP 2-17
2.13 Spraying in process 2-18
2.14 Steel strip system 2-19
2.15 Full-scale model and SMAs based protection system 2-20
2.16 Hold-down strap 2-22
2.17 Hold-down anchor bolts 2-23
2.18 Expansion type anchors 2-24
2.19 Drop-in and self-drilling 2-25
2.20 Self-Contained Adhesive Anchor 2-26
2.21 Site-Mixed Adhesive Anchor 2-26
3.1 In-plane failure 3-2
3.2 Out -of plane failure 3-2
3.3 The failure modes of URM under in-plane lateral loads 3-5
3.4 Axial displacement vs. axial load behaviour of cable with no prestress 3-6
3.5 Axial displacement vs. axial load behaviour of cable with prestress 3-7
4.1 The wall dimension of the specimen 4-3
xv
Preliminaries
4.2 The details of the reinforcement and dimensions for the footing 4-4
4.3 The forms the reinforced cages 4-5
4.4 A reinforced concrete foundation beam after concrete had been casted 4-6
4.5 Wall construction 4-6
4.6 Wall retrofitted using cable systems 4-7
4.7 The detailed of the cable system design 4-8
4.8 Connection details between cable system and masonry (TOP) 4-9
4.9 Connection details between cable system and masonry (BOTTOM) 4-10
4.10 The test specimen wall fixed with FRP 4-11
4.11 Thetestsetup(1) 4-13
4.12 The test setup (2) 4-13
4.13 Hydraulic jack (1) 4-14
4.14 Hydraulicjack(2) 4-14
4.15 The testjig 4-15
4.16 The loading history 4-16
4.17 Strain gauges 4-17
4.18 The locations of strain gauges for FRP 4-18
4.19 Kyowa displacement transducers 4-18
4.20 The locations of displacement transducers 4-19
5.1 Hysteretic behaviour of Wall-i (Unretrofitted wall) 5-3
5.2 The effective height H0 5-5
5.3 Assumptions for rocking strength evaluation of a wall falling

with crushing at the base corner 5-6
5.4 The load-deflection envelopes of three with cable 5-8
5.5 Variation in stiffness with lateral drift for all the walls 5-9
5.6 The load-deflection envelops of three walls 5-11
5.7 Hysteretic behaviour of Wall-2 (Retrofitted wall with cable) 5-12
5.10 Hysteretic behaviour of Wall-S (Retrofitted wall with FRP) 5-14
xvi
Preliminaries
5.11 Hysteretic behaviour of Wall-6 (Retrofitted wall with FRP) 5-14

5.12 Hysteretic behaviour of Wall-7 (Retrofitted wall with FRP) 5-15
5.13 The load-deflection envelope of Wall-2 (Retrofitted with cable) 5-16
5.16 The load-deflection envelope of Wall-S (Retrofitted with FRP) 5-18
5.17 The load-deflection envelope of Wall-6 (Retrofitted with FRP) 5-19
5.18 The load-deflection envelope of Wall-7 (Retrofitted with FRP) 5-19
5.19 Initial cracking of wall 1 5-20
5.20 Crack pattern of brick of wall 1 (1) 5-21
5.21 Crack pattern of brick of wall 1(2) 5-21
5.22 Failure mode of wall 1 5-22
5.24 Crack pattern of brick wall 2 5-23
5.25 In-plane diagonal shear cracking wall 2 5-24
5.26 The failure mode of wall 2 (Back side) 5-25
5.27 The failure mode of wall 2 (Front Side) 5-25
5.29 Crack pattern of brick of wall 3 5-26
5.30 In-plane diagonal shear cracking of wall 3 (1) 5-27
5.31 In-plane diagonal shear cracking of wall 3 (2) 5-27
5.32 Rocking failure of wall 3 5-28
5.33 Lift-up of wall 3 5-28
5.34 The failure mode of wall 3 (Back side) 5-29
5.36 FRP debonding of wall 5 5-30
5.37 Brick crack at rear of the wall of wall 5 5-30
5.38 In-plane diagonal shear cracking of wall 5 5-31
5.39 The failure mode of wall 5 (Back Side) 5-31
xvii
Preliminaries
5.41 First crack of wall 6 5-32

5.44 In-plane diagonal shear cracking of waIl 6 5-34
5.50 In-plane diagonal shear cracking of wall 7 5-37
6.1 Structural analysis methods 6-3
6.2 The difference between FDM and FEM 6-4
6.3 The finite element method concept 6-5
6.4 Modeling strategies for masonry structures 6-7
6.5 The state of stress of masonry 6-11
6.6 The typical unaxial stress-strain curve for masonry under compression 6-11
6.7 Uniaxial compressive behavior of masonry upon loading normal
to the bed joints 6-13
6.8 Test set-up for tensile strength of masonry parallel to the bed joints 6-15
6.9 Typical experimental stress-displacement diagrams for tension in
the direction parallel to the bed joints 6-15
6.10 Biaxial strength of solid clay units masonry 6-16
6.11 Uniaxial behavior of plain masonry 6-17
6.12 Post-peak tensile stress-strain models 6-21
6.13 Shear retention 6-24
6.14 Modes of failure of solid clay units masonry under biaxial loading 6-26
6.15 A typical failure surface of masonry 6-27
6.16 Simplified failure envelopes for masonry 6-29
xviii
Preliminaries
6.17 Masonry failure surfaces in plane stress 6-30

6.18 Smear crack model 6-36
6.19 [D1] matrices for different modes of failure 6-38
6.20 Discrete-cracking model 6-40

6.21 Masonry failure surface in the p-q plane 6-41
7.1 A simple truss model for the retrofitted wall with cable system 7-2
7.2 Free body diagrams 7-3
7.3 Stress-strain relationships 7-5
7.4 Simplified truss model 7-5
7.5 Force and displacement relationship in elastic analysis for two trusses 7-6
7.6 Force-displacement relationships for the retrofitted wall 7-9
7.7 Force-displacement relationships for the cable 7-9
8.1 The geometry and node locations for CPS4R 8-3
8.2 The geometry and node locations for S4R 8-4
8.3 The geometry and node locations for T2D2 8-4
8.4 The geometry and node locations for connector element (BEAM) 8-5
8.5 A stress-strain curve for clay-brick masonry 8-6
8.6 A two-dimensional failure surface for masonry 8-7
8.7 The dimension of the wall 8-9
8.8 The finite element model for the URM wall before cable reinforcing 8-9
8.9 The actual wall retrofitted with cable 8-10
8.10 The finite element model for the retrofitted wall with cable 8-10
8.11 The details of connection modeling (section A-A) 8-11
8.12 Loading and boundary condition of the retrofitted wall model 8-12
8.13 The mesh for the retrofitted wall model 8-13
8.14 The geometry and node locations for M3D4R 8-14
8.15 Surface-based tie algorithms 8-15
8.16 The actual wall retrofitted with FRP 8-16
8.17 The finite element model for the retrofitted wall with FRP 8-16
8.18 Loading and boundary condition of the retrofitted wall model 8-17
xix
Preliminaries
8.19 The mesh for the retrofitted wall model 8-18

8.20 Load-deflection plot for URM wall 8-22
8.21 Load-deflection plot for retrofitted wall with cable 8-23
8.22 F.E.M. results of load-deflection plot for URIvI wall before
retrofitting and after retrofitting 8-24
8.23 Comparison between experimental and ABAQUS for carried force
by cable 8-25
8.24 Load-deflection plot for retrofitted wall FRP 8-26
8.25 F.E.M. results of load-deflection plot for URM wall before
retrofitting and after retrofitting 8-27
8.26 Comparison between experimental and ABAQUS for carried force
byFRP 8-29
8.27 F.E.M. results of Load-deflection plot for the retrofitted walls by
cable system and FRP 8-30
8.28 Comparing the forces carried by the cable and by FRP from ABAQUS 8-32
9.1 Performance level 9-7
9.2 A global framework for Performance-based Design 9-8
9.3 Vision 2000 performance objectives 9-9
9.4 The performance-based seismic rehabilitation design procedure 9-11
9.5 Component force versus deformation curves 9-12
9.6 illustrative design example, floor plan 9-15
9.7 Illustrative design example, single-story building 9-15
9.8 Retrofitting of single-story building using cable system 9-16
xx
Preliminaries
LIST OF TABLES
4.1 Summary of masonry wall specimens 4-20

5.1 The summary of test results 5-2
5.2 The summary of test results for the retrofitted walls by cable system 5-8
5.3 The summary of test results for the retrofitted walls by FRP 5-10
6.1 Summarizes the physical properties that define the natural
characteristics of various problems 6-2
8.1 Comparisons between experimental and ABAQUS first cracking loads 8-24
8.2 Comparisons between experimental and ABAQUS loads at failure 8-25
8.3 Comparisons between experimental and ABAQUS first cracking loads 8-28
8.4 Comparisons between experimental and ABAQUS loads at failure 8-28
8.5 Comparisons between the retrofitted walls by cable system and FRP
for first cracking loads 8-31
8.6 Comparisons between the retrofitted walls by cable system and FRP
for ultimate loads 8-31
9.1 The definitions of performance level 9-3
9.2 Structural Performance level for URM walls 9-14
xxi
Preliminaries
NOTATION
A b net mortared area of the bed joints above and below the test brick
A area of net mortared/grouted section
B1 damping coefficient used to adjust one-second period spectral response for the
effect of viscous damping
B5 coefficient used to adjust short-period spectral response for the effect of viscous
damping
C1 modification factor to relate expected maximum inelastic displacements to

displacements calculated for linear elastic response
C2 modification factor to represent the effects of pinched hysteretic shape, stiffness

degradation, and strength deterioration on maximum displacement response
C3 modification factor to represent increased displacements due to dynamic P-A

effects
Cm effective mass factor to account for higher mode mass
numerical value for adjustment of period, T
D wall length
DCR demand capacity ratio
[D failed material constitutive matrix
Eme expected values of elastic modulus for masonry in compression
Fa factor to adjust spectral acceleration in the short-period range for site class
factor to adjust spectral acceleration at one second for site class
G shear modulus
xxii
Preliminaries
G compressive fracture energy
Gme expected shear modulus of masonry
G1 fracture energy
G mode I fracture energy
mode II fracture energy
H thickness of soil
H0 effective wall height
J force-delivery reduction factor
L length of wall or pier
N average SPT blow count in soil within the upper 100 feet of soil
SPT blow count in soil layer "i"
P axial load
lower bound vertical compressive strength of existing URM walls
expected vertical axial compressive force per load combinations
CE expected gravity compressive force applied to a wall
CL lower bound of vertical compressive force from load combination, pounds
probabilities of exceed ance
P1 plasticity index of soil
mean return period
Q CE expected strength of the component or element
xxiii
Preliminaries
Q CE expected strength of the component or element at the deformation level under

consideration for deformation-controlled actions
Q CL lower-bound strength of a component or element at the deformation level under

consideration for force-controlled actions
QD dead load effect
QE action due to design earthquake loads calculated using forces and analysis models
QG action due to design gravity loads
QL effective live load effect
Q effective snow load effect
Q UD deformation-controlled design action due to gravity loads and earthquake loads
Q UD force due to the gravity and earthquake loads
Q UF force-controlled design action due to gravity loads in combination with

earthquake loads
S1 long-period response acceleration parameter
s a response spectrum acceleration at the fundamental period
S spectral response acceleration at short periods
design response acceleration parameter at a one-second period
S design short-period response acceleration parameter
design response acceleration parameter at a one-second period
T fundamental period in the direction under consideration
T0 period at which the constant acceleration region of the design response spectrum
begins at a value = 0.2 T5
xxiv
Preliminaries
characteristic period of the response spectrum
V pseudo lateral load.
VbJS lateral strength of wall or pier based on bed-joint shear strength, pounds
lateral strength limited by diagonal tension stress, pounds
the maximum horizontal shear that can be resisted by a rocking failure under
static in-plane loading
V,. lateral rocking strength of wall or pier component, pounds
strength of a wall undergoing sliding, under seismic excitation, along a horizontal

joint
lateral strength limits by toe compressive stress, pounds
V test the load at first movement of a masonry unit
W effective seismic weight of the building
Y years
oç shear ratio
c initial cohesion
d parameter used to measure deformation capacity
d, depth of layer "i"
e parameter used to measure deformation capacity
fa upper bound of vertical axial compressive stress , psi
fae expected vertical compressive stress
fb normalized compressive strength of the masonry unit
f' dt lower bound of masonry diagonal tension strength, psi
xxv
Preliminaries
f'm lower bound of masonry compressive strength, psi
f'm lower bound masonry compressive strength
fme expected masonry compressive strength
compressive strength of the mortar
uniaxial tensile strength
fte expected flexural tensile strength
compressive strength of masonry
h eff height to resultant of lateral force
h,, height (in feet) above the base to the roof level
a coefficient which takes into account the vertical stress distribution at the
compressed toe
m component or element demand modifier (factor) to account for expected ductility

associated with this action at the selected Structural Performance Level, rn-factors
n number of layers of similar soil materials for which data is available
p mean vertical stress on the wall due to the axial load
PD+L estimated gravity stress at the test location
undrained shear strength in layer "i"
su undrained shear strength of soil
t wall thickness
v expected masonry shear strength
v me expected bond-joint sliding shear strength, psi
xxvi
Preliminaries
average shear wave velocity
v shear wave velocity of the soil in layer "i"
Vte average bed-joint shear strength
individual shear strength test values
w water content of soil
d maximum in-plane diaphragm displacement in inches
dilatancy angle
a factor equal to 0.5 for fixed-free cantilever wall, or equal to 1.0 for fixed-fixed
pier
aggregate interlock factor
factor of adjust fundamental period of the building
E Cr strain at the stress level of
9 rotation angle between the principal stresses and the material axes
a parameter which takes into account the boundary conditions at the typical test
layouts
a knowledge factor
p. sliding coefficient of friction of the masonry joint
initial internal friction angle
ør residual internal friction angle
(a0 } stress vector
1 n(S natural logarithm of the spectral acceleration parameter ("i" = "s" for short-period
or "i" =1 for 1 second period) at the desired probability of exceedance
xxvii
Preliminaries
natural logarithm of the spectral acceleration parameter ("i" = "s" for short-
period or "1" =1 for 1 second period) at a 1O%/50 year exceedance rate
natural logarithm of the spectral acceleration parameter ("i" = "s" for short-
period or "i" =1 for 1 second period) for the BSE-2 hazard level
natural logarithm of the mean return period corresponding to the exceedance

probability of the desired earthquake hazard level
xxviii
Chapter 1: Introduction
CHAPTER 1
INTRODUCTION
1.1 General
Australian civil engineers have not paid enough attention to earthquake resistant
design of structures until Newcastle earthquake in 1989. It was commonly believed
that the earthquake risk was insignificant thus earthquake resistant design was not
specially considered for most buildings. Earthquake resistant design of structures has
only become an important issue in Australia following Newcastle earthquake in 1989.
The Newcastle earthquake in 1989 led to the creation of a new set of guidelines for
earthquake resistant design. This new code (AS1 170.4, 1993) has resulted in the need
to systematically retrofit structures that no longer comply with the new guideline.
Un-reinforced masonry (URM) is one of the oldest and most widely used construction
methods in Australia. Inherent advantages of URM include: aesthetics, heat and sound
insulation, fire resistance, economical considerations and sound understanding of its
mechanical properties and contribute to its continuing appeal. Although the history of
past earthquakes has shown that URM buildings suffered both severe damage and
accounted for the maximum loss of life, they continue to be popular. In Australia the
majority of URM buildings were constructed with little or no seismic design
requirements. This has resulted in a large inventory of buildings with a poor capacity
to dissipate energy through inelastic deformation in an earthquake event. Further
more, most of the historical buildings throughout Australia are unreinforced masonry,
highlighting the need to improve their performance by retrofitting and strengthening
to resist potential earthquake damages.
1-1
Masonry buildings consist of several structural element types. However, the one most
likely to be subjected to earthquake damages is the load-bearing wall. These elements
are designed primarily to resist vertical loads. However, they are often subjected to in-
plane or out-of-plane loads resulting from lateral loads such as earthquakes. The in-
plane shear resistance in load-bearing URM walls is provided by the shear bond
strength of the mortar and the friction shear due to the vertical load. The aging and
often deteriorated mortar joints have little shear capacity. Under severe earthquake
loads the shear capacity of the mortar is often exceeded, which can cause the failure
of the wall. Effective retrofitting techniques are needed to strengthen such walls.
1.2 Statement of the research problems
Seismic retrofitting of structures has appeared recently in the literature. A review of

previous studies has revealed that most research has been focused on reinforced
concrete frame structures. Further more, many repair and retrofit techniques have
been developed by practicing engineers on an individual basis, there exists little
information or technical guidelines with which an engineer can judge the relative
merits of these methods. Also, the research has been carried out mainly in America
and Japan where the risk of major earthquakes is high. In Australia, although older
masonry buildings are often unreinforced and seismically vulnerable and should be
retrofitted to provide higher strength and ductility, research into seismic retrofitting of
masonry structures is rare. Therefore, there is an urgent need for research work in the
areas of seismic retrofitting of URM.
1.3 Research Objectives
The objective of this research is to develop a new alternative retrofitting technique to

upgrade existing non-ductile low-rise unreinforced masonry walls by using
experimental testing and numerical modelling.
From the review of seismic retrofitting techniques for URM, it can be seen that
adhesive techniques can be applied easily to practice. Therefore, the proposed
research will focus on adhesive techniques by using different materials such as steel
rods, Fibre Reinforced Polymer (FRP) and cables to attach on to the masonry walls to
1-2
retrofit or strengthen them. The literature review has also revealed that steel rods and
FRP have been used by other researchers/engineers; however, using cables will be an
innovation of the current project. It is expected that using cables to strengthen the
seismic resistance of URM will become one of the effective retrofitting methods.
Cable consists of a number of wires or strands and has high tensile strength, lightness
and high corrosion resistance. These materials can absorb tensile stress and increase
overall element stiffness, ductility and bearing capacity. Using cable system for
seismic retrofitting applications has other advantages such as architectural versatility,
durability, and no loss of valuable space. Furthermore, it does not add a significant
mass to the existing building and can remain unchanging dynamic properties for
structures. Therefore, Cable would be seen to be particularly suitable for seismic
retrofitting materials.
An innovative retrofitting technique is presented in this thesis using the cable system.
This cable system consists of adding diagonal cables tied to the walls. These cables
are connected to the wall through the steel plates and chemical bolts to transfer the
load to the foundation. This method appeared to be an efficient upgrading technique
worth investigating.
In order to be successful in this investigation, the following objectives need to be

achieved:
• Experimental testing of URM walls retrofitted with cable system

• Experimental testing of URM walls retrofitted with FRP
• Development of numerical model
• Verification and application of the numerical model
• Development of performance-based design procedure for URM walls
retrofitted with cable system
Parts of these objectives were achieved by designing, constructing, and testing of

nineteen wall specimens. Four walls were unretrofitted unreinforced masonry walls;
other eight URM walls were retrofitted with the proposed cable system and seven
1-3
walls were retrofitted with FRP to compare the results with cable system; All walls
were tested under combined constant gravity load and incrementally increased in-
plane lateral displacement reversals.
1.4 Thesis Organization
The thesis is presented in 10 chapters. The introduction and the objectives of the
proposed research project and the proposed retrofitting method were described in
Chapter 1. An extensive review of the literature on relevant and recent research of
seismic retrofitting techniques for masonry structures can be found in Chapter 2. The
overall behaviour of a retrofitted masonry wall cannot be anticipated without
clarifying the characteristic behaviour of each of its constituent part. Such knowledge
is prerequisite to the determination of the lateral load capacity of retrofitted walls.
Chapter 3 reviews the characteristics, behaviour, and failure mode of unreinforced
masonry walls and cables under reversed cyclic loading for the purpose of designing
retrofitted walls. Chapter 4 describes the design of the test specimens and the
expected behaviour of the retrofitted walls. The details of the test specimens and
testing procedure are also described in Chapter 4. Chapter 5 reports the behaviour of
the test specimens observed during the test and the experimental results. Chapter 5
also compares and discusses the observed behaviour and results between the
retrofitted and unretrofitted walls. An overview of the numerical models, the failure
criterion, the constitutive law and the behaviour after cracking for unreinforced
masonry are given in Chapter 6. A basic mechanical model for the retrofitted wall
with cable system is developed in Chapter 7. The finite element models for the
retrofitted walls by cable system and FRP are proposed in Chapter 8. The validation
of the numerical modelling results and some discussions are also described in Chapter
8. The development of the performance-based design procedure for URM walls
retrofitted with the cable system is presented in Chapter 9. Chapter 10 presents a
summary of research findings and conclusions.
1-4
Chapter 2: Review of the related literature
CHAPTER 2
REVIEW OF THE RELATED LITERATURE
2.1 Introduction
The basic concept of seismic retrofitting is to upgrade the structural strength and improve
the inelastic deformation capacity or ductility of the structure. In the last two decades,
several seismic retrofitting techniques for masonry structures have been developed and
practiced, but rarely validated with experiments and numerical modelling. According to the
literature search, the seismic retrofitting techniques for masonry structures offer the
following advantages (Mayes, 1989, FEMA, 1997, Cesteffi-Guidi, 1983):
1 Reduce the earthquake forces
2 Enhance/Improve the existing building to resist earthquake load
• Change the structural system

• Upgrade the element strength
All of the techniques could strengthen the structure to some extent. However, one could
not say which one is better than others because different techniques are suitable for
different cases. But it has been found out that the adhesive techniques can be applied easily
in practice. Therefore, this literature review will focus on the adhesive technique and
compare several materials such as FRP, reinforced bars, steel strips and steel plates to
2-1
provide some useful information for practising engineers involved in the design and
assessment of seismic retrofitting of URM buildings.
Older masonry buildings are often unreinforced and seismically vulnerable and should be
retrofitted to provide higher strength and ductility. Common causes of seismic inadequacy
are: insufficient design loads, structural damage sustained in a previous earthquake, flawed
design or construction, and changes in building occupancy (Badoux and Jirsa, 1990). For
seismic retrofitting, one could follow the general procedure to evaluate the building and
identify the reasons for poor seismic capacity of the building. Then based on this
evaluation, a suitable scheme to strengthen the building could be developed. In this
literature review, the basic concept and the general procedures of seismic retrofitting will
be reviewed and discussed first, followed by an extensive review and comparison of
various retrofitting techniques. It is expected that the review and discussion will provide
useful information to practising engineers on how to choose a suitable technique to retrofit
the building effectively.
2.2 Basic concept and general procedures for seismic retrofitting
The basic concept of seismic retrofitting is to upgrade the ultimate strength of overall
building and improve the inelastic deformation capacity (ductility). There are two ways to
improve the seismic capacity. One is to reduce the earthquake force and the other is to
increase the capacity of the existing building to resist the earthquake load. There are also
two ways to upgrade the existing building. One is to change the structural system or to
change the load paths to reduce the load to the specific element and the other one is to
upgrade the individual element strength.
The general procedures for seismic retrofitting can be described as follows (Wyllie, 1983):
• Seismic evaluation/assessment of the original building
The earthquake response of a building depends mainly on the stiffness, strength,

deformation capacity (ductility), hysteretic characteristics of the building, the
characteristics of the soil and the characteristics of the earthquake ground motions.
2-2
Therefore it is important to understand the seismic performance of the building and to

identify the reasons for poor seismic capacity. Then on the basic of this evaluation, a
suitable scheme to strengthen the building could be developed.
• Determination of required seismic capacity (Criteria for seismic strengthening)
The criteria for seismic strengthening selected must be suitable to accomplish the goal of
the complying with current building codes, or regulations such as AS3826-1998. The
criteria for retrofitting in conjunction with the importance of the structure, seismic activities
and seismic intensities expected at the site. If the goal is to comply with current building
codes or regulations, then those codes or regulations generally become the criteria, unless
the engineer feels that a more severe criterion is appropriate and the owner agrees.
• Development and selection of strengthening schemes
Based on the seismic evaluation, a retrofitting technique could be selected to conduct the
seismic retrofitting analysis, to develop and select the strengthening schemes. The design
of retrofit scheme for a building is generally constrained by the existing structural and
architectural configuration. To design a successful retrofit scheme, the engineer needs to
identify the structural deficiencies of the building and understand the local and global
mechanical characteristics of the building. Among these characteristics, at least the
following factors should be considered: global and local strength and stiffness supplies,
global and local deformation and energy dissipating capacities, existing structural
irregularities and the dynamic properties of the building.
• Design of connection details
A key point of all seismic retrofitting techniques is the connection between old and new
materials.
• Re-evaluation of the retrofitted building
Finally, the retrofitted structure is re-evaluated to confirm that the seismic capacity satisfies
the required capacity. Sometimes after local retrofitting, the weight increase and the
2-3
frequency change of the building will affect the dynamic behaviour and also increase the
earthquake force. The foundation design should also be checked carefully to see if it is
good enough for the new structure.
2.3 The seismic retrofitting methods for unreinforced masonry
2.3.1 Reducing the earthquake forces
• Installing Base isolation
Conventional construction can cause very high floor accelerations in stiff buildings and
large interstory drifts in flexible structures. These two factors cause difficulties in insuring
the safety of the building components and contents (Mayes, 1989) (Figure 2.1). Installing
base isolation into conventional construction can reduce both interstory drift and floor
accelerations and provide a viable economic solution to the difficult problem of reducing
non-structural earthquake damage (Mayes, 1989) (Figure 2.2).
Figure 2-1 Conventional structure Figure 2-2 Base isolation structure

(Mayes, 1989) (Mayes, 1989)
The concept of base isolation relies on decoupling the structure from the damaging
component of the seismic ground motion, which is practically obtained by supporting the
2-4
structure on a highly flexible layer. In this manner, the fundamental period of the
construction is shifted in the high-periods region, where the energy content of the
earthquake is usually lower. The base isolation technique is particularly suitable to low to
mid-rise buildings and applied mainly to strategic buildings where feasibility of the
structure must be guaranteed in emergency situations and the contents represent an
economical or cultural value to be safeguarded. Although the cost of installing a base
isolation system has been reduced recently, for example, the cost of installing a new hybrid
base isolation system is about 3% of the total construction cost (Takenaka, 2001), the
technique is still quite expensive compared to many other retrofitting techniques which will
be discussed later in the thesis.
• Installing Damper
Damping devices can reduce seismic induced vibrations and improve the overall behaviour
of the structure by increasing its internal damping through the energy dissipated by the
inelastic deformation of these special devices. Consequently, the dynamic structural
response is considerably reduced, particularly in the original members of the structure.
There are two classes of damper. One is passive damper such as fluid viscous damper,
friction damper, steel-hysteretic damper and visco-elastic damper and the other one is
active and hybrid damper such as active mass driver (AMD), hybrid mass damper (HMD)
and active variable stiffness (AVS) system.
In passive dampers, the modified properties of the structure (period and/or damping
capacity) do not vary with the seismic ground motion. Figure 2.3 shows a typical example
of friction damper system, which is installed in the building's diagonal bracing system
(Cheung et al, 2000). Friction damper system's function is similar to that of a shock-
absorbing system in a car. In general the passive dampers have proven economical for
retrofits, as an example, the cost comparison for retrofitting a three-story reinforced
concrete structure with unreinforced masonry infill walls (total area was 6,243m2) was
made by Cheung et al. It was found that the total cost was 60% less using passive dampers
than the conventional methods which inclñde the strengthening of existing shear walls,
addition of new shear walls and partial foundation upgrading.
2-5
In active and hybrid damper, the structural characteristics are modified just as a function of
the specific seismic input (Mele and De Luca, 1994). Apparently, it requires outside power
sources and gets less attention from structural engineers who worry about scarce electricity
in the event of an earthquake. Figure 2.4 shows a typical AIVID system.
Figure 2-3. Friction damper (Cheung et al, 2000)
aI-,d
Sensor
11th
6th
Figure 2-4. Active damper (Mele et al., 1994)
2-6
• Removing heavy items
Removing heavy items can reduce the weight of the whole building. Therefore, it can
reduce the earthquake force (FEMA-273, 1997).
2.3.2 Changing structural system
Changing structural system is to upgrade the whole building's strength and ductility to
resist earthquake force. It usually combines to use with upgrading the individual element
strength.
The designer can determine the force path in the retrofitted building and adjust the strength
and stiffness as needed. They can use the following ways to change the structural system to
achieve a higher strength of the building:
1 Providing additional vertical-resisting elements (shear walls, bracing, external

buttresses, parapet bracing or columns)
2 Providing/stiffening the horizontal diaphragms and ties (use extra new floor to resist
earthquake)
3 Interconnecting the masonry wall with steel ties
2.3.3 Upgrading the individual element strength
Upgrading the element strength is the way to strengthen key individual elements such as
beam, column, and wall element separately. There are several retrofitting techniques to
upgrade unreinforced masonry walls as described as below:
2.3.3.1 ConfinementlJacketing techniques
This is a conventional upgrading technique, confinement or jacketing existing masonry

walls with new steel section, reinforced concrete or FRP overlays may serve to increase
2-7
flexural strength and ductility, and in-plane shear strength. The following jacketing
techniques are the most common ways to retrofit masonry walls.
1 Confining masonry walls with new reinforced concrete columns (Comti Euro-
International Du Beton, 1997)
2 Confining masonry walls with steel sections (Comti Euro-International Du Beton, 1997)
3 Jacketing masonry walls by reinforced concrete (R.C. sheath) (Benjamin, 1995), steel
plate, fibre reinforced polymer (FRP), steel rings, closed stirrups and prestressed
diametral bars (Cestelli-Guidi, 1983)
Confinement or jacketing of masonry walls provides additional thickness to the existing

walls with reinforced concrete or steel sections anchored to the inside or outside face of the
existing walls. Confinement with steel section may be the best solution when urgent
retrofitting is required (Figure 2.5) and jacketing technique has proved to be an effective
scheme for increasing lateral resistance and ductility (Figure 2.6). The procedures for
jacketing with reinforced concrete include: attach the wire mesh to the masonry wall; apply
R.C. overlay cast on the surface of the wall and finally the reinforcement anchored with
clasps into the masonry. When using confinement or jacketing techniques, extra care must
be taken to the foundation designs as they may be inadequate to carry the additional weight
from retrofitting. Although jacketing can be a technically effective procedure, it may not
be cost-effective, and it can also be very disruptive to building occupants and affect the
space utilisation.
2-8
2.3.3.2 Additional thin surface treatments (plaster with wire mesh and
cement mortar, e.g. shotcrete, coatings, grouted steel bars)
This technique could be referred to as jacketing as well. It involves adding Shotcrete or

other coating materials to one or both faces of the wall and providing an additional
thickness to the existing wall. The methods are commonly used in the United States
(Abrams, 1998).
• Shotcrete overlays
Application of reinforced shotcrete to the surface of a masonry wall is a common method
for enhancing both in-plane and out-of-plane strength of the walls. The shear area of the
wall is increased and the height-to-thickness (hit) ratio is lowered. The thickness of the
shotcrete overlay varies with seismic demand forces, but generally ranges from 70mm to
150mm (Abrams, 1998). Commonly the shotcrete overlay is assumed to resist all of the
lateral force applied to the wall with the brick masonry being neglected all together.
Reinforcement embedded in the shotcrete layer substantially improves both shear and
flexural capacities. However, shotcrete is expensive because considerable materials and
labour are involved with placement (Abrams, 1998).
• Coatings for URM walls

Thin cement-plaster coatings can be put onto the surface of a masonry wall to enhance the
in-plane shear strength of a URM wall. Interface shear stresses between the coating and
the masonry wall surface need to be resisted by steel anchors embedded into drilled holes
with an epoxy or cementicious grout (Abrams, 1998). The size and spacing of these bolts is
dependent on the thickness of the coating. For example, a 25 mm thick coating, 6 mm
diameter bolts spaced at 0.45 m horizontally and vertically across a wall surface is adequate
(Abrams, 1998). Surface coating changes the appearance of a wall and may not be
desirable for particular architectural applications.
2.3.3.3 Repointing
Re-pointing is a traditional retrofitting technique, commonly used in the masonry industry.

This technique is to remove the damaged parts and replace by one of the same type
2-10
material or by a concrete casting and therefore to restore the physical and visual integrity of
masonry (Cestelli-Guidi, 1983, Tinazzi et a!., 2000). When reinforcement is used, the
technique is referred to as "structural re-pointing". The technique could be used when the
mortar is deeply eroded or has fallen out or small cracks formed in the joints.
University of Missouri-Rolla has conducted a series of research using FRP laminates and
rods for seismic retrofitting of masonry structures. In particular, using FRP Structural re-
pointing technique and combining with FRP laminates on particular surface has proved to
be the most effective retrofitting technique (Tinazzi et al., 2000)(Figure 2.7). This is
achieved by placing deformed FRP rods into the joints, which are bonded to the masonry
wall by a suitable paste. It has been proved from their experimental testing that the method
has shown to dramatically improve the shear and bending moment capacities of the
masonry under static or cyclic in-plane and out-of-plane loads. Also the residual load
bearing capacity remains remarkable even after high damage levels were introduced
(Tinazzi et al., 2000).
FRP laminate Masonry substrate
FRP rod Epoxy mortar
.r4
Figure 2-7. Structural re-pointing technique and combining with FRP laminates
(Tinazzi et al., 2000).
2-11
__
2.3.3.4 Internal reinforcement
The advantages of using steel bars to retrofit URM walls are cost-effective, little disruption
to users of the structure, avoiding any noticeable alteration to the original masonry finishes,
simple and quick installation and no obvious additional weight of the structure. The
technique mainly used to improve both in-plane and out-of-plane flexural load capacity of
the masonry walls.
Several researchers have successfully used reinforcing bars or rods to enhance the strength
and ductility of URM structures (Cesteffi-Guidi, 1983, Plencnik et al., 1984, FEMA, 1992,
Pierepiekarz, 2000). Different techniques have been used among these investigators. For
example, some investigators used the centre coring technique, which involves grouting
reinforcing bars in cores drilled through the wall height in the vertical direction with
adequate anchorage (FEMA, 1992)(Figure 2.8). This technique consists of removing 100
mm diameter vertical cores from the top of the wall into the footing at regular intervals
(about 0.9 m to 1.5 m apart) and placing reinforcing steel and grout in the cored holes. The
reinforcement could be used with and without post-tensioning. This technique provides a
reinforced vertical beam to resist flexural stresses (Plencnik et al., 1984, FEMA, 1992).
The advantages of the technique include the preservation of the aesthetic appearance of a
building and significant construction cost savings (50% compared to the conventional
shotcrete approach) (Pierepiekarz, 2000). However, the method is criticized as well since it
is irreversible.
.. I.-
I
-
—
rekntorcernent
4- rt.
Figure 2-8. Centre coring technique (FEMA, 1992)
2-12
Grouted steel bars with diagonal, vertical and horizontal configurations have been used by
several investigators to increase tensile strength in these directions (Garrity, 1995,Manzouri
et a!., 1996). A retro-reinforcement technique that install stainless steel reinforcement
horizontally into existing masonry structures to increase the tensile strength and ductility of
the structures was developed jointly by the University of Bradford and Proprietary
Reinforcement Engineering (P.R.E) (Garrity, 1995)(Figure 2.9). The main procedures for
the techniques are: grooves up to 10 mm thick are cut into the mortar bed joints with a
depth of 50 — 70 mm; remove dust; inject a layer of grout; push the steel bar into the layer
of grout and then a second layer of grout is injected over the reinforcing bar. The major
limitation of the technique is that it requires a relatively thick layer of mortar joints.
root timbers tied to

gable wall masonry
6mm dia. stainless

reinforced
steel reinforcing bars mason
C'refro-reinfofcemenr) lintel
N
- reinforced masonry
spandrel beam
reinforced masonry
edge beam
16mm dia. threaded stainless

steel tie bars bolted to timber
floor joists and resin bonded
into spandrel and edge beam
masonry
Figure 2-9 Retro-reinforcement technique (Garrity, 1995)
2-13
2.3.3.5 Tying
This technique could be applied to masonry buildings without proper horizontal

diaphragms and interconnection between the diaphragms or roof and the masonry wall.
The technique could enhance the composite behaviour of the structure as opposed to
independent response. In order to anchor the floor and walls, metal strips can be bolted to
the joists and anchored to the wall (Figure 2.10) (Mercer, 1998). These metal strips, called
tension ties, are what keep the floor from sliding off the portion of the wall on which they
sit. The tension ties are placed on both the joists parallel and perpendicular to the walls
(Mercer, 1998). The technique is cost effective but the exposed steel could be vulnerable to
corrosion.
Figure 2-10. Tension tie for wall to floor retrofit (Mercer, 1998)
2.3.3.6 Grout injection
Grout injection technique has been used to repair cracked/damaged masonry. It should be
noted that the selection of grout materials is important, because corrosion protection and
compatibility with the existing materials have to be ensured. Injection of an incompatible
grout mix has been known to cause considerable damage (Gigla and Wenzel, 1997).
Epoxy materials have excellent penetration and bond characteristics and are commonly
used for concrete retrofits. However, they are not suitable for masonry partially because
the high strength epoxy materials are not structurally compatible with soft, relatively weak
masonry, and large inclusions of injected epoxy could significantly change stress
2-14
distributions under loads (Manzouri et al., 1996). Cement based grout injection is one of
the most widely used retrofit methods for unreinforced masonry. The grout injection
procedure involves drilling small injection holes (around 10 mm) into mortar joints;
pumping a fluid cement-based grout mixture into cracks, voids, or cavities within masonry.
Low injection pressures should be maintained to prevent damage while strengthening
already fragile materials. The method is most cost-effective for repairing isolated cracks.
When steel bars are used, the technique is referred to as "reinforced injection" and it
provides an effective way to restore composite action and the integrity of previously
damaged masonry walls. Supplementary injection anchors are used as a repair system for
historical masonry for transferring additional tensile and shear forces that cannot be
sustained by the masonry alone (Gigla and Wenzel, 1997). The anchors perform as un-
tensioned steel reinforcement or as prestressed tendons. Common anchor materials are
standard reinforcing steel, threaded rods or special prestres sing bars with roll-formed
sections. After clearing and pre-wetting the drill hole, the anchor element is inserted,
centred with spacers and grouted (Figure 2.11). Although reinforced injection is an
effective technique to increase the strength and ductility of masonry walls, it requires a
similar level of effort and cost as traditional techniques such as reinforced shotcrete or
composite fibre overlays. The main advantage of the technique is that they are least
disruptive to the building's appearance and occupants (Manzouri et al., 1996).
bond length free anchor length
——I
anchor length
Figure 2-11 Diagram of supplementary injection anchor in laboratory (Gigla, 1997)
2-15
2.3.3.7 Filling in door or window openings
A conmion method of stiffening or strengthening an in-plane masonry wall is to fill

window or door openings with masonry. Infihling of an existing opening will stiffen and
strengthen a perforated shear wall. The restriction of opening length was intended to limit
the introduction of new masonry, which may be considered to exhibit behaviour equal to
that of the original masonry (FEMA, 1992). The method could only be used if the existing
foundation is adequate to carry the extra loads from the filling and the functional or
aesthetic aspects of the building are not affected.
2.3.3.8 Enlarged openings
Door and window opening in URM walls may be enlarged to alter the aspect ratio of an
adjacent pier. By removing a portion of masonry above or below an opening, the height to
length aspect ratio of the adjacent piers will be increased to such an extent that rocking
behaviour may govern their response. Although this approach will weaken a perforated
masonry wall, it will also increase its inelastic deformation capacity if a ductile rocking
mechanism can be invoked. Furthermore, if the method is used, excessive diagonal tensile
stresses can be relieved for a relatively stocky pier, thus lowering its vulnerability to non-
ductile "X" cracking (FEMA, 1997).
2.3.4 Upgrade the individual element strength — adhesive techniques
More recently, steel strips and new technologies in composites such as FRP laminates have
been adapted to attach to the surface of masonry walls to strengthen them. The techniques
will enhance strength and ductility of unreinforced masonry walls.
2.3.4.1 Seismic retrofitting of URM walls by FRP
Fibre reinforced polymers (FRP) consist of high resistance fibre impregnated with
polymeric resins and have high tensile strength, lightness and corrosion insensitivity. FRP
can be adhered to the surface of a masonry wall with epoxy to provide strength across
potentially weak planes. Because of the high tensile strength of the material, a fairly thin
sheet can provide substantial strength to the masonry (Figure 2.12) (Korany et al., 2001).
2-16
There are three common types of FRP being used, which are Glass, Aramid, and Carbon
Fibre Reinforced Polymers (GFRP, AFRP and CFRP, respectively) and FRP are
commercialised in different shapes: rods, tendons, laminates and three-dimensional
components. All of them have been successfully used to enhance the strength and ductility
for seismic retrofitting of unreinforced masonry structures (Ehsani et al., 1999, Kolsch,
1998, Tumialan et a!., 2001).
Using FRP in seismic retrofitting applications has advantages like rapid application,
durability, low cost of installation, no loss of valuable space and no additional weight and
therefore can remain unchanging dynamic properties for the structures. However, it is
found out from the research that the inelastic deformation capacity of the wall may be
limited as a result of de-bonding of the FRP sheets which occurs suddenly with little
ductility (Abrams, 1998, Ehsani et a!., 1999). Therefore, FRP materials should be placed
strategically on those piers that are likely to attract damage first (high demand to capacity
ratio) (Abrams, 1998). The aesthetic appearance of the building may be a concern as well
as the natural appearance of the masonry wall is replaced with the FRP sheets. In general,
due to high cost and no complete knowledge of their mechanical behaviour, FRP has not
become popular in Australia yet for seismic retrofitting of masonry walls.
Figure 2-12 Retrofitting masonry walls with FRP (Korany et al., 2001)
2-17
2.3.4.2 Sprayed FRP for seismic retrofitting
More recently, an entirely new method of retrofitting concrete using Sprayed FRP (SFRP)
coatings have been developed (Banthia and Boyd, 2000). The investigators argued that
FRPs with continuous fibres are highly anisotropic. Properties in the direction of fibre
alignment are generally much superior to those in the direction perpendicular to fibre
alignment. In order to overcome such disadvantage, a new technique which consists of
spraying polymer and short, randomly distributed fibres concurrently on the surface of
concrete to be retrofitted using a spray gun is developed (Banthia and Boyd, 2000)(Figure
2.13). It was found from the testing results that sprayed composites with randomly
distributed short fibres performed equally well better than wraps with continuous fibres.
However, the method has not been applied to masonry structures yet.
Figure 2-13 Spraying in process (Banthia and Boyd, 2000)
2.3.4.3 Seismic retrofitting of URM walls by steel strips
A steel strip system for seismic retrofitting of low-rise concrete block masonry and
reinforced concrete walls has been developed by Taghdi et a! (2000). This steel strip
system consists of diagonal and vertical strips that were attached using through-thickness
bolts. Stiff steel angles and anchor bolts were used to connect the steel strips to the
2-18
foundation and the top-loading beam (Taghdi et al., 2000) (Figure 2.14). The width of each
diagonal steel strip was 220 mm with a thickness of 3.81 mm. The steel strip width was
chosen to ensure yielding of the gross section in tension prior to net section fracture at the
bolt locations. Two 80 x 3.81 mm vertical steel strips were added as boundary elements on
each side of the walls. Through-thickness 9.5 mm and 15.9 mm diameter structural steel
bolts were used to fasten the vertical and diagonal steel strips to the wall. The experimental
testing proved the technique was effective in significantly increasing the in-plane strength,
ductility and energy dissipation capacity of low-rise unreinforced walls. The testing results
indicated that the steel strips contributed to the energy dissipation for about 50% of the
total. However, the method may not be suitable for brick masonry. Brick can be difficult
to drill and anchor into. Most brick is hard and brittle. Old, red clay brick is often very soft
and is easily over-drilled. Both of these situations can cause problems in drilling and
anchoring.
Figure 2-14 Steel strip system (Taghdi et a!., 2000)
2.3.4.4 Shape memory alloys for seismic retrofitting of URM
Shape Memory Alloys (SMA5) are characterised by super-elasticity allowing energy

dissipation through a phase change from Austenite to Martensite and vice-versa without
any material damage. In recent years, research work has been carried out in Europe to
2-19
investigate the possibility of using the SMAs for the realisation of mechanical and seismic
protection systems for cultural heritage structures based on devices having intrinsically
energy dissipation capabilities (Renda et al., 2000). A series of experiments on URM
shear walls retrofitted with SMAs device under the in-plane cyclic loading have been
conducted (Renda et al., 2000) (Figure 2.15). It was found that the pre-stressing capability
and the energy dissipation of SMAs give an effective improvement of earthquake
resistance of the URM structures. The results showed that the SMAs devices contributed to
the energy dissipation for about 30% of the total. However, this new technique has a
similar problem as for steel strip retrofit; the fragility of the masonry does not allow a
simple fixing of the SMA devices.
Figure 2-15 Full-scale model and SMAs based protection system (Renda et al., 2000)
2-20
2.3.4.5 A comparison of adhesive techniques
The use of adhesive techniques has proved to be one of the most exciting and effective
technologies for external strengthening of masonry structures. Many researchers have used
different materials for adhesive technique to strengthen masonry structures and the results
seem quite successful to improve the strength and ductility of the structures. But which
one is the best? Apparently there is no simple answer for the question. FRP has the
advantages of the ease of application and lightweight; better durability than steel (as they
are not subject to corrosion). Therefore these make FRP as one of the most suitable
materials for seismic retrofitting of URM. The reason why FRP is not too popular yet is
partially due to their high cost. The initial cost of FRP materials is about 10 times more
than steel (ASBI, 2000), however the maintenance cost could be reduced as the steel
requires protection from corrosion which increases its total cost. Another reason for FRP
being less popular is that many practising engineers have not obtained enough knowledge
on FRP; especially their long-term behaviour requires to be better understood. At the
present, FRP is perfectly suitable for heritage structures because the money is not an issue
for the case (Tumialan et a!., 2001).
2.3.5 Shear wall hold-down anchors
When the applied force at the top of the wall reaches a certain amount, the overturning
moment will equal the resisting moment. Even the slightest additional applied force will
cause the wall to turn over. Since we cannot allow walls to overturn, we must install
anchors to hold them down to the foundation. Hold-down anchors are the most common
way to attach a wall to its foundation to resist uplift and overturning, but the concrete
foundation must be strong enough to hold the anchors. If the concrete is weak or
deteriorating and drilling holes for anchors is likely to cause cracks or crumbling, the
foundation should be replaced.
Hold-down anchors can take many forms. Some look like straps that emerge from the
foundation and nail to edge or face of the wall (Figure2.16). Creating a connection between
the foundation and shear wall need not involve bolts. This connector relies on an embedded
strap similar to the wall anchors you would see used in concrete tilt-up construction. Nails
2-21
create the connection between the strap and wall. The strap may also deteriorate due to its
reduced cover by concert.
Figure 2-16 Hold-down strap (ARA, 1997)
Others connect foundation bolts or threaded rods to the walls via bolts or nails (Figure
2.17). These ase called hold-down anchor bolts. The depth that the bolt is embedded in the
concrete is important. The hold-down anchors create a path for forces to travel out of a
shear wall and into the foundation.
2-22
APA RATED SHEATHING

or APA RATED SIDING 303
Studs
to framing
size as required!
Sill Plate
Bolt to foundation
Concrete foundation
Figure 2-17 Hold-down anchor bolts (ARA, 1997)
Hold-down anchor bolts are manufactured in a variety of types and sizes. These types of
anchors can be divided into two generic categories: expansion-type anchors and adhesive
or chemical-type anchors. (BIA, 1986)
1. Expansion type anchors: Two different types of expansion anchors are most
commonly used in seismic retrofit applications: the wedge anchor and the sleeve
anchor (see Figure 2.18). These anchors develop their strength by means of
expansion into the base material. Wedge anchors develop their hold by means of a
wedge or wedges that are forced into the base material when the bolt is tightened.
The wedges create large point bearing stresses within the hole; therefore, this
anchor requires a solid base material to develop its full capacity.
2-23
Sleeve anchors develop their strength by the expansion of a cylindrical metal sleeve
or shield into the base material as the bolt is tightened. The expansion of the sleeve
along the length of the anchor provides a larger bearing surface than the wedge
anchor, and is less affected by irregularities and voids in the base material than is
the wedge anchor.
//
A) AJN NOR B) ANeJOR
Figure 2-18 Expansion type anchors (BIA, 1986)
Drop-in and self-drilling anchors (see Figure 2.19) are two other types of expansion
anchors available, but are typically not recommended for use in seismic retrofit
applications. The reason for this is due to the embedment and setting characteristics
of the two anchors. Both anchors are produced to allow shallow embedment depths
and are expanded or set by an impact setting tool. The combination of shallow
embedment and high stresses imparted by the expansion tend to cause cracking or
splitting in concrete. Depending on the extent of cracking or splitting, the anchor
could experience a reduction in load-carrying capacity or undergo complete failure
during installation.
2-24
A) SELFORILUNG ANCHOR B) DROP4M
Figure 2-19 Drop-in and self-drilling (BIA, 1986)
There ase several considerations that should be examined when contemplating the
use of expansion-type anchors. These are: 1) Expansion anchors should not be used
to resist vibratory loads. Vibratory loads tend to loosen expansion anchors. 2)
Specific torques are required to set expansion anchors. Excessive torque can reduce
anchor strength or may lead to failure as excessive torque is applied. 3) Expansion
anchors require solid, hard embedment material to develop their maximum
capacities.
2. Chemical (adhesive) anchors are a threaded rod that is set in epoxy. For most
retrofit work, should use chemical anchors. For hold-down, wedge anchors do not
generally perform as well as adhesive anchors during the cyclic uplift loading
caused by earthquakes. Therefore, chemical anchors were adopted in this study.
Two basic types of adhesive anchors are currently available. The major difference
between the two is that one anchor is manufactured as a pre-mixed, self-contained
system, whereas the second type requires measurement and mixing of the epoxy
materials at the time of installation. The more popular self-contained types use a
double glass vial system (see Figure 2.20) to contain the epoxy. The outer vial
contains a resin and the inner vial contains a hardener and aggregate. The glass vial
is placed in a re-drilled hole and a threaded rod or bar is driven into the hole with a
2-25
rotary hammer drill, breaking the vials and mixing the adhesive components. The other
type of adhesive anchor requires that the epoxy components be hand-measured and
mixed before the epoxy is placed into a pre-drilled hole. A threaded rod or bar is then
set into the epoxy mixture, as shown in Figure 2.21.
A) EPOXY CAPSULE
B) THREADED ROD
C) INStÔ1LIJED ANCHOR
Figure 2-20 Self-Contained Adhesive Anchor (BIA, 1986)
MIX
PLACE
Figure 2-21 Site-Mixed Adhesive Anchor (BIA, 1986)
2-26
There are special requirements and limitations that should be considered when
contemplating the use of adhesive anchors. They are: 1) Specially designed mixing and/or
setting equipment may be required. 2) Dust and debris must be removed from the pre-
drilled holes to insure proper bond between the adhesive and base material. 3) The adhesive
mixture tends to fill small voids and irregularities in the base material. 4) Large voids may
cause reductions in anchor capacities. This is especially true with the self-contained
adhesive anchors since a limited volume of epoxy is available to fill the voids and provide
a bond to the anchor. 5) The adhesive bond strength is reduced at elevated temperatures
and may also be adversely affected by some chemicals.
2.4 Conclusions
Several seismic retrofitting techniques for URM structures have been reviewed and
discussed in this chapter. In order to retrofit existing buildings, it is important to understand
the seismic performance of the building and therefore to identify the reasons for poor
seismic capacity of the building. Then based on this evaluation, a suitable retrofitting
technique/scheme could be developed. Seismic retrofitting of masonry buildings is
relatively new for many structural engineers. A good retrofitting solution requires
consideration of technical, economic and social aspects. Each retrofitting technique has its
own advantages and disadvantages, when a technique is appropriate for one building, it
may not necessarily be appropriate for another. The selected method must be consistent
with aesthetics, function and the strength, ductility and stiffness, and the cost requirements.
According to the literature review, the following comments could be made: different
seismic retrofitting techniques maybe applied to different cases. For the adhesive
technique, FRP seems to provide the best solution. However, due to the high cost of FRP,
it has not become a popular retrofitting method yet. Research is still required to better
defme non-linear hysteretic behaviour of plain and retrofitted unreinforced masonry walls
so that design parameters (strength and deformation capacity) can be quantified for various
rehabilitation methods (Abrams, 1998).
2-27
Chapter 3: Unreinforced masonry walls and cable system
CHAPTER 3
BEHAVIOUR OF UNREINFORCED MASONRY WALLS AND

CABLE SYSTEM
3.1 General
The overall behaviour of a retrofitted unreinforced masonry wall cannot be evaluated without
understanding the characteristic behaviour of each constituent part. This chapter reviews the
characteristics, behaviour and failure modes of unreinforced masonry wall and cables, under
reversed cyclic lateral loading.
3.2 Unreinforced masonry walls
Typical Unreinforced masonry (URM) buildings are likely to be composed of several load-
bearing masonry walls arranged in orthogonal planes, with relatively flexible floor
diaphragms.
Two types of failure are commonly observed in load bearing URM wall subjected to seismic
load. These are in plane-failure characterized by a diagonal tensile crack pattern and the out -
of -plane failure, where cracks are primarily along the mortar bed joint; some inclined cracks
may also be developed. The exact crack pattern will depend on the wall's boundary
condition.
These two types of failure mode are shown in Figures 3.1 and 3.2.
3-1
———7
--I /
/ S.
/
/
-p
—i-
/
/
/ /
Figure 3-1 In-plane failure Figure 3-2 out-of-plane failure
The reasons for causing the out-of-plane failure of URM elements are excessive deflections
of diaphragms and insufficient connections between them such as parapet walls, veneers, and
unanchored load bearing walls during earthquake events. However, out-of-plane failure could
be prevented easily by proper measures (e.g. reinforced concrete ring beams, steel ties at the
floor levels). To provide the stability for in-plane walls to avoid collapse becomes a very
important issue. Therefore, this thesis will focus on the in-plane failure of URM walls.
3.3 Parameters affecting the strength of unreinforced masonry walls

It is clear from many investigations of low-rise shear walls that the behaviour of these walls
is affected by various parameteEs. A review of the influence of these parameters is helpful in
providing an understanding of the behaviour at the wall and failure modes involved. The
following discussions will focus on important parameters relevant to this study.
3.3.1 Wall aspect ratio
A low-rise wall is a wall with aspect ratios (i.e., with a ratio of height to horizontal length) of
less than two, when subjected to large shear force reversals the wall may exhibit non-ductile
behaviour. The flexural capacity of these walls cannot be attained and diagonal tension,
compression or sliding-shear failures develop prematurely. These modes of shear failure may
be brittle. Therefore, inelastic deformability and hence dissipation of earthquake induced
seismic energy is reduced. Wall with aspect ratios of less than 1.0 will induce shear failure
and greater than 1.0 will induce flexural failure. Thus, the aspect ratio of a shear wall is a
major parameter that dictates its behaviour (Taghdi, 1998).
3-2
3.3.2 Types of loading
Reversed cyclic loading applied to low-rise shear walls produces crossing of diagonal shear
cracks created by lateral load reversals. These intersecting diagonal cracks reduce the
capacity of diagonal compressive struts that form a major part of the mechanism of shear
resistance. Moreover cyclic loading may cause shear walls to slide along their construction
joints. This sliding does not occur in monotonically loaded walls. Therefore, while
monotonic force-deformation relationships may be taken as representative of the envelope of
hysteresis loops in flexure dominant members, the same may not be true at later stages of
loading for low-rise walls where cyclic loading effects often result in strength degradation
(Doostdar, 1994). Up to certain values, the axial compression will decrease the shear capacity
because the high axial compression will crush the bricks and mortars (Zhuge, 1996).
3.3.3 Axial loading
Generally, axial compression tends to increase the strength of walls. Also, in presence of
axial compressive stress, the inclination of diagonal shear cracks tends to be greater than 45
degree (as would be predicted by Mohr circle). Higher axial compression results in increased
shear capacity, as well as increased flexural capacity except for compression-controlled
failures (Drysdale et a!, 1994).
3.3.4 Construction joints
A construction joint is the weakest point in a low-rise shear wall subjected to inelastic load
reversals. Due to the interaction of flexure and shear, sliding along the construction joint may
take place before the wall reaches its capacity. This sliding, accentuating the problem,
destroys the interface along the construction joint. This will severely lower the strength of the
wall and result in a poor energy dissipation mechanism. Walls with low aspect ratio are more
likely to develop sliding shear along their construction joints (Taghdi, 1998).
3.4 Failure mode of URM under in-plane loads
The capability of unreinforced masonry walls to resist lateral loads is limited by the strength
of both masonry units and bed joint. As mentioned in Section 3.3.3, axial loads affect the
behaviour of low-rise walls. At lower axial loads, two failure modes are possible for the wall
with aspect ratios of less than 1.0: (1) when earthquake forces act in the plane of the walls,
3-3
sliding may take place along bed joints or (2) a horizontal crack may develop along the base
and the wall may rock, i.e. overturn around its toe. Localized cracks will appear at the
comprised corner due to the overturning moment. These cracks increase until complete toe
crushing occurs. This toe crushing does not affect the overall stability of the building when
deformations are small, which is usually the case for non-slender walls. The rocking failure is
the typical ultimate behaviour of a wall when subjected to lateral in-plane force combined
with low vertical loads (Taghdi, 1998).
Above a certain level of axial load, lateral sliding is suppressed and diagonal cracks form
within the masonry bricks due to an increase in bed joint friction. Under seismic loading,
cracks will occur in directions perpendicular to principal tensile stress. These in-plane shear
cracks are commonly expressed by double-diagonal (X) shear cracking. Such unreinforced
masonry walls are generally incapable of withstanding severe repeated load reversals,
suffering from low energy-dissipation capacity and severe strength degradation
characteristics (Taghdi, 1998).
The failure modes of URM walls under in-plane lateral loads can be summarized into two
groups: flexure bending and shear failure (Zhuge, 1996).
The flexure bending failure is caused because horizontal load or displacement demand
increase, bed joints crack in tension and shear is carried by the compressed masonry; final
failure is obtained by overturning of the wall and simultaneous crushing of the compressed
corner (Magenes and Calvi, 1997).
The shear failure is caused because peak resistance is governed by the fonnation and
development of inclined diagonal cracks, which may follow the path of bed joints and head
joints or may go through the bricks, depending on the relative strength of mortar joints,
brick-mortar interface, and bricks. At low levels of vertical load andlor low friction
coefficients, potential sliding planes can form along the cracked bed joints due to the
formation of tensile horizontal crack in the bed joints (Magenes and Calvi, 1997).
Two failure modes may exist under shear failure:

(1) shear sliding
(2) diagonal shear
3-4
Three failure modes may exist under flexure bending:

(1) rocking
(2) toe crushing
(3) flexural cracking at the heel
These modes are shown in Figure 3.3
.1
shear sflding diagonal shear
!'
!1
I 1T/
'I
flexural cracking rockin toe crushing
Figure 3-3 The failure modes of URM under in-plane lateral loads
(Zhuge, 1996)
It should be noted that not all the above failure modes will involve collapse of the wall panels
and final failure may be the combination of several failure modes.
3.5 Basic behaviour of the cable
Figure 3.4a shows the deformation vs. force curve of an axially loaded cable. As shown, the
cable buckles elastically when subjected to compressive strains, while it can develop its
yielding strength under tension. Figure 3.4b illustrates the behaviour of the same cable
subjected to several cycles of deformation. As shown, all inelastic tension elongation
accumulates every time the cable yields in tension. Figure 3.5a shows the deformation vs.
3-5
force curve of an axially loaded prestressed cable. Figures 3.4a and 3.5a are similar, except
that there is an initial state of stress and strain in the prestressed cable, which is accounted for
in Figure 3.5a by shifting the origin of the force vs. displacement axes. Because of this shift,
the cable can now "resist" compressive force (reduction of its initial tension can be
interpreted as the cable developing a compressive force) (Teran-Gilmore et a!. 1995).
(a) one cycle (b) several cycles
Figure 3-4 Axial displacement vs. axial load behaviour of cable with no prestress (Teran-
Gilmore et al. 1996)
F F
C '9
I
/
&
ooy 8
(a) one cycle (b) several cycles
3-6
initial I
prestr& /
(c) behaviour beyond point C
Figure 3-5 Axial displacement vs. axial load behaviour of cable with prestress (Teran-
Gilmore et al. 1996)
The yielding of a prestressed cable implies a reduction in its initial prestress and thus a
reduction in its capacity to "resist" compression. Assume the cable yields following the path
OABC in Figure 3.5b: the cable remains elastic in OA; once it reached point A, it yields and
follows AB; finally, the cable unloads up to its original prestressed deformation and reaches
point C. Figure 3.5c, which shows the behaviour of the prestressed cable beyond point C,
illustrates schematically the loss of prestress in the cable by comparing the initial prestress in
the cable with the remaining prestress after the cable yields and unloads to its initial position
(i.e., to equal to 0). As shown, there is an important reduction in the cable's capacity to
"resist" compression due to this loss of prestress.
Buckling of a cable is highly undesirable. As suggested in Figure 3.5a, the contribution of a

cable to resist lateral loads can suddenly "disappear" once it buckles. A sudden change in the
behaviour of a cable can lead sudden and undesirable changes in the strength and
deformation. Excessive yielding of a cable should also be avoided. The yielding of a cable
results, as suggested in Figure 3.5c, in a reduction of its ability to "resist" compressive forces
in a permanent axial elongation. The elongation of the cable leads to a reduction in its lateral
stiffness. Therefore, the axial strength and the amount of prestress provided to the cables
should be designed to prevent their buckling and / or excessive yielding.
3-7
The use of high levels of prestress, so that the cables may moderately yield in tension at
relatively small drifts, has been discussed by Pincheira and Jirsa (1992). Under these
circumstances, the cables dissipate some energy through their non-linear behaviour.
Pincheira and Jirsa (1992) noted that letting cables yield is more effective than keeping them
elastic provided they do not become slack in the process. Although this is an attractive
option, it should be considered carefully since cables with high levels of prestress are likely
to yield and lose some prestress during small and moderate seismic events that usually occur
prior to the safety level ground motion.
3.6 Ductility and energy dissipation in unreinforced walls and cables
Ductility is defined as the ratio of a specific deformation at any instant to that at the
beginning of yield. Ductility in structural walls can be exhibited only if the constituent
material itself is ductile (Magenes et al. 1997).
Materials with good post-yield behaviour are usually ductile, and have a relatively large
hysteretic ioop indicating good energy absorption per cycle such as cable. Materials
exhibiting brittle behaviour, such as unreinforced masonry, have poor energy absorption
characterised by small pinched and degrading hysteresis loops and are therefore non-ductile.
Although masonry can carry considerable compressive stresses, their performance in tension
is weak. But the cable is a good ductile material with good tensile strength. Therefore,
detailing of this kind of composite structure becomes extremely important to achieve an
overall ductile behaviour.
Yielding of cables of laterally loaded cantilever wall is a major source of energy dissipation.
A retrofitted unreinforced wall will increase energy dissipation due to distributing the
cracking to entire wall if the following failure modes (diagonal tension or diagonal
compression caused by shear, sliding shear along construction joints) are prevented.
Although cables are good in durability, the issue wasn't a concern of this thesis and will not
be discussed in details.
3-8
3.7 Overview of proposed retrofitting techniques
In order to preclude the above types of failures that are typically observed in URM walls and
to make the walls behave in a ductile manner, diagonal cables will be used to cover the areas
where principal stresses develop. The diagonal cables will extend between the corners of the
wall. These cables will cover the diagonal tension zone. It will strengthen diagonal tension
resistance and will prevent diagonal tension failure. Finally, the wall will be secured against
sliding using very stiff angles and high strength masonry anchors.
3-9
Chapter 4: Experimental program
CHAPTER 4
EXPERIMENTAL PROGRAM
4.1 Introduction
The purpose of this research is to develop a new high strength seismic retrofitting
technique for masonry structures. An innovative retrofitting technique with cable
system is presented in this thesis. In this chapter, the set-up of the experimental
program will be discussed.
4.2 Test specimens
To simulate practical situations, solid clay bricks (medium sandstock, size of brick:
230x1 lOx76mm) which are commonly used as load bearing walls in Australia were
chosen from commercially available clay bricks and payers to build the test wall
panels. The mortar used had a mix design of 1:1:6 (cement: lime sand: sand, by
volume) ratio and a thickness of 10mm. To maintain consistency, all the wall panels
were built by the same bricklayer and were cured in air for 28 days, before testing.
Each test wall panel was built on a reinforced concrete beam which was bolted on to
the reaction floor. The beams acted as a foundation for the walls.
The type of FRP used for this experiment was Sika CarboDur S (width 30mm,
thickness 1.2mm and tensile strength 2800 MPa) and epoxy adhesive for bonding
Sika CarboDur Strips was Sikadur-30. The FRP was fixed for one side of the wall
only.
4-1
The type of cable used for this experiment was Rostan typical grade 316 stainless
steel wire rope (19 single strands, diameter 10mm, and breaking load 71 KN) and
ending for cable was seafast threaded swage terminals (RF1513M1O1O). The
anchorage for the connection plate to the foundation beam was Ramset Chemset
Anchor (M l6xl9Omm, design tensile and shear load 8.5KN per anchor and Chemset
800 series). The cable was fixed on one side of the wall only.
4.3 Wall dimension
Walls with aspect ratios of greater than 1.0 have shear strength higher than their
flexural strength. These walls should fail in shear in a non-ductile manner. The
proposed retrofitting system aims to improve the performance of walls by increasing
shear strength above their flexural strength and by increasing ductility and energy
dissipation capability, therefore all specimens were chosen with an aspect ratio of 1.0
to ensure that most of the unretrofitted walls would exhibit shear-induced damage.
Therefore, the in-plane failure would be dominated. The dimension of the wall is
940mm long x 940mm high x 110 wide (11 courses high and 4 bricks in each course)
(see Figure 4.1). The wall was constructed on top of a concrete foundation beam to
simulate a house footing.
4-2
II II II
II II
II II II
940mm [4 courses)
Figure 1 Dimension
Figure 4-1 The wall dimension of the specimen
4.4 Footing and top beam
A stiff reinforced concrete footing was built to provide full fixity at the base. To
achieve full foundation fixity, they were over-designed, both in terms of strength and
stiffness. The specifications for the reinforcement and dimensions are shown in Figure
4.2. Each footing was provided with two holes. These holes were located to match
with the holes in the strong floor. High-strength bolts were used to fix the foundation
to the strong floor. Also, a stiff steel top beam was placed at the top of each wall
specimen. Its function was to transfer vertical compressive stress uniformly to the
wall.
4-3
Chapter 4: Experimental pro gram
LOCATION OF 2 TIE DOWNS (SEE DETAIL)
Y2Oiiini bars
l0rui SHEAR LIGATURE

@10
FOUNDATION BEAM
Figure 4-2 The details of the reinforcement and dimensions for the footing
4.5 Fabrication of wall specimens
The footing forms and the reinforced cage for footing were built first.
4.5.1 Footing forms
The forms consisted of a wooden base with steel sides connected by means of wing
nuts. This allowed for easy removal of the sides to access the lifting points cast into
the beam. The ends of the forms were wood and attached to the steel sides with wing
nuts. The forms had significant reinforcement along the length of the sides to
counteract the bowing of the sides that would occur due to the force of the concrete.
The forms were completely oiled before each pour to ensure the forms would separate
easily from the concrete.
4-4
4.5.2. Reinforcement cages
Each cage consisted of four 20 mm diameter steel reinforced bars, two top and two
bottom. Shear ligatures were placed at locations of the greatest shear, namely the ends
of the beams and the points where the beam was to be fixed to the floor. The beams
were fixed to the floor by two 50mm nuts cast into the beam. The vertical
reinforcement bars that would form the reinforcement for the wall were welded to one
of the bottom bars and then bent to ensure that the bar would exit the top of the beam
in the middle. Figure 4.2 contains beam design diagrams.
4.5.3 Beam construction
The forms were oiled and the reinforced cages were placed in the middle, as shown in
Figure 4.3. A concrete mix was purchased from a local supplier and the concrete
strength specified was 30 MPa. A poker vibrator was used to ensure that the concrete
was adequately compacted. Figure 4.4 shows a reinforced concrete foundation beam
after concrete had been casted. Both slump and compressive tests were performed on
the concrete to ensure that it met the specifications required.
Figure 4-3 The forms and the reinforced cages
4-5
Figure 4-4 a reinforced concrete foundation beam after concrete had been casted
4.5.4 Wall construction
Once the foundation beams had been given 28 days to cure a professional bricklayer
was employed to construct the walls. Twelve walls were built as shown in Figure 4.5.
To simulate a realistic situation, solid clay bricks such as those that are commonly
used to build load-bearing walls were chosen.
Figure 4-5 Wall construction
4-6
4.6 Retrofitting details
The detailed design of two retrofitting methods, namely cable and FRP will be
discussed in this section.
4.6.1 The detailed design for the cable system
Retrofit was accomplished by adding two 10mm diameter cables (wire ropes) on only
one side of the wall face, as shown in Figure 4.6. Ideally cables should be added on
both sides of the wall to prevent an eccentric stiffness and strength distribution that
may cause twisting of the retrofitted walls and provide simultaneous retrofit against
out-of-plane failures (although this was not tested in this investigation). The details of
the cable system design are shown in Figure 4.7. The cable diameter was chosen to
ensure that the wall would fail earlier than the cable. The anchor was designed for
transferring the load from cable to foundation and the cable should fail first.
Connection details between cable system and masonry are shown in Figure 4.8 and
Figure 4.9.
Figure 4-6 Wall retrofitted using cable systems
4-7
Uee electrode E70 kel

steel plate Fy2500 kg/om2 (2SOMpa)
22mm A324 9olts
Steel Plates
Steel block
See detaile
nei
From both
(a) Front View of the wall
steel pl8te (2)

300i300*lorirm steel plete (11
steel (4)
C')
300*300*10mm steel ptete (1)

steel plote (3)
steel plate (21
(b) Bottom View of the wall
4-8
lay
(c) Side View of the wall
Figure 4-7 The detailed design of the cable system
Figure 4-8 Connection details between cable system and masonry (TOP)
4-9
Figure 4-9 Connection details between cable system and masonry (BOTTOM)
4.6.2 The detailed design of FRP
Retrofit was accomplished by adding two (width 30mm, thickness 1.2mm and tensile
strength 2800 MPa) FRP strips on only one side of the wall face, as shown in Figure
4.10. FRP should be added on both sides of the wall to prevent an eccentric stiffness
and strength distribution that may cause twisting of the retrofitted walls, and provide
simultaneous retrofit against out-of-plane failures (although this was not tested in this
investigation). Epoxy adhesive, Sikadur-30, was used to attach FRP Strips on the
masonry wall.
4-10
Figure 4-10 The test specimen retrofitted with FRP
4.7 Material properties
In order to ensure that the results could be as free from potential errors as possible all
materials used were tested to check if they possessed the properties as required. These
materials included the clay bricks, the concrete and the mortar.
Each of the three materials was tested for their compressive strengths. A standard
concrete compressive strength test was performed after the samples cured for 28 days.
An average compressive strength of 30.8 MPa was obtained from the testing.
A standard mortar compressive strength testing was also performed after the samples
were collected and cured for 28 days. The compressive strength of these samples
returned an average of 7.6 MPa.
Finally, a standard brick compressive strength testing was performed. The bricks
failed at an average of 20.3 MPa.
4-11
All the details of these testing results and material properties can be found in
Appendix A. These material properties are used in the ABAQUS numerical models
in this thesis.
4.8 Test setup
The testing equipment was designed and set up to house the wall. The testing
equipment was a quite complex. It consisted of a jig to house the wall and apply a
roof load, initially a hydraulic jack to apply the displacement, in planeload cell• to
measure the side load when displacement was applied and a series of transducer and
strain gauge to measure wall movement and micro strain in the FRP.
The Nuts had been cast into the concrete foundation to allow the wall to be bolted to
the strong floor effective fixing it in place. A series of chocks were set up between the
jig and the wall. Once bolted in, it will further ensure that there were no sideway
movements of the wall when the load was applied. A beam was set up at the top
course of brick to ensure that the wall did not move out of plane. The equivalent of an
axial pressure of O.21MPa was applied as appoint load onto a ball bearing car, onto a
beam that sat on the top of the wall, allowing the wall to move in plane as the load
was applied but maintaining the roof load. Figure 4.11 and Figure 4.12 show the test
set up.
One hydraulic jack was used to apply lateral loads to the specimen and another one
used to apply a vertical load. A steel reaction frame supported the jacks. The vertical
load was applied uniformly to the top of the wall through a stiff spreader beam. The
walls were braced by using two lateral steel beams to prevent the out-of-plane
deflection.
4-12
End steel plete
Figure 4-11 The test setup
Figure 4-8 The test setup
4-13
The load was applied via a hydraulic jack (figure 4. l3and figure 4.14) that was
support by the test jig (figure4. 15) and was attach to the wall so that the wall could be
moved at accurate increment of displacement in both directions.
Figure 4.13 Hydraulic jack
Figure 4.14 Hydraulic jack
4-14
Figure 4.15 The test jig
4.9 Loading history
Each specimen was subjected to horizontal displacement reversals while also

subjected to a constant axial compression.
In this testing program, axial loads were applied to all specimens in addition to the
lateral loads, to create a more realistic loading condition. As discussed in Chapter
3.3.3, the behaviour of the masonry walls are considerably affected by the presence of
gravity loads. For example, the rocking capacity of URM walls depends directly on
the magnitude of the applied axial load. If bearing loads are not applied on such a
masonry wall, its lateral capacity is theoretically zero (if the self weigh of the wall is
neglected). Identical axial loads were applied to all specimens of this testing program.
This simulated identical tributary floor areas, and made it convenient to compare the
results between specimens. A realistic axial load of 0.2 MPa was applied to the wall
to simulate an applied roof and second story onto the wall.
4-15
To simulate earthquake loading, a series of horizontal displacement cycles of

increasing amplitude were used on all the walls. The wall was cycled twice at each of
the incrementally increasing inelastic displacement amplitude until failure. Figure
4.16 shows the loading history.
Load history
4
E
E
0)
E
0)
0
(0
.ao
(I)
a 1 0
(0
-2
N
x0
-4 .
-6
-8
Cycle No.
Figure 4.9 The loading history.
4.10 Instrumentation and data acquisition system
The specimens were well instrumented for displacement, rotation and strain
measurements. Strain gauge as shown in figure 4.17 was applied to the FRP and
cables to measure the amount of micro strain that the FRP material was experiencing
during the cycle of displacement. A computer records all the deflection of the wall
due to cycle load until failure. This micro strain data could be use to determine the
amount of force that was being taken by the FRP and cables for each configuration.
Figure 4.18 illustrates the locations of strain gauges for FRP.
4-16
Displacements were measured by Kyowa displacement transducers (see Figure 4.19)

and linear variable displacement transducers (LVDT's). Transducers are placed to
measure the displacement at various points on the wall and the foundation during the
test process. Figure 4.20 illustrates the locations of displacement transducers. Data
acquisition systems were used to record forces and deformations.
Figure 4.17 Strain gauges
4-17
II II
LI
Strain gauge
I
I
II
I II
IL
II II
II II II
Figure 4.18 The locations of strain gauges for FRP
Figure 4.19 Kyowa displacement transducers
4-18
_________________ ________
II
1.1 L..2
ii
I II II IP
I II II II I
I II II II I
I I
I II I I
II II
II II
Figure 4.20 The locations of displacement transducers
4-19
4.11 Summary
During the testing, each wall panel was subjected to a constant vertical load. The
magnitude of vertical compressive stress for each wall, together with the geometry of
the all specimens, and retrofitting methods were used are indicated in Table 4.1.
Table 4.1 Summary of masonry wall specimens
Retrofitting
Specimens Size (m) Methods Roof load
Wall 1 0.94x0.94 URM 2OKN
Wall 2 0.94x0.94 CABLE 2OKN
Wall 3 0.94x0.94 CABLE 2OKN
Wall 4 0.94x0.94 CABLE 20KN
Wall 5 0.94x0.94 FRP 2OKN
Wall 8 0.94x0.94 CABLE Non-const.

Wall 9 0.94x0.94 FRP Non-const.
Wall 10 0.94x0.94 URM Non-const.
Wall 11 0.94x0.94 URM Non-const.
Wall 14 0.94x0.94 Cable Non-const.
Wall 15 0.94x0.94 Cable Non-const.
Wall 16 O.94x0.94 FRP 2OKN
WaIl 17 0.94x0.94 Cable 2OKN
Wall 18 0.94x0.94 Cable 2OKN
Wall 19 0.94x0.94 URM 2OKN
4-20
Chapter 5: Observed behaviour and test results
CHAPTER 5
OBSERVED BEHAVIOUR AND TEST RESULTS
5.1 General
In this chapter, results and observed behaviour of tested walls are presented. Cracking loads,
ultimate loads, crack patterns, and failure modes are identified. Interpretation and discussion of
the recorded data are also presented in this chapter.
In this chapter, discussions on the general behaviour of retrofitted and unretrofitted wall
specimens, including the comparisons of predicted and experimentally measured behaviour, are
presented. The Comparisons between the retrofitted and unretrofitted wall specimens in terms
of ultimate capacity, load-deformation characteristic, stiffness, shape of hysteretic curve,
energy dissipation and ductility are also covered. The summaries of test results are presented in
Table 5.1.
5-1
Table 5.1 The summaries of test results
Retrofitting
Specimens Methods Cracking Load Ultimate Load
Wall 1 URM 14.7KN 23.8KN
Wall 2 CABLE 17.1KN 39.13KN
Wall 5 FRP 25.5KN 56.79 KN
Wall 6 FRP 20.9 KN 60.39KN
Wall 7 FRP 24.72 KN 63.51 KN
Wall 8 CABLE N/A N/A
Wall 9 FRP N/A N/A
Wall 10 URM N/A N/A
Wall 11 URM N/A 9.2KN
Wall 12 4 FRP strips N/A 28.6KN
Wall 13 FRP N/A 40.5KN
Wall 14 Cable 10.9KN 67.7KN
WaIl 15 Cable 11.2KN 62.2KN
Wall 16 FRP N/A N/A
Wall 19 URM 10.5KN 21.2KN
Note: The reason why some of the loads are not available is attached in Appendix D.
5.2 Behaviour of unretrofitted walls
The un-retrofitted walls failed in a combination of the sliding developed in one direction,
and the rigid-body rocking (with some small amount of sliding) developed in the other
direction, at an ultimate force of 23.8KN, as evidenced by the unsymmetric hysteresis
loop shown Figure 5.1.
5-2
URM
z
Ct
0
U)
Displacement (mm)
Figure 5-1 Hysteretic behaviour of Wall-i (Unretrofitted wall)
As it can be seen from the hysteretic relationship of URM wall in Figure 5.1, the
hysteretic energy dissipated in a rocking mode response is generally smaller and the
hysteretic energy dissipated in a sliding mode response is higher.
The wall exhibited relatively large deformations with minor strength increasing before
failure. Rocking and sliding could only develop as a consequence of cracking along the
bed joint. Despite the low strength of this wall, which indicates a certain strength
deficiency, its sliding friction and rocking behaviour noticeably dissipates energy.
In hindsight, even though calculations prior to testing predicted that rocking would
develop at a lateral load of lOkN, assuming cracking along the base of the wall, cracking
above the base joint should not be surprising; the base joint was fully mortared in
compliance with construction requirements and was thus stronger, forcing the crack to
occur in the weaker joints above. However, the reason cracking occurred above the third
course of masonry rather than the first is due to the metal bracket to prevent the wall from
moving.
5-3
The lateral load necessary to produce rocking at the crack level can be calculated using
the diagram of Figure 5.3 and the rocking equations suggested by Magenes et. al. (1998).
The maximum horizontal shear which can be resisted by a rocking failure under static in-
plane loading may be approximated introducing a proper stress distribution for the
masonry in compression and neglecting the tensile strength of bed joints. With reference
to Figure 5.3, equilibrium leads to the following expression:
H0 is the effective wall height (distance from zero moment)

(see Figure 5.2), t the wall thickness, p=PI (Dt) is the mean vertical stress on the wall due
to the axial load P, is the compressive strength of masonry, ids a coefficient which
takes into account the vertical stress distribution at the compressed toe (a common
assumption is an equivalent rectangular stress block with K =0.85). The effective
height H0 is determined by the boundary conditions of the wall and is related to the shear
ratiooç:
MH ço'H
vVD D D
Considering typical test layout, the parameter assumes a value of 1 when the wall is
fixed on one end and free to rotate on the other.
Therefore, the lateral load resistance in the rocking direction is
P 20000
p=—=
Dt 940x110
=0.2MPa
—
M — H0 — ço' H — lx(94O — 2x76 —10) —
—0 8
D— 940 940
5-4
r
H02L
= 940x110 — 0.2
)= 12.8kN (Experimental result 18.15KN)
0.8 2 0.85x16.5)
ii
Lii J
I II
I
I II II
I II
I I
Figure 5-2 The effective height H0
5-5
IN-PLANE SEISMIC RESPONSE OF TRICK MASONRY WALLS
v
—
KX>
D/2 D/2
Figure 5-3 Assumptions for rocking strength evaluation of a wall falling with crushing at
the base corner (Magenes et. al. 1998)
The strength of a wall undergoing sliding, under seismic excitation, along a horizontal
joint is sometimes expressed as:
=,iP
where J.1. represents the sliding coefficient of friction of the masonry joint, and cohesion is
neglected invoking the fact the joint is already cracked in tension due to flexure.
And the lateral load resistance in the sliding direction is
V5 = O.8x20 = l6kN (Experimental result 23.8KN)
where ji = 0.8 (adopt from Magenes et. al. 1998)
These calculated forces are less than the experimentally obtained values for lateral
loading in the sliding and rocking directions. However, the reasons for the higher
experimental values maybe because of the different levels of axial load and different
boundary conditions, different aspect ratios of the walls for the experiment.
5-6
5.3 Behaviour of retrofitted walls
In general, all retrofitted wall specimens exhibited superior behaviour when compared
with those unretrofitted wall specimens. The effective of the cable system and FRP
retrofit solutions is demonstrated in this section.
The retrofitted walls eventually tended to slide or rock in all cases as the lateral load
gradually increased. However, this sliding or rocking would only appear after cracking of
the masonry, at or near the base, parallel to the direction of applied load. After a
horizontal crack had formed above the second row of blocks in Wall 1, the unretrofitted
wall would have rotated as rigid bodies. However, presence of the cable system or FRP
prevented development of this rigid body rotation due to spreading cracks more evenly
over the entire wall.
5.3.1 Retrofitted walls with cable system
The testing results of three out of eight walls are discussed in this section. The load-
deflection envelopes for three walls are compared with the un-retrofitted wall in Figure
5.4 and the comparison between the ultimate lateral load resistance of the unretrofitted
wall and retrofitted walls is presented in Table 5.2. It can be seen from the envelope and
the table, the dramatic increase of the ultimate strength for walls retrofitted with cable is
evident. The improvement of the ultimate lateral load resistance of the retrofitted walls
with cable system is about twice the capacity of unretrofitted wall.
5-7
Table 5-2 The summary of test results for the retrofitted walls by cable system
The
Retrofitting The Cracking ultimate
Specimens Methods Load Load Failure mode
Wall 1 URM 14.7KN 23.8KN Rocking and sliding
Rocking and sliding +

Wall 2 CABLE 17.1KN 39.13 KN diagonal cracking
Wall 3 CABLE 13.7 KN 52.3SKN diagonal cracking
Wall 4 CABLE 14.9 KN 47.72 KN diagonal cracking
60
50
z
0
a)
20
10
0
0 5 10 15 20
Displacement (mm)
Figure 5-4 The load-deflection envelopes of wall retrofitted with cable
5-8
Variation in stiffness with lateral drift is plotted in Figure 5.5 for wall 4 and wall 1. The
stiffness at every drift level was calculated using the two-peak value of each cycle. Only
the first cycle of each drift was considered. The stiffness of each cycle was computed
using the peak-to-peak method. The curves indicated the sensitivity of the stiffness of the
specimen with respect to the top horizontal displacement. As can be seen from the figure
presented, stiffness of retrofitted walls improved compared to the unretrofitted wall.
Variation in Stiffness for all the walls
U)
a)
15
10
5
0- I I I I I
0 2 4 6 8 10 12 14 16
Displacement (mm)
Figure 5-5 Variation in stiffness with lateral drift for wall 4 and wall 1
5.3.2 Retrofitted walls with FRP
The testing results of three out of seven walls retrofitted with FRP are discussed in this
section. The summaries of test results are presented in Table 5.3. The load-deflection
envelopes for three walls are compared with the un-retrofitted wall in Figure 5.6. It can
be seen from the envelope, the dramatic increase of the ultimate strength for walls
retrofitted with FRP is evident. The improvement of the ultimate lateral load resistance of
•
the retrofitted walls with two FRP strips strengthening is respectively three times the
capacity of un-reinforced wall.
5-9
Table 5-3 The summary of test results for the retrofitted walls by FRP
The The
Retrofitting Cracking ultimate
Specimens Methods Load Load Failure mode
Wall 1 URM 14.7KN 23.8KN Rocking and sliding

Wall 5 FRP 25.5KN 56.79 KN diagonal cracking
Rocking and sliding ÷
Wall 6 FRP 20.9 KN 60.39KN diagonal cracking
Wall 7 FRP 24.72 KN 63.51 KN diagonal cracking
From Figure 5.6, it shows that URM wall is strongly nonlinear at low level of load due to
the low tensile strength of bed and head joints. It also shows that URM wall decreased
both strength and stiffness as the damage due to cracking increases. However, the URM
walls retrofitted by the FRP system slowed down the cracking propagation and increased
both strength and ductility.
5-10
Wall (FRP)
70
60
z
V 40
0
-J 30
ci)
V
20
10
0
0 10 20 30 40
Displacement (mm)
Figure 5-6 The load-deflection envelops of four walls
5.4 Comparison of hysteretic behaviour
The amount of hysteretic energy per cycle dissipated by a specimen at a specific drift
level is equal to the area enclosed by a hysteresis ioop of the lateral load versus top
displacement curve. The hysteretic energy dissipated during the first cycle of each drift
level is as a function of lateral drift, for all walls. This makes it possible to compare the
relative energy dissipation capability of the retrofitted and unretrofitted walls.
5.4.1 Retrofitted walls with cable system
As it can be seen from the hysteretic relationship of URM wall in Figure 5.1, the
hysteretic energy dissipated in a rocking mode response is generally smaller and the
hysteretic energy dissipated in a sliding mode response is higher. The hysteretic
relationships of retrofitted walls are shown in Figures 5.7, 5.8, and 5.9, respectively. It
can be seen from these figures, the hysteretic energy dissipations are much higher than
URM
5-11
wall because the effectiveness of cable system in distributing the cracks over the entire
wall increased the whole wall energy dissipation capacity.
Wall 2(CABLE)
H.
Displacem ent(m m)
Figure 5-7 Hysteretic behaviour of Wall-2 (wall retrofitted with cable)
wall 3(CABLE)
/
E -1 () J
1'
I,,,
0)
Displacement(m m)
5-12
WaIl 4 (CABLE)
z
0 I I
10 15 20 25—C
Cl)
Displacement (mm)
The above figures indicate that the retrofitted URM walls exhibit good strength, ductility,
stiffness, and dissipation of energy compared with the unretrofitted wall.
The hysteretic relationships of wall specimens retrofitted with FRP are shown in Figures
5.10, 5.11, and 5.12, respectively. They indicate that the retrofitted URM walls exhibit
approximately symmetrical stable hysteretic behaviour with significant increase in
ductility, stiffness, and dissipation of energy.
5-13
Wall 5 (FRP)
0
-J
ci)
C,)
Displacement (mm)
Figure 5-10 Hysteretic behaviour of Wall-5 (wall retrofitted with FRP)
Wall 6 (FRP)
z
-a
c'j
0
-J
a)
D
C/)
Displacement (mm)
5-14
WaIl 7 (FRP)
D
0
-J
ci)
-o
Cl)
Displacement (mm)
All wall specimens retrofitted using the proposed FRP system exhibited superior
behaviour when compared with the un-retrofitted wall specimen. The above figures
indicate that the retrofitted URM walls exhibit good strength, ductility, stiffness, and
dissipation of energy compared with the un-retrofitted wall. From the hysteretic
relationship of URM wall, it shows that the wall behaved in a combination of rocking and
sliding as evidenced by the unsymmetrical hysteresis loops. It also shows that the flexural
response was occurred because very large displacements were obtained without
significant loss in strength. From the hysteretic relationship of retrofitted walls, they
show that the walls behaved in diagonal shear cracking response as evidenced by the loss
in strength at post-peak response of hysteresis ioops.
These figures also indicate the hysteretic energy dissipations are much higher than URM
wall because the effectiveness of FRP system in distributing the cracks over the entire
wall increased the whole wall energy dissipation.
5-15
5.5 Behaviour of retrofitting materials

5.5.1 Cable system
The forces carried by cable and the forces carried by the whole retrofitted wall are shown
in Figures 5.13, 5.14, and 5.15, respectively. They indicate that the cable carried about
the 50% of the force acted on the whole wall after major cracks occurred. In these figures,
they also show that the cables start to carry a small amount of the force at the very
beginning stage. The results proved that that URM wall retrofitted by cable system can
effectively increase the load capacity after the onset cracking of the wall.
2 - Whole wall force V.S. Cable force
45
40
35
z 30
a)
25
0 5 10 15 20
Displacement (mm)
Figure 5-13 Comparison of forces carried by the cable and the whole wall (Wall 2)
5-16
60
50
40
z
0 30
-J
U)
U)
20
10
0
0 2 4 6 8 10 12 14 16 18 20
Displacement (mm)
Wall 4-cable
50
40
z
30
-j0
2 10 12 14 16
Displacement (mm)
5-17
The forces carried by FRP and the forces carried by the whole retrofitted wall are shown
in Figures 5.16, 5.17 and 5.18, respectively. They indicate that the FRPcarried about the
50% of force acted on the whole wall even after early loading stage, which is different
from the cable system. This is because FRP is fully attached to the wall. The results
proved that the URM wall retrofitted by the FRP system becomes a whole new structure
to carry the force together.
Wall 5 (FRP)
z
V
0 — Force on whole wall
-J — Force on FRP
w
V
Cl)
Displacement (mm)
5-18
Wall 6 (FRP)
z
— Force on whole wall
-j0 —Force on FRP
0
(I)
0 2 4 5 8 10 12 14 15 18 20
Displacement (mm)
Wall 7 (FRP)
-5 0 5 10 15 20 25 30 35
Displacement (mm)
5-19
5.6 Crack pattern of wall specimens
For the ultimate failure patterns of all tested walls, refer to Appendix B. The following
series of photograph shows typical examples of the failure modes of the test specimens.
5.6.1 WALL 1 (UN-RETROFITTED WALL)
Figure 5.19 shows the initial cracking occurred at a displacement of 0.5mm along the
mortar joint near the steel plate restraint.
• .
•- ...., —
Figure 5-19 Initial cracking of wall 1
5-20
The brick cracks occur at the displacement of 5mm as shown in Figure 5.20 and Figure
5.21 during the test.
Figure 5-20 Crack pattern. of brick of wall 1 (1) (Displacement= 5 mm)
Figure 5-21 Crack pattern of brick of wall 1 (2) (Displacement= 5 mm)
5-21
Figure 5.22 shows the final failure pattern of URM wall. It was due to a combination of
rocking above the steel plate restraint and then sliding along the mortar joint that were
above the lateral restraint on the left and the right hand side of the wall.
Figure 5-22 Failure mode of URM wall
5-22
5.6.2 Wall 2 (wall retrofitted with Cable system)
The crack of bricks occurs at the displacement of 14mm as shown in Figure 5.24.
Figure 5-24 Crack pattern of brick of wall 2 (Displacement= 14 mm)
5-23
The in-plane diagonal shear cracks occur at the displacement of 14mm during the test as
shown in Figure 5.25.
Figure 5-25 In-plane diagonal shear cracking of wall 2
Figures 5.26 and 5.27 show the final failure pattern of wall retrofitted with cable. The
wall behaved in a combination of rocking, sliding and diagonal cracking. Generally, the
presence of the cable system prevented development of the rigid body rotation observed
in the unretrofitted wall and allowed cracks to spread more eventually over the entire wall.
The retrofitted wall obviously increased the load canying capacity and changed the load
paths and cracking patterns. In the retrofitted walls, they show the effectiveness of cable
system in distributing the cracks over the entire wall as opposed to cracks concentrating
only at certain locations for URM walls.
5-24
Figure 5-26 The failure mode of wall 2 (Back side)
—
-V
- i -
--
:1
I..
-4
4
*;
-J L'
I q —
—--— -
--— —'—. —-V
* -j
Figure 5-27 The failure mode of wall 2 (Front Side)
5-25
5.6.3 Wall 3 (wall retrofitted with cable)
..-/i
U
L
0 0
I.
Figure 5-28 Initial cracking of waIl 3
The crack of bricks occurs at the displacement of 9mm as shown in Figure 5.29.
1'
Ic-a
j.,.
Figure 5-29 Crack pattern of brick of wall 3 (Displacement= 9 mm)
5-26
shown in Figure 5.30 and Figure 5.31.
Figure 5-30 In-plane diagonal shear cracking of wall 3 (Front Side)
Figure 5-31 In-plane diagonal shear cracking of wall 3 (Back side)
5-27
The rocking failure was also observed at the final stage after diagonal cracking as
illustrated in Figure 5.32
Figure 5-32 Rocking failure of wall 3
The lift-up crack opened wider and wider after the rocking failure occurred as illustrated
in Figure 5.33.
Figure 5-33 Lift-up crack of wall 3
5-28
_______ _________________
Figure 5.34 and 5.35 show the final failure pattern for test specimen, behaved in a
combination of rocking, sliding and diagonal cracking.
I
•
• —. f— ——Orr'— - if
--e---W' -;
-
0
—'---
-
—
1
tu
Figure 5-35 The failure mode of wall 3 (Back side)
5-29
5.6.4 Wall 5 (wall retrofitted with FRP)
Figure 5.36 shows that the FRP debonding as the ultimate value of displacement during
the test and Figure 5.37 shows the cracking bricks at the rear of the wall.
Figure 5-36 FRP debonding of wall 5
Figure 5-37 Cracking of Brick at rear of the wall 5
5-30
Figures 5.39 and 5.40 show the final failure pattern for test specimen, behaved in a
Figure 5-39 The failure mode of waIl 5 (Back Side)
5-31
Figure 5.41 sbows the crack initiating at the first cycle of 0.7mm of displacement on the
front of the wall.
5-32
The crack of bricks occurs at the displacement of I 3mm as shown in Figure 5.42 during
the test.
Figure 5-42 Crack pattern of brick of wall 6 (Displacement = 13mm)
The Figure 5.43 shows that the FRP debonding as the ultimate value of displacement
during the test.
5-33
Figures 5.45 and 5.46 show the final failure crack pattern for test specimen, behaved in a
Figure 5-45 The failure mode of wall 6 (Back Side)
5-34
if
-
Figure 5-46 The failure mode of waIl 6 (Front Side)
5-35
Figure 5.47 shows the initial cracking occurred at a displacement of 0.8mm.
Figures 5.48 and 5.49 show that the crack of bricks occurs during the testing and the FRP
debonding at the middle of the wall.
Figure 5-48 Crack pattern of brick of wall 7
5-36
5-37
The Failure mode for the test wall 7 is quite similar to other walls that have been
discussed, behaved in a combination of rocking; sliding and diagonal cracking. Figures
5.51 and 5.52 show that the wall fail in plane diagonal shear at the ultimate value of
displacement during the test.
ii—
iF
-: II. -
-
J[
-_'
U
.7'
C)
>/ 1
Figure 5-52 The failure mode of wall 7 (back Side)
5-38
5.7 Comparison between two retrofitting techniques
The summaries of test results are presented in Table 5.1. All the test wall specimens had
the same height-to-thickness ratio, and under the same constant axial compression. As it
is shown in the Table, the cable retrofitting technique can not prevent the wall from
cracking, but it increases more than twice times of the ultimate load capacity of the wall
and ductility. The improvement of the ultimate lateral load resistance of the walls
retrofitted with FRP strips is about three times the capacity of un-retrofitted wall. The
cracking loads of the walls retrofitted with FRP were also increased compared to URM
wall.
The testing results indicate although both retrofitting techniques could significantly
enhance the ductility, stiffness, dissipation of energy and the ultimate load carrying
capacity of URM walls, FRP retrofitting technique is more effective than cable to
improve the ultimate load capacity of the URM walls. However, due to high cost and no
complete knowledge of their mechanical behaviour (there are some concerns that the
inelastic deformation capacity of the wall may be limited as a result of de-bonding of the
FRP sheets which occurs suddenly with little ductility) (Abrams 1998, Ehsani et al.
1999), there are some limitations to apply FRP as a seismic retrofitting method.
Therefore, the cable retrofitting technique could provide an alternative way for seismic
retrofitting of URM walls. The cable retrofitting technique also has advantages of
aesthetic appearance of the wall and low cost. However, drilling and anchoring to brick
must be careful to construct as most brick is hard and brittle.
5-39
5.8 Discussion and Conclusions
Experiments conducted in this study show that both FRP and cable system retrofitted to
low-rise masonry walls are effective in significantly increasing their in-plane strength,
ductility, and energy dissipation capacity. The improvement of the ultimate lateral load
resistance of the retrofitted walls is 2 and 3 times over the capacity of unreinforced wall
for cable and FRP retrofitting techniques respectively. Seismic retrofitting of
unreinforced masonry walls with FRP or cable system proved to be an effective and
reliable strengthening alternative. However as discussed earlier, each retrofitting
technique has its own advantages and disadvantages. Although FRP is more effective in
increasing the ultimate load carrying capacity of URM walls than cable, the high cost
may be a major concern. The initial cost of FRP materials is about 10 times more than
steel (ASBI, 2000). Another concern for FRP is that many practising engineers have not
obtained enough knowledge on FRP; especially their long-term behaviour requires to be
better understood. On the other hand, although cable system could be a low cost
alternative, the fragility of the masonry does not allow a simple fixing of the cable
system.
Note that, although no out-of-plane tests were conducted within the scope of this research,
it's believed that it is preferable to use the proposed cable or FRP system on both sides of
the wall, to provide greater out-of-plane strength and minimize out-of-plane
displacements. Seismic retrofitting of unreinforced masonry walls with cable system or
FRP proved to be an effective and reliable strengthening alternative.
5-40
Chapter 6: Numerical modelling strategies for unreinforced masonry walls
CHAPTER 6
NUMERICAL MODELLING STRATEGIES FOR
UNREINFORECED MASONRY WALLS
6.1 Introduction
In this chapter, the numerical modelling strategies for unreinforced masonry walls are
reviewed and presented.
In general, engineering problems are mathematical models of physical situations.

Mathematical models are differential equations with a set of corresponding boundary
and initial conditions. The differential equations are derived by applying the
fundamental laws and principles of nature to a system or control volume. These
governing equations represent balance of mass, force, or energy. When possible, the
exact solution of these equations renders detailed behaviour of a system under a given
set of conditions. The analytical solutions are composed of two parts: (1) a
homogenous part and (2) a particular part. In any given engineering problem, there
are those parameters that influence the way in which a system behaves. There are
those parameters that provide information regarding the natural behaviour of a given
system. These parameters include properties such as modulus of elasticity, thermal
conductivity, and viscosity. Table 6.1 summarizes the physical properties that define
the natural characteristics of various problems (Moaveni, 1999).
6-1
Table 6-1 summarizes the physical properties that define the natural characteristics of
various problems (Moaveni, 1999)
Examples of Parameters That

Problem Type Characterize a System
Solid Mechanics Examples
modulus of elasticity, E
a truss
HI modvlus of elasticity, £
an elastic plate
Load
modulus of elasticity, E; second moment of

area, I
abeam
modulus of rigidity, C; polar moment of inertia

of the area,)
a shaft
On the other hand, there are parameters that produce disturbances in a system.
Examples of these parameters include external forces, moments, and temperature
difference across a medium, and pressure difference in fluid flow.
The system characteristics as shown in Table 6.1 dictate the natural behaviour of a
system, and they always appear in the homogeneous part of the solution of a
governing differential equation. In contrast, the parameters that cause the disturbances
appear in the particular solution. It is important to understand the role of these
parameters in finite element modelling in terms of their respective appearances in
stiffness or conductance matrices and load or forcing matrices. The system
6-2
characteristics will always show up in the stiffness matrix, conductance matrix, or

resistance matrix, whereas the disturbance parameters will always appear in the load
matrix.
Structural analysis methods, as shown in Figure 6.1, may be divided into two major
categories: (1) analytical (classical) and (2) numerical methods. (Nagy, 1989) The
development of analytical methods goes back to several centuries. This method may
be further subdivided into the exact (closed formed), and the approximate solution
groups. Unfortunately, analytical techniques can only be used for simple structures
which severely limits their utilization, although they serve useful purpose in teaching
strength of material and engineering mechanics• courses, Most real structures have
complex geometrical shapes, as well as loading and boundary conditions for which
numerical methods usually give better results. Numerical techniques are relatively
new, but the popularity rapidly increased with the availability of digital computers.
Numerical methods can also be further subdivided into the following groups: (a) finite
differences, (b) finite element, and (c) boundary integral techniques. (Nagy, 1989)
The finite difference method is based on replacing the governing partial differential
equations and the equations of boundary conditions by the corresponding finite
difference equations. This reduces the problem to a set of simultaneous algebraic
equations. The finite difference solution gives a point-wise approximation to the
governing equations. It requires a uniform, square mesh, thus creates difficulties in
describing irregular shaped structure.
Finite
Element
Figure 6-1 Structural analysis methods (Nagy, 1989)
6-3
The finite element technique is currently the most popular numerical method. It is also
based on solving simultaneous algebraic equations, but it gives a piece-wise
approximation to the governing equations. Both of these methods are based on
discretization of the structure, however in the finite element method different element
type can be utilized and the mesh size can be varied, thus it provides a closer
approximation of the structure boundary. The difference between FDM and FEM is
show in Figure 6.2. (Nagy, 1989)
( -- I'
—
I II
It
/
—
-TA A
—---
Finite Difference Finite Element
Model Model
Figure 6-2 The difference between FDM and FEM (Nagy, 1989)
The boundary integral technique is still more or less in a research stage, no general-
purpose codes are in existence. By contrast, many general-purpose programs based on
the finite element method exist on the market.
The primary differences between classical methods and finite elements are the way
they "view" the structure and the ensuing solution procedure. Classic methods
consider the structure as a continuum whose behavior is governed by partial or
ordinary differential equations. The finite element method considers the structure to
be an assembly of small finite-sized particles. The behaviour particles and the overall
structure are obtained by formulating a system of algebraic equations that can be
readily solved with a computer.
The finite element method (FEM) is a computer based procedure that can be used to
analyze structures and continua. FEM is based on the idea of building a complicated
6-4
object with simple blocks, or dividing a complicated object into small and
manageable pieces. (See Figure 6.3)
21
Actual Structure Finite Element Mode!
Figure 6-3 The finite element method Concept
In other words, the concept of finite element method is to replace the infinite number
of interconnected points of continuum with discrete numbers. This is accomplished
by assembling several or many simple shaped small (finite) elements in such a way
that the overall structure is approximated. This procedure is similar to putting the
jigsaw puzzle pieces together to obtain the total picture. The main difference is that
the jigsaw puzzle pieces have complex shapes, while the finite elements are made
simple.
In this study the simple truss model and the finite element method are adopted to
simulate the structural behavior. The finite element method is a powerful numerical
method for the analysis of structures, and it has been extensively used for analyzing
the behaviour of masonry structures in the last decade. The proposed numerical
models of URM walls retrofitted by cable system and FRP are generated using
ABAQUS finite element program.
The model considers both material nonlinearity and progressive cracking. It is based
on a simple yield surface with isotropic hardening and associated flow when the state
of stress is predominantly compressive, and uses damaged elasticity (smeared
cracking) to account for cracking, the onset of cracking is defined by a so-called
"crack detection surface". Due to the limitation of ABAQUS, the reduction of
stiffness caused by inelastic straining, which is very important for masonry is
6-5
neglected. Therefore, the model is not capable of predicting cyclic response; besides
the use of associated flow usually leads to over-prediction of volume strain. In spite of
these limitations, the model has proven to be able to reasonably predict the masonry
behaviour in monotonic loading or the envelope of cyclic response, as long as proper
material definition is provided. However, the main objective of this research is to
prove a new retrofitting technique with experimental testing. These ABAQUS models
are for limit-state examination purpose only.
6.2. Modelling techniques for unreinforced masonry wall
Masonry is a composite material that consists of units and mortar joints. A detailed
analysis of masonry, hereby denote micro-modeling, must then include a
representation of units, mortar and the unitlmortar interface. This approach is suited
for small structural elements with particular interest in strongly heterogeneous states
of stress and strain. The primary aim of micro-modeling is to closely represent
masonry from the knowledge of the properties of each constituent and the interface.
The necessary experimental data must be obtained from laboratory tests in the
constituents and small masonry samples. Several attempts to use interfaces for the
modeling of masonry were carried out in the last decade with reasonably simple
models, see Anthoine (1992) and Lourenco (1996) for references. In particular,
gradual softening behavior and all failure mechanisms, namely tensile, shear and
compressive failure, have not been fully considered.
In large and practice-oriented analysis the knowledge of the interaction between units
and mortar is, generally, negligible for the global structural behavior. In these cases a
different approach can be used, hereby denoted macro-modeling, where the material
is regarded as an anisotropic composite and a relation is established between average
masonry strains and average masonry stresses. This is clearly a phenomenological
approach, meaning that the material parameters must be performed in masonry tests
of sufficiently large size under homogeneous states of stress. A complete macro-
model must reproduce an orthotropic material with different tensile and compressive
strengths along the material axes as well as different inelastic behavior for each
6-6
material axes. A reduced number of orthotropic material models specific for masonry
has been proposed by many researchers, see Anthoine (1992) and Dhanasekar et al.
(1984, 1985c) for references.
Masonry is a material which exhibits distinct directional properties due to the mortar
joints which act as planes of weakness. In general, the approach towards its numerical
representation can focus. on the micro-modeling of masonry as components, such as
unit (brick, block, etc.) and mortar, or the macro-modeling of masonry as a composite,
see also Rots (1991). Depending on the level of accuracy and the simplicity desired, it
is possible to use the following modeling strategies, see Figure 6.4.
Unit (brick, block, etc)
Li Ii I
(a) (b)
"Unit" "Joint" Composite
(c) (d)
Figure 6.4 Modelling strategies for masonry structures: (a) masonry sample ;( b)
detailed micro-modelling; (c) simplified micro-modelling; (d) macro-modelling
(Lourenco, 1996)
• Detailed micro-modeling — units and mortar in the joints are represented by

continuum elements whereas the unit-mortar interface is represented by
discontinuous elements;
• Simplified micro-modeling - expanded units are represented by continuum
elements whereas the behavior of the mortar joints and unit-mortar interface is
lumped in discontinuous elements;
6-7
• Macro-modeling — unit, mortar and unit-mortar interface are smeared out in the
continuum.
In the first approach Detailed micro-modeling, Young's modulus, Poisson's ratio and,
optionally, inelastic properties of both unit and mortar are taken into account. The
interface represents a potential crack/slip plane with initial dummy stiffness to avoid
interpenetration of the continuum. This enables the combined action of unit, mortar
and interface to be studied under a magnifying glass. In the second approach
LI Simplified micro-modeling, each joint, consisting of mortar and the two unit-mortar
interface, is lumped into an "average" interface while the units are expanded in order
to keep the geometry unchanged. Masonry is thus considered as a set of elastic blocks
bonded by potential fracture/slip lines at the joints. Accuracy is lost since Poisson's
effect of the mortar is not included. The third approach LlMacro-modeling does not
make a distinction between individual units and joints but treats masonry as a
homogeneous anisotropic continuum. One modelling strategy cannot be preferred
over the other because different application fields exist for micro- and macro-models.
Micro-modeling studies are necessary to give a better understanding about the local
behavior of masonry structures. This type of modeling applies notably to structural
details, but also to modem building systems like those of concrete or calcium-silicate
blocks, where window and door openings often result in piers that are only a few
block units in length. These piers are likely to determine the behavior of the entire
wall and individual modeling of the blocks and joints is then to be preferred. Macro-
models are applicable when the structure is composed of solid walls with sufficiently
large dimensions so that the stresses across or along a macro-length will be essentially
uniform. Clearly, macro-modeling is more practice oriented due to the reduced time
and memory requirements as well as a user-friendly mesh generation. This type of
modeling is most valuable when a compromise between accuracy and efficiency is
needed. Therefore, the macro-modelling is adopted to simulate the masonry behavior
in this study as the major objective is to study the retrofitting effect of the wall.
6-8
6.3 Summary of the proposed numerical model
The proposed numerical model is based on ABAQUS concrete model and is modified
to simulate the special characterises of masonry. The model employs oriented
damaged elasticity (smeared cracking) and isotropic compressive plasticity to
represent the inelastic behaviour of masonry. The model is intended to simulate the
masonry behaviour for relatively monotonic loadings under fairly low confining
pressures. The model adopts a smeared crack approach in the sense that it does not
track individual "macro" cracks. Constitutive calculations are performed
independently at each integration point. The presence of cracks enters into these
calculations by the way in which the cracks affect the stress and material stiffness
associated with the integration point. The anisotropy introduced by cracking is
assumed to be important in the simulations for which the model is intended, so the
model includes consideration of this anisotropy.
Due to the limitations of ABAQUS, the model makes no attempt to include prediction
of cyclic response or of the reduction in• the elastic stiffness caused by inelastic
straining. Nevertheless, it is likely that--even in such cases--the stress .trajectories will
not be entirely radial and the model must predict the response in such cases in a
reasonable way. An isotropically hardening "compressive" yield surface forms the
basis of the model for the inelastic response when the principal stresses are
dominantly compressive. In tension once cracking is defined to occur (by the "crack
detection surface" of the model), the orientation of the cracks is stored and oriented,
damaged elasticity is then used to model the existing cracks. Stress components
associated with an open crack are not included in the definition of the crack detection
surface for detecting additional cracks at the same point, and it only allowed cracks to
form in orthogonal directions at a point.
Since ABAQUS is an implicit, stiffness method code and the material calculations
used to define the behaviour of the masonry are carried out independently at each
integration point in that part of the model that is made of masonry, the solution is
known at the start of the time increment. The constitutive calculations must provide
values of stress and material stiffness at the end of the increment, based on the current
6-9
estimate of the kinematic solution for the response at the spatial integration point
during the increment that provides the (logarithmic) strain at the end of the increment.
Once cracks exist at a point, the component forms of all vector and tensor valued
quantities are rotated so that they lie in the local system defined by the crack
orientation vectors (the normals to the crack faces). The model ensures that these
crack face normal vectors will be orthogonal, so that this local system is rectangular
This use of a local system simplifies the computation of the damaged
elasticity used for the components associated with existing cracks.
The model, thus, consists of a "compressive" yield/flow surface to model the masonry
response in predominantly compressive states of stress, together with damaged
elasticity to represent cracks that have occurred at a material calculation point, the
occurrence of cracks being defined by a "crack detection" failure surface that is
considered to be part of the elasticity. The details of the model are discussed in the
following sections.
6.4 Constitutive law of masonry
A constitutive model is a mathematical description of material behaviour. There are

two major aspects to develop an accurate analytical model. One is to understand the
behavior of masonry which is the constitutive relations of the material. And the other
is the failure criteria of the material because the major nonlinear effect of URM under
in-plane lateral load is due to progressive cracking.
Most masonry walls subjected to in-plane loads are in a state of biaxial stress.
Therefore, it is necessary to consider a biaxial stress-strain model for masonry subject
to in-plane loading. The state of stress of masonry is shown in Figure 6.5.
The stress-strain relations of masonry depend on the state of stress. The typical
unaxial stress-strain curve for masonry under compression is shown in Figure 6.6.
6-10
Lln on
Figure 6.5 The state of stress of masonry (Zhuge, 1996)
a.
(÷) clu (-)
o-It
(+)
Figure 6.6 The typical unaxial stress-strain curve for masonry under compression
(Zhuge, 1996)
6-11
6.4.1 Properties of masonry
Masonry is a composite material. The uniaxial behavior of the material is described

in this section with regard to the material axes, namely the directions parallel and
normal to the bed joints.
6.4.1.1 Uniaxial compressive behavior of masonry
The compressive strength of masonry in the direction normal to the bed joints has
been traditionally regarded as the sole relevant structural material property, at least
until the recent introduction of numerical methods for masonry structures. A test
frequently used to obtain this uniaxial compressive strength is the stacked bond
prism, see Figure 6.7 (a), but it is still somewhat unclear what are the consequences
in the masonry strength of using this type of specimens, according to Mann and
Betzler (1994). It is commonly accepted that the real uniaxial compressive strength of
masonry in the direction normal to the bed joints can be obtained from the so-called
RILEM test, see Wesche and Ilanzis (1980), shown in Figure 6.7 (b). The RILEM
specimen is however relatively large and costly to execute, especially when
compared to the standard cube or cylinder tests for concrete. Since the pioneering
work of Hilsdorf (1969) it has been accepted by the masonry community that the
difference in elastic properties of the unit and mortar is the precursor of failure.
Uniaxial compression of masonry leads to a state of triaxial compression in the
mortar and of compression/biaxial tension in the unit. Mann and Betzler (1994)
observed that, initially, vertical cracks appear in the units along the middle line of the
specimen, i.e. continuing a vertical joint. Upon increasing deformation additional
cracks appear, normally vertical cracks at the small side of the specimen, that lead to
failure by splitting of the prism. Examples of load-displacement diagrams obtained in
500 X 250 X 600 [mm3] prisms of solid soft mud bricks are shown in Figure 6.7 (c).
Increasing strength leads to a more brittle behavior. The quite high value of the
compressive fracture energy G which equals approximately 45 [Nmm/mm2 can
6-12
be explained by the continuum fracture energy and the 600 [mm] height of the
specimen, see Vonk (1992). Atkinson and Yan (1990) collected information about
the compressive stress-strain relation for masonry, but these results are not energy
based.
(a)
hj
h 5/tb ,,'iii1 32
ii4 b
fl_c) ----i
b -. 0.0 4.0 6.0 S.0 10.0
(,nrnj
(b) (c)
Figure 6.7 Uniaxial compressive behaviour of masonry upon loading normal to the
bed joints: (a) stacked bond prism; (b) schematic representation of RILEM test
specimen; (c) typical experimental stress-displacement diagrams for 500 X 250 X 600
[mm3] prisms of solid soft mud bricks, Binda et al. (1988). Here, is the mortar
compressive strength.
Uniaxial compression tests in the direction parallel to the bed joints have received
substantially less attention from the masonry community. However, masonry is an
anisotropic material and, particularly in the case of low longitudinal compressive
strength of the units due to high or unfavorable perforation, the resistance to
compressive loads parallel to the bed joints can have a decisive effect on the load
bearing capacity. According to Hoffmann and Schubert (1994), the ratio between the
uniaxial compressive strength parallel and normal to the bed joints ranges from 0.2 to
0.8. These ratios were obtained for masonry samples of solid and perforated clay
6-13
units, calcium silicate units, lightweight concrete units and aerated concrete units.
The current model adopts the uniaxial compressive strength normal to the bed joints.
6.4.1.2 Uniaxial tensile behavior of masonry
For tensile loading perpendicular to the bed joints, failure is generally caused by
failure of the relatively low tensile bond strength between the bed joint and the unit.
As a rough approximation, the masonry tensile strength can be equated to the tensile
bond strength between the joint and the unit.
In masonry with low strength units and greater tensile bond strength between the bed
joint and the unit, e.g. high-strength mortar and units with numerous small
perforations, which produce a dowel effect, failure may occur as a result of stresses
exceeding the unit tensile strength. As a rough approximation, the masonry tensile
strength in this case can be equated to the tensile strength of the unit.
For tensile loading parallel to the bed joints a complete test program was set-up by
Backes (1985). The specimen consists of four courses, initially laid down in the usual
manner; see Figure 6.8 (a). A special device attached to the specimen turns it 90° in
the intended direction of testing shortly before the test time, see Figure 6.8 (b). The
load is applied via steel plates attached to the top and bottom of the specimen by
special glue (unspecified in the paper). The entire load-displacement diagram is
traced upon displacement control. Two different types of failure are possible;
depending on the relative strength of joints and units, see Figure 6.9. In the first type
of failure cracks zigzag through head upon increasing deformation. The post-peak
response of the specimen is governed by the fracture energy of the head joints and the
post —peak mode II behavior of bed joints. In the second type of failure cracks run
almost vertically through the units and head joints. A typical stress-displacement
diagram shows progressive softening until zero. The post-peak response is governed
by the fracture energy of the units and head joints. The current model adopts the
uniaxial tensile strength normal to the bed joints.
6-14
(h)
Figure 6.8 Test set-up for tensile strength of masonry parallel to the bed joints,
(Backes, 1985): (a) building the test specimen; (b) test specimen before 90°rotation
and testing.
0,
061)
I fnun)
Figure 6.9 Typical experimental stress-displacement diagrams for tension in the

direction parallel to the bed joints, (Backes, 1985): (a) failure occurs with a stepped
crack through head and bed joints; (b) failure occurs vertically through head joints
and units.
6-15
6.4.1.3 Biaxial behavior
The constitutive behavior of masonry under biaxial states of stress cannot be

completely described from the constitutive behavior under uniaxial loading
conditions. The influence of the biaxial stress state has been investigated by Page
(1981, 1983) up to peak stress to provide a biaxial strength envelope, which cannot
be described solely in terms of principal stresses because masonry is an anisotropic
material. Therefore, the biaxial strength envelope of masonry must be either
described in terms of the full stress vector in a fixed set of material axes or, in terms
of principal stresses and the rotation angle 9 between the principal stresses and the
material axes. The most complete sets of experimental data of masonry subjected to
proportional biaxial loading is shown in Figure 6.10, (Page, 1981, 1983). The tests
were carried out with half scale solid clay units. Both the orientation of the principal
stresses with regard to the material axes and the principal stress ratio considerably
influence the failure mode and strength.
MatesMi
Figure 6.10 Biaxial strength of solid clay units masonry,

(Page, 1981, 1983).
6-16
It is further noted that the strength envelope shown in Figure 6.10 is of limited
applicability for other types of masonry. Different strength envelops and different
failure modes are likely to be found for different materials, unit shapes and geometry.
A comprehensive program to characterize using full scale specimens, see Ganz and
Thurlimann (1982) for hollow clay units masonry, Guggisberg and Thurlimann
(1987) for clay and calcium-silicate units masonry and Lurati et al. (1990) for
concrete units masonry.
6.4.2 The constitutive law adopted by the current model
When the principal stress components are dominantly compressive, the response of
the masonry is modelled by an elastic-plastic theory, using a simple form of yield
surface written in terms of the first two stress invariants. Associated flow and
isotropic hardening are used. The cracking and compression responses of masonry
that are incorporated in the model are illustrated by the equivalent uniaxial response
of a specimen shown in Figure 6.11. The biaxial effect has been considered in the
model. The uniaxial compressive strength will be discussed in Section 6.4.2.2.
= In
Ineia*tlc
Figure 6.11 Uniaxial behaviour of plain masonry
6-17
6.4.2.1 Strain rate decomposition
The current model uses the classical concept of plasticity theory: a strain rate
decomposition into elastic and inelastic strain rates,
We begin with a strain rate decomposition:
de = ded + (6-1)
where de is the total mechanical strain rate, dee! is the elastic strain rate (which
includes crack detection strains--this elastic strain will be further decomposed when
we describe the elasticity), and is the plastic strain rate associated with the
"compression" surface.
We assume that the elastic part of the strain is always small, so this equation can be
integrated as
S= + (6-2)
6.4.2.2 Plastic Strain rate
The model uses associated flow, so if =0 and > o,
(6-3)
otherwise, =0.
In the definition of dec, c0 is a constant that is chosen so that the ratio of ef1' in a
monotonically loaded biaxial compression tests to in a monotonically loaded
6-18
uniaxial compression test is a value specified as part of the failure surface data.
The equation defining c0 from and the other constants in the compression surface
is derived next.
The gradient of the flow potential for the compressive surface is
(6-4)
Since
ac7 3
and
— 3 S
— 2 q
then
In uniaxial compression p = q , where is the stress magnitude
and S11 = — so Equation 6-3 defines
(de:' = + — (6-5)
This equation can be integrated immediately to give
(e:' = + — iJ (6-6)
so is known from and the constants a0 and c0.
6-19
The constant c0 is calculated from the definition of the ratio of to
the total plastic strain components that would occur in monotonically loaded biaxial
and uniaxial compression tests. In biaxial compression when both nonzero principal
stresses have the magnitude abc P = abc = q = 0bc = , and
1
S11 so the flow rule gives
(6-7)
Using this equation and Equation 6-5 then defines from and the other constants as
—
2 (-)
The constant a0 is chosen from the ultimate stress reached in biaxial compression to
the ultimate stress reached in uniaxial compression. The derivation of a0 can be found
in Section 6.5.2.
6.4.3 Tension softening

Softening is a gradual decrease of mechanical resistance under a continuous increase
of deformation forced upon a material specimen or structure. It is a salient feature of
quasi-brittle materials like clay brick, mortar, ceramics, rock or concrete, which fail
due to a process of progressive internal crack growth. Such mechanical behavior is
commonly attributed to the heterogeneity of the material, due to the presence of
different phases and material defects, like flaws and voids. Even prior to loading,
mortar contains micro-cracks due to the shrinkage during curing and the presence of
the aggregate. The clay brick contains inclusions and micro-cracks due to the
shrinkage during the burning process. The initial stresses and cracks as well as
variations of internal stiffness and strength cause progressive crack growth when the
material is subjected to progressive deformation. Initially, the micro cracks are stable
6-20
which means they grow only when the load is increased. Around peak load an
acceleration of crack formation takes place and formation of macro crack starts. The
macro cracks are unstable, which means that the load has to decrease to avoid an
uncontrolled growth. In a deformation controlled test the macro crack growth results
in softening and localization of cracking in a small zone while the rest of the
specimen unloads.
For tensile failure this phenomenon has been well identified, see e.g. Hordijk (1991).
For shear failure, a softening process is also observed as degradation of the cohesion
in Coulomb friction models. For compressive failure, softening behavior is highly
dependent upon the boundary conditions in the experiments and the size of the
specimen, Van Mier (1984) and Vonk (1992). Experimental concrete data provided
by Vonk (1992) indicated that the behavior in uniaxial compression is governed by
both local and continuum fracturing processes.
Figure 6.12 shows some classical models for the post-peak tensile behaviour, where
is the uniaxial tensile strength and cr is the strain at the stress level of
c ,
(a) Elastic - perfectly (b) Intermediate (linear) (C) discontinuous.

brittle (brittle-linear)
(d) Bilinear (e) Dugdale (f) Exponential

(plastic-softening)
Figure 6.12 Post-peak tensile stress-strain models (Born and Sorace, 1993)
6-21
Model (e) is commonly used to analyze ductile material and would not be sufficiently
accurate in representing the true material and would not be sufficiently accurate in
representing the true material behaviour of masonry.
Models (a) and (b) are the most commonly implemented methods in the analysis of
brittle material. Model (a) is called the "tension cut-off model", represents a brittle
failure, consisting of a total stress released in the direction orthogonal to the crack.
However, the adoption of this model causes difficulties in achieving solution

convergence. In fact, due to the crack-surface interlocking of the material, the
stresses do not actually drop to zero immediately at the onset of cracking. The
stresses would drop a certain amount, then to be released gradually. Therefore,
Models (c), and (d) provide a more reliable description of this behaviour.
Research has shown that the tensile softening effects have generally a limited effect
on the global response of plain concrete and masonry structures (Rots, 1998; Rots
and Blauwendraad, 1989). At the same time, the presence of softening mechanical
behaviour is very burdensome for computation. For tension softening, the bilinear
model (d) is adopted in the current model.
6.4.4 Damaged elasticity
Following crack detection the damaged elasticity could be used to model the failed
material. The elasticity is written in the form
(6-9)
where D is the elastic stiffness matrix for the masonry.
Let a represent a cracked direction, with corresponding direct stress a— and direct
elastic strain . In these expressions and in the remainder of this section, no
6-22
summation is implied by repeated indices with a bar over them. If the fracture energy
concept is used, the strains are related to the user-specified stress/displacement
definition for the tension stiffening behaviour bye = u / c, where c is the characteristic
length associated with the integration point.
Then, in D, D— is the usual elasticity of the masonry if e— 0.
ifs open >e—>0,

ar aa
open, open (6-10)

D
aaaa ax
open open
where is the stress corresponding to e (as defined in the tension
stiffening data), and
C
open
= max(C— I
(6-11)
open
(6-12)
D (6-13)
which follows from the tension stiffening data.
We also assume no Poisson effect for open cracks: for >0, D— =0 for
The shear terms in the elasticity associated with existing crack directions are
D =G,
ajkrfl
6-23
where G = G for e— 0, and G = G for e— >0. In these expressions G

aa
is the elastic shear modulus, is a constant defined by the user as part of the
shear retention data (see Figure 6.13),
I .0
Figure 6.13 Shear retention (ABAQUS, 2000)
and is a linear function of and is also defined by
the user in the shear retention data. Here where (and) are
\ax/ + \/313/ ,
Macauley brackets, defining
0
" ' otherwise
L0
for any functionf.
6-24
6.5 Failure criterion for unreinforced masonry
6.5.1 Introduction
The failure of masonry under in-plane lateral load has been investigated for many
years. Unlike other materials (concrete or steel), masonry is a non-homogeneous
material. Therefore, it is not possible to attribute failure of masonry to a single cause.
Many types of failure are possible and the one that gives the lower bound is the
critical one. Furthermore, it has been extensively observed that the strength of
masonry under multiple stress conditions is a function of the state of stress and
cannot be predicted using simple tensile, compressive, and shearing stress
independently of each other. Because masonry is a material, which exhibits
directional properties due to the mortar joints acting as a plane of weakness, a
realistic failure model should consider this effect as well. Depending upon the stress
state acting on the joints, failure may occur in the joints alone, or in some form of a
combined mechanism involving the mortar and masonry unit. Therefore, to develop a
rational failure criterion, all the above factors must be taken into account.
To formulate a failure criterion for masonry, it is necessary to allow for all the
possible modes of failure. According to the experimental results from the present and
other research, the failure modes of URM under in-plane lateral loads are illustrated
in Figure 6.14.
6-25
Figure 6.14 Modes of failure of solid clay units masonry under biaxial loading,
(Dhanasekar et al. 1985c)
For uniaxial tension, failure occurred by cracking and sliding of the head and bed
joints. The influence of the lateral tensile stress on the tensile strength is not known
because no experimental results are available. A lateral compressive stress decreases
the tensile strength, which can be explained by the damage induced in the composite
material, by microslip of the joints and microcracking of the units. In the tension
compression loading cases failure occurred either by cracking and sliding of the
joints alone or in a combined mechanism involving both units and joints. Similar
types of failure occurred for uniaxial compression but a smooth transition is found to
other type of failure mode in biaxial compression. In biaxial compression failure
6-26
typically occurred by splitting of the specimen at mid-thickness, in a plane parallel to

its free surface, regardless of the orientation of the principal stresses.
For principal stress ratios<< 1 and>> 1, the orientation played a significant role and
failure occurred in a combined mechanism involving both joint failure and lateral
splitting. The increase of compressive strength under biaxial compression can be
explained by friction in the joints and internal friction in the units and mortar.
For concrete, the failure envelope seems to be largely independent of the loading
path, Nelissen(1972), which confirms the presence of single failure mode, i.e.
continuous crack growth at the microlevel. Presently, it is not known if the masonry
failure envelope obtained by Page is valid for nonproportional loading, particularly
because different failure modes can be triggered. Another point is that experimental
data about the softening of masonry under biaxial stress state.
The common method to predicting the failure of masonry is the biaxial failure
criterion (defined in terms of principal stresses). Figure 6.15 shows a typical failure
surface of masonry, but it didn't consider the effect of joint. Von Mises yield surface
is used to predict a crushing type of failure, whereas a tension cut-off surface is used
for tension-tension or tension-compression type of failure.
Tension
Compression
1
1'rn
Crushing Failure
Regime
Tensile
Regime Mises yield envelope
Failure critenon:
Concrete
— 0- 02 ÷ — =
Compression
Figure 6-15 A typical failure surface of masonry
6-27
Masonry is a material that exhibits distinct directional properties of weakness.

Therefore, its in-plane failure cannot be simply defined in terms of the principal stress,
the influence of the bed joint orientation should be also considered.
A simplified failure envelope for masonry has been developed by Zhuge (1996). This
failure envelope, which combines both biaxial and Coulomb shear failure models, is
capable of predicting joint failure through the use of the ubiquitous joint
model(Zienkiewicz and Pande, 1977).
The failure envelope has been shown in Figure 6.16 (Zhuge, 1996) with various stress
ratios = 5, 10, 15) .To illustrate the failure criteria discussed in this study, the
biaxial failure criterion (biaxial tension-compression range) in terms of the principal

stress system is transformed to direct stressed normal (ci,,) and parallel (a,,) to the
bed joint and shear stress ('t) along the bed joint. As shown in this figure, the lowest
failure curve defines the governing failure condition for any stress combination. The
uniaxial compressive strength f,71 is assumed equal to 10 . (uniaxial tensile
strength) in this envelope. It is a simple matter to modify this for other

compressive/tensile ratios. The straight line describes the shear sliding and the curves
describe the cracking type of failure.
It can be seen from this figure that when 45 0 <0 900 (a low ratio of It), failure
is controlled by the shear failure criterion [straight line], which agrees well with the
test results of Page (1982), showing that the crack developed along the bed joints due
to shear slips in this range. When the bed joint angle 0 decreased, the shear strength of
the wall increased as increased. If the shear strength exceeds the principal tensile
stress 0 <450, the failure will be controlled by the biaxial failure criterion (principal
tensile stress).
6-28
'C
-4 -8 -10 .12 -14

2 0 -2
(Compression)
Figure 6.16. Simplified failure envelopes for masonry (Zhuge, 1996)
6.5.2 Model of compression yield surface
When masonry is loaded in compression, it initially exhibits elastic response. As the

stress is increased, some nonrecoverable (inelastic) straining occurs, and the response
of the material softens. An ultimate stress is reached, after which the material softens
until it can no longer carry any stress. If the load is removed at some point after
inelastic straining has occurred, the unloading response is softer than the initial elastic
response: this effect is ignored in the model.
In multiaxial stress states the failure criteria can be generalized through the concept of
surfaces of failure and of ultimate strength in stress space. These surfaces are defined
below and are fitted to experimental data. Typical surfaces are shown in Figure 6.17.
6-29
crack detQclion eufluco
ufli*xi&I tension a
/
corn
/ / A
/ bI Ia I
/
tø iii Ion
/.
com
/
rtace
/
bIaxIal compresalon
Figure 6.17 Masonry failure surfaces in plane stress (ABAQUS, 2000)
The "compression" surface is
(6-14)
where p is the effective pressure stress, defined as
1
p= ——trace (a) , (6-15)
and q is the Mises equivalent deviatoric stress:
(6-16)
6-30
where S = a + p1 are the deviatoric stress components; a0 is a constant, which is

chosen from the ratio of the ultimate stress reached in biaxial compression to the
ultimate stress reached in uniaxial compression; and is a hardening parameter
is the size of the yield surface on the q -axis at p =0, so that is the yield stress
in a state of pure shear stress when all components of a are zero except
a12 = = The hardening is measured by the value of the relationship
is user-specified.
This simple surface is a straight line in (p - q) space (Figure 6.16) and provides a
good match to experimental data over a fairly wide range of pressure stress values (up
to four to five times the maximum compressive stress that can be carried by the
masonry in uniaxial compression). This form of the surface means that, as the
hardening changes, the surfaces in (p - q) space are similar, so--for example--
the ratio of flow stress in biaxial loading to flow stress in uniaxial loading is the same
at all flow stress levels. This does not appear to be contradicted by any experimental
data, and it means that only one constant (a0) is needed to define the shape of the
surface. -
The value of a0 is established as follows. In uniaxial compression p = and
q= where is the stress magnitude. Therefore, on =0,
(6-17)
In biaxial compression p =2/3 abC and q = abC, where abC is the magnitude of each
nonzero principal stress. Therefore, on =0,
(1 2a0
(618)
abC 3
6-31
The value of / = is specified by the user as part of the failure surface data.
a0 can be calculated from Equation 6-17 and Equation 6-18 as
(6-19)
6.5.3 Model of tensile failure
Cracking dominates the material behaviour when the state of stress is predominantly
tensile. The model uses a "crack detection" plasticity surface in stress space to
determine when cracking takes place and the orientation of the cracking. Damaged
elasticity is then used to describe the post failure behaviour of the masonry with open
cracks.
Numerically the "crack detection" plasticity model was used for the increment in
which cracking takes place and subsequently damaged elasticity was used once the
crack's presence and orientation have been detected. As a result there is at least one
increment in which the crack detection "plastic" strains were calculated. Since these
are really just the outcome of a numerical device to treat cracking, they are recast as
elastic strains in the direction of cracking and as plastic strains in the other directions.
(This means that we retain the stresses calculated for equilibrium purposes, as well as
the strain decomposition of Equation 6-1.)
The crack detection surface is the Coulomb line
f, =0 (6-20)
where is the failure stress in uniaxial tension and is a constant that is defined
from the value of the tensile failure stress, in a state of biaxial stress when the
6-32
other nonzero principal stress, , is at the uniaxial compression ultimate stress

value, o is the equivalent uniaxial tensile stress.
The stress measures p and q are defined in the same way as p and q, except that
all stress components associated with open cracks (that is, if a or /3 is a crack
direction in which the direct strain >0 or >0) are not included in these
,8,8
measures: they are invariants in subspaces of the stress space.
This surface has a simple mathematical form but matches plane stress data quite well.
The hardening is introduced in the particular form shown in Equation 6-15 so that, as
—+ 0, the surface becomes q — = 0, which in (p - q) space is the cone containing
the principal axes of stress. This means that, as the tension stiffening is exhausted in a
plane stress test, the stress point will drop back onto the nearest principal stress axis.
The value of is obtained as follows. The failure surface data include a definition of
a ratio that states that, in a plane stress test cracking would occur when one
principal stress has the value = — is the magnitude of the ultimate stress in
uniaxial compression) and the other nonzero principal stress has the value =
Another value provided as part of the failure surface data is which defines
= (6-21)
Cracking would, therefore, occur at the point with principal stresses —

and 0. For these values
= (i — (6-22)
and
6-33
(6-23)
Therefore, with =
f1 =0 (6-24)
so
(6-25)
1+ (1—f)
The crack detection model uses the assumption of associated flow, so if f, =0 and
def1 = (6-26)
otherwise, def' =0.
6.6 Masonry cracking models
The major reason for the nonlinear behaviour of masonry is progressive cracking of
the material due to the tensile strength of masonry being lower that its compressive
strength.
In the finite-element analysis of masonry structures, three different approaches have

been employed for crack modeling:
1. Smeared-cracking model
2. Discrete-cracking model
3. Fracture-mechanics model
6-34
The particular cracking model to be selected from the three alternatives depends upon
the purpose of the analysis. In general, if overall load-deflection behavior is desired,
without regard to completely realistic crack patterns and local stresses, the smeared-
crack model is probably the best choice. If detailed local behavior is of interest,
adaptations of the discrete-cracking model are useful. For the special class of
problems in which fracture mechanics is the appropriate tool, a specialized fracture
model may prove necessary. For most structural engineering applications, the
smeared-cracking modeling is generally used.
6.6.1 Smeared-cracking model
In this approach, the cracked concrete is assumed to remain a continuum; i.e., the
cracks are smeared out in a continuous fashion. It is assumed that the concrete
becomes orthotropic or transversely isotropic after the first cracking has occurred,
one of the material axes being oriented along the direction of cracking. Such
formulations easily allow for a gradual build-down or sudden drop of strength in the
direction of tension. Also, shear-strength reserves becasue aggregate interlocking can
be accounted for by retaining a positive shear modulus. This shear capacity also
means that secondary cracking does not necessarily appear perpendicular to the first
direction of cracks.
In the smeared-cracking model, a crack is not discrete but implies an infinite number
of parallel fissures across that part of the finite element (Fig 6.18). After cracking has
occurred, the cracked masonry becomes an orthotropic material and new incremental
relationships must be derived. This is accomplished by modifying the tan gent-
material-stiffness or the tangent-elasticity matrix [Di]. For instance, for plane stress
the stress-strain relationships in the cracked directions (n and t axes in Fig.6. 18)
become
Ao = [D,1 (6-27)
Ar [Ay
6-35
Where the tangent-stiffness matrix is defined by
000
[D]=o E 0 (6-28)
00/3G
And the lateral has the value
As (6-29)
E
Figure 6-18 Smeared-cracking model (Chen, 1982)
In Eq. (6-28) the modulus of elasticity E of the masonry is reduced to zero in the
direction normal to the cracked (n axis). Further, a reduced shear modulus is
assumed on the cracked plane to account for the aggregate interlocking. For stiffness
calculations it is necessary to transform the I matrix into global coordinates, and
this is accomplished by using the well-known transformation rule for stress and strain
tensors
6-36
[D, I = [TIT [D, ][TI (6-30)
The smear crack model is used in the ABAQUS concrete model. Constitutive
calculations are performed independently at each integration point of the finite
element model. The presence of cracks enters into these calculations by the way in
which the cracks affect the stress and material stiffness associated with the
integration point.
If failure has occurred, the failed material constitutive matrix [D ] has to be

modified according to the mode of failure as shown Figure 6.19. Various failure
modes and the failure criteria for URM under in-plane lateral loads have been
discussed in the preceding section. The details are described as follows:
a) Cracking Failure
Once tensile failure occurs, the stress normal to the crack is reduced to a very small
value. It is assumed that this is released, with the other stress parallel to the crack
remaining unchanged. The shear modulus is assumed to have a small residual value
to simulate frictional resistance due to aggregate interlock and this can be achieved
through an aggregate interlock factor 13'.
b) Crushing Failure
Once crushing failure occurs, all local stresses are released completely and the
material is assumed to lose all its capacity to transmit further loads. All the terms of
the constitutive matrix are reduced to approximately zero (a very small value is
adopted to avoid numerical difficulty).
6-37
Figure 6-19 [D1 } matrices for different modes of failure (Zhuge, 1996)
6-38
c) Shear failure
When shear sliding type of failure occurs, the shear stress must be reduced to that of
the allowable shear stress and not allowed to increase with further loading. To
simulate the sliding deformation upon further application of load, the shear modulus
G is reduced to a small value by multiplying it by the aggregate interlock factor
Stiffness in other directions is kept unaltered (see Figure 6.19).
6.6.2 Discrete-cracking model
An alternative to the continuous smeared-cracking model is the introduction of

discrete cracks (Ngo and Scordelis, 1966). This is normally done by disconnecting the
displacement at nodal points for adjoining elements as shown in Fig. 6.20. One
obvious difficulty in such an approach is that the location and orientation of the cracks
are not known in advance. Thus geometrical restrictions imposed by the preselected
finite-element mesh can hardly be avoided. This can be rectified to some extent by
redefinition of element nodes; such techniques are unfortunately extremely complex
and time-consuming (Ngo, 1975). Furthermore, within the mechanics of finite
elements, the trend has to use higher-order elements. These elements, particularly the
isoparametric version, yield rather poor-quality corner-stress definition which does
not blend well with the edge cracking associated with the discrete-crack concept.
With the change of the topology in these models, the redefinition of the model points
destroys the narrow bandwidth in the structural stiffness matrix and greatly increases
the computation effort required for the solution. These difficulties and limitations
results in a very limited acceptance of the use of discrete-cracking representations for
crack modelling in general structural applications.
6-39
(a) (b)
Figure 6-20 Discrete-cracking model (Chen, 1982)
Further, the aggregate interlock can be simulated in the discrete-cracking

representation by using special linkage elements that cross the crack and control its
behaviour as it slides. The stiffness of these linkages can be decreased as the crack
opens, thereby decreasing the interlock forces with large cracks.
6.6.3 Fracture-mechanics model
The success fracture-mechanics theory has been used in solving various types of
cracking problems in metals, ceramics, and rocks leads to its use in finite-element
analysis of masonry structures. If one accepts the fact that masonry is a notch-
sensitive material, a cracking criterion based on tensile strength may be dangerously
unconservative and the use of fracture-mechanics theory will provide a more rational
approach to concrete cracking. In its current state of development, however, the
practical applicability of fracture mechanics to masonry is still questionable, and
much remains to be done (Kesler et al., 1972) At present, this area is being very
actively studied by several researchers (e.g., Hillemier and Hilsdorf, 1997; Bazant and
Cedoline, 1980).
6-40
6.6.4 Crack detection surface
Cracking is assumed to be the most important aspect of the behaviour and

representation of cracking and of post cracking behaviour dominates the modelling.
The current model adopts the concept of crack detection surface. Cracking is assumed
to occur when the a failure surface that is called the "crack detection
surface." This failure surface is a linear relationship between the equivalent pressure
stress, and the Mises equivalent deviatoric stress and is illustrated in Figure 6.21.
When a crack has been detected, its orientation is stored for subsequent calculations.
Subsequent cracking at the same point is restricted to being orthogonal to this
direction since stress components associated with an open crack are not included in
the definition of the failure surface used for detecting the additional cracks.
Cracks are irrecoverable: they remain for the rest of the calculation (but may open and
close). No more than three cracks can occur at any point (two in a plane stress case,
one in a uniaxial stress case). Following crack detection, the crack affects the
calculations because a damaged elasticity model is used.
q
U
crack detection surface
auilace
I
Figure 6.21 Masonry failure surface in the p-q plane (ABAQUS, 2000)
6-41
For the purpose of developing the model, it is assumed that the material loses strength
through a softening mechanism and that this is dominantly a damage effect, in the
sense that open cracks can be represented by loss of elastic stiffness (as distinct from
the nonrecoverable straining that is associated with classical plasticity effects, such as
what is used for the compressive behaviour model). The model neglects any
permanent strain associated with cracking; that is, we assume that the cracks can close
completely when the stress across them becomes compressive.
6.6.5 Crack direction assumptions
Various researchers have proposed three basic crack direction models (Rots and
Blaauwendraad, 1989): fixed orthogonal cracks; the rotating crack model; and fixed
multidirectional (nonorthogonal) cracks. In the fixed orthogonal crack model the
direction normal to the first crack is aligned with the direction of maximum tensile
principal stress at the time of crack initiation. The model has memory of this crack
direction, and subsequent cracks at the point under consideration can only form in
directions orthogonal to the first crack. In the rotating crack concept only a single
crack can form at any point (aligned with the direction of maximum tensile principal
stress). Thus, the single crack direction rotates with the direction of the principal
stress axes. This model has no memory of crack direction. Finally, the
multidirectional crack model allows the formation of any number of cracks at a point
as the direction of the principal stress axes changes with loading. In practice, some
limitation is imposed on the number of cracks allowed to form at a point. The model
has memory of all crack directions.
The multidirectional crack model is the least popular, mainly because the criterion
used to decide when subsequent cracks form (to limit the number of cracks at a point)
is somewhat arbitrary: the concept of a "threshold angle" is introduced to prevent new
cracks from forming at angles less than this threshold value to existing cracks. The
fixed orthogonal and rotating crack models have both been used extensively, even
though objections can be raised against both. In the rotating crack model the concept
of crack closing and reopening is not well-defined because the orientation of the crack
6-42
can vary continuously. The fixed orthogonal crack model has been criticized mainly
because the traditional treatment of "shear retention" employed in the model tends to
make the response of the model too stiff. This problem can be resolved by
formulating the shear retention in a way that ensures that the shear stresses tend to
zero as deformation on the crack interfaces takes place (this is done in the ABAQUS
model, as described later). Finally, although the fixed orthogonal crack model has the
orthogonality limitation, it is considered superior to the rotating crack model in cases
where the effect of multiple cracks is important (the rotating crack model is restricted
to a single crack at any point).
The fixed orthogonal cracks model is adopted in the current model so that the
maximum number of cracks at a material point is limited by the number of direct
stress components present at that material point of the finite element model (for
example, a maximum of three cracks in three-dimensional, axisymmetric, and plane
strain problems or a maximum of two cracks in plane stress problems). Once cracks
exist at a point, the component forms of all vector and tensor valued quantities are
rotated so that they lie in the local system defined by the crack orientation vectors (the
normals to the crack faces). The model ensures that these crack face normal vectors
are orthogonal so that this local system is rectangular Cartesian. Crack closing and
reopening can take place along the directions of the crack surface normals. The model
neglects any permanent strain associated with cracking; that is, we assume that the
cracks can close completely when the stress across them becomes compressive.
6.6.6 Cracked shear retention
As the masonry cracks, its shear stiffness is diminished. This effect is defined in the
* SHEAR RETENTION option by specifying the reduction in the shear modulus as a
function of the opening strain across the crack. This option also provides for the
definition of a reduced shear modulus for closed cracks. This reduced shear modulus
will also have an effect when the normal stress across a crack becomes compressive.
The new shear stiffness will have been degraded by the presence of the crack.
6-43
The modulus for shearing of cracks is defined as pG, where G is the elastic shear
modulus of the uncracked masonry and p is a multiplying factor. The shear retention
model assumes that the shear stiffness of open cracks reduces linearly to zero as the
crack opening increases:
p=O (6-31)
where is the direct strain across the crack and E is the value given on the data line
of the *SHEAR RETENTION option. The model also assumes that cracks that
subsequently close have a reduced shear modulus:
close
= for < 0, (6-32)
where p close is defined on the data line of the option.
6.7 Conclusion
In this study, the finite element method is adopted for numerical modelling to
simulate masonry behaviour. Masonry is a composite material that consists of units
and mortar joints. Macro-modelling is adopted to model this composite material due
to the practical purpose. To develop an accurate finite element macro-modelling, it is
very important to give an accurate composite material behaviour and composite
failure criteria.
The proposed numerical model is based on ABAQUS concrete model and is

modified to simulate the special characterises of masonry. The model employs
oriented damaged elasticity (smeared cracking) and isotropic compressive plasticity
to represent the inelastic behaviour of masonry. The model is intended to model the
masonry behaviour for relatively monotonic loadings under fairly low confining
pressures. Due to the limitations of ABAQUS, the model makes no attempt to
include prediction of cyclic response or of the reduction in the elastic stiffness caused
6-44
by inelastic straining. An isotropically hardening "compressive" yield surface forms

the basis of the model for the inelastic response. when the principal stresses are
dominantly compressive. In tension once cracking is defined to occur (by the "crack
detection surface" of the model), the orientation of the cracks is stored and oriented,
damaged elasticity is then used to model the existing cracks. For tension softening,
the bilinear model is adopted in the current model.
6-45
Chapter 7: A basic mechanical model for the retrofitted walls
CHAPTER 7
A BASIC MECHANICAL MODEL FOR THE
RETROFITTED WALLS
7.1 General
The results of the experimental study carried out in this investigation demonstrate that
the proposed cable system is an effective seismic retrofitting strategy. Analysis of this
retrofitted wall with cable system will be conducted in this chapter in an effort to
explain the mechanism of load resistance. Simple analysis tools, suitable for the
design office environment will be developed and the results will be compared with
those obtained from the experimental research. A simple truss model based on the
theory of plasticity is presented and verified by the experimental data.
7.2 Simple truss model
A simplified truss model is introduced for the analysis of retrofitted walls to obtain
their load-deflection relationships. An indeterminate truss with five members is used
to represent retrofitted walls as shown in Figure 7.1. The masonry strut in tension is
presented as member '1' in Figure 7.1. The width of this vertical tension strut is
assumed to be one tenth of wall length. The rigid horizontal member in the plane of
the floor or roof, or the upper rigid top beam in the case of wall specimens is labelled
'2'. The masonry strut in compression is referred to as member '3'. The width of this
vertical compression strut is assumed to be one tenth of wall length. The diagonal
member in tension and compression are labelled as member '4' and '5', respectively.
7-1
The cable in tension is labelled as member '4'. The section properties of this element
were taken equal to the cable section properties. Member '5' is a masonry diagonal
compressive strut with an effective width taken as one-thirty two of the wall length.
Members '1' and '3', as well as '4' and '5' swap places when the horizontal load
reverse.
1. Vertical lension masonry member

2. Rigid horizontal member
3. Vertical compression masonry member
4. Diagonal tension cable member
5. Diagonal compression masonry member
Figure 7.1 A simple truss model for the retrofitted wall with cable system
A step by step calculation procedure is required to conduct inelastic analysis of the

truss model described. This involves a series of elastic analyses. An elastic analysis is
performed first to determine member forces in terms of total applied force V. The
stiffness and strength of each member is reduced gradually, as appropriate in the step-
by-step analysis, to account for degradation due to masonry crushing. The first
analysis is conducted to determine all member forces and the corresponding lateral
deflection in terms of the total applied force V. This step is to isolate point "a" and
"b" as a free body (Figure 7.2 (b) and (c)) and consider its equilibrium.
7-2
F5
(
1
Rax Rbx
Ray Rby
(a) Forces in a simple truss model
t,F4 PS
P3
8 Jr
a
8
b
Ray
I
Rby
Rbx
(b) A free body for point "a" (c) A free body for point "b"
Figure 7.2 Free body diagrams
7-3
From static in Figure 7.2 (b) we then obtain:
Fx = 0
Rax=-F4 cose (7-1)
And in the Figure 7.2 (c)
Z Fx=O
Rbx = F5 cos e (7-2)
Then, from the equilibrium of forces in the horizontal direction in the Figure 7.2 (a),
we obtain
Fx = 0
V + Rax - Rbx 0 (7-3)
Substituting equations (7-1) and (7-2) into this equation (7-3), we find
V= cos e + F5 cos e (7-4)
From equation (7-4), it shows that the force V is contributed from F4 (cable) and F5
(masonry). Or on the other word, the stress-strain relationship for the retrofitted wall
can be obtained from combing the stress-strain relationship of the masonry and cable
as shown in Figure 7.3. Therefore, this structure can be simplified by dividing it into
two static structures as shown in Figure 7.4. The displacement A is caused by the
force V. Because element 2 is a rigid element, it will cause the two new structures
(Figure 7.4(b) and (c)) having the same amount of displacement.
7-4
0
E
0 0
E
(a) Retrofitted wall (b) Masonry wall (c) cable
Figure 7.3 Stress-strain relationships
(a) Retrofitted wall (b) Masonry wall (c) cable
Figure 7.4 Simplified truss model
From Figure 7.5, it shows that force and displacement relationships in elastic analysis
for two simplified trusses.
7-5
cos 0
M=Acose
4E5
=
p5
F4 144 P4
LI
(a) Masonry wall (b) cable
Figure 7.5 Force and displacement relationship in elastic analysis for two trusses
From Figure 7.5, it shows
A4=A5 (7-5)
From Hooke's law
F4L4 = F5L5
(7-6)
A4E4 A5E5
From which
F5L5 A4E4
F4= (7-7)
A5E5 L4
Substituting equation (7-7) into equation (7-4) and solving for F5, we find
7-6
F5= VAE (7-8)

L
(1+ x 4)cosO
A5E5 L4
The theory of plasticity is adopted to estimate the strength of the retrofitted low-rise
walls. This theorem is based on the following assumptions:
1. The material behaviour must be elastic-plastic

2. The material is assumed to follow Hooke's law up to the yield point, after which it
yields under constant stress
Therefore, it is assumed that the masonry is elastic-plastic material. As the load V is

gradually increased, the force in the elements will increase also, and as long as the
stresses remain below the yield stress the forces can be determined by an elastic
analysis.
Eventually, the masonry (F5) element will reach the yield stress This occurs when
the force F5 become equal to A5. With further increase in the load V, the force
F4 (cable element) will increase also, but the force F5 (masonry element) remains
constant because the masonry (F5) element has become plastic. Finally, the cable
element (F4) will also become plastic, and then the structure can not carry any
additional load. This load is called the ultimate load Vi,.
When the masonry (F5) element yields, the corresponding value of the load V is
called the yield obtained by setting F5 equal to A5 in the above equation:
A4E4
= A5 (1+ x ) cos 8 (7-9)
A5E5
7-7
When the cable element (F4) also yields, the corresponding value of the load V is
called the ultimate load obtained by setting F4 equal to A4:
A5)cosO (7-10)
Upon reaching this force level, a complete plastic collapse mechanism forms. The
ultimate strength of the retrofitted wall is given by equation (7-10). The resulting
load-deflection envelope for a retrofitted wall can be constructed using these values.
Note that gravity loads were not considered in the above analysis. If the cables are
added as part of a seismic retrofitting scheme, the gravity loads would already be
resisted by the existing masonry wall. This can be accounted for in the truss model by
reducing the compressive strength of Members 5 by an amount corresponding to
gravity-induced stresses. If the gravity loads are applied after the cables are installed,
as was the case in this experimental program, both the existing masonry wall and the
new cable systems will resist the entire gravity load together. A correction can be
made in the above procedure.
7.3 Validation of the Proposed Techniques
The force-displacement relationships from the simplified truss model described in this
chapter have been compared with experimentally obtained lateral force-horizontal
displacement relationships of the retrofitted wall.
The following calculations were used to obtain the force-horizontal displacement

relationships of the retrofitted wall from the simple truss model.
Substituting = a4 and a5 into equation (7-10) for elastic analysis, we have
V=(a4A4+a5A5)cosO (7-11)
7-8
From Hooke's law,
a4 =E4e4 (7-12)
And
O•5 =E5e5 = (7-13)
Substituting equations (7-12) and (7-13) into equation (7-11), we find
v= A4E4 A5E5
A4cosO+ A5cos8 (7-14)
L4 L5
Based on the experimental specimens, the following parameters have been adopted
into this equation.
The dimensions of the walls were 940mm X 940mm X 110mm
For cable,
E4 = 107500 MPa (cable Elastic Modulus)
A4 = = 78.5mm2 (cable cross-section)
L4 = 1330 mm (cable length)
For masonry, the stress-strain relationship which was shown in Figure 7.3 has been
simplified to a bi-linear curve (figure 7.6) based on the discussion in Section 7.2.
7-9
/ L
Figure 7.6 Simplified stress-strain relationship for masonry
The compressive strength (f' of masonry = 7.92 MPa

E5 = 5940 MPa (masonry Elastic Modulus)
A5 = 10 = 3231 mm2 (one-thirty two of the wall length X width of the wall)
940
L5 = cos45 = 1330 mm (masonry diagonal length)
These results are presented in Figs 7.7 and 7.8 for the whole wall and force carried by
cable, respectively. The model provides good correlations with test data and can be
considered as effective analytical tools. In these figures, cable force has been
considered 30% loss when the force transfers from the masonry to cable through the
steel plate connections.
7-10
__
The force-displacement relationship for the retrofitted wall

50
45
40
35
z
0 2 4 6 8 10 12 14
Displacement (mm)
Figure 7.7 Force-displacement relationships for the retrofitted wall
Forced carried by the cable
25
20 --- - - - -.
z
0
0 15 . - - - -. - -
10
-- - -- Experiment
- . - -. -
— Simple truss model
5 ---.
0 2 4 6 8 10 12 14
Displacement (mm)
Figure 7.8 Force-displacement relationships for the cable
7-11
7.4 Effects of Diagonal Strut widths on the Accuracy of

Simple Truss Model
The effective width of the masonry considered in calculating the strength and stiffness
of compression strut was taken equal to one-thirty two of the length of the wall for
diagonal element. For the effective width of diagonal struts, a range of values were
recommended by other researchers in studies of other types of structures. For
example, Drysdale et al (1994) recommended, diagonal strut widths equal to one-
tenth to one-third of the diagonal length of the wall for truss models used in analysis
of infihled frames. The New Zealand Code specifies a diagonal strut width equal to
one quarter of infill length. Paulay (1992) proposed a width equal to one-quarter of
the diagonal length for design of infihled frames. A parameter study was conducted in
this study to determine a reasonable diagonal strut width for use in truss models. The
results showed that the resultant force displacement curve can be matched well with
the experimental results when the diagonal strut was equal to one-thirty two of the
wall length. Therefore, one-thirty two of the length of the wall has been adopted in
this study.
7.5 Conclusions
A basic mechanical model was developed to predict strength and ductility of low-rise
unreinforced masonry walls retrofitted with cable system. A step by step calculation
procedure is also provided to conduct inelastic analysis and design of the retrofitted
wall with the model described. It was found that this simple truss model captured
accurately the lateral force-horizontal displacement relationships for the retrofitted
wall and the cable. The model provided analytical results that are in good agreement
with those obtained experimentally. The model described in this chapter allows the
formulation of a simple analysis calculation that engineers may employ in retrofitting
existing low-rise walls.
7-12
Chapter 8: Analytical study for Retrofitted walls
CHAPTER 8
ANALYTICAL STUDY FOR RETROFITTED WALLS
8.1 Introduction
In this chapter, two nonlinear finite element models have been developed to exam the
experimental results. One model is developed for unreinforced masonry walls
retrofitted by cable system and the other model is developed for unreinforced masonry
walls retrofitted by FRP. It has been found that both models agree reasonably well
with the experimental results. The models take into account the material nonlinearities
as well as damage due to progressive cracking. Compressive behaviour of masonry is
modelled using the theory of plasticity and cracking is modelled using smeared crack
model. The models are generated using ABAQUS finite element program. The results
of the numerical modelling are also discussed in this chapter. It is shown that the
numerical models are capable of predicting not only the load carrying capacity, but
also ductility of the retrofitted masonry wall. The main objective of this research is to
prove a new retrofitting technique with experimental testing. These models are for
limit-state examination purpose.
8.2 Macro-modelling for the retrofitted wall
Masonry is a composite material. It consists of bricks and mortar joints. The macro-
modeling does not make a distinction between individual units and joints but treats
masonry as a homogeneous anisotropic continuum. Masonry can be assumed to be a
homogeneous material if a relation between average stresses and strains in the
composite material is established. Errors are induced by this assumption due to mortar
8-1
joints acting as a plane of weakness. For the numerical analysis, bilinear plane stress
continuum elements with full gauss integration were utilized.
An accurate analysis of masonry structures in a macro-modelling (or composite)

perspective requires a material description for all stress states. It is necessary to obtain
the composite yield criterion to describe the behaviour of composite material and
calculate the macro-properties of composite material based on the actual discretization
of the components. The constitutive laws, failure criteria and the crack modelling of
URM have been discussed in the previous chapter.
8.3 The finite element model for the retrofitted wall by cable system
In this section, a nonlinear finite element model has been developed to validate the
experimental results from unreinforced masonry walls retrofitted by cable system.
Reasonably good agreement has been found between the analytical and experimental
results. The model takes into account material nonlinearities as well as damage due to
progressive cracking. Behaviour of the masonry is modelled using the theory of
plasticity and cracking is modelled using smear cracking approach. The model is
generated using ABAQUS finite element program. The validity of the model is
established by comparison with experimental results. It is shown that the numerical
model is capable of predicting the load carrying capacity of the URM wall and the
retrofitted walls.
For macro-modelling, the four-node bilinear two-dimensional plane stress element,

CPS4R, was used to model the masonry. The four-node shell element, S4R, was used
to model the steel plates that used to connect between the cable and masonry wall.
The two-dimensional truss element, T2D2, was used to model the cable. No
compressive option was used to make sure the cable only take tension. The connector
element, connection type BEAM, was used to model connection between the steel
plate and masonry wall and connection between the steel plate and cable. The details
of the numerical modelling are also discussed in this section.
8-2
8.3.1 Selection of element type
The selection of element for masonry walls, steel plates and cables are discussed in
this section.
8.3.1.1 Masonry
For macro-modelling, the four-node bilinear two-dimensional plane stress element,

CPS4R, was used to model the masonry. The two-dimensional plane stress element
has four nodes with two degrees of freedom at each node-translations in the nodal x, y
directions. The plastic deformation, cracking in two orthogonal directions, and
crushing could be modelled by this type of element. The geometry and node locations
for this element type are shown in Figure 8.1.
1 2
x
4node reduced
element
Figure 8.1 The geometry and node locations for CPS4R (ABAQUS 2000)
A more sophisticated element, such as the eight-node isoparametric element, was not
used because it has been shown from previous research (Bazant and Cedolin, 1980;
Au, 1987) that the use of higher order elements was not warranted for the analysis of
brick masonry (where nonlinearity is mainly due to progressive cracking and not non-
linear material characteristics), provided a relatively fine element mesh was adopted.
8.3.1.2 Steel Plate
The four-node doubly curved general-purpose shell element, S4R, was used to model
the steel plates that used to connect between the cable and masonry wall. The shell
8-3
element has four nodes with six degrees of freedom at each node-translations and -
rotations in the nodal x, y and z directions. The geometry and node locations for this
element type are shown in Figure 8.2
Y
a
x
1 2
4-node reduced
integration
Figure 8.2 The geometry and node locations for S4R (ABAQUS 2000)
8.3.1.3 Cable
The two-dimensional truss element, T2D2, was used to model the cable. The 2-node
straight truss element uses linear interpolation for position and displacement and has a
constant stress. The truss element is long, slender structural member that can transmit
only axial force. No compressive option was used to make sure the cable only takes
tension. The element has two degrees of freedom at each node and translations in the
nodal x and y directions. The geometry, node locations, and the coordinate system are
2
2
Y
1
x end 1
2- oleniertt
Figure 8.3 The geometry and node locations for T2D2 (ABAQUS 2000)
8-4
8.3.1.4 Connection between the steel plate and masonry wall
The connector element, connection type BEAM, was used to model connection
between the steel plate and masonry wall. The geometry, node locations, and the
coordinate system are shown in Figure 8.4.
8.3.1.5 Connection between the steel plate and cable
The connector element, connection type BEAM, was also used to model the
connection between the steel plate and cable.
2
1
b
a
I
e"3
Figure 8.4 The geometry and node locations for connector element (BEAM)
(ABAQUS 2000)
8-5
8.3.2.1 The stress-strain relationship of masonry
Development of a model for the behaviour of masonry is a challenging task. Masonry

is a quasibrittle material and has different behaviour in compression and tension.
Figure 8.5 shows the stress-strain curve for clay-brick masonry which was adopted in
this study.
Fa&j,u
Figure 8.5 A stress-strain curve for clay-brick masonry (ABAQUS, 2000)
masonry, ABAQUS requires input data for material properties as follow (please
see details in Appendix A):
Elastic Modulus (E =5940 MPa
Ultimate uniaxial compressive strength (f' =7.92 MPa
Tensile strength (f',) =0.2 MPa

Poisson's ratio (v) = 0.09
Failure Ratio = 1, 0.035
Plastic strain = 0.0046
8-6.
8.3.2.2 Failure criteria for masonry
The model is capable of predicting failure for masonry materials. Both cracking and
crushing failure modes are accounted for. The two input strength parameters, i.e.,
ultimate uniaxial tensile and compressive strengths, are needed to define a failure
surface for the masonry. Consequently, a criterion for failure of the masonry due to a
multiaxial stress state can be calculated.
A typical two-dimensional failure surface for masonry that is adopted in this study is
shown in Figure 8.6. In masonry, cracking occurs when the principal tensile stress in
any direction lies outside the failure surface. After cracking, the elastic modulus of the
masonry element is set to zero in the direction parallel to the principal tensile stress
direction. Crushing occurs when all principal stresses are compressive and lie outside
the failure surface; subsequently, the elastic modulus is set to zero in all directions
(ABAQUS, 2002), and the element effectively disappears.
"crack detection surface
biaxial
tension
Figure 8.6 A two-dimensional failure surface for masonry

(ABAQUS 2000)
8-7
8.3.2.3. Steel Plate
Steel plates were added at the connection between cable and masonry wall in this
finite element model. Properties such as elastic modulus and yield stress, for the steel
plate used in this FEM study follow the design material properties used for the
experimental investigation. The steel plate for the finite element models was assumed
to be a linear elastic material. Material properties for the steel plate for all models are
as follows:
Elastic modulus, E5 = 200,000 MPa

Yield stress, = 410 MPa
Poisson's ratio, V = 0.3
8.3.2.4 Cable
The type of cable used for this experiment was Ronstan typical grade 316 stainless
steel wire rope (19 single strands, diameter 10mm, and breaking load 71 KN) and
ending for cable was seafast threaded swage terminals (RF1513M1O1O). Material
properties for the cable for all models are as follows:
Elastic modulus, = 107,500 MPa
Breaking load = 71 KN
Poisson's ratio,v =0.3
8.3.3 Geometry
The dimensions of the walls were 940mm X 940mm X 110mm. Figure 8.7 illustrates
typical dimensions for all walls before cable reinforcing. Figure 8.8 shows the finite
element model for the URM wall before cable reinforcing. Figure 8.9 shows the
actual wall retrofitted with cable. Figure 8.10 shows the finite element model for the
8-8
retrofitted wail with cable. Figure 8.11 shows the details of connection modelling
(section A-A).
i11, (-
- II
F"''' II II
Li II II
I
C'
LIII
II
II
II
II
II
II It II
940mm (4 courses)
Figure 8.7 The dimension of the wall
I I ii I I I I I I I I I I I
Figure 8.8 The finite element model for the URM wall before cable reinforcing
8-9
Figure 8.9 The actual wall retrofitted with cable
I
Section A:A
Figure 8.10 The finite element model for the retrofitted wall with cable
8-10
Cable
constraint
bolt
Masonry
Figure 8.11 The details of connection modelling (section A-A)
In Figure 8.11, it shows the masonry wall and steel plate are connected by four
BEAM connector elements. There are 1 mm between the masonry wall and steel
plate. It also shows the steel plate and cable are connected by a BEAM connector
element which is 1mm long.
8.3.4 Loading and boundary condition
All walls were tested under combined constant gravity load and incrementally
increased in-plane lateral displacement reversals, as shown in Figure 8.12. The finite
element models were loaded and constrained at the same locations as the walls, but
the models were only loaded monotonically.
8-11
Roof load
Lateral
Displacement
Fix to
strong floor
Figure 8.12 Loading and boundary condition of the retrofitted wall model
8.3.5 Finite element discretization
As an initial step, a finite element analysis requires meshing of the model. In other
words, the model is divided into a number of small elements, and after loading, stress
and strain are calculated at integration points of these small elements (Bathe 1996).
An important step in finite element modelling is the selection of the mesh density. A
convergence of results is obtained when an adequate number of elements is used in a
model. This is practically achieved when an increase in the mesh density has a
negligible effect on the results (Adams and Askenazi 1998) after the convergence
study. Figure 8.13 shows meshes for the retrofitted wall model.
8-12
8.4.1 Selection of element type
8.4.1.1 Masonry
For this macro-modelling, the same four-node bilinear two-dimensional plane stress
element, CPS4R, was used to model the masonry. Please see the section 8.3.1.1. for
details.
8.4.1.2 FRP
The membrane element, M3D4R, was used to model the FRP. This type of element is
used to represent thin surfaces that offer strength in the plane of the element but have
no bending stiffness. The element has three degrees of freedom at each node and
translations in the nodal x, y and z directions. The geometry and node locations for
this element type are shown in Figure 8.14.
Y 3
x
4- node reduced
integration element
Figure 8.14 The geometry and node locations for M3D4R (ABAQUS 2000)
8.4.1.3 Connection between FRP and masonry wall
A surface-based Tie constraint was used to model connection between FRP and
masonry wall. This constraint type, TIE, makes two surfaces to be tied together. The
TIE constraint can make the translational and rotational motion as well as all other
active degree of freedom equal for a pair of surfaces. The surface-based tie algorithms
are shown in Figure 8.15.
8-14
s4ave sufface defined

on shell sttucture
.master surFace defined
on shell struclure
Dlsplaoement and rotation degrees

of freedom are tied, unless the
NO ROTATION parameter is used.
Figure 8.15 Surface-based tie algorithms (ABAQUS 2000)
8.4.2 The properties of the materials
8.4.2.1 Masonry
The same masonry model was used in this model. Please see the section 8.3.2.1 for
details.
8.4.2.2. FRP
The type of FRP used for this model was Sika CarboDur S (Width 30mm, thickness
1.2mm and tensile strength 2800 MPa) and epoxy adhesive for bonding Sika
CarboDur Strips was Sikadur-30. Material properties for FRP for all models are as
follows:
Elastic modulus, E5 = 62000 MPa

Tensile strength = 2800 KN
Poisson's ratio, V = 0.22
8-15
8.4.3 Geometry
The dimensions of the walls were 940mm X 940mm X 110mm. Figure 8.16 shows
the actual wall retrofitted with FRP. Figure 8.17 shows the finite element model for
the retrofitted wall with FRP.
H±il+HikJXW
Figure 8.16 The actual wall retrofitted with FRP
Figure 8.17 The finite element model for the retrofitted wall with FRP
8-16
8.4.4 Loading and boundary condition
All walls were tested under combined constant gravity load and incrementally
increased in-plane lateral displacement reversals, as shown in Figure 8.18. The finite
element models were loaded and constrained at the same locations as the walls, but
the models were only loaded monotonically.
Roof load
Lateral
Displacement
Fix to
strong floor
Figure 8.18 Loading and boundary condition of the retrofitted wall model
8-17
8.4.5 Finite element discretization
Figure 8.19 shows meshes for the retrofitted wall model.
Figure 8.19 The mesh for the retrofitted wall model
8.5 Material nonlinear analysis
There are two types of nonlinearities which need to be considered. One is "material
nonlinearity" which is caused by the nonlinear stress-strain relation or progressive
local failure, the other is "geometric nonlinearity" caused mainly by the large
displacement of the structure. Only material nonlinearity is considered in this
research, as geometric nonlinearity is not applicable to unreinforced masonry under
in-plane lateral load since only small deflections occurred.
The nonlinear behaviour of URIVI is caused by two-major effects, progressive local

failure (cracking or crushing of the masonry) and nonlinear stress-strain relations of
the material, particularly occurring in the biaxial compression-compression and
unaxial compression stress states. Both these two effects are considered in this study.
The details of the numerical modelling have been discussed in Chapter 6.
8-18
8.6 Solution techniques for nonlinear analysis
The finite element models generated in ABAQUS are usually nonlinear and can
involve from a few to thousands of variables. In terms of these variables the
equilibrium equations obtained by discretizing the virtual work equation can be
written symbolically as
(8.1)
where FN is the force component conjugate to the variable in the problem and
tiM is the value of the Mth variable. The basic problem is to solve Equation 8.1 for•
the tiM throughout the history of interest.
In nonlinear analysis, the total load applied to a finite element model is divided into a
series of load increments called load steps. At the completion of each incremental
solution, the stiffness matrix of the model is adjusted to reflect nonlinear changes in
structural stiffness before proceeding to the next load increment.
The ABAQUS program (ABAQUS, 2002) generally uses Newton's method as a

numerical technique for solving the nonlinear equilibrium equations for updating the
model stiffness. The motivation for this choice is primarily the convergence rate
obtained by using Newton's method compared to the convergence rates exhibited by
alternative methods (usually modified Newton or quasi-Newton methods) for the
types of nonlinear problems most often studied with ABAQUS.
The basic formalism of Newton's method is as follows. Assume that, after iteration i,
an approximation to the solution has been obtained. Let be the difference
between this solution and the exact solution to the discrete equilibrium equation
Equation 8.1. This means that
(8.2)
8-19
Expanding the left-hand side of this equation in a Taylor series about the approximate
solution u7' then gives
+ +... = 0 (8.3)
If u7' is a close approximation to the solution, the magnitude of each will be

small, and so all but the first two terms above were neglected giving a linear system
of equations:
= FN (8.4)
where
= aFN
KNP
is the Jacobian matrix and
F1N =F NI M
The next approximation to the solution is then
(8.5)
and the iteration continues.
Convergence of Newton's method is best measured by ensuring that all entries in
and all entries in are sufficiently small. Both these criteria were checked by
default in an ABAQUS/Standard solution. ABAQUS/Standard also records peak
values in the force residuals, incremental displacements, and corrections to the
incremental displacements at each iteration.
8-20
In this study, Newton's method is adopted to solve nonlinear problem. For the
masonry elements, convergence criteria were based on force and displacement, and
the convergence tolerance limits were initially selected by the ABAQUS program.
8.7 Validation of the numerical modelling results with the test results
This section compares the results from ABAQUS finite element analyses with
experimental results. The following comparisons are made: load-deflection plots; first
cracking loads; loads at failure and crack patterns at failure. The development of crack
patterns for each wall and summaries of the maximum force occurring in the cable for
the finite element models are also discussed. The data from the finite element analyses
were collected at the same locations as the experimental tests for the masonry wall.
8.7.1 URM walls retrofitted by cable system
8.7.1.1 Load-deflection plots
Force-displacement relationships from the finite element models described in this

section have been compared with experimentally obtained envelopes of the lateral
force-horizontal displacement relationships of retrofitted walls by cable system and
unreinforced wall. These results are presented in Figures 8.20, 8.21 and 8.22. The
models provide good correlations with test data and can be considered as effective
analytical tools.
> Load-deflection plot for URM wall
Figure 8.20 shows that the load-deflection plot from the finite element analysis agrees
well with the experimental data for the URM wall. The load-deflection plot from the
finite element analysis is slightly stronger than that from the experimental results.
This is possibly due to the relative homogeneity of the finite element models when
compared to the relative non-homogeneity of the actual walls that contain two
different materials and neglect the weaker mortar plane effect. The first cracking load
8-21
for the finite element analysis is 15.2 KN, which is higher than the load of 14.7 KN
from the experimental results by only 4%. Lastly, the final load of 24.4 KN from the
model is higher than the ultimate load of 23.8 KN from the experimental data by only
3 %.
30
0
-J
G)
210
U)
0
0 1 2 3 4 5 6
Displacement (mm)
Figure 8.20 Load-deflection plot for URM wall
Load-deflection plot for retrofitted wall with cable
well with the experimental data for the retrofitted wall with cable. The load-deflection
plot from the finite element analysis is stiffer than that from the experimental results.
This is possibly due to the fatigue of the material. The result from the finite element
model is for the monotonic loading, but the results from the experiment are a
consequence of cyclic loading. The first cracking load for the finite element analysis
is 17.5 KN, which is higher than the average first cracking load of 15.2 KN from the
experimental results by 15%. Lastly, the final load of 49.15KN from the model is
higher than the average ultimate load of 46.4 KN from the experimental data by 6 %.
8-22
URM retrofitted with cable
50
40
z
0
0
0 2 4 6 8 10 12
Displacement (mm)
Figure 8.21 Load-deflection plot for retrofitted wall with cable
Load-deflection plot for before and after retrofitting
Figure 8.22 shows the finite element results of load-deflection curve for the URM
wall before and after retrofitting with cable. The load-deflection curve for the
retrofitted wall is much stiffer than the load-deflection curve for the unretrofitted wall.
In this figure, it also shows that the retrofitted wall significantly increases the strength
and ductility. These results are similar to the experimental results. The improvement
of the ultimate lateral load resistance of the retrofitted walls with cables is 2 times the
capacity of unreinforced wall. The improvement of the, ductility of the retrofitted
walls with cables is about six times the capacity of unreinforced wall.
8-23
50
40
z 30
0
a)
V 20
Cl)
UR R
10
RETROFITTING)
If
0
0 2 4 6 8 10 12
Displacement (mm)
Figure 8.22 F.E.M. results of load-deflection plot for URM wall before and after
retrofitting
8.7.1.2 First cracking loads
The first cracking load from the finite element analysis is the load step where the first
signs of cracking occur for masonry in the model. Loads at first cracking from the
model and the experimental results are compared in Table 8.1.
Table 8.1 Comparisons between experimental and ABAQUS first cracking loads
Retrofitting Methods Experiment F.E.M. % Difference

URM 14.7KN 15.2KN 4%
URM+CABLE 15.2KN 17.5KN 15%
The first cracking loads from the finite element analyses and the experimental data are
within 15% for URM wall and retrofitted walls. In all cases, the first cracking load
from ABAQUS is higher than that from the experimental data. This is possibly due to
8-24
the relative homogeneity of the finite element models when compared to the relative
heterogeneity of the actual walls that contain a number of microcracks.
8.7.1.3 Ultimate loads
Table 8.2 compares the ultimate loads for the URM wall and retrofitted walls from
experiments and from the finite element simulations.
Table 8.2 Comparisons between experimental and ABAQUS loads at failure

URM 23.8KN 24.4KN 3%
URM+CABLE 46.4KN t 49.15KN 6%
8.7.1.4 Force carried by cable
For the actual retrofitted walls, there was no evidence that the cable failed before
overall failure of the walls. This is confirmed by the finite element analyses. In figure
8.23, the forces carried by the cable from ABAQUS are compared with the tensile
strengths for the cable measured from experimental tests. In this figure, it shows 2KN
carried by the cable at the beginning stage due to the pretension of the
cable.
w 30
25
w
— 20
0)
I..
CD
0
5
CD
0
-J
0 2 4 6 8 10 12
Displacement (mm)
Figure8.23 Comparison between experimental and ABAQUS for carried force by

cable
8-25
8.7.2 URM walls retrofitted by FRP
8.7.2.1 Load-deflection plots
Force-displacement relationships from the finite element models described in this

chapter have been compared with experimentally obtained envelopes of the lateral
force-horizontal displacement relationships of retrofitted walls. These results are
presented in Figs 8.24 and 8.25. The models provide good correlations with test data
and can be considered as effective analytical tools.
Load-deflection plot for retrofitted wall with FRP
well with the experimental data for the retrofitted wall with FRP. The load-deflection
plot from the finite element analysis is stiffer than that from the experimental results.
This is possibly due to the fatigue of the material. The result from the finite element
model is for the monotonic loading, but the results from the experiment are a
consequence of cyclic loading. The first cracking load for the finite element analysis
is 25.3 KN, which is higher than the average first cracking load of 23.7 KN from the
experimental results by 6%. Lastly, the final load of 65.0 KN from the model is higher
than the average ultimate load of 60.2 KN from the experimental data by 7 %.
Wall (FRP)
70
60
,4o
20
10
0
0 2 4 6 8 10 12 14
Displacement (nun)
Figure 8.24 Load-deflection plot for retrofitted wall with FRP
8-26
Load-deflection plot for before and after retrofitting
Figure 8.25 shows the finite element results of load-deflection curve for the URM
wall before and after retrofitting with FRP. The load-deflection curve for the
retrofitted wall is much stiffer than the load-deflection curve for the unretrofitted wall.
In this figure, it also shows that the retrofitted wall significantly increases the strength
and ductility. These results are similar to the experimental results. The improvement
of the ultimate lateral load resistance of the retrofitted walls with FRP is 3 times the
capacity of unreinforced wall. The improvement of the ductility of the retrofitted
walls with cables is about six times the capacity of unreinforced wall.
70
30 URM ONLY (BEFORE

RETROFITTING)
-
URM÷FRP (AFTER
20 RETROFITTING)
10
0- I
0 2 4 6 8 10 12 14
Displacement (mm)
Figure 8.25 F.E.M. results of load-deflection plot for URM wall before and after
retrofitting
8-27
8.7.2.2 First cracking loads
model and the experimental results are compared in Table 8.3.
Table 8.3 Comparisons between experimental and ABAQUS first cracking loads

URM 14.7KN 15.2KN 4%
URM+FRP 23.7KN 25.3KN 6%
The first cracking loads from the finite element analyses and the experimental data are
6% for URM wall and retrofitted walls. In all cases, the first cracking load from
ABAQUS is higher than that from the experimental data. This is possibly due to the
relative homogeneity of the finite element models when compared to the relative
heterogeneity of the actual walls that contain a number of microcracks.
8.7.2.3 Ultimate loads
Table 8.4 compares the ultimate loads for the URM wall and retrofitted walls from
experiments and from the finite element simulations.
Table 8.4 Comparisons between experimental and ABAQUS loads at failure

URM 23.8KN 24.4KN 3%
URM+FRP 60.2KN 65.OKN 7%
8-28
8.7.2.4 Force carried by FRP
In figure 8.26, the forces carried by FRP from ABAQUS are compared with the
tensile strengths for FRP measured from experimental tests. From this figure, it shows
that the forces carried by FRP are linear from ABAQUS and are nonlinear from the
experiments. The reason for that is due to FRP debonding during the experiment. The
current model is neglecting this effect.
35
30
20
15
n 10
0
0 1 2 3 4 5 6 7 8 9 10
Displacement (mm)
Figure8.26 Comparison between experimental and ABAQUS for carried force by

FRP
8-29
8.8 Comparison between the retrofitted walls by cable system and.

FRP
8.8.1Load-deflection plot for the retrofitted walls by cable system

and FRP
Figure 8.27 shows the finite element results of load-deflection curve for the retrofitted
walls by cable system and FRP. The load-deflection curve for the retrofitted wall by
FRP is which is agreed well with the testing stiffer than the load-deflection curve for
the retrofitted wall by cable system.
70
60
50
30
20
10
0
0 2 4 6 8 10 12 14
Dispalcement (mm)
Figure 8.27 F.E.M. results of load-deflection plot for the retrofitted walls by cable
system and FRP
8-30
8.8.2 First cracking loads
models for the retrofitted wall by cable system and by FRP are compared in Table 8.5.
Table 8.5 Comparisons between the retrofitted walls by cable system and FRP
for first cracking loads

URM-I-CABLE 15.2KN 17.5KN 15 %
URM+FRP 23.7KN 25.3KN 6%
The first cracking loads for the retrofitted wall by FRP are much higher than the first
cracking loads for the retrofitted wall by cable system. These results are similar as the
experimental results.
8.8.3 Ultimate loads
Table 8.6 compares the ultimate loads for the retrofitted walls by cable system and
FRP from experiments and from the finite element simulations.
Table 8.6 Comparisons between the retrofitted walls by cable system and FRP
for ultimate loads

URM÷CABLE 46.4KN 49.15KN 6%
URM+FRP 60.2KN 65.OKN 7%
The ultimate load for the retrofitted wall by FRP is much higher than the ultimate
loads for the retrofitted wall by cable system. These results are similar as the
experimental results.
8-31
8.8.4 Force carried by FRP and cable
Figure 8.28 compares the forces carried by the cable and by FRP from the finite
element simulations. The forces carried by FRP are much higher than the forces
carried by cable system. These results are similar as the experimental results
35
Displacement (mm)
Figure8.28 Comparing the forces carried by the cable and by FRP from ABAQUS
8.9 Conclusion
The main objective of this research is to prove a new retrofitting technique with
experimental testing. These models are for limit-state examination purpose only. Non-
linear finite element models were developed to predict strength and ductility of URM
wall, URM walls retrofitted with cable system and walls retrofitted with FRP. The
experimental results obtained from URM masonry walls retrofitted with the cable
system and FRP are compared with those obtained from analytical solutions. The
analytical results were obtained using the finite element program ABAQUS. The
results from these models also show good agreement with the experimental results.
8-32
The limitations of this model are that this current model couldn't be predicted well for
the retrofitted walls and couldn't model the FRP debonding effect. Due to the
limitations of ABAQUS, the model makes no attempt to include prediction of cyclic
response or of the reduction in the elastic stiffness caused by inelastic straining.
8-33
Chapter 9: Performance-Based Design for retrofitted wall
CHAPTER 9
PERFOEMANCE-BASED DESIGN
FOR RETROFFITED WALLS
9.1 Introduction
Historically, buildings have been designed and constructed in conformance with

buildings codes whose primary goal has been to minimize the potential for loss of life in
a major earthquake. Over the years, the codes have become more detailed in their
prescriptive requirements and expanded to address minimum requirements for the
anchorage of non-structural building components. Despite these continual improvements
and expansions in scope, the primary assumption of the building codes is that buildings
will be damaged by severe earthquakes, and through damage to structural assemblages,
energy from the earthquake will be sufficiently dissipated to prevent building collapse.
Experience in recent California earthquakes, including the 1999 Loma Prieta and 1994
Northridge events, indicates that modern code provisions in zones of high seismicity are
relatively reliable in avoiding life-threatening building damage. Both events, having large
magnitudes and epicentral locations close to large population centers, resulted in fewer
than 100 fatalities. However, the economic losses associated with these earthquakes, $7
billion for Loma Prieta and more than $15 billion for Northridge, were unacceptably
large, given that events of this size can be experienced relatively frequently in zones of
high seismicity (Hamburger, 1996). As a result, public officials and members of the
insurance and financial industries are calling for building construction practices that can
better limit damage in future earthquakes.
9-1
In response to these concerns, structural engineers have initiated the development of a

concept of seismic design called Performance Based Earthquake Engineering Design
(PBEE). Philosophically, this design approach permits a building owner/operator to select
an acceptable level of building damage for a given intensity of earthquake ground
shaking which can then serve as an objective or target for the seismic design effort (see
Table 9.1). Buildings continue to be designed for the minimum levels of life safety
protection offered by the code based design approach, but PBEE offers owners/operators
the opportunity to limit business interruption economic loss and other consequences for
less severe but more probable earthquake hazards.
The performance-based design is used to design the retrofitted wall. The concept of
performance-based design and the performance-based design procedure are presented in
this Chapter. A numerical design example, using this performance-based design and
Australian Standard, is also presented.
9-2
DESIGNATION DESCRIPTION
Guidelines Vision 2000

and (SEAOC,
Commentary 1995)
(ATC, 1995)
Operational Fully Only very minor damage has

Operational occurred. The building retains
its original stiffness and
strength. Nonstructural
components operate, and the
building is available for
normal use. Repairs, if
required, may be instituted at
the convenience ofthe
building users. The risk of
life-threatening injury during
the earthquake is negligible.
Immediately Functional Only minor structural damage

Occupancy has occurred. The structure
retains nearly all its original
stiffness and strength.
Nonstructural components are
secured, and if utilities are
available, most would
function. Life-safety systems
are operable. Repairs may be
instituted at the convenience
of the building users. The risk
of life-threatening injury
during the earthquake is very
low.
9-3
Life Safety Life Safety Significant structural and

nonstructural damage has
occurred. The building has
lost a significant amount of its
original stiffness, but retains
some lateral strength and
margin against collapse.
Nonstructural components are
secure, but may not operate.
The building may not be safe
to occupy until repaired. The
risk of life-threatening injury
during the earthquake is low.
Collapse Near A limiting damage state in

Prevention Collapse which substantial damage has
occurred. The building has
lost most of its original
stiffness and strength, and has
little margin against collapse.
Nonstructural components
may become dislodged and
present a falling hazard.
Repair is probably not
practical.
Table 9-1 The definitions of performance level (Hamburger, 1996)
94
9.2 Concept of Performance-Based Design
The primary objective of current building code provisions for earthquake resistive design
is to avoid building collapse and loss of life. Recently, as the reliability of design and
construction practice in meeting this objective has improved, demand for building
designs capable of meeting other objectives, including minimization of damage,
continuous occupancy, and business interruption, has increased. Programs for the
development of performance-based design procedures have been initiated to address this
concern. However, significant barriers must be overcome to successfully implement these
concepts. Performance-based earthquake engineering implies design, evaluation, and
construction of engineered facilities whose performance under common and extreme
loads responds to the diverse needs and objectives of owner-users and society. PBEE is
based on the premise that performance can be predicted and evaluated with sufficient
confidence for the engineer and client jointly to make intelligent and informed decisions
based on building life-cycle considerations rather than on construction costs alone
(Hamburger, 1996).
In the United States, the new. generation of seismic codes for bridges is moving towards
performance-based design as a rational and cost-effective approach to seismic design.
Current seismic codes are based on accelerations and forces, which result in stronger and
more rigid structures, which in turn attract higher forces. Instead, the performance-based
design approach uses displacements and deformations along with the proper detailing to
allow the deformations to develop, permitting structures to be somewhat more flexible. In
addition, the necessary details to meet given performance criteria usually cost less to
construct than those required to meet the criteria of force-based approaches — the money
is spent where it is needed (Jackson, 2001).
To implement a performance-based design, one or more performance objectives (see

Figure 9.1) must be selected. A performance objective is a statement of the desired
building behavior, given that it experiences earthquake demands of specified severity. In
concept, the building user community is given a choice with regard to the performance
9-5
objectives for each building. For such a choice to be useful, it is necessary to define the
various behavior alternatives in terms meaningful to the layperson. Important parameters
include: the potential for loss of life, the cost of repairing sustained damage, and the
amount of time the building is out of service for repair or, in extreme cases, replacement.
While these parameters are meaningful to the public, and therefore can serve as a basis
for selecting among building performance alternatives, they are not useful as a basis for
design. An engineer cannot design for such performance specifications as a business
interruption of two weeks or a repair cost that is 20% of replacement value. Therefore, as
a prerequisite to practical implementation of performance-based engineering,
corresponding relationships must be established among parameters that are meaningful to
building users and design professionals. Since a building can experience a wide spectrum
of potential behavior states ranging from a complete absence of damage and earthquake
effects to complete collapse, establishing corresponding relationships is not a trivial task.
To complete the specification of performance objectives, particular earthquake levels for

which the performance is to be attained must be selected. This can be done on either a
probabilistic or deterministic basis. Most people easily relate to the deterministic
approach in which a specific magnitude event on a defined fault or source zone is
selected as the basis for design. This approach is most useful for determining the "worst
case" design performance objective in regions with seismicity controlled by one or more
major active faults or source zones. As an example, if it is told that the worst earthquake
ever expected to affect a planned building is an M7.5 event on a fault located 15 Km
away, the average person is capable of determining what performance is acceptable
should that event occur. However, in regions remotely located from such active sources,
this approach is less meaningful. In addition, this approach is not particularly useful for
determining secondary performance objectives such as the design earthquake for which
the Immediate Occupancy or Operational performance levels should be attained. It is
difficult for people to determine what performance should be specified for an M6.0 event
on the same fault without knowing the likelihood of such an event. However, if people
were told that an M6.0 event has an average return period of 100 years, it would be
9-6
possible to determine an appropriate performance level for such an event, based on

economic analysis (Hamburger, 1996).
higher performance
less loss
Operational Level 4
Backup 'utility services
'maintain functions; very little
damage. (Si ÷NA)
Immediate OccLlpancy Level

The building receives a "green
tag" (safe to occupy) inspection
rating; any repairs are minor.
(51.-NB)
LIfe Safety Level

Structure remains stable and
has significant reserve
capacity; hazardous
nonstructural damage is
controlled. (83-+-NC)
Cal lapse Prevention Level

The building remains standing,
but only barely; any other
damage or loss is acceptable.
(S5÷NE)
lower performance
more loss
Figure 9-1 Performance level (FEMA-273, 1997)
Several conceptual frameworks for PBEE have been developed in recent professional
efforts (SEAOC Vision 2000, FEMA 273, ATC-40). They differ in details but not in
concepts. The Figure 9.2 illustrates a global framework that identifies processes,
concepts, and major issues.
9-7
Processes Concepts issues
Figure. 9-2 A global framework for Performance-based Design
The SEAOC's Vision 2000 document recommends that buildings be constructed, based
on their intended occupancies and uses, to meet the performance objectives indicated in
Figure 9.3. Each diagonal line in the figure indicates design performance levels and
earthquakes recommended for the design of buildings of different occupancies and uses.
Individual, informed building users, of course, could select more stringent performance
objectives, if desired. In the figure, each combination of an earthquake return period and
performance level, indicated by a red diamond, represents a specific design performance
objective. The intent is that ordinary buildings provide a low risk that life be endangered
as a result of the performance of the building in any earthquake likely to effect it; and that
for frequent earthquakes, the building user not be burdened with extensive repairs or loss
9-8
of use; that buildings required for emergency response and essential public function have
a low risk of being damaged beyond a level that would permit their use, and; that
facilities housing systems and materials that would pose a hazard to many persons if
released, have a low risk of damage resulting in such release.
Design Performance Level
(50% 50
w
C- (20% - :5Q
w
=
U)
Figure 9-3 Vision 2000 performance objectives (Hamburger, 1997)
9.3 Performance-based Seismic Rehabilitation Procedure
Because of lack of proper reinforcement in the past, older facilities don't meet newer
standards, and this is where retrofit comes into the picture. To start a Performance-based
Seismic Rehabilitation Program, owners need to review their inventory and understand
which facilities may be at risk. If there is risk, the owner must prioritize and determine
9-9
which facilities are most important in order to schedule the order of retrofits and ensure
that those structures will not suffer damage in the future.
It is important to note that any step taken improves the reliability index. It is the owner's
choice how far to go, depending on the level of risk, the performance level selected, and
the amount to be spent.
The performance-based seismic rehabilitation design procedure (see Figure 9.4) includes
the steps indicated below (NEHRP, 2000).
• Review Initial Considerations (Appendix C.1)

• Select a Rehabilitation Objective for the building (Appendix C.2) including
Target building performance level and seismic hazard
• Obtain as-built information on the building and determine its characteristics,
including whether the building has historic status (Appendix C.3).
• Select a Rehabilitation Objective for the building (Appendix C.4).
• Select an appropriate Rehabilitation Method (Appendix C.6).
- Select a Rehabilitation Strategy (Appendix C.4) and perform a preliminary

design of corrective measures.
- Select an appropriate Analysis Procedure (Appendix C.4. 1).
- Perform an analysis of the building; including the corrective measures, to
verify is adequacy to meet the selected Rehabilitation Objective
- If the design is inadequate, revise the corrective measures or try an
alternative strategy and repeat the analysis until an acceptable design
solution is obtained.
• Perform Rehabilitation Design (Appendix C.6).

• Verify Rehabilitation Design
• Prepare construction documents of acceptable rehabilitation
9-10
PnorSeisrrsc Evaluation (Section 1.2) Section 1.3,1: Review Initial Considerations

• Structural characteristics (Chapter 2)
* Site seismic hazarde (Chapters land 4)
• Results from prior evaluations (Section 1.2)
Interest ii reducing seisnrrcnsk • Occupancy (not considered in this standard)
• Historic status (Appendix A)
• Economic considerations (Section C12.B.2)
• Societal issues: (Appendix A)
• Local (urisdiclional requirements (not considered in this standard)
2 Section 1.3,2: Select Rehabilitation Ob(ective

• Target Building Performance level (Section 1.5)
• Seismic Hazard (Section 1.6)
3 Section 1.3.3: Obtain As-Built lntorrrration (Chapter 2)
4 - Section 1.3.4: Select Rehabilitation Method
4A Simplified Rehabiktation (Chapters 2.10

and 11) 48 Systematic Rehabilitation (Chapters 2 through 9 and 11)
4C Other Choices
• Identify building model type • Consider deficiencies (riot in this standard)
• Consider deticiencles • Select retrahitifatiori strategy (Ctrapler2). occupancy
• Select lull or partial reJhabitltation • Select analysis procedure (Chapters 2 and 3) Demolish
• Consider general requirements (Chapter 2)
(Note: Slrnplilied Rehabilitation shall be used
tar Umited Objectives
5A 56 Section 1.3.5: Perform Rehabilitation Design

Perform Rehabilitation Design • Develop mathematical model (Chapters 3 through 9 for stillness
Determine and design and strength)
rehabilitation measures to meet • Perform force and deformalion response evaluation
applicable FEMA 310 (Chapters 2through 9 and 11)
requirements
I
• Size elements, components, and connections
(Chapters 2,5 through 9. and 11)
6A Secison 65 Section 1.3.5:

Verify Design Verify Rehabilitation Design
• Reevaluate building to assure • Appiy component acceptance otileria (Chapters 2 through 9
that rehabitiltation measures and 11)
reronve at deficiencies without • Review for confonnance with requirements ol Chapter2
creating new ones • Review for economic acceptebiilty
I
• Review for economic acceptability
I
6,IA Section 1.3.6i:Recfesign 6.2A Section 1.3.6.2: Prepare 6.18 Seclion1.3.6.1: Redesign 6.28 Section 1.3.6.2: Prepare
Unacceptable Construction Documents of Unacceptable Construction Documents
Rehabilitation Acceptable Rtihabiiilation Rehabilitation olAcceptable
Rehabilitation
Return to 4A to • Develop construction Return to4B to revise
reconsider documents analysis and or to • Develop construction
Rehabittation Objective • Segin rehabilitation 2 to reconsider documents
I
or to SA to revise Rehabilitation Ob(ective • Begin rehabilitation
• Exercise quality control
corrective measures • Exercise quality
Figure 9-4 The performance-based seismic rehabilitation design procedure (NEHRP,

2000)
9-11
9.4 A performance-based design approach for retrofitting URM low-rise

walls
URM wall is representative a brittle behaviour. It belongs to Type 3 component force

versus deformation curve (Figure 9.5). Type 3 curve is classified as force-controlled
actions. Force-controlled actions in primary and secondary components and elements
shall satisfy Equation (9-1):
KQCL QUF (9-1)
(For more details about equation 9-1 and Type 3 curve, please refer to Appendix C.5)
QI
Figure 9.5 Component force versus deformation curves (FEMA-273, 1997)
9.5 A practical design guideline for engineer
The practical design guideline has been provided in this section. It involves the following
steps.
Step 1 Seismic evaluation/assessment of the original building. (Chapter 2.2)
9-12
Step2 Determine the lateral seismic forces using a performance-based design code, such
as the FEMA 273.
Step3 Use Eqs. 9.2 and 9.3 to find lower bound lateral strength of wall (original
unretrofitting capacity).
QCL = (9-2)
=
QCL =
-
(For more details about equation 9-2 and 9-3, please refer to Appendix C.5)
Step 4 Determine the increasing lateral strength that need to retrofit
Step 5 Find the size of cable to resist this force.
Step 6 Check the Drift requirement in Table 9.2. If not safe, the design must be
modified. (For more details about Table 9.2, please refer to Appendix C.2)
Step 7 Design the connections between cable and the wall. (Appendix C-2)
Step 8 Re-evaluation of the retrofitted building. (Chapter 2.2)
Some useful practical information has been provided in Appendix C including cable
system, chemical anchor and FRP.
9-13
Table 9.2 Structural Performance level for URM walls (FEMA-274, 1997)
Elements Type Collapse Life Safety S-3 Immediate

Prevention S-5
Occupancy S-i
Unreinforced Primary Extensive Extensive cracking Minor (<1/911

Masonry Walls cracking and and some crushing width) cracking of
crushing; but wall remain in masonry at a few
portions of face place. No falling corner openings.
course shed. units. Extensive
crushing and spalling
of veneers at corners
of openings.
Secondary Extensive Same as primary. Same as primary.
crushing and
shattering; some
walls dislodge.
Drift 0.6% transit or 0.5% transit; 0.3% 0.1% transit;
permanent. permanent. negligible
permanent.
9.6 Sample design of cable system
A sample design of the cable system developed in this research program is presented in
this section. The design problem involved the single-story building shown in Figure 9.6
and Figure 9.7. The wall is to be retrofitted using the cable system, as shown in Figure
9.8. The building is 9 m x 9 m in plan, and has 3m clear height. It is enclosed by four
unreinforced end walls in each direction. The length of each wall is 3 m. A 300 mm thick
reinforced (0.25%) concrete slab was used at the roof, for parking. The walls are 110mm
thick, each. The building is assumed to be in site class A (hard rock with average shear
wave velocity, v5 >5.000ftlsec) and Earthquake Hazard level is assumed using values of
=0.25 and S1 =0.1 taken from MCE spectral response acceleration contour maps
(FEMA356).The life safety performance level is selected for the target building
performance level and the basic safety objective is selected for the rehabilitation
objective for this building.
9-14
Figure 9.6 illustrative design example, floor plan
j3m
*
9m
Figure 9.7 illustrative design example, single-story building
9-15
j3m
9m
Figure 9.8 Retrofitting of single-story building using cable system
9.5.1 Using Performance-based design according to FEMA 274
The equivalent static seismic base shear, V, in the East-West direction must be
determined first. This requires the computation of weight, W. The weight of the roof is
computed as:
W1= yxV=24x(O.3x9x9)=593KN
(i = 24 KN/m3, specific weight for concrete)
The weight of the upper half of each wall is,
W2 =yxV=19x(!x3x3xO.11x9)=75KN
= 19 KN/m3, specific weight for masonry)
9-16
The weight due to 25% of the snow load is,
W3=0.2x9x9x2=32KN
The total weight is,
W= 593 + 75 +32 = 690 KN
The fundamental period of the building is estimated as, T = h , where h is the
height to the roof level and = 0.02, 13 = 0.75 for URM buildings. (Please refer to
Appendix C.4. 1 for details)
0.75
T = 0.02 x 3 = 0.046 second
Determine the lateral seismic forces from Equation 9-4 (Please refer to Appendix C.4.1
for details)
V=C1C2C3CmSaW (9-4)
From the Appendix C.4.l, the following parameters can be determined
C1 =1.5 (T<0.lOsecond.)
C2 =1.0 (For linear procedures)
C3 = 1.0 (from Appendix C.4.1)
Cm = 1.0 (from table C-5)
Sa = 0.15, from the following calculation
9-17
= F0 = 0.9 x 0.25 = 0.2 (9-5)
= = 0.9 x 0.1 = 0.09 (9-6)
where Fa =0.9 and are site coefficients determined respectively from Appendix
C.4.1, Tables C-13 and C-14.
where
=0.25 and S1 =0.1 taken from MCE 10%5oyears spectral response acceleration
contour maps (FEMA356)
= B1)= (0.09 x 1)/(0.2 x 1) =0.4 (9-7)
and where and B1=1 for 5% damping (form Appendix C.4.1, Table C-b)
T0 = 0.2 = 0.2 x 0.4 = 0.09 (9-8)
for 0<T<T0
0.046
= — = 0.2 x — = 0.15 (9-9)
+0.41
The minimum lateral seismic forces is:
V=1.5x lxix lxO.15x690=155KN
The seismic base shear tributary to each wall;
V U
4
9-18
Knowing that the original (unretrofitted) capacity, V is 10 KN (calculated from Eqs.

9.2 and 9.3)
Determine the increasing lateral strength that needs to retrofit
KQCL QUF
U0
K 0.75
The retrofitted cable configuration is 450 from the horizontal level so that the minimum
needed breaking load for cable is:
F=
B =59KN
cos45°
Using this value to choose the suitable size of cable from the cable company catalogue.
Then the connections between the wall and cable should be designed to make sure the
cable is broken before the connections fail.
9.5.2 Using Australian Standard according to AS 1170.4 - 1993
The total horizontal earthquake base shear (V) in the East-West direction shall be
determined from the following equation
V= (9-10)
From theAS 1170.4, the following parameters can be determined
I = important factor = 1 (for structural type I)
9-19
S = site factor = 0.67 (for rock profile)
R = structural response factor = 2.75 (The selection of the structural response factor, R,
requires an explanation because this type of the walls is not included in the structural
response factor table of AS 1170.4. Observations and results of this experimental study
show that walls retrofitted by means of cables are ductile structural members with
hysteresis curves representative of element between concentrically-braced frames ( R =
4.0) and unreinforced masonry shear walls (R = 1.5). Therefore, an average value of R,
2.75 was chosen.)
T = structural period = = = 0.065

46 46
a = acceleration coefficient = 0.15 (using the same value as section 9.5.1)
1.25a 1.25*0.15
C = earthquake design coefficient = 2 = 2
=1.17
T3
Gg=G+(PcQ
where
G = dead load (see AS 1170.1)
Q = live load (see AS 1170.1)
= live load combination factor (see AS 1170.1)
This requires the computation of weight, Gg .The weight of the roof is computed as:
W1= yxV=24x(0.3x9x9)=593KN
(y = 24 KN/m3, specific weight for concrete)
9-20
The weight of the upper half of each wall is,
W2 =yxV= 19x(!x3x3xO.11x9)=75KN
(y = 19 KN/m3, specific weight for masonry)
The weight due 40% of the snow load is,
W3= 0.4 x 9 x9 x 2 = 64 KN
The total weight is,
Gg=593 +75+64=732 KN
The minimum lateral seismic force is:
V=
2.75
The seismic base shear tributary to each wall;
V U
4
Knowing that the original (unretrofitted) capacity, V is 10 KN (using the same value
as section 9.5.1)
Determine the increasing lateral strength that needs to retrofit
9-21
The retrofitted cable configuration is 450 from the horizontal level so that the minimum
needed breaking load for cable is:
F=
B
cos45°
=59KN
Using this value to choose the suitable size of cable from the cable company catalogue.
Then the connections between the wall and cable should be design to make sure the cable
is broken before the connections fail.
9.5.3 The difference between Performance-Based Design and Australian

Standard
As indicated in 9.5.1 and 9.5.2, the two designs received the similar results. That is
because Australian Standard is only based on the Basic Safety Objective to design the
buildings to satisfy the Life Safety Performance level and the same objective and
performance level were selected when the performance-based design was used.
The difference between Performance-Based Design and Australian Standard is that one
has choices to decide the rehabilitation objective and target building performance level
when the performance-based design was adopted and such options don't exist in the
current Australian Standard.
If the different rehabilitation objective and target building performance level are selected,
the response spectrum acceleration will be different so that the lateral seismic force will
be different.
Buildings continue to be designed for the minimum levels of life safety protection offered
by the Australian Standard design approach, however, Performance-Based Design offers
9-22
owners/operators the opportunity to limit business interruption economic loss and other
consequences for less severe but more probable earthquake hazards.
9.7 Difficulty to implement PBD
Although performance-based earthquake engineering design is a great idea, but there are
some difficulties that need to conquer to implant PBD.
Some of the most significant barriers to the implementation of performance-based design

are the economic and competitive pressures faced both by the building designer and the
building developer. Most building developers in today's market are not interested either
in personal occupancy or long-term ownership of the property. As a result, they believe
earthquakes of sufficient size to cause severe damage are unlikely to occur during the
time the developer controls the building. Such developers are unlikely to invest in the
construction of a building with enhanced earthquake performance characteristics unless
they believe the added cost will increase either the rental or sales value of the property.
Unfortunately, neither is the case.
Despite statements to the contrary following damaging earthquakes, the public generally
believes that building codes provide adequate protection against earthquake-induced
losses. The public is generally unaware of the performance standards inherent in the
building codes or the likely behavior of buildings in earthquakes. Therefore, they are
unlikely to discriminate in their selection of a building based on potential earthquake
performance. Thus, there is limited market value added to a building designed and
constructed for such enhanced performance.
The exceptions are those buildings developed specifically for the occupancy of a business
or institution knowledgeable in earthquake risk management. Most large corporations and
institutions today have an in-house risk management group, responsible for protecting the
organization against large catastrophic losses. In recent years, many such organizations
with properties in regions of high seismicity have attempted to implement performance-
9-23
based design concepts in the development of new properties. In some cases, however,
more conventional design approaches were adopted because the additional cost inherent
in the development of buildings with enhanced earthquake performance characteristics
could not be justified by management.
The adoption of a performance-based engineering approach can result in a significant

increase in building development costs. Design procedures conforming to current code
requirements employ simple analytical techniques and evaluate building response to only
a single earthquake demand level. Performance-based earthquake designs typically
address multiple-performance goals. To reliably provide buildings capable of meeting
these multiple-performance objectives, it is not only necessary to employ more complex
and time-consuming analytical techniques, but also to evaluate the building's probable
performance for several earthquake demand levels. This translates to higher engineering
fees. As a result, structural engineers are unlikely to provide performance-based design
services in a competitive market unless the building developer specifically requests them,
or it is required in the building code. Since developers are unlikely to request
performance-based design, adoption of performance-based requirements by the building
codes will be necessary for their widespread use. Adoption of performance-based
approaches in the building codes will be difficult to accomplish, as significant resistance
will occur based on economic grounds.
Perhaps the most significant barrier to the adoption of performance-based earthquake

resistive design is the lack of control exercised by the structural engineer over the
building design process. The structural engineer's role is typically limited to designing
and detailing a building's structural components. Design criteria development and
discussions with the building developer are typically handled by, or through, the
architect, who has limited understanding of earthquake performance issues and
consequently, is not an effective advocate for performance-based design issues. The basic
site selection, building configuration, and even the framing system used for a building, all
highly important factors in determining the building's performance, are typically decided
by the architect and developer, often with minimal consultation by the structural
engineer. Many structural engineers find it difficult to persuade the architect that a
9-24
selected system is less than appropriate for a building, for fear that the architect will find
another structural engineer who will support the original design concept, regardless of
whether or not it will perform adequately.
Even if the necessary design procedures were available today and the engineer could
design structures with predictable and acceptable seismic performance characteristics,
this would not assure that the buildings themselves will meet the intended seismic
performance objectives. Buildings are a complex collection of systems-structural,
mechanical, fire protection, electrical, architectural, and others. The performance of the
individual components of the nonstructural systems can be as important to the overall
earthquake performance of a building as is the performance of the structure, particularly
for enhanced performance objectives that address building operability. Yet the structural
engineer has little participation in the design of these systems for seismic resistance and
is often unaware of any details for installing these systems. Consequently, poor
earthquake performance of these systems is common, even in buildings that have been
designed to very stringent structural standards.
An example of such behavior is the performance of the Olive View Hospital, in Sylmar,
California, during the 1994 Northridge Earthquake (see front and back covers). The Olive
View Hospital replaced a previous facility, destroyed by the 1971 San Fernando
Earthquake. It was designed to criteria that were substantially in excess even of those
enforced by the State of California for construction of hospitals, since the 1971 event.
This building was located within a few kilometers of the epicenter for the Northridge
Earthquake, with recorded accelerations at the roof of the structure exceeding 2g. The
building structure behaved very well and could be said to meet the performance
requirements for the Operational performance level. However, utility piping within the
building failed, resulting in extensive water damage to the facility and forcing its closure.
As a result, the building's performance, as opposed to the structure's performance, was
below the Immediate Occupancy level. While this performance was still excellent, it did
not meet the intent of the design-to provide immediate post-earthquake health-care
service.
9-25
Chapter 10: Summary and Conclusions
CHAPTER 10
SUMMARY AND CONCLUSIONS
10.1 Summary
Many existing unreinforced masonry low-rise shear walls, built before the 1989
Newcastle Earthquake, were not designed to resist earthquakes. As a result, their
seismic behaviour may not be adequate, and may need to be retrofitted. A new
retrofitting procedure has been developed in the current research project. The
retrofitting strategy includes a cable system, designed to raise both strength and
ductility to a level compatible with today's standards.
Cable consists of a number of wires or strands and has high tensile strength, lightness
and high corrosion resistance. These materials can absorb tensile stress and increase
overall element stiffness, ductility and bearing capacity. Using cable system for
seismic retrofitting applications has other advantages such as architectural versatility,
durability, and no loss of valuable space. Furthermore, it does not add a significant
mass to the existing building and can remain unchanging dynamic properties for
structures. Therefore, Cable would be seen to be particularly suitable for seismic
retrofitting materials.
Four unreinforced masonry walls and eight unreinforced masonry walls retrofitted
with cable systems were designed, constructed and tested under combined gravity
load and in-plane deformation reversals. Seven unreinforced masonry walls retrofitted
with FRP were also conducted in this research for the comparison purpose. The
testing wall specimens had a thickness of 110 mm, an aspect ratio of 1.0 and a height
10-1
of 1000 mm. All walls were tested under combined constant gravity load and
incrementally increased in-plane lateral displacement reversals.
The cable system consists of diagonal cables (19 single strands, Diameter 10mm),
attached to the walls using through-thickness bolts. Stiff steel angles and anchor bolts
were used to connect the cable to the foundation. The cable was fixed on one side of
the wall only.
The type of FRP used for this experiment was Sika CarboDur S (Width 30mm,
thickness 1.2mm and tensile strength 2800 MPa) and epoxy adhesive for bonding
Sika CarboDur Strips was Sikadur-30. The FRP was fixed for one side of the wall
only.
Analyses of walls were conducted using the nonlinear F.E.M model (ABAQUS). The
results computed using F. E. M. were compared with recorded experimentally.
The comparisons showed good agreement. Furthermore, the performance-based
design procedure was also applied to design URM walls retrofitted with the cable
system.
10.2 Conclusions
Experiments conducted in this study show that the cable system, proposed to retrofit
low-rise masonry walls under in-plane lateral loads, is one of the most effective ways
in significantly increasing the in-plane strength, ductility and energy-dissipation
capacity. For the particular wall specimens considered in this thesis, the addition of
cable system increased the lateral load resistance of the wall by approximately 25 KN,
making the retrofitted walls 2 times stronger that the unreinforced masonry.
Experiments also indicate that the cable carried about the 50% of force acted on the
whole wall after major cracks occurred. The results proved that that URM wall
retrofitted by cable system can effectively increase the load capacity after the onset
cracking of the wall.
10-2
For the addition of FRP, the improvement of the ultimate lateral load resistance of the
retrofitted walls is 3 times the capacity of un-reinforced wall. Experiments also
indicate that the FRP carried about the 50% of force acted on the whole wall even
after early loading stage, which is different to cable system. This is because FRP is
fully attached to the wall. The results proved the URM wall retrofitted by FRP system
becomes a whole new structure to carry the force together.
Seismic retrofitting of unreinforced masonry walls with FRP or cable system proved
to be an effective and reliable strengthening alternative. However as discussed earlier,
each retrofitting technique has its own advantages and disadvantages. Although FRP
is more effective in increasing the ultimate load carrying capacity of URM walls than
cable, the high cost and no complete knowledge of its mechanical behaviour may be a
major concern as well as the aesthetic appearance of the building. On the other hand,
although cable system could be a low cost alternative, the fragility of the masonry
does not allow a simple fixing of the cable system.
In this study, the finite element method is adopted for numerical modelling to
simulate the masonry behaviour. Masonry is a composite material that consists of
units and mortar joints. Macro-modelling is adopted to model this composite material
due to the practical purpose.
Two nonlinear F.E.M models were developed to predict the strength and ductility of
the walls by cable system and FRP and validate the experimental results. One model
is developed for unreinforced masonry walls retrofitted by cable system and the other
model is developed for unreinforced masonry walls retrofitted by FRP. It has been
found that the analytical and experimental results are agreed well.
The proposed numerical models are based on ABAQUS concrete model and are
modified to simulate the special characterises of masonry. The models take into
account the material nonlinearities as well as damage due to progressive cracking.
The model employs oriented damaged elasticity (smeared cracking) and isotropic
compressive plasticity to represent the inelastic behaviour of masonry. The model is
intended to model the masonry behaviour for relatively monotonic loadings under
fairly low confining pressures. Due to the limitations of ABAQUS, the model makes
10-3
no attempt to include prediction of cyclic response or of the reduction in the elastic

stiffness caused by inelastic straining.
Cracking is assumed to be the most important aspect of the behaviour and

representation of cracking and of post cracking behaviour dominates the modelling.
The current model adopts the concept of crack detection surface. Cracking is assumed
to occur when the stress reaches a failure surface that is called the "crack detection
surface." This failure surface is a linear relationship between the equivalent pressure
stress, and the Mises equivalent deviatoric stress.
The fixed orthogonal cracks model is adopted so that the maximum number of cracks
at a material point is limited by the number of direct stress components present at that
material point of the finite element model (for example, a maximum of three cracks in
three-dimensional, axisymmetric, and plane strain problems or a maximum of two
cracks in plane stress problems). Once cracks exist at a point, the component forms of
all vector and tensor valued quantities are rotated so that they lie in the local system
defined by the crack orientation vectors (the normals to the crack faces). The model
ensures that these crack face normal vectors are orthogonal so that this local system is
rectangular Cartesian. Crack closing and reopening can take place along the directions
of the crack surface normals. The model neglects any permanent strain associated
with cracking; that is, we assume that the cracks can close completely when the stress
across them becomes compressive.
An isotropically hardening "compressive" yield surface was adopted to model for the
inelastic response when the principal stresses are dominantly compressive. For
tension softening, the bilinear model is adopted in the current model. The effect of
cracked shear retention is also considered in the models.
The ultimate shear strength of the retrofitted wall can be simply obtained by the
F.E.M model. A F.E.M approach, described in this thesis, gives analytical results that
are in good agreement with experimental results. The global lateral force-
displacement behaviour of all retrofitted walls can also be captured with a reasonable
degree of accuracy using the F.E.M. model described in this thesis.
10-4
Finally, a performance-based design procedure is proposed as a design tool that

engineers may use to retrofit existing non-ductile low-rise walls. A numerical design
example using this performance-based design and Australian Standard to design the
walls retrofitted by cable system is alsp provided
The difference between Performance-Based Design and Australian Standard is that

you have choices to decide the rehabilitation objective and target building
performance level when you use the performance-based design and you don't have
other choice when you use the current Australian Standard.
If the different rehabilitation objective and target building performance level are
selected, the response spectrum acceleration will be different so that the lateral
seismic force will be different.
Buildings continue to be designed for the minimum levels of life safety protection
offered by the Australian Standard design approach, but Performance-Based Design
offers owners/operators the opportunity to limit business interruption economic loss
and other consequences for less severe but more probable earthquake hazards.
Although performance-based earthquake engineering design is a great idea, but there
are some difficulties such as the economic and competitive pressures faced both by
the building designer and the building developer that need to conquer to implant PBD.
As the conclusion, the results from both of experiments and numerical modelling
showed that both the strength and ductility of tested specimens were significantly
enhanced with this new technique. Seismic retrofitting of unreinforced masonry walls
with cable system proved to be an effective and reliable strengthening alternative. The
limitations of this model are that this current model couldn't be predicted well for the
retrofitted walls and couldn't model the FRP debonding effect. Due to the limitations
of ABAQUS, the model makes no attempt to include prediction of cyclic response or
of the reduction in the elastic stiffness caused by inelastic straining.
10-5
10.3 Reconimendations
The following recommendations àan be made based on the findings of current

research.
• The proposed retrofitting technique is limited to small size low-rise URM

walls to avoid local failure at the connection between the cable and wall.
• To avoid out-of-plane movement and to increase wall redundancy, it is highly

recommended that the cable systems are applied on both sides of the wall.
• The improvement of the ultimate lateral load resistance of the retrofitted walls
by FRP is higher than cable system, but the wall retrofitted by cable system is
cheaper than the wall retrofitted by FRP.
• The anchor bolts used to attach the cable system to the masonry walls must not
be fastened (torque) to a level that could damage the masonry wall.
• Model accuracy can be improved by using realistic material property values

and by properly defining boundary conditions.
• All cable should be lightly pretensioned during the installation.
• Drilling to brick should be carefully conducted to avoid further damage as

most brick is brittle.
10-6
Reference
REFEREN CE
Australian Standard, (1998). "Strengthening existing buildings for earthquake", AS 3826-

1998, Standards Association of Australia.
Australian Standard, (1993). "Minimum design loads on structures", AS 1170.4-1993,

Standards Association of Australia.
Australian Standard, (2001). "Masonry structures", AS 3700-2001, Standards Association

of Australia.
American Segmental Bridge Institute (ASBI). (2000)"FRP compared to stainless steel

reinforcement". Communication News.
Alimoradi. A. (2003). "State-of-the-art in performance-based design optimisation-inference

to consequence-based engineering optimisation." A report to the Consequences-Based
Engineering Institute of the Mid-America Earthquake Center, University of Illinosis at
Urbana- Champaign, 1-24
Abboud, B. E., Hamid, A. A. and Harris, H. G. (1996). "Flexural behavior of reinforced

concrete masonry walls under out-of-plane monotonic loads." ACI Structural Journal, 93
(3), 327-335.
Anthoine, A. (1992). "In-plane behaviour of masonry: A literature review." Report EUR

13840 EN, Commission of the European Communities, JRC — Institute for Safety
Technology, Ispra, Italy.
Abrams D.P. (1998). New perspectives on seismic rehabilitation. Pro. of Asia-Pacific

workshop on Seismic Design and Retrofit of Structures, Taipei.
R-1
Reference
Anand, S. C. and Rahman, M. Ayubar. (1994). "Accurate estimation of interface shear

stresses in composite masonry wails." Journal of Structural Engineering, 120 (3), 998-
1015.
Au., S. (1987). "Concentrated loads on solid masonry." PhD Thesis, The University of
Newcastle.
Albert, M. L., Elwi, A. E. and Roger Cheng, J. J. (2001). "Strengthening of unreinforced

masonry walls using FRPs." Journal of Composites for Construction, 76-83.
ABAQUS, (2000). ABAQUS theory manual, HKS
Arya, S. K., ASCE, A. M. and Hegemier, G. A. (1982). "Finite element method for
interface problems." Journal of Structural Division, 108 (ST2), 327-342.
ARA. (1997). "Homeowner's Guide — Earthquake safeguards." The Engineered Wood

Association.
Banthia N and Boyd AJ. (2000). "Sprayed fibre-reinforced polymers for repairs." Canadian
Journal of Civil Engineering, 27 (5), 907-9 15.
Badoux M. and Jirsa J.O. (1990). "Steel bracing of RC frames for seismic retrofitting."
Journal of Structural Engineering, ASCE, 116 (1), 55-74.
Barbosa, A. F. and Ribeiro, G. D. (1998). "Analysis of reinforced concrete structures using

ANSYS nonlinear concrete model." Computational Mechanics, 1-7.
Bailey, K. A. (1983). "The analysis of plain, brick masonry under in-plane loads." PhD
Thesis, University of the West Indies.
Benjamin B. (1995). "Reinforced concrete shear for repair of damaged masonry walls."
JABSE Sym. San Francisco, Extending the of structures, 373-378.
R-2
Reference
Beolchini, G. C., Grub, F. and Valente, G. (1997). "Analyses of repair and maintenance of
a masonry wall", Structural studies; repairs, and maintenance of historical buildings, 3,
367-376.
Bazant, P. and Cedolin, L. (1980). "Fracture mechanics of reinforced concrete", Journal of

the Engineering Mechanics Division, Vol. 106, No. EM6, 1287-1306.
Binda, L., Fontana, A. and Frigerio, G. (1995). "Mechanical behaviour of brick masonries
derived from unit and mortar characteristics." Proc. mt. Brick and Block Masonry Conf.,
J.W. de Courcy, Elsevier Applied Science, London, UK, 378-402.
Backes, H. P. (1985). "On the behaviour of masonry under tension in the direction of the
bed joins." Dissertation, Aachen University of Technology.
Bruneau, M. (1995). "Performance of masonry structures during the 1994 Northridge (Los
Angeles) earthquake." Canadian Journal of Civil Engineering, 22, 378-402.
Bruneau, M. (1994). "Seismic evaluation of unreinforced masonry buildings-a state-of-the-

art report." Canadian Journal of Civil Engineering, 21,512-539.
BIA. (1986). "Anchor Bolts for Brick Masonry." Technical Notes on Brick Construction
44, Brick Institute of America, Reston, Virginia, 1986.
Casabonne, C. (2000). "Masonry in the seismic areas of the Americas, recent investigation
and developments." Proceeding of structure engineering and material, 319-327.
Calvi, G.M. and Magenes G. (1991). "Experimental evaluation of seismic strength of old
masonry structures." Proceeding of the International Brick/Block Masonry Conference,
Calgary, .Canada, 319-327.
R-3
Reference
Cestelli-Guidi, C.(1983). "Strengthening of building structures- therapy." IABSE

Symposium Venezia 1983., Strengtheningof building structures-diagnosis and therapy, 81-
114.
Comti Euro-International Du Beton. (1997). Fastenings for seismic retrofitting-state of the

art report. London, Thomas Telford.
Cook, R. D., Malkus, D. S., and Plesha, M. E. (1989). Concept and applications of finite
element analysis, John Wiley & Sons, mc, Canada.
Chen, W.F. (1982). Plasticity in reinforced concrete. United States of America, McGraw-
Hill Book Company.
Cheung M., Foo S. and Granadino J. (2000). Seismic retrofit of existing buildings:
innovative alternatives. Proceedings of mt. Conf. on the Seismic Performance of
Traditional Buildings, Turkey.
Comite Euro-International Du Beton. Fastenings for seismic retrofitting-state of the art

report. 1997; London, Thomas Telford.
CSIRO Building, Construction and Engineering. (1999). "Future issues in performance-

based design in commercial buildings." CSIRO, 1-43
DAIS, (2001). "Strengthening existing buildings for earthquakes." Building Management,

Department for Administrative and Information Services, South Australia.
Dhanasekar, M., Page, A.W. and Kleeman, P.W. (1984). "Finite element model for the in-
plane behaviour of brick masonry." Ninth Australasian Conference on the Mechanics of
Structures and Materials, ACMSM 9. Sydney, 262-267.
R-4
Reference
Dhanasekar, M., Page, A.W. and Kleeman, P.W. (1985a). "Biaxial stress-strain relations
for brick masonry." Journal of the Structural Engineering, ASCE, Vol.111, No.5, 1085-
1100.
Dhanasekar, M., Page, A.W. and Kleeman, P.W. (1985b). "The failure of brick masonry
under biaxial stress." Proc. Instn. Civ. Engrs. Part 2, No. 79, 295-3 13.
Dhanasekar, M. (1985c). "The performance of brick masonry subjected to in-plane

loading." PhD Thesis, The University of Newcastle.
Dawe, J. L. and Aridru, G. G. (1993). "Prestressed concrete masonry walls subjected to

uniform out-of-plane loading." Canadian Journal of Civil Engineering, 20, 969-979.
Drysdale, R. G. and Essawy, A. S. (1988). "Out-of-plane bending of concrete block walls."

Journal of Structural Engineering, 1(114), 121-133.
Drysdale, R. G., Hamid, A.A., and Baker, L.R. (1994). Masonry structures, behaviour and
design, Prentice-Hall, Englewood Cliffs, N.J.
Ehsani M. R, Saadatmanesh H. and Velazquez-Dimasi JI. (1999). "Behaviour of retrofitted

URM walls under simulated earthquake loading." Journal of Comp for Constr, ASCE, 3
(3): 134-142.
Ehsani, M. R., Saadatmanesh, H., and Al-Saidy, A. (1997). "Shear behavior of URM
retrofitted with FRP overlays." J. Comp. for Constr., ASCE, 1(1), 17-25.
Ehsani, M. R., Saadatmanesh, H., and Velazquez-Dimasi, J. I. (1999). "Behavior of

retrofitted URM walls under simlulated earthquake loading." J. Comp. for Constr., ASCE,
3(3), 134-142.
R-5
Reference
Erbay, 0. 0. and Abrams, D.P. (2001). "Seismic rehabilitation of unreinforced masonry

shear walls." nisee, 373-384.
Fanning, P. (2001). "Nonlinear models of reinforced and post-tensioned concrete beans."

Electronic Journal of Structural Engineering, 2, 111-119.
FEMA. (1992). "NEHRP guidelines for the seismic rehabilitation of buildings." Report No.
FEMA-172, Federal Emergency Management Agency, Building Seismic Safety Council,
Washington, D.C.
FEMA. (1997a). "NEHRP guidelines for the seismic rehabilitation of buildings." Report
No. FEMA-273, Federal Emergency Management Agency, Building Seismic Safety
Council, Washington, D.C.
FEMA. (1997b). "NEHRP commentary on the guidelines for the seismic rehabilitation of
buildings." Report No. FEMA-274, Federal Emergency Management Agency, Building
Seismic Safety Council, Washington, D.C.
Fiorato, A.E., Oesterle, R.G., and Corley, W. G. (1983). "Behaviour of Earthquake

Resistant Structural Walls Before and After Repair." ACI Journal, Detroit, 403-4 13.
Foliente, G.C. (2000). "Developments in performance-based building codes and standards."
Ganju, T. N. (1997). "Non-liner finite element analysis of clay brick masonry."

Proceedings of the Australasian conference on the mechanics of Structures and Materials.
59-65.
Ganz, H. R and Thurlimann, B. (1982). "Tests on the biaxial strength of masonry." Report
No. 7502-3, Institute of Structural Engineering, ETH Zurich, Zurich, Switzerland.
R-6
Reference
Guggisberg, R and Thurlimann, B. (1987). "Experimental determination of masonry

strength parameters." Report No. 7502-5, Institute of Structural Engineering, ETH Zurich,
Zurich, Switzerland
Gigla, B., and Wenzel, F. (1997). "The repair of historic masonry structures by injection
thj
anchors." International Brick/Block Masonry Conference, Tongji University,
Shanghai, China, 1154-1167.
Garrity, S. (1995). "Strengthening masonry structures with retro-reinforcement." IABSE

Symposium San Francisco 1995.,Extending the lifespan of structures, 1399-1404.
Giordano, (2002). "Numerical modelling of masonry structures using the abaqus'

"concrete" model." ABA QUS users' Conference, 24 (8), 1057-1069
Hamburger, R. (1996). "Performance-Based Seismic Engineering: The Next Generation of

Structural Engineering Practice." Peiformance-Based Seismic Engineering, 1-9
Hamburger, R.O. (1997). "A framework for performance —based earthquake resistive
design" (http://nisee.berkeley.edu/lessonsIhamburger.html)
Hamburger, R.O. (1999). "The challenge of performance-based design."

(http://peer.barkeley.edu/news/l999jan/challenge.htm1)
Hamid, Ahmad A., Chia-Calabria, Catherine and Harris, Harry G. (1992). "Flexural
behavior of joint reinforced block masonry walls." ACI Structural Journal, 89 (1), 20-26.
Hamilton III, H. R. and Dolan, C. D. (2001). "Flexural capacity of glass FRP strengthened
concrete masonry walls." Journal of composites for construction, 170-179.
R-7
Reference
Heintz, J. A.,and Wosser, T. D.(1995). "Strengthening of an unreinforced masonry

building." JABSE Symposium San Francisco 1995.,Extending the lifespan of structures,
163- 168.
Hiraishi.H et al. (1998). "New framework for performance based design of building
structures —Design flow and social system-" Wind and Seismic Effects, 234-247
Horton, R. T. and Tadros, M. K. (1990). "Deflection of Reinforced Masonry Members."
ACI Structural Journal, 87 (4), 453-463.
Jackson, T.B. (2001). "How to meet seismic requirements for bridge."

(http://www.betterrods.com/articlesfbrfebOa.htm)
Kachlakev, D., Miller, T. and Yim, S. (2001). "Finite element modelling of reinforced
concrete structures strengthened with FRP laminates." Final Report, Oregon Department of
transportation Research Group, Washington.
Khattab, M. H. (1993). "In-plane behaviour of grouted concrete masonry under biaxial

states of stress." PhD Thesis, McMaster University.
Klingner, R.E. (1997). "performance-based design of masonry structures for seismic

loads." nisee, 33-40.
Korany, Y., Drysdale, R., and Chidiac, 5. (2001). Retrofit of Unreinforced Masonry
Buildings: The State-of-The-Art. The 9th Canadian Masonry Sym, Fredericton, New
Brunswick, June, CD Paper RETFITO8.
Kolsch, H. (1998). "Carbon fiber cement matrix (CFCM) overlay system for masonry
strengthening." J. of Comp. for Constr., ASCE, 2(2), 105-109.
Krawinlder, H. "Advancing performance-based earthquake engineering"

(http://nisee.berkery.edu/lessons/krawinkler.html)
R-8
Reference
Lourenco, P. B. (1995). "An orthotropic continuum model for the analysis of masonry
structures." Report 03-21-1-31-27, Deift University of Technology, Deift, The Netherlands.
Lourenco, P. B. (1996). Computational strategies for masonry structure, Delft University

Press, The Netherlands.
Lotfi, H. R and Shing, P. B. (1993). "Analysis of masonry walls with smeared and discrete
crack models." Proceedings of the Symposium on Structural Engineering in Natural
Hazards Mitigation, 1179-1184.
Lotfi, H. R and Shing, P. B. (1994). "Interface Model Applied to Fracture of Masonry

Structures." Journal of Structural Engineering, 120 (1), 63-80.
Lofti, H. R. (1992). "Finite element analysis of fractures of concrete and masonry

structures." PhD Thesis, Tehran University, fran.
Lurati, F and Thurlimann, B. (1987). "Test in concrete masonry walls." Report No. 8401-3,
Institute of Structural Engineering, ETH Zurich, Zurich, Switzerland
Magenes, G. and Calvi, G. M. (1997). "In-plane seismic response of brick masonry walls."
Earthquake engineering and structural dynamics, Vol 26, 1091-1112.
Manzouri, T., Schuller, M. P., Shing, P.B.,and Amadei, B. (1996). "Repair and retrofit of
unreinforced masonry structures." Earthquake Spectra., 12(4), 903-922.
Mayes RL. (1989). "Design of structures with seismic isolation." The Seismic Design
Handbook, 413-438.
Meichers R.E. (1990). Newcastle Earthquake Study, The Inst. Of Eng., Australia.
R-9
Reference
Mehta, S. and Saadeghvaziri, M. A. (1998). "Analysis of unreinforced masonry (URM)

walls and evaluation of retrofitting schemes for YRM structures." Structural Engineering
and Mechanics, 6 (7), 801-8 15.
Mele E and De Luca A. (1994). State of the art report on base isolation and energy
dissipation. STESSA, Proceedings mt Workshop, organised by the European Convention
for Constructional Steelwork, 631-658.
Mercer J. C. (1998). "Simple low cost retrofit procedures for historic unreinfroced masonry
buildings." Proc. of the 1998 ASCE Structures Congress.
Miranda, E. (1997). "Stremgth reduction factors in performance-based design."

(http://nisee.berke1ey.edu/1essons/miranda.htm1)
Moaveni, 5. (1999). Finite element analysis, Prentice-Hall, mc, New Jersey.
Nagy, L. I. (1989). Introduction to finite element analysis and msc/nastran utilization,

THE MACNEAL-SCHWENDLER CORPORATION, California.
Page, Adrian W. (1978). "Finite element model for masonry." Journal of Structural
Division, 104 (ST8), 1267-1285.
Page, A. W. and Shrive, N. G. (1990). "Concentrated loads on hollow concrete masonry."

ACI Structural Journal, 87 (4), 436-444
Page A.W. (1992). "The design, detailing and construction of masonry — the lessons from
the Newcastle Earthquake." Australian Civil Eng. Transactions, Vol. CE 34, No. 4, 1992,
343-353.
Pierepiekarz M. (2000). New research can help building owners who are shaken up by
quake hazards, Seattle Daily Journal of Commerce online edition.
R-10
Reference
Pincheira, J. A. and Jirsa, J. 0. (1992). "Post-tensioned bracing for seismic retrofit of RC

frames." Earthquake Engineering, Tenth World Conference, 5 199-5204.
Plencnik, J., Cousins, T., and O'Conner, E. (1984). "Strengthening of unreinforced

masonry buildings." J. Struct. Engrg., ASCE, 112(5), 1070-1087.
Potter R. J. (1994). "Earthquake and Masonry — an overview." Third National Masonry

Seminar, Brisbane, 15.1-15.10.
Pauley, T. and Priestley, M.J.M. (1992). Seismic design of reinforced concrete and
masonry buildings, Wiley, New York.
Rai, D. C., Goel, S. C. and Holmes, W. T. "Seismic strengthening of URM buildings with
steel bracing." 697-705.
Renda V. Tirelli D, Magonette G and Molina J. (2000). "Contribution of the joint research
centre to the charac.terisation of shape memory alloys and full-scale validation tests of
retrofitting techniques for masonry shear walls typical of the architectural heritage", mt
Millennium Congress — Archi, Paris, France.
Rahman, M. A. and Anand, S. C. (1994). "Empirical Mohr-coulomb failure criterion for

concrete block-mortar joints." Journal of Structural Engineering, 120 (8), 2408-2422.
Rot, J. G. (1991). "Numerical simulation of cracking in structural masonry." Heron, 36 (2),

49-63.
Rodriguez, R., Hamid, A. A. and Larralde, J. (1998). "Flexural behabior of post-tensioned

concrete masonry walls subjected to out-of-plane loads." ACI Structural Journal, 95 (1),
70.
R-11
Reference
Saliba, J. F., Al-Allad, Riad S. amd Sawaya, George E. (1996). "Use of finite elements in
masonry structures." The Masonry Society Journal, 9-16.
Sayed-Ahmed, E. Y. amd Shrive, N. G. (1995). "Numerical analysis of face-shell bedded

hollow masonry walls subject to concentrated loads." Can. J. Civ. Eng, 22, 802-818.
SC SOLUTIONS. "Nonlinear analysis using performanced-based design procedures."

(http://www. scsolutions .comlperformance.html)
Shing, P. B., Noland, J. L., Ekiamerus, E. and Spaeh, H. (1989). "Inelastic behavior of
concrete masonry shear walls." Journal of Structural Engineering, 115 (9), 2204-2225.
Shing, P. B., Schuller, M., Soskere, V. S. and Carter, E. (1990). "Flexural and shear
response of reinforced masonry walls." ACI Structural Journal, 87 (6), 646-656.
Shing, P. B., Schuller, M. and Hoskere, V. S. (1990). "In-plane resistance of reinforced

masonry shear walls." Journal of Structural Engineering, 116 (3), 6 19-640.
Sittipunt, C., Wood, S. L., Lukkunaprasit, P. and Pattararattanakul. P. (2001). "cyclic

behavior of reinforced concrete structural walls with diagonal web reinforcement." ACT
Structural Journal, 554-563.
Spyrakos, C. C. (1996) Finite element modeling in engineering practice, Algor, mc, the
United States of America.
Stasa, F. L. (1985). Applied finite element analysis for engineers, CBS, the United States
of America.
Structural Engineers Assn. Of California (SEAOC), Vision 2000 Committee. (1995).

"Performance based seismic engineering of building: Volume I, Part 1: interim
R-12
Reference
recommendations; Part 2 : conceptual framework; Volume H, Part 3: preliminary
Northridge lessons; Part 4: moving the Blue toward performance based engineering."
nisee.
Soroushian, P., Obaseki, K. and Choi, Ki-Bong. (1988). "Nonlinear modeling and seismic
analysis of masonry shear walls." Journal of Structural Engineering, 114 (5), 1190-1119.
Taghdi, M., Bruneau, M. and Saatcioglu, M. (2000). "Analysis and design of low-rise
masonry and concrete walls retrofitted using steel strips." Journal of Structural
Engineering, 126 (9), 1026-1032.
Takenaka Corp. (2001). "Hybrid base isolation system used in super high-rise office
building for the first time in the world." News Release, www.takenaka.co.jp.
Tena-Colunga, A., EERI, M. (1996). "Some retrofit options for the seismic upgrading of
old low-rise school buildings in Mexico." Earthquake Spectra., 12(4), 883-902
Teran-Gilmore, A., Bertero, V. V., and Youssef, N.(1996)."Seismic rehabilitation of

framed buildings infilled with unreinforced masonry walls using post-tensioned steel
braces." Earthquake Spectra., 12(4), 863-882
Tinazzi, D., and Nanni A.(2000)."Assessment of technologies of masonry retrofittingwith

FRP." Center for Infrastructure Engineering Studies, Univ. of Missouri -Rolla
Taghdi, M. (1998), "Seismic retrofit of low-rise masonry and concrete walls by steel
strips." PhD Thesis, Univ. of Ottawa.
Taghdi, M., Bruneau, M., and Saatcioglu, M. (2000). "Seismic retrofitting of low-rise
masonry and concrete walls using steel strips." J. Struct. Engrg., ASCE, 126(9), 1017-
1025.
R-13
Reference
Taghdi, M., Bruneau, M., and Saatcioglu, M. (2000). "Analysis and design of low- rise
masonry and concrete walls retrofitted using steel strips." J. Struct. Engrg., ASCE, 126(9),
1026-1032.
Teran-Gilmore, A., Bertero, V. V., and Youssef, N.(1995)."Seismic rehabilitation of

framed buildings infihled with unreinforced masonry walls using post-tensioned steel
braces." Report No. UCBJEERC-95/06, Earthquake Engineering Research Center, College
of Engineering, Univ. of California at Berkeley
Tinazzi, D., Tumialan, G., Myers, J. J., and Nanni A."Field evaluation of unreinforced
masonry walls strengthened with FRP composites subjected to out-of-plane loading."
Center for Infrastructure Engineering Studies, Univ. of Missouri -Rolla.
Tinazzi, D., Arduini, M., Modena, C, and Nanni A.(2000). "FRP-Structural repointing of
masonry assemblages." Proc., 3rd Inter. Conf on Advanced Composite Materials in Bridges
and Strucutres, Ottawa, Canada, 585-592.
Tinazzi,D., Modena C.,and Nanni A. (2000). "Strenthening of masonry assemblages with

FRP rods and laminates." International meeting on composite materials,PLAST
2000,Proceedings,Advancing with composites 2000, Ed. I. Crivelli-Visconti, Milan,Italy,
May 9-11, 2000,pp.4ll-4l8.
Tomazevic, M., Lutman, M. and Petkovic, L. (1996). "Seismic behavior of masonry walls:
Experimental: simulation." Journal of Structural Engineering, 1040-1047.
Tomazevic, M. and Weiss, P. (1994). "Seismic behavior of plane-and reinforced-masonry

buildings." Journal of Structural Engineering, 120 (2). 323-338.
Triantafihlou, T. C. (1998). "Strengthening of masonry structures using epoxy-bonded FRP

laminates." Journal of Composites for Construction, 96-105.
R-14
Reference
Tseng, T. and Djordjevic, W. (2001). "Use of performance-based design guidelines to

deduce risks of Taiwan earthquake damage." SEMI, 1-7
Tumialan, G., Tinazzi, D., Myers, J. J. and Nanni, A. "Field evaluation of unreinforced
masonry walls strengthened with FRP composites subjected to out-of-plane loading."
Tumialan G, Micelli F and Nanni A. Strengthening of masonry structures with FRP

composites. Structures 2001, Washington D.C., May 2001: 21-23.
Tumialan, G., Huang, P. C., Silva, P., and Nanni A.(2001) "Strengthening of masonry walls
by FRP structural repointing." Non-Metallic Reinforcement for Concrete Structures-
FRPRCS-5, Cambridge
Tumialan, G., Micelli, F., and Nanni A.(2001) "Strengthening of masonry structures with
FRP composites ." Structures 2001, Washington D.C., May 21-23,2001.
TOWN PARK CONSTRUCTION. "Foundation"

(http://www.toenparkconstruction.com/foundations.htm)
Velazquez-Dimas, J. I., Saadatmanesh, H., and Ehsani, M. R. (1999). "Behavior of

retrofitted URM walls under simulated earthquake loading." J. Comp. for Constr., ASCE,
3(3), 134-142.
Velazquez-Dimas, J. I., and Ehsani, M. R. (2000). "Modeling out-of-plane behavior of

URM walls retrofitted with fiber composites." J. Comp. for Constr., ASCE, 4(4), 172-181.
Walker, G. R. (1990). Interim report on Newcastle earthquake. Division of Building,

Construction and Engineering, CSIRO.
Weaver, Jr., William and Johnston, Paul R. (1984). Finite elements for structural analysis,
Prentice-Hall, mc, the United States of America.
R-15
Reference
Wyllie L.A. (1983). "Seismic strengthening procedures for existing structures."

Strengthening of building structures — mt. Assoc. of bridge and structural Eng. Sym., 363-
370.
Xue, Q. (2000). "Need of performance-based earthquake engineering in Taiwan: a lesson

from the Chichi earthquake." Earthquake Engrg. Struct. Dyn., 29, 1609-1627.
Zhuge, Y., Thabbiratnam, D. P., and Corderoy, J. (1998). "Nonlinear dynamic analysis of
unreinforced masonry." J. Struct. Engrg., ASCE, 124(3), 270-277.
Zhuge, Y. (1996). "Nonlinear Dynamic Response of Unreinforced Masonry under In-Plane

Lateral Loads". PhD Thesis, Queensland University of Technology.
Zienkiewicz, 0. C., and Taylor, R. L. (2000). The finite element method Volume 1 The
basis, Bitterworth-Heinemann 2000, England.
R-16
Appendix A
APPENDIX A - MATERIAL PROPERTIES
CONCRETE - COMPRESSIVE STRENGTH TEST RESULT
Table Al
Diameter Average Height Area Load
Diameter
Specimens mm mm mm mm2 N MPa
1 100.8 100.4 202 8011.85 241000 31
100.0
2 100.5 100.25 202 7972.23 245000 32

100.0
MORTAR MIX - COMPRESIVE STRENGTH TEST RESULT
Table A2
Diameter Average Height Area Load
Diameter
Specimens mm mm mm mm2 N MPa
1 101.2 101.0 204 8011.85 61000 7.60.
100.8
2 100.9 101.05 206 8019.78 61000 7.61
101.2
BRICK - COMPRESSIVE STRENGTH TEST RESULT
Table A3
Length Width Height Area Core Load Load •
. Diameter Bearing.
Area
Specimens mm mm mm mm2 mm mm2 N MPa
1 230 110 600 25300 40 21530.09 484000 22.480
2 230 110 600 25300 40 21530.09 484000 19.972
3 230 110 600 25300 40 21530.09 484000 18.579
A-i
Appendix A
MATERIAL PROPERTIES USED IN THIS THESIS
Table A4
Elastic Poisson's Yield Compressive Tensile Breaking
Modulus ratio stress strength strength load
Material MPa MPa MPa MPa KN
Masonry 5,940* 0.09 7.92 0.2
Cable** 107,500 0.3 . 71
Steel*** .
Plate 200,000 0.3 410

62,000 0.22 2800
Note: * Calculated based on compressive strength of masonry (AS3700 (2001)).

** Adopted from Rostan booklet.
*** Adopted from BHP booklet.
**** Adopted from Sika Booklet.
A-2
________
Appendix B
APPENDIX B - FINAL FAILURE PATTERNS
WaIl 1 (Unretrofitted URM wall, FRONT SIDE)

Final Load = 22.49KN
I II
I Ii II II Ii I
I II II II I
I II II II II I
I II II II I
I. F I I I II I
I II II II
II II II
H' III II
•II
111111!!
I! II
ii II II
B-i
Appendix B
WaIl 1 (Unretrofitted URM wall, BACK SIDE)

I II II Ii I
I ii II I
I I I
I I
I I
I I I I I II II I
I II II I
I II
I I
B-2
__H_______
Appendix B
Wall 2 (Retrofitted URM wall with P cable, FRONT SIDE)

I ii II I
B-3
Appendix B
WaIl 2 (Retrofitted URM wall with P cable, BACK SIDE)

I II I
I II I
II I
I II I
I II
I II
II
II II 1.
Appendix B
WaIl 3 (Retrofitted URM wall with cable, FRONT SIDE)

I II I
I I I
I I
I I
I I
I II
I I
I II II It
B-5
Appendix B
WaIl 3 (Retrofitted URM wall with cable, BACK SIDE)

I II I II I
I II II II II J
I II II
I I
I I I
I I
I ii I
I II II
I I
I ill • it it
B-6
Appendix B
Wall 4 (Retrofitted URM wall with cable, FRONT SIDE)

I II II I
I I
I I I
I II I
I II
II
I [I IL__
B-7
Appendix B
Wall 4 (Retrofitted URM wall with cable, BACK SIDE)

Final Load 47.72KN
I I
II
I ii II I
I I I
II
I I
IL
B-8
Appendix B
WaIl 5 (Retrofitted URM wall with FRP, FRONT SIDE)

I II II II 1
I Ii II II
I I
II
ii
I
B-9
_______
Appendix B
WaIl 5 (Retrofitted URM wall with FRP, BACK SIDE)

I hi ii Ii I
I II II II
II I
I I
ii
I hi II I
t. ii flu
H' lit II
H
I
F I I I I
B-1O
Appendix B
Wall 6 (Retrofitted URM wall with FRP, FRONT SIDE)

I II II Ii I
ii II II I
• I I
I I
II
II
II II
II
ii II
B-li
Appendix B

I ill II 1:1
lii I
iii
Lit 2' HI
I I L ii Ii 1
•—1t I
1:1
I I III
B-12
____________________ __________
Appendix B
Wall 7 (Retrofitted URM wall with FRP, FRONT SIDE)

Final Load = 63.51 KN
I II II II I
I II II II II I
I II
I 11 I I I
I II I
I II
H I'
B-13
Appendix B

I ii I
I II I
I
I
II
II I
I I
I II
B-14
Appendix C
APPENDIX C - PERFORMANCE-BASED
DESIGN GUIDELINES
C. 1 Review Initial Considerations
The design professional shall review initial considerations with the building owner or
code official to determine any restrictions that exist on the design of rehabilitation
measures. Initial considerations shall include structural characteristics of the building,
seismic hazards including geologic site hazards known to be present at the site, results of
prior seismic evaluation, building use and occupancy requirements, historic status,
economic considerations, societal issues, and local jurisdictional requirements.
C.2 Select Rehabilitation Objective

A Rehabilitation Objective shall be selected as the basis for design. Rehabilitation
Objectives are statements of the desired building performance when the building is
subjected to earthquake demands of specified severity. Figure C. I summarizes the
various Rehabilitation Objectives available in the NEHRP Guidelines.
Alternative Rehabilitation Objectives

I I
Limited Basic Safety Enhanced
Partial Life Safety for BSE-1 Life Safety br BSE-1 Life Safety br 855-1
Partial Collapse Prevention Cotlapse Prevention for BSE-2 (Collapse Prevention for BSE-2
for BSE-2
and
Partial tite Safety < 500 yearS Immediate occupancy at any return
Collapse Prevention 2.500 years period, or
limrted Safety at any return panod Damage Control at any return period, or
tile Safety al'. 500 years
General goal: To make the General goal: To provide General goal: To provide a low
building better than it was a tow risk of endangerment nsk of endangerment of life
before rehabilitation of life safety for any event safety for any event likely to
likely to affect Ihe building affect the building, and to further
protect building features and/or
contents against damage
Figure C-i Summary of the various Rehabilitation Objectives (FEMA-274, 1997)

Appendix C
Building performance can be described qualitatively in terms of the safety afforded

building occupants, during and after the event; the cost and feasibility of restoring the
building to pre-earthquake condition; the length of time the building is removed from
service to effect repairs; and economic, architectural, or historic impacts on the larger
community. These performance characteristics are directly related to the extent of
damage sustained by the building.
The extent of damage to a building is categorized as a Building Performance Level. A

broad range of Building Performance Levels may be selected when determining
Rehabilitation Objectives. Each Building Performance Level consists of a Structural
Performance Level, which defines the permissible damage to structural systems, and a
Nonstructural Performance Level, which defines the permissible damage to nonstructural
buildings components and contents.
Earthquake demands are a function of the location of the building with respect to
causative faults, the regional and site-specific geologic characteristics, and the ground
motion hazard level(s) selected in the Rehabilitation Objective.
In the NEHRP Guidelines, hazard levels may be defined on either a probabilistic or

deterministic basis. Deterministic demands are defined within a level of confidence in
terms of a specific magnitude event on a particular fault, which is most appropriate for
buildings located within a few miles of a major active fault. Probabilistic hazard levels
frequently used in this NEHRP Guidelines and their corresponding mean return periods
(the average number of years between events of similar severity) are as follows:
Earthquake Having Mean Return Period
Probability of Exceedance (years)
50%/50 year 72
20%150 year 225
lO%50 year 474
2%/50 year 2,475
C-2
Appendix C
They are termed a Basic Safety Earthquake 1 (BSE-1) and Basic Safety Earthquake 2
(BSE-2). The BSE-1 and BSE-2 earthquakes are typically taken as 1O%/50 and 2%/50
year events, respectively, except in regions near major active fault.
The Rehabilitation Objective selected as a basis for design will determines, to a great
extent, the cost and feasibility of any rehabilitation project, as well as the benefit to be
obtained in terms of improved safety, reduction in property damage, and interruption of
use in the event of future earthquakes. Table C. 1 presents a matrix indicating the broad
range of Rehabilitation Objectives that may be used in the NHERP Guideline. Each cell
in this matrix represents a single Rehabilitation Objective. The goal of a rehabilitation
project may be to satisfy a single Rehabilitation Objective—for example, Life Safety for
the BSE-1 earthquake—or multiple Rehabilitation Objectives—for example, Life Safety
for the BSE-1 earthquake, Collapse Prevention for the BSE-2 earthquake, and Immediate
Occupancy for an earthquake with a 50% probability of exceedance in 50 years. A
specific analytical evaluation should be performed to confirm that a rehabilitation design

is capable of meeting each desired Rehabilitation Objective selected as a goal for the
project.
C-3
Appendix C
Target Building Performance

Levels
o
0)
In
C 0 C),,
C
Cu
E
co
—
C
C)
C)
OS-s It 0-a
50%/5O year a b C d
I
,.
20%/SO year e 1 g h
-o
a-
BSE—1 J k
'Cu
(—iQ%/50 year)
a BSE-2
(—2%/50 year)
m n 0
Notes:
1. Each cell in the above matrix represents a discrete Rehabilitation
Objective.
2. The Rehabilitation Objectives in the matrix above may be used to
represent the three specific Rehabilitation Objectives defined in
St tions 1.4.1, 1.4.2, and 1.4.3, as follows:
= Basic Safety Objective (BSO)
Jc+ p i-any of a, e, i, I>, f, j, or n = Enhanced Objeciives
o alone or n alone or m alone = Enhanced Objective
k alone or p alone Limited Objectives -
c, g, d, li, 1= Limited Objectives
Table C-i Rehabilitation Objectives (FEMA-274, 1997)
In general, Rehabilitation Objectives that expect relatively low levels of damage for
relatively infrequent earthquake events will result in more extensive rehabilitation work
C-4
Appendix C
and greater expense than objectives with more modest goals of controlling damages.
Figure C.2 schematically presents the relationship between different Rehabilitation
Objectives and probable program cost.
Basic Safety Objective
cost
2% (BSE-2)
Increasing
performance
Figure C-2 Surface showing relative costs of various Rehabilitation Objectives (FEMA-
274, 1997)
• Basic Safety Objective
A desirable goal for rehabilitation is to achieve the Basic Safety Objective (BSO). In
order to achieve this objective, building rehabilitation must be designed to achieve both
the Life Safety Performance Level (3-C) for BSE-1 earthquake demands and Collapse
Prevention Level (5-E) for BSE-2 earthquake demands. Once a desired Building
Performance Level for a particular ground shaking severity (seismic demand) is selected,
the result is a Rehabilitation Objective. With the exception of the Basic Safety Objective
(BSO), there are no preset combinations of performance and ground shaking hazard. The
Basic Safety Objective is met when a building can satisfy two criteria: (1) the Life Safety
Building Performance Level, which is the combination of the Structural and
Nonstructural Life Safety Performance Levels, for the Basic Safety Earthquake 1 (BSE-
1), and (2) the Collapse Prevention Performance Level, which only pertains to structural
C-5
Appendix C
performance, for the stronger shaking that occurs less frequently as defined in the Basic
Safety Earthquake 2 (BSE-2).
• Enhanced Rehabilitation Objectives
Any Rehabilitation Objective intended to provide performance superior to that the BSO is
termed an Enhanced Objective. An Enhanced Objective must provide better than BSO-
designated performance at either the BSE-l or BSE-2, or both. Enhanced performance
can be obtained in two ways:
• Directly, by design for the BSE-1 or BSE-2 earthquakes. Examples include

designing for a higher Performance Level than Life Safety for the BSE-l or a
higher Performance Level than Collapse Prevention for the BSE-2.
• Indirectly, by having the design controlled by some other selected Performance
Level and hazard that will provide better than BSO performance at the BSE-l or
BSE-2. For example, if providing Immediate Occupancy for a 50% - 50 year
event controlled the rehabilitation acceptability criteria in such a way that

deformation demand was less than that allowed by the BSO, the design would be
considered to have an Earthquake Objective.
• Limited Rehabilitation Objectives
Any Rehabilitation Objective intended to provide performance inferior to that of the BSO
is termed a Limited Objective. A Limited Objective may consist of either Partial
Rehabilitation or Reduced Rehabilitation. Limited Rehabilitation Objectives should be
permissible if the following conditions are met:
• The rehabilitation measures do not create a structural irregularity or make an
existing structural irregularity more severe;
• The rehabilitation measures do not result in a reduction in the capability of the
structure to resist lateral forces or deformations;
• The rehabilitation measures do not result in an increase in the seismic forces,
unless this component's behavior is still acceptable considering overall structural
performance;
• All new or rehabilitated structural elements are detailed and connected to the
existing structure, as required by the NEHIRP Guidelines;
C-6
Appendix C
• An unsafe condition is not created or made more severe by the rehabilitation

measures; and
• Locally adopted and enforced building regulations do not preclude such
rehabilitation.
C.2.1 Target Building Performance level
Building performance is a combination of the performance of both structural and

nonstructural components. Table C.2 describes the overall levels of structural and
nonstructural damage that may be expected of buildings rehabilitated to the levels defined
in the NEHRP Guidelines for the seismic rehabilitation of buildings (2000). For
comparative purposes, the estimated performance of a new building subjected to the
BSE-1 level of shaking is indicated. These performance descriptions are estimates rather
than precise predictions, and variation among buildings of the same Performance Level
must be expected.
Building Performance Levels
Immediate
Collapse Prevention Life Safety Occupancy Operational
Level Level Level Level
Overall Damage Severe Moderate Light Very Light
General Little residual stiffness Some residual No permanent drift. No permanent drift;
and strength, but load- strength and stiffness Structure structure substantially
bearing columns and left in all stories. substantially retains retains original
walls function. Large Gravity-load-bearing original strength and strength and stiffness.
permanent drifts, elements function. No stiffness. Minor Minor cracking of
Some exits blocked, out-of-plane failure of cracking of facades, facades, partitions,
Infills and unbraced walls or tipping of partitions, and and ceilings as well
parapets failed or at parapets. Some ceilings as well as as structural elements.
incipient failure, permanent drift, structural elements. All systems important
Building is near Damage to partitions. Elevators can be to normal operation
collapse. Building may be restarted. Fire are functional.
beyond economical protection operable.
repair.
Nonstructural Extensive damage. Falling hazards Equipment and Negligible damage
components mitigated but many contents are generally occurs. Power and
architectural, secure, but may not other utilities are
mechanical, and operate due to available, possibly
electrical systems are mechanical failure or from standby sources.
damaged, lack of utilities.
Comparison with Significantly more Somewhat more Much less damage Much less damage
performance intended damage and greater damage and slightly and lower risk, and lower risk.
for buildings risk, higher risk
designed, under the
NEHPJ' Provisions, for
the Design
Earthquake
Table C-2 Building Performance level (FEMA-274, 1997)
C-7
Appendix C
• The Structural Performance Levels
The Structural Performance Levels are the Immediate Occupancy Level (S-i), the Life
Safety Level (S-3), and the Collapse Prevention Level (S-5). Table C.3 relates these
Structural Performance Levels to the limiting damage states for common vertical
elements of lateral-force-resisting systems.
Elements Type Collapse Life Safety S-3 Immediate

Prevention S-5 Occupancy S-i
H
Unreinforced Primary Extensive Extensive cracking Minor (<1/9
Masonry Infill cracking and some crushing width) cracking of
and
Walls crushing; but wall remain in masonry infills and
portions of face place. No falling veneers at a few
course shed. units. Extensive corner openings.
crushing and spalling
of veneers at corners
of openings.
Secondary Extensive Same as primary. Same as primary.
crushing and
shattering; some
walls dislodge.
Drift 0.6% transit or 0.5% transit; 0.3% 0.1% transit;
permanent. permanent. negligible
permanent.
Table C.3 Structural Performance level for URM walls (FEMA-274, 1997)
Inherent in the concept of Performance Levels and Ranges is the assumption that
performance can be measured using analytical results such as story drift rations or
strength and ductility demands on individual components or elements.
The structural Performance Ranges are the Damage Control Range (S-2) and the Limited
Safety Range (S-4). Specific acceptance criteria are not provided for design to these
intermediate performance ranges. The engineer wishing to design for such performance
needs to determine appropriate acceptance within the Damage Control Range may be
obtained by interpolating the acceptance criteria for performance within the Limited
Acceptance criteria for performance within the Life Safety and Collapse Prevention
Performance Levels.
C-8
Appendix C
Inunediate Occupancy Performance Level (S-i)
Structural Performance Level S- 1, Immediate Occupancy, means the post-earthquake

damage state in which only very limited structural damage has occurred. The basic
vertical-force and lateral-force resisting systems of the building retain nearly all of their
pre-earthquake strength and stiffness. The risk of life-threatening injury as a result of
structural damage is very low, and although some minor structural repairs may be
appropriate, these would generally not be required prior to reoccupancy.
Life Safety Performance Level (S-3)
Structural Performance Level S-3, Life Safety, means the post-earthquake damage state
in which significant damage to the structure has occurred, but some margin against either
partial or total structural collapse remains. Some structural elements and components are
severely damaged, but this has not resulted in large falling debris hazards, either within or
outside the building. Injuries may occur during the earthquake; however, it is expected
that the overall risk of life-threatening injury as a result of structural damage is low. It
should be possible to repair the structure; however, for economic reasons this may not be
practical. While the damaged structure is not an imminent collapse risk, it would be
prudent to implement structural repairs or install temporary bracing prior to reoccupancy.
Collapse Prevention Performance Level (S-5)
Structural Performance Level S-5, Collapse Prevention, means the building is on the
verge of experiencing partial or total collapse. Substantial damage to the structure has
occurred, potentially including significant degradation in the stiffness and strength of the
lateral-force-resisting system, large permanent deformation of the structure, and—to a
more limited extent—degradation in vertical —load-carrying capacity. However, all
significant components of the gravity-load-resisting system must continue to carry their
gravity load demands. Significant risk of injury due to falling hazards from structural
debris may exist. The structure may not be technically practical to repair and is not safe
for reoccupancy, as aftershock activity could induce collapse.
C-9
Appendix C
> Damage Control Performance Range (S-2)
Structural Performance Range S-2, Damage Control, means the continuous range of
damage states that entail less damage than that defined for the Life Safety level, but more
than that defined for the Immediate Occupancy level. Design for Damage Control
performance may be desirable to minimize repair time and operation interruption; as a
partial means of protecting valuable equipment and contents; or to preserve important
historic features when the cost of design for Immediate Occupancy is excessive.
Acceptance criteria for this range may be obtained by interpolating between the values
provided for the Immediate Occupancy (S-i) and Life Safety (S-3) levels.
> Limited Safety Performance Range (S-4)
Structural Performance Range S-4, Limited Safety, means the continuous range of
damage states between the Life Safety and Collapse Prevention levels. Design parameters
for this range may be obtained by interpolating between the values provided for the Life
Safety (S-3) and Collapse Prevention (S-5) levels.
> Structural Performance Not Considered (S-6)
Some owners may desire to address certain nonstructural vulnerabilities in a

rehabilitation program—For example, bracing parapets, or anchoring hazardous materials
storage containers—without addressing the performance of the structure itself. Such
rehabilitation programs are sometimes attractive because they can permit a significant
reduction in seismic risk at relatively low cost. The actual performance of the structure
with regard to NEHRP Guidelines requirements is not known and could range from a
potential collapse hazard to a structure capable of meeting the Immediate Occupancy
Performance Level:
• Nonstructural Performance Levels
Four Nonstructural Performance Levels are defined in these NEHRP Guidelines.

Nonstructural components addressed in these performance levels include architectural
components, such as partitions, exterior cladding, and ceilings; and mechanical and
electrical components, including HVAC systems, plumbing, fire suppression systems,
C-1O
Appendix C
and lighting. Occupant contents and furnishings (such as inventory and computers) are
also included.
> Operational Performance Level (N-A)
Nonstructural Performance Level A, Operational, means the post-earthquake damage

state of the building in which the nonstructural components are able to support the
building's intended function. At this level, most nonstructural systems required for
normal use of the building—including lighting, plumbing, HVAC, and computer
systems—are functional, although minor cleanup and repair of some items may be
required. This performance level requires considerations beyond those that are normally
within the sole province of the structural engineer. In addition to assuring that
nonstructural components are properly mounted and braced within the structure, in order
to achieve this performance it is often necessary to provide emergency standby utilities.
In addition, it may be necessary to perform rigorous qualification testing of the ability of
key electrical and mechanical equipment items to function during or after strong shaking.
> Immediate Occupancy Level (N-B)
Nonstructural Performance Level B, Immediate Occupancy, means the post-earthquake

damage state in which only limited nonstructural damage has occurred. Basic access and
life safety systems, including doors, stairways, elevators, emergency lighting, fire alarms,
and suppression systems, remain operable, provided that power is available. There could
be minor window breakage and slight damage to some components. Presuming that the
building is structurally safe, it is expected that occupants could safely remain in the
building, although normal use may be impaired and some cleanup and inspection may be
required. In general, components of mechanical and electrical systems in the building are
structurally secured and should be able to function if necessary utility service is available.
However, some components may experience misalignments or internal damage and be
nonoperable. Power, water, natural gas, communications lines, and other utilities required
for normal building use may not be available. The risk of life-threatening injury due to
nonstructural damage is very low.
C-li
Appendix C
> Life Safety Level (N-C).
Nonstructural Performance Level C, Life Safety, is the post-earthquake damage state in

which potentially significant and costly damage has occurred to nonstructural
components by they have not become dislodged and fallen, threatening life safety either
within or outside the building. Egress routes within the building are extensively blocked,
but may be impaired by lightweight debris. HVCA, plumbing, and fire suppression
systems may have been damaged, resulting in local flooding as well as loss of function.
While injuries may occur during the earthquake from the failure of nonstructural
components, it is expected that, overall, the risk of life-threatening injury is very low.
Restoration of the nonstructural components may take extensive effort.
> Hazards Reduced Level (N-D)
Nonstructural Performance Level D, Hazards Reduced, represents a post-earthquake

damage state level in which extensive damage has occurred to nonstructural components,
but large or heavy items that pose a falling hazard to n number of people—such as
parapets, cladding panels, heavy plaster ceilings, or storage racks—are prevented from
falling. While isolated serious injury could occur from falling debris, failures that could
injure large numbers of persons—either inside or outside the structure—should be
avoided. Exits, fire suppression systems, and similar life-safety issues are not addressed
in this performance level.
> Nonstructural Performance Not Considered (N-E)
In some cases, the decision may be made to rehabilitate the structure without addressing
the vulnerabilities of nonstructural components. It may be performed without interruption
of building operation. In some cases, it is possible to perform all or most of the structural
rehabilitation from outside occupied building areas, while extensive disruption of normal
operation may be required to perform nonstructural rehabilitation. Also, since many of
the most severe hazards to life safety occur as a result of structural vulnerabilities, some
municipalities may wish to adopt rehabilitation ordinances that require structural
rehabilitation only.
C-12
Appendix C
• Building Performance Levels
Building Performance Levels are obtained by combining Structural and Nonstructural

Performance Levels. A large number of combinations is possible. Each Building
Performance Level is designated alpha-numerically with a numeral representing the
Structural Performance Level and a letter representing the Nonstructural Performance
Level (e.g. 1-B, 3-C). Table C.4 indicates the possible combinations and provides names
for those that are most likely to be selected as a basis for design. Several of the more
common Building Performance Levels are described below.
Table 2-9 Building Performance Levels/Ranges

Structural Performance Level
Nonstructural
Performance S-5 Coflapse S-S Not
Levels Range Range prevention Considered
N-A Operational 2-A Not Not Not Not

Operational 1-A recommended iecommended recommended recommended
N-B Immediate 2-B 3-8 Not Not Not
$rnmediate Occupancy 1.8 Tecommended recommended recommended
N-C Life Safety 1-C 2-C Life Safety 3-C 4-C 5-C 6-C
N-OHazards Not 2-D 3-0 4-0 s-o 1 6-0

Reduced recommended
N-E Not Not Not Not 4-E 5-E Collapse No
Considered recommended recommended rerommended Prevention rehabilitation
Table C-4 Building Performance Levels / Ranges (FEMA-273, 1997)
Operational Level (1-A)
This building Performance Level is a combination of the Structural Immediate occupancy

Level and the Nonstructural Operational Level. Buildings meeting this performance level
are expected to sustain minimal or no damage to their structural and nonstructural
components. The building is suitable for its normal occupancy and use, although possibly
in a slightly impaired mode, with power, water, and other required utilities provided from
emergency sources, and possibly with some nonessential systems not functioning,
Buildings meeting this performance level pose an extremely low risk to life safety. Under
very low levels of earthquake ground motion, most buildings should be able to meet or
exceed this performance level. Typically, however, it will not be economically practical
to design for this performance under severe levels of ground shaking, except for buildings
that house essential services.
C-l3
Appendix C
> Immediate Occupancy Level (i-B)
This Building Performance Level is a combination of the Structural and Nonstructural

Immediate Occupancy levels. Buildings meeting this performance level are expected to
sustain minimal or no damage to their structural elements and only minor damage to their
nonstructural components. While it would be safe to reoccupy a building meeting this
performance level immediately following a major earthquake, nonstructural systems may
not function due to either a lack of electrical power or internal damage to equipment.
Therefore, although immediate reoccupancy of the building is possible, it may be
necessary to perform some cleanup and repair, and await the restoration of utility service,
before the building could function in a normal mode. The risk to life safety at this
performance level is very low.
Many building owners may wish to achieve this level of performance when the building
is subjected to moderate levels of earthquake ground motion. In addition, some owners
may desire such performance for very important buildings; under sever levels of
earthquake ground shaking. This level provides most of the protection obtained under the
Operational Level, without the cost of providing standby utilities and performing rigorous
seismic qualification of equipment performance.
> Life Safety Level (3-C)
This Building Performance Level is a combination of Structural and Nonstructural Life

Safety levels. Buildings meeting this level may experience extensive damage to structural
and nonstructural components. Repairs may be required before reoccupancy of the
building occurs, and repair may be deemed economically impractical. The risk to life in
buildings meeting this performance level is low. This performance level entails somewhat
more damage than anticipated for new buildings that have been properly designed and
constructed for seismic resistance when subjected to their design earthquakes. Many
building owners will desire to meet this performance level for a severe level of ground
shaking.
C-14
Appendix C
> Collapse Prevention Level (5-E)
This Building Performance Level consists of the Structural Collapse Prevention Levels
with no consideration of nonstructural vulnerabilities, except that parapets and heavy
appendages are rehabilitated. Buildings meeting this performance level may pose a
significant hazard to life safety resulting from failure of nonstructural components.
However, because the building itself does not collapse, gross loss of life should be
avoided. Many buildings meeting this level will be complete economic losses.
This level has sometimes been selected as the basis for mandatory seismic rehabilitation
ordinances enacted by municipalities, as it results in mitigation of the most severe life-
safety hazards at relatively low cost.
C.2.2 Seismic Hazard Level
Earthquake hazard is a representation of the demand that will be placed upon a structure
at a given site for an intensity of ground shaking. The selection of an appropriate intensity
of ground shaking is a critical step in the Performance Based Seismic Design process.
Earthquake hazard can be defined using probabilistic or deterministic approaches. For
example, if a known fault source exists proximate to the site that has historically
produced earthquakes of a characteristic magnitude; a scenario event can be defined as
the earthquake hazard for consideration. For this scenario, a faulting mechanism,
hypocenter, length of rupture, magnitude and duration can be selected and deterministic
methods used to estimate the intensity of ground shaking at a given site. Alternatively, a
ground shaking intensity can be probabilistically selected as the shaking associated with
an event that has a prescribed probability of exceedence in a prescribed period of time or
an average return period. For example, building codes have historically been based on
ground shaking intensities corresponding to a 10% probability of exceedence in a 50-year
period of time. The selection of an appropriate earthquake hazard permits an owner,
architect or other design professional to implement a rational strategy for earthquake
design that reflects the unique needs of an owner and the seismic climate of the site.
C-15
Appendix C
C.3 Obtain As-Built Information
Existing building characteristics pertinent to its seismic performance—including its

configuration, and the type, detailing, material strengths, and condition of the various
structural and nonstructural elements, including foundations and their interconnections.
The project calculations should included documentation of these characteristics in
drawings or photographs, supplemented by appropriate descriptive text. Existing
characteristics of the building and site should be obtained from the following sources, as
appropriate:
• Field observation of exposed conditions and configuration

• Available construction documents, engineering analyses, reports, soil borings and
test logs, maintenance histories, and manufactures' literature and test data
• Reference standards and codes from the period of construction
• Destructive and nondestructive examination and testing of selected building
components
• Interviews with building owners, tenants, managers, the original architect and
engineer, contractor(s), and the local building official
As a minimum, at least one site visit should be performed to obtain detailed information
regarding building configuration and condition, site and geotechnical conditions, and any
issues related to adjacent structures, and to confirm that the available construction
documents are generally representative of existing conditions. If the building is a historic
structure, it is also important to identify the locations of historically significant features
and fabric. Care should be taken in the design and investigation process to minimize the
impact of work on these features. The following sections are focused to discuss about
obtaining as-built information for masonry wall.
C.3.1 Material Properties for Masonry walls

The term "masonry" is used to define the composite of units, mortar, and possibly grout
and/or reinforcement. Whereas there are specifications to control the manufacture of each
of the constituent materials, the most basic engineering properties to consider for analysis
C-16
Appendix C
of a building system are those representing the composite. Thus, permissible values are
given in this section for compressive strength and elastic modulus of the masonry
assemblage, flexural tensile strength at the unit-mortar interface, and shear strength and
shear modulus of vertical components such as piers, panels, for estimating stiffness and
strength of masonry wall.
• Masonry compressive strength
Expected masonry compressive strength, fme' shall be measured using one of following
three methods. (FEMA 273, 1997)
1. Test prisms shall be extracted from an existing wall and tested per Section 1.4.B.3
of the Masonry Standards Joint Committee's Building Code Requirements for
Masonry Structures (MSJC, 1995a).
2. Prisms shall be fabricated from actual extracted masonry units, and a surrogate
mortar designed on the basis of a chemical analysis of actual mortar samples. The
test prisms shall be tested per Section 1 .4.B.3 of the Specification for Masonry
Structures (MSJC, 1995b).
3. Two flat jacks shall be inserted into slots cut into mortar bed joints and
pressurized until peak stress is reached.
For each of three methods, the expected compressive strength shall be based on the
net mortared area.
If the masonry unit strength and the mortar type are known, values may be taken
from Tables C.5 and C.6 of MSJC (1995a) for clay or concrete masonry constructed
after 1960. The fme value shall be obtained by multiplying the table values by a factor
that represents both the ratio of expected to lower bound strength and the height-to-
thickness ratio of the prism.
In lieu of material tests, default values for masonry prism compressive strength shall
be taken to not exceed 900 psi for masonry in good condition, 600 psi for masonry in
fair condition, and 300 psi for masonry in poor condition.
C-17
-
Appendix C
Masonry Condition1
Property Good Fair Poor

Compressive Strength (t'm) 900 psi 600 psi 300 psi
Elastic Modulus in Compression 5501'ni 5501'rn
Flexural Tensile Strength2 20 pSI 10 0
Shear Strength3 - .
Masonry with a running bond lay-up 27 psi 20 psi 13 psi
Fully grouted masonry with a lay-up other Ihan running bond 27 psi 20 psi 13 psi
Partially grouted or ungrouted masonry with a lay-up other 11 psi 8 psi 5 psi
than running bond
L Masonry condition shall be classified as good. fair, or poor as defined in this standard.
Table C-5 Default values for masonryprism compressive strength (MSJC, 1995a).
Property Factor
Compressive Strength (frne) 1
Elastic Modu'us in
Fiexural Tensile Strength 13
Shear Strength 1
Table C-6 The factor that represents both the ratio of expected to lower bound strength
and the height-to-thickness ratio of the prism. (MSJC, 1995a)
• Masonry elastic modulus in compression
Expected values of elastic modulus for masonry in compression, E me' shall be

measured using one of the following two methods:
1. Test prisms shall be extracted from an existing wall, transported to a laboratory,

and tested in compression. Stresses and deformations shall be measured to infer
modulus values.
2. Two flat jacks shall be inserted into slots cut into mortar bed joints, and
pressurized up to nominally one half of the expected masonry compressive
C-18
Appendix C
strength. Deformations between the two flat jacks shall be measured to infer
compressive strain, and thus elastic modulus.
In lieu of prism test, values for the modulus of elasticity of masonry in compression shall
be taken as 550 times the expected masonry compressive strength, fme.
• Masonry flexural tensile strength
Expected flexural tensile strength, fie' for out-of-plane bending shall be measured using
one of the following three methods:
1. Test samples shall be extracted from existing wall, and subjected to minor-axis
bending using the bond-wrench method.
2. Test samples shall be tested in situ using the bond-wrench method.
3. Sample wall panels shall be extracted and subjected to minor-axis bending in
accordance with ASTME 519.
In lieu of material test, default values of masonry flexural tensile strength for walls or
infill panels loaded normal to their plane shall be taken to not exceed 20 psi for masonry
in good condition, 10 psi for masonry in fair condition, and zero psi for masonry in poor
condition. For masonry constructed after 1960 with cement-based mortars, default values
of flexural tensile strength can be based on values from Table 8, 3.10.5.1 of BSSC
(1995).
Flexural tensile strength for unreinforced masonry (URM) walls subjected to in-plane
lateral forces shall be assumed to be equal to that for out-of-plane bending, unless testing
is done to define expected tensile strength.
• Masonry shear strength
For URM components, expected masonry shear strength, v me' shall be measured using
the in-place shear test. Expected shear strength shall be determined in accordance with
Equation C-i.
C-19
Appendix C
]
Vme = 1.5 (C-i)
where
CE = Expected gravity compressive force applied to a wall or pier component stress
considering load combinations given in Equations c-i 9 and C- 19
A,1 = Area of net mortaredlgrouted section, in.2
v te = Average bed-joint shear strength, psi
The 0.75 factor on the v te term may be waived for single wythe masonry, or if the collar
joint is known to be absent or in very poor condition.
Values for the mortar shear strength, v te' shall not exceed 100 psi for the determination
Of Vte in Equation C-i.
Average bed-joint shear strength, v fe' shall be determined from individual shear strength
test values, in accordance with Equation C-2.
Vtest
— PD+L (C-2)
= Ab
where V test is the load at first movement of a masonry unit, A b is the net mortared area
of the bed joints above and below the test brick, and PD+L is the estimated gravity stress
at the test location.
In lieu of material tests, default values of shear strength of URM components shall be
taken to not exceed 27 psi for running bond masonry in food condition, 20 psi for running
bond masonry in fair condition, and 13 psi for running bond masonry in poor condition.
These values shall also be used for masonry in other than running bond if fully grouted.
For masonry in other than running bond and partially grouted or ungrouted, shear
strength shall be reduced by 60% of these values. For masonry constructed after 1960
C-20
Appendix C
with cement-based mortars, default values of shear strength can be based on values in
BSSC (1995) for unreinforced masonry.
• Masonry shear modulus
The expected shear modulus of uncracked, unreinforced, or reinforced masonry,

shall be estimated as 0.4 times the elastic modulus in compression. After cracking, the
shear modulus shall be taken as a fraction of this value based on the amount of bed joint
sliding or the opening of diagonal tension cracks.
• Location and minimum number of test
The number and location of material tests shall be selected to provide sufficient
information to adequately define the existing condition of materials in the building. Test
locations shall be identified in those masonry components that are determined to be
critical to the primary path of lateral-force resistance.
A visual inspection of masonry condition shall be done in conjunction with any in situ
material tests to assess uniformity of construction quality. For masonry with consistent
quality, the minimum number of tests for each masonry type, and for each three floors of
construction or 3000 square feet of wall surface, shall be three, if original construction
records are available that specify material properties, or six, if original construction
records are not available. At least two tests should be done per wall, or line of wall
elements providing a common resistance to lateral forces. A minimum of eight tests
should be done per building.
Test should be taken at locations representative of the material conditions throughout the
entire building, taking into account variations in workmanship at different story levels,
variations in weathering of. the exterior surfaces, and variations in the condition of the
interior surfaces due to deterioration caused by leaks and condensation of water andlor
the deleterious effects of other substances contained within the building.
For masonry with perceived inconsistent quality, additional tests shall be done as needed
to estimate material strengths in regions where properties are suspected to differ.
Nondestructive condition assessment tests may be used to quantify variations in material
strengths.
C-21
Appendix C
An increased sample size may be adopted to improve the confidence level. The relation
between sample size and confidence shall be as defined in ASTM E22.
If the coefficient of variation in test measurements exceeds 25%, additional tests shall be
done. If the variation does not reduce below this limit, use of the test data shall be limited
to the Linear Static Procedures.
If mean values from in situ material tests are less than the default values, additional tests
shall be done. If the mean continues to be less than the default values, the measured
values shall be used, and shall be used only with the Linear Static Procedures.
C.3.2 Condition Assessment

• Visual examination
The size and location of all masonry shear and bearing walls shall be determined. The
orientation and placement of the walls shall be noted. Overall dimensions of masonry
components shall be measured, or determined from plans, including wall heights, lengths,
and thickness. Locations and sizes of window and door openings shall be measured, or
determined from plans. The distribution of gravity loads to bearing walls should be
estimated.
The wall type shall be identified as reinforced or unreinforced, composite or

noncomposite, and/or grouted, partially grouted, or ungrouted. For RM construction, the
size and spacing of horizontal and vertical reinforcement should be estimated. For multi
wythe construction, the number of wythes (the thickness of the collar joint or cavity), and
the placement of interwythes should be noted. For grouted construction, the quality of
grout placement should be assessed. For partially grouted walls, the locations of grout
placement should be identified.
The type and condition of the mortar and mortar joints shall be determined. Mortar shall
be examined for weathering, erosion, and hardness, and to identify the condition of any
repointing, including cracks, internal voids, weak components, and/or deteriorated or
eroded mortar. Horizontal cracks in bed joint. Vertical cracks in head joints and masonry
units, and diagonal cracks near openings shall be noted.
C-22
Appendix C
The examination shall identify vertical components that are not straight. Bulging or
undulations in walls shall be observed, as well as separation of exterior wythes, out-of-
plumb walls, and learning parapets or chimneys.
Connections between masonry walls, and between masonry walls and floors or roofs,
shall be examined to identify details and condition. If construction drawings are
available, a minimum of three connections shall be inspected for each general connection
type (e.g., floor-to-wall, wall-to-wall). If no deviations from the drawings are found, the
sample may be considered representative. If drawings are unavailable, or significant
deviations are noted between the drawings and constructed work, then a random sample
of connections shall be inspected until a representative pattern of connections can be
identified.
• Nondestructive tests
Nondestructive tests may be used to supplement the visual observations. One, or a

combination, of the following nondestructive tests, shall be done to meet the
requirements of a compressive evaluation:
• Ultrasonic pulse velocity

• Mechanical pulse velocity
• Impact echo
• Radiography
The location and number of nondestructive tests shall be in accordance with the
requirements.
• Supplemental tests
Ancillary tests are recommended, but not required, to enhance the level of confidence in
masonry material properties, or to assess condition. These are described in the
Commentary to this section.
• Knowledge (ic) factor
C-23
Appendix C
To account for uncertainly in the collection of as-built data, a knowledge factor, K, shall
be selected from Table C-7 considering the selected Rehabilitation Objective, analysis
procedure, and data collection process. Knowledge factors shall be applied on a
component basis as determined by the level of knowledge obtained for individual
components during data collection. A knowledge factor, K, equal to 0.75, representing a
minimum level of knowledge of the structural system, shall be used if a visual
examination of masonry structural components is. A knowledge factor, K, equal to 1.00,
shall be used only with a comprehensive level of knowledge of the structural system.
Data Collection Requirements

Level of Knowledge
Data Minimum Usual Comprehensive

RehabititatiO BSO or Lower BSO or Lower Enhanced Enhanced
ii Objective .
Analysis ISP. LDP All All All

Procedures
Testing No Tests Usual Testing Usual Testing Comprehensive Testing
Drawings Design Or Design Or Design Or Constrtwtion Or
Drawings Equivalent Drawings Equivatenl Drawings Equivalent Documents Equivalent
Condition Visual Compro- Visual Compre- Visual Conipre- Visual Compre-
Assessment hensive hensive hensive hensive
Material From From From From From From From From
Properties Drawings Default Drawings Usual Drawings Usual Documents Cornpre-
or Default Values and Tests Tests and Tests rests and .Tests herisive
-
- Values • Tests
Knowledge 0.75 0.75 1.00 1.00 [0.75 0.75 1.00 1.00
Factor I
Table C-7 The Knowledge Factor, K (FEMA-356, 2000)
• Engineering properties of masonry walls
This section provides basis engineering information for assessing attributes of structural
walls, and included stiffness assumptions, strength acceptance criteria, and deformation
acceptance criteria for Immediate Occupancy, Life Safety, and Collapse Prevention
Performance Levels.
Masonry walls shall be categorized as primary or secondary elements. Walls that are
considered to be part of the lateral-force system, and may or may not support gravity
loads, shall be primary elements. Walls that are not considered as part of the lateral-force-
C-24
Appendix C
resisting system, but must reniain stable while supporting gravity loads during seismic
excitation, shall be secondary elements.
• Stiffness
The lateral stiffness of masonry wall and pier components shall be determined based on
the minimum net sections of mortared and grouted masonry in accordance with the
guidelines of this subsection. The lateral stiffness of masonry walls subjected to lateral
in-plane forces shall be determined considering both flexural and shear deformations.
The masonry assemblage of units, mortar, and grout shall be considered to be a

homogeneous medium for stiffness computations with an expected elastic modulus in
compression, Eme•
For linear procedures, the stiffness of a URM wall or pier resisting lateral forces parallel
with its plane shall be considered to be liner and proportional with the geometrical
properties of the uncracked section.
For nonlinear procedures, the in-plane stiffness of URM walls or piers shall be based on
extent of cracking.
Story shears in perforated shear walls shall be distributed to piers in proportion to the
relative lateral uncracked stiffness of each pier.
Stiffness for existing enhanced and new wall shall be determined using the same
principles of mechanics.
C.4 Select Rehabilitation Method
The basic procedure to select Rehabilitation Method shall include the following:
• An analysis procedure shall be selected in accordance with the requirements and

limitations. (see Appendix C .4.1)
• The structure shall be analyzed to determine if it is adequate to meet the selected

Rehabilitation Objective(s) and, if it is not adequate, to identify specific
C-25
Appendix C
deficiencies. If initial analyses indicate that key elements or components of the

structure do not meet the acceptance criteria, it may be possible to demonstrate
acceptability by using more detailed and accurate analytical procedures.
• One or more rehabilitation strategies shall be developed to address the

deficiencies identified in the preliminary evaluation.
• A preliminary rehabilitation design shall be developed that is consistent with the
rehabilitation strategy.
• The structure and the preliminary rehabilitation measures shall be analyzed to
determine whether the rehabilitated structure will be adequate to meet the selected
Rehabilitation Objective(s).
• The process shall be repeated as required until a design solution is obtained that
meets the selected Rehabilitation Objective(s), as determined by the analysis.
C.4.1 Analysis Procedures
An analysis of the structure shall be conducted to determine the distribution of forces and
deformations induced in the structure by the design ground shaking and other seismic
hazards corresponding with the selected Rehabilitation Objective(s). The analysis shall
address the seismic demands and the capacity to resist these demands for all elements in
the structure that either:
• Are essential to the lateral stability of the structure (primary elements); or

• Are essential to the vertical load-carrying integrity of the building; or
• Are otherwise critical to meeting the Rehabilitation Objective and could be
subjected to damage as a result of the building's response to the earthquake
hazards.
The analysis procedure shall consist of one of the following:
• Linear analysis, including Linear Static Procedure (LSP), and Linear Dynamic
Procedure (LDP), including:
- Response Spectrum Analysis, and
- Linear Time-History Analysis, or
C-26
Appendix C
• Nonlinear analysis, including Nonlinear Static Procedure (NSP) and Nonlinear

Dynamic Procedure (NDP), or
• Alternative rational analysis
Linear Procedures
Linear procedures shall be permitted for buildings that do not have an irregularity. For
buildings that have one or more of the irregularities, linear procedures shall not be used
unless the earthquake demands on the building comply with the demand capacity ratio
(DCR) requirements.
The results of the linear procedures can be very inaccurate when applied to buildings with
highly irregular structural systems, unless the building is capable of responding to the
design earthquake(s) in a nearly elastic manner.
Method to determine limitations on use of linear procedures
The methodology presented in this section shall be used to determine the applicability of
linear analysis procedures based on four conditions of irregularity. The determination of
irregularity shall be based on the configuration of the rehabilitated structure. A linear
analysis to determine irregularity shall be performed by either an LSP or an LDP. The
results of the analysis shall be sued to identify the magnitude and uniformity of
distribution of inelastic demands on the primary elements and components of the lateral-
force-resisting system.
The magnitude and distribution of inelastic demands for existing and added primary
elements and components shall be defined by demand-capacity ratios (DCRs) and
components in accordance with Equation (C-3):
DCR = QUD
(C-3)
QCE
C-27
Appendix C
where:
Q UD = Force due to the gravity and earthquake loads calculated
Q CE = Expected strength of the component or element

DCRs shall be calculated for each action (such as axial force, moment, and shear) of each
primary component. The critical action for the component shall be the one with the
largest DCR. The DCR for this action shall be termed the critical element DCR at that
story. If an element at a particular story is composed of multiple components, then the
component with the largest computed DCR shall define the critical component for the
element at that story.
The applicability of linear procedures shall be determined as follows:
1. If all component DCRs 2.0, then linear procedures are applicable.

2. If one or more component DCRs exceed 2.0, and no irregularities.
3. If one or more component DCRs exceed 2.0 and any irregularity, then linear
procedures are not applicable, and shall not be used.
Basis of the Procedure
If the Linear Static Procedure (LSP) is selected for seismic analysis of the building,
the design seismic forces, their distribution over the height of the building, and the
corresponding internal forces and system displacements shall be determined using a
linearly elastic, static analysis in accordance with this section. Buildings shall be
modeled with linearly elastic stiffness and equivalent viscous damping values
consistent with components responding at or near yield level. The pseudo lateral load
shall be used to calculate internal forces and system displacements due to the design
earthquake.
Period Determination
The fundamental period of a building shall be calculated for direction under

consideration using one of the following analytical, empirical, or approximate
methods specified in this section.
. Method 1—Analytical
C-28
Appendix C
Eigenvalue (dynamic) analysis of the mathematical model of the building shall be

performed to determine the fundamental period of the building.
• Method 2—Empirical
The fundamental period of the building shall be determined in accordance with

Equation (C-4):
(C-4)
Where:
T = Fundamental period (in seconds) in the direction under consideration
C, = 0.035 for steel moment-resisting frame systems
= 0.0 19 for concrete moment moment-resisting frame systems
= 0.030 for steel eccentrically-braced frame systems
= 0.060 for wood buildings
= 0.020 for all other framing systems
h,, = Height (in feet) above the base to the roof level
= 0.90 for steel moment-resisting frame systems
= 0.90 for concrete moment-resisting frame systems
= 0.75 for all other framing systems
• Method 3—Approximate
For unreinforced masonry buildings with single span flexible diaphragms, six stories or
less in height, use of Equation (C-5) to approximate the fundamental period shall be
permitted.
C-29
Appendix C
T= (C-5)
where d is the maximum in-plane diaphragm displacement in inches, due to a lateral

load in the direction under consideration, equal to the weight tributary to the diaphragm.
• Pseudo Lateral Load
The pseudo lateral load in a given horizontal direction of a building shall be determined
using Equation (C-6). This load shall be used to design the vertical elements of the
lateral-force-resisting system.
V=C1C2C3CmSaW (C-6)
where:
V = Pseudo lateral load.
C = Modification factor to relate expected maximum inelastic displacements to

displacements calculated for linear elastic response, calculated as follows: C1 = 1.5
for T< 0.10 second. C1 = 1.0 for T second.
Linear interpolation shall be used to calculate C1 for intermediate values of T.
T = Fundamental period of the building in the direction under consideration.
T = Characteristic period of the response spectrum, defined as the period

associated with the transition from the constant acceleration segment of the
spectrum to the constant velocity segment of the spectrum.
C2 = Modification factor to represent the effects of pinched hysteretic shape, stiffness

degradation, and strength deterioration on maximum displacement response. For
linear procedures C2 shall be taken as 1.0.
C = Modification factor to represent increased displacements due to dynamic P-A

effects. For values of the stability coefficient 0, per Equation (3-2) less than 0.1,
C-30
Appendix C
otherwise C3 shall be set equal to 1.0, otherwise C3 shall be calculated as I + 5 (9-
0. 1)/T using 0 equal to the maximum value of 0, of all stories.
c m
= Effective mass factor to account for higher mode mass participation effects
obtained from Table C-8. Cm shall be taken as 1.0 if the fundamental period, T, is
greater than 1.0 second.
Values for Effective Mass Factor Cm1
Concrete Concrete Steel Steel Steel

No. of Moment Shear Concrete Moment Concentric Eccentric "
Stories Frame Wall Pier-Spandrel Frame Braced Frame Braced Frame Other
1—2 1.0 1.0 1.0 1,0 1.0 1.0 1.0
3 or more 0.9 0.8 0.8 0.9 0.9 0.9 1.0
1. betakeons I.Oith meotaiperiod, T, is gicaler than 1.0 second.
Table C-8 Values for Effective Mass Factor (FEMA-356, 2000)
Sa = Response spectrum acceleration, at the fundamental period and damping ratio of the
building in the direction under consideration. The values of S a shall be obtained
from the procedure.
W = Effective seismic weight of the building including the total dead load and applicable
portions of other gravity loads listed below:
1. In areas used for storage, a minimum 25% of the floor live load shall be
applicable. The live load shall be permitted to be reduced for tributary area as
approved by the code official. Floor live load in public garages and open
parking structures is not applicable.
2. Where all an allowance for partition load is included in the floor load design,
the actual partition weight or minimum weight of 10 psf of floor area,
whichever is greater, shall be applicable.
3. Total operating weight of permanent equipment.
C-31
Appendix C
4. Where the design flat roof snow load calculated in accordance with ASCE 7
exceeds 30 psf, the effective snow load shall be taken as 20 % of the design
snow load. Where the design flat roof snow load is less than 30 psf, the effective
snow load shall be permitted to be zero.
> Seismic Hazard
Seismic hazard due to ground shaking shall be based on the location of the building with
respect to causative faults, the regional and site-specific geologic characteristics, and
selected Earthquake Hazard Level. Assessment of seismic hazard due to earthquake-
induced geologic hazards shall be performed.
Seismic hazard due to ground shaking shall be defined as acceleration response spectra or
acceleration time-histories on either a probabilistic or deterministic basis.
Unless otherwise approved, the site-specific procedure shall be used where any of the
following conditions apply.
1. The building is located Type E soils and the mapped BSE-2 spectral response
acceleration at short periods (S exceeds 2.0g.
2. The building is located on Type F soils.

Exception: Where determined is less than 0.20g, use of a Type E soil profile
shall be permitted.
3. A time-history response analysis of the building will be performed as part of the
design.
> General Procedure for Hazard Due to Ground Shaking
The seismic hazard due to ground shaking shall be defined for any Earthquake Hazard
Level using approved spectral response acceleration contour maps of 5%-damped
response spectrum ordinates for short-period (0.2 second) and long-period (1 second)
response.
C-32
Appendix C
The short-period response acceleration parameter, S and the long-period response

acceleration parameter, S1. shall be determined as follows:
1. If the desired Earthquake Hazard Level corresponds with one of the mapped
Earthquake Hazard Levels, obtain spectral response acceleration parameters
directly from the maps. Values between contour lines shall be interpolated.
2. If the desired Earthquake Hazard Level does not correspond with the mapped
levels of hazard, then obtain spectral response acceleration parameters from the
available maps, and modify them to the desired hazard level, either by logarithmic
interpolation or extrapolation.
3. Obtain design spectral response acceleration parameters by adjusting the mapped
or modified spectral response acceleration parameters for site class effects.
4. If the desired Earthquake Hazard Level is the Basic Safety Earthquake 2 (BSE-2),
obtain spectral response acceleration parameters.
5. If the desired Earthquake Hazard Level is the Basic Safety Earthquake 1 (BSE-1),
obtain the spectral response acceleration parameters.
6. Using the design spectral response acceleration parameters that have been
adjusted for site class effects, develop the general response spectrum.
BSE-2 Response Acceleration Parameters
The design short-period response acceleration parameter, S , and design response
acceleration parameter at a one-second period, for the BSE-2 Earthquake Hazard

Level shall be determined using values of S and taken from approved MCE spectral
response acceleration contour maps and modified for site class.
Parameters and shall be obtained by interpolating between the values shown on the
map for the response acceleration contour lines on either side of the site, or by using the
value shown on the map for the higher contour adjacent to the site.
C-33
Appendix C
BSI- 1 Response Acceleration Parameters
The design short-period response acceleration parameter, S , and design response
acceleration parameter at a one-second period, S for the BSE-1 Earthquake Hazard

Level shall be taken s the smaller of the following:
1. The values of S and taken from approved 1O%'50 year spectral response
acceleration contour maps and modified for site class. Values between contour
lines shall be interpolated.
2. Two-thirds of the values of the parameters for the BSE-2 earthquake hazard level.
• Adjustment of Mapped Response Acceleration Parameters for Other Probabilities

Of Exceedance
Acceleration response spectra for earthquake hazard level corresponding to probabilities

of exceedance other than 2%/50 years and 1O%150 years shall be determined using the
following procedures.
• Probabilities of Exceedance Between 2%/50 Years and 1O%/50 Years
For probabilities of exceedance, EY' between 2%150 years and I O%/50 years, when the
mapped BSE-2 short-period response acceleration parameter, is less than 1 .5g, the
modified mapped short-period response acceleration parameter, S , and modified
mapped response acceleration parameter at a one-second period, S1. shall be determined

from Equation (9-7):
in(S1) = in(S110150) + — )}[O.606 ln(PR) — 3.7.3] (9-7)
where:
in(S1) = Natural logarithm of the spectral acceleration parameter ("i" = "s" for
short-period or "i" =1 for 1 second period) at the desired probability of

exceedance
C-34
Appendix C
I n(S ilO / 50) = Natural logarithm of the spectral acceleration parameter ("i" = "s" for
short-period or "i" =1 for 1 second period) at a lO%/50 year exceedance

rate
= Natural logarithm of the spectral acceleration parameter ("i" = "s" for
short-period or "i" =1 for 1 second period) for the BSE-2 hazard level
ln(P R) = Natural logarithm of the mean return period corresponding to the
exceedance probability of the desired earthquake hazard level

and the mean return period, at the desired exceedance probability shall be calculated
from Equation (C-9):
— (C-8)
ln(1—PEY)
Where P EY is the probability of exceedance (expressed as a decimal) in time Y (years) for

the desired earthquake hazard level.
When the mapped BSE-2 short-period response acceleration parameters, is greater

than or equal to 1 .5g, the modified mapped short-period response acceleration parameter,
S and modified mapped response acceleration parameter at a one-second period, S1, for
probabilities of exceedance between 2%/50 years and lO%/50 years shall be determined
from Equation (C-9):
si —s
— 110/50
C9
Where Si, 5j10/50 and are so defined above and n shall be obtained from Table C.9
C-35
Appendix C
Values of Exponent n for

Determination of Response
Acceleration Parameters at Hazard
Levels between years and
years; Sites where Mapped
BSE-2 Values of 1.5g
Region S5
California 0.29 0.29
Pacific Northwest 0.56 0.67
Intermountain 0.50 0.60
Central US 0.98 1.09
Eastern US 0.93 1.05
Table C-9 Values of Exponent n for determination of response acceleration parameters

between 10%/50 years and 2%/50 years (FEMA-356, 2000)
• General Horizontal Response Spectrum
A general horizontal response spectrum as shown in Figure C-3 shall be developed using
Equation (C-b), (C-i 1), and (C-12) for spectral response acceleration, S a versus
structural period, T, in the horizontal direction.
a
(C-b)
for 0<T<T0 , and
Sa for (C-li)
Sa (C- 12)
where and T0 are given by Equations (C-13) and (C-14)
B1) (C-13)
C-36
Appendix C
T0 = 0.2 T5 (C-14)
and where B and B1 shall be taken form Table C-10.
Damping Caefficients Bs and B1 as a

Function of Effective Damping 13
:Etfect:ive Viscous Damping 13
(percentage of critical
damping)1
.
8s B1
0.8 0.8
5 1.0 1.0
10 1.3 1.2
20 1.8 1.5
30 2.3 1.7
40 2.7 1.9
3.0 2.0
I. Damping coefficients shall be based on linear interpolation for
effective 'viscous damping values other than those given.
Figure C-3 A general horizontal response spectrum (FEMA-356, 2000)
Sa = Sxs _L
Bs I Ts + '0.41
0
Sa =
Sa = IB1T
I
U) Sxi/Bi
Ts 1.0
Period, T
Table C-10 Damping Coefficients and B1 (FEMA-356, 2000)
Use of spectral response accelerations calculated using Equation (C-l0) in the extreme
short period range (T<T0) shall only be permitted in dynamic analysis procedures and
only for modes other than the fundamental mode.
C-37
Appendix C
• General Vertical Response Spectrum
Where a vertical response spectrum is required, it shall be developed by taking two-thirds

of the spectral ordinates, at each period, obtained for the horizontal response spectrum.
• Damping Ratios
A 5% damped response spectrum shall be used for the rehabilitation design of all
buildings and structural systems.
• Probabilities of Exceedance Greater than 1O%150 years
For probabilities of exceedance greater than I O%150 years and when the mapped short-
period response acceleration parameter, S is less than 1.5g. the modified mapped short-
period response acceleration parameter, S and modified mapped response acceleration
parameter at a one-second period, S1. shall be determined from Equation (C-9), where
the exponent n is obtained from Table C-li.

Acceleration Parameters at
Probabilities of Exceedance 0,-eater
than 1O%/50 years;
Sites where Mapped BSE-2 Values
of 1.5g
Region .
S1
California 0.44 44
Pacific Northwest and 0.54 0.59
Intermountain
Central and Eastern US 0.77 0.80
Table C-il Values of Exponent n for determination of response acceleration parameters

at probabilities of exceedance greater than lO%150 years; sites where mapped BSE-2
Values of S < 1 .5g (FEMA-356, 2000)
C-38
Appendix C
For probabilities of exceedance greater than 1 O%/50 years and when the mapped short-
period response acceleration parameter, S is greater than or equal to 1 .5g, the modified
mapped short-period response acceleration parameter, S , and modified mapped

response acceleration parameter at a one-second period, S1, shall be determined from
Equation (C-9), where the exponent n is obtained from Table C-12.

Acceleration Parameters at
Probabilities of Exceedance Greater
than iO%/50 years;
Sites where Mapped BSE-2 Values
of S515g
Region S1
California 0.44 0.44

Pacific Northwest 0.89 0.96
Intermountain 0.54 0.59
Central US 0.89 •0.89
Eastern US 1.25 1.25
Table C-12 Values of Exponent n for determination of response acceleration parameters

at probabilities of exceedance greater than 1O%/50 years; sites where mapped BSE-2
Values of S 1 .5g (FEMA-356, 2000)
C-39
Appendix C
• Adjustment for Site Class
The design short-period spectral response acceleration parameter, S and the design
spectral response acceleration parameter at one-second, S shall be obtained from
Equations (C-15) and (C-16), respectively, as follows:
(C-15)
(C-16)
where .Fa and F,, are site coefficients determined respectively from Tables C-13 and C-
14, based on the site class and the values of the response acceleration parameters S and
S1 for the selected return period.
Values of Fa as a Function of Site

Class and Mapped Short-Period
Spectral Response Acceleration
Mapped Spectral Acceleration at Short-Periods
ss
Site Ss Ss=
Class 0.25 0.50 0.75 1.00 1 .25
:ij
NOTE: Straight-line interpolation shall be used for intermediate values of

S5.
* Site-specific geotechnical investigation and dynamic site response
anal! yses shall be performed.
Table C-13 Values of Fa (FEMA-356, 2000)
C-40
Appendix C
Values of as a of Site
Class and Mapped Spectral
Response Acceleration at
One-Second Period S1
Mapped Spectral Acceleration at One-Second

Per iod
Site
Class S1=O..2 S1=O.3 S1=O.4 SiO.50
A 0.8 0.8 th8 0.8
B 1.0 1.0 1.10 1.0 1.0
C 1.7 1.6 1.5 1.4 1.3
D 2.4 2.0 1.8 1.6 1.5
E 3.5 2.8 2A 2A
F * * * * *
NOTE: Straight-line interpolation shall be used for intermediate values of

SI.
* Site-specific geotechnical investigation and dynamic site response
analyses shall be performed
Table C-14 Values of (FEMA-356, 2000)
• Site classes
Site classes shall be defined as follows:
1. Class A: Hard rock with average shear wave velocity, >5.000ftlsec
2. Class B: Rock with 2,SOOftJseccv5 c5.000ftlsec
3. Class C: Very dense soil and soft rock with 1,200 2,SOOft/sec or with
either standard blow count N >50 or undrained shear strength >2,000 psf
C-41
Appendix C
4. Class D: Stiff soil with 600 ftlsec<v5 1,200 ftlsec or with 15<N 1,000
psf <2,000 psf

5. Class E: Any profile with more than 10 feet of soft clay defined as soil with
plasticity index P1>20, or water content w>40 percent, and <500 psf or a soil
profile with v5 <600 ftlsec.

6. Class F: Soils requiring site-specific evaluation:
6.1. Soils vulnerable to potential failure or collapse under seismic loading, such as
liquefiable soils, quick and highly-sensitive clays, collapsible weakly-
cemented soils
6.2. Feats and/or highly organic clays (H<10 feet of peat and/or highly organic
clay, where H = thickness of soil)
6.3. Very high plasticity clays (H>25 feet with P1>75)
6.4. Very thick soft/medium stiff clays (H>120 feet).
The parameters, , N and v5 are, respectively, the average values of the shear wave
velocity, Standard Penetration Test (SPT) blow count, and undrained shear strength of the
upper 100 feet of soils at the site. These values shall be calculated from Equation (C-17),
below:
(C-17)
where:
N, = SPT blow count in soil layer "i"
n = Number of layers of similar soil materials for which data is available
d, = Depth of layer "i"
= Undrained shear strength in layer "i"
C-42
Appendix C
v = Shear wave velocity of the soil in layer "i"
and
(C-19)
Where v data are available for the site, such data shall be used to classify the site. If
such data are not available, N data shall be used for cohesionless soil sites (sands,
gravels), and data for cohesive soil sites (clays). For rock in profile classes B and C,
classification shall be based either on measured or estimated values of Classification
of a site as Class A tock shall be based on measurements of v either for material at the
site itself, or for rock having the same formation adjacent to the site; otherwise, Class B
rock shall be assumed. Class A or B profiles shall not be assumed to be present if there is
more than 10 feet of soil between the rock surface and the base of the building.
• Distribution of Seismic Forces for Unreinforced Masonry Buildings with

Flexible Diaphragms
For unreinforced masonry buildings with flexible diaphragms for which the fundamental
period is calculated using Equation (C-5), it shall be permitted to calculate and distribute
the pseudo lateral loads as follows:
1. For each span of the building and at each level, calculated period from Equation
(C-5).
2. Using Equation (C-6), calculate pseudo lateral load for each span.
3. Apply the lateral loads calculated for all spans and calculate forces in vertical
seismic-resisting elements using tributary loads.
4. Diaphragm forces for evaluation of diaphragms shall be determined from the
results of step 3 above and distributed along the diaphragm span considering its
deflected shape.
5. Diaphragm deflection shall not exceed 6 inches for this method of distribution of
pseudo lateral loads to be applicable.
6.
C-43
Appendix C
C.4.2 Rehabilitation Strategies
A Rehabilitation Objective shall be achieved by implementing rehabilitation measures.

based on a strategy of addressing deficiencies identified by a prior seismic evaluation.
Each rehabilitation measure shall be evaluated in conjunction with other rehabilitation
measures, and the existing structures as a whole, to assure that the complete rehabilitation
scheme achieves the target Building Performance Level for the selected Earthquake
Hazard Level. The effects of rehabilitation on stiffness, strength, and deformability shall
be taken into account in an analyticaL model of the rehabilitated structure. The
compatibility of new and existing components and elements shall be checked at
displacements consistent with the demands produced by the selected Earthquake Hazard
Level and geologic site hazards present at the site.
• Local Modification of Components
Some existing buildings have substantial strength and stiffness, but some of their
components may not have adequate strength, toughness, or deformation capacity to
satisfy the Rehabilitation Objectives. An appropriate strategy for such structures may be
to perform local modifications of components that are inadequate while retaining the
basic configuration of the building's lateral-force-resisting system. Local modifications
that can be considered include improvement of component connectivity, component
strength, and/or component deformation capacity. This strategy tends to be the most
economical rehabilitation approach when only a few of the building's components are
inadequate. Local strengthening allows one or more under-strength elements or
connections to resist the strength demands predicted by the analysis without affecting the
overall response of the structure. This could include measures such as cover plating steel
beams or columns, or adding plywood sheathing to an existing timber diaphragm. Such
measures increase the strength of the element or component and allow it to resist more
earthquake-induced force before the onset of damage.
Local corrective measures that improve the deformation capacity or ductility of a

component allow it to resist large deformation levels with reduced amounts of damage,
without necessarily increasing the strength. One such measure is placements of a
C-44
Appendix C
confinement jacket around a reinforced concrete column to improve its ability to deform
without spalling or degrading reinforcement splices. Another measure is reduction of the
cross-section of selected structural components to increase their flexibility and response
displacement capacity.
• Removal or Lessening of Existing Irregularities
Removal or lessening of existing irregularities may be an effective rehabilitation strategy

if a seismic evaluation shows that the irregularities result in the inability of the building
to meet the selected Structural Performance Level.
The results of analysis should be reviewed to detect existing irregularities. Stiffness,

mass, and strength irregularities may be detected either by reviewing the results of a
linear analysis, examining the distribution of structural displacements and DCRs, or
reviewing the results of a nonlinear analysis by examining the distribution of structural
displacements and inelastic deformation demands. If the distribution of values of
structural displacements, DCRs, or inelastic deformation demands predicted by the
analysis is nonuniform with disproportionately high values within one story relative to
the adjacent story, or at one side of a building relative to the other, then an irregularity
exists.
Such irregularities are often, but bit always, caused by the presence of a discontinuity in
the structure, as for example, termination of a perimeter shear wall above the first story,
Simple removal of the irregularity may be sufficient to reduce demands predicted by the
analysis to acceptable levels. However, removal of discontinuities may be inappropriate
in the case of historic buildings, and the effect of such alterations on important historic
features should be considered carefully.
Effective corrective measures for removal or reduction of irregularities, such as soft or

weak stories, included the addition of braced frames or shear walls within the soft or
weak story. Torsional irregularities can be corrected by the addition of moment frames,
braced frames, or shear walls to balance the distribution of stiffness and mass within a
story. Discontinuous components such as columns or walls can be extended through the
zone of discontinuity.
C-45
Appendix C
Partial demolition can also be an effective corrective measures for irregularities, although
this obviously has significant impact on the appearance and utility of the building, and
this my not be an appropriate alternative for historic structures. Portions of the structure
that create the irregularity, such as setback towers or side wings, can be removed.
Expansion joints can be created to transform a single irregular building into multiple
regular structures: however, care must be taken to avoid the potential problems associated
with pounding.
• Global Structural Stiffening
Global stiffening of the structure may be effective rehabilitation strategy I the results of a
seismic evaluation show deficiencies attributable to excessive lateral deflection of the
building, and critical components do not have adequate ductility to resist the resulting
deformations.
Construction of new braced frames or shear walls within an existing structure is effective
measures for adding stiffness.
• Global Structural Strengthening
Global strengthening of the structure may be an effective rehabilitation strategy if the

results of a seismic evaluation show unacceptable performance attributable to a global
deficiency in structural strength. This can be identified when the onset of global inelastic
behaviour occurs at levels of ground shaking that are substantially less than the selected
level of ground shaking or large DCRs (or inelastic deformation demands) are present
throughout the structure. By providing supplemental strength to such a lateral-force-
resisting system, it is possible to raise the threshold of ground motion at which the onset
of damage occurs. Shear walls and braced frames are effective elements for this purpose,
but they may be significantly stiffer than the structure to which they are added, which
requires their design to provide nearly all of the structure's lateral resistance. Moment-
resisting frames, being more flexible, may be not become effective in the building's
response until existing brittle elements have already been damaged.
C-46
Appendix C
• Mass Reduction
Mass reduction may be an effective rehabilitation strategy if the results of a seismic

evaluation show deficiencies attributable to excessive building mass, global structural
flexibility, or global structural weakness. Mass and stiffness control the amount of force
and deformation induced in a structure by ground motion. Reductions in mass can result
in direct reductions in both the amount of force and deformation demand produced by
earthquakes and, therefore, can be used in lieu of structural strengthening and stiffening.
Mass can be reduced through demolition of upper stories, replacement of heavy cladding
and interior partitions, or removal of heavy storage and equipment loads.
• Seismic isolation
Seismic isolation may be an effective rehabilitation strategy if the results of a seismic

evaluation show deficiencies attributable to excessive seismic forces or deformation
demands, or if it is desired to protect important contents and non-structural components
from damage. When a structure is seismically isolated, compliant bearings are inserted
between the superstructure and its foundations. This produces a system (structure and
isolation bearings) with a nearly rigid body translation of the structure above the
bearings. Most of the deformation induced in the isolated system by the ground motion
occurs within the compliant bearings, which are specifically designed to resist these
concentrated displacements. Most bearings also have excellent energy dissipation
characteristics (damping). Together, this results in greatly reduced demands on the
existing elements of the structure, including contents and non-structural components. For
this reason, seismic isolation is often an appropriate strategy to achieve Enhanced
Rehabilitation Objectives that included the protection of historic fabric, valuable
contents, and equipment, or for buildings that contain important operations and functions.
This technique is most effective for relatively stiff buildings with low profiles and large
mass. It is less effective for light, flexible structures.
C-47
• Appendix C
• Supplemental Energy Dissipation
Installation of supplemental energy dissipation devices may be an effective rehabilitation

strategy if the results of a seismic evaluation show deficiencies attributable to excessive
deformations due to global structural flexibility in a building. Many available
technologies allow the energy imparted to s structure by ground motion to be dissipated
in a controlled manner through the action of special devices—fluid viscous dampers
(hydraulic cylinders), yielding plates, or friction pads—resulting in an overall reduction
in the displacements of the structure. The most commonly used devices dissipate energy
through frictional, hysteretic, or viscoelastic processes. In order to dissipate substantial
energy, dissipation devices must typically undergo significant deformation (or stroke),
which requires that the structural experience substantial lateral displacements. Therefore,
these systems are most effective in structures that are relatively flexible and have some
inelastic deformation capacity. Energy dissipaters are most commonly installed in
structures as components of braced frames. Depending on the characteristics of the
device, either static or dynamic stiffness is added to the structure as well as energy
dissipation capacity (damping). In some cases, although the structural displacements are
reduced, the forces delivered to the structure can actually be increased.
C-48
Appendix C
C.5 Acceptance Criteria for Performance-based Rehabilitation Design
C.5.1 General Requirements
Prior to selecting component acceptance criteria, components shall be classified as

primary or secondary, and actions shall be classified as deformation-controlled or force-
controlled.
Elements and components that provide the capacity of the structure to resist collapse
under seismic forces induced by ground motion in any direction shall be classified as
primary. Other elements and components shall be classified as secondary.
• Deformation-Controlled and Force Controlled Behaviour
The Type 1 curve depicted in Figure C-9 is representative of ductile behaviour where
there is an elastic range (point 0 to point 1 on the curve) followed by a plastic range
(points 1 to 3) with non-negligible residual strength and ability to support gravity loads at
point 3. The plastic range includes a strain hardening or softening range (points 2 to 3).
Primary component actions exhibiting this behaviour shall be classified as deformation-
controlled if the strain-hardening or strain-softening range is such that e>2g; otherwise,
they shall be classified as force-controlled. Secondary component actions exhibiting Type
1 behaviour shall be classified as deformation-controlled for any e/g ratio.
The Type 2 curve depicted in Figure C-9 is representative of ductile behaviour where
there is an elastic range (point 0 to point 1 on the curve) and a plastic range (points 1 to 2)
followed by loss of strength and loss of ability to support gravity loads beyond point 2.
Primary and secondary component actions exhibiting this type of behaviour shall be
classified as deformation-controlled if the plastic range is such that e 2g; otherwise,
they shall be classified as force-controlled.
The Type 3 curve depicted in Figure C-4 is representative of a brittle or nonductile

behavior where there is an elastic range (point 0 to point 1 on the curve) followed by loss
of strength and loss of ability to support gravity loads beyond point 1. Primary and
C-49
Appendix C
secondary component actions displaying Type 3 behavior shall be classified as force-

controlled.
Figure C-4 Component force versus deformation curves (FEMA-273, 1997)
• Deformation-Controlled Actions
Deformation-controlled design actions Q UD shall be calculated in accordance with

Equation (C-19):
QUD = QG ±QE (C-19)
where
Q E = Action due to design earthquake loads calculated using forces and analysis models.
Q G = Action due to design gravity loads.
The following component gravity forces, Q G' shall be considered for combination with
seismic loads.
When the effects of gravity and seismic loads are additive,
QG =1.1(QD+QL+Qs) (C-20)
When the effects of gravity counteract seismic loads,
QG =O.9QD (C-21)
C-50
Appendix C
where:
Q D = Dead load effect (action)
Q = Effective live load effect (action), equal to 25% of the unreduced design live load
but not less than the measured live load
Q = Effective snow load effect (action), equal to either 70% of the full design snow
load or, where conditions warrant and approved by the regulatory agency, not less
than 20% of the full design snow load, except that where the design snow load is 30
pounds per square foot or less, Q = 0.0
Evaluation of components for gravity and wind forces, in the absence of earthquake
forces, is beyond the scope of this document.
Q UD = Deformation-controlled design action due to gravity loads and earthquake loads

Because of possible anticipated nonlinear response of the structure, the design
actions as represented by Equation (C-19) may exceed the actual strength of the
component or element to resist these actions. The acceptance criteria take this
overload into account through use of a factor, m, which is an indirect measure of
the nonlinear deformation capacity of the component or element.
• Force-Controlled Actions
Force-controlled design actions, Q UF' shall be calculated using one of the following
methods:
1. Q UF shall be taken as the maximum action that can be developed in a component
based on a limit-state analysis considering the expected strength of the
components delivering load to the component under consideration, or the
maximum action developed in the component as limited by the nonlinear response
of the building.
2. Alternatively, Q UF shall be calculated in accordance with Equation (C-22).
C-51
Appendix C
± (C-22)
Q =Q
123
where:
Q UF = Force-controlled design action due to gravity loads in combination with

earthquake loads
J Force-delivery reduction factor, greater than or equal to 1.0, taken as the smallest
DCR of the components in the load path delivering force to the component in
question, calculated in accordance with Equation (2-1). Alternatively, values of J
equal to 2.0 in Zones of High Seismicity, 1.5 in Zones of Moderate Seismicity,
and 1.0 in Zones of Low Seismicity shall be permitted when not based on
calculated DCRs. J shall be taken as 1.0 for the Immediate Occupancy Structural
Performance Level. In any case where the forces contributing to Q UF are
delivered by components of the lateral force resisting system that remain elastic, J
shall be taken as 1.0.
The basic approach for calculating force-controlled actions for design differs from that
used for deformation-controlled actions because nonlinear deformations associated with
forced-controlled actions are not permitted. Therefore, force demands for force-
controlled actions should not exceed the force capacity (strength).
Ideally, an inelastic mechanism for the structure will be identified, and the force-
controlled actions, Q UF for design will be determined by limit analysis using that
mechanism. This approach will always produce a conservative estimate of the design
actions, even if an incorrect mechanism is selected. Where it is not possible to use limit
(or plastic) analysis, or in cases where design forces do not produce significant nonlinear
response in the building, it is acceptable to determine the force-controlled actions for
design using Equation (C-22).
Coefficients C1, C2 and C3 were introduced in Equation (C-6) to amplify the design
base shear to achieve a better estimate of the maximum displacements expected for
buildings responding in the inelastic range. Displacement amplifiers, C1, C2 and C3 are
C-52
Appendix C
divided out of Equation (C-22) when seeking an estimate of force level present in a
component when the building is responding inelastically.
Since J included for force-controlled actions, it may appear to be more advantageous to

treat an action as force-controlled when rn-factors are less than J. However, proper
application of force-controlled criteria requires a limit state analysis of demand and lower
bound calculation of capacity with yield a safe result whether an action is treated as
force- or deformation-controlled.
C.5.2 Acceptance Criteria for Linear Procedures
• Deformation-Controlled Actions
Deformation-controlled actions are primary and secondary components and elements

shall satisfy Equation (C-23).
mJ(QCE QUD (C-23)
where:
m = Component or element demand modifier (factor) to account for expected ductility

associated with this action at the selected Structural Performance Level, rn-factors
Q CE = Expected strength of the component or element at the deformation level under

consideration for deformation-controlled actions
K = Knowledge factor
Q CE' the expected strength, shall be determined considering all coexisting actions on the
component under the design loading condition.
• Force-Controlled Actions
Force-controlled actions in primary and secondary components and elements shall satisfy
Equation (C-24):
C-53
Appendix C
KQCL (C-24)
where:
Q CL = Lower-bound strength of a component or element at the deformation level under

consideration for force-controlled actions
Q CL' the lower-bound strength, shall be determined considering all coexisting actions on
the component under the design loading condition.
C.5.3 Strength acceptance criteria for masonry walls
Unreinforced masonry walls and piers shall be considered as deformation-controlled

components if their expected lateral strength limited by bed-joint sliding shear stress or
rocking (the lesser of values given by Equations C-25 and C-26) is less than the lower
bound lateral strength limited by diagonal tension or toe compressive stress (the lesser of
values given by Equations C-27 or C-29). Otherwise, these components shall be
considered as force-controlled components.
A. Expected lateral strength of walls and piers
Expected lateral strength of existing URM walls or pier components shall be based on
expected bed-joint sliding shear strength, or expected rocking strength, in accordance
with Equations C-25 and C-26, respectively. The strength of such URM walls or piers
shall be the lesser of:
QCE (C-25)
QCE (C-26)
where:
A = Area of net mortaredlgrouted section, in.2
C-54
Appendix C
h eff =Height to resultant of lateral force
L = Length of wall or pier
CE
= Expected vertical axial compressive force per load combinations in Equations C-
20 and C-21
v me = Expected bond-joint sliding shear strength, psi
VbJS = Lateral strength of wall or pier based on bed-joint shear strength, pounds
Vr = Lateral rocking strength of wall or pier component, pounds
a = Factor equal to 0.5 for fixed-free cantilever wall, or equal to 1.0 for fixed-fixed pier
Expected lateral strength of newly constructed wall or pier components shall be based on
the NEHERP Recommended Provisions (BSSC, 1995), with the exception that capacity
reduction factors shall be taken as equal to 1.0.
B. Lower bound lateral strength of walls and piers
Lower bound lateral strength of existing URM walls or pier components shall be limited
by diagonal tension stress or toe compressive stress, in accordance with Equation C-27
and C-29, respectively. The lateral strength of URM walls or piers shall be the lesser of
Q CL values given by these two equations.
f/h eff is larger than 0.67 and less than 1.00, then:
QCL =Vd( = f'dt (C-27)
QCL VIC aPCL[__j1 (C-29)

—
J
where:
C-55
Appendix C
A h eff , L and a are the same as given for Equations C-25 and C-26, and:
fa = Upper bound of vertical axial compressive stress from Equation C-20, psi
f' dt = Lower bound of masonry diagonal tension strength, psi
f'm = Lower bound of masonry compressive strength, psi
CL = Lower bound of verical compressive force from load combination of Equation C-
21, pounds
Vd, = Lateral strength limited by aidgonal tension stress, pounds
= Lateral strength limitsd by toe compressive stress, pounds.
For determination of the bed-joint shear strength, may be substituted for the
diagonal tension strength, f' dt' in Equation C-25.
The lower bound masonry compressive strength is frne divided by 1.6.
If the Linear Static Procedures are used, lateral forces from gravity and seismic effects
shall be less than the lower bound lateral strength, Q CL' as required by Equation C-22.
C. Lower bound vertical compressive strength of walls and piers
Lower bound vertical compressive strength of existing URM walls or pier components
shall be limited by masonry compressive stress per Equation C-29.
QCL = = 0.90 (0.95f'm (C-29)
where:
f'm is equal to the expected divided by 1.6.
C-56
Appendix C
If the Linear Static Procedures are used, vertical forces from gravity and seismic effects
shall be less than the lower bound lateral strength, Q CL' as stated in Equation C-22.
C.C.4 Deformation acceptance criteria for masonry walls
A. Linear procedures
If the linear procedures are used, the product of expected strength, Q CE' of those
components classified as deformation-controlled, multiplied by m factors given in Table
C-16 for particular Performance Levels and K factors , shall exceed the sum of unreduced
seismic forces, Q E' and gravity forces, Q G' per Equation 3-19. The limiting behavior
mode in Table C-15 shall be identified from the lower of the two expected strengths as
determined from Equations C-25 and C-26. For determination of m factors from Table C-
16, the vertical compressive stress, fae' shall be based on an expected value of gravity
compressive force given by the load combinations given in Equations C-20 and C-2 1.
Linear Static Procedure—rn Factors for UFiM In-Plane Walls and Piers
in Factors
Noic:
Table C-15 m factors for URM in-plane walls and piers (FEMA-273, 1997)
B. Nonlinear procedures
If the Nonlinear Static Procedure is used, deformation-controlled wall and pier

components shall be assumed to deflect to nonlinear lateral drifts as given in Table C-16.
Variables d and e, representing nonlinear deformation capacities for primary and
secondary components, are expressed in terms of story drift ratio percentages, as defined
in Figure C-5. The limiting behavior mode in Table C- 16 shall be identified from the
lower of the two expected strengths as determined from Equations C-25 and C-26.
C-57
Appendix C
1.0
eff
Figure C-5 Idealized force-deflection relation for walls (FEMA-273, 1997)
Nonlinear Static Procedure—Simplified Force-Deflection Relations for URM In-Plane Walls

and Piers
Note: is pennittvd
Table C-16 Simplified force-deflection relation for URM in-plane walls (FEMA-293,
1997)
For components of primary lateral-force-resisting elements, collapse shall be considered

at lateral drift percentages exceeding values of d in the table, and the Life Safety
Performance Level shall be considered at approximately 75% of the d value. For
components of secondary elements, collapse shall be considered at lateral drift
percentages exceeding the values of e value in the table, and the Life Safety Performance
C-58
Appendix C
Level shall be considered at approximately 75% of the e value in the table. Drift
percentages based on these criteria are given in Table C-17.
If the Nonlinear Dynamic Procedure is used, nonlinear force-deflection relations for wall
and pier components shall be established based on the information given in Table C-17,
or on a more comprehensive evaluation of the hysteretic characteristics of those
components.
C-59
Appendix D
APPENDIX D - TEST PROBLEMS
Specimens The reason why the loads are not available

Wall 1 No Problem
Wall 2 No Problem
Wall 3 No Problem
Wall 4 No Problem
Wall 5 No Problem
Wall 6 No Problem
Wall 7 No Problem
Wall 8 Out of range for Hydraulic jack
Wall 9 Failure of monitoring equipment
Wall 14 No Problem
Wall 15 No Problem
Wall 17 No Problem
Wall 18 No Problem
Wall 19 No Problem
D-1
Appendix E
APPENDIX E - PRATICAL INFORMATION

Appendix E
APPENDIX E-1 CABLE INFORMATION

RIGG I
A —
—.
TYPICAL GRADE 316 STAINLESS STEEL WIRE ROPE BREAKING LOAD SPECIFICATIONS
1x19 * 7x19 * 7x7

WIRE DIAMETER kN kg lb kN kg -- jb_ kN kg lb
1.2mm 3/64 in. 1.08 110 243 0.81 83 183 0.85 87 192
1.5mm 1/16 in. 2.11 215 474 1.44 147 324 2.01 205 452
2.0mm 5/64 in. 3.14 320 705 2.26 230 507 2.37 242 534
2.5mm 3/32 in. 4.90 500 1102 3.82 390 860 . 3.71 378 833
3.0mm 1/8 in. 7.06 720 1587 5.00 510 1124 5.34 544 1200
4.0mm 5/32 in. 12.60 1285 2833 8.89 907 2000 9.49 1089 2400
4.8mm 3/16 in. 18.90 1930 4255 12.60 1280 2822 14.10 1437. 3168
5.0mm — 19.60 1999 4407 13.90 1417 3124 14.80 1509 3327
5.6mm 7/32 in. 24.20 2470 5445 17.20 1750 3858 18.10 1850 4080
6.0mm — 28.00 2876 6340 20.00 2039 4495 21.40 2181 4810
6.4mm 1/4 in. 34.00 3440 7584 22.00 2280 5027 25.90 2642 5825
7.0mm 9/32 in. 35.00 3549 7824 36.00 2784 6138 29.10 2966 6540
8.0mm 5/16 in. 46.00 4640 10229 36.00 3630 8003 3874 8540
9.5mm 3/8 in. 65.00 6580 14506 51.00 5150 11354 57.20 5830 12850
10.0mm — 71.00 7250 15984 56.00 5670 12500 . 59.30 6045 13330
11.0mm 7/16 in. 86.00 8770 19335 68.00 6950 15322 76.70 7820 17240
12.0mm — 102.00 10401 22930 80.00 8158 17985 85.40 8700 19180
12.7mm 1/2 in. 119.00 12101 26678 90.00 9150 20172 107.00 10900 24030
14.0mm 9/16 in. 139.00 14174 31248 - 109.00 11115 24504 117.00 11930 26300
16.0mm 5/8 in. 182.00 18559 40916 133.00 13600 29983 — — —
19.0mm 3/4 in. 212.00 21618 47660 191.00 19500 44730 — — —
22.0mm 7/8 in. 285.00 29062 64071 - - - - - -

26.0mm 1 in. 398.00 40585 89475 - - - - - -
28.0mm - 516.00 52600 115963 - - - - - -
30.0mm — 660.00 67300 148371 — — — — — —
32.0mm 11/4 in. 751.00 76580 168830 - - - - - -

34.0mm — 848.00 88500 195109 — — . . — — —
36.0mm - 950.00 96875 213573 - - . - - - -

Note — Wire construction for sizes above 26mm diameter may not be 1 x 19
STRETCH IN WIRE ROPES

Elastic stretch (mm) =
Stretch is a characteristic of all wire ropes,
initially as permanent stretch when the load where:
TYPICAL VALUES FOR E ARE:-
is first applied and the individual wires bed
W = Applied Load (kN)
down, and then as conventional elastic stretch 1 x 19 ltJ7.5kN/mm2 (15.59 x 106 PSI.)
within the wires. Where stretch is critical L = Cable Length (mm)
to the application, initial stretch can be 7 x 19 47.5kN/mm2 (6.89 x 106 P.S.I.)
accounted for with cables pre-tensioned E — Strand Modulus (kNImmt)
or pre-stressed during swaging and cable 7x7 57.3 kN/mm2 (8.31 x 106 PSI.)
A=Areaof Cable =
manufacturing. Elastic stretch can be
calculated by the following formula: D = Nominal Diameter of Cable (mm)
SWAGING DIMENSIONS
Correct installation of a swage fitting requires that the shank of the fitting be formed down onto the wire with specialised dies and presses in accordance
with the following dimensions:
Wire mm 2.5 3.0 4.0 5.0 5.6 6.0 7.0 8.0 10.0 11.0 12.0 14.0 16.0 19.0 22.0 26.0
Diameter in. 3/32 1/8 5/32 7/32 1/4 9/32 5/16 3/8 7/16 1/2 9/16 5/8 3/4 7/8 1
Code mm 2.5M 03M 04M 05M . 06M 07M 08M 1OM — 12M 14M 16M 19M 22M 26M
Code in. 03 04 05 06 07 08 09 10 12 14 16 18 20 24 28 32
LenglhotWire mm 32.3 38.4 45 55 60.5 70 78.5 88 110 122 140 157 176 210 245 286
lnsideSwage in. 1.27151 1.77 2.17 2.38 2.76 3.09 3.46 4.33 4.80 5.51 6.18 6.93 8.27 9.65 11.26
Diameter 00 mm 5.54 6.35 7.54 9.12 10.85 12.55 16.13 17.86 19.84 21.44 24.99 8.17 34.52 40.46 46.02
Before swaging in. 0.218 0.250 0.297 0.359 0.427 0.494 0.563 0.635 0.703 0.781 0.844 0.984 1.109 1.359 1.593 1.812
DiameterOD mm 4.83 5.56 6.35 7.95 9.53 11.13 12.70 14.30 15.88 17.48 19.05 22.23 25.40 31.75 36.50 41.28
After swaging in. 0.190 0.219 0.250 0.313 0.375 0.438 0.500 0.563 0.625 0.688 0.750 0.875 1.000 1.250 1.437 1.625
ToleranceonOD mm +0, -0.13 - +0-0.20 +0-0.23

+0, -0.005 +0, -0.007 +0, -0.008 +0, -0.009 +0, -0.010 +0, -0.012
WIRE ROPE — STAINLESS STEEL (BMSA 29)
PRODUCT WIRE DIAM. REEL LENGTH BL (BSMA29) WEIGHT PER METRE
No. mm m (ft) kg (Ib) kg (Ib)
1 x 19 Grade 316
WR6119-1.5M 1.5 305 (1000) 180 (397) 0.011 (0.024)
WR61 19-02M 2.0 305 (1000) 320 (706) 0.019 (0.042)
WR61 19-2.5M 2.5 305 (1000) 500 (1103) 0.031 (0.068)
WR61 1 9-03M 3.0 305 (1000) 720 (1588) 0.045 (0.099)
WR61 19-04M 4.0 305 (1000) 1285 (2833) 0.079 (0.174)

1 x 19 Grade 316
WR61 19-05M 5.0 305 (1000) 2000 (4410) 0.124 (0.273)
6.0 305 (1000) 2876 (6342) 0.178 (0.392) CONSTRUCTION

WR61 1 9-06M
19 single strands
WR61 1 9-07M - 7.0 305 (1000) 3549 (7826) 0.243 (0.536)
Strongest
WR6119-08M 8.0 305 (1000) 4640 (10231) 0.317 (0.699) -
Lowest Stretch
95 305 (1000) 6546 (14434) 0.446 - (0:983)
Least flexible
WR6119-1OM 10.0 305 (1000) 7250 (15986) 0.495 (1.091)
ii.b 305 (1000) 8770 (19338) 0.624 (1.376) USES

Architectural, Structural
WR6119-12M 12.0 305 (1000) 10401 (22934) 0.713 (1.572)
*
rigging, Balustrading
WR6119-14M 14.0 305 (1000) 14174 (31254) 0.971 (2.141)
WR6119-16M 16.0 305 (1000) 18559 (40922) 1.270 (2.800)
WR6119-19M 19.0 305 (1000): 21618 (47668) 1.780 (3.925)
WR61 19-22M 22.0 305 (1000) 29062 (64082) 2.360 (5.204)

WR61 i9-26M 26.0 305 (1000) 40585 (89490) 3.300 (7.277)
7 x 19 Grade 316 7 x 19 Grade 316

WR6719-1.5M 1.5 305 (1000) 126 (278) 0.009 (0.020)
CONSTRUCTION
WR6719-02M 2.0 305 (1000) 275 (606) 0.017 (0.037) 7 strands of 19 wires
WR6719-2.5M 2.5 305 (1000) 428 (944) 0.027 (0.060) Most flexible
WR6719-03M 3.0 305 (1000) 612 (1349) 0.034 (0.075)
USES
WR6719-04M 4.0 305 (1000) 907 (2000) 0.061 (0.135) Lifting, Running rigging
WR6719-05M 5.0 305 (1000) 1418 (3127) 0.095 (0.209)
WR6719-06M 6.0 305 (1000) 2040 (4498) 0.138 (0.304)
WR6719-07M 7.0 305 (1000) 2785 (6141) 0.199 (0.439)
WR6719-08M 8.0 305 (1000) 3632 (8009) 0.243 (0.536)
WR6719-1 OM 10.0 305 (1000) 5673 (12509) 0.381 - (0.640)
WR6719-12M 12.0 305 (1000) 8163 (17999) 0.548 (1.208)
WR6719-14M 14.0 305 (1000) 11122 (24524) 0.746 - (1.645) - 7 x 7 Grade 316
WR6719-16M 16.0 305 (1000) 14387 (31723) 0.974 (2.148) CONSTRUCTION
7 strands of 7 wires
7 x 7 Grade 316
Semi-flexible
WR677-O3MA 3.0 305 (1000) 544 (1200) 0.035 (0.077) PVC
WR677 O4MA 40 305 (1000) 959 (2115) 0 063 (0139)
USES
3.0 305 (1000) 544 (1200) 0.072 (0.159)
Balustrading
WR677=O4MP** 4.0 305 (1000) 959 (2115) 0.082 (0.181)
* Standard imperial diameter wire. Refer imperial diameter fitting tables. ** Wire rope PVC coated.
HAND SWAGING TOOLS AND WIRE CUTTERS
PRODUCT CUTS WIRE TYPE WEIGHT

No. mm (in.) g (oz)
Wire Cutters - -
CLA6O6 4 (5/32) H.W.C.9 - - 310 (11.0)
CLA6O9 6 (1/4) H.W.C. 9 720 (25.4)
PRODUCT LENGTH TO SUIT FERRULE SIZE WEIGHT

No. mm (in.) mm (in.) g (oz)
Hand Swage Tools

CLA731 300 (12) 1.5, 2, 2.5 (1/16, 5/64, 3/32) 730 (25.8)
CLS763 600 (24) 2, 2.5, 3 (5/64, 3/32, 1/8) 2300 (81.3)
CLA775 750 (30) 4, 5 (5/32, 3/16) - 3960 (139.9)
CLA776 750 (30) 6 (1/4) 3930 (138.9)

______________
____________
__________ ___________
_____________
____________ __________________________
___________________________
______________________________
i 'ii u ii g :1; i i i II ru
Threaded swage terminals (2.5-26mm, 3/32"-l" wire) have rolled right Ronstan's Extra Long threaded swage terminals can be used to lengthen
hand UNF threads for greater strength and are manufactured from Grade cables to ensure adequate thread engagement is maintained.
316 Stainless Steel for maximum corrosion resistance. WIRE DIAMETER
Metric: 3mm — 26mm Imperial: 1/8" —1"
WIRE PRODUCT SUIT THREAD A B C D E WEIGHT

DIAM. No. TURNBUCKLE TYPE mm (in.) mm (in.) mm (in.) mm (in.) mm (in.) g (oz)
Metric_Swage
2.5mm RF1513M2.503 RF217,218,2193/16" UNF 4.0 (5/32) 78 (31/8) 32 (t)J4) 6.5 (1/4) 10 (0.4)
3mm RF1512M0304 RF1578M0304 1/4" UNF 4.0 (5/32) 39 (1 9/16) 133 (51/4) 75 (215/16) 7.0 (9/32) 25 (0.9)
4mm RF1512M0404 RF1578M0404 1/4' UNF 4.0 (5/32) 45 (13/4) 140 (51/2) 75 (215/16) 7.0 (9/32) 30 (1.1)
4mm RF1512M0405 RF1578M0405 5/16" UNF 5.3 (7/32) 45 (13/4) 152 (6) 89 (31/2) 8.7 (11/32) 50 (1.8)
5mm 1/4" UNF 4.0 (5/32) 139(51/2) 75 (215/16) 7.0
5mm RF1512M0505 RF1578M0505 5/16" UNF 5.3 (7/32) 55 (23/16) 166 (61/2) 89 (31/2) 8.7 (11/32) 51 (1.8)
5mm RF1512M0506 RF1578M0506 3/8" UNF 6.3 (1/4) 55(23/16) 105 (41/8) 10.5 (13/32) 81 (2.9)
6mm RF1512M0607 RF1578M0607 7/16" UNF 7.6 (5/16) 70 (23/4) 221 (811/16) 125 (47/8) 11.9 (15/32) 155 (5.5)
6mm RF1512M0608 RF1578M0608 1/2" (11/32) 241 (91/2) 143 (55/8) 13.9 (9/16) 205 (7.2)
7mm RF1512M0708 RF1578M0708 1/2" UNF 8.5 (1 1/32) 79 (31/8) 251 (97/8) 143 (55/8) 13.9 (9/16) 225 (7.9)
8mm 1/2" UNF 8.5 (11/32) 259 (101/4) 143 (55/8) 13.9 (9/16) 240 (8.5)
8mm RF1512M0810 RF1578M0810 5/8" UNF 11.5 (7/16) 88 (31/2) 313 (12 3/8) 191 (71/2) 16.7 (21/32) 420 (14.8)
10mm RF1512M1O1O RF1578M1010 5/8" UNF 11.5 (7/16) 110 (45/16) 336 (131/4) 191 (71/2) 16.7 (21/32) 460
12mm RF1513M1212 RF1579M1212 3/4" UNF 14.5 (9/16) 140 (51/2) 286 (11 3/8) 105 (41/8) 21.2 (27/32) 560 (19.7)
14mm RF1513M1414 Rf1579M1414 7/8" UNF 18.0 (23/32) 158 (61/4) 320 (125/8) 118 (45/8) 24.0 (15/16) 850
16mm RF1513M1616 RF1579M1616 1" UNF 18.0 (23/32) 177 (7) 360 (14 1/8) 130 (51/8) 25.5 (1) 1200 (42.3)
19mm RF1513M1920 RF1579M1920 11/4" UNF 25.5 (1) 211 (85/16) 421 (169/16) 150 (515/16) 33.5 (15/16) 2270 (80.1)
22mm RF1513M2220 RF1579M2220 11/4" UNF 25.5 (1) 226 (87/8) 460 (18 1/8) 150 (515/16) 33.5 (15/16) 2220 (78.3)
RF1579M2622 13/8" UNF 29.5 (15/32) 272 (103/4) 588 (233/16) 236 (95/16) 38.5 (11/2) 4680 (165.1)
28mm CUSTOM 0 - M39 36.0 (113/32) - - 458 (18 1/32) 179 (71/16) - - 4040 (141.2)
30mm CUSTOM 0 - 1/1 6) - - 507 (20)197 (73/4) - -
32mm CUSTOM0 - M45 46.0 (113/16) - - 529 (20 13/16) 205 (81/16) - - 6600 (232.8)
CUSTOMQ - - - - -
36mm CUSTOM - M48 46.0 (113/16) - - 595 (23 7/16) 231 (93/32) - - 8820 (311.1)
Imperial Swage - -
3/32 RF217,218,219 3/16" UNF 4.0 (5/32) 30 (13/16) 78 (31/8) (1/4) 10 (0.4)
1/8 in. RF151O-0404 RF1576-0404 1/4" UNF 4.0 (5/32) 39 (19/16) 132 (53/16) 75 (215/16) 7.0 (9/32) 30 (1.1)
5/32 in. RF151O-O504 RF1576-O5O4 1/4" UNF 4.0 (f3/4) 141 (51/2) 75 (215/16) 7.0 (9/32) 30 (1.1)
5/32 in. RF151O-0505 RF1576-0505 5/16" UNF 5.3 (7/32) 45 (13/4) 155 (61/8) 89 (31/2) 8.7 (11/32) 45 (1.6)
3/16 in. RF1576-O6O4 1/4" UNF 4.0 (5/32) 44 (13/4) 139 (51/2) 75 (215/16) 7.0 41 (1.4)
3/16 in. RF151O-0605 RF1576-0605 5/16" UNF 5.3 (7/32) 55 (23/16) 164 (61/2) 89 (31/2) 8.7 (11/32) 56 (2.0)
3/16 in. RF151O-06O6 RF1576-O6O6 3/8" UNF 6.3 (1/4) 55 (23/16) 179 (71/16) 105 (41/8) 10.5 (13/32) 81 (2.9)
7/32 in. RF1 510-0706 RF1 576-0706 3/8" UNF 6.3 (1/4) 60 (23/8) 185 (71/4) 105 (41/8) 10.5 (13/32) 91 (3.2)
1/4 in. RF151O-O8O7 - 7/16" UNF 70 (23/4) 221(811/16) 124 (47/8) 11.9 (15/32) 150 (5.3)
1/4 in. RF151O-08O8 RF1 576-0808 1/2" UNF 8.5 (1 1/32) 70 (23/4) 241 (91/2) 143 (55/8) 13.9 (9/16) 195 (6.9)
9/32 in. RF1 51 0-0908 RF1 576-0908 1/2" UNF 8.5 (11/32) 78 (31/16) 248 (93/4) 143 (53/8) 13.9 (9/16) 220 (7.8)
5/16 in. RF151O-1008 RF1576-1008 1/2" UNF 8.5 (11/32) 88 (31/2) 261 (10 1/4) 143 (55/8) 13.9 (9/16) 240 (8.5)
5/16 in. RF151O-1009 - 9/16" UNF 10.4 (13/32) 88 (31/2) 276 (107/8) 159 (61/4) 15.6 (5/8) 310 (10.9)
5/16 in. RF151O-1O10 RF1576-1010 5/8" UNF 11.5 (7/16) 88 (31/2) 306 (12 1/16) 191 (71/2) 16.7 (21/32) 410 (14.5)
3/8 in. RF151O-121O RF1576-1210 5/8" UNF 11.5 (7/16) 109 (45/16) 335 (133/16) 191 (71/2) 16.7 (21/32) 460 (16.2)
7/16 in. RF1511-1412 RF1577-1412 3/4" UNF 12.0 (15/32) 122 (413/16) 262 (10 5/16) 105 (41/8) 21.2 (27/32) 460 (16.2)
1/2 in. RF1511-1614 RF1577-1614 7/8" UNF 140 (51/2) 297 (1111/16) 118 (45/8) 24.0 (15/16) 670 (23.6)
9/16 in. RF1511-1816 RF1577-1816 1" UNF 18.0 (23/32) 157 (63/16) 341 (13 3/8) 130 (51/8) 25.5 (1) 1040 (36.7)
5/8 in. RF1511-2016 RF1577-2016 1" UNF 18.0 (615/16) 355 (14) 130 (1) 1210 (42.7)
3/4 in. RF1511-2420 RF1577-2420 11/4" UNF 25.5 (1) 219 (85/8) 419 (16 4/8) 150 (515/16) 33.5 (15/16) 2320 (81.8)
7/8 in. RF1511-2820 RF1577-2820 11/4" UNF 25.5 (1) 245 (95/8) 462 (183/16) 150 (515/16) 33.5 (15/16) 2720 (95.9)
1 in. RF1511-3222 RF1577-3220 13/8" UNF 29.5 (15/32) 281 (11 1/16) 586 (23 1/8) 236 (9 5/16) 38.5 (11/2) 4770 (168.2)
EXTRA LONG Imperial_Swage
1/8 in. RF151OLO4O4 UNF 4.0 (5/32) 234 75 (2 5/1 6r7.0 (9/32) 30 (1.1)
3/16 in. RF1510L0605 RF1576-O6O5 5/16" UNF 5.3 (7/32) 55 (23/16) 316 (12 7/16) 89 (31/2) 8.7 (11/32) 56 (2.0)
*Recommended for balustrade applications only. B.L is below the rated B.L of 3/16" and 5mm wire
O Available as a completed cable assembly only
Appendix E
APPENDIX E-2 CHEMICAL ANCHOR

INFORMATION
® Ramset
CHEMShI ® INJECTION 800 SERIES

Chemset InjectIon 800 Series are Application
a chemical anchor system based High strength anchoring where vibration is a
on epoxy mortar. The two parts are consideration, wind load structures and
dispensed and mixed in one action machinery loadings, balustrades, handrails,
through a static mixing nozzle, fans, motors, safety barriers and structural
which allows accurate mixing connections.
with no mess.
Chemset Injection 800 Series are tough, Storage
corrosion resistant products suitable for Store in a cool dry place between 5°C and
anchoring jobs close to edges where there 25°C. Keep away from direct sunlight.
is a need for anchoring jobs close to Shelf life is minimum of 12 months in original
edges where there is a need to avoid sealed package at 15°C, not exposed to light.
bursting stress on the surrounding substrate.
The high strength and toughness of Materials
Chemset 800 mortars allows higher loads on Epoxy resin and hardener
anchored fixtures. This enables designs to
be optimised by providing a broader range
of anchoring options. Chemset 800 mortars
also have exceptional corrosion resistance
making them ideal to use in areas where
there is frequent cleaning or where chemicals
are frequently used. Chemset 800 Series
mortars can be used to fix starter bars,
wall ties, threaded studs, bolts and large
screws into concrete, brickwork, masonry
and stone.
The Chemset 800 Series consists of two
products, each tailored to different climates.
Chemset 801 is designed to provide
rapid cure with adequate working time in
temperate climates. Chemset 802 has been
developed with a slower pot life for areas 1 OrtI correct diameter hole 2. Insert Mixer Nozzle and 3. PJlow Epoxy-Set to cure
where there is high ambient temperature or to recommended depth. inject mixture into hole. as per recommendations.
where a slowermortar is required. Push Stud Bolt to the Attach fixture.
bottom of the hole whilst
turning.
For fixing to
Concrete, solid brickwork, stone.
Features
— Two component epoxy technology with
one component dispensing system.
— Excellent resistance to alkali and moisture
in new concrete.
— Tough and durable for long term high
strength.
— No shrinkage during curing.
— Suitable for diamond cored holes.
- Non flammable.
- Low odour.
Phone 1300 180 063

www.ramset.com.au
® Ramset
CHEMSb I ® INJECTION 800 SERIES

INSTALLATION AND PERFORMANCE DETAILS
,,,,,
Anchor Thread Hole Installation Structural Limits Rec. Working Load (kN) See Safety Factors p11 & 12 No.01
Size Size 0 Embedded Fixture Fixture Tight Minimum Minimum Minimum Fixings
(mm) Depth Clearance Thickness Torque Edge Anchor Structural 2OMPa 3OMPa 4OMPa Per
(mm) 0 (mm) (Nm) Distance Spacing Thickness Cartridge
(mm) (mm) (mml (mm) Tensile •Shear Tensile •Shean Tensile •Shear
8 MO 10 80 11 15 8 30 50 100 5.6 4.4 6.9 4.4 8.1 4.4 119
10 MiD 12 90 13 25 15 40 60 120 7.6 7.1 9.3 7.1 10.7 7.1 81
12 M12 14 110 15 30 30 50 70 140 10.8 10.5 13.2 10.5 15.3 10.5 53
16 M16 18 125 19 40 70 65 100 160 15,8 19.8 19.3 19.8 22.3 19.8 34
20 M20 24 150 25 75 140 80 120 190 25.3 30.0 31.0 30.0 35.7 30.0 13
M20 24 **170 25 55 140 80 120 220 28.6 30.0 35.1 30.0 40.5 30.0 11
24 M24 26 160 30 105 230 95 145 200 29.2 43.4 35.8 43.4 41.3 43.4 15
M24 26 **210 30 55 230 95 145 270 38.3 43.4 47.0 43.4 54.2 43.4 11
• For shear loads acting toward the edge(s) of the concrete, the above edge distances and spacings are not applicable, please consult Ramset Technical Consultant. This table
should be read in conjunction with the Rarnset Engineers Design Manual.
Note: See pages 56 — 59 for anchors and accessories.
• SET1ING TIMES . FIXINGS P ER CARTRIDGE (400m1) Approximately

Base Standard Grade Standard Grade Anchor Hole 0 (mm) Hole Depth (mm) Number of Fixings
Temp. Gel Time Loading Time Size Per Cartridge
5 l20minutes 66hours 8 10 80 119
10 90 33 10 12 90 81
20 20 23 • 12 14 . 110 53
25 15 20 16 18 125 34
30 12 17 20 24 150 13
40 7 10 24 26 160 15
Temperatures refer to base material not air temperature. Shetf life 6 months.
Phone 1300 780 063

www.ramset.com.au
________
® Ramset
Concrete Results"
Chemical Anchoring
ChemSet" Injection 800 Series
20
ChemSetTM Injection 800 Series

GENERAL INFORMATION
PERFORMANCE RELATED MATERIAL SPECIFICATION INSTALLATION RELATED

I j I
Product
ChemSet" Injection 801 is a heavy true epoxy
injection anchor.
ChemSei" Injection 802 is a heavy duty, true epoxy
injection anchor.
Benefits, Advantages and Features

Suitable for structural applications:
— High bond strength.
Suitable for use in contact with drinking water:
Meets AS/NZ4020 - 1999.
Suitable for fire rated dowel bars:
Meets AS1530.4.
Suitable for diamond cored holes:
High bond strength.
Suitable for use in industrial environments where corrosion
and alkali resistance are required:
(See table 5.3 pages 22 and 23.)
Versatile:
Rapid cure for temperate climates and longer working time for deep holes
or hot climates.
Installation
Installation temperature limits:
— Substrate: 5°C to 40°C.
Mortar: 18°C to 35°C.
Load should not be applied to anchor until the
chemical has sufficiently cured as specified.
Service temperature limits:

-10°C to 80°C.
Setting Times
1. Dnll recommended diameter and depth hole.
2. Clean hole with hole cleaning brush.
Gel lime Loading lime Gel lime Loading lime
Remove all debris using hole blower. - (mins) (Ins) (mins) (Iris)
Hole must be dry.
3. Insert mixing nozzle to bottom of hole. 40°C — — I — 7 I 10
Fill hole to 3/4 the hole depth slowly,
ensuring no air pockets form. 3o:c — 3 I 10 12 I 17
4. Insert Ramset" ChemSet Anchor E 25 C - 4 12 20 20

Stud/rebar to bottom of hole while turning. 20°C —
6 14 25 23
5. CtiemSet" Injection to cure as per setting times. —
40 24 90 33
6. Attach fixture.
5°C - 75 36 120 66
0
Nate Cartridge temperature minimum 15°C.
® Ramset
Chemical Anchoring
Injection 800 Series
Installation and Working Load limit performance details:

ChemSetTM Injection 800 Series and Anchor Studs
Installation details Minimum_dimensions* Winking Load Limit
Anchor
DnIIed hole Fixture hole Anchor Tightening Edge Anchor Substrate Tension, Na (1(N)
size, db
diameter, db diameter, d1 effective depth torque, 1, distance, spacing, thickness, bm Shear, Va (kN) Concrete compressive strength, fc
(mm)
(mm) (mm) h (mm) (Nm) (mm) (mm) (mm) 20 MPa 32 MPa 40 MPa
MB 10 10 80 10 30 50 100 5.3 6.5 6.5 6.5
M10 12 12 90 20 40 60 120 7.1 8.6 9.9 10.3
M12 14 15 110 40 50 70 140 10.5 12.5 14.4 15.3
M16 18 19 125 95 65 100 160 19.9 16.9 19.6 20.8
150 190 30.0 24.3 28.1 29.9

M20 24 24 180 80 120
170 220 30.0 29.3 33.9 36.1
160 200 43.4 28.8 33.3 35.5

M24 26 28 315 95 145
210 270 43.4 43.3 50.1 53.4
* Note: For shear loads acting towards an edge or where these minimum dimensions are not achievable, please use the simplified strength limit state designprocess to verify capacity.
DESCRIPTION AND PART NUMBERS
Description Cartridge Size Climate Part No.
ChemSet" 801 Cartridge 400 ml Temperate C8O1C
ChemSet" 801 Jumbo Cartridge 750 ml Temperate C8O1J
ChemSet" 802 Cartridge 400 ml Tropical C802C
Mixer Nozzle for 800 Series — — ISNE
Effective depth, h (mm)
Preferred h = otherwise,
h = Le - t
h 6 * dh
t = total thickness of material(s) being fastened.
ENGINEERING PROPERTIES
Refer to "Engineering Properties" for ChemSet'" Anchor Studs

on page 113.
® Ramset
Chemical 4 jichoring
Strength Limit State Design Injection 800 Series
20.4
Select anchor to he evaluated
Table la Indicative combined loading — interaction diagram

70
Notes:
Shear limited by steel capacity.
— Tension limited by concrete capacity
60
using nominal depths.
— No edge or spacing effects.
2 50 MPa
0,
°' 40
=
0,
0,
1::
0
0 10 20 30 40 50 60 70 80 90 100
Design shear action effect, (kN)
Table lb Absolute minimum edge distance and anchor spacing values, em and am (mm)
Anchor size, db M8 M1O M12 M16 M20 M24
em, am 25 30 35 50 60 75
Step ic Calculate anchor effective depth, h (mm)

Refer to "Description and Part Numbers" table for
ChemSef" Anchor Studs (page 113).
Preferred h = otherwise,
h = Le - t
h 6 * dh
t = total thickness of material(s) being fastened.
[Checkpoint
LI Anchor size determined, absolute minima compliance achieved, effective depth (h) calculated.
® Ramset
Concrete
20 Chemical Anchoring
Strength Limit State Design
Verify concrete tensile capacity - per anchor
Table 2a Reduced characteristic ultimate concrete tensile capacity, (kN), = 0.6, = 32 MPa
Anchor size, db M8 M10 M12 M16 M20 M24
Drilled hole dia., dh (mm) 10 12 14 18 24 26

50
60 8.9
70 11.2 12.2
80 13.7 15.0
90 17.8 19.2
100 20.9 22.5
110 25.9 29.1
120 29.5 33.1
125 31.4 35.2 38.5
140 37.2 41.8 45.7
150 50.6 54.5
160 55.8 60.0
170 61.1 65.7
180 66.6 71.6
190 72.2 77.7
200 78.0 83.9
210 90.2
220 96.8
230 103.4
240 110.3
Bold values are at Anchor Stud nominal depths.
Note: Effective depth, h must be 6 x drilled hole diameter, dh for anchor to achieve tabled shear capacities.
Table 2b Concrete compressive strength effect, tension,

f' (MPa) 20 25 32 40 50
0.87 0.93 1.00 1.07 1.14
Table 2c Edge distance effect, tension, Xne
* Anchor size, db
Edge distance, e (mm)
M8 M10 M12 M16 M20 M24
fl
25 0.85
30 0.98 0.83
35 1 0.91 0.81
40 1 0.88
50 1 0.85
60 0.96 0.83
65 1 0.87
75 0.96 0.85
80 1 0.88
100 1
Table 2d Anchor spacing effect, end of a row, tension, Xflae
n n
*
a
Anchor size, db
Anchor spacing, a (mm)
M8 M10 M12 M16 M20 M24
25 0.76
30 0.81 0.75
35 0.86 0.79 0.74
40 0.92 0.83 0.78 .
50 1 0.92 0.85 0.76

60 1 0.92 0.81 0.75
15 1 0.89 0.81 0.76 •
100 1 0.92 0.85

120 1 0.92
150 1
® Ramset
Concrete
Chemical Anchoring
Strength Limit State Design ChemSet" Injection 800 Series
20
Table 2e Anchor spacing effect, internal to a row, tension, Xnai
I
II Anchor size, dh

M8 MiD M12 Mi6 M20 M24
25 0.52
30 0.63 0.50
35 0.73 0.58 0.49
40 0.83 0.67 0.56
50 1 0.83 0.69 0.52
60 1 0.83 0.63 0.50
15 1 0.78 0.63 0.52
100 1 0.83 0.69
120 1 0.83
150 1
Checkpoint
Verify anchor tensile capacity - per anchor
Table 3a Reduced characteristic ultimate steel tensile capacity, (kN), 0.8

Anchor size, db MO MiD Mi2 MiS M20 M24
ChemSet" Anchor Stud

14.2 22.7 33.8 64.1 96.5 139.8
Grade 5.8 Carbon Steel
ChemSet" Anchor Stud

16.5 26.1 37.9 70.7 110.3 158.9
A4/3i6 Stainless Steel
Step 3b Reduced characteristic ultimate bolt steel tensile capacity, ONif (kN)
Not appropriate for this product.
Design reduced ultimate tensile capacity, ONur
DNur = minimum of ONurc,
Check N* / ONur 1,
if not satisfied return to step 1
® Ramset
Concrete
20 Chemical Anchoring
Strength Limit State Design
Verify concrete shear capacity - per anchor
Table 4a Reduced characteristic ultimate concrete edge shear capacity, DV (kN), 0q = 0.6, =32 MPa
Anchor size, db M8 M10 M12 M16 M20 M24
Edge distance, e (mm)
25 1.6
30 2.2 2.4
35 2.7 3.0 3.2 3.6
50 4.6 5.1 5.5 6.2 6.9
60 6.1 6.7 7.2 8.2 9.0 9.8
75 8.5 9.3 10.1 11.4 12.6 13.7
125 18.3 20.1 21.7 24.6 27.1 29.5
200 40.6 43.8 49.7 54.9 59.7
300 80.5 91.3 100.9 109.7
400 140.5 155.4 168.9
500 . 217.2 236.1
600 310.3
Note: Effective depth. h must be 6 x drilled hole diameter. dh for anchor to achieve tabled shear capacities.
Table 4b Concrete compressive strength effect, concrete edge shear,

(IVIPa) 20 25 32 40 50
0.79 0.88 1.00 1.12 1.25
Table 4c Load direction effect, concrete edge shear,

Angle, a° 0 10 20 30 40 50 60 10 80 90- 180
Xvd 1.00 1.04 1.16 1.32 1.50 1.66 1.80 1.91 1.98 2.00
Load direction effect,

conc. edge shear, X0d
Table 4d Anchor spacing effect, concrete edge shear,
Note: For single anchor designs, Xva = 1.0
Edge distance, e (mm) 25 30 35 50 60 15 125 200 300 400 500 600

25 0.70 0.67 0.64 0.60 0.58 0.57 0.54
30 0.74 0.70 0.67 0.62 0.60 0.58 0.55 0.53
35 0.78 0.73 0.70 0.64 0.62 0.59 0.56 0.54 0.52
50 0.90 0.83 0.79 0.70 0.67 0.63 0.58 0.55 0.53 0.53
60 0.98 0.90 0.84 0.74 0.70 0.66 0.60 0.56 0.54 053 0.52
15 1.00 1.00 0.93 0.80 0.75 0.70 0.62 0.58 0.55 0.54 0.53 0.53
150 1.00 1.00 1.00 0.90 0.74 0.65 0.60 0.58 0.56 0.55
200 1.00 0.82 0.70 0.63 0.60 0.58 0.57
300 0.98 0.80 0.70 0.65 0.62 0.60
400 1.00 0.90 0.77 0.70 0.66 0.63
500 1.00 0.83 0.75 0.70 0.67
625 0.92 081 0.75 0.71
150 1.00 0.88 0.80 0.75
875 0.94 0.85 0.79
1000 • 1.00 0.90 0.83
1250 1.00 0.92
1500 1.00
® Ramset
Chemical Anchoring
Strength Limit State Design ChemSet' Injection 800 Series
Table 4e Multiple anchors effect, concrete edge shear,
Note: For single anchor designs, = 1.0
Anchor spacing
0.20 0.40 0.60 0.80 1.00 120 1.40 1.60 1.80 2.00 2.25 2.50
Edge distance, a / e
Number of anchors, ii
2 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
3 0.72 0.78 0.80 0.83 0.86 0.88 0.91 0.93 0.95 0.96 0.98 1.00
4 0.57 0.64 0.69 0.74 0.79 0.82 0.86 0.89 0.92 0.94 0.97 1.00
5 0.49 0.57 0.63 0.69 0.74 0.79 0.83 0.87 0.90 0.93 0.97 1.00
6 0.43 0.52 0.59 0.66 0.71 0.77 0.81 0.85 0.89 0.93 0.96 1.00
1 0.39 0.48 0.56 0.63 0.69 0.75 0.80 0.84 0.88 0.92 0.96 1.00
8 0.36 0.46 0.54 0.61 0.68 0.74 0.79 0.84 0.88 0.92 0.96 1.00
9 0.34 0.44 0.52 0.60 0.67 0.73 0.78 0.83 0.87 0.91 0.96 1.00
10 0.32 0.42 0.51 0.59 0.66 0.72 0.17 0.82 0.81 0.91 0.96 1.00
15 0.26 0.37 0.47 0.55 0.63 0.70 0.76 0.81 0.86 0.90 0.95 1.00
20 0.23 0.35 0.45 0.54 0.61 0.68 0.75 0.80 0.85 0.90 0.95 1.00
Checkpoint
Verify anchor shear capacity - per anchor
Table 5a Reduced characteristic ultimate steel shear capacity, (kN), 0.8

Anchor size, db MR M10 M12 M16 M20 M24
Threaded Stud
8.9 14.1 21.0 39.7 59.9 86.8
Grade 5.8 Carbon Steel
Threaded Stud
12.7 23.9 34.7 64.6 100.8 145.2
A4/316 Stainless Steel
Step 5b Reduced characteristic ultimate bolt steel shear capacity, OV5f (kN)
Not appropriate for this product.
Checkpoint Design reduced ultimate shear capacity,
OVur = minimum of OVurc, OVus
Check / OVur 1,
if not satisfied return to step 1
® Ramset
Concrete
20
STEP
® Ramset
Concrete
CHEMSETTM INJECTION 800 SERIES

Date of Issue: February 4, 2002 Page 1 of 6
*
PRODUCT Allows for close anchor spacing between
Chemset Injection 800 Series is a chemical anchor fixing parts.
*
system based on epoxy mortar. The two parts are Integrated system, no need to mix
*
dispensed and mixed in one action through a static Metered system reduces waste.
*
mixing nozzle, which allows accurate mixing with no For a partially used cartridge, simply change
mess. nozzle and re-use.
*
High strength.
*
DESCRIPTION Excellent chemical resistance.
*
Chemset 800 Series injection mortars are tough, Low shrinkage — allows for large anchors to
corrosion resistant products suitable for anchoring be installed.
*
jobs close to edges where there is a need to avoid Low odour.
*
bursting stress on the surrounding substrate. Non-flammable for transport and storage
The high strength and toughness of Chemset 800 PERFORMANCE PROPERTIES

Series mortars allows higher loads on anchored Typical properties after 7 days cure at 25°C and 50% RH
fixtures. This enables designs to be optimised by
providing a broader range of anchoring options. Appearance A: White thixotropic paste
B: Black thixotropic paste
Chemset 800 Series mortars also have exceptional Mixed colour appearance Mid grey thixotropic paste
corrosion resistance making them ideal to use in Viccosity (mixed material) Approx. 90,000 cps at 25°C
areas where there is frequent cleaning or where
Density (mixed material) Approx. 1.30 kf/litre
chemicals are frequently used.
Flammability Non flammable
Service temp. (cured) -10°C to +80°C
Chemset 800 Series consists of two products, each
Heat distortion temp. Approx. 80°C
tailored to different climates. Chemset 801 is
Hardness >80 Shore D
designed to provide rapid cure with adequate working
time in temperate climates. Chemset 802 has been Minimum installation Substrate temp. 5°C
developed with a slower gel time for areas where temperature Mortar temp. 18°C
there is a high ambient temperature or where a Recommended maximum Ambient temp. 43°C
slower mortar is required. installation temperature Mortar temp. 30°C
Water absorption ASTM <2% (10 days at 25°C)
RECOMMENDED USES D570
* Full cure 7 days at 25°C
Fixing to solid concrete, solid brick and
masonry. Modulus of elasticity >4 GPa
* Chemical composition Epoxy
Suited for structural fixings.
*
Holding down fixings. Tensile strength BS2782 > 25 MPa
Securing stainless steel fixings into corrosive Compressive strength > 75 MPa
environments or where a large degree of AS1O12.9
chemical resistance is required. AS2073 using 80mm
*
Suitable for fixing wall ties, starter bars, cubes
deformed rebar, studs, bolts and large Flexural Strength Approx. 22 MPa
screws. ASTM C348-86, BS 6319
*
Suitable vertical and horizontal applications.
*
Suitable for situations where anchor is
subjected to cyclic loading.
*
Suitable for use in areas subjected to
vibration and dynamic loading
*
Suitable for use in dry, diamond cored holes.
FEATURES AND BENEFITS

*
Stress free anchoring allows for close to
edge applications.
Technical Data Sheet

® Ramset
Concrete

PRECAUTIONS SOLID SUBSTRATES

o Do not install anchors o hen the Chemset mortar
temperature is less than 1. Drill hole using correctly rotary
o Do not install anchors o hen substrate temperature hammer drill bit to the specified
is less than depth. (see Chart No. 2 for threaded
o At temperatures beloo 801 studs refer to the
and 802 should be a armed or stored in temperatures specifications).
of for 24 hours prior to use to improve product
floo and cure.
o Avoid excessive pressure being applied to the
dispensing gun. Excessive pressure applied
to the dispensing gun may result in mortar loss
from the back of the cartridge. 2. Clean hole El ith correctly hole
o ose only the correctly mixed mortar a incorrect mix cleaning brush (see Chart No. 2) a ith
ratio a ill cause a reduction of the physical properties stiff nylon or a ire bristles. asing a
of the cured system. combination Push/Pull and ta isting
o Partly used cartridges must be used a ithin 5 a eeks (rotation-) motion, ensure the sides of
of initial use. the hole are scrubbed at least
El AID ays ensure hole is properly cleaned, dry and 3 times for the full depth of the hole.
no a ater present.
o Do not expose mortar to moisture or U ater for
at least 24 hours after installation to alloo for
maximum curing. 3. Remove debris, dust etc. from the
a Do not dilute mortar a ith any solvents and/or hole using a hole cleaning blon er
other chemicals. El ith at least 4 so ift pumps,
o Ala ays use Ramset other may alternatively a strong blast of

cause ineffective mixing and reduce the properties compressed air may be used.
of the mortar.
o Do not install into uncured concrete.
a Do not remove spiral mixer from
O Do not cut or shorten
o Not recommended for overhead applications
use Chemset 100.
INSTALLATION INSTRUCTIONS:
Read the aPrecautionso section of these instructions 4. Reinforcing bars, internally threaded sockets,
prior to use. - threaded rods or studs to be used should be cleaned
and free from oil, grease, flaking rust or debris.
Setting and technical data provided applies to: Threaded rods or studs should be chisel ended to
a Holes drilled o ith Rotary Hammers using drill prevent them unthreading from the cured mortar.
bits toleranced in accordance a ith Dd\18035,
and a here holes have been cleaned using a 5. Ensure that holes are dry. d holes have been left for
brush and air pump. a prolonged period since drilling, re-cleaning in
o Holes cored a ith diamond coring eauipment that accordance El ith 12 a 3oabove is recommended.
have been cleaned using a brush and air pump.
a AID ays El ear safety glasses or goggles.

_______
_____________
® Ramset
Concrete

6. asing only Ramset 11.ance the reauired fill is

mixing (other obtained release pressure
may cause ineffective mixing pressing rear trigger and
and reduce the properties of a ipe off excess material.
the mortar) remove nut from
the cartridge and attach a
mixing (screo doo n
tight).
12.Push the stud into the hole using a
7. Mount assembled cartridge slon to isting motion. Wipe an ay the
into the Ramset a niversal excess material.
Applicator.
I
8. open the orange valve.
(Turn so arroo points
fora ard). the handle 13.Do not touch the anchor until mixture
so the mortar is dispensed has gelled and do not load anchor
out of the until an until curing is complete as per Chart
even, uniform light grey No.1.
colour is achieved.
(Approximately 5-6 trigger pulls should be adeauate).
The initial mortar floc is unsuitable for fastening and
must not be used. This initial floa should be disposed
of into empty packaging or similar materials.
9. The mortar must be in01cted a ithout

creating air pockets. To achieve this 14.Attach fixture and tighten nut in
insert mixer to the base of the accordance a ith recommended
hole and In01ct from the base out tightening torQue (See Chart No 2)
10 SIoa ly a ithdraa the as the *1

hole fills a ith each ose an
extension tube for deep holes
Chart No. 3 lists the correct number
of trigger pulls to dispense 01st
enough resin for each hole/stud
Ramsew Fasteners (Aust) Pty aimited does not give any representation, guarantee or a arranty, express or implied that the information
is or a ill be complete or accurate or that it has been independently verified. To the extent permitted by statute, Ramseto Fasteners
(Aust) Pty aimited, its servants and its agents a ill not be liable (a hether in negligence or other tort, by contract or under statute) in
respect of any loss or damage (including consenuential loss or damage) arising by any reason of or in connection a ith the provision of
the information or by purported reliance on it.

® Ramset
Concrete

CHART 1 - SEllING TIMES
Gel lime Load lime

40 40
C) 0)
20
I-
C) -
10 CC
10
5-
.= 5=
= =
C#30 Cl)
0 10 20 30 40 50 60 70 80 90 100 110 120 5 10 15 20 25 30 35 40 45 50 55 60 65 70

Minutes Hours
C)
40
C) p'. 1 11111 1 11
- - =
—
ill
- CC c...
C) '?'.'' ———\- -8(U——- - -—
I-
20 E ————s — ————— — ——
C) — — — C) S
————— ————— — ——
—— —
5= — — — — — — — — — — — — —
=
Cl)O
0 10 2030 40 50 60 70 80 90100110120 0 5 10 15 20 2530 35 40 45 50 55 60 65 70
Minutes Hours
Note: For ambient temperature beloo the mortar must be conditioned to a minimum of See precautions, page 2.
ANCHOR STUDS
CHART 2 - SOLoD SuBSTRATE uNSTALLATnON DETAOLS
Anchor Thread Drill Hole Hole Maximum Maximum Tightening Minimum Minimum Minimum
Sine Sine Sine Cleaning Depth Fixture Fixture Tornue Edge Anchor Structural
(mm) Brush in Clearance Thickness (Nm) Distance Spacing Thickness
Sine Substrate n (mm) (mm) (mm) (mm)
(mm) (mm) (mm)
8 M8 10 13 80 11 15 8 30 50 100
10 M10 12 13 90 13 25 15 40 60 120
12 M12 14 13 110 15 30 30 50 70 140
16 M16 18 20 125 19 40 70 65 100 160
20 M20 24 20 150 25 75 140 80 120 190
24 M24 26 26 160 30 105 230 95 145 200

® Ramset
CHEMSETEI cNoECTdJN 800 SEREES
Date oncasuenoebruarn iE2002 aaoe a ono
CHART n aER
Anchor Nominal Nominal Number of Fixingsn Trigger Pulls per

Hole a Hole Depth Holea
(mm) (mm) 380m1 750m1 380m1 750m1
8 10 80 91 193 1 1
10 12 90 62 132 1 1
12 14 110 41 86 2 1.5
16 18 125 26 54 2 2
20 24 150 10 22 6 5
24 26 160 11 24 6 5
Note: aApproximately based on continuous installation a ithout interruptions or changes.

alrigger pulls using Chemseta nniversal Applicator (CaA). Provided as a guide and a ill vary a ith temperature.
CHART a a RECOMMENDED a LOADS a i1H CHEMSET ANCHOR STUDS

Anchor Thread Recommended Working Load (kN)
Sine Sine
2OMPa 3OMPa 4OMPa
Tensile Shear Tensile Shear Tensile Shear
8 M8 6.5 4.4 6.5 4.4 6.5 4.4
10 M10 8.6 7.1 9.7 7.1 10.3 7.1
12 M12 12.5 10.5 14.1 10.5 15.3 10.5
16 M16 16.9 19.8 19.2 19.8 20.8 19.8
20 M20 24.3 30.0 27.6 30.0 29.9 30.0
M20 29.3 30.0 33.3 30.0 36.1 30.0
24 M24 28.8 43.4 32.7 43.4 35.5 43.4
M24 43.3 43.4 49.2 43.4 53.4 43.4
For shear loads acting ton ards an edge(s) of the concrete the above capacities are not applicable.
Contact a Ramset technical consultant for advice.
Teoonioal Data Soeet

® Ramset
Concrete
CHEMSETO cNoECTcDN 800 SERi:ES

Date aaae a ona
CHART a On CHEMSET 800 EanOSED TO
Chemical! Conc. Resistant Not Chemical! Conc. Resistant Not

Linuid % Resistant Linuid % Resistant
Acetic Acid 10 a Lactic acid any a
Acetic Acid 30 a Machine oil 100 a
Acetone 25 a Methanol 10 a
Battery Acid a a Nitric acid conc. a
Beer a a Nitric acid 30 a
Butanol 100 a Nitric acida 10 0
Calcium chloride, an any a Petrol 100 a
Calcium hydroxide, an a a Petroleum 100 a
Carbon tetrachloride 20 a Phosphoric acid 30 a
Citric Acid any a Phosphoric acid 10 a
Common salt solution any a Rivera ater a a
Communial a aste a ate a a Sea a ater a a
Diesel fuel 100 a Sodium carbonate, an any a
Distilled a ater a a Sodium chloride, an any a
Engine oil 100 a Soduim hydroxide 20 a
Ethanol 40 a Sodium hydroxide 10 a
Formaldehyde, an 30 a Sodium silicate any a
Formic acid 40 a Sulphuric acid conc. a
Formic acid 10 a Sulphuric acid 20 a
Gasoline 100 a Sulphuric acid 10 a
Glycol 100 a Tapnater 0 a
Hydrochloric acid conc. a Toluene 100 a
Hydrochloric acid 10 a Turpentine a a
Hydrochloric acid 20 a Water a n
Hydrochloric acid 30 a Water glass any a
an n anueous solution (diluted) satusaturated %n% byneight
HEALTH a SAoETn For more detailed information refer to the Material

o Avoid contact a ith skin (R38) Safety Data Sheet No. 265]3 and 1056.
n Avoid contact a ith eyes (R36) Note: Numbers in brackets, example (R38) refer to
o Avoid breathing vapour (S23) (R20) the internationally risk phrases from the
o Wear protective gloves n hen mixing or using. aNational Code of Practice for the Labelling of
o d poisoning occurs, contact a doctor or Workplace Substances.n
Poisons mformation Centre.
o d so allon ed, do not induce vomiting. STORAGE
Give glass of a ater. Store in a cool, dry place beta een and
o d skin contact occurs, remove contaminated an ay from direct sunlight.
clothing and a ash skin thoroughly.
o d in eyes, hold eyes open, flood a ith o ater Kamseta Fasteners (Aust) Pty Limited does not give any
for at least 15 minutes and see a doctor. representation, guarantee or warranty, express or implied that
o Do not use in poorly ventilated or confined the information is or will be complete or accurate or that it has
space. (S38). been independently verified. To the extent permitted by statute,
Ramseta Fasteners (Aust) Pty Limited, its servants and its
agents will not be liable (whether in negligence or other tort,
by contract or under statute) in respect of any loss or damage
(including consequential loss or damage) arising
by any reason of or in connection with the provision of the
information or by purported reliance on it.
Teoonioal Data Soeet

Appendix E
APPENDIX E-3 FRP INFORMATION

Sika® CarboDur REVISED JULY 2002
Heavy-Duty CFRP Strengthening System
DESCRIPTION Sika CarboDur is a heavy-duty strengthening system for reinforced ôoncrete and tim-
ber. It consists of two components: Sikadur-30 adhesive for bonding reinforcement
and Sika CarboDur-laminates.
USES The Sika CarboDur System can be used to strengthen reinforced concrete and timber
structures due to:
— Loading increases.
— Damage to structural parts.
— Serviceability improvements.
— Change in structural system.
. — Design or construction defects.
ADVANTAGES • Low in weight.

• Available in any length.
• Low overall thickness.
• Easy to transport.
• No preparation of Sika CarboDur-laminates.
• Laminate intersections are simple.
• Economical application — no heavy handling and installation equipment.
• Very high tensile or flexural strengths can be achieved.
• High modulus of elasticity.
. • Outstanding fatigue resistance.
. • Can be coated without preparation.
•
• Alkali resistant.
STORAGE AND When not exposed to direct sunlight, Sika CarboDur-laminates have unlimited shelf
SHELF LIFE life. Sikadur-30, when stored in the original sealed containers w ithin t he temperature
range of +5°C to ÷25°C will ke ep for a minimum of three years.
INSTRUCTIONS Surface Preparation

FOR USE The concrete or timber surface must be clean and free from grease and oil, dry, and
have no loose particles or laitance.
This can be prepared by blast cleaning, scabbling or grinding. The concrete age should
be 3 to 6 weeks minimum, depending on thickness, curing conditions, etc.
The surface to be coated must be level, with steps and form work marks not greater
than 0.5 mm.
After cleaning remove all dust from the surface with an industrial vacuum cleaner.
Mixing
Sikadur-30 is supplied in factory proportioned units com-prising the correct quantities
of Part A (Resin) and Part B (Hardener). Thoroughly stir both components separately
using a slow running drill/stirrer with a helical paste mixer (max. speed 600rpm).
Decant all Part B into Part A and mix thoroughly together until a uniform colour is
achieved (typically 3 mins). A streaky colouration is indicative of inadequate or incom-
plete mixing. Apply immediately. Small units may be hand mixed provided an even
colour is achieved.
Application
If any necessary patching work needs to be done on the surface, this must be done
with Sikadur-41, on the day preceding the actual bonding operations.
Page 1 of 4
Apply the well mixed Sikadur-30 adhesive carefully to the prepared substrate with a
spatula to form a first layer of at least 1 mm. Place the Sika CarboDur-laminate on a
table and clean it with Sika® Colma Cleaner. Apply the Sikadur-30 adhesive across the
width of the laminate, ensuring a total coverage. The adhesive should be minimum 1
mm deep at each side, to minimum 2 mm deep at the centre. This is best done using
a plastic spatula shaped on site to achieve this profile.
Within the open time of the adhesive, depending on temperature, place the Sika
CarboDur-laminate onto the concrete surface. Using a roller, press the laminate into
the epoxy material until the adhesive is squeezed out on both sides of the laminate.
Remove surplus epoxy adhesive. Samples should be made up on site to check the
adhesive used in respect of curing rate and final strength. Measure the compressive
bending and adhesive strength after curing. As a final check, test the laminates for
drum miness by tapping lightly.
There is no need for mechanical equipment to press strips onto the substrate nor is it
necessary to provide clamps or supportive devices to keep overhead strips in place.
Once cured the top of the laminates can be painted with a coating material such as
Sikagard-62, Sikagard-670W or Sikagard-680S.
Cleaning
Clean tools immediately with Sika® Colma Cleaner. Wash hands and skin thoroughly in
warm soapy water. Cured material can only be removed mechanically.
TECHNICAL DATA A. SIKA CARBODUR-LAMINATES
Colour Black
Base Carbon fibre reinforced with an epoxy matrix
Apparent Density 1 .6g/cm3
Temperature Resistance Between 150°C and 500°C.
Sika CarboDur S Sika CarboDur M Sika CarboDur H

Elastic Modulus > 165,000 MPa > 210,000 MPa > 300,000 MPa
Teflsile Strength* > 2,800 MPa >2,400 MPa > 1,300 MPa
Mean Value of Tensile*

Strength at Break 3,050 MPa 2,900 MPa 1,450 MPa
Elongation at Break > 1.7% > 1.2% > 0.45%
Packaging Supplied in rolls of 250 m or palletised in pre-cut sections.
* Mechanical values obtained from longitudinal direction of fibres.
Availability Type Width Thickness Cross Sectional

mm mm Area mm2
Sika CarboDur S Sika CarboDurS512* 50 1.2 60

Sika CarboDur S612 60 . 1.2 72
Sika CarboDur S812* 80 1.2 96

SikaCarboDurSlOi2 100 1.2 120
SikaCarboDurSl2i2 120 1.2 144
Sika CarboDur 5614 60 1.4 84
Sika CarboDur S914 90 1.4 126
Sika CarboDur Si 214* 120 1.4 168
Sika CarboDur M SikaCarboDur M614 60 1.4 84

Sika CarboDur M914 90 1.4 126
Sika CarboDur M12i4* 120 1.4 168
Sika CarboDur H Sika CarboDur H514 50 1.4 70
(°Readily available, other types available on demand.)
- . . -.
- Sika® CárboDur
Page 2 of 4
________
B. SIKADUR-30 ADHESIVE FOR BONDING REINFORCEMENTS
Appearance Part A White paste

Part B Black paste
Part A + B Light grey when mixed
Mix Ratio A: B = 3: 1 (parts by weight and volume)
Density 1.77 kg/L (A + B)
Pot Life* 40 minutes (at 35°C)
Open Time* 30 minutes (at 35°C)
Sag FIow* 3-5 mm (at 35°C)

Shrinkage* 0.04% .
Glass Transition Point* 62°C
Static 12,800 MPa
Adhesive Strength (wet)* 4 MPa (Concrete Failure)
Shear Strength* 15 MPa (Concrete Failure)
Coefficient of Expansion 9x per °C (-10°C to 40°C)
Packaging 5 kg tins Part A 3.75 kg

Part B 1.25 kg
* To F.l.P. Fédération Internationale de a Précontrainte

Note: The values given may vary according to amount of air entrained during mixing.
Compressive strength Development of compressive strength

(DIN 1154.7) MPa 120
100
— +35C
80
60
40
20
L
Hours
0 50 100 150 200 250 300 350
Tensile strength Development of tensile strength

(DIN 43455) MPa 4C—
— *35C ——
". r_-- .----
I
ioL_______
Hours .0
0 0 150 200 250 300 350
Tensile slant shear Development of tensile slant shear strength

strength (Sika Test) MPa
20
::
;;
Hours
0 50 100 150 200 250 300 350
Sika® CarboDur
Page 3 of 4
DESIGN NOTES • A Sika CarboDur-laminate has no plastic deformation reserve.
• Maximum bending resistance of a strengthened section is reached when laminate failure occurs
during steel yield and before concrete failure.
• Mode of failure is influenced by the laminate cross section.
• To limit crack widths and deformation the yield point should not be reached in the reinforcing bars
under service conditions.
• Any shear cracks which occur must be prevented from causing displacement on the strengthened
surface and shearing of the laminate.
• Stress and deformation calculations can be made by the normal methods.
• When assessing the condition of the structure, look at the dimensions, quality of existing
construction materials, climatic conditions and agreed conditions of service.
• Design aspects to be verified in a Sika CarboDur Strengthening System:
Bearing safety — non strengthened structure (total safety factor g=1 .2)
— strengthened structure (mode of failure mentioned above, check on strains)
— shearing of laminate must not occur
— anchorages
Fatigue resistance — check concrete and steel stresses
Serviceability — deformation (with average strains, elasti behaviour of the structure and
time-based strain changes in concrete)
— steel stresses (no plastic deformation in service conditions)
— limit crack widths (by limiting the steel stresses to less than the yield point
under service conditions).
A full Sika Design Manual is available. Contact our Technical Department for details.
CONSUMPTION Laminate type Sikadur-30

RATES S512 / H514 .- 0.31 kg/rn
S612 / S614 / M614 -0.38kg/rn
S812 - 0.50 kg/rn
S914 / M914 —0.56kg/rn
S1012 .- 0.62 kg/rn
S1212/S1214/M1214 —0.74kg/rn
IMPORTANT NOTES • Do not apply Sikadur-30 to surfaces with standing water. Maximum moisture content of the con-
crete 10%.
• Always mix a full kit to avoid mix ratio error.
• Only mix as much material as can be applied within the stated potlife.
• Do not dilute the product with solvent as this will affect the cure and in-service performance.
• Constant exposure to service temperatures >50°C may affect the performance of the product.
• Sika CarboDur can be firerated if required using standard fire rating materials.
• The temperature at which the Sikadur-30 is stored at during the 24 hours before it is mixed will
govern its potlife when mixed.
.
• For shear strengthening, refer to SikaWrap technical data sheets.
• Compressive strengths, etc., of epoxy resins must be qualified by the testing method, eg. Test Standard
or size of specimen under test and the rate at which the test piece is loaded under test, as these
factors will affect the result markedly. Faster loading rates will generally give higher ultimate loads
and vice versa. Also, a specimen at lower temperature will show higher strengths and vice versa.
• Sikadur-30 Parts A and B are a water pollutant and should not be discharged into drains, waterways
or soil.
HANDLING • Avoid contact with the skin, eyes and avoid breathing its vapour.
PRECAUTIONS • Wear protective gloves when mixing or using.
• If poisoning occurs, contact a doctor or the Poisons Information Centre.
• If swallowed, do NOT induce vorniting. Give a glass of water.
• If skin contact occurs, remove contaminated clothing and wash skin thoroughly.
• If in eyes, hold eyes open, flood with water for at least 15 minutes and see a doctor.
IMPORTANT The information, and, in particular, the recommendations relating to the application and
NOTIFICATION end-use of Sika products, are given in good faith based on Sika's current knowledge
and experience of the products when properly stored, handled and applied under normal
conditions. In practice, the differences in materials, substrates and actual site conditions are
such that no warranty in respect of merchantability or of fitness for a particular purpose, nor
any liability arising out of any legal relationship whatsoever, can be inferred either from this
information, or from any written recommendations, or from any other advice offered. The
proprietary rights of third parties must be observed. All orders are accepted subject to our
current terms and conditions of sale. Users should always refer to the most recent issue of the
Technical Data Sheet for the product concerned, copies of which will be supplied on request.
PLEASE CONSULT OUR TECHNICAL DEPARTMENT FOR FURTHER INFORMATION.
Sika Australia Pty Limited 55 Elizabeth Street

ABN 12001 342 329 Wetherill Park NSW 2164
Tel: (02) 9725 1145
www.sika.com.au Fax: (02) 9725 2605
Sika® CarboDur
Page 4 of 4
Revised March 2004
Sikadur®-30
Thixotropic adhesive for bonding reinforcements
Description Sikadur-30 is a thixotropic adhesive mortar based on a 2-component solvent
free epoxy resin. Sikadur-30 is used primarily to bond structural
reinforcements to other substrates. It can also be used to bond and fill a wide
variety of building and construction materials.
Uses For structural bonding of:

-
• Sika CarboDur to concrete, brickwork, stone, timber and blockwork.
• Steel plates to concrete.
- Concreteelements.
- Bridge segments.
For bonding of:
• Starter bars.
• Wall anchors.
• Fixings.
T For vertical and overhead filling of:
• Holes
• Dimensional inaccuracies.
• Excellent green-strength — non sag in vertical and overhead applications.
Advantages
High creep resistance under permanent load.
• Good mechanical strength.
• Good chemical resistance.
• Low shrinkage during curing stage.
• Application at low temperatures.
- Excellent adhesion to most building material even when damp.
• High abrasion resistance.
• High early strength.
• High tensile and flexural strength.
• Supplied in factory proportioned units.
Storage and Shelf Life Minimum shelf life is approximately two (2) years. Store under controlled
conditions in original containers (minimum 5°C, maximum 35°C temperature
range).
Instructions for Use

Surface Preparation All surfaces must be clean, sound and free from dust, ice, oils, grease or
other surface contaminants such as form release residues and curing
membranes.
Concrete, Mortar, Stone: Mechanically abrade the surface with a needle gun,
water blast, grit blast or grind. All surface laitance must be removed. Cement
and concrete should be at least 3-4 weeks old (max. moisture content 4%).
Minimum adhesive tensile strength of the substrate 1 .5MPa.
Metals: Remove any paints, oils, grease, rust and oxide films by grit blasting.
Apply Sikadur-30 without delay.
Timber: Sandblast the surface or grind the timber, leaving a rough surface on
the substrate.
Epoxy: Abrade and rinse with Sika Colma Cleaner.
Mixing Sikadur-30 is supplied in factory proportioned units comprising the correct

quantities of Part A (Resin) and Part B (Hardener). Thoroughly stir both
components separately using a slow running drill/stirrer with a helical paste
mixer (max. speed 600 rpm). Decant all Part B into Part A and mix
Sikadur®-30
Page 1 of 4
— —...- -
—.——- -j
Mixing (continued) thoroughly until a uniform colour is achieved (typically 3 mins.). A streaky
colouration is indicative of inadequate or incomplete mixing. Apply
immediately. Small units may be hand mixed provided an even colour is
achieved.
Application When using Sikadur-30 as a structural reinforcement adhesive, apply well
mixed epoxy to the substrate, at maximum 1mm thickness, using a spatula,
trowel, or float. Adhesive should then be applied to the Sika CarboDur
laminate or steel plate in a roof shape (minimum 1mm thickness at the sides,
2mm in the centre).
Steel plates must be fixed to the concrete substrate using specially prepared
supports.
Sika CarboDur laminates are pressed into the substrate with hard rubber
rollers by hand. It is important that laminates are evenly pressed onto the
substrate until the adhesive is squeezed from both sides of the laminate.
Cleaning Uncured material may be cleaned from application tools, etc. by using Sika
Colma Cleaner (flammable solvent). Cured material can only be removed
mechanically.
Technical and Physical Data
Appearance Part A White paste
Part B Black paste
Part A + B Light grey when mixed
Mix Ratio A: B 3 : 1 (parts by weight & volume)
Density 1.77 kg/L (A + B)
Pot Life * 40 minutes (at 35°C)
Open Time* 30 minutes (at 35°C)
Sag Flow* 3 - 5mm (at 35°C)
Shrinkage 0.04%
Glass Transition Point* 62°C
Static 12,800 MPa
Adhesive Strength* Concrete: 4MPa (Concrete Failure)
Steel 33MPa (Sandblasted Substrate)
Shear Strength* 15 MPa (Concrete Failure)
Coefficient of Expansion 9x1 per °C (-10°C to 40°C)
Packaging 5 kg tins Part A 3.75 kg
Part B 1.25 kg
*10 F.I.P Federation Internationa le de Ia Precontrainte
Note: The values given may vary according to amount of air extrained during mixing.
Compressive strength Development of compressive strength

toe. ,,so.7)
i°0
00
.330
=
20
40.at 0
0 '50 200 250 3 0 3
Tensile strength Development of tensile strength

(044 43450) MPO 40
I
+35.C
3 50 1100 1 200 250
Tensile slant shear Development of tensile slant shear strength

strength Test)
T —
•33•C
,
0 50 100 1150 200 250 300 350
Sikadur®-30
Page 2 of 4
Important Notes • Do not apply Sikadur-30 to surfaces with standing water. Maximum
moisture content of the concrete 10%.
• Always mix a full kit to avoid mix ratio errors.
• Only mix as much material as can be applied within the stated potlife.
- It is important to entrain as little air as possible when mixing Sikadur-30.
Air entrainment will result in lower compressive strength development.
• Do not dilute the product with solvent as this will affect the cure and
service performance.
- Constant service temperature >50°C may affect the performance of this
product.
- Sikadur-30 can be heated up to 70°C during curing to speed up the
setting time.
If in doubt, consult our Technical Department.
• Compressive strengths etc. of epoxy resins must be qualified by the
testing method e.g. Test Standard or size of specimen under test and the
rate of which the test piece is loaded while under test, as these factors
will effect the result markedly. Faster loading rates will generally give
higher ultimate loads and vice versa. Also, a specimen at lower
temperature will show higher strengths and vice versa.
Handling Precautions • Avoid contact with the skin, eyes and avoid breathing it's vapour.
• Wear protective gloves when mixing or using.
• If poisoning occurs, contact a doctor or Poisons Information Centre.
• If swallowed, do NOT induce vomiting. Give a glass of water.
• If skin contact occurs, remove contaminated clothing and wash skin
.
thoroughly.
• If in eyes, hold eyes open, flood with water for at least 15 minutes and
see a doctor.
• For more detailed information refer to Material Safety Data Sheet.
Important Notification The information, and, in particular, the recommendations relating to the
application and end-use of Sika's products, are given in good faith based on
Sika's current knowledge and experience of the products when properly
stored, handled and applied under normal conditions. In practice, the
differences in materials, substrates and actual site conditions are such that no
warranty in respect of merchantability or of fitness for a particular purpose,
nor any liability arising out of any legal relationship whatsoever, can be
inferred either from this information, or from any written recommendations, or
from any other advice offered. The proprietary rights of third parties must be
observed. All orders are accepted subject of our terms and conditions of
sale. Users should always refer to the most recent issue of the Technical
Data Sheet for the product concerned, copies of which will be supplied on
request.
PLEASE CONSULT OUR TECHNICAL DEPARTMENT FOR FURTHER
INFORMATION.
Page 3 of 4
}
r --—
A U 0
®
Sika Australia Pty Limited
ABN 12 001 342 329
www.sika.com.au
55 Elizabeth Street
Wetherill Park NSW 2164
Tel: (02) 97251145
Fax: (02) 9725 2605
Sikadur®-30
Page 4 of 4

Reading PDF

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Reading PDF

Încărcat de

Drepturi de autor:

Formate disponibile

SEISMIC RETROFITTING TECHNIQUES FOR

EXISTING UNREINFORCED MASONRY

STATEMENT OF ORIGINAL AUTHORSHIP

Chuang, S.W. and Zhuge, Y. (2002). "State-of-the-art report on seismic retrofitting of

Chuang, S.W. and Zhuge, Y. (2003). "Seismic Retrofitting of Unreinforced Masonry

Chuang, S.W. (2003)."The development of seismic retrofitting techniques for existing

Chuang, S.W. , Y. and McBean P.C. (2004). "Seismic Retrofitting of

Chuang, S.W., Zhuge, Y. and McBean P.C. (2004). "Seismic Retrofitting of

Chuang, S.W., Zhuge, Y. and McBean P.C. (2004). "Seismic Retrofitting of

TITLE PAGE (i)

2. REVIEW OF THE RELATED LITERATURE

2.3.3.3 Repointing 2-10

3. BEHAVIOUR OF UNIREINFORCED MASONRY WALLS

5. OBSERVED BEHAVIOUR AND TEST RESULTS

5.5.2 FRP 5-18

6. NUMERICAL MODELLING STRATEGIES FOR UNREINFORECED

6.6 Masonry cracking models 6-34

7. A BASIC MECHANICAL MODEL FOR THE RETROFITTED WALLS

8. ANALYTICAL STUDY FOR RETROFITTED WALLS

8.3.2.2 Failure criteria for masonry 8-7

8.8 Comparison between the retrofitted walls by cable system

9. PERFORMANCE-BASED DESIGN FOR RETROFFITED WALLS

10. SUMMARY AND CONCLUSIONS

2.1 Conventional structure 2-4

5.3 Assumptions for rocking strength evaluation of a wall falling

5.11 Hysteretic behaviour of Wall-6 (Retrofitted wall with FRP) 5-14

5.41 First crack of wall 6 5-32

6.17 Masonry failure surfaces in plane stress 6-30

6.20 Discrete-cracking model 6-40

8.19 The mesh for the retrofitted wall model 8-18

4.1 Summary of masonry wall specimens 4-20

A area of net mortared/grouted section

C1 modification factor to relate expected maximum inelastic displacements to

C2 modification factor to represent the effects of pinched hysteretic shape, stiffness

C3 modification factor to represent increased displacements due to dynamic P-A

Cm effective mass factor to account for higher mode mass

numerical value for adjustment of period, T

DCR demand capacity ratio

[D failed material constitutive matrix

Eme expected values of elastic modulus for masonry in compression

factor to adjust spectral acceleration at one second for site class

G compressive fracture energy

Gme expected shear modulus of masonry

G mode I fracture energy

mode II fracture energy

H0 effective wall height

J force-delivery reduction factor

L length of wall or pier

SPT blow count in soil layer "i"

lower bound vertical compressive strength of existing URM walls

expected vertical axial compressive force per load combinations

CE expected gravity compressive force applied to a wall

CL lower bound of vertical compressive force from load combination, pounds

probabilities of exceed ance

P1 plasticity index of soil

mean return period

Q CE expected strength of the component or element

Q CE expected strength of the component or element at the deformation level under

Q CL lower-bound strength of a component or element at the deformation level under

QD dead load effect

QG action due to design gravity loads

QL effective live load effect

Q effective snow load effect