Widodo Thesis

A thesis
submitted in partial ..........,..........,..,......

of the requirement of the degree
of
Doctor of Philosophy in Civil En:glnleermg
at the
University of Canterbury
by
Widodo
-;:;:
DEPARTMENT OF CIVIL ENGINEERING

UNIVERSITY OF CANTERBURY
CHRISTCHURCH, NEW ZEALAND
February 1995
To my beloved. wife,
Ninik Sumll'tiningsih
and my sons,
Titan Danar Raharjo
Stevan Chondro Suryono
Abstract
In the past few decades, engineers have realized that an appropriate estimation of
energy dissipation on the structural system is one of the important roles in aseismic design of
building structures located in hazardous seismic areas H26 ,P26,V9. The inelastic load-deformation
behaviour of the structural members and vibration of the foundations on the flexible soil
medium are two important features of the energy dissipation. Soil-structure interaction is the
way to refine the existing common methods in structural analysis as it able to take into account
the soil-foundation energy dissipation from the structural system.
Study of the response of building structures supported by a soil medium using lumped
parameter methods have been carried out by some researchers. However, most of these studies
used unrealistic physical or structural responses and soil-foundation models which did not have
real application in modern building aseismic design. The current New Zealand code NZS
4203: 1992 states that a special study should be carried out where energy dissipation in the
structural system is primarily through the rocking of foundations.
Analytical investigations using the same methods in more realistic applications of

aseismic design in building structures was carried out. The investigations cover several inelastic
damage indicators for both frames and fram~-wall structures with a different numbers of
storeys, numbers of frames, hysteresis rules, rigid joint lengths, supported by different-
foundation types, soil models, soil stiffnesses and exited by different earthquake inputs. It was
found that rocking structures exhibit advantages or disadvantages and show the inadequacy of
the current wall moment design envelopes for frame-wall structures.
Methods to overcome the disadvantages need to be developed. A new simple wall

moment design envelope for different types of foundation and soil-foundation hysteresis rules
has been proposed. In addition, the minimum required static bearing capacity factor for soil-
under minimum wall gravity load and the approximation of the building'S lateral fixity are also
discussed.
ii
Acknowledgements
This research work on the Rocking of Multistorey Buildings reported in this thesis, was
carried out in the Department of Civil Engineering at the University of Canterbury, under the
overall guidance of Professor Robert Park and Dr. Nigel Cooke, the former and the present
Heads of the Department.
I wish to express my deepest gratitude to my supervisor, Dr. Athol James Carr for his
enduring patience, invaluable evaluation, advice and encouragement. His warm communication
due to his enthusiasm and cheerful heart is greatly appreciated; a great experience.
I would like to thank to Emeritus Professor Thomas Paulay for his private and special
lecture in Capacity Design Philosophy of Frame and Wall-Frame Structures, Dr. John Berrill
for his assistance in Soil Mechanics, Dr. Peter J. Moss and Dr. Jose I Restrepo-Posada for
their evaluations and suggestions.
The Fees Scholarship provided by the Ministry of External Relations and Trade
(MERT), New Zealand is greatly appreciated. I wish also to deeply thank the Universitas Islam
Indonesia, Y ogya:karta, Indonesia for granting me this opportunity for this study and fully
funding the living costs. To Kopertis Wilayah V, Yogyakarta for his great assistance in
employee affairs during my study in the Department of Civil Engineering, University of
Canterbury, New Zealand, is also acknowledged.
My thanks are also extended to the Technical Staff of the Department, especially for
Mrs. V.I. Grey for her drawing of figures, Mr. B. Hutchison, P. Coursey for their assistance
in computer facilities and Mr. G.E. Hill for his assistance in use of general Department
facilities. The friendly service of the Engineering Library Staff is also appreciated. Thanks also
are due to fellow students for their support and friendship especially Dr. Li Bing, Dr. A. Mori,
Tabuchi Masamichi, Li Xin Rang, S. Turner, W. Munro, Iman Satyarno, Lystiadi and all
fellow Indonesian Students.
Finally, I wish to express my deep gratitude to my parents for their prayers, to my

sister and brothers for their support, to my beloved wife Ninik S. for her special deep feeling
and moral SUppOlt and to my sons Titan D.R and S.Chondro S., who have provided much
happiness.
iii
TABLE OF CONTENTS
ABSTRACT . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ACKNOWLEDGEMENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
TABLE OF CONTENTS .................................. iii
NOTATIONS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. xiv
Chapter I
Introduction
1.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Objectives of the Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

1.3 Scope and Outline of the Thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Chapter n
Review of the Usage of Lumped Parameter Methods in Rocking of Building
Structures
2. 1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 Experimental Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2.1 Clough and Huckelbridge (1977) . . . . . . . . . . . . . . . . . . . . . . . 7
2.2.2 Priestley, Evison, Carr (1978) . . . . . . . . . . . . . . . . . . . . . . . . 9

2.2.3 Huckelbridge and Clough (1978) . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Analytical Investigations

2.3.1 Chopra and Gutierrez (1974) . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3.2 Yim and Chopra . . . . . . . . . . . . . . . . . . . . ". . . . . . . . . . . . 19
2.3.2.1 Flexible Elastic SDOF Structure on the Winkler Foundation .. 19
2.3.2.2 Flexible Elastic SDOF Structure on a Two Spring-Dashpot

System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3.2.3 Flexible Elastic MDOF Structure on a Two Spring-Dashpot

System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.3.3 Psycharis (1982) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.3.4 EI-Hifnawy and Novak (1986) . . . . . . . . . . . . . . . . . . . . . . . . 33
2.4 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

iv
Chapter III
Conceptual Overall Design, Preliminary Analysis, Design and Evaluation
3.1 Introduction.. . . . . . . . . . . . . . . . . . . . . . 38
3.2 Building Overall Conceptual Design (BO-CD)
3.2.1 Design of Building Configuration . . . . . . . . . . . . . . . . . . . . 39
3.2.2 Structural materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.2.3 Non-structural Elements. . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.2.4 Building Prototypes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.2.5 Structural Systems . . . . . . . . . . . . . . . 43
3.3 Preliminary Analysis
3.3.1 Building Gravity Loads. . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.3.1.1 Dead Load (DL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.3.1.2 Live Load (LL). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.3.2 Lateral Seismic Load (E) . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.3.3 Reduction of Live Loads and Tributary Area. . . . . . . . . . . . . . . 49
3.3.4 Member Forces and Capacity Design Procedures . . . . . . . . . . . . 50
3.3.4.1 Effective Member Flexural Regidity (EIe). . . . . . . . . . . . . . 50
3.3.4.2 Beam and Column Forces. . . . . . . . . . . . . . . . . . . . . . . 51
3.3.4.3 Moment Redistribution. . . . . . . . . . . . . . . . . . . . . . . . . 51
3.3.4.4 Strength Definition and Their Inter-relationships. . . . . . . 52
3.4 Preliminary Sizing and Reinforcement
3.4.1 Design of Beam . . . . . . . . . . . . . 52
3.4.1.1 Effective Flange Width of T Beam under Positive Moment. . . 52
3.4.1.2 Effective Flange Width of T Beam Under Negative Moment. . 53
3.4.2 Design of Columns. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3.4.3 Design of Stmctural Walls. . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.4.3.1 Shape and Section of Stmctural Walls. . . . . . . . . . . . . . . . 55
3.4.3.2 Ductility Relationship in Structural Walls . . . . . . . . . ; . . . . 55
3.4.3.3 Stability of Structural Walls. . . . . . . . . . . . . . . . . . . . . . 57
3.5 Design Evaluation
3.5.1 Moment Redistribution of Beams . . . . . . . . . . . 58
3.5.2 Tension Reinforcement Ratio. . . . . .. . . . . . . . . . . . . . . . . . . 58
v
3.5.3 Design of Structural Walls. . . . . . . . . . . . . . . . . . . . . . . . . . 59

3.6 Summaty and Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Chapter
Dynamic Properties of Soils, SOlii Shear Strains and Rheological Model of
Soils
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 64
4.2 Dynamic Properties of Soils
4.2.1 Shear Modulus of Soils. . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.2.1.1 Effect of Soil Shear Strain . . . . . . . . . . . . . . . . . . . . .. 69
4.2.1.2 Effect of Void Ratio. . . . . . . . . . . . . . . . . . . . . . . . . . 69
4.2.1.3 Effect of Confining Pressure. . . . . . . . . . . . . . . . . . . . . 71
4.2.1.4 Effect of Soil Plasticity Index . . . . . . . . . . . . . . . . . . . . 71
4.2.1.5 Effect of Cyclic Loadings. . . . . . . . . . . . . . . . . . . . . . . 73
4.2.2 Material Damping of Soils. . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.2.3 Poisson's Ratio of Soils . . . . . . . . . . . . . . . . . . . . . . . . . . " 76
4.2.4 Shear Wave Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 76
4.3 Shear Strain Induced by Earthquakes. . . . . . . . . . . . . . . . . . . . . . . . 81
4.4 Threshold Shear Strain and Cyclic Pore Water Pressure . . . . . . . . . . . . 85
4.5 Simple Rheological Model Of Soil
4.5.1 Elastic State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ' 87
4.5.2 Viscous State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 88
4.5.3 Coulomb State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 88
4.5.4 Elasto-Plastic. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
4.5.5 Viscoelastic State. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.5.5.1 Simple Composite Models. . . . . . . . . . . . . . . . . . . . . . 89
4.5.5.1.1 Voigt-Kelvin Model. . . . . . . . . . . . . . . . . . . . . . 89
4.5.5.1.2 Maxwell Model . . . . . . . . . . . . . . . . . . . . . . . .. 91
4.5.5.2 Complex Composite Models . . . . . . . . . . . . . . . . . . . . 92
4.5.5.2.1 Standard Linear Model. . . . . . . . . . . . . . . . . . . . 92
4.5.5.2.2 Burger ModeL . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.6 Summary and Conclusions . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . 96
vi
Chapter V
Substructure Modelling
5.1 Introduction........... . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
5.2 Substructure General Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . 98
5.2.1 Soil Water Content and Water Table . . . . . . . . . . . . . . . . . . . . 98
5.2.2 Isotropic Material. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
5.2.3 Backfill, Soil Added and Foundation Mass Effect . . . . . . . . . . . . 100
5.2.4 Foundation Footing Flexibility . . . . . . . . . . . . . . . . . . . . . . . . 101
5.2.5 Coupling Between Rocking and Horizontal Vibration Modes . . . . . 101
5.2.6 Coupling Effect Between Adjecent Footings . . . . . . . . . . . . . . . 101
5.3 Soil Modelling
5.3.1 Multiple Layers over An Elastic Half-Space 104
5.3.2 Layers over Rigid Boundary . . . . . . . . . . . . . . . . . . . . . . . . . 105
5.3.3 Layer over Elastic Half-Space . . . . . . . . . . . . . . . . . . . . . . . . 105
5.3.4 Elastic Half-Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
5.3.5 Fictitious Layer over Elastic Half-Space . . . . . . . . . . . . . . . . . . 109
5.4 Modelling of the Foundation Footing
5.4.1 Modelling of Footing Shapes . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.4.1.1 Circular Foundation Footing . . . . . . . . . . . . . . . . . . . . . . 110
5.4.1.2 Rectangular Foundation Footing . . . . . . . . . . . . . . . . . . . . 110
5.4.1.3 Arbitrary Shape Foundation Footing . . . . . . . . . . . . . . . . . 112
5.4.2 Modelling of Footing Vertical Embedment
5.4.2.1 Surface Foundation . . . . . . . . . . . . . . . . . . . . . . . . . . . , 112
5.4.2.2 Fully Embedded Foundation. . . . . . . . . . . . . . . . . . . . .. 112
5.4.2.3 Partially Embedded Foundation. . . . . . . . . . . . . . . . . . .. 113
5.5 The Soil Foundation Model Used. . . . . . . . . . . . . . . . . . . . . . . . .. 113
5.6 The Soil-Foundation Mechanical Model. . . . . . . . . . . . . . . . . . . . .. 114
5.7 Mathematical Representation of Soil-Foundation Model . . . . . . . . . . .. 116
5.8 Modelling of Soil-Foundation damping
5.8.1 Modelling of Soil Material Damping. . . . . . . . . . . . . . . . . . . . 118
5.8.2 Modelling of Soil-Foundation Radiation Damping. . . . . . . . . . . . 118
5.9 The Substructure Parameter Quantities Used in this Study. . . . . . . . . . . 119
5.10 Summary and Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
vii
Ch.apter VI
Superstructure Modelling
6.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 122

6.2 Identification of Problem in Inelastic Structural Analysis . . . . . . . . . . . 122
6.3 General Assumption in the Analysis of the Superstructure
6.3.1 The Effect of Non-Structural Components. . . . . . . . . . . . . . . .. 123
6.3.2 Floor Diaphragms and Location of Structural Walls. . . . . . . . . .. 125
6.3.3 Torsion Effect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 126
6.3.4 Earthquake Orientation. . . . . . . . . . . . . . . . . . . . . . . . . . .. 127
6.4 Modelling Analysis of the Structural System. . . . . . . . . . . . . . . . . . . 127
6.4.1 Modelling of In-plane Wall-Frame Interaction. . . . . . . . . . . . .. 127
6.4.2 Modelling of Out-of-plane Frames and Wall-Frame Interactions. . .. 128
6.5 Inelastic Modelling of Reinforced Concrete Members .. 128
6.5.1 Modelling of the Member Element 130
6.5.1.1 Two-component Model . . . . . . . 130
6.5.1.2 Multi-component Model . . . . . . . . . . . . . . . . . . . . . .. 131
6.5.1.3 One-component. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 131
6.5.1.4 Discrete Element Model . . . . . . . . . . . . . . . . . . .. . .. 132
6.5.1.5 Distributed Flexibility Model. . . . . . . . . . . . . . . . . . . .. 132
6.5.2 Hysteresis Modelling of RC Frame Member. . 133
6.5.3 Modelling of Member Flexural Rigidity
6.5.3.1 Reinforced Concrete Frame Member. .. .......... 137
6.5.3.2 Cantilever Structural Wall. . . . . . . . . . . . . . . . . . . . .. 139
6.6 Member Element and Hysteresis Model Used in this Study. . . . . . . . . .. 140
6.7 Modelling of Rigid Zone and Plastic Hinge Lengths. . . . . . . . . . . . . .. 141
6.8 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 143
Ch.apter VII
Soil Foundation Responses and Impedances
7.1 Introduction ... . . . . . . . . . . . . . 144

7.2 Response of Soil-Foundation Systems
7.2.1 Static Response of Soil-Foundation System. 144
viii
7.2.2 Dynamic Response of Soil-Foundation System

7.2. 2.1 Winkler-Voigt Soil-Foundation Model . . . . 145
7.2.2.2 Elastic Half-Space Soil-Foundation Model 146
7.2.2.3 Lumped Parameter Soil-Foundation Model 149
7.3 Stiffness and Damping of Surface Foundation
7.3.1 Static Stiffness and Damping Coefficient. . . . .. ........... 151
7.3.2 Dynamic Siffness and Damping Coefficient. . . . . . . . . . . . . 152
7.4 Stiffness and Damping of Embedded Foundation
7.4.1 Frequency Independent Stiffness and Damping Coefficient. . . . 157
7.4.2 Frequency Dependent Stiffness and Damping Coefficent . . . . . . . . 158
7.5 Stiffness and Damping Coefficient of Arbotrary Shapes and Partially
Embedded Foundation
7.5.1 Vertical Stiffness and Damping Coefficient 165
7.5.1.1 Vertical Static Stiffness. . . . . . . . . . . . . . . . . . . . . . . 165
7.5.1.2 Vertical Dynamic Stiffness. . . . . . . . . . . . . . . . . . . . . .. 168
7.5.1.3 Vertical Damping Coefficient. . . . . . . . . . . . . . . . . . . . . 169
7.5.2 Horizontal Stiffness and Damping Coefficient 170
7.5.2.1 Horizontal Static Stiffness . . . . . . . . . . . . . . . . 170
7.5.2.2 Horizontal Dynamic Stiffness . . . . . . . . . . . . . . . . . . . . . 172
7.5.2.3 Hirizontal Damping Coefficient . . . . . . 174
7.5.3 Rocking Stiffness and Damping Coefficient 176
7.5.3.1 Rocking Static Stiffness . . . . . . . . . . . . . . . 177
7.5.3.2 Rocking Dynamic Stiffness . . . . . . . . . . . . . . . . . . . . . . . 178
7.5.3.3 Rocking Damping Coefficient . . . . . . . . . . . . . . . . . . . . . 178
7.6 Effect of Material Damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
7.7 Effect of Soil and Foundation Mass . . . . . . . . . . . . . . . . . . . . . . . . 181
7.8 Machine Foundation and Earthquake Engineering . . . . . . . . . . . . . . . . 182
Chapel' VIII
Analysis Procedures
8.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 185

8.2 The Methods of Analysis of Soil-Stmcture Interaction . . . . . . . . . . . " 185
ix
8.2.1 Lumped Parameter Method. . . . . . . . . . . . . . . . . . . . . . . . .. 185

8.2.1 Finite Element Methods
8.2.1.1 Direct Finite Element Method. . . . . . . . . . . . . . . . . . .. 188
8.2.1.2 Finite Element with Artificial Boundaries. . . . . . . . . . . . . 189
8.3 The Method of Analysis Used in This Study. . . . . . . . . . . . . . . . . . . 192
8.4 The Differential Equations of Motion of the Structural System. . . . . . . . 193
8.5 Overall Damping of Soil-Structure Interaction Systems. . . . . . . . . . . . . 198
8.5.1 Damping in Linear Elastic Analysis. . . . . . . . . . . . . . . . . . . . 198
8.5.2 Damping in Elastic Response Analysis . . . . . . . . . . . . . . . . .. 201
8.5.3 Effects of Foundation Flexibility on the Damping Quantity. . . . . . 202
8.6 Choice of the Numerical Integration Scheme. . . . . . . . . . . . . . . . . .. 203
8.7 Choice of the Earthquake Input. . . . . . . . . . . . . . . . . . . . . . . . . .. 204
8.8 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 205
Chapter IX
The Seismic Response of the 12 and IS-storey Fl'ames and Frame-Wall
Structures Subjected to the 1940 Et Centro NSC Earthquake
9.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 207

9.2 The response Quantities of Inelastic Structures . . . . . . . . . . . . 207
9.2.1 Single Maximum Value of the Damage Parameters
9.2.1.1 Ductility Factor . . . . . . . . . . '.' . . . . . . . 208
9.2.1.2 Member Forces, Nodal Displacements and Inter-storey Drifts. 208
9.2.2 The Accumulation of Member Deformations and Energy Dissipation
9.2.2.1 The Accumulation of Member Deformations. . . . . . . . . .. 209
9.2.2.2 The Accumulation of Energy Dissipation. . . . . . . . . . . . . 210
9.3 The Member Damage Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
9.3.1 The Deformation Model. . . . . . . . . . . . . . . . . . . . . . . . . .. 213
9.3.2 The Low-Cycle Fatigue Model. . . . . . . . . . . . . . . . . . . . . .. 213
9.3:3 The Combination Model. . . . . . . . . . . . . . . . . . . . . . . . . .. 214
9.4 The Storey and Structural Damage Indices. . . . . . . . . . . . . . . . . . . .. 215
9.5 The Seismic Response of Frame and Frame-Wall Structures on Fixed Bases
and Flexible Foundations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 217
9.5.1 Soil Shear Strains. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 217
x
9.5.2 The Base Node Rotations and Vertical Displacement of Footing Tips 219
9.5.3 Fundamental Period of Structures . . . . . . . . . . . . . . . . . . . . " 222
9.5.4 The Damage Para metes
9.5.4.1 Maximum Roof Horizontal Displacement . . . . . . . . . . . . . . 224
9.5.4.2 Base Shear and Storey Shear Ratios . . . . . . . . . . . . . . . . . 226
9.5.4.3 Beam and Column Moment Ratios and Wall Moments . . . . . . 230
9.5.4.4 Inter-storey Drifts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
9.5.4.5 Beam Cmvature and Plastic Hinge Rotations . . . . . . . . . . . . 234
9.5.4.6 Plastic Hinge Rotation of Column and Wall Bases . . . . . . . . . 237
9.5.4.7 Beam Cmvature Ductility Demands . . . . . . 238
9.5.4.8 Accumulation of Plastic Hinge Rotations ... 238
9.5.4.9 Viscous and Hysteretic Energy Dissipation . . . . . . . . . . . . . 241
9.5.5 Member, Storey and Structural Damage Indices . . . . . . . . . . . . . 243
Chapter X
Effects of Rigid Joint Lengths, Number of Frames and Hysteresis Rules on
the Response of the 12 and 1S-storey Frames and Frame-Wall Structures
10.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . " 250

10.2 Effect of the Rigid Joint Lengths. . . . . . . . . . . . . . . . . . . . . . . .. 251
10.2.1 Fundamental Period an Roof Horizontal Displacement. . . . . . . . 251
10.2.2 Base Shear and Momber Forces. . . . . . . . . . . . . . . . . . . . . . 254
10.2.3 Inter-storey Drifts and Plastic Hinge Rotations. . . . . . . . . 254
10.2.4 Column, Wall Plastic Hinge Rotation and Beam Cmvature
Ductility Demands . . . . . . . . . . . . . . . . . . . . . . . . . 258
10.2.5 The Accumulation of Plastic Hinge Rotation and Energy
Dissipation .. . . . . . . . . . . . . . . . . . . . . . . . . . . 259
10.2.6 Member, Storey and Structural Damage Indices. . . . . 260
10.3 Effect of Number of Frames in Frame-Wall Structures. . . . . 264
10.3.1 Fundamental Period and Roof Horizontal Displacement. . . . . . . 265
10.3.2 Base Shear and Member Forces. . . . . . . . . . . . . . . . . . . . .. 268
10.3.3 Plastic Hinge Rotations and Beam Curvature Ductility Demands. . 272
xi
10.3.4 Total Plastic Hinge Rotations and Energy Dissipation. . . . . . . .. 273

10.3.5 Member, Storey and Structural Damage Indices. . . . . . . . . . .. 274
lOA Effects of Different Hysteresis Rules in Flexure . . . . . . . . . . . . . . " 276
10.4.1 Base Shear. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 277
10.4.2 Member Forces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 277
10.4.3 Inter-storey Drifts and Plastic Hinge Rotations. . . . . . . . . . . .. 283
10.4.4 Base Plastic Hinge Rotations and Beam Curvature Ductility
Demands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . " 283
10.4.5 Total Plastic Hinge Rotation and Energy Dissipation. . . . . . . .. 288
10.4.6 Member, Storey and Structural Damage Indices. . . . . . . . . . .. 289
10.5 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . " 293
Chapter XI
Effects of Soil Shear Strains, Shear Moduli and Soil Models on the Response
of the 12 and IS-storey Frames and Frame-WaIL Structures
11.1 Introduction.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 296

11.2 Effects of Soil Shear Strains and Soil Moduli . . . . . . . . . . . . . . . . .. 297
11.2.1 Fundamental Period and Roof Horizontal Displacement. . . . . " 297
11. 2. 2 Base Shears and Member Forces . . . . . . . . . . . . . . . . . . . " 297
11.2.3 Inter-storey Drifts and Plastic Hinge Rotations. . . . . . . . . . . .. 300
11.2.4 Column, Wall Base Plastic Hinge Rotations and Beam Curvature
Ductility Demands . . . . . . . . . . . . . . . . . . . . . . . " . . . . .. 301
11.2.5 Total Plastic Hinge Rotations, Energy Dissipation and Damage
Indices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 305
11.3 Effects of Winkler Soil-Foundation and Layer over Rigid Boundary
Soil Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 309
. 11.3.1 Winkler Foundation . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 309
11.3.2 Soil Layer over a Rigid Boundary . . . . . . . . . . . . . . . . . . .. 315
11.4 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 319
Xll
Chapter XII
Effects of Earthquake Excitations on the Response of the 12 and IS-storey
Frames and Frame- Wall Structures
12.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 321

12.2 Roof Horizontal Displacement and Base Shear. . . . . . . . . . . . . . . .. 323
12.3 Wall Moments, Inter-storey Drifts and Plastic Hinge Rotations. . . . . . . 327
12.4 Column, Wall Base Rotations and Beam Curvature Ductility Demands. . 329
12.5 Total Plastic Hinge Rotations and Energy Dissipation. . . . . . . . . . . . 333
12.6 Member, Storey and Structural Damage Indices. 333
12.7 Overall Response Quantities of Structures
12.7.1 Base Shear. . . . . . . . . . . . . . . . 341
12.7.2 Plastic Hinge Rotation of Column and Wall Bases. . . . . . . . . . . 341
12.7.3 Vertical Displacement of Wall Footing Tips. . . . . . . . . . . . . . 344
12.7.4 Fundamental Period of Structu res. . . . . . . . . . . . . . . . . . . .. 344
12.7.5 Coefficient of Wm 347
12.7.6 Overall Drifts and Plastic Hinge Rotation Relationship . 348
12.7.7 Overal Drift and Structural Damage I ndex Relationship . 349
12.8 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . 350
Chapter XIII
The Analysis and Design Recomendations for Rocking Structures
13.1 Introduction . . . . . . . . . . . . . . . . 352

13.2 The Building's Lateral Point of Fixity 353
13.2.1 Frame Structures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 353
13.2.2 Frame-Wall Structures . . . . . . . . . . . . . . . . . . . . . : . . . .. 358
13.3 The Wall's Design Moment Envelopes of Frame-Wall Structures. . . . . . 363
13.4 The Hysteresis Rules of Soil-Foundation Impedances. . . . . . . . . . . . . 369
13.5 Static Bearing Capacity Safety Factor of Soil. . . . . . . . . . . . . . . . .. 374
13.6 Soil-Foundation Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 379
13.7 SummalY and Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 380
xiii
Chapter XIV
Conclusions and Recommendation for Further' Research
14.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 381

14.2 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 381
14.3 Recommendation for Fmther Research. . . . . . . . . . . . . . . . . . . . .. 385
14.3.1 Laboratory Work. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 386
14.3.2 Analytical Work .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 386
Rererences . 388
Appendices
xiv
Notations
ao = dimensionless frequency
b == one-half of footing foundation length
be critical wall thickness
beff = effective slab width of T -beam
bw = thickness of structural waH
bWb required wall thickness
c = damping coefficient of SDOF system
c = cohesion of soil
Cbb = horizontal damping coefficient of pile-cap foundation system
cb!f = horizontal-rocking damping coefficient of pile-cap foundation
c's = dimensionless dynamic damping coefficient
Cu = compression zone length of structural wall
c. frequency, shape and embedment s ide coefficient of horizontal damping
Cae = frequency, shape and embedment front coefficient of horizontal damping
Crx dynamic damping coefficient of sidewalls induced by rocking with respect to x-axis
Cry = dynamic damping coefficient of sidewalls induced by rocking with respect to y-axis
Corx = dynamic damping coefficient of base mat induced by rocking with respect to x-axis
Cory = dynamic damping coefficient of base mat induced by rocking with respect to y-axis
c. f = damping coefficient of soil-foundation
c.r damping coefficient of dynamic vertical sub grade reaction
Cw) damping coefficient per unit length of footing foundation
Cz = frequency and shape dependent coefficient of vertical damping
ex = frequency and shape dependent coefficient of horizontal damping in x-dir.
(\ = frequency and shape dependent coefficient of horizontal damping in y-dir.
d = breadth contact of soil-footing foundation
d = effective contact area between soil and sidewalls
dm = depth of concrete member
db diameter of longitudinal bar
d. r coefficient of dynamic vertical subgrade reaction
e = void ratio of soil
e = eccentricity
f = frequency
xv
f = coefficient for second moment of inertia of flanged section

fc == specified compressive strength of concrete
1;, = specified yield strength of steel reinforcement
fm = frequency of the model
f" == frequency of the prototype
f,. rocking frequency
fhf2 = displacement function
g = acceleration of gravity
h = height of SnaF structure
~ = height of the centre mass of superstructure
~ = height of storey j
hw height of structural wall
hI = height of first storey
h* equivalent height as SnaF structure
= mode index
layered soil index
= impact force index
j = floor number index
k stiffness of SnaF system
k coefficient of soil plasticity index
khh = horizontal stiffness of pile-cap foundation system
khl/- horizontal-rocking stiffness of pile-cap foundation system
k,..;,emb = dynamic embedment coefficient for rocking stiffness with respect to x-axis
kry.emb = dynamic embedment coefficient for rocking stiffness with respect to y-axis
k,..;,sur = dynamic surface coefficient for rocking stiffness with respect to x-axis
~,sur = dynamic surface coefficient for rocking stiffness with respect to y-axis
k's dimensionless dynamic stiffness coefficient
= spring stiffness of soil-foundation
k,~ coefficient of static vertical subgrade reaction
k",emb = dynamic horizontal stiffness coefficient in x-direction
Js.,emb = dynamic horizontal stiffness coefficient in y-direction
= vertical stiffness dynamic coefficient
= vertical stiffness dynamic coefficient of embedded foundation
:= vertical stiffness dynamic coefficient of trench foundation
= vertical static stiffness per unit length of footing foundation
xvi
lw = length of structural wall

m = mass
m mass of SDOF system
m = coefficient of soil confining pressure
fib = mass of base-mat
~ mass of storey j
mo mass of footing foundation
In" modulus of vertical subgrade reaction

fit total mass of structure
m* equivalent mass as SDOF structure
p = normal pressure
q = coefficient in soil over-consolidated ratio
r = radius of footing
rb = base nodal points at the soil interface
re = equivalent radius of rectangular footing foundation
rk = kinetic energy reduction factor
ro radius of footing
r. = vertical side-wall reaction per unit length of embedded foundation
r, = structural nodal points at the base
r" = rigid joint zone reduction factor

s distance between two adjacent footing foundations
= time
= soil degradation parameter
u = horizontal displacement
Ub horizontal displacement of base footing
uj = horizontal displacement of storey j
Uo = horizontal displacement of footing with respect to reference point
ug = free field horizontal displacement
u, = complex frequency displacement response of superstructure
uof complex frequency displacement response of footing foundation
iig = horizontal ground acceleration
Umax = maximum horizontal displacement
y = displacement
Ym horizontal rocking displacement at the centre of mass of model

yp horizontal rocking displacement at the centre of mass of prototype
xvii
displacement of dashpot element

= displacement of spring element
= yield displacement
= vertical displacement
= ground vertical displacement
= water content of soil
Ab base area of footing foundation
Ar = tributary area
Ag gross section area
A, soil-sidewall contact area of embedded foundation
Aw = effective soil-sidewall contact area of embedded foundation CAw < A.)
Awb = required area of wall boundary element
Awg - gross section area of wall
Aw. = soil contact area of side-wall which is parallel with horizontal motion
Awoe = soil contact area of side-wall which is perpendicular with horizontal motion
B - one-half of footing width
Bb = width of building plan
BSu = ultimate base shear
BSCu = ultimate base shear coefficient
BS c = critical base shear
BSC~ - critical base shear coefficient
BSCm = base shear coefficient corresponds to maximum horizontal displacement
C = damping matrix of MDOF system
C - basic seismic coefficient
= rocking damping coeff. induced by compression-extension waves at base footing
= horizontal damping coefficient generated by base shear wave in x-direction
Cby horizontal damping coefficient generated by base shear wave in y-direction
Cd,i = dynamic damping coefficient corresponds to vibration mode i

Ceq _. equivalent damping coefficient of soil-foundation system
Cg = a constant
Cbb = horizontal damping coefficient of soil-foundation system
C hr = horizontal-rocking damping coefficient of soil-foundation system
Crr = rocking damping coefficient of soil-foundation system
Ccx,emb = total damping coefficient of rocking embedded foundation with respect to x-axis
rocking damping coeff. induced by compression-extension waves at sidewalls
xviii
Cwrx2 = rocking damping coefficient induced by shear wave at sidewalls

CWrx3 = rocking damping coefficient induced by torsion wave at sidewalls
Cv = vertical damping coefficient of soil-foundation system
C1 ,C 2 = damping parameter of embedded foundation
DL = Dead Load
DT = time step of integration scheme
Dc cracking displacement
Dm maximum displacement experienced
DI damage index
DIm = member damage index
DIs = structural damage index
DR = damage ratio
Dy yield displacement
E = earthquake load
Ee = modulus of elasticity of concrete
Eeq = equivalent modulus of elasticity of soil
Eb = hysteresis energy
E" = kinetic energy

En = dissipation hysteretic energy under positive forces
Ep = dissipation hysteretic energy under positive forces
E. modulus of elasticity of soil
E, strain energy
Et total energy
Ey = viscous energy
EI member flexural rigidity
Fe = cracking force
Fe = damping force
FDR Kr/Kf
Fi inertia force
Fk elastic force
Fy yield Force
G shear modulus of soil
Ge shear modulus of elasticity of concrete
Gem complex soil shear modulus
Gm6X . = maximum soil shear modulus
xix
Go = soil shear modulus at first cycle of cyclic loading

Go -- soil shear modulus at ground surface
G. - soil shear modulus at embedded portion
G oec. = apparent soil shear modulus
Gz = soil shear modulus at depth z from ground surface
GN = soil shear modulus after ~ cycles of cyclic loading
Hb = total height of building
Ib mass moment inertia of base-mat
I"" = second moment of inertia of footing with respect to x-axis

I"y = second moment of inertia of footing with respect to y-axis
~ = effective second moment of inertia
Ig = gross second moment of inertia
Ij = mass moment inertia of storey j
10 - mass moment of inertia of footing foundation
~ = total mass moment inertia
Itre = coefficient of trench effect
IWoll = coefficient of side-wall effect
Iwe - effective wall second moment of inertia
lwg - gross second moment inertia of wall

l.p = mass moment inertia with respect to horizontal axis
K = stiffness matrix of MDOF system
K..vu = average unloading stiffness
K d1 = dynamic stiffness corresponds to vibration mode i
Kw = effective relative stiffness of beam
~ = equivalent soil-foundation stiffness

Kf = initial stiffness
Khb = horizontal static stiffness of soil-foundation system
Kbr = horizontal-rocking static stiffness
Ka = coefficient earth-pressure at rest
~ = post elastic stiffness

K... = rocking static stiffness of soil-foundation system
Kdv,emb = dynamic vertical stiffness of embedded footing foundation
Ksv,emb = static vertical stiffness of embedded footing foundation
Kdx,emb = dynamic horizontal stiffness of embedded footing in x-direction
Kdy,ernb = dynamic horizontal stiffness of embedded footing in y-directioll

xx
K,,,,emb = static horizontal stiffness of embedded footing in x-direction
Ksy,emb = static horizontal stiffness of embedded footing in y-direction
K"I = static stiffness corresponds to vibration mode i
Kuxtsur = surface static rocking stiffness with respect to x-axis
K'IY"ur -- su rface static rocking stiffness with respect to y-axis

Ksv.sur = static vertical stiffness of surface foundation
Ksx.lliur = static horizontal stiffness of surface fou ndation in x-direction

Ksy,sur = static horizontal stiffness of surface foundation in y-direction
Kunld = reloading stiffness

KHH = horizontal soil-foundation impedance
KRR = rocking soil-foundation impedance
KHR - horizontal-rocking soil-foundation impedance
Kr = secant stiffness
~ initial stiffness
vertical stiffness of soil-foundation system
L = span of the beam
L one-half of footing length
~ = length of beam plastic hinges
Lr = live load reduction factor
Lu minimum uniformly distributed live load

LL = live load
M = material factor
M = mass matrix of MDOF system
M moment
M earthquake magnitude
Me = critical overturning moment

MDOF = multi degree of freedom
ME = moment induced by earthquake
Mj = ideal moment
M d +e moment due to dead and earthquake load
Mo moment amplitude of footing foundation
Moo moment after redistribution

M = mass off soil medium
Mu ultimate moment
N = number of storey
xxi
N = number of blows in SPT soil test

NCR :::: normalized cumulative rotation
Nz total vertical side-wall reaction of embedded footing foundation
OCR = Over Consolidated Ratio of clay soil
OD = overall drift
P = axial load
Pumi = minimum column axial load
Puma = maximum column axial load
Po initial axial load
Pu = ultimate axial load
Py = axial yield force
Py :::: yield strength
PGA peak ground acceleration
PGV = peak ground velocity
PH~ = total plastic hinge rotation
PI = plasticity index of soil
R seismic risk factor
R = epicentre distance
~ = reduction coefficient
R.n :::: moment Reduction factor

R :::: distance from centre of mass to rocking point
Rs :::: the shorter distance between site to fault rupture
R, = axial load reduction factor
Rz :::: base vertical reaction of soil
S :::: structural type factor
S :::: a number of vibration modes
Si :::: ideal strength
So over-strength
Sp :::: probable strength
Su :::: required strength
Sy :::: shape dependent coefficient of soil-foundation horizontal stiffness
Sz :::: shape dependent coefficient of soil-foundation vertical stiffness
SF safety factor
SDOF = single degree of freedom
SPT :::: standard penetration test
xxii
Sl>S2 = stiffness parameter of embedded foundation

T = structural period
T, = fundamental period of structure
V jt total horizontal displacement of storey j
Dg == ground acceleration
Vo displacement amplitude of footing foundation
Uu = ultimate horizontal displacement
V horizontal reaction at footing base
Vb = base shear
Y re = compression-extension wave velocity
Y La = Lysmer's analog velocity
Yo = horizontal force amplitude of footing foundation
V, shear wave velocity
ye. m = shear wave velocity taking into account of the effect of soil material damping
Yj = generalized displacement of mode i
W = total dead load of the structure
Wo weight of the footing
Wom = total weight of footing and machine
Z soil depth
Z = rigid joint zone length ratio with respect to member depth
a = angle between radius of rocking with vertical side of rigid body
a = coefficient of post elastic stiffness
a = soil shear modulus parameter with respect to soil depth
a h/lw
a = slenderness ratio
{3 width/thickness coefficient of structural wall
{3 = w/wv = frequency ratio

f3 = deterioration coefficient
{3 = coefficient of reloading stiffness
{3 parameter of Newmark integration scheme
15 soil shear modulus degradation index
15 ratio of embedment to radius of footing
Dc = calculated damage parameter
15 m maximum deformation
Dt threshold damage parameter
xxiii
Ou = ultimate damage parameter

Ou = ultimate deformation
E = strain
E,m = maximum predicted steel strain
~ = internal friction of soil
~f = under-capacity factor
QJ; = coordinate of mode i
~m = maximum CUlvature
I/Jo == flexural over-strength factor
CPp = probable strength factor
l' = unit weight of soil
"I = mlmo = mass ratio
"I = parameter of Newmark integration scheme

"10,"11 dimensionless coefficient fol' dampers
"I" = soil shear strain
'Y.t = threshold soil shear strain
'Y.v = soil density

Ao = over-strength factor
A - weighting factor of member/structural damage indices
/LA - displacement ductility factor

/LLI,m = member displacement ductility factor
/La" = structural displacement ductility factor
/L1> - curvature ductility factol'

/Lo,/Ll = dimensionless coefficient for masses
1] viscosity coefficient
0 - rocking angle
0 - time step factor of Wilson-O method
0 = amplitude rocking angle of footing foundation
00 initial angle of rocking block
Ou = ultimate rocking angle i.e when footing commence to lift-off
Of = complex rocking displacement l'esponse
~ = damping ratio
~ = axial load eccentricity factor of structural wall
~m = damping ratio effect of material damping

~m = damping ratio of the model
xxiv
~'Yq bearing capacity ratio between interfering and isolated footing

p = mass density
P = soil Poisson's Ratio
Pc = poisson's ratio of concrete
CPo = initial displacement of rocking a rigid bodies
/Pi = displacement after the ilb impact
(T = stress
(Th = horizontal geostatic effective stress of soil
110 = confining pressure of soil
(Tv = vertical geostatic effective stress of soil
1/; = rocking angle
W natural frequency of superstructure
Wm = dynamic moment magnification factor
Wm = coefficient of member damage index
w, dynamic shear magnification factor
WI = coefficient of plastic hinge rotation
Wv = natural frequency of the structural system
r rx,lre = coefficient of trench effect on the static rocking stiffness with respect to x-axis
r ry,tre coefficient of trench effect on the static rocking stiffness with respect to y-axis
r rx,wall = coefficient of sidewall effect on the static rocking stiffness with respect to x-axis
r ry,wall coefficient of sidewall effect on the static rocking stiffness with respect to y-axis
Chapter I
Introduction.
1.1 GENERAL
Efficiency and safety are two important considerations in the seismic resistant design
of structures. Sometimes these considerations compromise each other because they are
negatively correlated. To maximize the efficiency is the general goal in the seismic resistant
design and construction of the building structures, however, this goal tends to increase the
structural risk since maximizing the efficiency tends minimizing the apparent structural safety.
An optimization between efficiency and safety in aseismic design and construction, therefore,
needs to be achieved although the optimum value itself is relative and difficult to define
quantitatively.
Evidence from recent earthquakes A21 ,B28,H28,M31,W32, indicates that the building damage is
mainly due to defective application of the principles of seismic resistant design of structures.
These defects involve unclear application of the design philosophy, poor building
configurations, energy dissipation, detailing and construction. These all lead to a minimum
value of the structural safety factor. Therefore, one of the most effective ways of mitigating
the destructive effect of earthquakes is to improve the existing methods of structural analysis,
to develop new or better methods in designing, detailing, constructing and maintaining
structures. This also includes the retrofitting of existing hazardous facilities B2.5.
The structural analysis may be able to estimate the response demands, while the
capacity response can be shown by laboratory tests. The seismic design limit states, consisting
of serviceability, damage control and survival limit states, currently have been clearly
defined Pl8 The estimation of the seismic responses of the structures has improved significantly
since the development of more rational inelastic structural analyses. This is aided by
considering more rational member models and energy dissipation models. However, the
quantitative safety margin is still difficult to define because some uncertainties are involvedK22 ,
1
2
i.e modelling the input ground motion and modelling of structural system. Ref. PIS states that
the input motion corresponding to the boundaries between these limit states, cannot be defined
precisely, and the analysis commonly ignores the coupling effect between soil-foundation and
superstructure. In an uncoupled system, the superstructures are assumed to be fixed at the soil
base. However, soils are not a rigid material and are not bonded to the superstructure.
An improvement of the common methods of structural analysis is the use of soil-

structure interaction which has been developed over the last few decades. Two different
approaches have been made to the solution, either a continuum-impedance approach or discrete-
numerical methods can be utilized. By considering the coupling effect between the soil-
foundation and superstructure to be a whole structural system, more. confidence in the results
from the structural analyses in estimating the structural safety can be achieved. The apparent
factors of safety can be reduced when this confidence increases, enabling optimization of the
cost of constructionH29
Two important phenomena that distinguish the uncoupling and coupling effects between
soil-foundation and the superstructure in soil-structure interaction are the effects of the
kinematic and inertial interactions. These phenomena are due to the fact that the earthquake
ground motions are not directly transmitted to the superstructure but pass through the soil-
foundation system. The kinematic interaction refers to the effect of spatial variation of the
seismic motions at different points due to differing elevation or incoherent seismic motions with
respect to a common motion at the relatively stiff basement. The inertial interaction refers to
soil deformation inducing rocking, vertical and horizontal displacement vibration modes of
superstructures. In addition to soil-material damping due to the non-linear inelastic behaviour
of soil, a radiation, spatial or geometrical damping occurs during these vibration modes i.e an
amount of energy is dissipated through the vibration energy transmitted away into the
surrounding foundation footings.
Several aspects affect the incoherent seismic motions and the studies carried out in
Ref. V 10, Vll and VI2 concluded that the kinematic interaction is less significant in
comparison with the inertial interaction, especially for relatively shallow foundations. The
kinematic interaction in the continuum-impedance method of soil-structure interaction on a
stmcture with shallow foundation is often neglected. The foundation input motions are assumed
to be the same as the free field motions.
3
The improvement in the analyses using the continuum-impedance method in soil-

structure interaction can be achieved by using more rational soil models. These models involve
incorporating the layering of the soil over a half_spaceH17 ,L8,L9,W22, the fictitious layer due to non-
linear behaviour of soil adjacent to the footing surfaceP11l the homogeneous layer over the
horizontal rigid boundaryGS, using more degrees of freedom and elements of lumped
models D14 ,w9 and using more refined analysis such as boundary element and partially embedded
arbitrary shaped foundations D3 ,F4,H7.G1,G2,G13.
Both analytical and experimental investigations of building structures supported by

foundation compliance to simulate effect of soil-foundation flexibility have been reported in
some references. CHOPRA & GUTIERREZ (1974) presented the analytical investigation of
elastic MDOF structures resting on rigid circular disk footings supported by an elastic half-
space soil model. CLOUGH & HUCKELBRIDGE (1977) reported the results of laboratory
tests of three storey steel frames where the foundations were allowed to lift-off.
HUCKELBRIDGE & CLOUGH (1978) carried out similar analytical and experimental
investigations of nine-storey steel frames. PRIESTLEY, EVISON & CARR (1977) conducted
laboratory tests of a SDOF structure supported by both rigid and flexible rubber pad
foundations. YIM & CHOPRA (1983, 1985) reported the analytical investigation of elastic
SDOF structures supported by Winkler foundation and elastic MDOF frame structures
supported by two spdng-damper element foundations. PSYCHARIS (1983) discussed the
response of elastic MDOF structures supported by two spring-damper element foundations.
HIFNAWY & NOV AK (1986) investigated an elastic MDOF structure supported by pile
foundations.
1.2 OBJECTIVES OF THE RESEARCH
Some research, either analytical or experimental investigations on the effects of the

foundation compliance on the response of building structures has been carried out. However,
most ofthese investigations are based on gross assumptions i.e by using very simple structures,
simple soil and foundation models, elastic response of superstructures and as a result the
investigations place greater emphasis on the scientific aspects rather than the engineering
assessment. More realistic superstructures, soil and foundation models, structural responses and
essential guidance for both the analysis and the design of rocking structures are needed.
4
The first objective of this research is to ascertain the advantages and disadvantages of
rocking structures when compared to the system where rocking is prevented, using more
realistic superstructures, structural responses, modelling of soil and supporting foundations and
including effects of partial lift-off of the footing foundations.
Using the results obtained in the first objective, the second objective will be to attempt
to develop the essential guidance for analysis or design of rocking structures including the
partial lift-off phenomenon of the footing foundations.
1.3 SCOPE AND OUTLINE OF THE THESIS
More uncertainties are found in the nature of soils than that in superstructures. Part of
the rationale in engineering methods is to minimize the uncertainties by the use of logical
assumptionsH29 so that the expected solution to the problems can be achieved without any
significant loss of accuracy. A number of assumptions were made in this thesis.
To evaluate the developments, the results, and the proper selection for the further
research on rocking structures, a review of the previous research needs to be carried out. This
research review is presented in Chapter II. The selection of several aspects including the
building prototypes, configurations, number of stories and the preliminary design of the
corresponding building prototypes are presented in Chapter ill. The preliminary design and
selection of these aspects are based on the principles of aseismic design of building structures
and the New Zealand building codes.
Soil-foundation impedances are a function of soil-shear modulus which is strain

dependent. Chapter IV discusses the dynamic properties of soils and the maximum possible soil
shear strain induced by earthquakes and the simple soil rheological models in terms of
engineering applications.
Soil-foundation impedances are also functions of the soil and footing foundation model
being used. A number of possible uses of these models for substructure modelling are presented
in Chapter V and followed by the selection of the more realistic soil-foundation models. The
corresponding soil mechanical model and the selected soil properties are also presented.
Superstructure modelling is presented in Chapter VI. This chapter involves the modelling of
5
horizontal load resisting systems, modelling of inelastic reinforce concrete members, effective
stiffness, inelastic load-deformation relationships and the modelling of plastic hinge and rigid
joint zone lengths.
Several soil-foundation impedance formulae have been proposed by researchers based

on the various soil and foundation models, methods of analysis and level of soil shear strain.
The soil-footing foundation responses and impedances are discussed in Chapter vn. The
procedures of the analysis of soil-structure interaction is presented in Chapter VIII. This
chapter involves the mathematical expression of the motions of the structural system, the
overall damping of the structural system, the choice of the numerical integration schemes,
earthquake excitations and the methods used in the analyses.
Chapter IX discusses the inelastic response parameters, damage indicators and these
response parameters on the corresponding building prototypes supported by the flexible soil-
foundation system in comparison with the more common fixed base structures, when subjected
to the 1940 El Centro N-S component earthquake. This earthquake has often been used as the
standard earthquake excitation in many codesA2,x4.
Some parameters may show diverse effects on the structural responses. The effects of
different rigid joint lengths, number of frames and flexural member hysteresis rules on the
seismic responses of frames and frame-wall structures supported by flexible foundations and
fixed base foundations are presented in Chapter X, Chapter XI investigates the effects of
different soil moduli, soil shear strains and soil models on the same structural system. The
effects of different earthquake excitations on the same structural system are presented and
discussed in Chapter XII.
Based on the results obtained from the four foregoing chapters, the suggested analysis
methods and proposed design guidance for rocking structures is presented in Chapter XIII.
Finally, some conclusions and recommendations for further research are presented in Chapter
XIV.
Chapter II
Review of the Usage of Lumped Parameter Methods in
Rocking Olf Building Structures
2.1 INTRODUCTION
The analysis and design of multistorey buildings subjected to earthquake excitations is

very important in seismically hazardous regions. The response of the structures cannot be
computed unless the method of transmission of the ground excitation from the soil into the
superstructure has been determined.
At the present time, the most common assumption is that the free-field strong motion
accelerogram may be applied directly at the base of the building. The superstructure is also
assumed to be firmly bonded to a perfectly rigid baseeJ ,C6.M3 i.e the building is assumed to be
fixed at its foundations. This assumption is usually adopted in the analyses of multi-degree of
freedom structuresX4
Depending on the site conditions, the actual soil may be soft or firm, but it is not
infinitely stiff. In addition, the structures may rest on the ground under the force of gravity
only and are not restrained from rocking. A more accurate response analysis may be achieved
when the soil is modelled as a flexible material and not assumed to be infinitely rigidMJ H2
Taking into account the effect of the soils on the analysis of building structures is usually
referred to as a soil-structure interaction analysis. A lumped parameter method is often used
in this soil-structure analysis, where the flexibility of the soil is represented by a system of
springs and dashpots. The earthquake excitation is not directly transmitted to the superstructure
but passes through the soil-foundation interaction system. The effects of the soil-structure
interaction on the superstructure response may be of great significanceWIO for building structures
resting on soft soilsxl.
6
7
The effects on the responses of building soil-structure interaction have been reported
by researchers. Major research results, especially using lumped parameter methods, are
described in the following sections.
2.2 EXPERIMENTAL INVESTIGATIONS
2.2.1 Clough and Huckelbddge (l977)C7
In some earthquakes the overturning moment at the base of the structure due to the
lateral inertia forces may exceed the available overturning resistance available from the weight
of the structure. Partial transient separation of the foundation from the ground may occur when
additional anchorage systems are not provided. However, greater effort to provide overturning
constraint to the structure in which the separation is prevented under a major earthquake would
seem uneconomicalC7 . Therefore, a better understanding of the behaviour of the rocking
structure is necessalY.
CLOUGH & HUCKELBRIDGE (1977), investigated simulated earthquake excitation

induced dynamic response of a simple structure in which the columns are allowed to lift-off.
A three dimensional three-storey single bay steel frame structure was tested where the total load
of 8 kips was equally distributed on each floor. The structural test model is shown in Fig. 2.1.
The structure model was placed on a shaking table. The experiments showed that the structural
model would result in a relatively high lift-off response. The column's bases were pinned so
that the columns could rotate freely. Two types of impact pads with different stiffnesses were
provided beneath the column bases during the lift-off responses.
Two strong motion accelerograms were used to excite the table, the 1940 EI Centro N-
S and the 1971 Pacoima Dam S47W records. The El Centro record was used both with and
without the corresponding vertical component of the input motion. Each signal was run at a
wide range of scale factors, from a maximum acceleration of 0.1 g to 1.1 g. Hundred's of data
channels recorded the time histories of table acceleration, floor displacements, floor shears,
overturning moments, column responses and column axial forces. Typical results from these
tests are shown in Fig. 2.2 and Fig. 2.3.
8
Fig. 2.1 Structural Model
I.'
1.1
~
::
~
-I
l.0l-'-:,,~
7 7" ,~,
- :
- :: - ' ;
-- - :.t..:-
: : . -:-:-
. ~
. ~
.~.. ~.lJll~
~
. . . .,.,-,U
-:-
....,...,.JU
11"[ lit 'HCOHO$
Jrd Floor Accelerat.ion ....,..---,---,------,-----.---.
i::\i .:f*~ r::........... ;:~:~}~:;: ". ". ". . . . . .

'.' 1'.' ~.O I" .ot
TIt1
Li
IN

HCDHIl5
1.1 lI.t ILl It 11.t U.t &J.I
ln4 Floor Accelerat.Lon
Fig. 2.2 Floor Accelerations and Storey Shears of the Fixed Bases
JNftru.~iVV~ ;::E3..!.t::i~: ',:r.~

I.S .
;. j~
1*' ", 11.0--' '
AL,lhlr.\ IJ AII A r\ : I.' . ",hi ... AI\fH II
l:::\-I.,. I.G 1.1l I,' td.l '.I '.0 'II,. 14.' 11.11 ILl ,1 't.G U.I T II" I!. " . ', ..
11t1 U S((i).KOS
led Floor Shl!lar
led Floor AcceleJ;'Q,tion
2nd floor Accelerat.ion
Fig. 2.3 Floor Accelerations and Storey Shears of the Lift-off Structures
9
The floor accelerations and storey shears from the scaled El Centro N-S excitation with
maximum acceleration of 0.74 g with lift-off of the column bases prevented, are shown in Fig.
2.2. The maximum floor accelerations were approximately 1.46 g and 1.95 g with the
corresponding storey shears being approximately 28 kips and 18 kips for the second and third
floors respectively. Fig. 2.3 shows the floor accelerations and storey shears when the base-mat
is allowed to lift-off. In this case, the building model is subjected to a scaled 1940 N-S El
Centro earthquake with maximum acceleration of 0.67 g. Maximum floor accelerations were
approximately of 1.03 g and 1.16 g, and the corresponding storey shears were approximately
of 18.1 kips and 10.8 kips for the second and third floors respectively.
Even though these structural responses are not exactly comparable because of the
slightly different applied table accelerations, it is clear that when column base is not allowed
to lift-off, the floor accelerations are greater than when column base lift-off is permitted. The
storey shears of the structure without column lift-off are greater than those when lift-off is
allowed. It was also found from this experiment that for a low level applied loading, only
slightly different responses resulted from these two different base mat stiffnesses. The general
result of the investigation showed that lift-off results in a definite reduction of the structural
response.
2.2.2 PRIESTLEY, EVISON AND CARR (1978)1':3,1'4
The possibility of the foundation rocking of shear wall structures designed to NZS
4203: 1976 was reported by Priestley, Evison and Carr (1978). When using the building code
NZS:4203, to derive the physical properties of a structure in particular seismic zones, the
product of the Structural Type Factor S, Materials Factors M, and capacity factor cf>jcf>f may
be greater than 2.0. The code states that:
/I flO -foundation system need to be designed to resist the overturning

moments greater than those resulting from hon'wntal forces corresponding
to SxM = 2.0. 11 clause 3.3.6.1
Because of this limitation, some ofthe building systems will be allowed to rock on their
foundations under seismic excitations.
10
Laboratory tests were conducted to describe the mechanism of the rocking process and
finally a simple design method was proposed to predict rocking displacement at the centre of
mass of the structures under an earthquake excitation. The experimental structural model was
a single degree of freedom stiff steel column placed on a mild steel plate. A similar steel
member was used as a foundation beam and rigidly welded to the base plate and column. The
model was placed on the shaking table as shown in Fig. 2.4.
The model was a 1/6 scale model of the prototype, and was sUbjected to free vibration
or earthquake excitations. The example of the response to 0.6 EI Centro 1940 N-S is shown
in Fig. 2.5. The laboratory measurement of the reduction of horizontal response due to
foundation impact was compared with the formulas derived for the Rocking of a Rigid Block
on a Rigid Base by HOUSNER (1963). After i impacts, the horizontal displacement of the
centre mass in terms of the initial displacement IPo may be written as,
(2.la)
in which rand lPo respectively are,
mRl( l - cos2a)
r = 1 - T
[
r (2.th)
(2.k)
where m is the mass of the rigid block, 10 is mass moment inertia of the block about the
rocking point, ~ is the distance of the line between the centre of mass to the rocking point,
~ is the angle between this line and vertical plane of the block and 60 is the initial angle of the
rocking block.
Several test runs were made and the horizontal displacement was plotted against the
measured frequency as shown in Fig. 2.6. Good agreement was found between the measured
reSUlts and the theoretical response. It was found that the damping ratio ~ was 0.1 fn where
fr is the rocking frequency. A relationship between the damping ratio ~ and the frequency f and
horizontal rocking displacement of the centre of mass Ym against the frequency was then
obtained. From an Elastic Response Spectrum and these relationships, it was possible to
predict the horizontal rocking displacement of the prototype for a given damping ratio ~. The
prediction procedures are follows:
11
In!l.lrUmfllnllltiofi ;....
-=.: DI;;~LlltJr
--
-INPUT- -QUTPUT-
Fig. 2.4 Strucrural Model
-~- -~-
I- IOmrn I Fronl Fool IJplift
I---t O.:Jg V~j<., A<c.lfooO
I----l SOmm Horiron'.'

O;,p/.'''ffl.o1
1---II.Olf I/orilonl.'
.4ta:.'Hali(Jn
l---i 'am", ToOl". 'llilP',
~
Back ,:"", IJpllft
Fig. 2.5 Structural Responses
..
J:
'2 3 4 5 I'> 1 8
Model PClok Honzonlal OtsplocG1Q:nt IlTIml- 1h!Xlrgllc(l1 Equlvohmt VISCOUS Damping.. )., t%gt
Fig. 2.6 Displacement vs Rocking Frequency Fig. 2.7 Viscous Damping vs Rocking Frequency
12
1. Assuming the peak horizontal rocking displacement Ym for the model, then the
model rocking frequency foo may be determined from Fig. 2.6,
2. The damping ratio ~ro (A in the figure) can be found from Fig. 2.7,
3. The resulting rocking frequency fro of the model is then scaled to the
equivalent value for the prototype frequency ~ by using the scale factor ( in
this case equal to 6). Based on these values ~ro and 4 and using an elastic
response spectrum, the prototype horizontal rocking displacement Yl' may be
found. The resulting Yp is then scaled to the horizontal rocking displacement
of the model Yml,
4. Yoo from step 1 and Yml from step 3 are then compared. IfYm and Yool do not
agree with each other, a second trial must be made by using the new horizontal
rocking displacement Yml and re-starting from step 1. Trial and error-iteration
are repeated until Ym i-I = Ym i and tm i-I = tooi, where i = 1,2 ... n, are the
iteration numbers.
2.2.3 HUCKELBRIDGE AND CLOUGH (1978)H8.
Having been successful on their first experiment -in the three story steel frame
structure, HUCKELBRIDGE & CLOUGH (1978) undertook a similar experiment and
analytical investigation on the behaviour of a more complicated structure. The model was three-
bay nine-story steel structure with open-frames, and approximately one-third scale of the
prototype. A gravity load of 10 kips is applied equally distributed on each floor. This structural
model was placed on the shaking table as shown in Fig. 2.8. In this experiment, the column
bases were designed to allow lift-off or act as fixed-bases. In order to carry out the fixed base
test the elements to accommodate column lift-off were removed and the column bases were
bolted rigidly to the foundation.
The experimental investigation had two primary objectives; first, to obtain data
describing the lift-off behaviour of the frame and to obtain comparative data when column base
lift-off was allowed or prevented under similar ground excitations. Meanwhile the analytical
13
., "
"
Fig. 2.8 Structural Model Fig. 2.9 Mathematical Model
:;;:; u . , - - - -.......- - - - , . - - - - - , - - - - - . . , . . . - - - - ,
4"
: ~ ~ ...,
_:: ~u,.J-.----:".-----:':...----:,'". ...----:,t".---..,.}
's IHlmU
::
=g::::.'!-----,
tt . , - - - - - . , . - - - - - , , . . . . . . - - - - , . - - - - , - - - - ,
~~:::=I::.1.=~&::B::~::M::~=~.
lJrot: In HfC_OI
l.JJUA( Ill ... us. fUr. ,.n flU
.-
tUf( U
1, Th'A( UAG
uenes
fae "lIlT fur
10 Responses of the Bases Fig. 11 Responses of the Lift-off Structu res

14
program was to evaluate the available non-linear numerical techniques for predicting the lift-off
response of these open frame structures. The second objective of the analytical program was
to evaluate the current computer program capabilities for predicting non-linear frame response
through the use of concentrated plastic hinges located at the end of the members. The
mathematical model used in the analytical investigation is shown in Fig. 2.9.
Free-field 1940 EI Centro NS and Pacoima Dam S74W ealthquake ground motions
were used as input to the shaking table. Each record was applied over a wide range of
intensities both with and without the vertical component of the ground motion. The structural
response quantities recorded included horizontal floor accelerations, floor displacements,
member forces and deformations and column lift-off displacements. The typical records of story
shear and overturning moment for both fixed and lift-off cases are shown in Fig. 2.10 and 2.11
respectively.
As shown, the story shears and overturning moments when lift-off is allowed in general
are smaller than those when lift-off is prevented. The maximum storey shears of the sixth floor
when base lift-off was allowed was approximately 45 Kips compared with 49.5 kips when lift-
off was prevented. Many short-duration transient responses and impacts are seen on the lift-off
response shown in Fig. 2.11.
Computed results provided very good agreement with the responses from the
experimental tests both with and without lift-off. As shown in Fig. 2.10, the structural response
is dominated by the first mode. The horizontal floor displacement for the fixed base case is
slightly less than when lift-off is allowed. The structures in which lift-off is allowed showed
damping ratios ~ as being quite small. For steel structures it is possible to assume that they are
undamped. The equivalent damping ratio ~ of the structure without lift-off was approximately
3.2 % which is considerably greater than when lift-off was allowed.
The final conclusions of this investigation are similar to those for the earlier three story
bUilding. The structural responses and ductility demands are reduced when the column base is
allowed to lift-off. Providing a large ductility in a reinforced concrete structure or braced steel
frame can be costly.
15
2.3 ANALYTICAL INVESTIGATIONS.
2.3.1 Chopra and Gutierrez (1974)C6.
An efficient method of soil-structure interaction analysis was proposed by CHOPRA

& GUTIERREZ (1974) for a multistorey building where its foundation rested on a linear
homogenous elastic half space as shown in Fig. 2.12. The dynamic stiffness and damping force
of the soil-structure interaction depend upon the frequency of excitation. The governing
differential equations of motion of the soil-structure interaction must therefore be written in
term of the frequency domain. Analytical solution of these differential equations leads to a large
amount of complex computation effort because the solution has to be repeated for many
excitation frequencies c6 , A lumped parameter method was used in that the flexibility of the soils
was represented by spring and dashpot systems.
The building is assumed to have N rigid floors resulting in a single degree of freedom
per floor. The foundation is assumed to be a rigid circular footing and its thickness is
neglected. Coupling between the horizontal and rocking vibration modes was considered, thus
the system has N +2 governing differential equations of motion. Those equations can be
determined by considering dynamic equilibrium of the corresponding vibration modes. The
dynamic equilibrium of the N floor masses due to ground acceleration fig are,
(2.2)
j :::: 1, 2, ... N
where U/ = u& + U o + hj 8 + Uj is the total horizontal floor displacement at level j.
The governing equations of motion corresponding to horizontal translation and rocking

are,
j=N
(2.3)
Em U/+ m" (ug+u)
j=l
j + V(t) :::: 0
j=N (2.4)
Em j hj iJ/ + Ite + M(t) :::: 0
j=1
where M, C and K are the mass, damping and stiffness matrices respectively, ID.i = mass of
floor j, mo = mass of the base-mat, Uo = horizontal translation of the footing in addition to the
16
free field motion, u g = free field horizontal ground motion, hj = height of the floor j, Ij =
moment of inertia of mass j, 10 = moment of inertia of base and 0 is rocking angle.
Eg. 2.2, 2.3 and Eg. 2.4 may be rewritten in form,
(2.5)
(2.6)
(2.7)
where,
(2.8)
j=N j=N (2.9)

It = 1(} + Ll} L; = Em j hj
j=l j=l
A steady state harmonic loading with frequency w is applied to the system. The
horizontal reaction V and moment M acting at the foundation can be expressed in terms of
V = Vo e iUA and M = Mo eiUA . The resulting horizontal displacement and rotation are uo(t) =
Vo e iUA and 8(t) = e
ei<d. The linear force and displacement relationship of the system can be
expressed in matrix form,
== [(k M +ia./::;,)Khh (2.10a)

(k'h +iaocrh)Krh
(2.10b)
where Khh = horizontal static stiffness, KIT is static rocking stiffness, Khr = Krh is translation-
rotation static stiffness, k'g and C'g are dimensionless dynamic coefficients, ~ = dimensionless
17
frequency, 1'0 = radius of footing foundation. The values of k's and c's may be determined
from available graphV4 as shown in Fig. 2.13, or from formulaeV2.WlO.
For a given ground acceleration Og(t) ::: eitA , the response quantity u(t) can be expressed
in term of a complex frequency response u(t) = u.(w) eilAt and similarly for the response
quantities u,,(t) = uo~w)eiWi, O(t)= OrCw)ei<.t, V(t)= V~w)eiWl and M(t) = M~w)eiWi. Substituting
these expressions into Eqs. 2.5, 2.6 and 2.7 and after dividing by eitA lead to,
(2.11)
--i(o)1 11'Mu,(i(o) -cil-m,uo!i(o) -i(o)'J.L;Ojf.J +K" ..o)uoff.J + (2.12)

K.,m(,)Oj(o) =-m t
-f.t)
2h I'Mu (,) -(,)2L;uof(,) -6lI1I Ojf.J -(o)+Krnif.Juoff.J + (2.13)
g h
K_(w)OJ6lI) = - ,.
where Kvv, Kvm, K",y and K",m are from Eq. 2.10. The complex frequency response u.(w) , uarCw)
and OrCw) for any particular applied frequency w may be determined by solving the
simultaneously equations of motion of Eq. 2.11, 2.12 and Eq. 2.13. The response of the system
due to an arbitrary ground motion Ug(t) may then be determined by Fourier Synthesis.
The computational effort to determine the unknown complex frequency responses from
Eq. 2.11, 2.12 and Eq. 2.13 is quite expensive because the calculation has to be repeated for
many values of w for the N +2 equations. An efficient approach to the analysis is introduced
by reducing the number of degree of freedom of the superstructure by using only S vibration
modes, where S is much less than N. Assuming that the structural system has classical
vibration modes similar to those of a fixed base structure, the complex frequency response of
the superstructure u.(w) may be written in terms of the first S modes contribution as,
t~
(2.14)
ug ( w) = I: Y;( (ol)
~4
where = unknown generalized displacement, i = coordinate of mode i.

18
V ELASTIC HALF SPACE (P ,C. ,I' )

~
Fig, 2.12 Structural Model 13 Dynamic Stiffness and

Damping Coefficients
9
8
10
7
t5
.. 6
15
4
'3 Number of Madea Ine'Ud.dl
Numbar of Madea Included.
, 2 Exo<:t 5
------1
3
~~~~~~~~~~~~5
:5 >1 II I;
W/W
.7
.6 Number of Modes InCIUd~dl

.5 ~\
E~a(;1 1 \15
I
.4
I
I
I
I
..
."
.3 I "
0.
E
.z. <t
.1
o l~~~~~-L~-L~-L~~~~
o 2 7 2 4 1\ 7
WIW,
Fig. 2.14 Acceleration of the l<t and 25 th Fig. 2.15 Acceleration of the Base and
floor vs Frequency Ratio 13 th floor vs Frequency Ratio
19
Substituting Eq. 2.14 in Eq. 2.11, 2.12 and 2.13 and pre-multiplying by <PiT and
utilizing the orthogonality properties results in S +2 simultaneous linear algebraic equations
with the unknown Y1(w), i = 1,2 .. S, UoAw) and 0Aw),
An example of this approach applied to a 25-story building with a fundamental period

of 0.5 sec was presented. It must be noted that this fundamental period of free vibration is not
realistic for such a tall buildings. However, the model requires a large number of stories Nand
the building needs to be stiff relative to the foundation so that the soil-structure interaction is
significant.
The building has a damping ratio of 5 % , radius of footing foundation ro = 50 ft,

Poisson's ratio 1I = 0.333, and shear wave velocity Vs = 1000 ftlsec and was subjected to unit
ground motion Uit) = 1 eic.t . The accelerations for the 25th , 13th, pt floor and base mat are
shown in Fig. 2.14 and Fig. 2.15. These results indicate that the complex frequency response
using the limited contribution of S vibration modes are relatively close to the exact value.
Although this system does not possess classical normal modes, it does show advantages in
simplifying the analysis.
2.3.2 Yim and Chopra (1984)C3C4
2.3.2.1 Flexible Elastic SDOF Structure on tbe Winkler Foundation
Foundation vibration whether generated by machine vibration or earthquake

excitation usually exhibit three vibration modes, vertical translation, horizontal translation and
a rocking oscillation. Among these vibration modes, the rocking oscillation may be the most
critical mode for foundation stability M;lI7, as a partial lift-off of the foundation base may occur,
especially for an unbonded soil-foundation system.
Effects of transient lift-off of the foundation on a single degree of freedom

flexible structure was investigated by YIM & CHOPRA (1984). The structure with mass m
was supported by a rigid foundation-mat of mass IDa resting on a Winkler foundation system
as shown in Fig. 2.16. Vertical vibration and rocking oscillation were considered in this
analysis, resulting in 3 differential equations of motion. The structural system is excited by a
horizontal ground acceleration Uit). The horizontal and vertical displacements are denoted by
20
u(t) and vet) respectively. When the system is vertical a static load (m + mo)g is applied in the
centre of the foundation mat, the static displacement v. is,
(2.15)
where g is the acceleration of gravity, b is one-half of the foundation width and ~I is the
foundation spring stiffness per unit length of footing foundation.
When the soil-foundation system is unbonded, one edge of the foundation mat
is at incipient lift-off when the ultimate foundation rotation au reaches as the value,
(m+m,) g
(2.16)
2 k wl b 2
The corresponding ultimate base shear BS u , base shear coefficient BSCu and
horizontal displacement Vu are,
(m+m,)g b (2.17)
3h
(2.18)
(m+m,)g b
(2.19)
3kh
where I = mjm is the mass ratio, ex = h/b is the slenderness ratio and W = m g is the total
weight of the system.
The critical forces correspond to the eccentricity reaching one-half of the

footing length and the building is in the point of overturning. The limiting value for the critical
overturning moment Me. base shear BSe and base shear coefficient BSCe are,
(2.20)
21
/
- - fiOf;ltr[tlcrulTlt'f
.lJF\.IF."JIICW.>rTCO /
- (J";;(JC<VlO CC.. 1"AtT.
U'LII'T P1Al:ilfTITtl
--.~:--.~A--:'4~~--I~-*--~-*.--~.~'-
Gf(J.,;tJ,. p/2""b~
fbi
Fig. 2.16 SOOP Structure on Winkler Foundation

and Moment-Rotation Relationship
---- .. - BONDED CONTACT. UPLIFT PREVENTED

UN BONDED CONTACT. UPLIFT PERMITTED
a
:::I
::I
"- 0
::I
-8
10
"- 0
Ib
-10
Fig. 2.17 Displacement and Rotation

Response Ratios
20
20
1,0
10
o. , oa
~.
o.
0,'
0,'
~i ~1
.
~
''"~"
..
z
W
,..~ 0,2.
0 10 0.10
U
;;:
:::.., OJ.')U 0,09
'".,u
Olio "-
006
:3
...'"
<l
00' U'fi"'U~
AT
::0"
!OC.''''luT~uPt.."
Vii
.....'" 0.04
'"
<II
002
.,:r 0.02
..
~
CD
'"'"
"'"
0010 0,011]
0000 0.000
, - -__ ,,~,o-iYfi"PQfffl~ tnT,,,,
0006 000.
Fl.[ItUlL( supPotiTIG Sl$Tt64
0004 l- 'Q11Ii_t!N 4NO f'tl(l*tUlld[It' STS'f(bll 0,004
e.o",olOCOf4f"'C"T, U~I'" f'll1t'J(Hf(():
J --- "I'#tf\,{tt
~oHOlO
'Vi f(tf
COHT4CT YWLll"f !>1""I4ITT (0
OOOl OJJOt :-
q ~~~ .. - - [OUfY41"lkl fW()"(t.(fd(~T '!Y5ftt1;
Ultl!)OWO[G eatlsl'ACr_ UPLII"T P[1UfffT(O
I
4
VIBRATION Pft,QO T. SEC
Fig. 18 Base Shears Coefficient for Fig. 2.19 Base Shear Coefficient for
Different Foundation Types Different Frequency Ratios
22
(m+m) g b (2.21)
BSc
h
BSc (m+m) g b = 1+y (2.22)

BSCt; - - - ==
W hmg a:
At the maximum displacement U max ' the corresponding maximum base shear
coefficient BSC m is,
= BS... (2.23)
BSC".
W mg g
The governing differential equations of motion of the system can be derived by

taking the dynamic equilibrium of the corresponding vibration modes. The resulting differential
equations are highly non-linear. Considering horizontal equilibrium at the free body of the top
mass, including the P-delta effect, yields,
m ii + m he + c U + k u -m u. (A '" ) (u+h6)
'J-m~..v +g -'---":'"
M
(2.24)
K g It
This differential equation of motion of the structural system is now written in

a general form so that it is valid for both full contact and partial lift-off of the foundation. The
reaction R(t) and moment M(t) that are generated by partial lift-off of the foundation can be
written in terms of the contact-width of the foundation d. Substitution of the reaction and
moment to the vertical and moment equilibrium equations of the structural system leads to,
(2.25)
(2.26)
23
where kwl and Cwl are the soil-foundation stiffness and damping per unit length and I: J = dlb,
where d is the width of paltial contact of the foundation. Value of 1:1 and 1:2 depend on whether
the foundation is in full contact or under partial lifted-off. The complete derivation of the above
two equations may be found in Appendix B and the modal calculation procedure in Appendix
C of Ref.C3.
Response of the structure subjected to the N-S El Centro 1940 earthquake can
be found by the numerical solution of the equations of motion of Eq. 2.24, 2.25 and Eq. 2.26.
The displacement and rotation response ratio for a relatively slender structure with 0: = 10,
'Y= 0, f3 = 8 ({3 = wlw w = Vklm , W = V(2 kwl b)/(m + mJ ), I; = c/(2 m w) = 0.05,
y , y
tv = 2 Cw1 b/{(2 ww(m + mJ} = 0.4, with fundamental period T = 1 sec is shown in Fig.
2.17. In full contact, the structural response is dominated by the first mode, but the response
of the structure during foundation lift-off is very complicated due to the presence of higher
frequency modes.
Effects base fixity or lift-off on the maximum base shear coefficient ESc.n is
shown in Fig.2.18. The BSCm of the structure on the Winkler foundation is always smaller than
that on the rigid foundation, especially where T > 0.4 second. The base shear also reduces
when the foundation is allowed to lift-off. The structures with shorter periods of free vibration
( smaller T ) have a greater tendency to have lift-off of the foundation.
The effect of the slenderness ratio 0: = hlb, frequency ratio {J, mass ratio 1',
and P-delta on the maximum base shear coefficient were also presented in the analysis. The
effect of frequency ratio (3 on the maximum base shear coefficient is illustrated in Fig. 2.19.
For smaller values of (3 (smaller value of kwl or relatively soft soil), a smaller maximum base
shear coefficient is found. The final conclusion of this investigation is that, there is a great
tendency for structures of relatively long fundamental period to develop a smaller base shear
when compared to the static shear value at incipient lift-off. In general, where foundation lift-
off is allowed a smaller base shear is observed when compared to the rigid foundation.
2.3.2.2 Flexible Elastic SDOF System on a Two Spring-Dashpot System
A similar investigation where the foundation was supported by two spring-

dashpot systems was repOited by YIM & CHOPRA (1984). The spring and damping values
24
are assumed to be constant and independent of the excitation frequency. The properties of the
structural system are shown in Fig. 2.20.
Initial static displacement v, , ultimate foundation rotation ()u, base shear BSu,
base shear coefficient BSCu, horizontal displacement U u, critical overturning moment Me, base
shear BS e and the base shear coefficient BSC e respectively, are similar to Eq. 2.15 to Eq.
2.n. The governing differential equations of motion corresponding to the horizontal, vertical
and rocking vibration modes are,
(2.27)
b b (2.28)
(m+m.)v + ~ c V + tEzc(-) hO + ~ kv + tEz k(-)hO= -(m+m)g
It It
(2.29)
m b1...
-" b[E.C (-)h8+F-C
- htl-cu-ku+- b b
V+E1k (-)h8+~k v] = 0
3h 1. h r hA -.z. & S h .. t
where k, and c. are the soil-foundation stiffness and damping respectively. Values of E1 and Ez
depend upon whether the foundation is in full contact or in partial lift-off. Those values can be
found in Appendix B of Reference C3.
The response of the structural system subjected to the N-S component of El

Centro 1940 ground motion was computed by numerical integration ofEq: 2.27, 2.28 and Eq.
2.29 respectively. Those equations are nonlinear as indicated by the dependence on the
coefficients El and Ez. The structural responses-ratios are similar to those of the previously
investigation. During full contact, the displacement response-ratio is dominated by the first
mode, whereas during partial lift-off the displacement response becomes very complex due to
a higher frequency response.
The effect of the slenderness ratio ~ and frequency ratio .13 on the base shear
coefficient are shown in Fig. 2.21. It is clear that these effects resemble the investigation of
the structure supported by a Winkler foundation. When the foundation-mat is permitted to lift-
off, the base shear coefficients are smaller than when up-lift is prevented. Slender structures
have a greater tendency to uplift and have a greater reduction in the base shear coefficient. This
is because smaller ultimate base shear BS u is required when one edge of the foundation-mat is
nm
_ he
I
I
h
P-i
"""'......, ~Ot(qTAC'.
U(I>!JI'1' ""''' ..... ifftO
-~g~liI.c.,..
Ifh,.I,r 1"'l,,"1I1((I.
/
/
/
/
/
"
I
I
I k,c 1M fl[ACiION 'OfiC[--OJ!:lPLAClW(liIf
~[LAffON "Or4 S'UPPC)A:l'tNG tll':MNf
Ie
" /
I
I
(el WO"[NT-~OTATION ft[LAT10N FOR

FOUNDATiON WAf
Fig. 2.20 SDOF Structure on Two Spring~Dashpot Foundation and

Moment-Rotation Relationship
1.0 r-------------------,
,.& cotf' ~#()'t.'$

,J 0.'
... ...,.
~
u
...u 0'0
COX'f', Y"
;;:
... ...;;: Q,O";
D03 i-
8'"
W
"..."x
00'
.S'" 0,04
..,.
w oll.
w
~
., Oo~
..
w
.,
0.010
0,000
000.
0004 0004 rLXIIH. ~uPPQAfUl(; SY~T[Jd

- - - ' II" 10. ooqtl{n(~1.r.f.VIlLlrTPl!ltvr"'lr(o - - - IJ. ". fIo.'I(,"-Dmf&CT.Lljt ..J'I"~"t"UtlD
- - .. fOllJ~tJ.tDCQQlnC',~P'lI'l'p"Nln(a - - D 4. \,IMIliMiOo(OCONT,tr.f ,IJ""I" YUlHlfl(.O
........ " a I 9., eowMDCDI!II"'(.',un,'''-f'''tV~TtI Q.002 .......... ' U . CQIItOI[OCOUTaef.u.P\.I" VIllIWIlITI(O
0.001
-~~-- f.j. 'I~ lJi='IM~D{IH.DWf'U,~f'\..'l'r P{(H.tITTlO - - _ .... /:J. rt, UWaOtiMOCOiIfToI.U.U"l"T fl(Fl"'lntO
<I' o.t 0.4 D.B 0.6 t ~ I)' ot 0,4 0.606 ( t .-

VUHtATIOM PERIOO T.. !ire. VHlRATIClH PERIOD T, SEC.
1 Base Shear Coefficient for Different Slenderness and Frequency Ratios

26
at incipient lift-off (Eq. 2.17). In general, the conclusions of this investigation are similar to
the previous investigations. Because foundation lift-off provides positive effects in reducing the
structural deformations and forces, foundations should be allowed to lift-off.
2.3.2,3 Flexible Elastic MDOF Structure on a Two Spdng-Dashpot System
The response of a flexible N-floor system to the vertical and rocking vibration
modes was investigated as the continuation of the two previous studies. The structural system
was supported by spring-dashpots at each edge of the rigid rectangular massless foundation as
presented in Fig. 2.22. Spring stiffnesses and damping coefficients are assumed to be
independent of the excitation frequency.
The N +2 governing differential equations of motion can be determined by

considering dynamic equilibrium of the corresponding vibration modes. As in Ref. C6, the
structure is assumed to possess the classical normal modes of the building on a rigid fixed base
foundation. The horizontal floor displacement vector u(t) due to structural deformation of mode
i, therefore is conveniently defined as,
1_
(2.30)
u(t) == L VP) ci>l
1=1
By introducing Eq. 2.30, and pre-mUltiplying by ~T and utilizing the

orthogonality properties, the uncoupled governing differential equations for horizontal, vertical
and rocking modes are respectively,
(2.31)
(2.32)
m,v + R(t)
l=l'I (2.33)
L LtYi + Ib 8 + M(t) == - L: (;/t)
1=1
27
(2.34)
j=N (2.35)
Lr : : Em) ~iI'g)
}=1
The overturning moment M(t) and the vertical reaction R(t) are similar to these
of the Winkler foundation model described in Section 2.3.2.1. Eq. 2.31 consists of n
uncoupled differential equations of motion. This equation can be transfonned to a single
differential equation of motion by pre-multiplying by ~h)2/(MlL.h) and introducing,
/;n (2.36)
u(t) = L Yt- (t) .p;
1=1
where,
(2.37)
and
(2.38)
The transformation of Eq. 2.33 can be carried out using Eq. 2.37. Finally,
Eqs. 2.31 and 2.33 leads to ,
(2.39)
1=,. (2.40)
L mt htit + If) a + M t = -L: Ug(t)
14
lithe foundation rotation term in Eq. 2.39 is transferred to the right hand side,
then this equation will represent the equation of motion of a single degree of freedom system
28
U~!ilR(s.no
PO$ITIOr~ - - - - -
--"
Fig. 2.22 MDOF Structure on Two Spring- Fig. 2.23 Equivalent SDOF Structure
Dashpot Foundation
- nEXIBl[ SUPPOR.Tlp.lCi SYSTfJA.lJPliF"T pER/AITTEO

-- - - RIGID SUPPORTING; SYSTUI. UPLIFT PREIIEPolTto
~c " '" I
tI
~o,"JO
0,.
0,.
0,'
0,'00
.,,"
III IalOO[ R(SPo.'-lS(
,~ /\ l\ [\, 0.>
\!\i\j_l" z
'" 0,1
:::.., 0".
D
U
E
000
..8'"
a: 0.<>1
"VI
:::.. 0,02
" 001
0,OQ8
0.000
" "'
0.,.,. 'LI.IIL JUlQIfJIUO Ul1,)1
-0.010 ___ oo~el~c"..f"'~.~'"""("IU.r'D
_ .... I<>.. D"rD~""'UCI ...,...'-'fIJI .... TTII>
I~
~ lIl~dJ'~\'~rli"'~' ~... IiJ~~~
B.'" MCO: 11E,5PONSE.
0.002
'" -0.00' L_=~'V'~~_~~~_III,~,~~_1I11~1~_"'~c'_J'''~"'_~~:,,_.I'.-_:r,"~'I,

o 0
o 0.2 Z 4 20
T)lt{E. SECQNOS fUNDAMENTAL VII3RATION PERIOD T] SEC.
Fig. 2.24 Mode Contributions on the Base Fig. 2.25 Base Shear Coefficient of the
Shear Coefficient MDOF Structures
10 I RIGID SUPPOATING SYSTEM Ibl FLEXIBLE SUPPORTING SYSTEM 10' FLEXIBLE SUPPORTING SYSTEM
...n. J UPLIFT PREVENTED BONDED CONTACT, UPLIFT PREVENTED UNBONDEO CONTACT, UPLlF T
PERMITTED
0.<
i
~.; (l.1
8~
~
8,
;",
...
-
0_04
0......
.......
CfUTtCAL BASE SHEAR
o~~~<=O~F~F~IO~IE~Nl~"~<~~--~~--------~---------------*'~---------+------------==~~--------~
,-~
\');;,.. 2: IJOO(S
;.~,
\ \
\.f~1
'i-V
w IUOOE "'. \ .t4()P( ..........
~~ ~ ~
=
000'
\ ~\
'.
FLINDAMENTAL 'VIBRATION PEfUOO T SE.C
Fig. 2.26 Modal Contributions on the Base Shear of MDOF Structure

29
with mass m" and height h' as shown in Fig. 2.23. However, these modal equations are
coupled through the unknown base rotation 0 and vertical displacement v. Hence, the N +2
equations (Eq. 2.32, Eq. 2.39 and Eq. 2.40) must be solved simultaneously to determine the
unknowns Y-+(t), OCt) and vet) respectively. Solution ofthose equations may be carried out by
using numerical integration for a specified ground motion. The displacement and horizontal
inertia force at floor j are,
1=<1 (2.41)
up> == E Yi~(t) ~;
1=1
i=i<l (2.42)
Fit) = E fI)/' mJ ~UY4~(t)
4=1
Critical values of the base shear BSc are similar to Eq. 2.21 if Il\ replaces (m
+ mo)' Responses of a ten-story building with 5 % critical damping ratio, slenderness ratio O!
= 10, massless foundation m.. = 0 subjected to 1940 N-S El Centro earthquake allowing for
the effects of foundation lift-off, foundation flexibility and the effects of higher mode
contributions on the structural response were investigated. The base shear coefficient shown
in Fig. 2.24, in which the higher mode contribution have been plotted to an amplified scale.
The base shear coefficients were dominated by the first mode as found in previously
investigations. As in the earlier investigations, lift-off results in smaller base shear coefficients
than when lift-off is prevented, except for fundamental period Tl = 4 to 9 seconds as shown
in Fig. 2.25. In these longer periods, lift-off results in greater base shear coefficients than when
lift-off is prevented because the higher vibration modes contribute significantly to the base
shear. Again, foundations of the slender structures have a greater tendency to lift-off.
As shown in Fig. 2.26, the contribution of higher modes of the buildings

supported by flexible foundation are more significant than for those buildings with a fixed base.
In this case, the mode shapes and ratios of the natural vibration periods are independent of the
fundamental period T1 The increased contribution of the higher modes is due only to the
relative values of the response spectrum coordinates, which in tum depends on the spacing of
the vibration periods. This shows that the contribution of higher vibration modes becomes more
significant in a more flexible structureA2 ,AM4. The simplified analysis method considered only
the contribution of the first mode.
30
2.3.3 Psycharis (1932?5,P6,P7
PSYCHARIS (1983) presented the dynamic behaviour of rocking structures where the
foundation~mat was allowed to lift-off. The rigid block and flexible superstructure were
supported by either two springs or a Winkler foundation-mat. A vertical acceleration Vg and a
horizontal acceleration Xg applled to the system resulting in vertical and rocking vibration
modes. Horizontal translation of the base was usually very small when compared with the
rocking effect, and could be neglected.
Only structural systems with a flexible elastic superstructure will be discussed herein.
The superstructure is a multistorey frame with N -floors and N + 1 rigid masses and is supported
by a two spring-dashpot system as illustrated in Fig. 2.27. The mass of the superstructure was
assumed to be concentrated at each rigid floor, so that there is one degree of freedom per floor.
The spring stiffnesses and damping coefficients are assumed to have constant values and are
independent of the excitation frequency. P-delta effects were taken into account in this analysis.
The corresponding vertical and rocking displacements, as functions of time, are denoted
by vet) and OCt) respectively. The governing differential equations of horizontal motion can be
determined by taking the dynamic equilibrium of the corresponding vibration modes,
(2.43)
M ii {- M he {- M ge {- c u {- K u -MiK
where u is floor horizontal displacement of the superstructure.
Taking the dynamic equilibrium in the vertical and rocking modes during full contact
leads to the equations,
(2.44)
- m t Vg
(2.45)
where,
31
m
-------- -- (II)
---.]-
I
Hn Kfl 0.. I I
I
--_. __ ..... -
mil_I 2 '.2
tn-II ,
I
I
lun"
Pn
I
Pn- I I
Ki'l Ci.a
I
-2-'2 I
i
h:~r r
0)
K, C'
I I
_ I .:::::.L I 4d
I
2 :2 t--=-I/
I I
n, P. I l
III
c ~ ~.s.. I
ml 2 :2
(I) I
f I 6,'I
hi S,S I
Z 2 I
(OJ
Fig. 2.27 MDOF Structure on Two Spring- Fig. 2.28 Free Body of the Rocking Structure
Dashpot Foundation
10,0 WITH LIFT -OFF 0,8 WITH LlF T -OFF
5,0
0,0
-5,0
l-
I
~IO,O APPROXIMATE SOLUTION C)
w
0,8 APPROXIMATE SOLUTlO~
Z 5,0 ~ 0.4
o fl:
f; 0,0 I--"""",A+4-4'-J,,-J'-lf- ~ 0,0 1--1lr+-+-..jI-:-.fr-.jJ.:...!-ttt4~++++t-A4-\-A,~"'"
YJ I
.;! . 5,0 V1 -0.4
w w
o -10,0 '-----'-_-'----'-_J......--'-_.l.......""""'-_..I...-.....J.---J '1l -0,8 '---'-_'----'-_J......""""'-_..I...---I._-1---'_-l
D
WiTHOUT LIFT-OFF O.B WITHOUT LIFT-OFF
0,4
0,0
-0,4
t_-,

! I I I t
-10,0 ' - 0, 8,'=-0---':~..L--L_..J'---'----'-_l....-.--'-----':--'
1,5 3,0 6,0 7.5 9,0 10,5 12,013,5 15,0 o 1.5 3,0 4.5 6,0 7.5 9,0 10,5
TIME (sed TIME!sed
2.29 Roof Displacement Fig, 2.30 Base Shear Coefficient

32
j=N (2.46)
"', = Emj' j = 1,2."N
j=O
where Vg is the ground vertical displacement, mj , hj and Ij respectively are the mass, the height
and the centroid moment of inertia of floor j. Meanwhile k.f and C.f are spring stiffness and
damping coefficients of the soil-foundation system, h., is the height of the centre mass of the
superstructure and ~ is the total mass of the structure.
During lift-off of the foundation, Eq. 2.44 and Eq. 2.45 become:
(2.47)
(2.48)
j=N m$ m$
~-u = - -~-mih
L.J 2 j 2 r-g c
j=l
Eq. 2.43 consists of N-coupled differential equations of motion. Analytical procedures

using Laplace Transforms were used to solve Eqs. 2.45, 2.47 and 2.48 resulting in a very
complicated equation. Eq. 2.44 was solved directly using Duhamel's Integral. Approximate
solutions considering only the contribution of the first mode in the case of an undamped and
a damped structure was presented. In this case, it is assumed that the classical fixed base mode
shapes exist in the structure.
A numerical example for a ten-story building subjected to the Pacoima Dam S61E
earthquake ground motion was carried out. Damping ratios were assumed to be 6.5 % for the
first mode and 5 % for the other modes. The differential equations of motion were integrated
numerically using a Runge-Kutta method, taking the first five-modes only. The roof horizontal
displacement relative to the base and the base shear coefficient history are shown in Fig. 2.29
and Fig. 2.30.
Fig. 2.29 shows that the approximate structural response was close enough to the total
response, and that the structural response is dominated by the first few modes. Meanwhile, a
very short transient response exists in the base shear coefficient as depicted in Fig. 2.30. The
33
maximum base shear coefficient where the foundation is allowed to lift-off, in general, is
slightly smaller than when foundation lift-off is prevented. Other results of this investigation
are that the apparent fundamental of period of the system will increase with increasing lift-off
and that the contribution of the higher modes on the structural response are not significant. The
final conclusion of this investigation was that, in general, it was not evident whether base-mat
uplift was beneficial to the structure. The result depends on the structural parameters of the
structure and the characteristics of the ground motion.
2.3.4 El-Hifnawy and Novak (1986)E2
The elastic response of five and ten-story shear buildings supported by pile foundations
in which the pile caps were allowed or not allowed to lift-offwas reported by EL-HIFNAWY
& NOVAK (1986). Three pile foundations were considered; a), pile heads are fixed to the cap
and pile tips are socketed, b), pile heads are fixed to the cap but the pile tips are end bearing
and c), the pile heads are not rigidly connected to the cap or the connection is secured only due
to the dead load of the building. These three pile foundation types are shown in Fig. 2.31.
For socketed or floating piles, the stiffness, damping coefficient and pile reaction was
the same when the pile was in tension or in compression. For end bearing piles, the pile
tension reaction depends only on a relatively small of pile skin resistance. In the case of free
head with end bearing piles, the pile does not have the ability to provide a tension reaction. The
three types schematic of pile vertical stiffness Ie.. and damping Cy are shown in Fig. 2.33.
In this analysis, the S90W horizontal component of San Fernando Valley earthquake
record scaled by factor of 1,2 and 4 were applied to the structural system. Horizontal, rocking
and coupled horizontal and rocking vibration modes were considered. Vertical vibration modes
were not considered in the analysis. The governing differential equations of motion can be
determined by taking dynamic equilibrium of the corresponding vibration modes of the
mathematical model shown in Fig. 2.,32. Taking the dynamic equilibrium in the horizontal
forces of the elastic superstructure yields,
(2.49)
M U + M ii" + M h V+ C Ii + K u - M U,;<t)
34
Dynamic equilibrium for horizontal displacement and rocking of the whole structural
system leads to the equations.
(2.50)
(4.51)
J=N J=N (2.52)

nat = m!J+Em t
j=1
, It = I[) + Em} h/
}=1
The terms kllb. chh , ~f and cff are the stiffness and damping coefficients of the pile for the
horizontal and rocking modes respectively. The terms khf (~h = k hf) and C bf (cfh = C hf )
respectively are the pile stiffness and damping coefficients due to coupling between horizontal
and the rocking modes. The mass and height of floor j is denoted by mj and hj , ub and 1J; are
the translation of the base-mat in addition to the free-field horizontal motion and the base-mat
rotation respectively. Stiffnesses and damping coefficients of the pile were calculated according
to Ref.N3.
The Wilson-OcE numerical integration technique was used to solve Eq. 2.49, 2.50 and
Eq. 2.51 with a time step of 0.02 and 0 = 1.37. In general, the story displacement of the
building supported by floating piles is greater when compared to the building on end bearing
piles, except for the foundation with a 15 pile rigid connection. A consistent relationship was
found in the case of pile forces, where the pile forces in floating piles are always greater than
those of the pinned end bearing piles. The story displacements generally increase if the pile cap
is not rigidly connected to the pile head, except for the building supported by 27 pinned
floating piles. It is clear that, the type of pile cap connections, pile tip conditions and the
number of piles significantly affect the building response.
Base shear coefficients for the building supported by end bearing and floating piles is
shown in Fig. 2.34. The relationship between the base shear coefficient and number of
supporting piles in both cases is not clear. The rigid connection of the pile caps may reduce
or increase the base shear, depending upon the number and types of pile. Reduction of the base
shear in the case of rigidly connected socketed piles may appear surprising because in a linear
. ','.
, -Rr
...
, ' .. .
'
. :.
tol fiXED HEAD I bl ,,"IKED HEllO Itl mEE HEAD

SOCKt':TED PILES END BEARING PILES END BEARING PILES
of Pile Foundations 2.32 Mathematical Mode1
eV kyv (\I kyv
ell
-- - --, "Ay
L
y
Tip UPlif,,, tv ey
\I II
Ysiol
+
(0) Sockeled (]I Flooling Pile
Flooling
(b) Endbeoring Pile>

Endbt'Clring PilI"
Eliminated
'" Pile Acting
ec) EndbearlnQ or Floatlnll PilE'
Fig. 2.33 Pile Force-Deformation Relationship
TEI\ISION AlLOWED {RlGttJ CONNEC.TJO~" TENSION ALLOWED (RK)IO CoNNECTION t

T(NSION ooT ALLOWED {OO CONNECTfOO} TLN5!OfII NOT ALLOW[O
2000
...
'"
;
,.:.
i"
800
400
0
(0)
l' I. 24 21
!
NUMBEr:?
PINNFD END BEARING PILES PINNED ELOllTlNG PILES
2.34 Base Shear for Different Pile Types

36
analysis an increase in the stiffness usually results in an increase in the seismic loading.
Finally, the effect of the foundation mat lift-off may reduce the seismic loading but also
substantially increase the compressive load on the piles. In the case of socketed piles with a
rigid connection, the building seismic base shear may also increase or decrease. Many factors
such as the number of piles, the size of the building, the type of pile cap connections and the
intensity of the ground shaking appears to affect the building response.
2.4 SUMMARY AND CONCLUSIONS
Effects of foundation rocking on the response of elastic superstructures subjected to

either harmonic or earthquake excitation have been reported by the aforementioned researchers.
Rocking of the foundation may result in a partial lift-off of the foundation-mat or maintain full
foundation contact. All of the researchers assumed that the circula~6 or rectangular
foundation C7 ,H8,P4,C3,C4,S9,slO was perfectly rigid and rested on an elastic half space. The
foundation was supported by a Winkler foundation eJ ,CA,P7 system, equivalent two spring-dashpot
systemC6,S9,P6,F1, rubber pad materials El ,P4, a spring systemH8 , or a pile foundation systemE2 . In
the case of a spring-dashpot foundation system, the stiffnesses depended on the excitation
frequencyC6 or was frequency independent, both cases being modelled by a linear spring
system.
The superstructure was a single degree of freedom systemC3 ,CA,EJ.F4, or a multi-degree

of freedom building systemC7 ,E2,HB,SIO, The structure was a concrete masonry shear wall
structureEJ ,P4, a steel structureC1 ,E2,H8, or not defined C3 ,C'A,s9.slo. All of the analyses assumed that
each floor has only one degree of freedom, with the superstructure response being either non-
linearc7 ,H8 or linearly elasticc6.c7.E2,P6. To simplify the solution of the problem, the response of
the linear multi-degree of freedom building was calculated by approximate methods, using
either fe-normalized as an equivalent single degree of freedom systemS10 , or considering only
the contribution of the first or the first few modesC6,PS,P6.P1. Solution of the governing
differential equations of motion was carried by analytical procedures P6,P7 and by numerical
integration techniques. The investigations were experimentalC7 ,El,P4, an experimental and
analytical investigationHB , and analytical investigations onlyC3,C4,c6,s9,siO.
The experimental and analytical dynamic soil-foundation responses indicate that the
foundation mat may lift-off during earthquake excitation. The superstructures are generally
37
assumed to be elastic and only a few went into the inelastic range C7 Hll The superstructure
responses of the building may be categorized by three types of result. Firstly, partial lift-off
of the rocking foundation, in general, gave a positive effect on the building
responseC3C4C6E3HB.P4.S9.SlO. Secondly, the effect of a partial lift-off of the rocking foundation
on the superstructure response was not clea~P7. Third, partial lift-off of the foundation may
increase or decrease the superstructure response.
Chapter III
Conceptual Overall Design, Preliminary Analysis, Design
and Evaluation
3.1 INTRODUCTION
For many engineers and students the design of reinforced concrete structures covers
only the estimation of the size of the building members, the calculation of the member strengths
and the amount and distribution of the required bar reinforcements. However, these steps are
only a part of the complex Earthquake Resistant Design (EQ-RD) processB17 .
The conceptual methodology for EQ-RD is an essentially trial and error iteration
procedureB17,BI8,F6,T18consisting ofthree main interacting aspects involving both architectural and
structural engineering tasks. Firstly, Earthquake Ground Motions (EQ-GM); secondly, the
Building Conceptual Overall Design (BC-OD) and thirdly, the Building Conceptual
Methodology for Numerical (BC-MN) Earthquake Resistant Design. The last aspect involves
preliminary analysis, preliminary sizing and reinforcing, modelling of structural system and
analysis for the preliminary design of the building. The flow-chart for the Earthquake Resistant
Design (EQ-RD) of a reinforced concrete structure is shown in Fig. 3.1.
The first aspect covers the selection and processing of the earthquake ground motion
for both serviceability and safety limit states. The results are presented in the form of Smoothed
Linear Elastic and Inelastic Design Response Spectra denoted by SLEDRS and SIDRS
respectively. The selection of the earthquake ground motion and the methods of structural
analysis will be discussed in Chapter VIIl.
This chapter discusses the building overall conceptual design, preliminary analysis,
member sizing and reinforcement. Modelling of the sub-structure and superstructure will be
described in Chapters V and VI respectively.
38
39
It is necessarily to recognize the relationship between the analysis and design for
identifying all aspects of the earthquake resistant design. This basic design equationD17
expressed as,
Demand Supply (3.1)
The demand and supply covers all structural propelties such as stiffness, strength,
energy absorbtion and energy dissipation. Determination of the demand can be carried out by
the numerical analyses of a mathematical model of the entire building system. Prediction of the
supply is carried out during the preliminary design procedures. Evaluation of the demand and
the prediction of the supply, in general, are not straightforward because of many uncertainties
such as the critical seismic excitation, the entire building system, and the difficulty in predicting
the real structural propertiesB18
3.2 BUILDING CONCEPTUAL OVERALL DESIGN (BC~OD)
The Building Conceptual Overall Design (BC-OD) involves four main steps which are
the selection of the building configuration, structural system, structural material and type of the
non-structural elements.
3.2.1 Building Configuration
An important aspect of the conceptual design is the configuration of the building. The
configuration can be interpreted as the size, shape, and the proportions of the three-dimensional
form of the buildingAls The design includes the shape, size, location and orientation of the
structural elements such as frames and structural walls as well as the non-structural elements.
The architect is a main palticipant in the seismic design because the earthquake excitation acts
on the whole building and does not distinguish between the structural system and the non-
structural elements. The structural engineer is in a position to advise the architect on the most
desirable location and orientation of the structural elements, because the configuration of the
building will influence its response under dynamic loading A17 ,P18. Factors to be considered
during the design of the building configuration are follows.
40
1. The Building Plan: Defining of the building plan is the first step in the building design.
Simple and regular building plans are preferable such as symmetry in one or two horizontal
axes. Building plans with T, L, U or Z shapes should be avoided or else subdivided into
simpler forms. Lack of symmetry may lead to a significant torsional response during the
earthquake excitation. It is also preferable that the building should not too long in plan so as
to avoid the possibility of different soil responses along the building foundation during the
earthquakes.
2. Height to Base-Width Ratio: As the building height Hb to the base-width Bb ratio increases
the outer columns and foundations will receive the higher overturning forces during seismic
excitation. This is because the buildings has a reduced resistance to overturning moments. The
higher the ratio the larger will be the inertial effects on the soil-structure interaction.
3. Lateral-Load Resistance Systems: Lateral-load resisting systems of frames, structural walls

or a combination of frames and walls are usually employed. The location and orientation of
these system should be made so that, at every floor level, a symmetry in the lateral stiffness
will be achieved.
4. Distribution of Mass and Stiffness: The distribution of the masses and stiffnesses are ideally
uniformly distributed in both the horizontal and vertical directions. Increasing the floor mass
with the building height will result in greater overturning moments and sensitivity to higher
mode effects. The vertical distribution of the inter-storey stiffnesses and strength should be
made such that no soft-storey effect exists in the building.
3.2.2 Structural Materials
Reinforced concrete, steel or timber may be used for multistorey buildings. In this
study, however, reinforced concrete structures are considered. A relatively low concrete
compressive stress fe = 25 MPa is used. A nominal diameter of 28 mm (D28) for the
reinforcing bars with a specified yield stress fy = 300 MPa is used for the reinforcement of all
beams, columns and walls. A similar steel yield stress with a nominal diameter of 10 mm
(DlO) is used for the floor slab reinforcement. The concrete Poisson's Ratio Yc = 0.19 is also
used. The concrete elastic modulus Ee and concrete shear modulus G c can be computed as,
41
EQ-RD
,
PRELIMINARY DESIGN
I
f t
, ,
..
BC-OD EQ-GM
I
.....
Planning: Build. Config., I

Processing Data: Acuisition Data: Time
Str. Material, Non-Struct. EDRS and IDRS ~ History of EQ-GM, for
Elem. and Struct. System Different Limit States
-- I
F-,
... 1
t
I PRELIMINARY DESIGN PROCEDURE
t
Preliminary Analysis: Tl, Preliminary Sizing: Modelling of Structural
Design Criteria, Gravity Member Sizing and System, Energy Dissipa-
--- ~
,
Load, Lat. Forces, and Reinforcement tions and Method of Str.
Member Actions Analysis
I
I
~ AnalysIs of Preliminary Design ()f the BUilding.

! Structural Responses and Acceptability of Design Criteria
Check
Acceptability
Not Accepted
Accepted
FINAL DESICJN
Earthquake Resistant Construction (EQ-RC~

Fig. 3.1 Flow-Chart for Earthquake Resistant Design (EQ-RD) (819)
42
(3.2)
Gc (3.3)
2 (1 + v~ )
where Ee and fc are in Mpa.
For fa == 25 MPa, the concrete elastic modulus Ee and concrete shear modulus G c are
23500 Mpa and 9874 MPa respectively.
3.2.3 Non-Stmctl.lral Elements
The majority of the analyses of moment resisting frames assume that the structural
frames are open so that they have no infill elements. In reality, non-structural elements are
nearly always present in the form of masonry or brick walls, partitions, doors and windows.
Frames infilled by non-structural walls are invariably stiffer than the open-frames and reduce
the possible excessive lateral deformations under small earthquakes. In this case it may be
unnecessary to separate the non-structural elements from the lateral-force resisting systems P18
However, under moderately severe earthquakes the damage to the non-structural element is
difficult to avoid and may cause failure of the columns in the infilled frames s28
3.2.4 Building Prototypes
To satisfy the above considerations, a simple plan building prototype as shown in Fig.
3.2 is used in this study. The width of the building plan Bb is 18.40 metres with the building
length depending on the number of frames being used. The building plan is symmetric in both
the x and y-axes where the main lateral-load resisting system consists of parallel reinforced
concrete frames and structural walls. The number of frames and structural walls are adjusted
to provide the different responses for each structural combination. The distance between the
parallel frames and structural walls is 9.20 metres as was used in previously studies J6 ,G7,G8,G9.
The earthquake excitation is assumed to act parallel to the main lateral-resistant systems.
A two bay symmetry frame with beam span of 9.20 metres is used in the regular in-
plane frame as shown in Fig. 3.3.a. The in-plane wall-frame consists of a single structural wall
43
centrally placed between two columns as shown in Fig. 3.3.b. The building prototype has a
typical floor height hj of 3.65 m. The first storey height for the LFlex, LEx and SCFlex
(hi +hJ building is 5.0 m and is 3.65 m for the SFix bUilding.
The ineltial interaction will be significant in relatively high building structures. For
this purpose, the 12-storey and IS-storey building prototypes are used in this study. These
buildings have a total height Hb of 45.15 and 67.05 metres corresponding to H.,IBb ratios of
approximately 2,45 and 3.64 respectively. The secondary and transverse beams are considered
to have the same size of 0.35 by 0.70 m with the slab thickness of 0.14 m.
3.2.5 Structural Systems
Three structural systems were considered in this study; first, the building prototypes
were assumed to be fixed at the ground surface as shown in Fig. 3.4; second, the building
prototypes were assumed to be fixed at the footing level as depicted in Fig. 3.5 and third, the
building prototypes were suppOlted by footing foundations partially embedded in the soil
system. In the last case the analyses were carried out by considering the effects of the
foundations on the whole structural system and denoted by a structure with flexible foundations
as shown in Fig. 3.3. The response and impedance of the soil-foundation system will be
discussed in Chapter VII. The methods of analysis of the whole structural systems will be
described in Chapter VIII.
3.3 PRELIMINARY ANALYSIS
Preliminary structural analyses estimate the structural member actions under applied
gravity and horizontal loads. The gravity loads usually consist of the structural dead load (DL)
and live-load (LL). These loads are discussed in the following sections.
3.3.1 Building Gravity Loads
Different utilizations of the structure will influence to the magnitude of the building live
load. Office building's are considered in this study. The building prototypes were designed
according to Code of Practice for Design of Concrete Stl1lctures NZS 3101: 1982 and New
Zealand Standard Code of Practice for General Structural Design and Design Loadings for
Buildings NZS 4203: 1984 followed the publication of NZS 4203: 1992. No significant
44
Earthquake
direction
Main
! A
I
t I I Secondary
Column A B beam
y
Fig. 3.2 Plan of the Frame-Wall Building Prototypes
Frame 12 @
"""'--+---1 3 . 85m
G d Ground
faun floor
surface \
---_\_----
(a) Section A-A (b) Section
Fig. 3.3 and Wall-Frame with Flexible Foundation (LFlex)

45
- n
Fra me 12 @
3.6 Sm
Gro und 4
flo 0 r\ 3
2
1\
1
- '- -- - -- -- 1-
O~~~~~~~~
Fixed
Fig. 3.4 ~rarne and Wall-Frame with Fixed Bases on the Ground Surface (SFix)
Fra me 12 @
"'" 3.6Sm
hw
und ==rhl
'\
\
- - --
~ -_\- -- - - - -- 1-=-=-=-= __ Jtil _____
I. ~ , "
"
I.'-
the
Fig. 3.5 Frame and WaH-Frame with Fixed Bases on the Footing Level (LFix)
46
difference on loading effect between NZ 4203: 1984 and NZS 4203: 1992. The building dead-
load, live load and equivalent static ealthquake loads are described in the following sections.
3.3.1.1 Dead Load (DL)
The dead load of the building consists of the self weight of the structural and
non-structural elements such as partitions, brick-walls, ceiling, ducting and piping which are
permanently installed in the buildings. Normal weight reinforced concrete structures of 2.4 tlm3
was used. Floor finishing, ceiling, ducting and piping are considered to be uniformly
distributed over the floor area with unit weight of 0.122 tlm2 , The curtain wall and partition
are assumed to have self weight of a 0.051 t/m2
3.3.1.2 Live Load (LL)
The live load varies with the usage of the buildings and is considered to be
uniformly distributed over the floor areas. Both dead and Jive load are normally specified by
Codes. As mentioned above, the building prototypes are office buildings and according to Table
no.2 NZS:4203,1984, the basic minimum floor distributed live load is 2.5 kPa.
3.3.2 Lateral Seismic Load (E)
Various techniques has been used for simulating the effects of earthquake excitation for
preliminary design of the buildings. These mayan equivalent static lateral load or more
complex procedures using approximate mode shapes and Smoothed Elastic and Inelastic Design
Response SpectraP17 Pl9 The first approach is preferred by most designers and is used for the
preliminary analyses because its simplicity and familiarity in the structural engineersP12 ,P18.
Using Section 3.4.2 NZS 4203: 1984, every building was designed to withstand a total
horizontal seismic force Vb in each direction under consideration according to the formula,
(3.4)
in which,
47
Cd = C R S M (3.5)
where Vb is the base shear, W is the total dead weight of the building, R, 5, M are the
basic seismic coefficient, the seismic risk factor, the structural type factor and structural
material factor respectively.
In computing the base shear force Vb it was assumed that only a fraction of the live
load 0.333 LL was used in calculating the total weight of the building WPlR The difference
in the total dead weight of the plane frames and frame-walls was small and they were assumed
to have the same total dead w~ight.
The basic seismic coefficient is a function of the seismic zone, the subsoil flexibility
and the fundamental period of the structure in the direction under consideration according to
Fig.3 in NZS:4203,1984, while the R, S, and M factors can be determined Table noA, 5 and
6 of NZS:4203,1984 respectively. The building prototypes were assumed to be on an
intermediate subsoil in seismic zone A of NZS:4203, 1984. The value of structural type factor
S for frame-wall structures is not specified in Table 5 NZS 4203:1984, and an S of 1.0 based
on an engineering judgmentG1 was used in this study.
Several formulae exist for the calculation of an approximate value of the building
fundamental periods T1. According to SEAOC (1988), the value of T1 can be approximated
byB20,
(3.6)
where Ct is equal to 0.030 and 0.035 for reinforced concrete frame and steel structures
respectively. Hb is height of the building, in feet, from the base.
BERTERO et a1.(1988) evaluated the current approximate formula and compared with
the observed results from the data recorded from the 1971 San Fernando earthquake in several
existing reinforced concrete moment resisting frame buildings. The variation between 1'1 and
building height are shown in Fig. 3.6 and 3.7 respectively. They shown that the current
approximate formula with Ct = 0.035 represents a lower estimate of the fundamental period
T 1
48
3.00 ... ............... .

3.00 ..
's_ '"
. . ~ ~
p P :: ~o
" Ii
~ 2.00 E
R 2.00
I I
o o
o 1.00 D 1.00
1.00 . . .
1.00 .
. .
. . . . ~ ~,;t # ~ . . . . . . ~
~ .........:.........:.. ..... .; ....... ....... .

~ ~
0.00 ' "
. . .
:~ ~ + :. ~ : ~ ... ~
.
00 100 .\50 200 250 100

TOTAL HEIGHT ABOVE BASE (IN FEET)
TOTAL HEIGHT ABOVE BASE (IN FEET)
Fig. 3.6 Building Period vs Height of the Fig. 3.7 Building Period vs Height of the
Reinforced Concrete Frame Reinforced Concrete Frame
l.$ ................ -r-" -.. G __ - - : ...... ~ - _ .. - - f ................ "'f --- .. -_ ....

~ r-.. . . --.. . f"- ~ ........ -:
r.............. .. . . . . . . . . "r....................r.......... ... . ----. ,t......... -_ . . "':,

I
,
I
J.O l.O ................... --~ ~ -~--
2.:' ................ "' .. f. . . . . . . . . :f. . . . ----..:

:
. . . . r. . .
f. . . . . : :
1~o fr..--f----~ .....}. ..

. -l,f
'"o...
'1:
t.,$
"0
.~ \." : "' . . . . . . . . . r. . . . -.. .. f. --.. . _. -: . . f-.. . . . . -.. . f. . -.. ---.. -fM .
i ! ~
~--.. ~M_~ ....... ;
n. a.. 1.0 ~
....... _ ...... _ .. ....._ .t:...!~_ . . ~_" ..............~ . _
1 .. I . . . . . ~_ ~_ # _ .. .. - - .. - - _ _ .......... - .... i
1.0 , I I , ,
: : : : : : :
0.5 : : : :
fi' .................. p .................. ,. ................... ,. .................. ,. ................................... _ .. .
: :
0.5 , I J I
:
I
:
:
: l
:

:
:

0.0 +--"';':---i't---'j---j-.'_ _-t'o---t'--:-""'1'

0.0 $O,Q 100.0 lW.O 200.0 7$0.0 lOO.O l~.O
500.0 ,100.0 1~.O 200.0 2~O.O JOO.Q 3.)0.0
Height of Structure (feet)
Height of Structure (feet)
Fig. 3.8 Building Period vs Height of the Fig. 3.9 Building Period vs Height of the
Reinforced Concrete Frame Frame-Wall Structures
49
BENDIMERAD at a1.(1991) extended the previous investigation to cover 10 moment

resisting ductile frames, 8 moment resisting non-ductile frame buildings and 5 frame-wall
buildings. The fundamental period was calculated from the data recorded in the 1971 San
Fernando and the 1989 Lorna Prieta earthquakes. The relationship between Tl and building
height for moment resisting frame and frame-wall buildings are shown in Fig. 3.8 and 3.9
respectively. The value of Ct :::::: 0.035 was suggested for the design procedures even though
this value shows a lower bound compared with the results of the investigation. Fig. 3.9 shows
that this suggested value of C t gives good agreement with the investigation results. Fig. 3.7
shows that when the building height is less than 45 m. the value of Tl = 0.1 N represents a
lower bound compared that based on ~ = 0.035. Accordingly> the value of Tl = 0.1 N was
used in this study because it gives a higher value of the basic seismic coefficient Cd- The
parameters used in Eq. 3.5 are given in Table 3.1.
3.3.3 Reduction of Live-Loads and Tributary Area
The self weight of the slabs, ceilings and the live load are uniformly distributed over
the floor area and are transmitted to the columns through the secondary and main beams.
Assuming that the beam strength capacities are greater than the slab capacity, this load is
distributed to the column using the floor tributary areaP18,Xl.
Past experience indicates that the probability of the floor area being subjected to the
maximum intensity of an uniformly distributed live-load decreases as the floor area increases.
The floor slab reinforcement, however, is based on the maximum intensity of the live-load
specified by the Code. The main and secondary beams are designed for a tributary area with
a reduction ofthe live-load Lr when the tributary area is greater than 20.0 m2 , NZS 4203:1984
recommends that for all occupancies of the floor and roof, except for retail store, storage and
garage, the reduction of live-load Lr can be computed by,
JOT At> 20.0 m 1 (3.7)
For the building under consideration,
+-
3.0
(3.8)
~
50
where Rd is the reduction coefficient, Lu is the minimum uniformly distributed live-load

specified by the Code and Ar is the area of the tributary floor under consideration.
Table 3.1 C, R, S, M and Cd Parameters.
Type of structures C R S M Cd = CRSM
1. 12-storey buildings
a. Frame structures 0.075 1.00 0.80 0.80 0.0480
b. Frame-wall structures 0.075 1.00 1.00 0.80 0.0600
2. 1 B-storey buildings
a. Frame structures 0.075 1.00 0.80 0.80 0.0480
b. Frame-wall structures 0.075 1.00 1.00 0,80 0.0600
The base shear force V" shall be distributed over the height of the structure with the
formula according to Section 3.4.6 NZS 4203: 1984.
3.3.4 Member Forces and Capacity Design Procedures
Given the building configuration, the gravity load and the lateral seismic force, the
structural member flexural moments, shear and axial forces can be computed. The Capacity
Design Procedure was used in this analysis in which the corresponding design philosophy and
step-by-step application is briefly described Ref.PI8. The total base shear V" and the
distribution of the lateral seismic load acting at every floor level was calculated from the
equivalent lateral static loads specified by NZS 4203: 1984. 'The following steps are required
to determine the actions of the structural members.
3.3.4.1 Effective Member Flexural Rigidity (EI)
The building design may be based on an elastic or inelastic analysis. Inelastic

response of the structures allows for significant cracking of the structural members due to large
deformations. Test on a full scale 7-storey building indicated that the cracking of the structural
members is not concentrated at a particular section but is spread out along the membersO IO As
the member cracks the flexural rigidity EI reduces significantly.
Other parameters affecting of the beam flexural rigidity is the contribution of

the monolithic floor slab. During an inelastic response the effect of the monolithic floor slab
51
on the beam flexural stiffness is represented by the effective width of a T beam, beff and its
effect in the flexural strength is not necessarily the same. This effect is discussed further in
Section 3.4.1. PAULAY & PRIESTLEY (1992) discussed the calculation of the member
actions in the preliminary analysis using an effective member flexural rigidity EI" taking into
account the effect of the slab. The modelling of the effective member flexural rigidity EI" as
discussed in Chapter VI was adopted from ReLP18 and was also used in the calculation of
member actions in this preliminary analysis.
3.3.4.2 Rerun and Column Forces
As presented by Ref.P1S, the beam moments from the gravity loads are simply
the beam fixed end moments. The column moments due to gravity loads were ignored. The
member actions from the lateral loads acting at floor level were computed by the approximate
method developed by MUTO (1975). This method is similar to the cantilever methods but the
location of the column inflection points has already been tabulated as functions of the storey
number and the relative stiffnesses between the beams and columns at the storey under
consideration. The beams moments are calculated using the equilibrium of forces at every joint.
The total member actions are obtain by superimposing these two load combinations.
3.3.4.3 Moment Redistribution
One of the important characteristics of the capacity design philosophy is that

the superstructure dissipates energy by developing plastic hinges in the members. For frame
structures, the beam sway-mechanism in which plastic hinges occur at the beam-ends and
column bases was selected. The negative moments computed from the combination gravity and
lateral loads are usually much greater than the positive moments. During preliminary analysis,
these negative moments are redistributed such that no significant difference occurs between the
negative and positive moments. The amount of redistribution moments is restricted and is
discussed in Section 3.5.1.
The essential requirement of the moment redistribution process is that the

equilibrium of the total moment from the seismic and gravity loads before and after
redistribution is maintained. During moment redistribution no loss of member demand from the
beam-end moments occurs at all storey levels. By applying the moment redistributions and
52
utilizing the beam moments at the column faces the design of the beam members may be more
efficient than that from elastic design.
3.3.4.4 Strength Definition and Their Inter-relationships
Strength expresses the resistance of the structure, member sections or structural

materials. The material strength or the resistance of the member cross section, however, is not
precisely known but varies about a certain probable limif'lll. The strength developed at different
levels involves the Required Strength (5..), the Ideal Strength (5;), the Probable Strength (Sp)
and the Over-strength (50)' Inter-correlation factors for these strengths are the Strength
Reduction Factor (cp), Flexural Over-strength Factor (CPo), Probable Strength Factor (cpp) and
Over-strength Factor (Ao) and can be found in Ref. PI8.
The behaviour of the structure under dynamic loadings is much more complex
than for static loadings and factors need to be introduced in the application of the capacity
design procedures. Such factors are The Dynamic Moment Magnification Factor (UlnJ, The
Dynamic Shear Magnification Factor (Ul.), Moment Reduction factor (R.,,) and Axial Force
Reduction Factor (R). The definition of these level of strengths, factors and their inter-
relationships can be found at Ref.PI8.
3.4 PRELIMINARY SIZING AND REINFORCEMENT
3.4.1 Design of Beams
All parameters used in the beam design such as minimum distance between the
longitudinal bars in one layer, minimum distance between layers, minimum concrete cover,
minimum bar reinforcement, and ultimate concrete compression strain were taken from NZ5
3101: 1982. Other important parameters are as follows.
3.4.1.1 Effective Flange Width of T .Beams under Positive Moment
The beam cross section with a monolithic slab is T -shaped rather than a
rectangular. When a beam is subjected to a positive moment the floor slab undergoes
compression and contributes to the beam stiffness. The effect of the slab width, beff on the beam
53
strength and stiffness are reported in Ref.CI9, P32, X2 and X3. Ref.PlS reported that the
flange contribution to the stiffness of T or L-beams is typically less than its contribution to the
flexural strength. The effective flange width, berr in T-beam stiffness, as presented in Section
3.3.6.2 NZS 3101: 1982, is adopted in this study. For the building prototype described in
Section 3.2.4, an effective flange width, belT of one-folllth of the beam span was used to
calculate the T-beam stiffness.
304.1 EtIective Flange Width of T~beams under Negative Moment
In seismic-force dominated ductile reinforced concrete frames the plastic hinges

are expected to occur at the beam-ends at the column faces. When a T-beam is subjected to a
negative moment the upper part of T -beam will be in tension and makes little or no contribution
to the beam stiffness or beam strength. The top bar reinforcement at the location of the plastic
hinges will be expected to yield in tension. Moreover, the slab reinforcement which is parallel
and adjacent to the beam may also yield extensively. Accordingly, during large inelastic cyclic
displacement the flexural strength of a T-beams section can increase significantlyl'l8.
Even though some intensive experiments. have been conducted, the exact value
of the contribution of slab reinforcement to the flexural strength of T-beams is still difficult to
predict because it is influenced by many factors PlE PARK & PAULAY (1975) suggest that an
approximation of the flange steel reinforcement within a width of four-times the slab thickness
each side of the web may be considered as contributing to the T-beam flexural strength.
CHEUNG (1991) recommends that the effective flange width beff within which effectively
anchoraged slab bars should be assumed to contribute fully to the negative flexural strength of
T -beam is the lesser of :
a. one-fourth of the beam span under consideration, extending each side from the
centre of the beam section,
b. one-half of the span of the slab transverse to the beam under consideration,
extending each side from the centre of the beam section,
c. where the beam extended to an exterior column, one-fourth of the span of the
transverse edge beam extending each side from the centre of the beam section.
However, a more conselvative effective flange width berr from which the slab
reinforcements contribute to the T-beam flexural strength is given in Section 6.S.3.2.e NZS
3101:1982 and was adopted in this study. The effective flange width berr in each side is four
54
and two times of floor thickness extending from the column face for interior and exterior
columns respectively.
The ideal flexural strength Mi = Mi at the critical section of the T-beams were
obtained using ultimate state design principles that satisfy Eq. 3.1. Mu and 1> are the ultimate
required flexural strength of the beam section after applying the moment redistribution and
flexura I strength reduction factor respectively. For practical purpose, the beams of a series of
floors were designed using the same reinforcement. An example of the beam moments before
and after redistribution and the required beam flexural strength was presented in Ref. J6 and
G7.
3.4.2 Design of Columns
The determination of the column actions according to Capacity Design Procedures has
been clearly defined at NZS 3tOl: 1982. In the case of frame-wall structures the step-by-step
procedures for the column actions is presented Ref. P18. These steps are similar to those for
frame structures except that there is a modification of the dynamic moment magnification
factor.
The dynamic moment magnification factors for frame-walls are smaller than these for
frame structures. This is because the effect of higher modes on the column moment of frame-
wall structures decrease as the floor deformations are controlled by the wallG7 ,G8.G9,P18. A
smaller value of the dynamic moment magnification factor implies no'reduction of the column
moment, or Rn = 1.0 for a column with a relatively smaIl axial load. The calculation of the
column axial load and bending moment are obtained from steps 9, to and 11 in Ref, P18. An
example computation of the column actions is given in Ref. J6 and G7.
Square columns reinforced on four sides were used in the preliminary design. The size
of the columns and number of bars in the strong and weak axes are arranged such that the
columns provide a satisfactory strength and the reinforcement content is with in the code limits.
For convenience, the columns over a series of stories have the same reinforcement. Since the
plane-frames are symmetric, the two outer columns have the same reinforcement. The load
combinations between maximum column axial load Puma or minimum column axial load P umi
with column moment Mu have ben considered. The combination between Purni and Mu governs
in most cases.
55
3.4.3 Design of Structural Walls
During a small intensity earthquake the column bases of the frame structures are
designed to be able to resist the overturning moment and shear without excessive stress. Under
this condition the use of the structural wall in the building is uneconomic. Under larger
earthquakes, large member deformations at the critical sections will occur folloWing the
development of the plastic hinges. The main aim the structural wall is to control the plastic
deformation of the structural membersPlB
Stiffness, strength and ductility are three important structural properties required in
seismic design of reinforced concrete structures. A structure with an adequate stiffness will
limit lateral deformation to tolerable levels. A structural wall is one of the lateral resisting
system possessing stiffness, strength and ductility, if appropriately designed. Aspects of the
design is discussed in the following sections.
3.4.3.1 Shape and Section of the Structural Walls
Various structural wall sections can be used as presented in Ref.PI8. When the
earthquake forces become significant the shear strength and stability requirements may govern
and a minimum wall thickness is necessary to avoid premature out-of-plane lateral buckling.
BERTERO (1980) investigated the effect of the shape of the wall cross-section on the wall
stability and energy dissipation capacity. It was found that when compared with the rectangular
shape the barbel cross section or a wall with boundary elements provided higher stability
against premature lateral buckling and dissipates more energy with a stable hysteresis loop over
a wide range of displacement ductilities. The stability against out-of-plane lateral buckling is
the critical mode as discussed by PAULAY & PRIESTLEY (1992). The use of the barbel
section to increase the wall stability was also suggested by Ref.X2 and X3 and such a cross
section was used in this study.
3.4.3.2 Ductility Relationship in Structural Walls
During a large earthquake the structure might sustain damage but must not
collapse. Thus, the structure must sustain large plastic deformation without loss of strength.
The curvature and displacement ductility are two parameters describing the wall plastic
deformations. The displacement ductility Jl,tJ, is related to the curvature ductility Jl,q,. In a wall,
56
20
.?~ 18
::1.
".. 16
~<- Il

5
Q:
..
0-
12
@ 10u
::0 8
C)
.2'"
<-
6
a
C
a 2
a
a 2 t, 6 8 10 12 /l, 16
Wall Height to Length Ratio (hw/1w =ArJ
Fig. 3.10 Curvature Ductility vs Wall Fig. 3.11 Out-of-piane Buckling Model of
Height to Length Ratio the Structural Wall
Curvature
(a) Moment curvature relationship (bl First.yield (cl 'Ultimall?' curvaturl?

curvotUN'
Fig. 3.12 Definition of the Curvature Ductility
0.10
--
0.08
0.06 '---4-..9'~
br/lw
o.ot. I------+_"""l=--!--+--i--,-:::-;-II-
!
0.02 c-! -+---+__0_0'--+-
b ... b w b bcfw
, ~Tiib
b]
b,,, if-
b,,, ",/16
Fig. 3.13 Minimum Dimension of the WaWs Fig. 3.14 Critical Wall Thickness vs the
Boundary Element Wall Height to Length Ratio
57
the parameters effecting this relationship are the wall length lw and the wall plastic hinge length
lp. The relationship between curvature ductility Iloll and displacement ductility /-LA can be
expressed by the approximate formula PIO ,Pl8,
(3.9)
The variation between the curvature ductility demand Ilq. with the wall height-
length ratio for any required displacement ductility /-Lt;. of a cantilever wall based on the plastic
hinge length in the range of 0.30 < l/lw < 0.80 is presented in Fig. 3.10.
3.4.3.3 Stability of Structural Walls
Instability of the structural wall occurs when parts of a thin wall are subjected
to large compressive strains. To avoid unexpected premature buckling an appropriate wall
thickness bWb normally should be taken as not less than one-tenth of the height of the first
storeyP1S,P33. To evaluate the buckling mechanism and the critical web thickness for structural
walls PAULAY & PRIESTLEY (1992) presented a simplified model analysis as shown in Fig.
3.11. Based on the theory of stability, out-of-plane buckling of the waH will occur when the
wall thickness bWb is less than the critical wall thickness b e PlB ,
b
wb
~b c
= /* ~
~8TP
(3.10)
where 1* ~ 0.5 lw is the length of the buckling, lw is the wall length, em is the predicted
maximum steel strain, ~ is the axial load eccentricity and fJ is the effective wall thickness.
The out-of-plane buckling is associated with an axial load with an eccentricity

e = ~ b Wb and the neutral axis is located at a distance of {3 bWb from the wall surface as shown
in Fig. 3.11. The in~plane neutral axis is located at a distance of {3 lw from the boundary
element surface. Using m = /-Lq, y from the curvature ductility relationship shown in Fig. 3.12
and Cu = {3 lw from the wall cross section, the maximum predicted steel strains esm can be
expressed as,
58
(3.11)
Ref.P18 stated that ~ = 0.333, {3 = 0.80 for the two-faces of the longitudinal
wall reinforcement, 1$ = 0.5 Iw and t{>y = 0.0032/1w can be taken as an approximate values.
Substituting these values in Eq. 3.11 and re-substituting in Eq. 3.10 leads to,
Jor !l = 0.80 (3. 12a)
for p = 0.50 (3.12b)
The value of J.I.<b can be determined by using Fig.3.10 for a given value of wall
height to length ratio hwllw. When the waH thickness bWb is less than the critical wall thickness
according to Eq.3.12 the boundary element as shown in Fig. 3.13 with a cross section area of
Awb should be provided from the approximate formula,
(3.13)
The bwb should be greater than h,/16 where hi is the height of the first storey. The variation
between be versus the wall height to length ratio for any wall displacement ductility J.l.A is shown
in Fig. 3.14.
3.5 DESIGN EVALUATION
3.5.1 Moment Redistribution of Beams
From NZS 3101:1982, the maximum value of the redistributed beam moment is
restricted to not more than 30 % for two main reasons. Firstly, when the required beam
flexural strength reduces because of the moment redistribution the section will begin to yield
at a moment somewhat less than the moment computed from an initial elastic analysis. The
limitation ofthe redistributed moment is to ensure that the member plastic hinge does not occur
prematurely. Secondly, the consequence of the smaller beam flexural strength is that the
rotational ductility demand will increase. An example of the redistributed beam moment is
presented in Appendix B.
59
The average of the ratios of the redistributed moments of the exterior and interior
columns for frames and frame-wall structures were less than 30 %. The percentage of the
redistributed moments decreased as the beam moment induced by the equivalent static load ME
increased. The superimposed negative and positive moments have almost the same value
because the dead and factored live-loads are only small fractions of ME'
3.5.2 Tension Reinforcement Ratio
The Tension reinforcement ratio may be used to determine whether an appropriate

member preliminary sizings has been achieved. The column tension reinforcement ratio varied
from O.S to 2.1 % and satisfies Ref. X2 and X3. Some of the top floor beams were given a
minimum tension reinforcement ratio and the maximum tension reinforcement ratio is 1.2 %.
All the beam and column reinforcement ratios are in an acceptable range, meaning that an
appropriate preliminary member sizes were achieved. The requirements ofEq. 3.1 are assessed
over whole structural systems.
3.5.3 Design of Structural Walls
As mentioned earlier, barbel shape cross section structural walls are used. To avoid
premature buckling, a wall thickness bwb and the size of the barbel was arranged to satisfy the
requirements of Eq. 3.12 and 3.13. It was found that most critical load combinations was the
combination of minimum wall axial load Pwumi and wall ultimate moment Mwu.
The selection of the wall length is carried out by trial and error and depends on the
amount of reinforcement steel. When the wall length is too small it is difficult to satisfy the
required flexural strength with a relatively high steel reinforcement ratio. In this case the
flexural strength will govern. In contrast, when the wall length is too large it is difficult to
provide the required axial load as only a relatively small concrete compressive zone is
necessary to satisfy the required flexural strength. In this case the axial load will govern. It is
necessary to develop an approximate formula for the wall length from which both the axial load
and the wall flexural strength are easily satisfied in the design process.
For this purpose, a number of barbel wall strength analyses for IS-storey buildings have
been carried out to determine the relationship between reinforcement content p versus the
combination of material strength or wall axial load eccentricity for different values of the
4-
e = hwllw = 12.77 "
3.5 3.S
3
c: 45/300, hwllw = 12.8
3
~ D: 451430. hwliw = 14.4
*
~. 2.5 hwllw = 1
0
.;:: 2.5
m
a: ...cr'"
.5 2
c: 2 OJ
a:
'0;
a: c: 1.5
0
'OJ
.;:; 1.5 hw!lw = iii
c: l- 1
Il)
l-
O.S
0.5
0
1.5 :2 2.5 3 3.5 4 4.5
(1
45/300 45/430 Eccentricity e = MN

25/300 25/430
fig. 3.15 Tension Reinf. Ratio vs. Material Strength Fig. 3.16 Tension Reinf. Ratio vs. Eccentricity. 0')
o
3T-~------------------~~~~~~~~~~~--------~---'
3.5 ~------------------------------------------------------------~
2.5 3
2.5
2
2
1.S
1.5
0.5
0.5
o +-------+_------+-------+-------+-------+-----~
0+----+----r---4----4----+---~--~r_--~--_+----+_--~ 2 :3 4 5 6 7
251300 30 35 40 45 25/430 30 35 40 45/430 Eccentricity e M/V
Fig. 3.17 Tension Reinf. Ratio liS. Material Strength Fig. 3.18 Tension Reinf. Ratio \/s. Eccentricity
5 4.5
4.5 4 A: 25/300, "wliw = 9.72

. ....: 3.5
S: 25/430, hw/lw = 10.32
C: 45/300, hw/iw = 11.76
~ 3.S ~ 3 D: 45/430, 11""11",, = 13.02
~" :3 "<:I
':l
-
II: II: 2.5
E 2.5
., :g.
II: II: 2.
C
2. c:
.s., '"
';;; 1.5
c: 1.5 l:
OJ
{! f-
II' 0.5
0.5
0 0
8 9 10 11 12 13 14 1.5 2 2.5 :3 3.5
Wall Height to Length Ratio I\w/lw. Eccentricity ii MN.
Fig. 3.19 Tension Reinf. Ratio liS. W<:JII Height to length Ratio Fig. 3.20 Reinf. Content \IS. Eccentricity.
Il- 3T-------------------------------------------------~
3.5 !
~
A: 1111 iw ., 10
B: I'll/I""
C: 111/lw
c 2.5
~ 3t ~
At.
~ 2.5
'"
a:
C 2
'w
II:
c: 1.5
0
'iii
.,
c:
f- Rect1Sngull'lr W,,!!
0.5 45/430 0.5
~'.""" ,1
0
9 10 12 13 14 O+----r--~~--~---r--~~--+_--_+--~~--+_--_+--~
8
25/300 30 35 40 45 25/430 30 35 40 45/430
Wall Height to length Rltia.
Fig. 3.21 Tension Reinf. Ratio IfS. W!311 Haight to length Ratio Fig. 3.22 Tension Reinf. Ratio liS. Material Strength
62
required wall height to wall length ratio as shown in Fig. 3.15 and 3.16 respectively. For a
given wall height to length ratio hw/1w the required reinforcement contents decrease as the
material strength increases and will increase as the eccentricity e = M/V increases where M
and V are an applied flexural moment and axial load respectively. The strength analyses were
carried out for the eccentricity ranging from e = 3.5 to 3.8, the combination of concrete
compressive strength fo = 25 to 45 MPa and steel bar yield strength fy = 300 and 430 MPa
as shown in Fig. 3.15 and for a wide range of wall height to length ratios of hw/lw == 8.94 to
14.4 as shown in Fig. 3.16.
An approximate formula for barbel walls with satisfy both Eq.3.12 and 3.13 is,
(3.14)
where fo and t;, in Mpa, e, llw and lw is in metres.
Equation 3.14 was based on providing a minimum wall web reinforcement and taking
into account the moment contribution of the yield web reinforcement. Analyses on the
performance of Eq. 3.14 in the design of the structural wall on the I8-storey frame-walls was
carried out for the required barbel reinforcement ratio for given combination of material
strength and axial load eccentricity as shown in Fig. 3.17 and 3.18 respectively. They shown
that the required tension reinforcement ratio varies from 1.5 to 2.5 % and satisfies the
reinforcement limits according to Ref.X2 and X3.
Similar analyses have been carried out in the case of a rectangular wall cross-section
for the I8-storey frame-wall structures with the results shown in Fig.3.19, 3.20 and 3.21
respectively. As mentioned earlier, a minimum wall thickness of one-tenth of the first storey
height or hJllw ratio as suggested by Ref.P18 and P32 is adopted here. Fig.3.21 shows that the
reinforcement ratio decreases as hJ/lw decreases. An approximate formula for wall height to
length ratio can be computed from,
(3.15)
::: 12.75 + _f_c--=- - 050 ~
1500 e
where C{ = h/lw is the desired value of the first storey height to wall length ratio.
Fig. 3.22 shows the numerical test results of the required wall reinforcement content
for a 12-storey wall with e = 2.6 metres and an 18-storey wall with e = 3.20 metres for a
63
wide range of material strengths. The tension reinforcement in the wall was assumed to be
placed in the 1.25 b to 1.75 b distance from each wall with minimum reinforcement ratios
between these end segments. The main wall tension reinforcement ratios are nearly 2.0 %.
3.6 SUMMARY CONCLUSIONS
The relationship between The Conceptual Overall Design, Preliminary Analysis and
Design aspects in an integrated Earthquake Resistant Design (EQ-RC) for buildings has been
discussed. The selection of the building plan and type, stiffness distribution, location and
orientation of the building lateral-resistant systems affect the building response during
earthquake excitationA15 ,P18 and needs to be emphasized in the designing process.
Twelve and eighteen storey symmetrical, reinforced concrete frames and frame-wall
structures were selected as building prototypes. These were designed to NZS 3101:1982 and
NZS 4203: 1984. Three types of foundation were considered; the building was assumed fixed
at the ground surface, at the footing level and the building was supported on a flexible
foundation system.
Capacity design procedures were used in the preliminary design . An appropriate

design for the beams and columns was achieved as indicated by an acceptable reinforcement
ratio according to NZS 3101:1982. The beam and column flexural strengths were computed
using an ultimate strength analysis.
The structural walls were designed for the flexural actions considering not only the
required wall ductility but also wall stability against lateral buckling. To ease the selection of
wall length 1w approximate formulae for the wall height to length ratio hjlw for barbel and
rectangular cross-sections were developed. Numerical tests to evaluate the performance ofthese
formulae were carried out and the required tension reinforcement ratio p are in the range
indicated in NZS 3101:1982.
Chapter IV
Dynamic Properties of Soils, Shear Strains and Rheological
Model of Soils
4U INTRODUCTION
In structural engineering the dynamic loading may be due to earthquakes, vibration

effects of unbalanced machines, wind loads, blast loads, ocean waves or vibration resulting
from vehicle movement. Patterns of wave propagation, cyclic amplitudes and frequency content
of these loadings vary with the type of generating mechanism. Strong motion accelerograms
recorded from earthquakes are very random both in amplitude and frequency content. Based
upon experience, the frequency range of the energy release by earthquake may be significant
between 0.1 Hz - 10 Hz01 ;n. Waves resulting from steady state loading such as machine
foundation vibration, may be considered as a sinusoidal loading. For low speed, moderate
speed and high speed machines, the generated frequencies may be considered in the range of
1-10, 10-25 and 25-50 Hz respectively-u which are much higher than those for earthquakes.
This input frequency is one of the important parameters in the analysis offoundation vibrations.
Under static loading with constant soil stress, soils may undergo creep where the soil
strains may increase or decrease in strain-rate with time. Under dynamic loads, either from
earthquakes or steady state loadings, the stress-strain behaviour of soils are non-linear and
hysteretic. Stress strain hysteresis loops of soils indicate that even though relatively small, soils
have internal material damping, The non-linearity depends on the soil type. Under cyclic
loading sands tend to show a high non-linearity, whereas non-linearity of clay soils strongly
depends upon the plasticity index pF6. All of these characteristics such as the stress-strain
relationship, material damping and the non-linearity of soil properties basically represent a
viscoelastic behaviour of the soilsCS ,K3,Wll. A large number of factors may affect the behaviour
of soils and a comprehensive general stress-strain relation of soils is very complexH5.H6.Z1.
64
65
41.2 DYNAMIC PROPERTIES OF SOILS
The analysis of soils under dynamic loading is one of the most interesting problems
discussed by geotechnical engineers over the past few decades. The dynamic properties of soils
are the soil shear modulus G, the soil material damping C, the soil Poisson ratio /I and the
shear wave velocity V. These properties are very important parameters in the analysis of soil-
foundation vibrationsH3 ,Rl.
In general, the dynamic properties of soils may be categorized in two groupsID . The
first group covers dynamic loading on soils where the shear stress and shear strains are
relatively small. In this group are soil vibrations due to unbalanced machines, conventional
blasting, traffic loadings and small earthquakes. The second group are problem where soils are
subjected to nuclear explosions and large earthquakes which produce strong waves propagating
with relatively large soil shear stress and shear strain.
4.2.1 Sherur Modulus (G) of Soils
As mentioned above, the shape of two dimensional cyclic load-deformations of soils

are both non-linear and hystereticD5 ,H5,M2,ol. In the case of soil vibration problems, the shear
stress-shear strain relationship is the most important. The soil shear modulus and the soil
material damping can be computed directly from this relationship. Examples of the shear stress-
shear strain relationship is depicted in Fig. 4.1 and a simplified model is shown in Fig. 4.2.
There are three important parameters that can be defined from the shear stress-shear
strain relationship; first, the initial slope of the loading curve represents the maximum value
of the soil shear modulus Gmax ; second, the slope of a line through its end point of any
hysteresis loop indicates the secant or apparent shear modulus G see and third, the area of the
hysteresis loop indicates the amount of energy dissipated during the corresponding loading
cycle. Generally, the measurement of maximum soil shear modulus Gmax , may be defined at
a low level of shear strain as small as 0.25x1(J4 and frequency about of 0.5 Hz. The main aim
of measurement at low frequency is to avoid the effects of the inertia forces on these three
parameters. The shear modulus and material damping of soils are affected by a number of
important parameters. Those important parameters are tabulated in Table 4.1 below.
66
Table 4.1 Parameter Effecting Shear Modulus & DampingHS
[mportant to
Parameters Shear Modulus Damping
Clean Sllnds Cohes. Soils Clean Sands Cohes.Soils
1. Strain Amplitude V V V V
2. Confining Pressure V V V V
3. Void Ratio (e) V V V V
4. No. of Load Cycles Rb R V V
5. Degree of Saturation R V L U
6. Overconsolidated Ratio R L R L
7. Effective Strength Envelope L L L L
11. Oetahedml Shear Stress L L L L
9. Frequency of Loading R R R L
10. Time Effects R L R L
11. Grain Chametenstic, Size, Shllpe,
Gradation, Mineralogy R R R R
12. Soil structure R R R R
13. Vol. Change Due To Shear Strain
(strain less 0.5 %) U R U R
V : Very Important, L : Less Important, R : Relatively Unimportant except as it may effect

another parameter, U : Relatively Important, but is not clearly known at present time, Rb :
Except for satumted clean sand where the number of cycles of loading is less important
parameter.
The parameters are grouped into 3 categories; i.e very important, less important and
relatively unimportant. For aU types of soil, the shear strain amplitude 'Y , the effective mean
principle stress (confining pressure) Uo , and the void ration e are the most important
parameters influencing the soil shear modulus and damping. Recently, it was also found that
the plasticity index PI of cohesive soils gives an even more significant effect on the soil
modulus and damping when compared with that of the confining pressureV6,S13,Zl.
HARDIN & BLACK (1969) and HARDIN & DRNEVICH (1972) have shown that for
sand and cohesive soils, the maximum soil shear modulus may be expressed in the respective
forms,
Gmax (4.1)
Cg F(e) OCR'l (4.2)

67
SHEAR STRESS
T
D B
I
I
~c
SHEAR STRAIN, Y
Fig. 4.1 Shear Stress-Strain Relationship of Soil
Sheor Shear
Siress Stress
(0) ( b)
Fig. 4.2 Simplified Soil Stress-Strain Relationship (T13)
400
1 10- 6
;500~rr~~~~~~~~~~6~
1'=10- III

1000r--r~c-,-+-----~------~ .... 330 --- --- 1--- --I --III
~u 300
-...till
~
500

l1li
200 ,
i
I
1.0 1.5 2.0 I
: : -l -- --
Void ratio e i--- --- -- - -
-
N.P. 0 10 20 30 40 50
Plastic inde~ Ip (%)
Fig. 4.3 Shear Modulus vs Void Ratio and Coefficient of Soil Shear Modulus vs
Plasticity Index of Clay Soils
68
The upper and lower bound of the constant Cg respectively are,
(4.3)
Cg = 1576" 620 150 (ha

""1' ,
kips , kg) (4.4)
inch 2 cm Z
and the value of F and m respectively are,
== (2.972 - e)2 (4.5)

F(e)
1 + e
m = 0.5 (4.6)
Cg and m are constant, F is constant function of void ratio e, 0'0 is the mean effective
principle stress (confining pressure).
If soils samples are normally consolidated, Eq. 4.2 is equal to Eq. 4.1. In the case of
overconsolidated clay soils, the value of q may be found to depend on the soil plasticity index,
as represented in Table 4.2.
ZEN & UMEHERA (1978) examined the validity of the constant Cg by in situ and
laboratory tests. A number of undistributed and remoulded clayey soil samples were tested and
the results are represented in Fig. 4.3. As shown the results of the laboratory test, in genera],
agree with Eq. 4.2. It was also found that the value of constant Cg ( coefficient A in Fig. 4.3)
is not always equal to Eqs. 4.3 or 4.4 but varies with the plasticity index PI. Fig. 4.4 is
represents the shear modulus of clayey soil presented by some researchers and shows that the
shear modulus computed by Eq. 4.2 is relatively high.
Results of tests on cohesive soils was reported by KIM & NOVAK (1981). Several
different undisturbed soil samples were used. The properties of soils varied from hard, highly
overconsolidated, normally consolidated to very soft glacial soils. Plots of normalised shear
modulus F(e) against confining pressure for a normally consolidated different clay soil is shown
in Fig. 4.5. Compared with the HARDIN & BLACK (1969) results, the range of these soils
shear moduli lie near a lower bound. It was also found that constant Cg varies from 893 to
1726 with an average value of 1395 (shear modulus in kPa).
69
Table 4.2 Coefficient qRS

Plasticity Index (PI) Coefficient q
o o
20 O.lB
40 0.30
60 0.41
80 0.48
>100 0.50
4.2.1.1 Effect Soil Shear Stain "I..
The maximum shear modulus GmOll computed from Eq. 4.1 and Eg. 4.2 was
based upon low levels of soil strain. Vibration in soils due to some types of dynamic loadings
do not necessarily generate low levels of shear strain. As mentioned in Section 4.3 the soil
shear strains induced by earthquake loading varies from approximately 10 -4 to 10 .3. It is
important therefore, to know the soil shear modulus G for any soil shear strain.
Through intensive laboratory tests it was found that the soil modulus G
decreases very rapidly with increasing shear strainH5 .H5.K2, especially after the shear strain
exceeds a threshold level. Fig. 4.6 and Fig. 4.7 represents the soil shear modulus reduction
curve for clean dry sand and various clayey soils respectively. The variation of the soil shear
modulus with strain from these figure seems to differ. This is because of different abscissa
scales, i.e linear and logarithmic scales respectively. Fig. 4.7 shows the normalised soil shear
modulus (G/Gm,J variation with strain amplitude.
4.2.1.2 Effect of Void Ratio (e)
Soils where the void ratio e is relatively small can be considered dense or stiff
soils. Results of laboratory tests indicate that the soil shear modulus is inversely proportional
to the void ratio e as represented in Eq. 4.5. HARDIN & BLACK (1968) proposed an
empirical formula for the effect of the void ratio e on the soil shear modulus Gmox Laboratory
tests were carried out for a range of void ratio e from 0.5 to 1.90. For clayey soils" three soil
samples were tested namely Leda Clay, Lick Greek Silt and Brown Loam. During the tests,
applied confining pressures (To were 1.0 kg/cm2, 0.89 kg/cm2 and 1.0 kg/cm2 , with void ratios
1.9,0.91 and 0.61 respectively. The respective maximum shear moduli G mox were 1.7 lesi, 6.5
ksi and 11. 0 lesi. It appears that the value of dynamic shear modulus is a strongly inverse
function of the void ratio.
70
'00/:)
eO.a
.....
2 40.0
~Z ~
""
'J .
..., 2:0,0
Ci '":;
.. 2 "
Q
"
:; '"
:l!
~ '"
l" it.oP{:~Nr CP;
I
.. '11'r.lOOL t< "<HL
!g
{fd l:C..i:
0 .0 4W\', 0', ITzt. On!! wmo5rOQ tLA'"
..
i'
~
o ,wo,
o .,MI
0'1<)
01,
16.'a olin
liB:] ()M~
Wj\.ll.ltE&uRC (LA'"
SILT
'HATH"lIi!
;; 4.0 ,:)'3T
V IWA) l2'tli HAU,ILtCW4 Sli. T
S. o ,sa, 06. ~l' 'SARHh\ (LA.'"
o !l0) 0" 8'] U7 K)NA t:LA'
o 'P~l 0 .... 140) poor 'STANl!:V C",A,"
2.0.ro
40 00 r)Q 200 40Q {OO 0000
Void ratio it
Fig. 4.4 Shear Modulus vs Void Ratio of Fig. 4.5 Effects of Confining Pressure on
Cohesive Soils (16) the Soil Shear Modulus (K2)
tu..... 1)11< $A.ND

VOID IUllO 'O3l'
,R:01JHCY 'VI.! u.s..
J
\:J.O 1:1. WIHOWGl 'l.Ay
o W.'tUAC,!!luRG CLAY
o Ct1"TI4"'" Sil. T
V ttAMIt.TOI't SkLT
Ll.
ZQ 0 $A,QklA. Ct.A'I'
o tONA. CL"Y
o peAl" STANLEY CLAY
)Q~l iOi'l
!itftAff 'HRAIk ("".. 1
Fig. 4.6 Effects of Confining Pressure and Fig. 4.7 Normalized Shear Modulus (K2)
Void Ratio on the Shear Modulus (HS)
1.0
0.8
'-.
0.5 <0
00.4
1 .. 0 t.o
2.0 t.o
J.O to
0.2
0.0
10-4 10 ... 10 -, 10 -. 10
Shear Stroin. percent
Fig. 4.8 Nonnalized Soil Shear Modulus for Different Void Ratio (S13)
71
Effects of void ratio e to normalised shear modulus of clay silt was also
reported by SUN eLall (1988). Data from laboratory tests for about 70 cohesive soils from
many countries were collected. Most of the samples were normally consolidated clayey soils.
The relationship between the normalised soil shear modulus reduction curve G/Gm"" against the
soil shear strain "I for any value of void ratio e is shown in Fig. 4.8. Again, the soil shear
modulus is strongly dependent on void ratio e.
4.2.1.3 Effect of Confining Pressure (0'0>
The effect of the confining pressure (Jo on the shear modulus G and normalised
shear modulus G/Gmwr. for sand and clayey soils are clearly shown in Fig. 4.9 and Fig. 4.10.
As shown, higher applied confining pressure (Jo gives higher soil shear moduli. No.2, No.3
and No.6 in Fig. 4.9 refer to clay soils with 22 %, 41 % and 70 % fractions of sand
respectively. As shown in Fig. 4.9 sand is more sensitive to the effect of confining pressure
rather than clay soils. The value of m in Eq. 4.1 or Eq. 4.2 is affected by the corresponding
soil shear strainOl For low levels of soil shear strain the value of m = 0.5 can be usedH3.
Fig. 4.5 shows the effect of confining pressure on the maximum soil shear
modulus G m J3J[' This results were conducted for a low level of soil shear strain of "I., = 0.002
%. The slope of the relationship varies from 0.46 - 0.61 with an average value of 0.55. This
slope indicates the value of min Eq. 4.1 or 4.2. The effect of applied confming pressure on
the soil shear modulus is very significant. This result is significantly different from Fig. 4.9,
ZEN et a1.(1988).
4.2.1.4 Effect of Soil Plasticity Index (PI)
The effect of the plasticity index PIon the shear modulus is shown in Eq. 4.2.
The constant k in this equation is a function of the plasticity index PI. However, Eq. 4.2 is
used only for calculating the maximum soil shear modulus Gmax for low levels of soil shear
strain. The effect of the plasticity index on the shear modulus for any value of soil shear strain
is not available.
Only a few researchers have reported this effect. SUN et a1.(1988) found that
the plasticity index strongly influences the shear modulus of clayey soils as illustrated in Fig.
4.11. In soils with a high plasticity index (Mexico clay) the shear modulus seems to remain
72
.J"INOSQR (LAY
o lff-v.,'jtli.fi .. J1g
o Q/Lv. 12') <i "PIl
.
l
. 0
A
rlA,v' l~11. P"
diN" 591' l _Po
~~O76~~1~n-~----JoJ-~4~~--
1
4.9 Effects of Confining Pressure on Fig. 4.10 Variation of Shear Modulus

the Soil Shear Modulus (Z2) with Confining Pressure (K2)
1.0
0.8
~ 0,6
E
0
......... Typo
o 0.4 Cl
C2
Cl
0.2. C4
c,
0.0
1 a -.
10
Shear Strain, percent
Fig. 4.11 Normalized Shear Modulus for Different Soil Plasticity Index (S13)
0.8
0.6
PI=200
30 S0100~
0,2
o~
(),O "oi)l'--~OO1----;!;.~----:-!;--.---.::::::..L....-
Qoom ~ ru __..J
10
CYCLIC SHEAR STRAIN,Yc{%)
4.12 Normalized Shear Modulus for Different Soil Plasticity (V6)
73
constant for a wide range of shear strains. This result was based on data as mentioned in
Section 4.2.1.2 . ZEN et a1.(1988), also reported an effect of the plasticity index Plan the
shear modulus was even greater than the effect of the confining pressure. DOBRY &
VUCETIC (1977) and VUCETIC & DOBRY (1991) presented shear modulus reduction curves
for normally and overconsolidated soils as illustrated in Fig. 4.12. As shown, the soil with a
high plasticity index tends to behave more linearly-elastic even at a high soil shear strain. The
effect of the plasticity index PI on soil parameters are presented in Table 4.3. PENDER et
a1.(1992) reported that dynamic properties of Auckland clay follows closely the curves C2 and
C3 in Fig. 4.11. The soils plasticity index PI varies in the range from 20 % to 30 %. The
shear modulus reduction curve for Auckland clay is illustrated in Fig. 4.13.
Table 4.3 Effect of Plasticity Index to Soil Parameters
Plasticity Index. (PI)

Soil Parameters
Low PI High P
l.Initial Shear Modulus" Low High

2.Modulus Degradation High Low
3.Damping High Low
4.Soils Amplification Low High
5.Non linearity of Soils High Low
.. For overconsolidated clay soils
4.2.1.5 Effect of Cyclic Loadings
Fatigue phenomena may reduce the strength or stiffness of the materials. Under
a large number of cyclic loadings, the soil shear modulus may reduce significantlyll.v5.v6. An
example of the effect of the number of cycles of loading N on the normalised modulus
reduction curve is illustrated in Fig. 4.14. It is clear that the effect of cyclic loading on the
shear modulus is more significant for soils with low plasticity index than for those with medium
to high plasticity indices.
The effect of cyclic loading on the soil shear modulus can be evaluated by a,'
so called, degradation index OIl.V6. This index can be written as,
(4.7.3)
74
~ Stress control
II Strain control
0.8
0.6
"~
t?
a 0.4 Starting points ~---J
0.2
o L -_ _~'.~'~I~"~"LI~LiI~lil~II~!lLI~~~1ilillt~IlLI~~~~L-LJ~~
10.4 10. 3 10.1 10. 1 1.0 10.0

Shear strain amplitude (%)
Fig. 4.13 Normalized Shear Modulus for Auckland Clays (P20)
0.8
0.6
0.4
IOCR= 11
0.2
0.0 '::-=:::-_ _---L___----~----------'.-----::::..---~

0.0001 0.001 0.01 0.1 10
CYCLIC SHEAR STRAI N. 1;: (%)
Fig. 4.14 Effect of Cyclic Loading on the Normalized Shear Modulus (V6)
0.4
El:
- PI = 0
w (SAND)
I-
~
-< O.J
.c:: PI = 15
0::
~
1':-,.
z 0.2
-<
C
Eo-
C
< 0.1
r.::::
c.;
[;;J
c
0.0
0 2, J
CYCLIC SHEAR STRAL"I, y, (%)
Fig. 4.15 Effect of PIon the Cyclic Degradation Index (V5)

75
(4.7.b)
where GN is the secant or apparent shear modulus after N cycles. Go is the shear modulus at
the first cycle and t is a degradation parameter. Eq. 4.7.b can be expressed as,
Il = - (4.8)
log N
Variation of the degradation parameter t versus the cyclic shear strain is shown
in Fig. 4.15. The smaller the degradation parameter t the smaller the degradation of the shear
modulus. As shown, sand has a higher degradation parameter t than clay which means that
sand is more sensitive to cyclic loading than clay.
4.2.2 Material Damping of Soils
Even at relatively small levels of shear strain the cyclic stress-strain relationship of soils
is not linear elastic but possesses a hysteresis loop as illustrated in Fig. 4.1. Energy is
dissipated by a process of particle slippage within the soils. The amount of energy dissipated
at every loading cycle is indicated by the area of the corresponding hysteresis loop. Energy
dissipated by this mechanism is usually called material damping or internal damping of the
soils.
As for the soils shear modulus, the material damping of soils is dependent upon the
level of soil shear strain 'Y. Unlike the soil shear modulus, material damping of soils increases
rapidly with shear strain as shown in Fig. 4.16. The effect of the void ratio e on the soil
material damping is not clear. Ref. H5 reported that material damping decreases as the void
ratio e increases.
Accompanying Fig. 4.13, PENDER et a1.(1992) reported that the damping ratio for
Auckland clay lay within the range of the SEED & IDRIS (1970) results, as shown in Fig.
4.17. It was also found that the stiffness and damping was not greatly affected by the number
of cycles, meaning that the soil shear modulus and damping ratio depended only on the level
of applied soil shear strain. The impOltant aspects on the soil shear modulus and the soil
material damping are illustrated in Fig. 4. 18G12.
76
4.2.3 Poisson's Ratio (1') of Soils
Poisson's ratio J,I may be defmed from the strain ratio of the axial compression and the
lateral expansions21 Poisson's ratio of the soil has a direct relationship on the stiffness of the
soil-foundation interaction and an appropriate value of the Poisson's ratio must be used in the
analysis.
Soil deposits are non-homogenous both in the vertical and horizontal direction. This is
because of natural sedimentation of rock debris and other soil particles. The strength and the
elastic properties of the soil deposits may not be the same in the vertical or horizontal
directionsill Accordingly, soil is really an anisotropic material, i.e the elastic properties of soil
are different in any direction. For simplicity, however, it is common to assume that most of
the engineering materials are homogeneous and isotropic. Typical values of soil Poisson's ratios
presented in the literature are presented in Table 4.4.
Table 4.4 Typical Value of Soil Poisson Ratio
Types and Condition of Soils Poisson ratio
1. Clay, saturated 0.50

2. Clay with sand and silt 0.30 - 0.42
3. Clay, unsaturated 0.35 - 0.40
4. L1ess 0.44
5. Sandy Soils 0.15 - 0.25
6. Sand 0.30 - 0.35
4.2.3 Shear Wave Velocity (Vs)
The shear wave velocity V. near the ground surface has been recognised as one of the
important soil properties in earthquake engineering. The dynamic stiffness and damping
coefficients are dependent on the dimensionless frequency a., which is inversely correlated with
the shear wave velocity V. The relationship between the shear wave velocity and the shear
modulus G is written as,
(4.9)
where p is the mass density of soils

77
o
WINOSCR eLA'(-
WALlACCI3u:RG -CLA"
I 1:' ... ~ 1.]& I
l;... ~ 1.051
.-
40
. 1OOkPa consolidation
""
200kPa consolidation
o CH"rt4AM 'SILT 4 Il~ o~n I
tJ kAMILTON s-tl T l II ... a'lll I
'" Damping limits rrom Fig. 3
o URNIA (Lf>Y 11'1.. O,$-g I 30
o tONA (LAY tll:,, 0,(;2 I
o PORT STANLEY (LAy ~ t., ~ 0 __ 50 I /
UAV ~ ,..)011 kPu /
1000 CYCLE"S / ~ 2.0
0
/
.0 '"
0
10
dOl
O~----'~--~------~--------~
10'\! 10" 10 10'
Shoar strain ampliludo ("!o)
4.16 Variation of Soil Damping Ratio Fig. 4.17 Variation of SoH Damping Ratio
with Strain Amplitude (K2) of Auckland Clays (P20)
G"'Oo-G"",.
Mooolooic loodin\) OJfVII
Yo l'C
10-11 10- 6 '0~4 10d S 10'" 10- 3 1~2
I I
" ""
--
G
~O.5 O.!i
"
"Allefog8.WfVlIIfor ~
GAAVELLY lIOil6 "
'"
CLAYS SANDS
"
0.001 0.01 0.1 0.00' 0.01 0"'
ZO ZO
A PIII'I:OO.
to 10
SANOSAND
GRAVELS
0 0
0.001 0.01 O.t 0.001 0.01 0.1
Ye .lXlr~nI rc parclml
Fig. 4.18 Effect of Parameters on the Non~linear Hysteretic Stress-Strain of Soils (G12)
78
SITE VESIGHATION
D E F
5.0
10.0
15.0
20.0
25.0
4 . 19 Variation of the Shear Wave Velocity in Christchurch Sites (Bll)

F 19.
Ys-M
(II=UHI
, vs=i7.Dl 1. J14
~
OJ III It5
lIhlu Hilllu)
Fig. 4.20 Variation of Shear Wave Velo- Fig. 4.21 Variation of Shear Wave Velo-
city wit N-SPT Value (12) city with N-SPT Value (13)
Z
Iml
10
Fig. 4.22 Variation of Shear Wave Velocity with Depth (C23)

79
Other expressions for the shear wave velocity have been reported by HARDIN &
BLACK (1969) and KIM & NOVAK (1981)K2 respectively,
~ = (103.6 - 34.93 e) OCRlf1. o/lj.l (4.10)
(4.11)
The expressions for the shear wave velocity in Eq. 4.10 and Eq. 4.11 are similar to Eq.
4.2, where the shear wave velocity is a direct function of the void ratio e, over-consolidated
ratio OCR, and confining pressure (fo. Compared with Eq. 4.10, the shear wave velocity
computed from Eq. 4.11 is significantly smaller.
The shear wave velocity may also be obtained from in situ downhole or crosshole tests.
A crosshole test is probably the best geotechnical method for determining the variation with
depth of in situ low strain shear wave velocity Gl2. The crosshole method is based on a very
simple concept. A shear wave is generated in a borehole and the arrival time at the same
elevation is measured in a neighbouring borehole. The shear wave velocity V. is computed
from the travel time and the distance between boreholes. The measurement of the shear wave
velocity in clayey soils using this method has been reported by Ref. BlO and C12. In general,
the results of the tests indicate that the shear wave velocity will increase with depth.
Shear wave velocities in the upper 30 metres from the ground surface was also reported
by BERRILL et al.(1993). Six sites around Christchurch, New Zealand were investigated. The
soil composition of each site and the variation of shear wave velocity with depth is illustrated
in Fig. 4.19. As shown, the shear wave velocities depend upon the type and elevation of the
soil layers. The variation of shear wave velocity with depth in some sites is not clear, but at
site D and F this variation is more obvious.
Site investigations of shear wave velocity using Standard Penetration Test (SPT) are
also reported by some researchers. Even though the SPT is mainly used for granular soils, this
method may also used for clayey SOilS D6I2.I2.14,09. From this method, several empirical formulas
for shear wave velocity have been proposed. By using more than 940 data samples, lMAI
(1977) presented an empirical formula for soil shear modulus as a direct function of the number
of blows N of the SPT value. The data was collected over 10 years for clay, sand and gravel
for a range of geological soil ages. The correlation is plotted in Fig. 4.20 and mathematically
80
expressed in Eq. 4.12. It is clear that the shear wave velocity is well correlated with the N-
SPT value, especially for relatively stiff soils. The coefficient correlation was 0.889.
(4.12)
(mlsec)
IMAl & TONOUCHI (1982) also presented an empirical formula for soil wave velocity
V. The SPT data were collected from approximately 250 sites, using 386 boreholes distributed
throughout Japan. The soils consist of peat, clay, sand and gravel, covering more than 1650
data items. Empirical correlation between the shear wave velocity and the N-SPT value is given
in Eq. 4.13 and presented in Fig. 4.21.
v:r -- 970 N.314 (m/sec) (4.13)
An empirical formula for the shear wave velocity based upon soil type, geological age,
N-SPT value and soil depth Z was also investigated and reported by OHTA & GOTO (1979).
An empirical relationship was formulated by using 300 sets of available data from alluvial
plains in Japan. This empirical formula is given in Eq. 4.14, where the coefficient of
correlation is 0.856.
vs = 68 79 ~.171 Z 0.199 (mlsec) (4.14)
Similar investigation was also carried out in Florence, Italy as reported by

CRESPELLANI et al.(1989). More than 200 observations of N-SPT values were collected on
clayey soils. Empirical formula for shear wave velocity resulting from this investigation is
expressed in Eq. 4.15 and illustrated in Fig. 4.22.
Vs = 261.0 N-{Jo1 ZO.24 (m/sec) (4.15)
No.1 and 2 in Fig. 4.22 are variations of the shear wave velocity with depth computed
,
from Eq. 4.15, while No. 3,4 and 5 are adopted from the OHTA & GO TO (1979)
investigation. As shown in this figure, Eq. 4.15 gives a shear wave velocity greater than that
from Eq. 4.14.
81
4.3 SOIL STRAIN INDUCED BY EARTHQUAKES
When soils are in an rest condition, there are only two types of stresses acting on the
soil element at a particular depth beneath the ground surface_ These stresses are the vertical and
horizontal geostatic effective stresses, which are generally denoted by (1" and (1h
respectivelyMl1.H12.V5. Without any disturbance, both of these stresses remain in an equilibrium

condition. The equilibrium of soil stresses may change whenever an external applied loading
such as earthquakes, vibration of machine foundations, blasting, pile driving or vibration by
a soil compactor acting on the corresponding soil element.
If the soil shear strain 1'.. generated by the normal static load isextremely small, it does
not have significant effect. This is because deformation associated with very small shear strains
will not produce distress l7 However, the combination of small shear strain and the inertia force
initiated by high frequency excitation may have a significant effect on the soil response. The
response of soil deposits under earthquake excitation, for instance, is a complex phenomenonVS.
This is because the soil deposits have nonlinear properties, are nonhomogenous and anisotropic
materials, and the waves generated by the earthquakes propagate in three directions. Upward
propagating earthquake shear waves will normally generate soil shear stress and shear strain
'Yss which are amplified as the ground surface is reached.
The estimation of the soil shear strain induced by earthquakes is essential, especially
for soil-foundation interaction or soil-structure interaction. As mentioned before, the dynamic
properties of the soils depend on their stress-deformation relationship. These properties vary
to a large extent depending upon the magnitude of the corresponding soil shear strain.
Having an estimation of the soil shear strain induced by earthquakes, enables the choice
of appropriate soils models and analysis methods. This is shown systematicaHyI6,l7 in Fig. 4.23.
As shown, within the range of infmitesimal shear strain lcrs up to approximately lO-4. the linear
elastic soil models and linear methods of analysis are fully justified. For soil shear strains of
approximately 10-4 to 10-2 , viscoelastic modelling of soils and equivalent linear methods may
be used. For soil shear strains greater than 10-2 and up to the failure strain, step by step
integration methods are suggested.
How large are the actual soil shear strains induced by an earthquake or seismic ground
motion? Several studies have been carried out by researchers. A study to estimate the soil
82
1O~ 10-~ 10-' 10-) 1O- l 10- 1
Srooll
strain
1 MNJ,um
!Strain
I L(lr~
!!.train
IFClllu~
strain
Elastic
Elosto -P(OS!IC
Foilul"l1l'
Ei~t'Ct oi
lellen
Efft'Ct oi
100d1"O rotf
Modl!'l
LmlKll'"
f!ost,e
mode-l
\'51;0- <!'Iostic
moot'l
~ history
lmcing lylX'
modE'l
Mll'thOO oi lln('(lr \~UlValm
tlllt'Or
\ SIe-p-by-sli'p
intll"9'"otion
O~IVSt$ mf'lhod mtthod mE'thod
Changes in Soil Properties with Strain Amplitude (16)
\0 ..........- - - - , - - - - , - - - - . ,
. _---.,. -. .....
.... ~
-.. _-- ... -- "'"- ... .. ,

..
F 'Ig .424
. Soil Strain vs Epicentre Distance (516)
1.0 A I I.. ~~~~

J). VP'A Afters;,;cj{ .. &-- -- II SHIOGAMA
o --=--::::::... ~ '-" ~ . . . . . '-
,,~,~ J::,,>.... ",,-
A HOSOSHIMA
OHIROO .
~A" J:,. "- AKUSHIRO J.MA.
VDM 'Aftershock $-."
...

"-
"
....
..... ,
J::,. ",
" "
0...... '\. "'" ",
0.5 Ma' hock" '\.. ". "-
"nA
vrr. lOS (ib ~ "A . "
VDM MahShock "0- ...... . . "'-
.......
...... "'-",- .....
-
.... ,-'
---RANGE OF lM30RATORY DATA FOR CLAy..... "
FUR SAND ...
10-4 10-3
EFFECTIVE SHEAR STRAIN Yeff
Fig, Normalized. Shear Modulus vs Strain Amplitude (MU)
83
shear strain induced by earthquake ground motions has been reported by SUGIMURA (1977).
Classical wave propagating theory was used in this study. The analysis was conducted by using
the May 1940 Imperial Valley earthquake and the 1952 Kern County earthquake. This study
concluded that the soil shear strains can be estimated to be in the range of about 1O..~ (%) for
the EI Centro site to about 10-4 (%) for the Taft site . These results are much smaller than the
maximum soil shear strains as shown in Table 4.5.
SHINOZUKA et.al (1983) estimated the expected maximum ground strain as function
of emthquake magnitude M and epicentral distance R. A wave propagation model were used.
The relationship between the soil strain and the epicentre distance R for any level of earthquake
magnitude M is depicted in Fig. 4.24 . O'ROURKE et.al (1984) reported the soil strain between
pairs of recording stations during the 1971 San Fernando earthquake. By using the different
ground displacements at particular times between pairs of stations laying along the same radial
line from the epicentre and knowing the distance between them, the average ground strain was
computed. Result of this study show that the maximum ground strain fell in the range of
0.6x1Q4 to 6.37xlO-4. MIDORIKAWA (1992) also reported a case study during the April 5,
1985 Chile earthquake. Acceleration time histories from Almendral Valparaiso (VPA) and Vila
Der Mar (VDM) were used. The acceleration time histories are recorded from accelerographs
which were installed on the soil deposits. A rough estimation of the soil shear strain 'Y..
proposed by TOKIMATSU et.al (1989) reported in this study may be computed from,
= 0.4 (PGV) (4.16)

Yss V
fI
where PGV is the peak velocity of horizontal ground motion, and V. is shear wave velocity.
The soil shear strains obtained from VP A and VDM site during main and after shocks
is illustrated in Fig. 4.25. It was found that during the main shock, the maximum soil shear
strains are approximately 3xl0-4 and 5xl0-4 for VPA and VDM respectively.
Research to estimate soils shear strain induced by earthquakes near the ground surface
were also carried out in Japan. The research was conducted by using ground acceleration
record data observed by a dense instrument array installed at the Public Work Research
Institute (PWA Campus)A12,TIB,T9. The site profiles where the accelerographs were installed is
composed of clay, sand, loam and gravel. The data has been collected during the period from
1979 to 1989 covering more than 100 earthquakes, with magnitudes from 4 to 7.9 and
84
Fig. 4.26 Variation of Max. Ground Ace. Fig. 4.27 Max. Soil Shear Strain vs
with Epicentre Distance (fg) Epicentre Distance (f8)
60~5r~======~------r-------~r-------~
II1II Dohry<1 aI. (19821
LINEAR
_Drvil< <I a1.1I9B4)
THRESHOLD
I I
.... 50 NRC(l98.l) SHEAR STRAIN, y. I I
P. M.l.Sui <I aJ. (l9S0) J
tf4o~~~t-_A_n_d_<m_n_(~19_B~)--4 I . /
/I!/ rr, ~/
~ MIC\;y And Sud, (1984)
~
~30
..... KoulSo(,.. {l9871 0
U
..... SOLmOhThPOINTS: I
Z
F;; 20 OIlHfXT MEASUREMENTS VOLUMJ:..RIC / I
< OPEN OAT;. PO INTI: / THRESHOLD
SHEAR STRAIN, "Iff
/:{
't'"
....l INOI E ESTl
p..
10 R CT lMATIOH I ..;1-----@-..
PARTIALLY $ATL'RATEn
I / COMPACT .. I) CL.\\'
.
O~i
/
____~L-__~__________./ .~--- .
o.oooi 0.001 0.01 0.1 1.0

THRESHOLD SHEAR STRAINS 1,1 AND y,. (%)
Fig. 4.28 Effect PIon the Threshold Shear Strain (V5)
PI =0
15
30
OCR=I-ll
50
100
AI'I'IIOXIMATE
200
HJ\NGE OF y.. _--..r;.","",;
0001 orn 01
001 0.1
(Ycuc SHEA~ STRAIN. ')',,1%1
(YCLIC SHEAR STRAIN. '1'",[%1
4.29 Effect of PIon the Modulus Reduction and Damping Curve (V5)
85
epicentral distance from 3 to 758 km. The observed maximum ground accelerations have been
plotted against epicentre distance R for any earthquake magnitude M as shown in Fig. 4.26.
By using a double integration technique, the three components of acceleration history were
converted into three components of displacement history. Knowing the ground displacement
history, the estimation of ground strain history may be computed by using standard tetrahedron
finite element models. The placement of the instrument array and the finite element calculation
procedures are discussed in Ref.T8 and T9. The proposed empirical formula for the maximum
soil shear strain 'Ys. near the ground surface is,
== { 0.894 10548 M (R + 30)--iI714 } 10--6 (4.17)
where M is the earthquake magnitude and R is the epicentral distance in Km.
The maximum soil shear strain '1.. versus epicentre distance R for any value of
earthquake magnitude is summarized in Fig. 4.27. By using empirical formula expressed in Eq.
4.17 and assuming that each earthquake mechanism is independent of the local or global
geological conditions, the approximate maximum soil shear strain for particular earthquakes
are presented in Table 4.5.
4.4 THRESHOLD SHEAR STRAIN AND CYCLIC PORE WATER PRESSURE
Under cyclic loading, soils show different stress-deformation relationships. VUCETIC

(1992) slated that under certain values of soil shear strain, the soil volume does not change
during applied cyclic loading. This level of soil shear strain is called the threshold shear strain
and is denoted by 'Y.t. If the applied soil shear strain is greater than the threshold shear strain,
the volume of the soil will change by either compaction or dilation. Dry cohesionless granular
soils tend to compact, while in the saturated granular soils develop an excess of pore water
pressure, sometimes followed by soil liquefaction. Several studies of the threshold soil shear
strain have been carried out by some researchers and the results are shown in Fig. 4.28. The
approximate range of threshold shear strain in the modulus degradation curve and the damping
curve for any plasticity index are presented in Fig. 4.29. As shown, sand has a thresbold shear
strain 'Yst of approximately 0.01 % which is much smaller than that for clay soil at around 0.1
%. In Fig. 4.14, if the applied shear strain 'Yu is less than the threshold shear strain 'Ys1;, there
is no degradation of the soil stiffness during cyclic loadings.
86
If saturated soil deposits are subjected to an upward propagating shear wave, the
hydrostatic pressure or pore water pressure between particles of soil may change. This depends
upon the type of soil and the level of the applied shear strain. As shown in Fig. 4.28 sand has
a smaller threshold shear strain does than clay soils, which means that sand has greater
tendency to change its volume than clay. If the volume of the soil does not change during cyclic
loading i.e with the applied shear strain less than the threshold shear strain, no permanent pore
water pressure remains after the cyclic loading has stopped. In contrast, the pore water pressure
remains permanent if the applied shear strain is greater the than threshold shear strain.
Commonly, clay soils show smaller changes in pore water pressure than do sandsM9
Table 4.5 Maximum Ground Shear Strain 'Y..
Earthquakes Ground Type M R (Km) Max. Shear Strain
1. El Centro, 1940 Soil 6.7 9 0.000246
2. Taft, 1952 Soil 7.2 49 0.000282
3. Parkfield, 1966 Soil 5,6 10 0.000060
4. Mexico, 1962 Soil 7,1 260 0.000245
5. Chile, 1971 Soil 7.5 120 0.000238
6. Rumania, 1977 Soil 7.0 165 0.000104
7.Tokachi-Oki,1968 Soil 7.8 200 0.000280
8.Miyagi-Ken-Oki,78 Soil 7.4 100 0.000234
9.New Zealand,1966 Soil 6.0 77 0.000046
Pore water pressure increases with the number of cycle of loading. When the pore
water pressure increases, the effective confining pressure or inter-granular pressure of the soil
particles will reduce. It has been mentioned earlier that the soil confining pressure has a
significant effect on the soil shear modulus, particularly for clay sands or sands. Soil shear
moduli reduce with reducing confining pressures.
When the applied soil shear strain 'Yss is less than the threshold shear strain 'Y.l,
therefore, the soil shear modulus G will not degrade under cyclic loading and there will be no
degradation of the soil confining pressure lJo
87
4.5 SIMPLE RHEOLOGICAL MODELS OF SOIL
Deformation of soils under static or dynamic loading may be rapid or relatively slow.
Creep due to static loading or hysteretic response due to cyclic loadings indicate that soils
deform from their original state with time. The study of the time-dependent deformation of
materials, in general. is called rheology from the Greek word rheas, means moving flowKl,TI.
During the analysis of soil-foundation interaction, the actual physical soils need to be
modelled both mechanically and mathematically to describe their behaviour under cyclic
loadings. A mechanical model is really much more easily understood A1 ,TI than a mathematical
representation.
The basic elements of the mechanical model are usually a linear-elastic spring, a
dashpot and Coulomb friction modeL A single linear-elastic spring mechanical model is called
a Hookean model, since its behaviour exactly follows Hooke's rule, whereas a single dashpot
model is called the Newtonian modeL Describing cyclic time-dependent behaviour of non-linear
materials only using single elements of a mechanical model is not normally reasonable.
Utilizing a combination of the elements is usually recommended, even though this may lead to
complex problems. Ref.K3 stated that as a result of laboratory tests, two important parameters
should be considered during selection of the soil mechanical modeL These parameters are the
stress level and the soil water content. Simple mechanical and mathematical model for soils will
be described as follows.
4.5.1 Elastic State
If materials behave purely elastically, a Hookean model can be used to represent their
stress-strain or load-displacement relationships. Soils will behave purely linearly elastically
whenever the applied stress and strain is relatively small. The Hookean model is represented
by single spring as shown in Fig. 4.30. The relationship between stress-strain or load-
displacement is linear and independent of time as described by Hooke's rule. Stress (J and strain
E or load P and displacement y has a linear relation, then
(4. 18a)
u == E11 IE
88
(4. 18b)
P = k y
where Es is the elastic modulus and k is the spring stiffness.
4.5.2 Viscous State
If .the level of the water content is greater than the liquid limit W'm then the soils will
perform as a viscous materiafO. The Newtonian model is used for this condition and is
sometimes called viscous model, since to generate any stress or force the model may use the
viscosity of the liquid material 'YI. Visualization of this mechanical model is generally
represented by motion of a piston in a cylinder (dashpot) as shown in Fig. 4.31. In this model,
the relationship between stress and strain is non linear, but is a linear function of the change
of the displacement with respect to time. Its relationship may be written as,
(4.19a)
p = 11 dy
dt
or
p (4. 19b)
dt 11
4.5.3 Coulomb State
This model is shown in Fig. 4.32, in which the strain or displacement will occur if only
if,
(4.20)
where P y is the yield force below which strain or displacement will not occur.
4.5.4 Elasto-plastic State
This model is composed of series of Hookean and Coulomb models to represent the
elasto-plastic behaviour of materials. When the applied force or load is less than yield force Py ,
89
the behaviour of this model is purely elastic. If the applied load is equal or greater than the
yield load, purely plastic behaviour will occur. This model is shown in Fig. 4.33. The
corresponding load-deformation relationships are therefore,
(4.21. a)
P = k y for y < y,
(4.21.b)
P = k Yy for Y ~ Y,
where yy is the yield displacement.
4.5,5 Viscoelastic State
When the water content of soils is less than the plastic limit, the soils may be semi-solid
or solid. An increase in the water content and stress may change the physical state of the soils
from elastic to plastic or even to a liquid state. In terms of engineering materials, however, soil
is not necessarily purely linear elastic or purely plastic. Hysteretic loops of the soils indicate
that soils have viscosity. In engineering practice the soils can be assumed to be viscoelastic
materials L2 ,v2.
4.5.5.1 Simple Composite Models
As mentioned before, use of a single element cannot represent the actual stress-
strain behaviour of soil materials. Combinations of single mechanical elements to form
composite models may appropriate. Voigt-Kelvin and Maxwell are the simplest composite
mechanical models. Both mathematical and mechanical models will be described in the
following.
4.5.5.1.1 Voigt-Kelvin Model
This model is represented by a parallel combination of a single linear

spring and a single dashpot and shown in Fig. 4.34. In this model, the principle of equal strain
90
k CI
4.30 Hookean Model 1 Newtonian Model
(]
Or
4.32 Coulomb Model Fig. 4.33 Elasto-Plastic Model
k (]
Strain Stress
iC.-_--1.._ _ Time .E
L - - L_ _ _ _ Time
Fig. 4.34 Voigt-Kelvin Model Fig. 4.35 Maxwell Model
CI (]
4.36 Standard Model Fig. 4.37 Bulger Model

91
'between two elements must be satisfied. Total stress or force in the model can be calculated
from the summation of individual stresses or forces, therefore,
(4.22.a)
P "'" k y + 'I] dy
dt
(41.22.b)
The solution of Eq. 4.22.b is,
p _!., (4.23)
y:=.-!.(l-e 'I)
k
The displacement y of Eq. 4.23 versus time t will indicate a creep

phenomenon in the soils, where the strain or displacement increases rapidly with time as
indicated in Fig. 4.34. The detail solution of Eq. 4.22.b can be seen in Appendix A.I.
4.5.5.1.2 Maxwell Model
The Maxwell model is series combination of a single-linear spring and

a dashpot as represented in Fig. 4.35. The equality of the stress or force in the spring and
dashpot is required. Total strain or displacement of the model is the summation of the
individual strain contributions of the spring and the dasbpot, thus,
(4.24)
or
dy = !dP p
+ - (4.25)
dt kat 'I]
Eq. 4.25 may also be written in the form,

92
dP +!p :=: k dy
dt 11 dt (4.26)
where y. is associated with spring displacement and yp refers to the displacement in the dashpot
element. The solution of Eq. 4.26 is,
(4.27)
If Eq. 4.27 is plotted against time t, the shape of the curve is an

exponential function as shown in Fig. 4.35, meaning that the stress or force will decrease
rapidly when t is small. This is a relaxation phenomenon of the materials. Accordingly, the
Maxwell model is used to describe the relaxation stress behaviour of materials. The detail
solution of Eq. 4.26 can be found in Appendix A.2.
4.5.5.2 Complex Composite Models
To achieve a better approach in describing the cyclic behaviour of soils, the complex
composite mechanical model is introduced. Various complex composite models have been
developed A" C5,K3. Some of these more complex mathematical expressions are not always
applicable in practice. Two complex composite models are described below.
4.5.5.2.1 Standard Linear Model
This model is composed by parallel single elastic spring and a Maxwell model
as shown in Fig. 4.36. Again, the concept of equal strain between the two parallel elements
must be satisfied. The generating stress or force initiated by this model can be computed by
the summation of the forces from the individual parallel elements, therefore,
(4.2S.a)
(4.28.b)
93
(4.28.c)
where PI and P 2 are the stress or force contribution of the elastic spring of the Maxwell model
with displacement YI and elastic spring of the Hookean model with displacement Y2' Note that
YI = Y2 = y. The force contribution from the Maxwell element may be computed from the
following relationship.
(4.29.a)
(4.29.b)
(4.29.c)
The subscript s1 and d1 in EqA.29.a indicate the spring and the damper for the
Maxwell element. Substituting of Eq. 4.29.c to Eq. 4.28.a results in the total stress or force
generated by the two parallel elements yields.
(4.30)
If one assumes the stress at any time Pt = Po = P is a constant value, the first
derivative of dP/dt is equal to zero, and from Eq. 4.28.a leads to,
(4.31)
Substituting Eq. 4.31 into Eq. 4.30, and after rearrangement leads to,
(4.32.3)
dy + (k2 1) Y = A P
dt
where,
94
(4.32.b)
The solution of Eq. 4.32.a is,
(4.33)
The detail solution of Eq. 4.32.a can be found in Appendix A.3.
4.5.5.2.2 Burget' Model
The Burger model consists of series of two simple composite mechanical

models, a Voigt-Kelvin model and Maxwell model in series. This model is represented in Fig.
3.37. Stress equality between the Maxwell and Voigt-Kelvin element must be satisfied. The
total strain or displacement of the system is the summation of the strain contribution from each
element. Thus,
(4.34)
Subscript K and M are indicate the Voigt-Kelvin model and Maxwell model
respectively. If the stress or force at any time Pl = Po = P is constant, then the total strain of
the system is the summation of the strain developed by the Voigt-Kelvin model from Eq. 4.23
and strain contribution of the Maxwell model from Eq. A.2A, therefore
(4.35)
_ 1
Y - P - + -
t 1
+ -(1 - e
kl
1l1t
)
1
[
k2 112 11
Eq. 4.35 may be able to describe the correlation between the strain and the
constant stress as a function of time, with kl , kz, 111> 11z taken as constant values. Taking the
first derivative of Eq. 4.34 gives,
95
(4.36)
The first and second terms of the right hand of Eq. 4.36 have been defined in
Eq. A.1.3 and Eq. 4.25 respectively. Substitution of these equation into Eq. 4.36 leads to,
(4.37)
After rearrangement Eq. 4.45 can be expressed as,
(4.38.a)
where,
(4.38.b)
Eq. 4.38 is similar to Eg. 4.26 thus the corresponding solution may be found
in a similar way. All of the mathematical derivation can be found in Ref. TI.
The soil rheological models have been used by researchers to describe the time
dependent behaviour of soils. CHRISTENSEN & WU (1964) used the standard mechanical
model as shown in Fig. 4.36 to simulate a soil under creep behaviour. Analytical and
laboratory tests were carried out. A good agreement between the experimental and theoretical
stress and strain have been achieved and followed by the evaluation of the spring stiffnesses
and dashpot coefficients. KOMAMURA & HUANG (1974) proposed a new rheological model
to simulate the creep behaviour in sliding soils. The mechanical model consists of the series
of the Bingham and Voigt-Kelvin models. Laboratory creep tests have been carried out and
followed by procedures to determine the parameters of the proposed mechanical model.
PENDER (1983) proposed a new mechanical model to simulate the behaviour of the soil
96
beneath a foundation under earthquake excitation. The model consist of series of non-linear
springs and the Voigt-Kelvin model. The use of a soil mechanical model in the soil-foundation
response may depend on the soil state. For a shallow soil layer over a rigid boundary or an
elastic half-space may require different soil-mechanical models. Unfortunately, the validity of
this proposed mechanical model has not been shown from laboratory tests.
The dynamic properties of soils from laboratory tests has been reported for both
cohesive and cohesionless soilsHJ ,H4,H5,H6,J(2,zz. Tests were conducted over wide range of soil
shear strains. It was found that several parameters had significant effects to the soil shear
modulus G. The most important parameters are the void ratio e, the confining stress 0"0 , the
shear strain amplitude ')'SS and the plasticity index PI. Because the soil-foundation stiffness is
a direct function of soil shear modulus ,),,,, an appropriate estimation of the soil shear modulus
is essential.
Research to estimate the soil shear strain induced by earthquakes has been carried out
by analysisSl6 and back analysis using earthquakes ground motion dataMl3 ,04,TB,T9. Empirical
formulae for the maximum ground shear strain induced by earthquakes have been proposed,
either as functions of the peak ground velocity and the shear wave velocity or as functions of
the earthquake magnitude M and epicentral distance R. Results show that the approximate
maximum soil shear strain ')' induced by several earthquakes varies in the range of O.5xlQ4
to 3xlO-4.
Meanwhile, to describe the cyclic behaviour of soils, several mechanical models have
been proposedK3 ,M6. Composite soil mechanical models can be relatively simple or complex. If
a complex mechanical model is being used laboratory tests are needed to show whether the
proposed mechanical model is in good agreement with the real behaviou~l. It is shown that
parallel elements will model the relaxation stress, and a series elements will represent creep
behaviour. The use of these models to describe the dynamic behaviour of soils beneath a
foundation depend on the stress level and the soil conditions.
Chapter V
Substructure Modelling
5.1 INTRODUCTION
Depending on the site conditions, Ii building may rest on soft, intermediate, stiff, or
very stiff soils. When the building foundations rests on and are bonded to very stiff soils, the
absolute motion of the footings subject to earthquake excitation will follow closely the free field
ground motion. The buildings can be assumed to be fixed at their bases. When stiff structures
are supported by relatively soft soils, the conventional fixed based assumption cannot be
expected to yield a satisfactory result, but for the case of a flexible frame, such an assumption
may still be justifiedB4 In the case of fixed base structures, the simple structural response
analysis can be carried out by ignoring the effect of the soil flexibility. The motion from an
earthquake is directly transmitted to the superstructure. Furthermore, the structure has no
ability to radiate energy away into the soil base.
When the supporting soil is not stiff, a difference in the displacements of the footing
foundation and the free field ground motion will be observed. The soil flexibility will affect the
response of the superstructure. The structural analysis must be carried out for the whole
structural system leading to the called soil-structure interaction analysis. The motion initiated
by an earthquake is not transmitted directly to the superstructure but passes though the soil-
foundation system. The input energy is dissipated within the superstructure and in the soil-
foundation system as well as being radiated back into the soil system. This last effect is often
referred to as radiation damping.
The effect of the interaction is to modify the structural stiffnesses as well as the
effective damping and the response to the seismic ground motion. The stiffness and damping
matrices of the governing differential equations of motion must include both the stiffness and
damping of the soil-foundation interaction or substructure.
The dynamic soil structure interaction is a complex systemJ7 ,Sl,W14. This is because of
the complexity of the soils beneath the building foundation, the nature of earthquake ground
97
98
motion and the superstructure behaviour. The complexity of the soils is not only because of its
physical dynamic properties but also the local soil geometry and the interaction between the
soils and the foundation. A major effort is required to solve the problems related to the
H1 H2
SOilS ,SI. It is necessary to make major simplifying assumptions to get a clear substructure
modelling, This substructure modelling is not only the physical modelling of the soil and
foundation but also the mechanical and mathematical modelling of the soil-foundation
interaction.
5.2. SUBSTRUCTURE GENERAL ASSUMPTIONS
Soil as an engineering material is generally formed by nature and the composition of

a soil deposit is generally non-homogenous. This inhomogeneity of the soil deposit affects its
physical properties such as soil density, water content, soil strength and its geometry such as
position and the thickness of the layers. Examples of soil deposit profiles are shown in Figs.
5.1 and 5.2. where the soil layers and the stiff boundaries have an irregular form. The soil
layers may have variable thickness and may not be horizontal.
In dynamic soil-foundation interaction, the analysis cannot easily cover all problems
such as the local site conditions as shown in the figures. Simplifications and assumptions are
made so that the soil-foundation interaction analysis be modelled appropriately. These
simplifications and assumptions are described in the following sections.
5.2.1 Soil Water Content and Water Table
In nature, the soil water content We will vary with depth. The soil will be dry, semi
saturated or saturated and the degree of saturation of the soil usually changes with the seasons.
This variation will affect the physical properties of the soi\08 including the dynamic stiffness
and damping coefficients of the soil-foundation interaction. Laboratory tests indicated that the
peak soil shear modulus Gp , will be achieved at an optimum water content we,oPtGll The soil
shear modulus G will decrease when the soil water content we is greater or less than this
optimum water content we,opt.
The presence of a water table may make the soil model appear to be a layered system.
This is because soil below the water level will have different properties. For simplicity, the
99
AII",l.t .ml colic fill

C<nlr,t
Calcium nrhon.ltrtl Will
o
10
40
,\0
1;0m
Vertical
\Cate
Fig. 5.1 Basin of the Valley of Mexico-Geologic Profile E-W
,'" . ",t,
:'
-'".',
~',
eOAST
V., mn (mi.)
O+.:--_-,lS0L,.0_ _I:.,:0L:,00.:....-_jI500
OIJ) TOWN
em,ITER
o
I
km
\
aecreasmg damage
5.2 Epicentre and N-S Geological Cross Section of Kalamata Earthquake

100
presence of the water table is often ignored, and the water content is assumed to be same for
the entire depth of the soil beneath the foundation.
5.2.2 Isotropic Material
As shown in Fig. 5.1 and Fig. 5.2 the physical properties of the soil are very irregular.
The physical properties in a particular soil layer change with depth. The irregularity of the
properties, either in a layer or in the entire depth of the soil profile may lead to a complex
analysis problem. For simplicity, the soil is often assumed to be an homogeneous material,
either within a soil layer or over the entire depth of the soil profile. If the assumption of
homogeny material is accepted, the propelties of the soil are often assumed to be independent
of direction, i.e isotropic materialss21
5.2.3 Backfill, Soil Added and Foundation Mass Effect
During foundation construction, the excavated space is not usually totally replaced by
the footing foundation, but soil is backfilled as shown in Fig. 5.3. Moreover, the footing
foundation is often not fully embedded but rather paltially embedded as shown in Fig. 5.4. The
soil backfill may be on both sides or above part of the footing foundation. The physical
properties of the soil backfill may be different from that of the original soil. The effects of the
soil backfill on the soil-foundation static stiffness have been reported by JOHNSON &
EPSTEIN (1977). The static stiffnesses decrease as the shear modulus of the soil backfill
decreases and as the width of the soil backfill increases.
As discussed in previous chapters, most research has ignored the effects of the backfill
on both the dynamic stiffnesses and damping coefficients of fully embedded foundations. In the
case of a partially embedded foundation as described in Section 7.5, the effects of side and
upper soil backfill on the foundation response was also ignored.
During foundation vibration, the soil mass above and below of the footing foundation
will vibrate in harmony with the footing P19 The soil shear cone modelling as introduced by
Ref.M5 indicates a similar phenomenon. The footing foundation also has its own mass. This
footing and soil added mass can be taken into account during the analysis of soil-foundation
response. As repOlted by Ref.W4, this added mass will depend on the Poisson's ratio of the
soil and the mode of vibration. However, as discussed in Section 7.7 the effects of these
101
masses were small. This is because the mass to stiffness ratio as written in Eq. 7.105 is
usually very small. As discussed earlier, these effects are usually ignored.
5,2,4 Footing Foundation Flexibility
In the theory of an elastic half-space the stress distribution beneath of the footing was
assumed to be equally distributed over the entire width of the footing. The stress distribution
may be parabolic, unifol'm or an inverted parabola for flexible, intermediate or rigid footing
respectively. Ref.R2 and R5 reported that the change in the stress distribution beneath the
footing foundation will affect the response of the soil-foundation system.
As discussed in the Chapter VII, the footing foundations were assumed to be rigid
bodies. This assumption leads to a simple soil-foundation response analysis. In a rigid footing
under vertical or horizontal dynamic force, the displacement will be the same over the entire
width of the footing. Under a rocking oscillation the vertical displacement will depend linearly
on the distance from the centre of rotation of the footing.
5.2.5 Coupling Between Rocking and Horizontal Vibration Modes
When the centre of the rotation coincides with the centre of mass, coupling between
rocking and horizontal modes will not develop. However, this case is rare and means that
rocking motions will develop a coupling between the rocking and horizontal displacements. The
degree of the coupling depends on the eccentricity between the centre of rotation and lhe centre
of mass. The coupling stiffnesses and damping coefficients of a fully embedded foundation is
represented by Eq. 7.43 or Eq. 7.48.
In the case of partially embedded foundation as described in Section 7.5, the coupling
between rocking and horizontal modes was neglected. Taking into account the effect of this
coupling will lead to a more complex form of the governing differential equations of motion.
The coupling effect is not considered herein.
5.2.6 Coupling Effect Between Adjacent Footings
Multi-storey buildings in a city are generally built very close to each other. The footing
foundations in a building or the footing foundations of two adjacent buildings may be close to
102
Backfill
=h ~
H,
IsotropIc "'" ,.
tl2b I
Layer 1
homogeneous
HalfspBce v/scoelas tic
haJfspace
v;"cr,v,p
Fig. 5.3 SoB Backfilled Fig, 5.4 Partially Embedded Footing
I~ ~I
~1
Tie ef1 H1
2b 1.. .1 beam I ~ 2b ~I v,
Layer 1 P1
Layer 2 V,2
H2
Fig. 5.5 Two Adjacent Footings Fig. 5.6 Surface Failure on two
Adjacent Footings (S25)
LIl,... 4:
0------0 11t'l'1 1'!>Dd.1I S a H1".o#- .. JI", DIB II I
t::I'---~ w~~rt. ,s.~ .. JS... D;8"1
+-... -t- .. .. ~.o(>. U". (JIB.
:l',
I
.\ \.
.~
---t: ~.Ia.I('U"iI'IW.
- .. , - .'... A ....
!~: 1:1: g;g: ~
{J, \
, Xi., \
/ \\ .
I
'N\
: -" ",
\~
\~
, 1'-".
\
\
-
'~.:':: '~.\ ::.>.'~ :~; rt.:).: ~',: >.; ,:' .
I
~
~
~" . . . . :~.~.:.>. --
: ', : . ,:: ,',!.':' \ : .':.:':;;
1'"
," " :,.':, Homol/olUOUS Stratum :;:,
-...,

::~ ~.:,,\.: 1:.: :'<.<:. :'. :"\':: .,~.,.:.. v:' :~:,.:'.: '. :,:., ~:,: .:.: ::,::",
I
c. 10 20
'l.pa ....l ':\.'
10 ~i:.0:\;'i~~;:~;';'":;7'~/I\~;;I.l:W;,i0~~7/~;'~~J-'
i{"'''lff10 "'''Uli .rigid lormation
Fig, 5,7 Bearing Capacity Ratio (S25) Fig. 5.8 Soil Layer over Rigid Base
103
each other. The dynamic effects of these adjacent footings is sometimes called the cross-
interaction effect. Due to this interaction, both the soil-foundation stiffnesses and damping
coefficients may be different from those of an isolated footing. The effects of this interaction
among adjacent footings should not be ignored in seismic designKlO. Two adjacent footings with
a distance's' apart are shown in Fig. 5.5.
Several studies on the behaviour of an adjacent footings have been reported. STUART
(1962) reported the theoretical and experimental bearing capacity of adjacent strip footings on
sand. The nearly static loading was applied on both smooth and rough surface footings. The
types of surface failure and the comparison between tests and analytical results are shown in
Fig. 5.6 and Fig. 5.7 respectively. ~yq is the bearing capacity ratio between the interfering and
isolated footing. In general, good agreement was obtained between the bearing capacity from
the laboratory tests and the theoretical results. When the distance between two footings's' is
less than four times the footing width (4 B), the bearing capacity ratio increases significantly.
Similar results were also reported in Ref.W16.
Cross-interaction effects between two adjacent footing foundations was also reported
by KOBORI et al.(1979). Both theoretical and laboratory tests were carried out on relatively
close (s = 2 to 4 B) adjacent square surface footings. Steady state harmonic loads were applied
on one footing called the active footing while the other footing was called the passive footing.
The velocity amplitudes were plotted over a wide range of frequency excitations. This study
concluded that the cross-interaction effects of adjacent footings were considerable over a wide
frequency range as the separation becomes small. The consideration of the cross-interaction
among adjacent footings would be important in a seismic design of structures in large city.
KAWAKAMI & TASAKI (1988) repOlted the analysis of c ross- interaction between
two surface circular footings by using the boundary element method (BEM). Three types of
incident waves SV, SH and P wave were considered which correspond to parallel, transverse
and vertical excitations respectively. To reduce the complexity of the problem, the foundation
was assumed to be supported on an elastic half space and the coupling between rocking and
horizontal modes was not considered. The amplitude of the foundation motion normalized by
the amplitude of the applied incident waves was plotted against the dimensionless frequency.
Results of this analysis indicated that the normalized amplitudes were dependent on excitation
frequency> foundation separation, mass of the foundation and the type of vibration mode. The
closer the two adjacent foundations, the greater the cross-interaction between them.
104
Dynamic interaction between adjacent foundations was also reported by ROESSET &
T ASSOULAS (1987). Rectangular surface and fully embedded foundations were analyzed using
finite element methods. Five vibration modes respectively, vertical, horizontal, rocking,
torsional and coupling between the rocking and horizontal modes were considered. The analysis
parameters were foundation separation, embedment effect and foundation arrangements and
foundation mass. When the frequency excitation was less than the fundamental frequency of
the soil stratum, the displacement response increased as the footing foundation separation
decreased. A substantial coupling between the horizontal and vertical and between rocking and
torsion modes was found that would not occur for a single foundation. The foundation
embedment emphasized the effect of the foundation mass, especially for a side-by-side
foundation arrangement. It was also found that coupling between rocking and horizontal
excitations becomes more significant with both embedment and the presence of an adjacent
footing.
The coupling between two adjacent footing foundations as discussed above was limited
only in its effects such as footing displacements or rotations. Substantial coupling between
horizontal and vertical displacement and coupling between rocking and torsion modes lead to
complex problems. The effects of dynamic interaction between two adjacent identical
building Al2 or buildings with different numbers of stories F3 yields a more complex problem.
The dynamic stiffnesses and damping coefficients incorporating the effect of adjacent fully or
partially embedded footings are not available. The coupling effects due to an adjacent footing
is therefore not considered in this study.
5.3 SOIL MODELLING
5.3.1 Mu1tiple~layers Over an Elastic Half-Space
For simplification during an analyses, the soH layers as depicted in Fig. 5.1 and the
rigid boundary can be assumed to be a layered soil as in Fig. 5.9. Each layer i can be
characterised by the thickness Hi> shear wave velocity V sj soil shear modulus G j , Poisson's
Ration Pi and soil density Pi'
Under static or dynamic loading acting on the footing foundation, the stress in the soil
spreads out in a radial three dimensional direction which is usually referred to as the effective
105
depth of influenceS19 This effective depth of influence depends mainly upon the. radius of the
footing. PENDER (1983) reported that the zone of soil involved in nonlinear strains is
approximately one footing diameter deep. SRIDARAN & GANDHI (1990) stated that the zone
of soil with negligible soil stress is apPfoximately a depth of three times the width of the
footing. The effective depth of influence of the footing under static loading presented by
HAJIAN & ELLISON (1985) is shown in Fig. 5.9. Properties of the first layer will be
important for both footing A and B. However, layers 2, 3 and 4 can significantly affect the
impedance functions for foundation B only. Layer 5, would not have any effects on either
foundation. Fig. 5.10 shows the stress distribution in the soil from which the effective depth
is approximately up to 2 ro for horizontal and rocking and more than 3 fo for vertical
displacement modes. If the footing foundation rests on a layered soil such as foundation B in
Fig. 5.9, the impedance functions of the foundation are very complicatecfIl7L8 ,L9.
5.3.2 Layers Over a Rigid Boundary
The number of soil layers may be as shown in Fig. 5.9, two layers as shown in Fig.
5.5 Of even a single layer over a horizontal rigid boundary as shown in Fig. 5.8. The numbers
and physical properties of the soil layers will affect the impedance functions of the footing
foundation.
GAZETAS (1983) presented appfoximate formulae for static stiffnesses and impedance
functions of a circular surface foundation on a single layer over a horizontal rigid boundary as
shown in Fig. 5.8. The static stiffnesses are linear functions of the footing radius fo and an
inverse function of the layer thickness. The stiffnesses increase as the layer thickness decreases
because the soil system becomes stiffer. In contrast, the damping coefficients are inverse
functions of the static stiffness. The damping coefficients are much smaller than in the case of
an elastic half-space soil model, particularly at low excitation frequencies. The static stiffnesses
and the impedance functions of a strip foundation show similar results. Thus, foundation and
soil models greatly affect the foundation impedance.
5.3.3 Layer over Elastic Half-Space
Soil layering conditions naturally vary from site to site. A soil layer over an elastic
half-space model was discussed by LUCO (1974). The footing foundation was assumed to be
106
I
I
lAVER 1 \ I
' ..... -----/
LAVER 2.
lAVER l
LAVER ~
LAYlR 5
Fig, 5.9 Idealized Stress Distribution for a Layered Soils (Hll)
'v HORIZONtAL
DO 10 00 05 10 00
Io Io ------:;10
Io
~g
......r- JO .-t-
I
ill
47 41
I ,d/
Y
tt' 7
1%
I
b' 96.IS'
17
[ 1126.1l'
S9
V
159
I I'
2DB /
I
I lB
b - If !b
200
I lB f
334
lb 20B ~ J 1 aa
1M (BI (el
My ROCKINO - I TO SIG
00 05 10 10
o Io
r-F" l81
l' R' ...r 1
I I
JA' ~
- R' B2.Gl'
I 59
2b - - -
SR'
Jb ' - - - - - - - ; - I O ! - : - I - - - - - - - ' 200 f' JB' GR' 310

III Ifl
Fig. 5.10 Stress Distribution Beneath the Foundation Footing (Hll)

107
a massless rigid circular disk resting on the surface. The soil-foundation model is shown in Fig.
5.11. The solution of the equations of motion in cylindrical coordinates finally leads to the
force-displacement relationships. Plots of the stiffness and damping coefficients k's and c's
versus dimensionless frequency ~ for V./V.z = 004, p/pz = 0.85, VI = 0.30, Vz = 0.25 are
shown in Fig. 5.12 and Fig. 5.13 respectively. In comparison with Eq. 7.34 and Eq. 7.35 or
Fig. 7.6, 7.7 and Fig. 7.8, these stiffness and damping coefficients are more complicated.
HADJIAN & LUCO (1979) also discussed a layer over an elastic half-space soil model.
To simplify the problem, a rigid circular surface foundation was considered. The elastic half-
space and the soil layer were assumed to have a material damping of 3 % and 5 % of critical
damping respectively. The resulting stiffness and damping coefficients were similar to those
of Fig. 5.12 and Fig. 5.13 respectively. The effects of the layered system were; first, the
vertical and rocking vibration modes were the most and the least sensitive modes of the layered
system; second, the effect of the layered system would increase with increasing difference
between Vsl and V.z; third, the effect of the layered system decreases as the layer thickness
increases and fourth, if the layer thickness is more than 4.5 times the footing radius r o ' an
approximate elastic half-space could be used to replace the layered system.
GAZETAS (1983) also presented the response of a layer over a half space soil modeL
The approximate formulae for static stiffnesses based on the results provided by HADJIAN &
LUCO (1979) were presented. The corresponding dynamic stiffness k's and damping
coefficients c's can be found in Ref. H17.
S.3.4 Elastic Half-Space Soil Model
Several researchers have presented the dynamic response of a surface foundation

supported by a layer over a half-space soil model for circular foundationsH17.L8.L9 or rectangular
foundations W9,W22. These analyses cannot be generalized because of the number of parameters
that affect the soil-foundation response. These include the number, thickness and soil-properties
of the layers, footing shape and footing size relative to the layer thickness. As discussed in the
Chapter vn the effects of embedment of the foundation impedance should not be neglected.
The foundation impedance of an arbitrary shaped and partially embedded foundation has been
developed only for the foundation resting on an elastic half-space. The analysis of the layered
lOS
p.'lIt1
Fig. 11 Circular Surface Foundation on Layered Soils (L8)
'. '. '.

Fig. 5.12 Dynamic Stiffness Coefficients (L8)
"
5.13 Dynamic Damping Coefficient (L8)

109
soil system becomes more complex when the effects of the foundation embedment is taken into
account. Because of this complexity> the impedance coefficients of an arbitrary shaped and
partially embedded footing foundation resting on a layered soil-system have not yet been
developed.
To simplify the effects of a layered soil system on the foundation response, two
alternative approaches, namely an equivalent soil elastic modulus Ee<jSI9.HIl, and an equivalent
soil foundation stiffness K"qJ2,J3,J4,J5 were introduced. In the first approach, the elastic soil
modulus of the layered system as shown in Fig. 5.9 or Fig. 5.14 is transformed to an
equivalent single elastic modulus Ee<).' The soil model is equivalent to an homogenous elastic
half-space as illustrated in Fig. 5.15. The soil-foundation impedance is based on this equivalent
soil elastic modulus. In the second approach, the soil-foundation static stiffness for a Jayered
system is transformed into an equivalent elastic half space soil-foundation static stiffness Keq
by using an approximate formulae. This approach is limited to the calculation of an equivalent
static stiffness. The equivalent damping coefficients and the impedance coefficients are not
available.
The impedance of an arbitrary shaped and partially embedded foundation resting on a

layered soil model can be developed when the elastic properties in this layered system shown
in Fig. 5.14 are transformed to an equivalent soil elastic properties Eeq as shown in Fig. 5.15.
Therefore, the soil beneath an embedded footing foundation can be modelled as an equivalent
homogeneous elastic half space.
5.3.5 Fictitious Layer Over Elastic HaIf-Space
In Fig. 5.10 it is shown that even under static loading large soil stress distributions
occur immediately beneath a footing foundation and these stresses diminish with depth. The
depth where the higher stress occurs depends on the mode of vibration. Similar soil stress
distributions are discussed in Ref.S19.
PENDER (1983) reported that the zone of soil immediately beneath the footing
foundation will deform non-linearly as a fictitious layer, while the soil further away will deform
elastically as an elastic half-space. Non-linearity of this dynamic soil response corresponds to
the higher soil shear strain 'Y when compared with the smaller soil shear strain associated with
110
the elastic deformation. Thus, there are two zones of soil which exhibit different soil shear
strains. This fictitious layer over the half-space is shown in Fig, 5.20.
5.4 FOUNDATION FOOTING MODELLING
5.4.1 Modelling of Footing Shape
The modelling of shape of the footing foundation basically follows the development of
the footing foundation response analyses. In the early applications of the elastic half-space only
a simple shape of footing foundation could be used, but recently more complex modelling of
the footing shape has been employed.
5.4.1.1 Circular Foundation Footing
Historically, the soil-foundation vertical stiffness constant as expressed in Eq.

7.15 was derived from the theory of elasticity based on Boussinesq's stress distribution
assumptionsRS. These assumptions was based on the mathematical theory of elasticity for the
simple case of a vertical point load acting on the elastic half-space soil model. Among the
assumptions are that the load-deformation relationship obeys Hooke's rule and the stress
distribution in the soil will be distributed symmetrically with respect to the vertical axis. The
equilibrium of forces acting on an element soil a distance of r from the vertical axis and depth
z from the ground surface subjected to a.vertical loading leads to a set of differential equations
of equilibrium. These equations are generally more conveniently expressed in cylindrical
coordinates. Ref.L8 and L9 also used a circular footing for deriving the impedance functions
for a layer over an elastic half-space soil model. The circular footing is the simplest footing
shape.
5.4.1.2 Rectangular Foundation Footing
Most foundation footings are not circular but rectangular in shape. One
approach is to use an equivalent circular shape. This approach is used because of the difficulty
of finding the analytical solution for other shapes. However, results of experimental
investigations of rectangular foundations show that the use of this approach on the foundations
where the length to width aspect ratio LIB > 3 is not necessarily correctG1 , There are fewer
111
Surface
footing
H. v r=footing
ratio
n
v H=,,H,
1..<1
EGqtv
V
Fig. 14 Multiple Soil Layers Fig. 5.15 Equivalent Soil Properties
0.8 III
Fig. 16 Trench Effect Fig. 5.17 Trench Settlement Reduction

Factor vs Shape Dependent
18 Side-Wall Effect Fig. 5.19 Side-Wall Settlement Reduction

Factor vs Wall Base Area Ratio
112
results available for rectangular footing than for circular foundation shapes. WONG & LUCO
(1978) presented the impedance function of a rectangular surface foundation using a mixed
boundary value problem. Evaluation of the surface displacement vector by applying boundary
conditions and prescribing the surface stress components leads to a system of linear integral
equations. Due to a lack ofaxisymmetry, the solution of these integral equations may be
achieved by an approximate numerical method where the soil-foundation contact is divided into
a numbers of sub-regions. Assuming uniform contact stresses in each region are prescribed,
the force-displacement relationship for the footing foundation can be computed.
5.4.1.3 Arbitrary Shape Foundation Footing
The foundation footings can also be modelled as an arbitrary

shapeD3 ,D4,F4,Gl.Gl,H7,W2.1. It is not convenient to use analytical solutions to develop the soil-
foundation impedance because the soil-foundation contact surface cannot be generalized to the
regular coordinate systems. The approximate numerical solutions are generally used either as
a mixed boundary problem as described above or by using finite element or boundary element
methodsG2,H? ,P8.
5.4.2 Modelling of Footing Vertical Embedment
5.4.2.1 Surface Foundation: As for the modelling of foundation shapes, the

embedment modelling ofthe foundation footing follows the development ofthe analyses of soil-
foundation interaction. In early applications of the theory of an elastic half-space the footing
foundation was assumed to rest on the ground surface.
5.4.2.2 Fully Embedded Foundation: Knowing the type of continuum and using
the [mite element formulation G5 , a fully embedded foundation model was developed. The
embedded foundation transmits load to the stiffer soil than does the surface foundation.
However, in the soil foundation interaction, there is no guidance at what depth the soil shear
modulus G should be calculated. The variation of soil stiffness with depth for cohesive soils
in tenns of the soil shear modulus can be calculated by approximate formulae Gl2 ,
(5.1)
or more generally,
113
Gz = ( 1 + a~ )"I'i (5.2)
lJ
where G z and Go are soil shear modulus at depth z and at the ground surface respectively, B
is one-half of the foundation width, a is a parameter dependent on the soil condition and the
exponent m is a coefficient.
5.4.2.3 Partially Embedded Foundation: This foundation model was introduced

because the footing thickness is not necessarily the same as the depth of embedment and part
of the footing sidewalls might not be perfectly bonded to the soil backfill. It was found that the
effect of the degree of the embedment on the soil-foundation static stiffnesses was very
significant for rocking and torsion modes, substantial for the horizontal mode and only
secondary for the vertical vibration modeGS. The trench and side wall aspects G21 affect the
settlement of the partiai1y embedded foundation footing as shown in Figs 5.16 and 18
respectively. Figs. 5.17 and 5.19 correspond to the trench and side-wall settlement reduction
factors versus footing shape-dependent. Effect of these aspects on the soil-foundation
impedances will be discussed in Section 7.5.
5.5 THE SOIL-FOUNDATION MODEL USED IN TillS STUDY
As discussed in Section 5.3 the soil underlying the footing foundation can be modelled
as multiple-layers over an elastic half-space, a single layer over a rigid boundary, a layer over
an elastic half-space, an elastic half-space or a fictitious layer over an elastic half-space. Using
multiple-layers over an elastic half-space soil-model seems to be the most comprehensive
model. Unfortunately, the soil-foundation impedance of this soil-model is available only for a
surface foundation. In Section 5.3.4 techniques to treat the multi-layered soil model as an
equivalent elastic half-space soil-model are discussed. For simplicity, the soil beneath the
foundation in this study will be assumed to be an equivalent elastic half-space.
Most building footings are square or rectangular and are partially embedded
foundations. The soil-foundation impedance for this foundation model resting on an elastic half-
space will be discussed in Section 7.5. During dynamic loading the soil immediately below the
footing foundation will undergo higher shear strains than does the soil remote from the footing
base P19 This condition leads to a fictitious two soil layered system with an upper and lower
layer. The soil shear modulus G in the upper layer may degrade because of the higher soil
114
shear strains occurring in this layer. The lower layer can be assumed to possess its initial shear
modulus Gmllx> because the soil shear strain is relatively small. This soil-foundation model will
lead to a rectangular partially embedded foundation resting on a fictitious layer over an elastic
half-space as shown in Fig. 5.20. For the comparison, the soil-layer over a rigid boundary as
shown in Fig. 5.8 and the Winkler soil-foundation model is also considered.
5.6 THE SOIL-FOUNDATION MECHANICAL MODEL
As presented in Section 4.5, at least two important parameters should be considered

during the selection of soil mechanical model, the soil stress level and the soil water content.
For the simplified analysis of soil-foundation interaction discussed in Section 5.2.1 the soil
water table does not lie in the soil under consideration, so only a single soil water content was
considered. In Chapter VII, the soil-foundation impedance is assumed to be independent of the
soil water content. The soil mechanical model, therefore can be selected by considering only
the stress level in the soil model under consideration.
The higher soil stress that occurs in the soil immediately beneath the footing foundation
as shown in Fig. 5.10 is associated with the higher soil shear strain 'Y exhibited in that soil
layer. The degrading normalized soil shear modulus G/G max with soil shear strain 'Y .. for a
typical Auckland, New Zealand, clay soil was presented in Fig. 4.13. Under dynamic loading
such as in earthquakes, the soil shear modulus will degrade and the soil material damping will
increase with an increase in the level of soil shear strain. The classical question is, how high
are the soil shear strains induced by earthquake excitations? If an approximate maximum soil
shear strain 'Y.s of Eq. 4.17 is adopted, the likely approximate maximum soil shear strains
induced by a particular earthquake can be computed from Table 4.5. From the approximate soil
shear strain, the approximate soil shear modulus G can be estimated.
As shown in Table 4.5 the approximate maximum value of soil shear strain 'Y varies
from 0.46x1(J4 to 2.82xlO-4. As discussed in Section 4.3 and illustrated in Fig. 4.23 the level
of soil shear strain indicates what method should be used during the analysis of soil-structure
interaction. When the soil shear strain is in the range of 10-4 to 10'2, the use of an equivalent
linear method is justified. An equivalent linear or apparent shear modulus G 1 as suggested by
Ref.G12, P19 and P20 is used in this study. As discussed in Section 4.3 and as illustrated in
Fig. 4.28 and Fig. 4.29, the threshold soil shear strain for clay soil is around 10'), greater than
1
P,
Inelastic; soli ~ ,0, , V D P
5.20 Fictitious layer over Fig. 5.21 Mechanical Model of Soil

an Elastic Half-Space
Isotropic equivalent homogeneous

1I1$(;o ..l8$tlc halfspacs
Fig. Equivalent an Elastic Half-Space Fig. 5.23 Equivalent Soil Mechanical Model
12,----,-.......,..---,--......--.,....--,---..-....,...--, 100 , - - - - - - . - - - - -............., - - - - - - ,

LICK CRE~ StLT
80 I- a
60 I-
~ 40 I- / / Cyefes 1 & <40

6. 20 I-
il:! 0.0 f--------~
:<l 20
'"
&:.
(f) .40 -
60 I'
I1i G 10 It 14 ,S IQ
~~: L---'-_..L.-_ '_~_....L____'_....L..._ L..- - , _
..
$ffiAJN AlMI'UTUiJ IN 104 IN/IN 3.0 2.0 .1.0 0.0 1.0 2,0
StlOlf IlrO/ln (%)
Effect of Loading Frequency Soil Stress-Strain Relationship

on the damping Ratio (lI5) at Small Soil Strain (P20)
116
the soil strains in Table 4.5. As a consequence, the effect of the number of cycles of loading
on the degrading soil shear modulus can be ignored and the reduction of soil confining pressure
(Jo during cyclic loading would not be significant.
In this study the frequency independent response analysis was used where the response
analysis considers only a particular value of excitation frequency. By considering the use of the
soil model as described in Section 5.5, the proposed soil-foundation mechanical model is shown
in Fig. 5.21. The mechanical model consists of a series of two Voigt-Kelvin models. The upper
Voigt-Kelvin model is associated with the behaviour of the fictitious upper layer, and the lower
Voigt-Kelvin model refers to the behaviour of the lower soil-layer. This model is the
simplification of the soil-foundation model as presented by PENDER (1983). As shown, Pi is
associated with the applied load corresponding to the ilh displacement mode. The terms K J and
~ are the upper and lower soil-foundation stiffnesses associated with the apparent shear
modulus G J and the maximum soil shear modulus Go = Gmax. respectively. C 1 and Co are the
damping coefficients for the upper and lower layers respectively. In this case, the damping
coefficients include the radiation damping and material damping ofthe corresponding soil layer.
The effects of soil material damping on the soil-foundation stiffness and damping coefficients
will be discussed in Section 7.6.
5.7 MATHEMATICAL REPRESENTATION OF SOIL-FOUNDATION MODEL
The soil models shown in Fig. 5.20 and Fig. 5.21 are not yet mathematically tractable
for practical use in soil-foundation interaction analyses. Analyses of soil-building interaction
using lumped parameter models need soil-foundation impedances. These soil-foundation
impedances can be computed by transforming the proposed soil-mechanical model shown in
Fig. 5.21 to those shown in Fig. 5.22 or Fig. 5.23. Terms K., and Ce in Fig. 5.23 are the
equivalent soil-foundation stiffness and damping coefficients respectively.
To be able to transform to the equivalent soil-mechanical model, the equilibrium

between series of two Voigt-Kelvin models as shown in Fig. 5.21 must be satisfied. This leads
to,
dYl (S.3a)
Pi = P = Kl Yi + C1-
dt
117
(S.3b)
where K1 , 1(." C 1 , Co are the stiffness and damping coefficients for the upper and lower layers
respectively and Yt. Yo are the displacements of the upper and lower layers respectively.
Taking the first derivative of the displacements of both Voigt-Kelvin models in terms
of the element displacements with respect to time yields,
1 liP (S.4a)
1 dP (S.4b)
The total displacement of the system is the summation of the displacements of each
element yields,
Y = Yl + Yo (S.Sa)
fly = dYl (S.Sb)

dt dt
Substituting Eq.5.4.a and 5.4.b in Eg.5.5.b and after rearrangement in terms of the
global displacement y leads to,
dP Xl Ko dY-I- Ko + Co Xl (}/-y
(5.6)
dt Xl -I- Xo dt Xl + Xo &1
Integrating Eg. 5.6 yields,
P=K y+C dy (5.7)

61J 6g dt
where Keq and Ceq are the equivalent stiffness and damping coefficients of the system as shown
in Fig. 5.23 from which,
118
(5,8a)
C~q (5.8b)
The series of Voigt-Kelvin mechanical models of the foundation system as shown in

Fig. 5.21 then can be transformed to the single equivalent Voigt-Kelvin model as shown in Fig.
5.23. Eqs. 5.8.a and 5.8.b show the mathematical expression of the equivalent single Voigt-
Kelvin modeL Given K" Ko, C, and Co, the equivalent soil-foundation impedance then can be
computed.
5.8 MODELLING OF SOIL-FOUNDATION DAMPING
5.8.1 Modelling of Soil Material Damping
Soil material damping can be modelled as either hysteretic or viscous damping. The
choice will be determined by the dependency of the cyclic soil damping on the loading
frequency. If the soil cyclic damping is frequency dependent the use of the viscous model is
justifiedc8.P19. If the soil cyclic damping is frequency independent the hysteretic model is
suggested.
PENDER (1983) stated that the energy dissipated per cycle in the soil is independent
of the loading frequency. HARDIN & DRNEVICH (1972) reported that the damping ratio of
dry cohesionless soils were almost unaffected by loading frequency, but were slightly affected
in the case of the cohesive soils as shown in Fig.5.24. Fig. 5.25 shows that at a shear strain
'Y" :::::: 0.01 % the cyclic behaviour of volcanic ash soil with plasticity index PI = 25 % is
nearly elasticP2. Accordingly, the material damping of the soil can be modelled as a viscous
damping.
5.8.2 Modelling of Soil-Foundation Radiation Damping
The radiation damping generated by the soil-foundation system has been discussed in
the proceeding chapter. It was clear that this damping is strongly frequency dependent and is
represented by terms of dynamic stiffness damping coefficients k's and c's respectively. These
119
coefficients are dependent on the dimensionless frequency a., in which is dependent on the
loading frequency.
The soil-foundation radiation damping is modelled as viscous damping by all

researchers. Since this damping is radiated to infinityl'19, the damping phenomenon is
represented in the proposed soil-foundation mechanical model as illustrated in Fig. 5.21. The
difference between C 1 and Cfj in this figure is because of the different contributions of the
material damping in the upper and lower layers. The contribution of the soil material damping
on the soil-foundation stiffness and the damping coefficients given by Eq. 7.98 and Eq. 7.99
is adopted.
5.9 SUBSTRUCTURE PARAMETER QUANTITIES USED IN THIS STUDY
The substructure parameters can be categorized in three groups, the soil properties, the
physical dimensions of the footing foundation and the formulae used in the design of the
foundations. Obtaining the representative values for the critical soil properties is the most
difficult task in the substructure des ignRS , requiring support by intensive laboratory or field
testing. In general, the soil may not be purely cohesive or purely cohesionless but possesses
both an angle of internal friction cb and cohesion c. This type is usually called a c-cb soil and
is assumed in this study.
The Poisson's Ratio of the soil is one of the important parameters as the stiffness is
directly affected by this parameter. The Poisson's Ratio of the soil" = 0.38 was used in this
study. This value is associated with partly saturated c-cb soils. Another soil parameter is the
angle of internal friction cb. For a c-cb soil, this parameter is affected by the value of the soil
plasticity index PI. Ref.Mll shows that for PI = 20 % - 30 %, the angle of internal friction
cb varies in the range of 19o to 36. Plots of the angle of internal friction 4> against the soil
plasticity index PI was presented by Ref.B3 and Ref.Bg. From these graphs and for the
Auckland clay soilPW with the plasticity index of PI ;::: 20 - 30 %, the angle of internal friction
cb = 22 was selected.
There is no clear guidance on how to choose the soil cohesion c. The soil cohesion may
as low as 0.20 kg/ cm2 and as high as 1.14 kg/cmz J3. A relatively high value of soil cohesion
c = 0.7 kg/crrr was taken in these analyses. A similar difficulty was also found during the
120
choice of soil density 'YIIV with depends on the type of soil and the water content. Data from
Ref.E5 indicated that the soil density varies from 1350 kg/m3 to 1740 kg/m] and 1815 kg/m3
to 2050 kg/m3 for dry and wet soils respectively. The soil density 'Ysv = 1800 kg/mJ was taken
in these analyses. The frequency excitation w = 15.71 rad/sec is used.
The static safety factor SF for the foundation covers the uncertainties of the structural
simplifications in the structuralmooelling. the loading, the analysis, the construction and the
maintenance. Foundation design has a greater degree of uncertainty than in man-made materials
such as steel or concrete. Ref.B3 states that the factor of safety for cohesive soil is considerably
higher than that for cohesionless soil because the former is more state dependent. Ref.M19
suggests that for building foundations the safety factor SF """ 2 - 3 may be used for normal
static service load conditions. The static safety factor SF =3 was used in this study.
The footing foundations of the structures under consideration were assumed to have
rectangular shapes with a length-width ratio less than 1.75. The ultimate bearing capacity of
the soil beneath of the foundation was calculated using Terzaghi's equation as presented in
Ref. WS. The size of the foundation was adjusted so that under the applied vertical load the
static safety factor SF =3 was maintained.
The soil shear wave velocity V. was calculated by Eqs. 4.9 and 4.10 and by the
approximate formula of Eq. 4. 13. The maximum soil shear modulus GmllX was calculated by
using Eq. 4.2 and the approximate formula presented by Ref.B. The soil confining pressure
(Jo of the normally consolidated soil was calculated by the method presented in Ref.P23, where
the coefficient earth-pressure at rest Ka was calculated fromD9.Mll,
(5.9)
The use of some of the formulae for the soil shear wave velocity V. and the maximum
soil shear modulus GUlIlX was to cross-check the computed values for a given soil void ratio e
and an approximate N-SPT blow number. The soil void ratio e for the 12 and 18-storey
foundations varied between 0.7 to 1.2 which is within the range of the validity for the
corresponding formula discussed in Section 4.2.1.2.
In the upper layer, the soil material damping of 1 % and 10 % are corresponding to
the maximum of shear modulus G = Gmax and G = 0.50 G max respectively. Ko and Co were
121
computed from G max and the damping coefficient at small shear strain. Kj and C 1 were
computed for any value of shear modulus and damping coefficient as a function of the shear
strain. The equivalent stiffness and damping coefficient ~ and Ceq were computed using Eq.
5.8.
5.10 SlJMl\iIARY AND CONCLUSIONS
The soil-building interaction becomes significant when the superstructure is relatively

stiff while the supporting soil is relatively soft and when these conditions are met the effects
cannot be ignored.
Unfortunately, the soil usually has a very complex geometry and properties. This
complexity ofthe soil-foundation interaction may also arise from both the foundation shape and
the degree of the embedment. Accordingly, some simplifications and assumption have been
made including the soil water content, effects of the soil backfill, coupling between vibration
modes, footing flexibility and coupling between adjacent footing.
Among the possible soil models, the fictitious layer over an elastic half-space was the
main soil model used for this study. The foundation selected is a rectangular partially embedded
foundation model. The complete soil-foundation model, therefore, is a rectangular partially
embedded foundation resting on fictitious layer over an elastic half-space. The soil layer over
horizontal rigid boundary and the Winkler soil model were used for a comparison with the
principle soil-model.
The proposed soil mechanical model consists of a series of two Voigt-Kelvin models
representing the fictitious upper layer and the lower half-space respectively. The mathematical
expressions for this series is conveniently transformed to a single equivalent Voigt-Kelvin
model with equivalent spring stiffness ~ and damping coefficient Ceo The material damping
of the soil and the radiation damping of the soil-foundation system are modelled as a viscous
damping model.
The soil parameters included the angle of internal friction , the cohesion c, Poisson's
Ratio P, the soil density "tsv, the static safety factor SF and the coefficient of earth-pressure at
rest Ko were based on the appropriate data from many references.
Chapter VI
Superstructure Modelling
6.1 INTRODUCTION
As described in Chapter III, the earlier steps in the building design will be emphasized
in the building configuration especially if the building is in a seismic area. Among the
impOltant parameters in the building configuration are the selection and location of the lateral-
load resisting systems. It is desirable that both moment resisting frames and the combination
of structural frames and walls (frame-walls) are used as the horizontal force resisting systems
in the building prototypes. SCHUELLER (1977) and ARNOLD & REITHERMAN (1982)
briefly discuss building configurations for seismic areas.
If the combination of two different horizontal load resisting systems such as cantilever
walls and frames is utilized, it is necessary to clarify both the individual and combined
behaviour of these systems. The problems arising from inelastic analysis is more complex than
for elastic analysesM2, and the identification of these problems is important.
6.2 IDENTIFICATION OF PROBLEMS IN INELASTIC STRUCTURAL ANALYSES
Assuming that the floor diaphragms are rigid in their planes, the lateral displacements
of the frames is a storey shear mode, implying that the floors virtually remain horizontal while
the beam column joints can rotattf6. The typical deflected shape of a frame structure is shown
in Fig. 6.2. In contrast the cantilever wall deflected shaped is much more influenced by a
flexural deformation as shown in Fig. 6.3. Therefore frames and cantilever wall structures have
different deformation characteristics.
If these individual deformation characteristics are linked together as a hybrid frame-wall

structure, the interaction between the two different horizontal deformations cannot be avoided.
The defected shape is shown in Fig. 6.5. Redistribution of the lateral load occurs, often with
large interactive forces at top and bottom levels of the structure as shown in Fig. 6.6. The final
122
123
deformation shape of the overall structure, generally depends on the ratio of stiffness between
the two components. The interaction between the frames and structural walls leads to complex
analytical problems RB
Building codes generally prescribe design static lateral loads which are considerably
smaller than those determined from elastic dynamic analyses. Under earthquake excitation, the
response of building structures is often assumed to be inelastic rather than elastic. The
structures will survive provided the individual members are ductile enough to sustain the
deformations without premature failure. Intensive research on the inelastic behaviour of sub-
assemblages of concrete members have been carried out over the past few decades.
In the inelastic analysis of building structures, several fundamental problems need to

be identifiedM20,M2S.M26. These problems include modelling of inelastic behaviour ofthe concrete
members, yield capacity reduction, p-delta effects, damping of the structural systems, rigid
zones at the joints and plastic hinge lengths, selection of critical ground motions, damage
parameters and damage indices. The selection of the numerical integration technique is
important because an inelastic dynamic analysis requires a more refined numerical integration
technique than is necessary for elastic analysis. This covers the accuracy and stability of the
integration schemeB5.W15.
The uncertainties arising from the building model and earthquake excitation mechanism
makes the inelastic dynamic analyses more complex. Additional problems include the effect of
non-structural members on both the structural stiffness and damping, floor slab flexibility,
torsional effects in the building, earthquake orientation and local site responses.
6.3. GENERAL ASSUMPTIONS IN THE ANALYSIS OF THE SUPERSTRUCTURE
6.3.1 The Effect of Non-structural Components
The non-structural elements can be defined as those parts of the building that
theoretically do not contribute to the structural strength and stiffness. These components include
ceilings, ducts, piping and components such as infil1 walls, doors, windows, internal partitions
or cladding. The infill components may make the frame stiffer than expected. The stiffening
effect will depend on their flexibility and construction details. Light internal partitions, doors
and windows may not be significant, but the stiffness or strength of the materials such as
124
Earthquake
direction
PI
Frame
!
Beams
:3
2
1 I I I
Column
l!1.
,. .n P! """'1' .l ~ J,
Fig. 6.1 Plan of the Frame-Walls Fig. 6.2 Deflected Shape of the Frame
'---.
me
Fig. 6.3 Flexural Deflected Shape of the Structural Wall
n
.
:3
2
1
Fig. 6.4 Free-standing Wall in the Fig. Deflected Shape of Frame-Walls

of Frame-Wall Structures
125
concrete masonry may significant. The infill components may be fully integrated or separated
from the structural resistant system as discussed in Ref.D9. If the separation between the infill
panel and the structure is insufficient then contact may cause damage. The cost of damage to
the non-structural elements often exceeds that of the structural elements D9E6
Effects ,of infill walls on the building responses have already reported by some
researchers. Ref.Lll reported the effect of infill walls on the vibrations of 10 and 4-storey
buildings. It was found that the fundamental period of the building with infill wans are much
smaller than those without infill walls. Ref.W25 reported that the effects of infill walls on the
building response were more pronounced in buildings with small fundamental periods. Both the
horizontal floor displacement and hysteretic energy absorbtion for buildings with infill walls
were much smaller than those without infill walls. Ref.E6 and S28 discussed the column
damage mechanisms in infilled frames during earthquakes. It was demonstrated that the
stiffness and storey shear of the storey with infill panels can be several times greater than that
without infills. The sudden shear failure of the infill may lead to impact loads being applied
to the columns causing damage to the columns. The effects of infill panels may also lead to
failure of the columns in shear because of reduced effective column length.
BERTERO & BROKKEN (1983) summarized of the results of experimental

investigations in Ref.B16 and K12. The infill walls increased the effective inter-storey lateral
stiffness by a factor of 4.66 and 10.94 for reinforced clay brick and lightweight concrete infills
respectively. The fundamental period ratios of infill walls and open frames of 0.4 to 0.6 were
found. PRIESTLEY & CALVI (1992) and PAULAY & PRIESTLEY (1992) presented that the
effects of the infill wall at relatively large lateral deformations become complex because the
frame would deform in a flexural mode while the infill attempts to deform in a shear mode.
The separation between frame and wall at the comers of the tension diagonal would develop
a diagonal compression strut. In routine analyses and design including that in this study, the
contribution of the non-structural element on the structural stiffness and strength is often
neglected.
6.3.2 Floor Diaphragms and Location of Structural Walls
In reinforced concrete frame design the floor slab is often assumed to carry gravity load
only. These gravity loads are specified by Codes. This floor loading is transferred to the
supporting beams using a tributary area. The floor diaphragm must be stiff enough to distribute
126
the lateral floor loads. In wall-frame structures the effect of diaphragm stiffness on the floor
horizontal displacement depends on the location and orientation of the structural walls in the
building as discussed by MUTO (1975). It was found that the floor horizontal displacement
decreased as the distance between walls decreased and the stiffness of the frames increased.
The plan and location of walls in the building prototypes are arranged
symmetrically as shown in Fig. 6.1. The distance between two structural walls is relatively
small so that the difference in the floor slab horizontal displacements along the building can be
neglected.
6.3.3 Torsional Effect
Torsion is one of the problems in the 3-dimensional response of building structures.

A precise 3-dimensional analysis is only required in a limited number of building structures,
such as buildings which are irregular in plan. The analyses of 3-dimensional buildings have
been conducted by many researchers, investigating the torsional provisions in building
codes H16 ,T15,T16 and torsional effects on the response of building structures C24,Kll,W18,W20. Ref. W18
presented the distribution of lateral load resisted by the frame and wall components in a 3-
dimensional building analysis when the building was subjected to a static lateral load acting at
every floor level. Ref. W20 presented the 3-dimensional analysis of a lO-storey steel structure
with varied concrete structural wall eccentricity ~. The building was subjected to applied
horizontal lateral load acting at every floor level. This study showed that torsion of the building
would reduce the ultimate load-carrying capacitY of the structure, increasing the horizontal
displacement of the outer frames and increasing the p-delta moments. CHOPRA & HEJAL
(1988) also discussed the earthquake response of asymmetric building frames. The building
prototype was a 5-storey reinforced concrete frame. Lateral-torsional coupling effects, in
general, would decrease the base shear, base shear moment, top storey lateral displacement of
the centre of rigidity but increase the base torque. The effects on multistorey buildings would
depend on the beam to column stiffness ratios.
In regular and symmetric multi-storey building frames, the 3-dimensional analysis can
be replaced by a series of 2-dimensional planar frame analyses in the x and y
directions C21 ,Dlo,M25. In frame-wall structures, the role of wall location and orientation is very
important. Orientation of the wall shows whether the walls will act along the weak axis or the
strong axis of the building for a particular direction of earthquake excitation, The location of
127
walls within the building plan is determined by the function of these walls. To perform well
in resisting torsion, the location of the walls should be symmetric and relatively far away from
the centre of mass. In this study, the frames and walls are located symmetrically and assuming
that the floor slab is infinitely rigid, torsional effects will not be evident in the wall-frame
structures.
6.3.4 Earthquake Orientation
Forces generated by an earthquake act in all three dimensional axes. Building are
designed to sustain normal gravity loads. Generally, the building vertical resistance possesses
a sufficient factor of safety so that it can sustain the extra vertical seismic forces elastically.
Only the effects of the horizontal earthquake forces are taken into account in design. For
buildings with simple symmetric geometries a simple analysis may be used. The forces can be
resolved into two sets of perpendicular lateral forces in the planes of the lateral load resisting
system.
6.4 MODELLING ANALYSIS OF THE STRUCTURAL SYSTEM
In frame-wall structures the structural wall is normally located as parts of plane frames
as shown in Fig. 6.1. There are two types of structural system, in-plane frames and in-plane
wall-frames. Under applied lateral loads, both of these systems will interact with each other
and the deflected shape represents the behaviour of the total building systems. In analysis,
however, interaction between frames and walls has to be defined.
6.4.1 Modelling of In-plane Wall-Frame Interaction
In-plane interaction between the frames and walls was presented by CARDAN (1961),
using inverted triangular, uniform, and concentrated horizontal loads acting on the structures.
Considering equilibrium conditions between external and internal forces acting on the wall leads
to a second order differential equation of unknown wall rotation at any height of the walL The
deflected shape of the wall may be determined by solving this differential equation. From the
wall rotations, the internal forces can be calculated. Similar results were presented Ref.RB.
KHAN & SBAROUNIS (1964), presented in-plane interaction between frames and
walls using iteration procedures. The frame-wall structure was represented by two different
128
systems, namely wall and frame systems. During the analysis, the compatibility of the
structural deformation must be satisfied. Applying an assumed joint displacement and rotation
to the frame gives the resulting shears and moments. These shears and moments are then
applied to the wall system giving new wall displacements and rotations. The net structural
deformations are found from the difference between the assumed and resulting deformations.
The iterations continue until compatibility of the structural deformations is achieved so that the
wall deformations are same as the frame deformations.
Interaction between frames and walls was also discussed by MUTO (1974), using a
slope deflection compatibility approach. Finite difference equations were used to develop the
Principle of the Three Moment Equation. Application to regular N-storey wall-frames structures
gives N equations of unknown wall joint rotations. The deflected shape of the wall may be
determined by solving simultaneously these N non-homogeneous equations. The wall bending
moments and beam moments were then able to be calculated.
All of the frame-wall interaction described above are for static elastic analyses. In
dynamic analyses where the stress of the material may exceed the elastic range, the application
of the approaches mentioned above are no longer valid. In this study, the in-plane wall-frame
possess only one column as shown in Fig. 6.3. The stiffnesses of the columns are very small
when compared to the wall stiffness and the contribution of the columns to the frame-wall
stiffness is insignificant and may be neglected. The frame-wall structure is represented by the
free standing waIl G8 ,G9.PI2,PI8 as shown in Fig. 6.4 and is modelled as a beam line element.
6.4.2 Modelling of Out-or-plane Frames and Frame-Walls Interaction
If the building has rigid floor slabs and the location of the frames is symmetric, then
under an applied horizontal excitation all frames will be subjected to equal displacements at any
given floor level. In this case, the entire building can be analyzed by considering single plane
frames connected together by rigid IinksC2I ,Dlo at all storey levels as depicted in Fig. 6.8. In
frame-wall structures, similar analyses for the out-of plane modelling may be used.
6.5. INELASTIC MODELLING OF REINFORCED CONCRETE MEMBERS
Earthquake resistant design philosophy assumes that under large earthquakes repairable
damage of the structural members is justified but the building must not collapsed. To achieve
Moment frame
\
Interacting
forces
/
Fig. 6.6 Interacting Forces of the Frame and Wall
II io
to
"
N
N
@I
M
22" X 22" Exterior Columns

20" X 24" Spandrel Beams All Around
Fig. 6.7 Frame-Walls Building Plan
3@:;>2'= 66'
Frame T1 Frame "2 Frame T3

(4 Inlerior Framll"'l (2 E~lerior Frames) (2 Interior Frames)
Fig. 6.8 Out-of-Plane Modelling Analysis

of Frame-Walt Structures
130
this goal the premature or brittle failure of the structural members should be avoided and the
lateral-load resisting systems should be ductile. The structure should be capable of developing
sufficient ductility and energy absorbtion capacity without significant degradation of its strength
and stiffnessP18 ,P21,
Under a large earthquake the concrete structural members may undergo large inelastic
deformations followed by the developing of member-end plastic hinges. This inelastic
deformation is usually associated with transverse cracking along the length of the member as
shown in Fig. 6.905 ,P1B The member flexural rigidity (EI) now varies along the member
length. Under cyclic loading, energy will be dissipated by member plastic hinges. The amount
of the energy dissipated will depend on the intensity of loading, loading rate, geometry of the
section and material propertiesP21
The member inelastic flexural rigidity and the energy dissipated by inelastic members
are important and should be clearly modelled mathematically. Parameters are the modelling of
a member element, modelling of the plastic hinge hysteresis loop and the distribution of
inelastic member flexural rigidity.
6.5.1 Modelling of the Member Element
The modelling of the member element and its hysteresis loop were a major discussion
point among civil engineers at the beginning of the 1960s. Why were the analytical elastic
structural response quite different from observed response. cThe dynamic lateral forces
developed in the structures during an earthquake were considerably in excess of the design
static lateral loads, but no significant damage was found in the structures even when they were
subjected to relatively large earthquakes. The engineers concluded that the great differences
between the observations and the elastic analyses were due to the fact that the members and
their hysteretic energy dissipation were not properly defined P25
6.5,1.1 Two-component Model
CLOUGH et.al (1965) developed the bi-linear member force-deformation

(hysteresis loop) modeL This model physically consist of two fictitious members as shown in
Fig. 6.10. One of the members was assigned to remain elastic so as to represent the strain
hardening behaviour and the other was elasto-perfcctly plastic to represent the yielding
131
phenomenon. When both of the members were in the elastic range the initial member flexural
rigidity is the sum of the flexural rigidity of two elements. When a member-end yielded the
resistance-deformation at this plastic hinge is governed by the elasto-perfectly plastic spring
behaviour. The member flexural rigidity is then only represented by the elastic member
representing the strain hardening effect. After unloading the yielded member-end returns to the
elastic state giving an unloading flexural rigidity as the same as the initial flexural rigidity.
The hysteresis associated with this member is a bi-linear loop as illustrated in

Fig. 6.17. In terms of structural analysis, this member model is realistic because the inelastic
deformation of a member-end is dependent on the moment acting at both member-ends.
However, the disadvantage of this I:Jlember model is that the only the bi-linear resistance-
deformation relationship can be represented. Application of this member model on the dynamic
response of two 20-storey open frame structures with fundamental periods of 1.6 and 2.77 sec.
was carried out. In general, the maximum horizontal displacements of the inelastic structures
were greater than those of the elastic structures, which is contrast with the results in Ref. V9.
6.5.1.2 Multi-Component Model
As reported by Ref.OS and Ref.OlD, this model was introduced by AOYAMA

& SUGANO (1968). The member model consists of three or four imaginary parallel elements
connected to a rigid bar at each end shown in Fig. 6.11. This model was developed from the
two component model as described above. Normally only one member is assigned to be elastic
and when the yield level is reached at the member-ends the other members are treated as elasto-
perfectly plastic. This model can represent flexural cracking, different yield levels at each end
of the member ends and the strain hardening effect.
6.5.1.3 One-component Model
This model was introduced by Giberson (1967) for the dynamic response of
multistorey buildings subjected to earthquake excitation. This model was introduced because
the above models could not account for varying member stiffnesses depending on the member
stress level. The model consists of a single elastic member with independent inelastic springs
at each member-end, each of which is assigned the resistance-deformation behaviour of the
plastic hinge at the member-end. This model is illustrated in Fig. 6.12.
132
In this model inelastic deformations at the member-ends are independent of each

other, which means that the inelastic deformation of a member-end is not affected by the
moment acting on the other member-end. Hence, any flexural hysteresis loop can be assigned
to the inelastic spring 05 and this is the advantage of the model. This is also the disadvantage
of this model because the member-end deformation should depend on the moment acting at both
member-ends. Furthermore, it is not always rational to lump all of the inelastic deformation
at the member-ends 05
6.5.1.4 Discrete Element Model
This model consists of either lumped inelastic stiffnesses as introduced by

Ref.W26 or distributed inelastic stiffnesses introduced by TAKA YANAGI & SCHNOBRICH
(1976)05. These member models were introduced because the actual inelastic member
deformation is spread out along the member rather than lumped at the member-ends. A
concrete member may be subdivided into several short segments with high inelastic
deformations and long segments for relatively low or nearly elastic member deformations as
represented in Fig. 6.13. The inelastic springs which represent the short line segments are
usually modelled as degrading stiffness hysteretic models. In the distributed inelastic stiffnesses
the short and long line segments can be assigned different flexural rigidities to represent the
degree of cracking under dynamic loading. These models may give more accurate results
especially in wall members, but the computational effort will increase with the number of
segments.
6.5.1.5 Distributed Flexibility Model
As mentioned before, in reality the cracking of the member is distributed along

the member length. Instead of dividing the member into several segments a parabolic
distribution of flexural flexibility is assumed as shown in Fig. 6.14. The inflection point of the
parabolic distribution of the member flexibility is assumed to remain elastic while any type of
hysteresis model may be assigned to the plastic hinge at each end of the member.
Several models representing the inelastic behaviour of concrete members have been
described. Each model has advantages and disadvantages, and the choice depends on many
factors. Ref.05 stated that a single element model can be expected to give only reasonable
results for relatively low rise frame structures, where the inflection point of the member is
133
nearly at the member-midspan. But Giberson (1973)10 stated that one-component model was
more versatile than the two component model. The main reason was that any type of hysteretic
model could be used for the inelastic springs, while the two-component model could only
represent a bi-linear hysteresis.
Because of its characteristics, the one-component model was used by many researchers
to represent the inelastic behaviour of the concrete members for SDOF concrete structure06 ,
laboratory tests of SDOF concrete structureS14 , or analysis and laboratory tests of multistorey
concrete framesslS.s29. The one-component model is employed in several computer programs
such as SAKE (Otani,1974), DRAIN 2D (Kanaan & Powel,1973), SHARPE (1973), or
RUAUMOKO (Carr,1993).
6.5.2 Hysteresis Modelling of Reinforced Concrete Member
As discussed in Section 6.5.1.3 the inelastic springs in the one-component model can
represent the hysteresis loop of the member plastic hinges. Modelling of the member hysteresis
loop then becomes very important because of its affect on the structural hysteresis energy
dissipation. This structural hysteresis energy is not only an important aspect in inelastic
resistant design of building structuresC25 but also can be used for damage parametefZ-3 and
damage index identificationC29 M23 ,M32.P27,l'2ll.
The member hysteresis loop observed from laboratory tests must be idealized in the
theoretical hysteresis model as shown in Fig. 6.15. The shape of the real hysteretic loop of the
concrete member depends on many factors P21 ,OlO, and the idealized member hysteresis loop
simulates the real behaviour will also depend on those factors. Most of the concrete member
hysteretic loops represent tlexural05 ,OIO, rather than shear or slip behaviour. A number of the
concrete member flexural hysteresis models are described below.
Elasto-Perfectly Plastic: This hysteretic model was used for the inelastic response of building
structures in the late 1950's and the early of 1960's as ,described in Section 6.5.1. The
hysteretic model is very simple and is shown in Fig. 6.l6. When the member-end moment
reaches yield, the member-end is assumed to undergo perfectly plastic deformation without any
increase in resistance. On unloading, the member returns to the elastic state with the unloading
stiffness equal to the initial stiffness. A similar behaviour will occur when the member-end
reaches yield in the opposite direction. This model has no strain hardening. This hysteretic
134
model does not dissipate energy until yielding occurs and there is no hysteretic energy
dissipation during small amplitude motion. As a result this hysteretic model does not properly
simulate the nonlinear behaviour of reinforced concrete membel'ssl4.
Early studies of the inelastic response of building structures were repOlted in Ref. V9,
ReLP25 and Ref.P26. The resistance-deformation ofthe SDOpV9 ,P25 and MDOP P26 columns was
assumed to be elasto-perfectly plastic. Ref. V9 concluded that the maximum horizontal
displacement for elastic structures with fundamental period TI = 1 sec. is the same as that of
an inelastic system. Ref.P25 and Ref.P26 concludes that the plastic deformation of the member
provided much structural damping. This increase in structural damping may reduce the
response of the structure during strong motion earthquakes.
Bi-linear Model: This hysteretic model was a development of the elasto-perfectly plastic model
by considering the effect of member strain hardening. This model is also represented by the
inelastic two component model introduced by CLOUGH et al.(1960). This hysteretic model is
illustrated in Fig. 6.17. The post-yield stiffness ~y is the effect of the strain hardening and is
usually expressed in terms of the initial stiffness Ky by the coefficient a as,
Kpy = a Ky , a: ~ 0.02 - 0.10 (6.1)
The bi-linear model, in general tends to dissipate much more hysteretic energy than observed
in tests. Ref.05, 010, 011, S14 and Ref.S29 stated that this model did not satisfactorily
simulate the inelastic behaviour of reinforced concrete members.
Clough Model: This hysteretic model was introduced in 1966 after considering the tests of
beam-column assemblies by the Portland Cement Association (PCA). Results of the tests
indicated that even though no strength degradation appeared during repeated loading, significant
degradation of the reloading stiffness was found. It was also evident that less energy dissipation
per cycle of loading resulted from this test as compared with an ordinary elasto-perfectly plastic
system. As the result, the ductility demand might increase. This hysteretic model is shown in
Pig. 6.18. The unloading stiffness was assumed to be the same as the initial stiffness, and the
degrading reloading stiffness indicated by the slope from the residual plastic deformation to the
maximum opposite displacement. The performance of this model was an improvement
compared with the hi-linear model, because under small amplitude loading cycles energy can
135
Possible hinge
location
Fig. 6.9 Inelastic Response of the Rein- Fig. 6.10 Two-component Member Model
forced Concrete Beam (05)
(loslle (cmpon~nts.
"--~,,-- L ------1
Fig. 6.11 Multi~component Member Model Fig. 12 Singe-component Member Model
lol
to!
6.13 Lumped and Distributed In- Fig. 6.14 Distributed Flexibility Model
elastic Member Model
Restoring
loree
R
Fig. 6.15 Real and Ideal Hysteresis Loop Fig. 6.16 Elasto-plastic Hysteresis Model
Fig. 6.17 Bi-linear Hysteresis Model Fig. 6.18 Clough Hysteresis Model
136
be dissipated. The model is relatively simple and has been used extensively in the nonlinear
analysis of reinforced concrete structures0 5,SI4.
Takeda Model : This hysteretic model was based on the results of quasi-static laboratory tests
on flexural of concrete members at the University of Illinois 1970T17 The primary curve
includes not only the post-yielding stiffness but also uncracked and cracked stiffnesses. The
reloading stiffness continues as an extension of the unloading degraded stiffness up to the
cracking load point in the other direction. The reloading stiffness then proceeds to the yield
displacement in the other direction. The hysteretic rule is shown in Fig. 6.19. The unloading
stiffness can be expressed as,
(6.2)
where Dc and Dy are the cracking and yielding displacement corresponding to cracking and
yielding force Fc and Fy respectively, Dm is the maximum displacement experienced, fJ is
coefficient normally in the range of 0.4 - 0.5.
Compared to the bi-linear hysteresis, this model is more complicated but relatively
realistic in simulating the nonlinear behaviour of concrete members. Results of several
laboratory tests either in the form of sub-assemblagesT17 , MDOF reinforced concrete
models s15 ,s29 indicate that this hysteretic model performs very well in simulating the nonlinear
behaviour of concrete membersO ll ,S15.S29,T17, The disadvantage of this hysteretic model is the
complexity of the rules. A simplification of this model is sometimes called Otani modelS15 as
shown in Fig. 6.20. This simplified Takeda model also performed very well in simulating the
nonlinear behaviour of a 10-storey reinforced concrete model as reported by Ref.S15.
Degrading Bi-linear : Despite the fact that the Takeda hysteretic model is relatively complex,
the Degrading Bi-linear was introduced By IMBEAULT & NIELSEN (1974). This hysteresis
model is relatively simple as shown in Fig. 6.21. This model is a modification of the hi-linear
model where after maximum yielding displacement the average unloading and reloading
stiffness Ravu is represented by,
= K { D:1
' D In
}Il ' P t:> 0.50 - 0.60 (6.3)
137
ReU5 stated that the complete hysteretic energy of this model was around 60 % of the energy
dissipated by the bi-linear model. Therefore, its application on the MDOF inelastic analysis
caused significant increases in member ductility demands.
Q-Hyst Model: The Q-Hyst model was developed in 1979s1s . This hysteresis model is basically
a modification of the bi-linear model, but may also called a modification of the Otani model
and is shown in Fig. 6.22. The reloading stiffness is the main difference from the Otani model.
This reloading stiffness is defined by the slope of the line from the residual displacement to the
maximum displacement in other direction. The unloading degradation is defined as in the
Takeda hysteresis model. Compared with the Takeda model, this hysteresis has a thinner loop
because of a smaller reloading stiffness and dissipates less energy. This result was reported by
Ref.S14 in which the Q-Hyst model resulted in peak displacements greater than those of the
Takeda model. Meanwhile, Ref.]5 and Ref.S29 stated laboratory tests on a 10 storey concrete
building model using the Q-Hyst model gave better performance at small amplitude responses
when compared with the Takeda model. but the opposite was obtained during large amplitude
response. In general, the performance of the Q-Hyst model was very close to the Takeda
modelsl4
Flexural hysteretic models for inelastic response of concrete members have been
described. The performance of the concrete hysteresis models have been investigated by several
researchers by analysisC20 ,IS,M1S,S13 and laboratory test;AJ!,S15,W21. The laboratory tests indicate that
tensile cracking of the concrete and yielding longitudinal of the reinforcement will cause
pinching or stiffness degradation immediately after reloading P21 ,S13 where the pinching stiffness
is much less than the initial stiffness. The pinching behaviour is largely because of shear
effects Il ,P21,S13 unbalanced reinforcement or bar slip. The pinching effect will decrease as the
applied axial load increase. The Sina (l979)Sl5 and Takeda-Takayanagi (1976)5 hysteretic
models incorporate the pinching behaviour of concrete members.
6.5.3 Modelling Member Flexural Rigidity (EI)
6.5.3.1 Reinforced Concrete Frames
In reinforced concrete buildings, every member usually has a constant cross

section along its length. Even though the degree of cracking may differ along the member,
using a constant cross section means that modelling of the member flexural rigidity E1 is easier.
138
Fig. 6.19 Takeda Hysteresis Model Fig. 6.20 Modified Takeda Hysteresis Model
u;.
Fig. 6.21 Degrading Hi-linear Hyst. Model Fig. 6.22 Q-Hysteresis Model
Fig. 6.24 Relocation of the Plastic Hinges

ofRC Beams (PI8)
5
Ratio of blb w
Fig. 6.23 Coefficient for Moment Inertia Fig. 6.25 Rigid Joint Zone Length of the
of Flanged Section (P18) Reinforced Concrete Frame (T18)
139
In the modelling of the concrete member flexural rigidity, simplifying

assumptions are made. The simplest modelling takes the fraction of gross section of concrete
member flexural rigidity EJg to give an equivalent flexural rigidity EJe. With constant value
of the concrete modulus of elasticity E e , PAULAY & PRIESTLEY (1992) recommended
typical values for the effective member moment of inertia 4 given in the following table.
Table 6.1 Effective Member Moment of Inertia (lJP18
Member Type Range Recommended Value
1. Rectangular Beams 0.30 - 0.50 Ig 0.40Ig
2. T & L beams 0.25 - 0.45 Ig 0.35Ig
3. Columns, P > 0.5 fo' Ag 0.70 - 0.90 Ig 0.80Ig
4. ColUmns, P = 0.2 fe' Ag 0.50 - 0.70 Ig 0.60Ig
5. Columns, P = -0.05 fe' Ag 0.30 - 0.50 Ig 0.40Ig
The effective width of monolithic T -beams gives both an additional sti ffness and
strength to the beams. The effective relative stiffness of the T and L concrete member.
therefore, may be expressed asPl8,
(6.4)
where Kw is an effective relative stiffness for T and L beams, La is beam span, l"e is an
effective moment of inertia of the concrete member, f is an coefficient for the moment of
inertia of the flange sections as shown in Fig. 6.23.
6.5.3 Structural Walls
During moderate or relatively large earthquakes, the base response of the fixed
cantilever wall may no longer remain elastic, but will most likely exhibit an inelastic response.
During this response, the base fixed wall may crack and the wall effective moment of inertia
Iwewill reduce. This will affect the stiffness and the fundamental period of vibration T of the
stmcture. Hence, the effect of the wall on the inelastic response and fundamental period of the
stmcture should be taken into accountP18 X1 If the cantilever wall is subjected predominantly
140
to flexure deformation, the effective wall moment of inertia ~ may be calculated from the
following formula P'B ,
1 - { 100 PM} I (6.5)

we - Iy + It: Awg wg
where fy, fc are the steel yield strength and the concrete compressive strength respectively (in
Mpa), Pu is the wall axial load during the earthquake (positive during compression), Aw& is the
gross section area of the wall and Iwg is the gross wall moment of inertia.
6.6 MEMBER ELEMENT AND HYSTERETIC MODEL USED IN THIS STUDY
As described in Section 6.5.1 each of the member element model has its own
behaviour, advantages and disadvantages. In this study the one-component model was used.
OTANI (1980) discussed the calculated and observed responses of a three-storey one-bay frame
structure. Laboratory tests to examine the reliability of the one and two-component models in
simulating the observed response of inelastic model structures indicated that the one-component
model gave fair agreement between the computed and observed responses. Similar results were
found for a modified two-component model. BAN ON et al.(1981) reported that tests of 32
subassemblage RC specimens showed that the one-component model was sufficient accurate in
simulating the inelastic cyclic behaviour of RC members.
FIUPPOU & ISA (1988) discussed the reliability of the one-component and inelastic
finite-length element models to simulate the inelastic behaviour of concrete sub-assemblages
under quasi-static loads. Both joint-rotation and load-displacement responses of the one-
component model gave good agreement with the experimental results, while the computed beam
moment-rotation was slightly different from the observed results. This may be because of the
limitation of this model, in that it cannot follow the gradual change the spreading inelastic
deformation along the beam length, the shift of the point of inflections and it does not has the
ability to account for the coupling effect between the moments acting at the two member-ends.
The computed response of an inelastic finite-length element model gave very good agreement
with the observed response.
As discussed in Section 6.5.2 flexural member hysteretic models have been introduced.
Some claims for the best model have been made without an understanding in earthquake
141
response0 5 Ref. 05 and Ref.012 briefly discussed the important hysteretic parameters for an
inelastic SDOF building structure. These parameters include uncracked stiffness, cracking force
level, yield resistance, post-elastic stiffness and the unloading stiffness degrading parameter.
The hysteretic models used included the Clough, Takeda, Bi-linear, Degrading Trilinear and
Ramberg-Osgood models. The effects of these parameters, in general, is more sensitive on
stiffer structures than for more flexible structures. The yield strength was the most important
parameter followed by the unloading stiffness degrading parameter and the post elastic stiffness.
The cracking force and uncracked stiffness did not significantly affect the structural response.
In this study, the unloading stiffness degrading parameter and the reloading stiffness
were used to control the fatness of the hysteretic loop. The bi-linear model is one of the
hysteresis models used in this study and is the fattest model used. Although this model
sometimes does not simulate the inelastic behaviour of concrete members it was used for
comparison with the other models such the Clough hysteresis model. The original Takeda
hysteretic model is very complex but the modified Takeda (Otani model) is an alternative and
is used in this study. This model has a thinner loop than the Clough model because the
modified Takeda unloading stiffness is smaller than Clough model. The Q-Hyst model is also
used in this study. Comparing with the modified Takeda model, the Q-Hyst loop is thinner.
This is because in the Q-Hyst model the reloading stiffness is smaller than that of the modified
Takeda rule. Thus, for the four hysteretic models used, the degree of hysteretic fatness is
controlled by both unloading and reloading stiffnesses.
6.7 MODELLING OF RIGID ZONE AND PLASTIC IDNGE LENGTH
The joint at the intersections between beams and columns performs a very important
role in generating stiffness of the moment resisting frame structures. Such joints are usually
required to be rigid so that they can accommodate all of the forces acting on the beams and
columns without any joint deformation. There seems to be some evidence from past
emthquakes that some stmctural damage was caused by joint failure but was mainly due to soft-
storey effects with the column failures due to shear failure or inadequate column lateral
confinementP18
The usual ealthquake resistant design philosophy for ductile moment resisting frame
structures allows for the formation of beam plastic hinges adjacent to the beam-column joints.
142
The intemal forces transmitted from these beam or column members result injoint shear forces
in both vertical and horizontal directions leading to diagonal compression and tension stresses
within the joint core. However, Ref.X2 and Ref.X3 stated that the joint strength should not
normally govern during the development of the full strength of the adjoining members and
energy dissipation within the joint core is undesirable. It means that during moderate
earthquakes the joints should not show any strength degradation.
Results of laboratory tests, show that after a few cycles of beam plastic hinging, it is
not possible to prevent some inelastic deformation occurring in the beam-column jointsA17.P1B.
An effort to prevent further development of joint inelastic deformation was suggested by
Ref.AI6, AI7 and P30. This can be done by relocating the beam plastic hinges away from the
column faces as shown in Fig. 6.24 even though this action causes an increase in the beam
curvature ductilities. If the joints remain elastic the beam rigid-zone length can be measured
from the column centre-line to the column face as shown in Fig. 6.25. Ref.H19 and TI8 stated
that an analysis based upon full rigid offset length underestimates the structural deflection but
the displacements are overestimates if the analysis is based on centre-line to centre-line
geometly. Modification of the concrete rigid-zone length is necessary by considering a rigid-
zone reduction factor rz Ref.HI9 recommends a rigid-zone reduction factor rz = 0.5, while
Ref.AI7 used rz = 0.5, rz = 0.25 and rz = 0 for their analyses. Rigid zone reduction factors
r z = 0, r z = 0.5 and rz = 1 are used in this study.
The plastic hinge length Lp has a direct bearing on the curvature and displacement
ductility. Laboratory tests indicate that such factors such as the depth of the member, the
longitudinal bar diameter c\, and the length of the member from the maximum moment point
to inflection point L. affect the plastic hinge length. Based on data compiled from several
laboratory tests an approximate plastic hinge length can be expressed asC26.P31.S31,
p
Lp == 0.08 L + 6 db , for ~ 0.3 (6.6)
Ie Ag
L 6 J, [0.5 + 1.67~) ,
p
Lp = 0.08 for < 0.3 (6.7)
Ie Ag Ie Ag
whereP is the member axial load, fc is the compressive concrete strength and Ag is the gross
member cross section area.
143
For typical beam and column proportions, Ref.P18 and P3I recommended the
approximate average value of member plastic hinge as,
(6.8)
where d m is the depth of the concrete member.
In this study the plastic hinge lengths given by Eq. 6.8 are adopted.
The problems associated with the inelastic analyses of multi-storey reinforced concrete
building have been discussed. These problems cover not only the uncertain nature of the
building such as effects of infill walls, floor slab flexibility but also inelastic characteristics of
the concrete members. These characteristics include modelling of the inelastic member,
member energy dissipation, joint rigid zone length and plastic hinge length. Other problems are
the selection of earthquake ground excitations.
The modelling of members and the modelling of member energy dissipation have been
proposed by many researchers. The member elements include the two-componentC25 , multi-
component, one-component, discrete element and distributed element models. All of these
models have their own characteristic and appropriate applications. The modelling of member
flexural energy dissipation is represented by hysteresis loops including the elasto-perfectly
plastic, bi-linear and degrading bi-linear, Clough, modified Takeda, Q-Hyst. and Sina models.
Member flexural rigidity EI choices differ among researchers010.PIS. The analysis of symmetric
and simple 3-dimensional structures can be modelled as a series of 2-dimensional plane
structural analysesM25 ,M26.
The one-component model is used in this analysis where the bi-linear, Clough, modified
Takeda and Q-Hyst models are used to represent the behaviour of the inelastic springs at each
end of the inelastic member. The member flexural rigidity E1 model of Ref.P18 is adopted.
The joint rigid zone reduction factors r2. = 0, rz = 0.5, rz = 1.0 and plastic hinge length Lp
= 0.50 dm where dm is the member depth are used in this study.
Chapter VII
Soil-Footing Foundation Response and Impedance
7.1 INTRODUCTION
Over the last few decades, there has been an increased interest in geotechnical
engineering on the dynamic response of foundations. The actual soil beneath the foundation
may be physically complex and no exact mathematical solution seems possible for the general
problem of soil-foundation interactionGI. The different approaches used by researchers, include
analytical methods using the theOlY of an elastic half-space, an approximate or lumped
parameter methods and the finite element method.
The application of elastic half-space theory to soil-foundation response analysis is

limited to simple soil-foundation systems. This is because the mathematical analysis based on
the theory of elasticity is sometimes not easy to folloW>6. The Lumped parameter method in
which the response of the soil-foundation system is analyzed using a spring-dashpot mechanical
system has become very popular because the behaviour of mechanical system is easy to
understand AI .
The soil-foundation response in the lumped parameter model can be classified into two
different groups. The first group emphasizes the frequency independent response where the
spring and damping coefficients are indepenpent of the frequency of excitation. The second
group involves the effect of the frequency of excitation in the foundation response. Recently.
soil-foundation responses also cover the effects of other important parameters, such as the
degree of embedment and the shape of the foundations. The determination of the both soil-
foundation stiffness and the damping coefficient involving all of these parameters is important.
7.2 RESPONSES OF SOIL-FOUNDATION SYSTEMS
7.2.1 Static Response of Soil-Foundation System
The analysis of a footing-foundation originally follows the development of the analysis

and design of machine foundations. In the past, machine foundations were frequently designed
144
145
without any appropriate analysis. The design rule was very simple, that is the total weight of
the massive concrete foundation normally should be taken at least 2 to 5 times the weight of
the supported machines X1 GS Although this design rule seems rational, the important design
parameters such as the type of excitation, generated frequency and the dynamic behaviour of
the supporting soil are not considered. It was later found that increasing the mass of the
foundation would reduce the effective damping coefficient of the supporting soil, whereas
increasing the soil-foundation contact area would increase the effective damping
coefficientC22.F4.G1.G13 .
Initially, a Winkler model of the soil-foundation was used. This foundation model
consists of a rigid beam resting on an elastic soil base. To be able to develop the relationship
between the applied load and the corresponding foundation deformation a simple mechanical
model can be defined. This consists of a bed of independent elastic springs supporting the
corresponding rigid beam. The stiffness of the springs represents the load-deformation
compatibility of the soil-foundation model. This model is only reasonable for low-frequency
or nearly static loading as the damping of the soil has not been included. If the vertical
deflection of the ground z is assumed to depend only on the normal pressure p acting on the
beam, the load-deformation relationship can be expressed as,
(7.1)
where z is ground vertical displacement and k.. is coefficient of subgrade reaction or sometimes
called the coefficient of elastic uniform compression.
7.2.2 Dynamic Response of Soil-Foundation System
7.2.2.1 Winkler-Voigt Soil-Foundation Model
The Winkler-Voigt foundation model is the development of the Winkler

foundation. The difference is in the application of a new soil mechanical model consisting of
bed of independent viscous dampers in parallel with the independent elastic springs to represent
the dynamic subgrade reaction of the soil-foundation system. The presence of the viscous
dampers in this model means that the soil-foundation system is no longer a static loading
system but the effects of the dynamic loading have been incorporated. The dynamic subgrade
reaction dsr of this soil-foundation model can be expressed as D15 ,
146
(7.2)
where ksr is the coefficient of subgrade reaction and Cor is the damping coefficient.
The disadvantage of this model is the difficulty in determining these

coefficients. Some IiteratureS24 ,VB discusses the empirical formulae for the modulus of subgrade
reaction ITl,;r ( m.. = ksr x b, where b is the width of beam), but the appropriate value of
damping coefficient is not clear. Extension of this model may lead to the elastic beam which
is supported by either independent Kelvin, Maxwell or Standard Viscoelastic mechanical
models.
7.2.2.2 Elastic Half-Space Soil-Foundation Model
Most geotechnical engineers agree that the work by REISSNER (1936) has
stimulated the development of the theory of foundation vibrations. This work was the first
mathematical formulation of the problem of a rigid body on an elastic half space where the
soils are modelled mathematically as a semi-infinite, isotropic, homogeneous and elastic
medium. A semi-infinite medium means that the mass of the soils under consideration is
bounded by the ground surface, isotropic indicates that elastic properties of soils are
independent of direction, homogeneous means that no different physical characteristics exist
within the soils and elastic means that the load-deformation of the soils follows Hooke's rule:
A surface foundation resting on an elastic half space is illustrated in Fig. 7.1.
The solution of foundation response has been developed by many researchersN1

If a steady state vertical harmonic loading P = Po e1wt acts on the centre of the foundation with
radius ro then the corresponding vertical displacement z(t) can be written as,
z(t) (7.3)
where Po is the amplitude of the applied loading, G is the soil shear modulus, w is the applied
angular frequency and f] and fz are displacement functions.
147
Force equilibrium of forces from Eq. 7.3 leads to
P",
=--D (7.4)
G I:
where
(7,5)
m(} Wo Wo
b(} - - -3 = - - -3 (7.6)
P re g(X) ': y r(J
g
(7.7)
where Zz is the amplitude of the foundation motion, Dz is a dimensionless magnification

amplitude, a" is dimensionless frequency, bo is the ratio between footing mass m., and mass of
the elastic medium p ro\ g is the acceleration of gravity, p is the mass density ofthe soils, Wo
is the weight of the footing foundation and 'Y is the weight per unit volume of the soils.
The determination of the displacement functions f1 and fz is mathematically

complexlli G4 and examples of the procedure can be found in Ref.G2 and LA. Plots of the
displacement functions f1 & f2 against the dimensionless frequency a., for a surface strip and
a circular foundation are shown in Figs. 7.2 and 7.3 respectively.
An improved representation of the behaviour of a surface footing resting on an

elastic half space was made by HSIEH (1953) and reported by RICHART et a1.(1970). Taking
the derivative with respect to time of Eq. 7.3 and after mathematical re-arrangement leads to
the relationship,
dz (7.8)
p = C - + K":t
~ dt v
148
Block
tmi Mass=m
III 17
G,v,p
+z
m =Q-P
d
(0) (b)
Fig. 7.1 Surface Footing on an Elastic Half-Space (R5)
0.3,.----,-----....,...-----.
----lUCO' W(S1AAHh. 1968
I
0.< \
F,
0.1
~.l
.0.\ ' - - _ - ' -_ _,.L...._--''--_-'-_--'

O. 0.1 1.0 I.S 1.0 l.~
Fig. 7.2 Compliance Function of Strip Fig. 7.3 Vertical Displacement Function of
Footing on Half-Space (G4) Circular Footing on Half-Space (R5)
(al
---I
v
:
I,
lViVltl!- lei
~_Ht
..,"""
n~~si
c.
Fig. 7.4 Lumped Parameter Model of Soil- Fig. 7.5 Spectral Comparison for Footing on
Footing Interaction (03) Half-Space and Lumped Model (L5)
149
where,
(7.9)
Kv = G (7.10)
If the total weight of the footing and machine is denoted by Worn, and the
foundation is subjected to steady state loading Q = Q, ei~. then the dynamic equilibrium of the
system shown in Fig. 7.1.b yields,
(7. 11.a)
or
(7.n.b)
Q=P+m-
d"z
it"
Substituting ofEg. 7.8 in Eg. 7.11.b yields,
(7.12)
Eq. 7.12 is similar to the dynamic equilibrium equation of a damped single

degree of freedom system. Terms Cv and 1(.. in Eg. 7.9 and Eq. 7.10 are frequency dependent
and refer to the damping coefficient and stiffness of the equivalent SDOF system. Eq. 7.12 also
shows that an improved representation of the theory of an elastic half space has been achieved
by the ability of this equation to express the geometrical damping Cv of the soil-foundation
system.
7.2.2.3 Lumped Parameter Soil-Foundation Model
The proceeding section showed that the dynamic vertical response of surface
footing resting on an elastic half space could be computed from Egs. 7.4 or of Eq. 7.12 and
150
the response is frequency dependent. Response of the foundation subjected to horizontal,

rocking or torsional steady state harmonic loading can be found elsewhereD6,R.'i.
The actual soil is seldom an elastic half space. Other parameters include size
and shape of the foundation, the variation of soil stiffnesses with depth, stress level, water
content and foundation embedment. Researchers concluded that to accommodate these
parameters it was more convenient to substitute the behaviour of simple spring-dashpot systems
than to modify the theory of an elastic half space. This substitution is possible since the
numerical quantities such as masses, damping coefficients and stiffness constants for all
vibration modes have been developed. Since the quantities involved are defined as constant
values, the entire soil-foundation system is the well known Lumped Parameter ModelR.'i. This
model is shown in Fig. 7.4.
LYSMER & RICHART (1966) developed a simplified equation for motion of

a lumped parameter model. If a rigid circular surface foundation is loaded by vertical steady
state force Q = Qo eiwt , the equation of motion can be expressed as,
(7.13)
where,
3.4 ,.{} rrv::: 4 G,.o

(7.14)
c = - - - vGp , k=--
1 - " 1 - "
where c is the damping coefficient and k is the stiffness constant.
Eq. 7.13 is similar to the differential equation of motion of a SnOF system,

and the response of the foundation can be obtained by solving this equation. The differential
equation of motion of the foundation subjected to other vibration modes can be written by
replacing the corresponding damping coefficient c, stiffness constant k and loading mode.
Attempts to prove the validity of the lumped parameter foundation model have been
carried out either by analytical methodsLS as illustrated in Fig. 7.5 or by laboratory testsR.'i. It
shown that the response of the lumped parameter foundation model provides close agreement
with the response computed from the theory of an elastic half space.
151
7.3 STIFFNESS AND DAMPING OF SURFACE FOUNDATIONS
7.3.1 Static Stiffness and Damping Coefficients
After substitution of the theory of an elastic half space by the lumped parameter
method, the application of the later method becomes very popular. The accuracy of this
foundation model depends on how appropriate are the selection of stiffness and damping
coefficients. Some important guidance on the choice these values is presented in Ref.W2.
Initially, the stiffness and damping coefficients in the lumped parameter model were
assumed to be constant and independent of excitation frequency. The stiffness constant was
based on static load and the damping coefficient was derived for a particular loading frequency.
It was later found that both of these quantities were frequency dependent. In a frequency
independent response the solution ofEq. 7.13 is carried out only at a particular frequency of
interest. The stiffness and damping coefficients of a rigid circular surface foundation for
vertical, horizontal and rocking vibration mode are written as,
4 G T. 3.4 "0 (7.15)

C =--{GP
1-v " 1-v
(7.16)
rJ
K,.=8G--"- (7.17)
3(1-\1)
where
(7.18)
where K." Kh , K,., C", C h and Cr are spring stiffness and damping coefficient for vertical,
horizontal and rocking modes respectively, P is soil Poisson's Ratio, fo is the radius of the
footing, G is the soil shear modulus, p is the soil density and I.p is the mass moment of inertia
of footing with respect to a horizontal axis.
152
If the case of a rectangular footing when the ratio of footing length 2L and footing
width 2B is less than 3G1 ,G2, the equivalent radius for vertical re v horizontal rc b and rocking re.r
respectively can be computed as,
(7.19)
7.3.2 Dynamic Stiffness and Damping Coefficients
The dynamic differential equations of motion of the surface foundation can be

developed either by using the theory of the elastic half space, lumped parameter model or shear
cone theoryMS.v3. The solution ofthese equations describes the frequency dependent relationship
between applied force Pj(t) and the corresponding displacement amplitude u;(t) to be described
in compact form V2,v3,
(7.20)
where Q; is the complex-impedance (stiffness and damping coefficient) value at vibration mode
i to be expressed in term of the static stiffness and complex impedance functions as,
(7.21)
where K .i is the static stiffness at vibration mode i and the term in the bracket in Eq. 7.21 is
usually called the complex-impedance function.
The real part of the complex-impedance function, k; will be called the dynamic stiffness
coefficient and Ci> which is proportional to the imaginary part, will be assigned to the dynamic
damping coefficient. Both k; and Cj are frequency dependent. The product of the static stiffness
and complex-impedance function leads to the dynamic stiffness Kdi and damping coefficient C d1
respectively thus,
(7.22)
CAl'
r" c,
= K . (-) (7.23)
.., $,1 V '
s
153
where kj and ci in Eq. 7.22 and Eq. 7.23 not only depend on the excitation frequency but also
depend on the soil Poisson ratio" as written in Eq. 7.25. The dynamic damping coefficient Cd,i
in Eq. 7.23 is also frequency dependent and depends upon the shear wave velocity V .
VELETSOS & WEI (1971) presented the frequency dependent response of a rigid
circular surface foundation resting on an elastic half space. Coupled horizontal. and rocking
modes were considered. If the applied horizontal force and moment are denoted by Hand M
respectively, the relationship between the force and displacement is,
(7.24)
where K.,h and K. are the stiffness constants for horizontal and rocking vibration modes
respectively, k's and c's are dynamic coefficients for stiffness and damping, u and cp are
horizontal and rocking displacements respectively, ao is the dimensionless frequency and ro is
the radius of the footing. The dimensionless dynamic coefficient k's and c's can be determined
by using Fig. 7.6, 7.7 and Fig. 7.8 respectively.
VELETSOS AND VERBIC (1973) also presented the dynamic response of a rigid
surface footing resting on an elastic half space in the frequency domain. Effects of the
foundation mass on the foundation frequency find damping coefficient were discussed. UPo and
Uo are the steady state harmonic force and displacement amplitude respectively, then the
relationship between the force and the displacement can be written as,
(7.25)
where K..! is the stiffness constant in vibration mode i, and kh Cj respectively are the dynamic
stiffness and damping coefficients of vibration mode i.
The value ofk; and ci may be determined by using Fig. 7.6, 7.7 and Fig.7.8. Based
on the results of the shear cone analysis, the coefficients ki and Cj for the vertical, horizontal
and rocking modes of a rigid circular surface foundation were approximated by,
154
(7.26.a)
(7.26.11)
(7.26.c)
(7.27.3)
(7.27.b)
(7.27.c)
Values of ai' b's and d's for any value of soil Poisson's ratio II were tabulated in Ref. V2 and
V3.
WOLF & SOMAINI (1986) presented the dynamic impedance of a rigid circular
surface foundation resting on an elastic half space. An improvement of the discrete model was
used as shown in Fig. 7.9. This discrete model was introduced in Ref.M5 and is represented
by a parallel spring of stiffness K and a dashpot with coefficient Co. In addition, a foundation
mass Mo is attached in series of the damper C 1 and mass MI' The system is loaded by P =
pew) and the corresponding displacement of the mass Mo and M t are uo(w) and ut(w)
respectively. Taking the dynamic equilibrium both of elements and after dividing by z ei~ leads
to,
(7.28)
155
(7.29)
Eliminating the value of ul(w) in 7.28 and substituting in Eq. 7.29 leads to,
(a)lMl (a)lMl
K (a)2M" Ml K C<f}
-------+1(,) +-
ll.+(~Ml)l K Cl1+(~M/' K (7,30)
~ ~
The dimensionless coefficients of the dampers 'Yo. 1'1 and of the masses ILo , ILl are then
introduced as,
1"0 (7.31)
Co = (-)K'V
V I., ,
r :& I" :1 (7,32)
(~) M = (~)
M (} = V K IlIII 1 V, K ..'-1
II
Eq. 7.30 can be transformed from uo(w) to uo(aJ by using the relationships in Eq. 7.31
and Eq. 7.32 yielding,
(7.33)
where
k(a) =1 - (7.34)
c(a) = III (7.35)

11
The values of 'Yo> 'Y1 and ILo. ILl have been tabulated in Ref.W9 or Ref.WlO. Sample
plots of k's and c's versus dimensionless frequency ~ for the vertical and rocking modes are
156
1.0 <0.45
Q .:; -- k ..
g ""
vo
---~- e"
"
q
.i2 .it
."
'0 0", ."
'0
V3
"
1i
>
.i!
~
7.6 Dyn. Stiffness and Damping Coeff. 7.7 Dyn. Stiffness and Damping
for Horizontal Vibration Mode (V4) Coeff. for Rocking Mode (V4)
lp
O.Mo~o
1 0.1 ~
.J ~ \
~, ' "'
o~~~==~==~~~==~~~~~~~
K
~
-01
Fig. 7.8 Dyn. Stiffness and Damping Coeff. Fig. 7.9 Two DOF Discrete Model for
for Coupling Modes (V4) Vertical Vibration (W9)
LO,.,...----------------,
> a) ~
..>I. ..>I.
t- C.S t- a.s
Z z
W w
~ 0.6 .. .. + .. .. .. ~ 0.6
lL.. lL..
lL.. lL..
W W
o o
u o.~ u 0
I.!) I.!)
z
~
z
g: 0.2 g: 0.2
(f) (f)
0.01--..,---,...--,;-.-,---..---..--..,..----,---1
o t 2 ~3 4 5 6 7 t 2 3 '9 6 g
DI HENS ION"e.ESS FREQUENCY ao=oal cs DI HENS IONLESS FREQUENt.~.. a o=(')91 cs
1.50..,..---------------- 0.5
> b) ~
b)
()
() i.25
I- t- 0.4
Z :z
w _ _ ; - . - ; - - . - . -.-"T - ; - : - : w
. ..
~1.00
U u
t' . ~ -".~ ~ ,,-. a~'."~ .. ~" 0.3
lL.. lL..
t:J ().7S ...
lL..
W
., =a
o ._,_4 "\) 0 o
u =. U 0.2
0.50 I.!)
1.l = if3
1/3
-!'l:
I.!)
1.l Z
Z ~ ... ~ . .. '" 1.l if2
(L ~ ... ~ ... ' '\) = 112 0.1
::!: 0.25
a: a:
o o
o.oo-!--,.--,--------;,---.,,.--.....-,-.----,-........,----j o.o-jL.-,--..---r--r----,r----,-~--I
o i 23456 a I 234 S 6 7
DIHENSIONLESS FREQUENCY ao=t>a/cs DIMENSIONLESS FREQUENCY ao=G>a/cs
7.10 Dyn. Stiffness and Damping Coeff. 7.11 Dyn. Stiffness and Damping Coeff.
for Disk in Vert. Mode (W9) for Disk in Rocking Mode (W9)
157
shown in Fig. 7.10 and Fig. 7.11 respectively. For a small value of ao the rocking damping
coefficient will be much smaller than the veltical damping coefficient.
DOBRY & GAZETAS (1986) presented the dynamic impedance of a rigid arbitrarily
shaped surface foundation resting on an elastic half space. The foundation is loaded by a
harmonic loading in any vertical. horizontal or rocking mode. The discrete mechanical model
is shown in Fig. 7.4. If Pi and Uj are the force and the displacement amplitude at vibration
mode i, then the dynamic equilibrium can be expressed in a form similar to Eg. 7.22 and Eq.
7.23. The dynamic stiffness and damping coefficients for vertical, horizontal and rocking
modes were systematically tabulated in Ref.D3.
7.4 STIFFNESS AND DAMPING COEFFICIENTS OF EMBEDDED FOUNDATIONS
Under dynamic loading, interface friction between the soil and the adjacent side-wall
footing is unavoidable and the responses of surface and embedded foundations under dynamic
loading may differ significantly. During dynamic loading a shear stress will occur at the soil
foundation interface. Dissipation of energy will be observed unless slip occurs between the soil
and the footing.
The estimation of the energy dissipated on the embedded foundation is important in the
foundation vibration analysis. From laboratory test results, the damping forces generated
depend upon the vibration modes, embedment depth, the area of the contact. physical
characteristics of the soil, and the contact between the foundation and surrounding soilsA'i. In
general, foundation embedment will increase the soil-foundation stiffness substantially if the
side-wall and the surrounding soils are relatively stiffly bonded or if no slip
occurs AS ,Gl,H7)2,l(2,Nl,N2,R2.S7.
7.4.1 Frequency Independent Stiffness and Damping Coefficients
Effects of the embedment on the dynamic foundation response observed in laboratory

tests was reported by NOVAK (1970). Two important parameters such as the degree of the
embedment and types of the back fill were considered. Results of the vertical vibration tests
are shown in Fig. 7.12 where these parameters gave significant effects on the foundation
response. The vertical response amplitude ofthe foundation reduces as the embedment increase.
158
A,B and C in the figure refer to a fully undisturbed embedded foundation, a foundation
surrounded by fill material and a foundation surrounded by an air gap respectively. The vertical
response of the foundation also reduces as the back fill density increases.
ANANDAKRISHNANAND KRISHNASWAMY (1972) also present the effects of the

embedment on the vertical steady state responses of an embedded foundation from both analysis
and laboratory tests. The lumped parameter method was used to predict the amplitude of the
frequency response, where the spring and damping coefficients were assumed to be constant
and independent of the excitation frequency. The relationship between the normalised frequency
and the embedment factor is shown in Fig. 7.13 where the normalised frequency increases with
the embedment factor. This study concluded that the correlation between test results and
predicted values of the foundation responses was quite satisfactory.
Effects of the embedment on the response of two existing different machine foundations
was also studied by STOKOE AND RICHART (1974). Harmonic loading either in the
horizontal or rocking vibration modes were applied to the foundation systems. The foundations
are rectangular and an equivalent radius ro was used to predict the foundation response. The
measured steady state foundation response was 25 % less than that predicted for an equivalent
circular foundation resting on the surface. It was also found that the measured damping ratio
was at least 30 % more than that predicted from an equivalent circular surface foundation.
JOHNSON AND CHRISTIANO (1974) analyzed the load displacement behaviour of

embedded cylindrical and strip foundations under static loading. The foundation was embedded
in an elastic stratum underlain by a rigid base. The vertical, horizontal, rocking, torsional and
coupled horizontal and rocking modes were considered. Stiffness constants of the embedded
foundation were evaluated by using the finite element method and were normalized with the
stiffness of the surface foundation. These normalized stiffnesses are shown in Fig. 7.14, 7.15
and Fig. 7.16 respectively which show that the stiffness constant of an embedded foundation
increases significantly with embedment ratio.
7.4.2 Frequency Dependent Stiffness and Damping Coefficients
An approximate dynamic response of an embedded massive foundation in the frequency

domain was reported by NOVAK AND BEREDUGO (1972). A cylindrical embedded footing
159
J "
"
"J
E:lt)Et30.,.,.,;o.j!'AL tt14fb.
G NaVAl(
.. IJ.
e
fP'tt
AV~HO~$
Fig. 12 Comparison of Vertical Response of Fig. 7.13 Frequency of Excitation vs

Embedded Footing from Tests (N1) Embedment Factor (AS)
lal \le-liCOI Mode Ib) Sliding Mode
Fig. 7.14 Vertical Stiffness Ratios for Fig. 7.15 Horizontal Stiffness Ratios
Circular Footing (J2) for Circular Footing (12)
PCt)
7.16 Rocking Stiffness Ratios for Fig. 7.17 Embedded Footing Excited
Circular Footing (12) Vertically (N2)
160
was assumed to be resting on an elastic half space. Vertical harmonic loading was applied at
the centre of the foundation plan as shown in Fig. 7.17. pet) and z(t) are the applied force and
the corresponding vertical displacement respectively. The base reaction ~ was assumed to be
equal to that of an elastic half space which means that this reaction was independent of the
depth of the embedment. Based on the theory of an elastic half space, the base reaction ~ can
be expressed as,
(7.36)
in which,
(7.37)
where f\ and f2 are frequency dependent displacement functions and G and ro are the shear
modulus and the radius of footing respectively.
If r. is the dynamic reaction per unit length of the embedment, then the total side
reaction can be expressed as,
Nit) = Jr,(z,t) dz (7.38)

o
The dynamic reaction r, is a function of the depth of embedment, soil shear modulus
G, dimensionless frequency 30 and mass density p. An approximate assumption that r. is
independent of the depth of embedment was then adopted. Accordingly, the dynamic reaction
r. = G. (51 + i 52) u(t) can be utilized and the total side reaction is,
1
(7.39)
Nz(t) == f G&(S1 'I- i SJz(t)dz = GI-(SI + i SJZ(f)
~
where 51 and 52 are functions of first and second kinds of Bessel functions 1. (a.) and Y..(a o)
respectively, L is the depth of embedment and Gs is the shear modulus of the embedded soil.
The values in the bracket in Eq. 7.39 are impedance function similar to those in Eq. 7.21. 51
refers to the soil-foundation stiffness and the damping coeftlcient is proportional to 52'
161
The final differential equation of motion of the foundation system is the summation of
Eq. 7.38 and Eq. 1.39 thus,
+ i SJ} = P
(7.40)
Eq. 7.40 is similar to Eq. 7.13, where the differential equation of motion of the soil
foundation system is represented in the form of the lumped parameter model. Hence, the
frequency dependent stiffness and damping coefficient respectively are,
(7.41)
K=
(7.42)
The values of C'. and S', as functions of the dimensionless frequency a" can be
determined by using Fig. 7.18. Experiments indicated that the approximate theory yielded
reasonable results.
The dynamic response of embedded massive foundation subjected to coupled horizontal

and rocking vibration modes was also presented by BEREDUGO AND NOVAK (1972). An
approximate theory based on the assumption that the dynamic reaction of the footing base was
equal to that of an elastic half space and the reactions acting on the footing sides were
independent of the depth of embedment. These assumptions were similar to the earlier
investigation of vertical response of an embedded footing foundation. The mathematical model
is shown in Fig. 7.19. Q(t) and M(t) are the applied horizontal and moment, u(t) and tJ;(t) are
the corresponding horizontal and rocking displacements respectively. The relationship between
the applied force and displacement in the lumped parameter method can be expressed in the
complex coupled equations,
(7.43)
where k" and c gs respectively are,

162
(4.44a)
(7.44. b)
(7.44.c)
(7.45.a)
(7.45.b)
e" = - [C + d~(z< - 0.5 I ) s]
(7.45. c)
where k... and c, respectively are the dynamic stiffness and damping coefficients, G is the soil
shear modulus, Ie is the rotational mass moment of inertia about the centre of mass, I is the
depth of embedment, 0 = llro is the depth of embedment to radius of footing ratio and Ze is the
distance from the centre of the foundation mass to the base of the footing,
The values of C .. and S' are dependent on the dimensionless frequency a.,. For several
values of soil Poisson's ratio JI, these parameters can be computed by approximate polynomials
as shown in Appendix I, Ref. BI. The corresponding frequency dependent stiffness Kss and
damping coefficient C. OS respectively are,
(7.46.a)
(7,46.b)
163
(7.46.c)
(7.47.3)
(7.47,b)
3 (7.47. c)
Grl
C = - (;
" (,) "
An approximate response of a fully embedded foundation was also presented by WOLF

AND SOMAINI (1986). Both cylindrical and prismatic embedded foundations were presented
for coupled horizontal and rocking vibration modes. Only the response of prismatic foundations
will be discussed. The foundation geometry and discrete mathematical model for the coupled
horizontal and rocking modes is illustrated in Fig. 7.20. P(ao) and M(ao) are the applied
horizontal force and moment respectively and the corresponding displacements respectively are
u(ao) and cPo(ao). The relationship between the applied force and the corresponding displacement
is similar to that in Section 7.3.2. This relationship is,
(7.48)
where,
(7.49)
(7.50)
164
Ik = 0.250 t2 for cylindrical loundatio" (7.51)

I,. = 0,333 /Z for prismatic loundation
(7.52)
where Ks.h,K. r are the static stiffnesses for horizontal displacement and rocking respectively,
kh' 1<:., ch, Cr are dimensionless stiffnesses and damping coefficients for horizontal and rocking
modes respectively, c rh = C hr are dimensionless damping coefficients for coupled horizontal
displacement and rocking, fk' fc are horizontal spring and damping eccentricities measured
from the level of the foundation base, 'Yoh is the damping dimensionless coefficient for
horizontal vibration and e is the depth of embedment.
The values of k's and c's can be computed by using Eq. 7.34 and Eq. 7.35 in which
the values of 'Yo, 'Yl and p.o. p., can be found in Ref.W9 or Ref.WlO. The approximate formula
for static stiffness K. ss and damping coefficient C.ss for vertical, horizontal and rocking modes
can be computed respectively from,
(7.53)
Ks,v = ~_: [ 3.1 (-f)O.7S + 1.6 ] [ 1 + { 0.25 + 0.2: b } (i)0.8 ]
K L
G-
= b{ I 0.6S + 0.8 (-)
6.8 (-) e 0.8]
I + 1.6 } 1 + { 0.33+ 1.34 } (-) (7.54)
s",y 2 -\I b b i b
1 +-
b
KS,hJl = - -
G b {
2-v
I 0.65
6.8 (-)
b
+ 2.4
} 1.34}
I
e 0.11
(1;)
] (7.55)
1+-
b
G-
Ks,n: = - b {
I-v
3
I + 0.80 }
3.20 (-)
b
+ (1.6) (!!../'] (7.56)
0.35 + (1) b
b
(7.57)
165
where 2b = foundation width, and 21 = foundation length, where 1 > b. If coupled horizontal
displacement and rocking is considered, then the coupling stiffness constants respectively are,
(7.58)
(7.59)
(7.60)
7.5 STIFFNESS AND DAMPING COEFFICIENTS OF ARBITRARY SHAPED AND

PARTIALLY EMBEDDED FOUNDATIONS
In the earlier discussions, the dynamic responses of the cylindrical and prismatic
foundations were covered for fully embedded foundations. The footing foundations, however
are not necessarily fully embedded but both arbitrarily shaped and partially embedded. In
Section 7.4.1 it was shown that the foundation embedment effects the foundation response.
Therefore, the responses of arbitrary shaped and partially embedded foundation are to be
discussed briefly.
The dynamic responses of arbitrary shaped and partially embedded foundations for all
vibration modes were reported by GAZETAS et al. (1985), GAZETAS & TASSOULAS (1987),
HATZIKONSTANTINOU et al.(1989) and FOTOPOUW et a1. (1989). Only the vertical,
horizontal and rocking vibration modes will be discussed.
7 .S.1 Vertical!. Stiffness and Damping CoefficientsD4 ,Gl,Gl1
7.5.1.1 Vertical Static Stillness
The method used in this study is an extension of the dynamic response of the
surface foundation resting on the elastic half-space by DOBRY & GAZETAS (1986) as
166
12
-- -- ~.:..S. IV,;., /J.O
--:t1
Me (1)
II ----_ .25
'1--
-- -
~
i - .......
10 0(1)
~- -, Ij'
9
8 C,
__C~
C, :2
0<>-
0,5
t"'- t--
I {:~
(J~~ &-~r-::!
-
r;:<_
...
Ze I { --,
I
"!7
u
6
.... " .... .3 ......
...
r-- k (J~17
Ki
~
......
I;:: JYt-
lL
o
UJ
:>
..J
~
5
4
c,
!
00
S21A ~2 . /
" Q2
K -
K
."., K
t;::> ~"--
5,
", -.;;;;:
'1'-.
r-- t-...... 1"-
H
CG
I {
3 ,
2
i-P
.h-
/" I
1/0y ~v
~ _s?- ::1::' 5.1r1--
_ s$~~i:-:.:~
o~
o 0.5 1.0 1.5
DIMENSIONLESS FREQUENCY a.
Fig. 7.18 Stiffness and Damping Para- Fig. 7.19 Embedded Footing Excited
meters (N2) Horizontally (Bl)
I---lB----1
T_
2L
Good eontoct betwuo
., .., .... ,.". ': '.
Fig. 7.20 Horizontal and Rocking Coupling Fig. 7.21 Arbitrary Shape Embedded
Discrete Model (Jl9) Footing on Half-Space (G 12)
,.!I
1[<i.""'\3f.~1 Ci.(IV
.~Pn2''''QII~''1
~~~O~.'--~02--7~''-~Q~4~O~'--O*'--~~'~~O~~~O~~~
A./{4L I,
Fig. 7.22 Vertical Static Stiffness Para- Fig. 7.23 Trench and Side-WaH Effects
meter vs Shape Factor (D3) for Embedded Footing (G 1)
167
described in Section 7.3.2. Included in this study are several results compiled from the
references. The results are applicable to a variety of cross sectional shapes, a wide range of
embedment depths and types of contact between the vertical sidewall and the surrounding soils.
The soil-foundation system is shown in Fig. 7.21.
The vertical static stiffness of an arbitrary shaped surface foundation resting on

elastic half space can be written as,
(7.61)
where K.v,lJIIf is the static stiffness of an arbitral)' shaped surface foundation, Sz is the vertical
static stiffness parameter, G is the soil shear modulus, L is one-half of foundation length and
71 is the soil Poisson's ratio.
The values of Sz are obtained for a wide variety of a rigid foundation shapes
from several references. The variation between Sz and a normalized base shape is shown in
Fig. 7.22. The best fit of these values yields,
(7.62.a)
Sz = 08

A A (7.62.b)
S = 0.73 + 1.54(_11_)0.75' for _b_ > 0'()2
I 4, 2 4 1
In reality, a footing foundation is placed at a particular depth below the free

ground surface. Generally the soil stiffness increases with the depth. The vertical displacement
of the trench footing foundation is therefore considerably smaller than that a surface foundation.
This is not only because of the increase in the soil stiffness but also the effect of the normal
and shear stresses generated at the horizontal plane of the foundation base, as shown in Fig.
4.23, reducing the vertical displacement of the foundation base. The effect of the trench
foundation on the vertical stiffness leads to the trench coefficient which is denoted by ~re'
Results of this study and compiled results from references leads to the relationship between ILre
and the shape-dependent depth parameter as plotted in Fig. 7.25. This trench coefficient can
be written as,
168
(7.63)
where D is the depth of embedment, B is one-half of foundation width where L > B and At,
is the area of the foundation base.
Another aspect of the embedded foundation response is the sidewall effect.

When the vertical sidewall is in contact with the surrounding soil, part of the applied load is
transmitted to the ground through shear stresses acting along the vertical sidewall. The vertical
foundation displacement will be sufficiently reduced because of the increase in the vertical
stiffness. This effect leads to the sidewall coefficient which is denoted by lwall' A plot of Iwall
against sidewall-base area ratio is illustrated in Fig. 7.24. The best fit for this coefficient leads
to,
IWall :: 1 + 0.19 { A:A }2fJ (7.64)
where A. is the area of the sidewall-soil contact.
The static vertical stiffness of the partially embedded foundation K.v,emb
therefore, can be computed by considering all of the coefficients and yields,
(7.65)
7.5.1.2 Vertical Dynamic Stiffness
Taking into account the effect of excitation frequency on the fully embedded
foundation leads to the dynamic stiffness coefficient denoted by k...emb' The vertical dynamic
stiffness of a fully embedded foundation can be computed by multiplying this coefficient by the
vertical static stiffness in Eq. 7.65 to get,
(7.66)
For Poisson's ratio /I < 0.4 plots of the dynamic stiffness coefficient of surface
foundation, kv .ur against the dimensionless frequency ao for any values of LIB can be
determined by using the graph shown in Fig. 7.26. For wide range of dimensionless frequency,
169
kv"ur is very sensitive to high values of the LIB ratio. Plots of kv,emb/kv,fiUr against the
dimensionless frequency for any value of D/B are shown in Fig. 7.27. The best fit of this
relationship yields,
(7.67)
For the foundation placed in a trench (without sidewall) the dynamic stiffness
coefficient, kv,lre can be expressed as,
(7.68)
In the case of an partially embedded foundation, the corresponding dynamic

stiffness coefficient can be computed by interpolating between Eq. 7.67 and Eq. 7.68.
7.5.1.3 Vertical Damping Coefficient
During vibration of the foundation, energy is transmitted away to the soil by

the spreading upwards-downwards waves and shear waves as shown in Fig. 7.28. The upward-
downward propagating waves is close to the "Lysmer's analog" velocity, VLa rather than P-
waves and the sidewalls mainly transmit shear wave (S-wave) to the surrounding soil. The
energy dissipation of this mechanism is called radiation damping denoted by C(w). This
damping coefficient is frequency dependent and increases with the contact area of the soil-
foundation. By assuming that the two waves generated at the base and sides of the embedded
foundation are independent of one another the total radiation damping 4 can be computed by,
(7.69)
where,
3.4 VB
v:fA == (7.70)
n; (1 - v)
in which p is the soil mass density, V. is shear wave velocity, VLa is Lysmer's analog wave
velocity, Ab is area of the foundation base, A. is the soil-sidewalls contact area and ~z is the
frequency and shape dependent coefficient.
170
The variation of coefficient Cz with the dimensionless frequency, ao is shown in Fig.

7.29.
7.5.2 Horizontal Stiffness and Damping CoefficientsOO ,Gl,Gll,G13
The horizontal vibration mode arises when the foundations or the foundations support
structures are subjected to dynamic lateral forces generated by periodical loading such as wind
load, water wave pressure, machine vibrations or non periodical loadings such as earthquake
excitation. A number of studies of the horizontal dynamic response of embedded foundations
have been reported for circular or rectangular fully embedded foundationsBl.P8.w9.
The dynamic response of arbitrary shaped fully or partially embedded foundations have
been presented. by GAZETAS & TASSOULAS (1987). An efficient boundary element
formulation was used in the study. Results were presented in the form of dimensionless
parametric graphs for a wide range of embedment depths, base shapes, dimensionless
frequencies and values of soil Poisson's ratio.
7.5.2.1 Horizontal Static Stiffness
The soil-foundation system is shown in Fig. 7.21. From the study of the
dynamic response of an arbitrary shaped surface foundationD3 , Xthhorizontal static stiffness of
the surface foundations in the x-direction K.x,Gur and the y-direction Koy.nur respectively can be
computed by,
K 2: GL S (7.71)
8J',s1U = (2 - v) y
K :::: K - 0.21 G L {1 _ (B) } (7.72)

U,IUT ty,IUT (0.75 - v) L
The value of 5y is dependent on the base-shape parameter. The data from

Ref.D3 is compiled with the results of this study and is presented in Fig. 7.30. The value of
Sy can be approximately expressed as,
A }0.ll5 (7.73)
SJI = 2: .... 2.50 { 4 ~ 2
171
",r------------------
Fig. 7.24 Sidewall Contact Factor vs Fig. Trench factor vs Shape Depend-
Sidewall-Base Area Ratio (01) ent and Depth Aspect (01)
0 .
iii _.
_ f~Z;1
t.diiI 4:1
__ .. 6:1
0,' (o!
.SO 40
o 0,5 ',~ o O,S
Fig. 7.26 Dynamic Stiffness Coefficient Fig. 7.27 Embedded to Surface Dyn. Stiffness
vs Dimensionless Frequency (01) Coefficient Ratio vs a"
t,
Fig. 7.28 Radiation Damping Model (01) Fig. 7.29 Radiation Damping Parameter
vs Dimensionless Frequency (Gl)
172
The schematic effects of the trench and sidewall of the embedment on the
foundation response is shown in Fig. 7.31. These effects are similar to those discussed in
Section 7.5.1.1. Results for rectangular foundations indicate that both of ~re and Iwwl are
independent of the loading directions or,
(7.74. a)
(7.74.b)
It was found that the Poisson's ratio p of the soil has no effect on ~re and I wall
Itre is affected by a single parameter D/B, Lwl is affected by dimensionless parameters hlB and
Av,IV as shown in Fig. 7.32 and Fig. 7.33 respectively. The best fit of the results to those
figures leads to,
1.. " 1 + 0.15 ~~ < 1.20

(7.75)
11llall =
h Aw
1 + 0.52 - -
}OMl (7.76)
{ B Ll
where Aw is the effective contact area of the soil-sidewall (Aw < A.) and h is the distance from
the mid-hight of the sidewall to the ground surface.
The static horizontal stiffness of an arbitrary shaped embedded foundation can

be computed by considering both coefficients in Eq. 7.75 and Eq. 7.76 thus,
(7.77.a)
(7.77.b)
7.5.2.2 Horizontal Dynamic Stiffness
The effects of the excitation frequency on the dynamic horizontal stiffness of

the embedded foundation are represented by the dynamic horizontal stiffness coefficients kx,emb
173
,
za
J
"
0
0 0Z O 0& 06 '0
A.
<'IL'
7.30 Horizontal Static Stiffness Fig. 7.31 Trench and Sidewall Effects (02)
I'ameter vs Shape Factor (G2)
u.------------------........,
l'O
f
(WO.lfo05-2(~,,,, .
a IT
II K"0{1
llCiiI K
." o tie ..
C;
1
l
I.~ o
~}
Cl 10
DIB
h AW)""O
( s'LT
Fig. Trench Factor vs Depth Ratio (02) Fig. 7.33 Sidewall Factor vs Shape and
Depth Parameter (G2)
1.18:2 LIB .. 2
kId/D)
kld/O,,"
OIl! .0
...... ,~"""-......".....t~
.......?......=. .
C! ;)
Fig. 7.34 Horizontal Dynamic Stiffness Co- Partial to Full Horizontal Dyna-
efficients (G 12) mic Stiffness Coefficient 12)
174
and Is,emb' In general, there are four parameters that influence these coefficients, the
dimensionless frequency, the aspect ratio LIB, Poisson's ratio and the embedment aspect ratio
diD. The effect of the soil Poisson's ratio is relatively small and its effect is neglected. Plots
of kx , ley versus the dimensionless frequency .for LIB = 2 and for different values of DIB are
shown in Fig. 7.34. The values of k~,emb and Is,em!> can be determined by considering the effect
of the embedment aspect ratio dlB as illustrated in Fig. 7.35. The horizontal dynamic stiffness
are therefore,
(7.78.a)
(7.7S.b)
where Kdy,emb and K.u,emb are the dynamic horizontal stiffnesses III the y and x-directions
respectively where x is the long direction.
7.5.2.3 Horizontal Damping Coefficient
The horizontal damping coefficient of an embedded foundation is initially

generated by both shear waves V. and compression-extension waves V00' Some references take
the compression-extension waves Va:> as equal to Lysmer's analog velocity VLa , When the
horizontal motions oscillate in the x-direction the compression-extension waves Voe propagate
in the x-direction and shear waves propagate away in a direction perpendicular to the
foundation motions. Other shear waves initiated at the foundation base propagate away to the
surrounding soil with velocity V._ The equivalent energy dissipating at the foundation base can
be expressed as,
(7,79.a)
(7.79.b)
cby -- 'y- P A b V8
where Cbx , Cby are the damping effects generated by a shear wave at the foundation base in the
x and y-directions respectively and ~x, ty are frequency and shape dependent coefficients in the
x and y-directions respectively.
175
r
Plan
T
- K
21.
1
--
v
Cy 28-
'SWAVIIH.i I'll Hom09"'''OUS HillispacOl

Shear Modulus: G
POisson's r~tIO: v
Mass llenSIly: P
a
Fig. 7.36 HaL Radiation Damping Parameter Fig. 7.37 Rocking Mode (H7)
vs Dimensionless Freq. (012)
r..,wart r.:,w.u
8 f..w.n _lm gH(Vns(~),?:'y 8
7 III 7
6 j/ 6
5
.}
7 t::
::>
4
"
3
7
,
3
d
(]
2 3 4 5 6 2 3 <1 56-7 8
~ l1+ ~(rt'(~}'''J (tt[t5 (ft(~tJ
Fig. 7.38 Sidewall Factor vs Shape Para- Fig. 7.39 Sidewal1 Factor vs Shape Para-
meter for Rocking Mode (H7) meter for Rocking Mode (H7)
t!
...." 0.5
1.5r-~--':""'1-----'
us 0.5
1.5 2
a.r~
Fig. "7040 Rocking Dynamic Stiffness Coeffi- Fig.7.41 Rocking Damping Coefi'.
cient of Surface Footing (H7) of Surface Footing (F4)
176
Both these radiation damping coefficients are functions of the dimensionless frequency,
the LIB ratio and the soil Poisson's ratio. Plots of ~y versus the dimensionless frequency for
various LIB ratio is shown in Fig. 7.36. It was found that the dimensionless damping
coefficient in the x-direction or longitudinal direction can be approximated by
(7.80)
The other energy dissipations generated by the sidewalls Cw are due to shear waves V.
propagated away in a direction perpendicular to the horizontal motions and compression-
extension waves with velocity Vce propagate away in direction of the horizontal motions. Total
energy dissipated by sidewalls Cw can be computed by
(7.81)
where C. and ~ce are the dimensionless damping coefficients referring to the propagating
directions of V. and Vre respectively, Aws is the soil-sidewall contact area parallel to the
horizontal motions and Awce is the soil-sidewall contact area perpendicular to the horizontal
motions.
Values of the dimensionless damping coefficients ~s and i\" depend on

dimensionless frequency and the aspect ratios of diD, DIB and LIB. Numerical results suggests
that for DIB < 2, an approximate value of ~s = (;ce = 1 can be used for practical applications.
Finally. assuming that the foundation base and sidewalls radiate independently. total energy
dissipation of the embedded foundation in term of radiation damping coefficients is,
(7.82)
4.5.3 Rocking Stiffness and Damping Coefficient!...II7,F4,G12
Among of the vibration modes, rocking is recognized as the most critical mode of
oscillation for a foundation, since it is associated with very small radiation damping. Rocking
modes arise when the structure suppOlted by the foundations is subjected to dynamic lateral
loadings.
177
The dynamic responses of partially embedded rocking footing foundations were reported
by HATZIKONSTANTlNOU et.all (1989) and FOTOPOULO et.al (1989). An efficient
boundaLy element formulation was used in this study. Results were presented mainly for
rectangular foundations with a base aspect ratio LIB < 10, a wide range of dimensionless
frequencies ao < 2, a wide range of embedment depths D/B < 2 and three values of Poisson's
ratio, Results were presented as dimensionless parametric graphs and in approximate formula
resulting from the best fits tothe available data.
7.5.3.1 Rocking Static Stiffness
The rocking static stiffness as discussed in Ref. H7 was the extension of the
study on Dynamic Response of Arbitrary Shaped Foundations presented by DOBRY &
GAZETAS (1986). The rocking foundation model is shown in Fig. 7.37. For the static
rocking stiffness of a surface foundation loaded in the longer axis or in the x-direction K.rx.5IJr
is always smaller than that respect to the y-direction K,.I)',.ur' These static stiffnesses can be
computed by,
K - G (1 )0.'15 S (7.83)
lIYX,fur - (1 _ v) bx n
(7.84)
where,
(7.85)
(7.86)
where I bx and Iby are the foundation base moments of inertia with respect to the x and y-axis
respectively.
The effects of foundation embedment on the static stiffness are similar to those
of the previous discussion. The rocking static stiffness of an embedded foundation increase
mainly because of the trench effect rr.lre and sidewalls effect rr,woIl' Normally r IX,lre or r 1)'.1", are
greater than 1, but numerical results indicate that for practical applications,
178
(7.87)
The coefficient rr,wall depends on d,D,B and L in a complicated way. It was

found that the best correlation of this coefficient and the dimensionless parameters d/B, B/L,
DId and d/L are given in Eq. 7.88 and Eq. 7.89 or as shown in Fig. 7.38 and Fig. 7.39
respectively.
r n,wali ~ 1 +
d [ 1
1.26 11 +
d
B {D}O.20{B}O.56
d L 1
"
(7.88)
r ""wall 0;$ 1 + 0.92

f.lLd}O.66 [ 1.50 + {tl}I.90{D}O.60
L d 1 (7.89)
7.5.3.2 Rocking Dynamic Stiffness
The dynamic rocking stiffness of an embedded foundation can be computed by

taking into account the effect of the excitation frequency. This effect leads to the dynamic
stiffness coefficients k.x,emb and k.r,emb respectively. The numerical results indicate that depth of
embedment have no substantial effect on either ~x and kJ)" These coefficients can be taken as
the values of k.x,sur and kJ)',our shown in Fig. 7.40. Accordingly, the dynamic rocking stiffness
of embedded foundations can be computed as
(7.90.a)
(7.90.b)
7.5.3.3 Rocking Damping Coefficient
The rocking damping coefficient of an embedded foundation is relatively

complex. There are four main components of damping that contribute to the total rocking
damping coefficient. Two damping components were generated by energy transmitting away
179
by compression-extension waves at base and sidewalls, shear waves and torsional shear waves
at the sidewalls.
When a footing foundation rotates with respect to the x-axis, the compression-
extension waves will be initiated at the foundation base and at sidewalls parallel to the x-axis.
The energy dissipation of these waves are denoted by Cbrn and Cwrx,l respectively. The energy
dissipation can be written as,
(7.91.a)
(7.91.b)
where d is the effective contact area between the soil and sidewalls, cOO[' cory, Crx and c..r are the
dynamic rocking damping coefficients associated with the base and sidewall energy dissipations
with respect to the x and y-directions respectively.
The values of Corx and ~ory for any value of dimensionless frequency and the base
aspect ratio LIB can be determined from the graphs in Fig. 7.41. The coefficients Crx and Cry
can be computed by the approximate formulae,
(7.92.a)
(7.92,b)
The other two waves are shear waves generated away from the sidewalls parallel with
the y-axis and torsional shear waves parallel to the rocking motion. The torsional shear waves
is generated with the rocking angle. The dissipation energies are denoted by Cwrx,2 and Cwrx ,3
respectively. These energy dissipations can be expressed as,
(7,93.a)
180
(7.93.b)
Finally, the total rocking damping coefficient in the x-direction C rx emb of an embedded
foundation and can be computed by summation of the damping components in Eq. 7.91 and Eq.
7.93 to yield,
C",,.,,,,, =cbrx +Cwn,l +Cwrx,2 +Cwn:,3 (7.94)
The rocking damping coefficient with respect to the y-axis can be computed in a similar
way.
7.6 EFFECTS OF MATERIAL DAMPING
As described in Chapter IV the cyclic behaviour of soils is nonlinear and hysteretic.

The degree of non-linearity depends on the plasticity index PI, while the size of the hysteretic
loop depends mainly on the level of soils shear strain 'Yos' At very low shear strains the
hysteretic loop may be very small and sometimes is often assumed to be linear. In general,
even though the amplitudes are very small, soil possess a hysteretic loop. The area of this
hysteretic loop indicates the amount of energy dissipated per loading cycle. This energy
dissipation is often referred to as the material damping of the soils denoted by ~m'
The dynamic stiffnesses and damping coefficients discussed previously have not been
incorporated in the effects of the soils material damping. It is also assumed that the material
damping is independent of the soil Poisson's ratio. The effect of soil material damping ~m on
the shear wave velocity V\m can be expresses in complex form asPs,
VC = V,. (7.95)
B,m (1 -
'J:)
i ... ",
The complex soil shear modulus Gem can be written as,
(7.96)
The dynamic impedance of the soil-foundation interaction can be written as,

181
(7.97)
where K. and Kd are the static and dynamic stiffnesses respectively.
The effects of soil material damping ~m on the soil-foundation stiffness and damping
coefficients are discussed in Ref.D4, Gl, G2, G13, H7 and Ref.F4. These effects on the
dynamic stiffness ~(~m) and damping coefficient Ci~J can be expressed by approximate
formulae,
(7.98)
(7.99)
The effects of material damping on the soil-foundation stiffness and damping were also
presented by Ref.W9 and Ref.WlO. The complex soil shear modulus and complex
dimensionless frequency were introduced. Substituting this complex dimensionless frequency
into Eq.7.30 and after rearrangement leads to the approximate values of the dimensionless
coefficients of the dampers 'Y' s and masses JL'. as follows,
(7.100.a)
(7.100.b)
The stiffness and damping coefficients are computed from Eqs. 7.34 and 7.35.
7.7 EFFECTS OF SOIL AND FOUNDATION MASS ON THE STIFFNESS

COEFFICIENTS
During calculation of the stiffness and damping coefficients described above, the
foundations were assumed to be massless. The actual foundation possesses its own mass. The
presence of the foundation mass will affect the dynamic stiffness coefficient k's. When a
foundation with mass M*, spring stiffness K" and damping coefficient C* is subjected to
harmonic excitation pet) = Po eio.(, the corresponding steady state response is u(t) = U o eio.( and
the force-displacement relationship at vibration mode i can be written as,
182
(7.101)
The values of M*, I~* and C;* in Eq. 7.101 are chosen such that the real and
imaginary part ofEq. 7.101 are equal to Eq. 7.25 hence,
('1,102)
(7.103)
The right hand side of Eq. 7.103 contains two parameters Kt and M*, an unique
relationship may be achieved by choosing K;" = K.,j then,
(7.104)
Eq. 7.104 indicates that if the mass of the foundation is considered, the dynamic
stiffness coefficient k j decreases depending on the foundation mass-stiffness ratio. Normally this
ratio is very small ( 1.5 %), therefore in most analyses of soil-foundation systems the effect
of the foundation mass on the dynamic stiffness is considered to be negligible. If the amount
of soil mass M, is assumed to contribute during the soil-foundation vibrations, the dynamic
stiffness coefficient becomes,
(7.105)
For the system subjected to vertical or horizontal excitations, M" represents the mass
of the foundation. In the case of a rocking oscillation mass M" is replaced by mass moment
of Inertia Int"' with respect to the axis of rotation.
7.8 MACIDNE FOUNDATIONS AND EARTHQUAKE ENGINEERING
The dynamic stiffness and damping coefficients of the footing foundations as discussed
above are derived from the steady state harmonic loading applied to the footing foundations.
In earthquake engineering, however, the foundation of the building structures is subjected to
183
non-harmonic transient loadings through the base of the foundations. WILSON & DOVEY
(1972) stated that the research associated with machine foundation problems has only a little
value in earthquake engineering because of the different loading types and sources. Following
this statement, GAZETAS (1991) stated that the information from machine-loaded foundations
could also be used to assess the dynamic soil-foundation structure interaction during seismic
shaking. Because of a lack of information regarding the calculation of soil-foundations stiffness
and damping coefficients subjected to earthquake loadings. the information from machine-
loaded foundation can be adopted.
The dynamic response of foundation vibrations have been investigated for surface
foundations resting on an elastic half spaceG4N1L5R2V2V3.V4.W2.W6.W9. or embedded
foundations AS 1l1 G2 )2.H7,N2.w9.wlo. The soil-foundation stiffness and damping coefficients become
more important when lumped parameter methods are used during analyses. The stiffness and
damping coefficients of surface foundations may be frequency independent or frequency
dependent. Results from the analytical investigations and laboratory tests, however, indicate
that under dynamic loading both the stiffness and damping coefficients are frequency
dependent.
The analysis of circular or rectangular fully embedded foundations have been

reportedBI.P8.W9.WIO. Most of the analyses consider that the soil-foundation responses were
frequency dependent. In reality, the building foundations are not fully embedded but partially
embedded. It was found during laboratory tests that the degree of the foundation embedment
has a significant effect on the foundation responses. Using formulae for fully embedded
foundations for partially embedded gives a considerable overestimation of the coefficients.
The dynamic responses of arbitrary shaped fully or partially embedded foundations have
been reportedD3F4.GI.G2.H7. The effective boundary element method was used for the analyses.
The results were presented in dimensionless parametric graphs for a wide range of
dimensionless frequency, base shape aspect ratios LIB, a wide range degree of the embedment
depth and some values of the soil Poisson's Ratio Il. For convenience, some of the results are
presented mathematically by an approximate formula resulting from curve-fitting of the
available data.
184
In general, soil material damping decreases the dynamic stiffness, but increases the
damping coefficient.
For simplicity during the analyses of soil-foundation response, the foundations are
assumed to be massless. The effect of the foundation mass is usually relatively small and for
practical purposes can be neglected.
Machine foundation vibrations and building footing foundations are subjected to

different sources of energy. However, information from machine-loaded foundations can be
used to assess the dynamic soil-foundation structure interaction during seismic excitationG12
Chapter VIII
The Analysis Procedures
8.1 INTRODUCTION
The soil-structure interaction effects not only stiff structures supported by soft soil but
also embedded structures, where an increase in soil-foundation contact causes modification of
the structural frequencies and increases the radiation damping. This latter area has received
great attention in the design of nuclear power plants. Greater confidence in the dynamic soil-
structure interaction has been achieved since the state of the art of structural dynamic and the
dynamic finite element was developed at the same time to accompany the previous methods of
the theory of the elastic half-space or continuum solutionsH2
The understanding of dynamic soil properties as presented in Chapter IV is one aspect

of the development of the analysis of soil-structure interaction. Another aspect refmement of
the soil-foundation modelling and impedance as discussed in Chapters V and VII. Further
research on the aspects of soil-structure interaction involved the new methodologies,
implementation and efficiency of the numerical solutionsRlO These aspects will be briefly
discussed in this chapter.
8.2 THE METHODS OF ANALYSIS OF SOILSTRUCTURE INTERACTION
The methods of analysis of soil-structure interaction can be divided broadly into two
main groups, The Lumped Parameter and The Finite Element Methods. Some techniques have
been developed to improve the accuracy of these methods.
8.2.1 Lumped Parameter Method
This method sometimes called the simplified substmcture method GlO or the impedance
methodN9 Hl The structural system is divided by two main parts, the substructure consisting of
the soil-foundation interaction and the supported superstructure. For a given earthquake
excitation the analysis procedures has two main steps; firstly, determination of the soil-
foundation impedance without the superstructure and secondly, combining this impedance into
185
186
the equations of the structural system. In most applications the input motion is assumed to be
at the level of the foundation and the same as the free field motionH20 with the reason as
described in Section 1.1. The simplest structural system is the stick model C3 C4 c1 ,S9,Tl9,Z5
supported by a rigid foundation. An example of this structural system is shown in Fig. 8.1.
In this method the soil-foundation impedance is represented by a spring and dashpot to

simulate the soil-foundation stiffness and radiation damping respectively. In the early
application of this method, the soil-foundation impedance was assumed to be frequency
independent, but recently these have been improved both in the soil and foundation modelling
and in the analysis methods as presented in Chapter VII.
Once the frequency dependent soil-foundation impedance is computed the overall

differential equations of motion of the structural system is written in the frequency domain
requiring the compatibility of the displacements and the equilibrium of forces at the foundation
interfacecB as shown in Fig. 8.2. The response of the structural system can be usually computed
by a numerical integration of the governing differential equations of motion. This frequency
domain analysis required a great of computation effort as discussed by Ref.C6. In the case of
a structure resting on the homogeneous elastic half-space, a drastic reduction of the
computational effort can be achieved by using constant impedance coefficients as suggested by
Ref.P34, T19 and ZS.
For a layered soil system this simplified analysis is not justified because the impedance
function as shown in Fig. 5.12 and 5.13 are very sensitive to the dimensionless frequency a.,.
ROESSET (1991) pointed out that the dimensionless frequency a., is only less significant for
seismic excitation than for machine foundations because the excitation frequencies are lower
as discussed in Section 4.1.
This method has several limitations; the assumption of a rigid foundation over the entire
building foundation is not appropriate in many situations where the coupling effects of the
neighbouring footing foundation cannot be incorporated, the effect of the non-linearity and
variability of the soils beneath the foundation in general cannot be directly included and the
impedance coefficients for layered medium are restricted to a horizontally layered soil model.
The non-linearity of the dynamic properties of soil was discussed in Chapter IV. Using
the maximum likely soil shear strain 'Yss as discussed in Section 4.3 the use of the equivalent
187
,
m,l
I, "'.J
T t: I
fi
r U""
h
1
k
2
.EI,
~
ill
",
1.41
1f IJI(t)
I'IAlFSPOCE
(l(t)
Fig. 8.1 StI1lcture with Rigid Foundation Fig. 8.2 Forces and Displacements Compatibi..:
on an Elastic Half-Space lity of Lumped SDOP on Half-Space (C8)
s - DEGREES OF
FREEDOM (DOF)
STRUCTURE
b - DOF
I
wlrr
I I I I I !I,
.Irr I I I I lf~
n""
I i

I I !Ii'
"I7r I J I I I I Ii},.
>I'" I I ! ! I I I TIl"
I .! ! I Ti~
,r.~;~ Vlt .'11;>'\ 'WilllYl!',
Fig. 8.3 System and Excitation in Direct Fig. 8,4 Substructure Method in Soil
Finite Element Method (G 17) Structure Interaction (G 18)
1--1---1
\
\
\
\
\
\
\
\
BJ(I<!rio( \
reQion \
"" .... InlerlO(

reQion
" .... ~-
Fig. 8.5 Typical Infinite System (L12) Fig. 8.6 Incident P-Wave (L12)
188
linear method where the soil-foundation stiffness and damping coefficients are computed from
the equivalent-linear soil shear modulusG12.P20 at a particular value of soil shear strain is still
justified. Attempts to simplify the effect of a layered soil on the soil-foundation impedance was
discussed by Ref.Hll , J2, J5 and S19.
3.2.2 Finite Element Methods
8.2.2.1 Direct Finite Element Method
The direct finite element method for the analysis of soil-structure interaction
is an obvious application of finite element technique. The system to be analyzed is a
combination of the superstructure and portion of a semi-infinite soil domain modelled as a finite
domain. Both of the superstructure and the soil domain are discretized into small elements and
excited by an earthquake eccelerogram at the base of the finite element domain as shown in
Fig. 8.3. The standard procedures for the formulation of the element stiffness matrix can be
found in many textbooks. The accuracy of the finite element technique, in general, depends on
the type and size of the elements and the selection of the displacement fields within each
elemenfI2.S1.Tl9.
One of the limitations of this method is the large computational effort required
during the analysis because of the small elements required to achieve a appropriate degree of
accuracy. The whole structural system may have several hundred to a few thousand degrees
. of freedom. The usual application of the finite element method simplified from 3-dimensional
to a 2-dimensional analysis leads to significant errors as indicated by an underestimation of the
peak response C27,GIO.H2.Tl9.
An effort to increase an efficiency of the numerical integration of a 2

dimensional structural system has been proposed by GUTIERREZ & CHOPRA (1977,1978)
by introducing a new method of analysis called The General Substructure Method. The
equations of motion are partitioned in terms of vector displacements of the structural nodal
points r, and base nodal points at the soil-interface rb as shown in Fig. 8.4. By considering
equilibrium of interaction forces and compatibility of displacements of the base nodal points
leads to differential equations of motion involving the complex value of dynamic stiffness
matrix Sru.J)(w) corresponding the soil-interface and the soil-domain displacements rr and fe
respectively. The analysis is computationally efficient because of an approximate application

189
of the first few structural modes only and assuming the dynamic stiffness matrix is frequency
independent. The effect of the finite soil domain is represented by a relatively sman rectangular
Sf(ij)(UJ) matrix with i =j = 1,2 ... n where n is the number of nodal points at the soil-
structure interfaceGlO ,G17.Glfl. The derivation of this matrix can be found in Ref.GlS.
The general finite element method, however, can be used to solve a relatively
complex and large structure involving non-regular foundation shapes, inclination and non-
linearity of the layered soil system, embedded foundations and dynamic coupling between
neighbouring rigid or flexible footing foundations.
8.2,2,2 Finite Element witb Artificial Boundaries
In many dynamic problems such as acoustics, fluid dynamics, blast effects or

foundation vibrations, wave propagation is the main difficulty in a rational approach to the
analysis ofthe system. An example of wave propagation in an infinite system is shown in Fig.
B.5 where an exited zone is enclosed by an imaginary convex artificial boundary. The wave
energy propagates from the excited zone in the interior region and all energy arriving at the
boundary will pass into the exterior region to infinity. Thus, no wave energy is reflected back
by the artificial boundary into the interior region. This phenomenon is a limitation of the
application of the direct finite element method to the analysis of soil-structure interaction. The
problem is the difficulty in modelling of a far-field infinite medium in the finite domainK17 in
which no wave energy is reflected back into the finite element domain. If the artificial
boundaries are selected far enough away from the exited zone, the number of degrees of
freedom becomes large and a huge computational effort is required in the analysis. Proposed
artificial boundaries are described in the following sections.
Elementary Boundaries: These boundaries, either fixed, free or rolling boundaries are called
perfect reflectorsK15 since no energy is absorbed or transmitted at the boundary. In the case of
a bounded soil medium a fixed boundary is usually used at the base while the a free or rolling
boundary is usually used at the vertical boundaries. These types of boundaries are usually
utilized in the direct finite element method.
Local Boundaries: If the artificial boundary is able to absorb or transmit the impinging wave
energy then it is called a transmitting boundary. These boundaries are called imperfect
190
transmitters because the boundary only performs a perfect transmitter under a certain
conditions. The term local implies that in the time or frequency domain, coupling between
viscous dashpots occurs only between the neighbouring boundary pointsXI.KI5. The first local
boundary is the standard viscous boundary presented by LYSMER & KUHLEMEYER
(1969)Kll'i. This viscous boundary is frequency independent consisting of independent normal
and tangential forces generated by a viscous damper as shown in Fig. 8.7. It was demonstrated
that this boundary performed as a perfect transmitter when the angle of incident wave 0 is
greater than 40 as shown in Fig.8.6.
Another artificial viscous boundary called the Unified Boundary was presented
by WHITE et al.(1977). This boundary is also frequency independent as modelled by a normal
and tangential dashpot at the boundary. If the direction of the arriving wave is known the
viscosity matrix of these transmitters can be evaluated such that the perfect energy transmittion
will be achieved. It was found that this artificial boundary is more efficient than the standard
viscous boundary. Several local transmitting boundariesKl6.wlO were proposed such as Ang-
Newmark Boundary (1972), Smith Boundary (1974), Paraxial or Engquist-Majda Boundary
(1977) and the extrapolation or Liao-Wong Boundary (1984). Ref.K6 showed that although
these transmitting boundaries appear different they are mathematically related. The choice of
boundary can be based on ease of use, accuracy and stabilit:yK16.
Consistent Boundaries: WOLF (1988) stated that the local transmitting boundaries are just
approximations and more rigorous transmitting boundaries would be frequency dependent where
the damping coefficient can be expressed as a function of the dimensionless frequency ~. The
Consistent Boundary is frequency dependent and requires a frequency domain analysis for a
linear system and is difficult to use in a non-linear problemw27. As presented by KAUSEL et
al.(1975) this transmitting boundary is discretized so that the boundary nodal points are fully
coupled KI5 in the form of a frequency dependent boundary stiffness matrix. Because this
boundary performs as a perfect absorber the transmitting boundary can be placed directly on
the edge of the structure greatly reducing the number of degrees of freedom in the system. This
transmitting boundary is frequency dependent and in order to get the simple eigenvalue
problem, it often requires the far-field horizontal rigid base boundaryKls. For most engineering
applications, the non-linearity of the far-field may not affect significantly the structural
responsezs and only an insignificant reflection will occur on the horizontal boundaryL13. The
effect of this vertical radiation was ignored in some soil dynamic studiesK1s .
191
Fig. 8.1 Viscous Boundary in Finite Element for Footing on Half-Space (L12)
"online.f
SoU""Stnu::lura IWILhndl'I r,o'
2n:hr6cthu'l
, UplH' (' Jalnt E1 ... nt

~nJd.. C j ' nheUve Slreu
S.pI':tclIn4\ ~~~~
( . Ltqu.rllC liOn
M"l1'1odIJ
'8t" Soli
Tf.A.n'IJeilting 8nundc.ry ~ -OfiiBhpot
( ,Sup4rhpClllnr
'wv ...
Infinite :hdl
Fig. 8.8 Transmitting Boundary in Soil Fig. 8.9 Discretization with Finite and
Structure Interaction Boundary Elements (E1)
Layer
Hoff
space
Fig. 8.10 The Strip Foundation and the

Finite Element Mesh
192
The comparison of the performance of elementary, local and consistent
boundaries on the surface footing compliance was reported by ROESSET (1977). An excellent
agreement between the consistent transmitting boundary and the analytical solution whereas the
solutions for the elementary and the local boundary depend strongly on the location of the
vertical boundary and the degree of soil material damping. Similar comparisons were reported
by KAUSEL & TASSOULAS (1981). It was found that the third degree approximation of the
Engquist & Majda transmitting boundary were similar to the consistent boundaly and if the
exciting frequency is less than the fundamental frequency of the soil stratum, the transmitting
boundary is not important.
Recent developments of the transmitting boundary have led to more powerful

analyses such as hybrid model GI9 ,R12 and boundary element methods A18 ,E7,K19,K20,M27,SJ2,ZS for use
in soil-structure and soil-foundation interaction analyses as shown in Fig. 8.8, 8.9 and 8.10
respectively. The near-field is modelled by an ordinary non-linear finite element method while
the far-field is usually assumed to behave linearly. ZHAO (1989) used the boundary element
method to analyze the compliance coefficients of a strip foundation with an horizontal viscous
boundary for the response of elastic and inelastic superstructures resting on stiff and soft soils
using a fixed horizontal boundary. Compared with analytical solutions, a good agreement of
the compliance coefficients was found. If the fundamental period of the soil-structure is less
than the period of the site a soil-structure interaction should be carried out while if the
fundamental period of the structure is larger than that of the site the soil-structure interaction
will reduce the structural response.
8.3 THE METHOD OF ANALYSIS USED IN THIS STUDY
The methods of analysis for soil-structure interaction have been discussed. The
comparison between lumped parameter and finite element methods on the soil-structure
interaction analysis were discussed in Ref.C27, H2, Sl, T19 and W28.
In this study, however, the Lumped Parameter Method was used for the analysis for
the foHowing reasons; firstly, this method is relatively simple, requires less computational
effort and is fundamentally correct"9, secondly, this method can easily be incorporated in
general purpose structural dynamic program working in time domain w19 such as
RUAUMOKO c14 , thirdly, the calculated foundation impedance can be used in the interaction
193
response of a of variety of building structuresTl9 and this method is easily used in a 3-

dimensional analysisH2
8.4 THE DIFFERENTIAL EQUATIONS OF MOTION OF THE STRUC1URAL

SYSTEM
The flexibility of the soil-foundation system will exhibit three in-plane vibration modes,
veltical, horizontal and rocking modes. If the superstructure possesses N degrees of freedom
as shown in Fig. 8.11 then the structural system have N +3 degrees of freedom as indicated
by the corresponding differential equations.
The differential equations of motion can be derived in three steps; firstly, taking an
equilibrium of inertia, damping and elastic forces of the superstructure, secondly, considering
dynamic equilibrium of the whole structural system in horizontal, rocking and vertical vibration
modes and thirdly superimposing these equations of motion. The building is assumed to deflect
in a shear mode and the structural mass is assumed to be lumped at every floor level as shown
in Fig. 8.11. If the inertia, damping and elastic force are denoted by F j , Fe and Fk: respectively,
the dynamic equilibrium of the fixed-base structural can be written as,
(S.l)
where,
(S.2a)
(S.2b)
(8.2c)
The dynamic equilibrium of the structural system in the horizontal direction yields,
N
FI.,h == E
}=1
mj {uj + ug + ub + h) o} + mf; {Ug + Ub} (S.3a)
(8.3b)
194
(S.3c)
The dynamic equilibrium of the structural system in the rocking modes yields,
The dynamic equilibrium of the structural system of the vertical direction yields,
The governing equations of motion of the structural system leads to,
[ M] fJs~ + [ C] URI + [ K] Uee = Pg (8.6)
where,
[m] mj m}j 0
N N
{ mj } mb + E mj E mjhJ 0
/=1 j=1
(8.7)
[ M] == N
{m}j} {m}j}T Ib + E
j=l
(I.t + mjhf) 0
N
( 0 }T 0 0 E(mj -I- m b)
j4
[c ] [ c ]{ 1 } [ c l{ hj } 0
{l}T[c] Chh ... { 1 }T [ C ]{ 1 } Chr ... { 1 }T [ C ]{ 1 } 0
[ C]
{ hj}1' [ C ] CFh -I- { 1 }T [ c J{ hj } err + { 1 }T [ c 1{ hI' } 0
{ 0 }T 0 0 C"
(8.8)
195
fixed Reference
Fig. 8.11 Mathematical Model of MDOF

Structure on Half-Space
Fig. 8.12 Stiffness and Damping Coefficients Fig. 8.13 Stiffness and Damping Coefficients
in Global Coordinates (09) in Local Coordinates (09)
196
[ k ] [ k ]{ 1 } [ k 1{ hJ } (j)
{l}T[Ic] Kkk + { 1 } l' [ k ]{ 1 } KIu' + { 1 } l' [ k ]{ 1 } 0

[K]
{ hj} T [ It; 1 Krio + { 1 } T [ k 1{ h} } Krr + { 1 }T [ k 1{ If} } 0
{ 0 }T 0 0 K"
(8.9)
- { [ m] + [ c ] + [ Ie ] } iII?
N
- 1: mjUg - { 1 }T [ C ]{ 1 }Ug - { 1 }T [ Ie ]{ 1 } Ug
}=1
(8.10)
pt = N
-E mj hj fI,. - { 1 }T [ C ] { hj } il, - { 1 }T [ Fe ] { hi } Ug
j=l
and the displacement vector of the structural system U.s is,
Usa = (8.11)
where [ m], [c] and [ k ] are mass, damping and stiffness matrices of the fIXed base system,
Ug and Ogv are horizontal and vertiCal ground acceleration respectively, U b, Uj' hJ and mj are
horizonta.l base and floor displacement, height and mass of floor level-j respectively, (J, Ib and
~ are the rocking angle, mass moment of inertia of the base and floor-j respectively, Cbb , C hr
= C rh , Crr , Khh , Kbr = K.h and K.r are the soil-foundation damping coefficients and stiffnesses
with respect to the horizontal, the coupling between horizontal and rocking and rocking modes
respective ly .
If the vertical vibration mode is ignored, the mass, damping and stiffness matrix of the
structural system are full populated matrices. In this analysis the response of the superstructure
is considered to the inelastic where the vertical vibration mode and the coupling between
horizontal and rocking modes are neglected as presented by Ref.C6, N4 and TIL When the
damping and elastic forces are only due to the differential velocity and displacements between
lumped floor mass. Eq. 8.7, 8.8, 8.9, 8.10 and 8.11 lead to,
197
[m] mj
N N
[M]
{ Inj } L
)=1
mj L
)=1
(8.12)
N
{ } {mfij}T L (lj + 2)
)=1
[ C ] {9 } { 4) }
[ C] ::::: { {8 }T Clio 0 (8.13)

{ {) }I' 0 Crr
[ Ie ] {0 } { (I }
[K] "" { 0 }:r K;u. 0 (8.14)
{ 0 }T 0 K,.,.
- {m } Uf{
N
P t ""
-E )=1
m) UK (8.15)
N
-E mj h) Uif
)=1
and the displacement vector of the structural system U" is,
(8.16)
In lumped parameter methods the building prototype is often assumed to have a rigid
foundation over the entire building base so that the soil-foundation impedance in the vertical,
horizontal and rocking stiffnesses are presented as global stiffness as shown in Fig. 8.12. In
rea1ity, however, the entire foundation is not necessarily rigid especially if the building has a
rigid core or has individual footings. In this case the soil-foundation impedance is represented
in local coordinates as shown in Fig. 8.13.
The eqUilibrium of forces of the structural system in the horizontal direction is fully
satisfied because the global soil-foundation stiffness and damping coefficients are simply
198
represented by the summation of the local coefficients. A problem arises from equilibrium of
forces in the rocking mode because the vertical and rocking soil-foundation stiffnesses and
damping coefficients are represented in local coordinates. To satisfy Egs. 8.13 and 8.14, the
local vertical stiffness and damping coefficient should be chosen so that both the global vertical
and rocking impedance are compatible with each other. These conditions are difficult to achieve
. simultaneously because the local vertical and rocking impedance are computed individually
while the tie-beam cannot be assumed to be rigid. Ref.D9 stated that this discrepancy may not
matter in the analysis where the horizontal and vertical excitations are not applied
simultaneously.
8.S OVERALL DAMPING OF SOIL-STRUCTURE INTERACTION SYSTEMS
Most researchers realize that the prediction of the real damping in dynamic problems
is less well developed when compared of masses and stiffnesses. This problem involves the
mechanism of damping, damping systems, damping types and the magnitude of the damping
forces. Depending in whether the structural response is linearly-elastic or inelastic, the
structural system has a fixed base or a flexible base, the analysis is in the time or frequency
domains will influence the damping systems, damping types, damping quantities and the
solution of the equations of motion. An analysis approach based on these parameters should be
defined so that the method of solution of the equations of motion can be selected. These
approaches are systematically shown in Fig. 8.14 and is described in the following sections.
8,5.1 Damping in Linear-Elastic Analysis
In elastic analyses the damping mechanism choices cover damping types such as viscous
or coulomb damping and either classical or non-classical damping system. HUMAR (1990)
stated that the present assumption that the viscous damping may be linear for small velocities
or non-linear for high velocities, whereas the real damping systems depend on magnitude and
distribution of the damping ratiosC22D7
As shown in Fig.8.14, for the fixed base linear elastic response the mass, damping and
stiffness matrices remain constant. If the damping ratios of the structure are relatively small
and equally distributed, a modal superposition method is usually justified. In this case the mode
shapes are independent of time. The discrete coupled equations are transformed into a set of
199
uncoupled or independent modal equations by assuming that the mass, stiffness and damping
matrices satisfy the Olthogonality condition. The damping model which satisfies the
orthogonality condition is called a classical or proportional dampingC22.w13 and the
corresponding structure possesses classical modes. The response of the structure can be
computed by a numerical integration of the uncoupled modal equations where the damping is
represented by a viscous damping ratio. This approach is represented by the first path in Fig.
8.14.
When an elastic structure is supported on a flexible soil-foundation with frequency

independent impedances, the mass, stiffness and damping coefficient are constant with time.
However, the damping is no longer equally distributed over the entire structural system because
the soil-foundation interaction possesses higher damping than the superstructure. The damping
causes significant coupling of the discrete and the modal equations and resulting non-diagonal
damping matrix as written in Eq. 8.8. In general, the mode shapes are different from those of
the undamped free vibration and the damping is not orthogonalDB The standard modal
superposition method is invalidatedC22 because the structure does not possess standard classical
modesC28. This type of damping is called non-classical or non-proportional damping and follows
the second path in Fig. 8.14.
Several approaches have been proposed where the non-proportional damping has mode
shapes, by using the multiple transformations where the damping matrix is diagonalizedC2s, by
using multiple coordinate transformations and approximate modal equationsTIl, by ignoring the
off-diagonal coefficient matricesC22.W13 and by solving a damped free vibration system leading
to complex eigenvectors and eigenvalueslll.S20.V7. The latest approach requires a greater
computation effort. Approximate modal shapes ignoring off-diagonal coefficient matrices seems
to be the good choice because the computation is relatively cheap compared with the case of
complex eigenvaluesw22 . Moreover, results from the numerical integration of the modal
coordinates with coupled damping by using only the first few modes with closely spaced natural
frequencies provides good agreement in the response amplitude when compared with the
response ,from the discrete coupled equationsC22W13.
In the case of a linear-elastic structure supported by a frequency dependent soil-

foundation impedance, the stiffness and damping matrices will be frequency dependent. This
structural system possesses non-proportional damping and does not have classical modes.
Accordingly, the analysis of the structure should be carried out by a direct numerical integration
200
Struct. Response Frequency Domain Time Domain
-.J I
1. Freq. Independ.
Fixed Bases
Constant of Mass,
Damping and
Stiffness Matrices
Linear Elastic --JiIIo- 2. Freq. Independ. ~ [m],[c], [k]
Analysis Flex. Found.
-...!3.Fre Q Dependent
Flex. Found. 1,2
4.Freq. Independ.
--JiIIo- ~
Fixed Bases
Changes of Mass ~ Damping Ratio
and/or Stiffness and
and/or Damping Its Distribution
Matrices r-
Inelastic 5.Freq. Independ.
I~ ~ [m],[c],[k]
Analysis Flex. Found.
M~
Equally
High or Equally
6.Freq. Dependent High and Small
L..JJ..
Flex. Found. Small
,
3,5,6 2
, 1,4
Non Proportional or Non Classical Damping """ill- Pro port. Damping
2,3,5,6 4 1 ,
Non-Classical Mode Shapes """ill- Classical Modes
2,4,5
Num. Integration of :
1. Mod. Equations (') ...
2. Coupled Eq. (2,4,5)
in Time Domain
3,6
,
Numerical Integration of the Coupled Equations
in Frequency or Time Domain (3,6)
Fig. 8.14 Flow ReI. between tha Str. Response and Damping Systems. Methods and Domain of Analyses
201
of the original coupled equationsC22.TIl in the frequency or time domain as indicated by the
third path in Fig. 8.14. To reduce the computational cost Ref.C6 used the contribution of
the first few modes in the integration of the modal coordinates in which the structure was
assumed to possesses the classical modes of the fixed base structure. The result of this
approximation was presented in Section 2.3.1.
8,5,2 Damping ill Inelastic Response Analysis
The stiffness matrices of an inelastic fixed base structure change with time. If the
structure possess an equally distributed and a relatively small damping ratio, proportional
damping is commonly used L14 The proportional damping may be time independent such as
in mass-proportional damping and initial stiffness proportional damping or time dependent
such as in instantaneous stiffness-proportional damping and time dependent Rayleigh
damping in which the damping matrix is proportional to the mass and instantaneous stiffness
matrices.
OTANI (1981) and FAJFAR et al.(1993) investigated the effect of damping types
on the structural response of SDOF system while LEGER & DUSSAULT (1992)
investigated the MDOF structures. For SDOF structures, stiffness-proportional damping
type provided less energy dissipation than mass-proportional damping but more effective
damping in small amplitude load reversals. In the case of MDOF structures, if Tl :s; 0.5
sec. the use of instantaneous stiffness-proportional damping was suggested, when Tl ;;.:: 0.5
sec. the Rayleigh damping should be used. For relatively flexible structures with Tl ;;.:: 1.S
sec., the seismic response was not affected by the type of Ray leigh damping. In the inelastic
response, the system does not have classical mode shapes because the mode shapes are
independent of time and frequency of excitation. The standard direct numerical integration
of the discrete coupled equations as indicated by fourth path in Fig, 8.14 is then required.
In the case of inelastic response analysis with frequency independent soil-foundation

impedance, the structural system possesses non-proportional time dependent damping matrix
even though the superstructure may use time dependent proportional Rayleigh dampingC22.HB,
It is clear that this structural system will not have classical modes because both of the
damping and stiffness matrices are time dependent. The response of the structural system
202
is computed by direct integration of the coupled equations of equilibrium as suggested by

Ref.C22 and as indicated by the fifth path in Fig. 8.14.
More complex and expensive analyses will result when an inelastic response
superstructure is supported by flexible foundation with frequency dependent soil-foundation
impedance .. The structural system will be both dependent on time and frequency of
excitation as the stiffness and proportional damping of the superstructure is time dependent
and the soil-foundation impedance is frequency dependent. The analysis will be carried out
both the time and frequency domains as indicated by the sixth path in Fig. 8.14. To reduce
the computational cost the analysis can be carried out using soil-foundation impedances with
constant impedance coefficients especially for non-layered soil model as suggested by
Ref.T19. The computation cost reduces because the analysis is only carried out in the time
domain.
8.5.3 Effects of Foundation Flexibility on the Damping Quantity
Part of the laboratory test results carried out by Ref.H8 indicated that the structural
damping of a structure with lift-off of part of the foundations was significantly reduced
when compared with the fixed-base structure. The structural steel model was nearly
undamped when the structural motion approached a rigid body rotation. Ref.N7 also
reported that the structural damping is always reduced by foundation flexibility. Compared
with the fixed-base structure, the structural damping ratio of the structure with a flexible
foundation was reduced by the cube of its circular frequency ratio. For the higher modes,
however, the structural damping is more complicated, it can be smaller or higher depending
on the soil-foundation flexibility. In the case of an inelastic response of the fixed base
structure, Ref. Ul reported that an inelastic behaviour has the effect in increasing the
effective damping ratio of the structure. The effect of the foundation flexibility on the
structural damping of an inelastic structural response, unfortunately, is not yet clear.
In this soil-structure interaction analysis the tangent stiffness Rayleigh damping was
used for the superstructure as used by Ref. C22, H8 and Z5. Because of the uncertain effect
of the foundation flexibility on the structural damping of the inelastic structural response,
the corresponding Rayleigh damping was specified by using 5 % of viscous damping ratio
for the fifth and tenth modes. To simplify and reduce the computational cost, the frequency
203
dependent soil-foundation impedance with constant imped,anre coefficients was used.
8.6 CHOICE OF THE NUMERICAL INTEGRATION SCHEME
In general, the structural response in dynamic problems may be calculated from the
governing equation of motion. The mathematical evaluation of the structural response can
use Duhamel's Integral for a linear-elastic systemH22 For a non-linear inelastic response
analysis with a relatively complex non-periodic dynamic loads such as earthquake loading
the numerical integration technique must be employed C!l,H2z,H23. The direct numerical
integration of the equations of motion at equal intervals of time gives displacements,
velocities and accelerations of the structural system.
The commonly use of numerical methods in structural dynamic are Newmark-~,
Wilson-O, Houbolt, Hilber and Central Difference methods H22 ,H23. To provide an acceptable
structural response, the minimum requirements of numerical stability, accuracy and
efficiency should be achieved. The stability of the integration scheme means that the errors
in the response parameters which may due to round-off in the numerical algorithm do not
grow during the time integration. The stability of the integration is classified as
unconditionally stable, conditionally stable and un-stable. The accuracy of the integration
can usually be achieved by utilizing a relatively small time step of integration .6.1. The
efficiency of the integration can be identified by the computational cost and storage
requirements.
NEWMARK (1959) introduced the numerical integration scheme for structural

dynamic with the parameters 'Y and ~. The value of ~ equal 1/4 and 1/6 corresponds to an
constant average acceleration and linear acceleration during the time interval, ~ = I/s is a
step in the acceleration from the initial value to the final value at the centre of the time step.
These numerical techniques are the well known Newmark-~ methods. For the best results
the time step interval 1/5 - 1/6 TI and the value of ~ ~ 1/6 - 1/4 were suggested.
Efforts to examine the stability and accuracy of the these integration schemes have
been reported in many references. Refs. B5 and H22 reported that for 'Y = 1/2 and ~ = 1/"
the numerical integration scheme is unconditionally stable, while Ref. H23 stated that this
integration scheme is unconditionally stable if'Y ;::: I/ z and {3 ;::: Y/ z. The stability of the
204
numerical methods is measured by the values of the spectral radii p(A)B5,H15,HZZ,H2.1.H~ When
its value is smaller than 1 the integration scheme is stable while if is greater than 1 the
integration scheme will be unstable. The value of l' = I/Z and f3 = 1/4 are commonly
used B5 ,CB,HZZ. The WILSON-O is unconditionally stable for () ~ 1.37 where the value of 0
= 1.4 is usually taken.
Providing these integration schemes are stable the selection of the numerical
integration scheme then will be based on whether the schemes have errors of period
elongation or amplitude decay. Investigations have been reported by BATHE & WILSON
(1973), HILBER (1976), HILBER et a1.(1977) , HILBER & HUGHES (1978) and HUMAR
(1990). Result of these investigations indicate that the Newmark-f3 family with the constant
average acceleration has no amplitude error and a relatively negligible period elongation.
The Newmark-f3 constant average acceleration is one of the options in RUAUMOKO
computer program and was selected for the numerical evaluation in this study. The time
interval At = 0.01 second was used.
8.7 CHOICE OF THE EARTHQUAKE INPUT
As mentioned in Section 3.1 the earthquake ground motion (EQ-GM) was an

important parameter in earthquake resistant design (EQ-RD). In this step, great uncertainties
arise in attempting to define the characteristics of future earthquakesBZI . In general, an
earthquake resistant design of buildings depends on two main aspects; firstly, the
characteristics of the earthquake such as the frequency of occurrence, intensity, magnitude
and ground accelerations and secondly, the design criteria depending on whether the
building is designed for a serviceability, damage control or survival limit stateBZ1 ,P18.
The peak ground acceleration PGA is an important factor in the selection of the
design earthquake even though this factor provides a poor estimation of the damage
potential, especially for a relatively flexible structureP1&.TIo. Damage observed and calculated
due to earthquakes indicated that the ground velocityPl8,M16, the velocity spectrumHZ5 or
acceleration to velocity ratio TIO,Z6,Z7 and earthquake durationZl provide better estimations than
ground acceleration.
205
Attempts to correlate the PGA with other earthquake characteristics refer to an

attenuation relationship. This is generally expressed as function of earthquake magnitude
M and epicentral distance RC29 and whether the site is on soil or rock. DOWRICK &
SRITHARAN (1993) stated that due to a lack of suitable ground motion data a reliable
attenuation of PGA for New Zealand has not been developed. The Japanese attenuation
model was adopted to fit earthquake data because of the similar geological and tectonic
similarities. For earthquake magnitudes M ~ 5.8 - 6.7 with the shortest distance between
site and fault rupture R" = 25 km, the PGA is around 0.10 - 0.21 g.
PAULAY & PRIESTLEY (1992) reported that for Western USA regions the
earthquake magnitude M = 7.5 possess a return period between 130 to 200 years.
DOWRICK (1987) stated that for buildings designed for a damage control limit a return
period of 50 to 100 years may be used. DOWRICK (1991) reported that the earthquake
magnitude ofM = 7.8 was the maximum earthquake magnitude recorded in New Zealand
since 1901. For example, an earthquake magnitude M = 7.5 and the shortest distance
between site and fault rupture R = 25 kIn gave a calculated mean PGA based on the new
Japanese attenuation as presented by Ref.F7 is approximately 0.315 g. Ref.X4 stated that
for class III buildings on rigid or intermediate soils in seismic zone A, the 1940 El Centro
N-S component can be used for the earthquake motions.
The analysis procedures for soil-structure interaction have been discussed. Both the
lumped parameter and finite element methods are useful tools for the response of soil-
structure interaction. Each method has its advantages and limitations. The primary
advantage of the lumped parameter method is its simplicityH2wI9. This method is used in this
study.
To simplify the analysis, the effect of the vertical earthquake excitation and coupling
between horizontal and rocking modes are neglected. The input of the soil-foundation
stiffnesses and damping coefficients on the structural system are represented in local
coordinates where the horizontal dynamic equilibrium of forces in the structural system is
fully satisfied. Since only horizontal earthquake excitation is considered, the discrepancy
206
between the local vertical spring stiffnesses and global rocking stiffness should not be
significantD9
The approach to the response analysis, structural system, damping type, damping
system, domain of analysis and solution of the governing equations of motion was
presented. The solution of the governing differential equations of motion depends on the
overall damping system of the structure. The numerical integration of the independent
modal equations of the linear-elastic structural system which proportional damping
numerical is fully justified. In the case of the structural system with non-proportional
damping, the direct numerical integration of the discrete coupled equations must be
used c22 ,TU. The soil-structure interaction analysis based on the frequency dependent soil-
foundation impedance with a constant impedance coefficient is used.
A number of the numerical integration schemes have been proposed by researchers.

The choice is based on its stability, accuracy and efficiency. It was found that the
Newmark-{3 with 'Y = 1/2 and {3 = 1/4 provided no amplitude decay and only a relatively
negligible period elongation. This integration scheme is used in this stUdy.
The choice of the earthquake input is one step in the earthquake resistant design
(EQ-RD). The design earthquake input based on the maximum acceleration is commonly
used even it gives a relatively poor prediction on the structural damage Pl8. Due to lack of
suitable data in New Zealand, the PGA attenuation derived from Japanese earthquake data
was used as suggested by Ref.D9. It was found that for M = 7.5 and R = 25 km the PGA
was approximately 0.315 g slightly smaller than the PGA of 1940 EI Centro N-S
component.
Chapter IX
Seismic Response Olf 12 and 1S-storey Frames and
Frame-Wall Structures subjected to the 1940 Centro NSC
INTRODUCTION
In the conventional earthquake resistant design of an elastic structure, the evaluation

of structural response quantities are commonly established in tenns of the maximum nodal
displacements and member forces M24 This is because these responses quantities are directly
related with the structural stiffness. The structural design criteria is commonly based on a
assumption that if the provided member strengths are greater than the corresponding seismic
forces, the survival of the structure can be expectedK18 . Elastic analysis, in general, cannot
predict accurately the post-elastic response quantitiesM20,M28,P3S.
Under a non-linear inelastic response, the structural displacement and force levels are
no longer directly related since the structural stiffness is amplitude dependent. The structural
response quantity is no longer based on a single parameter such as member force but more on
the defonnability of the members and structural systems. The response quantities of an inelastic
response then become more complexM23 ,M28,P35 and cannot be characterized satisfactorily by only
a single parametett23 These response quantities will be briefly discussed in the following
section.
9.2 THE RESPONSE QUANTITIES OF INELASTIC STRUCTURES
In earthquake resistant design of reinforced concrete structures, it is common to pennit

some degree of damage because of inelastic member defonnations, otherwise the design would
be too costlyA19,P27. In the structural analysis, these deformations are indicated by the response
quantities. The inelastic response quantities sometimes are termed as damage parametersP28 or
damage indicatorsM16 of the section, member or substructure. In general, the damage parameters
can be categorized into two groups; firstly, those based on a single maximum value such as
207
208
displacement, force, base shear, inter-storey drift and ductility factor and secondly, the
cumulative values such as a cumulative rotational deformations and dissipated energy.
9.2.1 Single Maximum Value of the Damage Parameters
9.2.1.1 Ductility Factor
In inelastic structural response subjected to an earthquake ground motion the

damage parameters are characterized by a maximum displacement and ductility
factor'1 16M23 ,M24,U1.Z3 defined as the ratio of the maximum to the yield deformations. For the
member, these factors include the maximum curvature, displacement, cyclic displacement and
accumulative ductility factor.
(9.1)
where i may be curvature, displacement, cyclic or accumulative ductility factor, Dm and Dy

correspond to the maximum and yield values respectively. The definitions of cyclic and
accumulative member displacement ductility are shown in Figs. 9.1.a and 9.1.b respectively.
The ductility factor describes the inelastic behaviour of the weakest parts of the
member where the damage is accumulated. For monotonic loading, the displacement ductility
factor "' .. ,m is a good parameter to describe the capacity of the member to deform inelastically,
but it is not able to describe the severity of inelastic deformation under cyclic loading
histories M24,M28. Later, it will be shown that the cyclic loading damage parameters such as the
number of yield excursions, yield reversals and dissipated energy provide more valuable
parameters. MAHIN & LIN (1983) and MESKOURIS & KRATZIG (1990) stated that to
establish the severity of damage to an inelastic structural member under cyclic deformation
histories, the concept of cyclic displacement and accumulative ductility factors can be used.
9.2.1.2 Member Forces, Nodal Displacements and Inter-storey-Drifts
The maximum member forces such as flexural strength, axial and shear forces
are commonly used to represent the damage parameters in the inelastic response analysisF8 The
209
stmctural displacement ductility factor 1'lJ.,8 can be used to predict the preliminary damage index
of the inelastic structural response as presented in Ref.P36. While CARR & TABUCHI (1993)
used the relationship between maximum top storey displacement and base shear in an elastic
and inelastic structural response to calculate an approximate value of the structural displacement
ductility factor 1'lJ.,8'
The inter-storey drift damage parameters become more important at the ground
floor level where the columns may be subjected to the maximum axial load. The column axial
load effect on the inter-storey drift is usually called the P-delta effect and was discussed by
CARR & MOSS (1980). The inter-storey drift limits have been clearly defined in some codes.
In NZS 4203: 1984 the inter-storey drift is based on the seismic regions and in NZS 4203: 1992
is based on the method of analyses and the response of the structures.
There are two basic concepts of the column design, firstly, based on the
member displacement ductility factor 1'lJ.,m and secondly based on the permissible inter-storey
drift. BERTERO (1988) presented a clear insight into the structural deformability, defining it
as the maximum displacement history without any significant reduction of strength,
displacement ductility and displacement ductility factor. A structure or member can have a
smaller ductility factor than another but its deformability can be larger. Because of this reason
and also by the fact that it is difficult to defme the yield displacement, PARK (1988) suggested
that the concept of using maximum inter-storey drift criteria has considerable merit.
9.2.2 The Accumulation of Member Deformations and Energy Dissipation
9.2.2.1 The Accumulation of Member Deformations
The damage parameters as represented by a single maximum value, as described

above, seem to be independent of time and cannot describe the cumulative damage effects under
cyclic loadings. Observation of the physical behaviour of the structural members in the field
and laboratory tests indicate that the cumulative damage parameters are better indicatorgMn1'36.
The number of yield excursions and yield reversals are damage parameters
suggested by ZAHRAH & HALL (1984). These parameters are defmed as the number of times
the member yields and the number of times where the member yields in opposite directions.
It was demonstrated in a SDOF structure with the same displacement ductility demand, the
210
1940 EI Centro N-S earthquake caused more damage because there was a higher number of
yield excursions and yield reversals than when subjected to the 1966 Parkfield N-E earthquake
excitation.
BANON et a!. (1981) showed that low-cycle fatigue damage was possible
under cyclic loadings such as earthquake excitation. Two damage parameters called Flexural
Damage Ratio FOR and Normalized Cumulative Rotation NCR were introduced. The FDR is
defined as the ratio of the initial flexural stiffness Kt to the secant stiffness Kr as shown in Fig.
9.4, and was developed from the damage ratio DR introduced by SOZEN (1977). NCR is
defined as the ratio of the sum of all plastic hinge rotations to the yield rotations. The FDR and
NCR are expressed as,
K
FDR =-L (9.2)
Kr
(9.3)
NCR
where j = 1,2 ... N are the number of members, 6mj and 6yj are the maximum and yield
member rotations at point j respectively
POWEL & ALLAHABADl (1988) stated that the damage parameter from this
cumulative member rotations has disadvantages because the damage caused by a large number
of small inelastic rotations is . likely to be less than that caused by a smaller number of large
inelastic rotations.
9.2.2.2 The Accumulation of Energy Dissipation
The damage parameters as previously described have advantages and

disadvantages. HOUSNER (1956,1975) introduced the velocity spectrum computed as the
square root of the energy dissipation and spectrum intensity defined as the area under computed
velocity spectrum. Recently, the damage parameter based on an amount of dissipated energy
has been commonly used and suggested by researchersDI2.D13,B24,F9,P36,UI.ZJ. The total energy input
on a structure subjected to an earthquake ground motion can be expressed as IJ24 ,TJ2.U1,
211
R
(RESISTANCE I R
(RESISTANCE I
__~-T~________~~__~~L-____~U .
(DISPLACEMENT 1
____~~______~L-__--~--L---~-u
(DISPLACEMENT!
ACCUMULATIVE DUCTILITY :E (uplll /u y + (

I 1+ 1+lu.I+ludl)
'--~-'-~..;...;~..:....;:..:..:. +I
Uy
CYCLIC DUCTILITY"
Fig. 9.1.a Cyclic DispL Ductility (M28) Fig: 9.1.b Accumulative Displ. Ductility (M28)
I 6 UI HI
(J"'11.
,. 3
M
TOTAL
DISSIPATED
EIRGV
~ ~ ro ~ ~ ~ ~ ~ ~ ~ 00
TlIoII:: (SECOHDSI
,
00
Fig. 9.2.<1 Secant and Initial Stiffness Fig.9.2.b Energy VI". Time and Yield Sequence (Z3)
212
E , =E/r;+EB +EV +Eh (9.4)
where subscript t, k, s, v and h refer to the total, kinetic, strain, viscous and hysteretic
respectively.
During the dynamic motion, part of the absorbed energy is temporarily stored
in the structure in the form of kinetic ~ and strain energy E._ The remaining energy is
dissipated by viscous damping Ev and irrecoverable hysteretic energy E h At the end of the
motions the amount of unrecoverable strain energy may exist caused by the permanent plastic
deformations. If this stored energy is relatively small, the energy input is almost equal to the
energy dissipation. At this state the viscous and hysteretic energy dissipation is very important
in the measurement of the damage parameter as represented in Fig. 9.2.b.
Using the same structure as previously mentioned, ZAHRAH & HALL (1984)
used the energy-based damage model in the SDOF structures. It was found that the 1940 EI
Centro N-S component caused more damage than that subjected to the Parkfield N65E ground
motion because the first ground motion causes more hysteretic energy demand than that by the
second ground motion. The dissipation energy is accumulated as a function of time. Since this
damage parameter is time dependent the duration of an earthquake input becomes important.
A similar damage model is also presented in Refs. B25, M23 and M32.
9.3 TIm MEMBER DAMAGE INDEX (DI)
All of the damage parameters as described above are represented in the quantitative
form. Recently, a qualitative interpretation of the structural response to severe earthquake
excitation led to the concept of damageability of the structuresM2J. The qualitative definition
of the damage itself is not clear or is difficult to formulateP36 To be able to define a qualitative
interpretation of the structural damage, the definition of structural damage then has to be
quantified in terms of the damage index Dr.
The parameters used in the calculation of the structural damage index DI result from
the structural analysis. Due to several uncertainties during the structural analysis, a rational
prediction of damageability of the structures is probabilistic and a complex problem. These
unceltainties include modelling of the structural system, the choice of earthquake ground
motions and identification of the types of the damage modes. There are three possible damage
213
modes; firstly, the damage mode may be subjected to a single yield excursion such as blast
load, secondly, a fatigue damage mode is subjected to cyclic loads and thirdly the damage
mode based on the combination between stress excursions and yield reversals. Several models
of damage indices have been proposed and some of them are described in the following
sections.
9.3.1 The Deformation Model
The damage index DI presented by POWEL & ALLAHABADl (1988) can be used to
establish the damage index of the structure subjected to a single damage parameter. This
parameter may be top floor displacement or total base shear. It is necessary to identify three
components namely threshold, calculated and ultimate values. The yield state, either yield base
shear or yield displacement, is generally used to represent the threshold values. However, the
estimation of the yield and ultimate value is not straightforward, because these values are
approximate as discussed in Ref.P36. The proposed damage index Dr is expressed as,
(9.5)
where oe, 0t, Ou are the calculated, threshold and ultimate values respectively and m is a
coefficient. This damage index is similar as proposed by ROUFAIEL & MEYER (1987)
9.3.2 The Low-Cycle Fatigue Model
This damage model is calculated by an assumption that the damage index increases with
the number of cyclic loadings. There are two possible damage modelsM23 ; firstly, high-cycle
fatigue, where the damage is caused by thousands or millions of elastic cyclic loadings and
second, low-cycle fatigue results from a number of cycles approaching or exceeding the yield
stress. McCABE & HALL (1989) proposed the damage index Dl based on the low-cycle
fatigue test of metals. The damage index is the ratio of the actual energy to the theoretical total
hysteretic energy Et corresponding to complete loading reversals. Thus the damage index is
based on normalized energy expressed as,
(9.6)
214
where Ep and En are the dissipation energy under positive and negative forces respectively as
the portion of the hysteresis loop. A similar damage index according to energy-based concept
is presented by MEYER (1992).
A damage index based on low-cycles fatigue metal tests was also presented by
STEPHENS & YAO (1987). This concept was based on the cumulative inelastic deformations
and may be valid for a variety of structural systems with different materials and loading
histories. The damage is evaluated for every loading cycle and the total damage associated with
n-cyc1es of response is given by,
(9.7)
where AO!,l and AO pf are the positive change in plastic deformation and the positive change in
plastic deformation at failure, a is the coefficient computed based on Ref.S30.
9.3.3 The Combination Model
Another damage mode is caused by the combination between high stress excursions and
repeated stress reversals. PARK & ANG (1985) developed the damage concept of the
reinforced concrete member as a function of maximum deformation and absorbed hysteresis
energy. The damage index DJ is calculated as the sum of the normalized structural deformation
and absorbed energy as expressed in,
(9.8)
where Om' Ou are the maximum deformation during seismic loading and ultimate deformation
capacity under monotonic loadings respectively, fl is the strength deterioration parameter, Py
is the yie1d shear strength and dE is an incrementa1 absorbed hysteretic energy.
The values of Om and dE are the response quantities dependent on the loading history,
while fl, Ou and Py are specified by structural capacity. The value of {1 is slightly affected by
shear span ratio, axial stress and longitudinal steel ratio. The approximate formula for this
value can be found in Ref.K1 3, P27 and P29. Similar member damage index based on sum of
the normalized member deformation and hysteretic energy dissipation is discussed by
COSENZA et al. (1993).
215
9,4 THE STOREY AND STRUCTURAL DMvlAGE INDICES
The damage indices, as described above, correspond to damage quantities at member

ends under flexural deformations. In a particular storey the highest damage index will occur
at the weakest member-end while for the whole stmcture it will occur in the weakest storey.
Determination of the storey and whole stmcture damage indices based in the maximum member
and storey damage index are not reasonable. The storey and the stmctural damage indices,
therefore, should be based on the distribution of the damage throughout the storey and the
building respectively. The storey and the structural damage indices can be calculated by using
weighting factors such as,
N
D :::; E { 1. D.} (9.9)
J~1 J J
where D is the storey or structural damage indices, Dj is the damage index of a member j or
storey j respectively, Aj is the weighting factor of the member j or storey.i respectively and N
is the number of member elements or the number of storeys respectively.
WEN et a1.(1988) and KUNNATH et a1.(1990) reported that the weighting factor ~
depends on the failure mode mechanism of the stmctures. For weak-column and strong-beam
failure mode mechanisms, the damage of the structure is closely related to the absorbed energy
and the weighting factor can be based on the energy ratio as,
E
1'1 :::; -N' - (9.10)
EE.
J=1 J
where Ej is the total absorbed energy of member j or storey j respectively.
In the case of the strong-column and weak-beam failure mode mechanism, it is

necessary to extend the foregoing concept. For strong-column and weak-beam type buildings
the member damage is strongly correlated with the member curvature demand, as will be
discussed in Chapter XII. The damage parameters and damage indices analysis of inelastic
building structures can be summarized in the systematic form as shown in Fig. 9.3.
Damage Properties Parameters Procedures Analyses
~ Member Forces
Base Shear Single

Parameter
Parameters Roof Hor. Displ. Model
~
Maximum Inter-storey Drifts
Single DEMAND
Cyclic/Ace. Ductility
Value vs
Storey Ductility
Dimensionless CAPACITY
Sum of Displ. Duct.
PROCEDURE
Displ./Curv.Ductility
Stress
Combina- Reversal DAMAGE
tion INDICES ""....
0')
Dimensional Hyst.Energy Dissip.
Parameters
Norm.Cumulative
~
Rotation (NCR)
Cumulative
d Revers.
Yield Excurs.
Vaiue
Dimensionless Cycies
DEGRADATION
Parameters Fatigue
PROCEDURE Model
Flex .. Damage Ratio

(FOR), Stiff. Oegr. I
Fill. 9.3 FlowoChart of the Peuameter lind Damage Index Analysis of the Building Structure.:;
217
9.5. THE SEISMIC RESPONSE OF FRAME AND WALL-FRAME STRUCTURES

WITH FrKED-JBASE AND FLEXIBLE FOUNDATIONS
In this case, the responses of two types of building with different types of foundations
are compared. In the first building, the foundation footings are assumed to be fixed at the
footing levels and in the second building, the foundation footings are supported by flexible soil-
foundation interactions. The foundation types are called fixed-base (LFix) and flexible
foundations (LFlex) respectively. In this case, the building prototypes were subjected to the
North-South component of the May 1940 Imperial Valley earthquake recorded at EI Centro.
This is because this earthquake has been used as the basis for the design response spectra in
many countries including New ZealandT2l The structural displacements and strains are
assumed to be relatively small compared with the dimension of the structures. These
assumptions imply that the stiffness of the structure is based on the initial geometry and
remains unchanged during loading history.
The building prototypes are assumed to have a strong column and weak beam failure
mode mechanism, where energy is absorbed. on the formation of plastic hinges at the beam ends
and column and wall bases. The frame-wall structures are analyzed as a series of 7-frames and
2-walls. P-..1. effects are not considered in this analysis. In this case, the soil shear modulus G
= 685 kg/cm2 corresponds to a soil shear wave velocity Vo = 190 mlsec is used. Other
parameters used in calculations of the soil-foundation impedances are presented in Section 5.9.
Based on the possible soil shear strain ),,, induced by earthquakes as discussed in Section 4.3,
the soil-foundation impedances are based on the soil-shear modulus G = 0.70 Gmax A simple
computer program SPRING was written to calculate the stiffnesses and damping coefficients
of partially embedded foundation as discussed in Section 7.5. The soil-foundation impedance
were based on the soil-mechanical model as discussed in Section 5.6. Effects of the material
damping to foundation impedance according to Eqs. 7.98 and 7.99 were adopted.
9.5.1 Soil Shear Stnlins
As discussed in Chapters IV, V and VII that the role of soil shear strain on the soil-
foundation is very important. The approximation of soil shear strain 1' on the lumped mass
idealization of horizontal soil layer over horizontal rigid boundary was discussed by Ref.D6.
The layer soil shear strain can be calculated using the inter-layer horizontal displacement and
the corresponding inter-layer thickness.
0.0001 0.0002
~
: 0420. F12RZ 10420, FW12RZ
0.00015
:Takeda. EeNS :T5keda, EcNSC i
0.00005 0.0001
~
'" '"
E 0.00005
III iii ~
~ ~
"" '""
0 0
~ I i'
.r:. .r:.
~~
III 4 III 4
"0 0 -0.00005
<II
III
.0.00005 -0.0001
-0.00015
-0.0001
\ -0.0002
Fig. 9.4 Soil Shear Strain of the 12.-atorey frame Fig. 9.S Soil Shear Strain of the 12-atofey Frame-Walls
0.0001 0.0002
i 0635,F1SRZ 0.00015
'Takeda. feNS
0.00005 0.0001
E'" '"
E 0.00006
Vi iii
~ ~
., 0
"'"
0
.&;
VI
"III
.r:.
:; 0 .0.00005
III en
0.00005 -0.0001
-0.00015
-0.0001 ~----------------------------------------------------~ -0.0002
Fig. 9.6 Soil Shear Strain of the 18-1'1torey Frame Fig. 9.7 Soil Shear Strain of the 1B-storey Frame-WEill'll
219
In the case of the fictitious layered soil over an elastic half space soil model used in this
study, the problem is to determine the location of soil with zero stress under dynamic loading.
Under static loading, HADJIAN & ELLISON (1985) and SRIDARAN et al.(1990) showed that
the stress distributions with soil depth depends on the loading modes. In section 5.3 the depth
of soil to zero stress is termed the effective depth of influence. For horizontal loading the
effective depth of influence is approximated by 2B where B is the footing width perpendicular
to the horizontal loading. PENDER (1983) stated that under dynamic loading, the zone of soil
involved in non-linear strains is about one footing diameter. Thus, this effective depth under
dynamic loadings is not clear yet.
If the effective depth of influence of the soil under dynamic loading is assumed to be
same as under static loading and the soil stress distribution is linear with depth, the possible
soil-shear strain history of the 12-storey frames and 12-storey frame-walls are shown in Figs.
9.4 and 9.5 respectively. The maximum soil shear strains are 0.95xlO-4 and 1.74xlO-4 for
frames and frame-walls respectively. These soil shear strains are comparable with the possible
soil shear strains 'Yss induced by earthquakes as given in Table 4.5. In relationship with the soil
shear modulus degradation curve for Auckland clay as shown in Fig. 4.13, these soil shear
stains correspond to G = 0.88 Gmax and G = 0.72 G max respectively.
The soil shear strain 'Ys. of the 18-storey frame and frame-wall structures are shown in
Figs. 9.6 and 9.7 respectively. The maximum soil shear strains are 0.48x1Q-4 and 1.43xlQ-4
for frames and frame-walls respectively. Compared with the soil shear strains above, these
results are smaller. The fundamental period of the structures seems to affect the soil shear
= 0.92 Gmax and G = 0.75 Gmax respectively.
strains. These soil shear strains correspond to G
Thus, the soil-foundation impedances based on G = 0.70 Gmax as mentioned in Section 9.5 are
still reasonable.
9.5.2 Base Node Rotations and Vertical Displacement of the Footing Tips
The important response parameters of the footing foundations are the base-node vertical
displacements and rotations. Given the footing size and knowing the base node vertical
displacements, the vertical displacements of the outer-tip of the foundation footing can be
calculated. These results are very important in identifying whether or not the entire width of
the footing foundation is in full contact with the supporting soil. The lift-off of the footing
foundation significantly effects soil-foundation impedances.
0.0005 0
]
~
I'I .
0.0004 I
'"
1= .0.005
""
-:g"
-d -0.01
0.0002 .....<:>
~
c: ..
:is
~
:;
a:"
0 ..
Q -0.015
c:
!:> ~
"
Z -0.0002 T "" ..c.n
Q
11.. "
7i
, "
Q
-O.0004.l. 'i -0.025
'E'"
>"
..0.0006 -0.03
Fig. 9.S Footing Rotation of the 12-15torey Frame Fig. 9.9 Vertical l)ispJl:io",ment 01 thill 1Z..,.torey Frillme's Footing Tip\!
0.0003 0
~
2. B 10 12 14 16 18
!
O.OOOZ { ...'"
j::: ..0.005
..,
E 0.0001
t .,
c:II
I:
<:>
c: ....." -0.01
~
'0" 0
..
'0
c:
a:
<:>
S.. 0.015
c:
~ -0.0001 '.,"
7i
....<:> Q"
Ci ..0.02
c
-0.0002 1:
>" Long OelShed Un~
-0.0003 0.025
Fig.9.10 Footing Rotation of the 1S-etofe!Y Frame! Fig.9.11 VerticlIl Displecsmenl of ~he 1S.storey'Frame'C! Footing Tips
o.oos 0
2 4 8
G420, FW12RZ
T15kedll, EcNSC
0.004
-().O1
";;
!
"'"
'0 0.002 ~
o
E-
.,'" ..IE -0.02
';;
<'=
a:"
0
.is''""
11.
,
I
-0.03
'Ii.,
c
~
co
-0.002 t: \(V
"
LI.. >" V
-0.04- Une
-0.004
-0.006 -0.05
Fig. 9.12 Footing Rotation of the: 12-etofey Fram<!!-waIl5 Fig. 9.13 Ven. Displacement of thG 'I2.-etO(6,( Frllme-wal! Footing Tips
0.006 0.006
0
Q.004 ,\
Walle :z
:~ 14 16 is
.. r, n
'I
!
-0.005
'5
E.
~
"
0.002
01
t
f
II
I'
,
,,
I'
II I'
I'
I Ii
II
I'
"
"
,
I
'C
E
"
"'" .().O15
1i.
-0.01
~ ;5
~
c::"
;; -0.02
'"
tt
:= u II
';;;
0
"
"-
-O.OOl
.
t:
> -0.025 1
'\ I
I,
II
"II
-0.004 T -0.03
\, 106815, FW1SRZ
y....-
II
lOllg Dashed Lllle ,Takedei, EcNSC
-0.035
Fig. 9.14 Footing Rotation of the K...,'t ...... " Frame-walla Fig. 9.15 Vert. Displacement of the
222
The base node rotations and the footing-tip vertical displacement histories of the 12 and
1S-storey frames are shown in Fig.9.S, 9.9.9.10 and 9.11 respectively. As depicted in Fig.9.8
and 9.10, the left, right and internal node rotations are presented in solid, dashed and dash-
dotted lines respectively. Their node rotations are not significantly different. At the beginning
of the rotation history, there is initial rotation of the left and right column footing due to
differences of initial vertical displacement as shown in Figs. 9.9 and 9.11 respectively. In these
figures, the solid and dashed lines indicate the veltical displacement of the outer-tips of the left
and right column footings, the long dash and long dash-dotted lines represent the vertical
displacement oUhe left and right-tips of the internal footing respectively. The zero value oUhe
y-axis represents the unloaded soil position. It is clear that the footing-tip vertical displacements
for both buildings are far from the lift-off limit. This means that during the loading history the
footing foundations are still in full contact with the supporting soil.
Base node rotations for the 12 and IS-storey frame-wall buildings are shown in
Fig.9.12 and 9.14 respectively. The solid and long-dashed lines indicate the rotation of the left
column and the wall footing node respectively. The rotation of the wall footing nodes are 8
times greater than that of column footing nodes. This result implies that it is difficult to stop
the stmctural wall rotating. Rotation of the stmctural wall has a significant effect on structural
responses as discussed in next sections. Foundation tip vertical displacement of the 12 and 18-
storey wall-frames are shown in Fig.9.13 and 9.15 respectively. The solid and dashed lines
represent the outer tip of the left and right column footings, while the long-dashed and long-
dash-dotted lines indicate the vertical displacement of the left and right tips of the wall footings
respectively. The outer tip vertical displacement of the wall footing of the 12-storey wall-frame
is in full contact with the supporting soil during the response history. A very short duration and
small footing lift-off occurred in the 1S-storey frame-walls, as shown in Fig.9.15. This is
caused by the relatively high rotation of the wall footing node. The effects of this lift-off on
the stmctural responses are ignored.
9,5.3 Fundamental Period of Structures
In structural analysis the fundamental period Tl is commonly used to describe the

stiffness of the structure and it is also useful for preliminary analysis. In the equivalent static
analysis specified by Codes, the base shear force is computed from the structural fundamental
period, therefore the magnitude of the design force input depends upon this period. In the
223
elastic dynamic analysis using the response spectrum approach, the response contribution of
every mode will depend upon its own period. The response of the structure is largely dependent
on the contribution of the first few modes. Studies to determine the approximate fundamental
period for Moment Resisting Frame and Frame-Wall Structures subjected to several
earthquakes has been reported by Ref.B20, and B26.
The fundamental period of the 12 and IS-storey buildings with fixed bases and flexible
foundations are shown in Tables 9.1 and 9.2 respectively. The fundamental period of the 12
and 18 storey frames with flexible foundations are 4.9 and 5.7 % longer than those for fixed
base foundations. In terms of number of stories N, these periods correspond to 0.23 Nand 0.19
N respectively. Ref. B13 reported that the fundamental period for 4 and 8-storey inelastic MRF
of 0.22 Nand 0.25 N respectively were found.
Table 9.1 Response Quantities of the 12-Storey Frame
Structural Systems
No. Pammeters Remark
Fixed Base Flex.. Found.
1. Fundamental Period
T. (sec). 2.754 2.886 (+)
2. Roof Displacement (m)
a. Positive Envelope 0.1873 (0.415 % Ho) 0.1900 (0.421 % Ho) (+)
h. negative Envelope 0.2997 (0.664 % Ho) 0.3009 (0.666 % Ho) (+)
3. Base Shear (Tonne)
a. Positive Envelope 104.00 102.50 (-)
h. Negative Envelope 94.80 94.20 (-)
4. Col. Base Curv. Ductility
a. Left Column + 1.364, -0.914 +1.233, -0.824 (-)
b. Internal Column + 1. 723, -0.954 + 1.270, -0.921 (-)
c. Right Column +1.929, -0.937 + 1.017, -0.884 (-)
5. Col.Base Rot.(x 0.001 Rad.)
a. Left Column + 1.639, -0.000 +1.481, -0.000 (-)
h. Internal Column +2.240, -0.000 +1.769, -0.000 (-)
c. Right Column +1.900, -0.000 +1.221, -0.000 (-)
6. Total P.H.Rol.(x 0.001 Rad.)
a. Positive Envelope 188.244 181.722 (-)
h. Negative Envelope 159.701 160.099 ( +)
c. Total Rotations 347.945 341.821 (-)
7. Energy Dissipation (Tm.)
a. Viscous (V.) 21.800 21.800 (=)
h. Hysteresis (H.) 21.300 19.200 (-)
c. V. + He 43.100 41.000 (-)
Hb : Total Height, (+) : Increase, (-) : Decrease, (=) : Equal, (.) : Not Clear
The fundamental period for the 12 and 18-storey fixed frame-wall structures as given
in Tables 9.3 and 9.4 and correspond to 0.19 Nand 0.17 N respectively. These periods are
smaller than those for the MRF because the frame-wall structures are stiffer. These
224
fundamental periods were based on a rigid zone length Z = 0.50, meaning that the length of
the rigid zones of columns and beams are 114 of the beams and column's depth respectively.
Chapter X describes the effects of the rigid zone length on the structural responses.
In the case of flexible foundations, the fundamental periods correspond to 0.24 Nand
0.20 N for the 12 and 18-storey MRF. In comparison with the fixed base structures, these
fundamental periods are 19.9 % and 18.60 % longer. These results imply that due to the
foundation flexibility, the frame-wall structures decrease in stiffness much more than do the
frame structures. Increasing the fundamental periods due to foundation flexibility is also
reported in Ref.C3, C4, E2 and P6. The change in fundamental period due to different
foundation system affects the structural responses especially for frame-wall structures.
9.5.4 The Damage Parameters
9.5.4.1 Maximum Roof Horizontal Displacement
The member response quantities from the inelastic analysis are the stresses,
strains, forces, deformations, degradations, ductility and work done. The structural responses
may be the inter-storey drift, storey shear, storey ductility, base shear and nodal responses. The
nodal responses consist of the nodal rotation and the nodal displacements in the x, y and z-
direction respectively. The roof horizontal displacement is often used as a damage parameter.
If the demand and capacity procedure as shown in Fig.9.3 is used, Ref. P36 reported that-the
roof horizontal displacement can be used to provide damage index for the structure.
Using elasto-plastic hysteresis behaviour, GUANGQIAN et al.(1984) reported

that the maximum roof horizontal displacement in an inelastic response analysis may be taken
as 0.9 times the sum of the maximum inter-storey drifts. The maximum displacement of the
12-storey frames with fixed and flexible foundations are 0.86 and 0.82 times the sum of the
inter-storey drifts respectively. These results are relatively close to the above approximation.
However, this approximation does not agree well with the results from the I8-storey frames.
The roof horizontal displacement of the I8-storey frames with fixed and flexible
foundations as given in Table 9.2 correspond to 0.74 and 0.68 of the sum of storey drifts. As
shown in Table 9.1 the roof horizontal displacement of the 12-storey frames is even larger than
225
that for the l8-storey as given in Table 9.2. This depends on the structural fundamental period
with respect to the peak spectra of the input motion. The roof displacement to building height
ratios of the 12-storey with fixed bases and flexible foundations are 0.664 % and 0.666 % Hb
and only 0.41 % and 0.395 Hb for the IS-storey frames respectively. This depend on the
natural period with respect to the peak in the response spectra.
Table 9.2 Response Quantities of the lS-Storey Frame
Structural Systems
No. Parameters Flex. Found. Remark
Pixed Base
T, (sec). 3.338 3.527 (+)
a. Positive Envelope 0.2229 (0.332 % Hu 0.2289 (0.341 % HJ (+)
b. negative Envelope 0.2752 (0.410 % Rt, 0.2648 (0.395 % Hi,) (-)
n. Positive Envelope 97.40 96.50 (-)
b. Negative Envelope 128.30 122.80 (-)
I!. Left Column +0.000, -0.000 +0.000, ..{).ooo (=)
h. Internal Column + 1.416, -0.753 +1.065, -0.763 (-)
c. Right Column +0.000, ..{).ooo +0.000, -0.000 (=)
5. CoI.Base Rot.(x 0.001 Rad.)
a. Left Column +0.000, -0.000 +0.000, -0.000 (.)
b. Internal Column + 1.481, -0.000 +1.114, -0.000 (-)
c. Right Column +0.000, -0.000 +0.000, -0.000 (.)
6. Total P.H.Rot.(x 0.001 Rad.)
ft. Positive Envelope 161.580 146.101 (-)
b. Negative Envelope 197.030 176.970 (-)
a. Viscous (VJ 26.320 25.180 (-)
b. Hysteresis (HJ 21.800 19.360 (-)
c. V. + He 48.120 44.340 (-)
: Total Building (+) : Increase, (-) : Decrease, (==) : (.) : Not Clear
The maximum displacement for fixed bases and flexible foundations for the 12-
storey frame-walls are 0.88 and 0.84 of the sum of the storey drifts respectively. For the 18-
storey frame-walls, the maximum displacements are 0.86 and 0.68 of the sum of the storey
drifts for the same types of foundations. The relationship between displacement to total drift
ratio versus fundamental period is shown in Fig. 9.39 which shows that this relationship is not
constant as reported by Ref.G20 but has a negative slope. The variation of roof displacement
to Hb ratios versus fundamental periods as shown in Fig.9.40 is not clear.
226
Table 9.3 Response Quantities of the 12-Storey Frame-Wall Structures
Structural Systems
No. Parameters Remark
Fixed Base Flex. Found.
1. Fund. Period T\ (see). 2.312 2.771 (+)
a. PosiLive Envelope 0.2962 (0.656 % Ht. 0.2691 (0..596 % Ht. (-)
b. negative Envelope 0.2683 (0.594 % H!.) 0.3138 (0.695 % 1\ (+)
1. Frames
11. Positive Envelope 809.00 (42.94 %) 912.00 (60.80 %) (+)
b. Negative Envelope 780.00 (36.90 %) 926.00 (50.80 %) (+)
2.Walls
a. Positive Envelope 1075.00 (57.06 %) 588.00 (39.20 %) (-)
b. Negative Envelope 1334.00 (63.10 %) 897.00 (49.20 %) (-)
3.Total Base Shear
a. Left Column +1.046, -0.870 +2.166, -0.937 (+)
b. Internal Column + 1.430, -1.439 +2.219, -1.841 (+)
c. Right Column +0.939, -1.261 + 1.349, -1.769 (+)
d. Wall +3.918, -3.971 +0.000, -0.000 (-)
5. Col.Base Rot.(x 0.001 Rad.)
a. Left Column +1.378, -0.000 +2.855, -0.000 (+)
b. Internal Column +1.895, -1.907 +2.940, -2.439 (+)
c. Right Column +0.000, -1.675 + 1. 777, -2.332 (+)
d. Wall +4.772, -4.837 +0.000, -0.000 (-)
6. Total P.H.Rol.(x 0.001 Rad.)
a. Positive Envelope 184.790 147.640 (- )
b. NegaLive Envelope 152.180 179.120 (+)
a. Viscous (Y.) 306.000 355.000 (+)
b. Hysteresis (HJ 306.000 172.000 (-)
c. V. + H. 612.000 527.000 (-)
Hb : Total Building Height, (+) : Increase, (-) : Deerease, (=) : Equal, (.) : Not Clear
9.5.4.2 Base Shear and Storey Shear Ratios
In both elastic and inelastic analyses, the total base shear is one of the most
important response design quantities. In most loading codes, the base shear force is then
distributed to evelY floor over the building height. POWELL & ALLAHABADI (1988)
reported that the base shear can be utilized to predict the structural damage index even-though
estimates of yield and ultimate base shear are only approximate. CARR & TABUCHI (1993)
used the elastic and elastic base shear to estimate the structural yield displacement.
227
The base shear of the 12 and IS-storey frames with fixed and flexible
foundations are shown in Tables. 9.1 and 9.2 respectively. The base shear of the 12 and 18-
storey buildings with flexible foundations is only 0.95 % and 2.65 % smaller than that for fixed
base foundations respectively. These results indicate that effects of the soil-foundation
interaction on the structural response are less significant for the flexible structures. The storey
shear ratios of the 12 and is-storey frames are depicted in Fig.9.16 and 9.24 respectively. In
the lower storeys, the storey shears for flexible foundations are less than those for fixed base
foundations. The effect of the higher modes are clearly shown in the upper storeys.
Table 9.4 Response Quantities of the i8-Storey Frame-Wall Structures
Structural Systems
No. Parameters Remark
Fixed Base Flex. Found.
T. (sec). 3.041 3.607 (+)
2. Roof Displaeement (m)
a. Positive Envelope 0.2379 (0.354 % lIt, 0.2945 (0.439 % 1\) (+)
b. negative Envelope 0.3067 (0.457 % 1\ 0.2987 (0.445 % 1\) (-)
1. Frames
a. Positive Envelope 670.00 (29.32 %) 1257.00 (49.16 %) (+)
b. Negative Envelope 838.00 (26.49 %) 1301.00 (49.59 %) (+)
2.Walls
3.Total Base Shear
a. Positive Envelope 2288.00 (100.0 %) 2557.00 (100.0 %) (+)
4. CoL Base Curv. Ductility
a. Left Column +0.000, -0.000 + 1.449, -1.044 (+)
c. Right Column +0.000, -0.000 +1.112, -1.396 (+)
d. Wall +1.954, -2.248 +0.000, -0.000 (-)
5. CoLBase Rot.(x 0.001 Rad.)
a. Left Column +0.000, -0.000 +2.060, -1.484 (+)
c. Right Column +0.000, -0.000 + 1.582, -1.985 (+)
d. Wall +2.870, -3.030 +0.000, -0.000 (-)
6. Total P.H.Rot.(x 0.001 Rad.)
a. Positive Envelope 126.080 163.700 (+)
b. Negative Envelope 135.160 168.690 (+)
c. Total Rotations 261.240 332.390 (+)
a. Viscous (V J 304.500 385.000 (+)
b. Hysteresis (HJ 223.900 138.000 (-)
c. Ve + He 528.400 523.000 (-)
Remark: (+) : Increase, (-) : Decrease, (=) : Equal, (.) : Not Clear.
228
12
Storey
8+.. - .... ......+.. - . .... ~.... i.. ....-......t ....~~ ........ 1
Storey
O~~~~~~~~~~~~~~~~
0.9 0.95 1.05 1.1 0.85 0.9 0.95 1.05

Shall. R<ltlo (Flex.to FIlled! Mornet Ratio (Flell. to Flxedl
Fig. 9.16 Storey Shear Ratios Fig. 9.17 bam Moment Ratios
10 10
Storey
B B
Square : St. Drit1

Circle : St.P.H.Rot.
6
lied Fixed
lalank Flex.
4 4
2 ............................... ~.-............-.......~. 2 ............. ....-..,-..-....,~.... ;...-.... ~..--..i..-....c>~ ..-I
0.8 0.9 1.1 1.2

Column Moment Retio (FleK. to Fix.1 Drift Ix 0.1 % HI & P.H.Rot.(K 0.001 Rnd.1
Fig. 9.18 Column Moment Ratios Fig. 9.19 inter-storey Drifts &: Storey P. Hinge Rot.
229
2.5
10 +, ....1.... ,.., .. ,+...... ..,,, ..,...................., ............ j

2
"0
10
tc
B
....
0
J:
0 1.5
~ ci
.!$
ci It)
c
.!$ .~
:l.: 6
to
+'
0 0
tc
.;
I/iI :E'"
I/iI Fix. 0:
4
Square: Fixed
o Flex. 0.5
Circle Flex.
Filled Pos.Env.
Blank Neg.Env.
o 5 10 15 20 25 o T-------~------~J-------Lr------~
Storey P.H.Rot. Ix 0.001 Rad.l Left.Col. Int. Col. Right Col.
Fig. 9.20 Interstorey Drift vs. Storey P.H.Rotations Fig. 9.21 Column's Base P.Hinge Rotations
12 12
Storey Store Square: St.P.H.Rot.

Circle : Cum.P.H.Rot.
10 10
Filled Fixed
Blank Flex.
B B
4+,,.. .... .......... i ...... .. ........ ...... C~ .., ..........~....+, .............,....... j
2+.... .. ........ ....,.... .. , ........ , .... i......0 ...... ,.......+............... ,.........
2 4 6 B 10 o 50 100 150 200

Curvature Ductility Demands PI. Hinge Rotations (x 0.001 Rad.l
Fig. 9.22 Beam Curvature Ductility Demands Fig. 9.23 Accumulation of P. Hinge Rotations
230
The base shear of the 12 and 18-storey frame-wall structures are presented in
Table 9.3 and 9.4 respectively. Base shear of for the flexible foundations are 17.0 % and 17.1
% smaller when compared to the 12 and 18 storey frame-walls with fIxed base foundations
respectively. The wall base shear reduces when the footing foundations rotate. In contrast, the
base shears resisted by the frames with flexible foundations increase by 15.7 % and 55.2 %
for the 12 and IS-storey respectively when compared with the frames with fixed bases. When
the structural walls rotate part ofthe base shear is redistributed to the frames. In Figs. 9.33 and
9.42, the frame and wall storey shears of the 12 and 18 storey frame-wall structures are
normalized with respect to the ftxed base foundations. The wall storey shear ratios of the 12
and 18 storey frame-walls are less than unity in most stories with the smallest ratios being
0.547 and 0.570 respectively at the first floor. On the other hand, the frame storey shear ratios
in most stories are greater than unity with the highest ratio of 1.19 and 1.87 respectively at the
first floor. The ability of the wall to resist the shear force decreases as the wall bases rotate.
The shear force is then redistributed to the frames resulting in a greater shear in the frames.
9.5.4.3 Beams, Column Moment Ratios and Wall Moments
As shown in Figs. 9. 17 and 9.25, the beam moments of the 12 and 18 storey
frames with flexible foundations are normalized with respect to the ftxed base foundation and
are termed as beam moment ratios. The outer ends of the beams of the buildings with flexible
foundations seem to suffer more than that of the internal ends although the moments are only
increased by a maximum of 5 %. Similar normalization of the left and internal column
moments of the 12 and IS-storey frames are shown in Fig. 9.18 and 9.26 respectively. Fig.
9.18 shows that the moment ratios are not consistent. Due to the foundation rotation, the
column moments in the lower levels of the 18-storey frame are redistributed to the middle
level. As a result, the column moment ratios at the middle level are greater than unity as shown
in Fig.9.26. This is evidence that the redistribution of forces due to the foundation flexibility
are more significant in relatively flexible structures.
The column moment ratios for the 12 and 18 storey frame-wall structures are
depicted in Fig.9.34 and 9.43 respectively. In approximately one-half of the upper part of the
building, the column moment ratios are greater than unity. At the first storey, these ratios are
1.16 and 2.07 for the 12 and 18-storey respectively. The wall moments for the 12 and 18-
231
18 18
16 16
14 14
Storey Storey
12 12
10 10
B Il
: Left Bay
4 : Right Bay
2 2
0 O~~~~-+~~~~+-~~~~~~~~
O.B 0.9 1.1 1.2 0.95 0.975 1.025 1.05

Storey Shear Ratio IFlex. to Fix.1 Moment Ratio (Flex to Fh'edl
Fig. 9.24 Storey Shear Ratios Fig. 9.25 Beam Moment Ratios
18
16
Storey
10 -.-~~~~~~ ,~~ 10~"",,,, ....,,+ ...~..,, .... ..t.......~....... ~:
8 ......"......"....,,......,,..11::J,,..,,..............01.... .-~-- .... ""......1

Square
Circle
6+..-~r ....,,+.. .. ~~~.--+-,, ~I 6
Filled : Fixed
Blank : Fie)!.
0.ll5 0.9 0.95 1.05 1.1 o 5 10 15

Moment Ratio (flex.to Flx.l Drift (xO.1 % H), P.H. Rot.lx 0.001 Rad.1
Fig. 9.26 Column Moment Ratios Fig. 9.21 Inter-storey Drifts & Storey P.Hinga Rot.
232
2,,5
Square : Fixed
Circle : Flax,
Filled : POS.fRV.
: Neg.fRv.
2
'ti
:t
..'"
a:::
0
0 1.15 -
d
*0
.... \!!l @
!:!
.,c
~ 0 0 .,8
.t:
'1: \iii @
...
OJ
.,"
0 IX:
4 at
C
:f
e i'i
Fix.
0.5
Flex.
2+-~~~+-~~~+-~~~+-~~~
0.00 5.00 10.00 15.00 20.00 o +-----~r-----_+------~----~

Storey PI. Hinge Rotations left Col. Int. Col. Right Col.
Fig. 9,28 Interstorey Drifts vs, Storey P. Hinge Rot Fig. 9.29 Column's Base PI. Hinge Rotations
1B~-------.----D-~D-~-----.-------,
: Cum.P.H.Rot.
16 : Storey Drift
Square: LO,End 2( + I Fixed

14 14 Flex.
... Clrele : IUI.End 1 H
: Fixed
12 : Flex. 12
10
o 2 6 B o 50 100 150 200

Curvature Ductility Demands PI. Hinge Rotations {x 0.001 Rad.'
Fig. 9.30 Beam Curvature Ductility Demands Fig. 9.31 Accumulation of Plastic Hinge Rotations
233
storey buildings are shown in Fig.9.35 and 9.44 respectively. The wall-base moment reduces
when the wall base is allowed to rotate because of the soil-foundation flexibility. As shown in
these figures. the redistribution of wall moments from the base to the upper storeys are more
significant in the IS-storey wall compared to the 12-storey wall. The wall moment ratios at the
first storey are 0.70 and 0.65 for the 12 and 18-storey walls respectively. These results are
similar to the redistribution of the column moments in the l8-storey frames, as mentioned
earlier.
From the column moment and storey shear ratios discussed above, it is clear
that the columns in the first storey of frame-wall structures with flexible foundations suffer
more than those of fixed base foundations. The columns of the first storey of the 18-storey
frame-wall structures with flexible foundations resist 87 and 107 % greater shear and moment
when compared with the fixed base foundations respectively. This is because of the
redistribution of shear and moment due to the soil-foundation flexibility. Further effects of the
wall rotation will be discussed in the following sections.
9.5.4.4 Inter-storey Drifts
The effect of gravity load induced axial load in the column on the storey drifts
will produce additional forces, moments and reduce the effective stiffnessI>U9T21. These
additional effects are commonly referred to as P-.1. effects, where P refers to the gravity
loading and ;:,. the inter-storey drifts. When the building responds elastically, the inter-storey
drifts are relatively small and these effects are commonly negligibleC1J ,Plll.M21. When the building
responds in a highly inelastic manner with a relatively large column axial load, these effects
cause a significant increase in the plastic deformationTIl. To minimize these effects, the inter-
storey drifts under seismic loading in many design codes must not to exceed specified limits.
Inter-storey drifts and storey plastic hinge rotations of the 12 and 18 storey
frames are shown in Fig.9.19 and 9.27 respectively. Inter-storey drifts of the flexible
foundation structures tend to be greater than that for fIxed base structures in almost all of floor
levels and especially so for the 18-storey frames. The maximum inter-storey drifts for the fixed
and flexible foundations are 0.85 % and 0.926 % H for the 12 storey and 0.63 % and 0.69 %
H for the 18-storey frames respectively, where H is the storey height. These inter-storey drifts
are less than the 1 % level allowed for Seismic Zone A, in Section 3.8.3.1 NZS 4203:1984.
234
These inter-storey drifts are also much less than the limit values of Section 2.5.4.5, NZS
4203: 1992. When an inelastic numerical integration time history method is used, the maximum
permissible inter-storey drifts is 2.5 % (including P-delta and torsion effects) of the
corresponding storey height.
For the 12 and 1S-storey frame-wall structures, the inter-storey drifts are
presented in Fig.9.38 and 9.47 respectively. The maximum values for the fixed and flexible
foundations are 0.95 % and 0.89 % H. for the 12 storey building and 0.61 % and 0.84 %
for 18 storey building respectively. Again, these maximum storey drifts are less the code limit
value. In relation to the P-.6. effects, CARR & MOSS (1980) pointed out that if the inter-storey
drifts are less than 1. 0 %, the P-Ll effect may be justifiably ignored.
9.5.4.5 The Maximum Beam Curvatures and Beam Plastic Hinge Rotations
In the strong column and weak beam failure mode mechanism, the structural
energy dissipation is expected to be due to the formation of beam plastic hinges. A large
amount of energy absorbtion is expected when the moment-curvature hysteresis loops are
relatively stable over several cycles without significant strength deterioration. Hence, the
maximum member curvature sustained during cyclic loading is an important aspect. PAULAY
(1979) and BOOTH (1994) reported that the maximum sustainable plastic hinge rotation was
0.035 radians or approximately 2 degrees. WATSON et a1.(1994) and WATSON & PARK
(1994) reported that based on tests of columns with low levels of axial load, the maximum
measured member curvature was 0.312 radians/m corresponding to 0.0624 radians of plastic
hinge rotation.
The maximum beam plastic hinge rotation for fixed bases and flexible
foundations are 0.0102 and 0.0099 radians for the 12 storey frames and 0.0068 and 0.0066
radians for the 18 storey frames respectively. In the cases of fixed and flexible foundations for
frame-wall structures, the maximum beam plastic hinge rotations are 0.0099 and 0.0071 radians
for the 12 storey buildings and 0.0083 and 0.0069 radians for the 18-storey buildings
respectively. The member rotations of the structure with flexible foundations are smaller than
those for fixed bases. These beam plastic hinge rotations are much lower than that the possible
maximum limit. The storey plastic hinge rotations for the 12 and 18 storey frames are shown
in Fig.9.19 and 9.27 respectively. The storey plastic hinge rotations for fixed bases seem
235
3000
2500 10
Square Frames
Total Base Shear Circle Walls
~
B ..
2000 Filled
Blank
E
.
05
.t::
U)
1500
6
"'"
OJ
Storey
'" 4
1000
2
500 Square : Frames
Circle : Walls
Filled Pos.Env.
Blank : Neg.Env.
o +---------~----------~---- 0.5 0.75 1.25

Fixed Flex. Shear Ratio (Hex. to Fix.)
Fig. 9.32 Total Base Shears Fig. 9.33 Storey Shear Ratios
Storey
10 + ... ..i..fH-O
Square:
Circle
6
Filled
Blank
o "'I-~~,+~~-'--l~~~1--'-~~+~~'--I
0.25 0.5 0.75 1.25 1.5 -10000 -5000 o 5000 10000

Moment Ratio (Flex. to Fix.). Moment (TmL 2-Walls
Fig. 9.34 Column Moment Ratios Fig. 9.35 Wall Moments

236
12 7
1
IG420, FW12RZI
Takeda, ECNS Square Fixed
Circle Flex.
10 Filled Pos.Env.
Dlank lIIeg.Env.
0

0
<Do
0
0
" ".6
0 01111( I!II
I!II .,,;
OJ
C<:
5
:r
;Ii.
B
o 0 o om ~
..-
0
0
,.
... ci
0
.!!
:i0 (;
.- _ II
t
0
lit
0
l1li '" .!.$
15
a:
OJ
""c
:f
3
.. \Ill
li
2
o Flex.
4 .. _._._.........
II Fix.
1
2
I
0
G 11 16 21
PI. Hinge Rot. he 0.001 Rad) left Col. Int.Col. Right Col. Walls
Fi!l. 9.36 Inter-storey Drifts vs. Storey P. Hinge Rot. Fig. 9.37 Column 8< Wall Buses P.Hinge Rotations
12
Storey
10
0.9 ._._.......
Square lB,End 2(+1

..,." "-
B
Circle
Triang.
RB,End 1 1-1
St. Drift
'"
a:
:iC
II

....<> .i.
II
Ailed Fixed
E ._....-......
6 Blank Flex. O.B
'"
III
S
i5.
5
4 Q.
~
0.7
2
Square:
Circl" :
12 St.Frm
1B St.Frm

Triang .: 12 St WF.
Olamd. : 18 St.WF.
j
0.6
o 2 4 (; II 10 2.5 3 3.5
Curv.Ductility Demand & Drift (x 0.1 % HI. FUnd"mental Period (""c.)
Fig. 9.38 Beam Curv. Duct.DemBmls & Int.st. Drifts Fig. 9.39 Sum of Drift to Hb Ratios vs. Fund. Period
237
greater than those for flexible foundations. The accumulation of beam plastic hinge rotations
for the 12 and IS-storey frames are 0.0184 and 0.0127 radians respectively. These values are
less than the results repOlted by TSO et a1.(1993). In the following section the importance of
storey plastic hinge rotations in the calculation of a structural damage index will be discussed.
Figs. 9.20, 9.28, 9.36 and 9.45 represent the variation ofinter-storey drifts and
plastic hinge rotations for the 12 and 18-storey frames and frame-wall structures respectively.
The storey plastic hinge rotations tend to increase as the inter-storey drifts increase, although
the relationships are not clear.
9,5.4.6 Plastic Hinge Rotations of Column and Wall Bases
The column plastic hinge rotations for the 12 and 18 storey frames are shown
in Fig.9.21 and 9.29 and are given in Table 9.1 and 9.2 respectively. Due to the rotation of
the foundations during the earthquake, the column plastic hinge rotations are reduced by an
average of 22.2 % and 24.8 % for the 12 and 18 storey frames respectively. The zero plastic
hinge rotations in these tables imply that the column bases remain elastic. These results confirm
the reduction of the moment at the column bases as indicated in Fig.9.18 for the 12 storey
frames and in Fig.9.26 for the 18 storey frame respectively.
In the case of the frame-wall structures, interesting results were found. Fig.
9.37 shows that the column base rotations for the 12-storey frame-walls with flexible
foundations increase significantly when compared with the fixed base foundations. Fig. 9.46
shows that the column bases of the 18-storey frame-walls with fixed bases remain elastic but
response inelastically in the case of the flexible foundations. In contrast, the wall bases of the
12 and IS-storey frame-walls with fixed bases response inelastically and remain elastic in the
case of flexible bases. This is because the rotation of the walls are several times greater than
the rotation of the column footings, as represented in Fig.9.12 and 9.14 respectively. The soil-
footing foundation to column stiffness ratios are relatively greater than the soil-footing
foundation to wall stiffness ratios. As a consequence, when the wall bases are highly rotated
the column bases displace with the same mode with the walls and resulting in highly increase
of plastic hinge rotation at the column bases.
238
9.5.4.7 Beam CUl'Vature Ductility Demands
In the strong column and weak beam type of failure mechanism, the member
ductility demand is the most common damage parameter used in inelastic analyses. The
limitation of this damage parameter is its inability to describe the member damage under low-
cycle fatigue type loadings.
As shown in Figs. 9.22 and 9.30, the beam CUIvature ductility factors over the
building height are smaller in the frame buildings with flexible foundations than those for fixed
base foundations. Similar results are found for the 12 and 18 storey frame-wall structures as
shown in Figs. 9.38 and 9.47 respectively. The maximum curvature ductility demands are 9.52
and 7.86 for the 12 storey frame and 6.58 and 5.93 for the 18 storey frame with fixed bases
and flexible foundations respectively. The equivalent values are 8.99 and 7.45 for the 12 storey
frame-wall structures and 7.15 and 6.2 for the 18 storey frame-wall structures. These show that
the maximum value of curvature ductility demand in the buildings with flexible foundations are
approximately 15 % smaller than those for fixed based buildings.
PRIESTLEY et al.(1981) reported that tests of spirally-confined concrete

columns with a relatively low axial load ratio P/Ag C where P, Ag and fc are the axial load,
gross area and compressive cylinder strength of concrete respectively, show that the maximum
available curvature ductility factors range between 16.3 to 25.4. Similar tests have been
conducted by PARK et a1.(1982) with the maximum curvature ductilities varying between 20
to 21. ZAHN et al.(1986) and WATSON et al.(1994) discussed a proposed curvature ductility
design chart. The chart shows that if the axial load ratio is less than 0.15, a curvature ductility
even greater than 30 may be obtained. Hence, the beam curvature ductility demand obtained
in this analyses can easily be achieved.
9.5.4.8 The Accumulation of Plastic Hinge Rotations
The maximum plastic hinge rotation and ductility demand as discussed earlier
represent a single value damage parameter. Accordingly, the structure with a higher plastic
hinge rotation and curvature ductility which only occurs in one member has a considerably
greater damage than the structure with only slightly smaller plastic hinge rotations and
239
, -.... 5000
G685, FW18R~1
Takeda, EcNSC Square 7Frames
4500
Circle 2Walls
:+ Triang. Tt.B.Shear
4000
Filled Neg.En\!.
gIII 0.8
Blank Pos.En\!.
r:c 3500
.a
J:
B A
I/)
.t: A 1111 '* E
3000
d l\'i
>- 0.6 to
!! .s::
CII
2500
B OJ
':'
2 "co
00
2000
.!:
....0
E
"
!f) 0.4- 1500
Square: 12 St.Frm
Circle : 1B St.Frm 1000
Tfiang. : 12 St.WF.
Oiamd. : 18 St.WF.
600
0.2
2 2.5 3 3.5 0
Fundamental Period (sec.) Fixed Flex.
Fig. 9.40 Sum of Storey Drifts to Hb Ratios vs. Period Fig. 9.41 Base Shears
18 18~-----,------{LHU~-.------r-----~
16 16 ... ...."""l"..",,i~...,,,,,,l,,,,...."" .. ~.. " .... " ........

Storey
14 14 ... ..+.... ....IT)'. .,,,,! .. ......T .... ..............
12 ..... Square: Frames 12+...."",, ...j......,,.. ,,fc.. ....j .... -t.... "".. 1
Circle : Walls
: Pos.Enll.
10 10 ....".. ".... ...;........ """,, .. fnc-:. :I~"i"'''''''''''''''''''''i'''''''''''''''''''''1
: Neg.Enll.
B .ilt""""..",,,,,,+,,,,,,.....,,,,..,,... B +."......"..". +.......""."....... IL).. ""..""i"....""".."'i....,,..,,j
Square: Left Col.

6 6 +......... ",, ..1111 ... '-1-17 ... ""...., Circle : Int. Col.
Filled : Pos.En\!.
4 4 ............."." ... l".. ,,IIT:H ...."1Blank : Neg.En\!,
2 2+,,.. <l~~~!~,,"... " ....
0
0.5 1.5 2 o 0.5 1.5 2 2.5
Shear Ratio (Flex. to Fhe.) Moment Ratio (Flex. to Fix.!
Fig. 9.42 Storey Shear Ratios Fig. 9.43 Column Moment Ratios
240
1'111.
o FIIllt.
1--__;-.. _...0+_..14... ........,. ...-<Ji
6 ........_. . ........
=-a L.... ~ ......... O.'b~i . . ~~. ........j
btorev 00
I_..,......+..u !. .
......... _42 ..+......... ,..... .... ..(.
!
.......... , .........10..
..
jfilled : Fix.
! IBlonk: Flex. II II
4 +..... . , ....1....,........ _....+_....__ ...... -+~ .. -......-......1

30000 -20000 -10000 o 10000 20000 2 6 10 14 18
Moment ITml, 2-Walls Storey P.H. Rot.lx 0.001 Red.1
Fig. 9.44 Wall Moments Fig. 9.45 Inter..storey Drifts vs. Storey P.Hinga Rot.
5 18
,Squllre Fixed 16
Cllcle Flel(,
Storey
4 Filled : Pos.Env.
14
Blank : NClg.Env.
12
,,;
...e
0
3
10
0
ci
~
..:
0
Ie 8
OJ
'"'"
:f
2
6
;;::
1
2
o
o 4 B '12
I.ott Col. Int. Col. Right Col. Wall Drift Ix 0.1 % HI & Beam CUrI/.Duct. Demands
Fig. 9.46 Column &. Wall Bases PI. Hinge Rotations Fig. 9.41 Inter-storey Drifts &. Beam C.Duct. Demands
241
curvature ductility demands even though they occur in all members. The concept of cumulative
plastic hinge rotations has been introduced to take account of cumulative effect of the structural
damage under dynamic loadings. The use of cumulative plastic hinge rotations to measure the
severity ofRC fixed frames under earthquake excitations were also reported TSO et al.(1993).
The cumulative plastic hinge rotations in the positive envelope of the 12-storey
frame and in the negative envelope of the 18 storey frames are shown in Figs. 9.23 and 9.31
respectively. The upper storeys of the 12-storey frame and an the storeys of the IS-storey
frame with flexible foundations suffer greater damage than the stmctures with fixed base
foundations. As given in Tables 9.1 and 9.2, the total plastic hinge rotations of the 12 and 18-
storey frames with fixed base foundations are 2 % and 11 % higher than those for flexible
foundations. Total plastic hinge rotations are 0.3479 radians and 0.3586 radians for the 12 and
IS-storey frame with fixed bases and 0.3418 radians and 0.3231 radians for the same frames
with flexible foundation respectively. A relatively flexible structure with flexible foundations
suffers less damage than a stiffer structure with fixed base foundations.
Total plastic hinge rotations for the 12 and I8-storey frame-wall structures are
presented in Tables 9.3 and 9.4 respectively. They show that the total plastic hinge rotations
of the 12-storey frame-walls with fixed base foundations are higher and accordingly suffer more
damage than those for flexible foundations. It is interesting to note that in the IS-storey frame-
walls with fixed base foundations the total plastic hinge rotation is 21.5 % smaller than for the
flexible foundations. This means that the 18-storey frame-walls with flexible foundations suffer
more damage than when it has a fixed base foundations. The fundamental period of the
structure and the spectral characteristics of the earthquake excitation seem to effect the
structural response. This is an exceptional result and in Chapter XII it will be shown that, in
general, the flexibility of the foundations reduce the structural responses.
9.5,4,9 Viscous and Hysteretic Energy Dissipation
As with the cumulative plastic hinge rotations, the hysteretic energy absorbed
by the structures can be used to represent the cumulative value of the damage parameter. This
hysteretic energy absorbtion is a function of time, hence, a longer duration of the earthquake
excitation may result in a greater absorbtion of hysteretic energy dissipation than that for a
shOtt duration earthquake. The 20 seconds of earthquake excitation from the 1940 EI-Centro
60 60
40 4Q
E E
1:: !::
c Hysteretic En. J:
,
c 30 C 30 Hysteretic En,
;;:; .,
';'
'"
"-
;;;
"-
'i
~ m
5 20 5
...!? ...
E'
20
C'"
I.!J
'"
C
I.U
10 10
0 0
0 s 10 12 14 16 18 20 0 10 12 14 19 20
Time Isec)
Fig. 9.48 Visco Bnd Hyst. Energy Dissiplldons of the 12-etQrey Fixed Frame Fig. 9.49 Visco snd H'(6t. Energy Dissipatiol'l!l 01 the 12-etorey Frame
700 700
Q420, FW12f\Z
SOO 600 Takeda, EcNSci
E 500
E500
Hysteretlc En.
"c
;;:; 400
"c
:0 400
I
Hystaretlc Ell.
""- '"
Q.
'iii
<c ..
'i
is 300
is 300
...E' ...eo
""
OJ
-200 200
"
!.U
En.
Ilol
En.
100 100
0 0
0 2 .8 10 12 14 115 13 20 0 2 10 14 18 20
'nme lseci Time (sec)
Fig. 9.50 Visco and Hyst. Energy Dissipations of the 1~-storey Fixed Frlllme-Walls Fig, 9.51 .Vlsc, IIInd Hyst. Energy Dissipations ~f the 12-etorey Fram<J-Wails
243
NSC earthquake is used. Ref. TID used this parameter to describe the severity of damage to RC
frames with fixed base foundations under earthquake excitations.
As given in Tables 9.1 and 9.2, the hysteresis energy demand of the 12 and 18-
storey frames with fixed base foundations are 10.9 % and 12.5 % higher than those for the
structures with flexible foundations respectively. The hysteretic energy absorbed by the 12
storey frame is nearly equal to that of the 1S-storey frames. Thus, the energy absorbed per unit
mass of the 12 storey frame is greater than that for the 1S-storey frame. It can be concluded
that the 12-storey frames are considerably more damage than the 18-storey frames.
The hysteretic energy absorbed in the 12 and 18-storey frame-walls with

flexible base foundations is much lower than that for the fixed base foundations. Again, this
result is because of the rotation the structure due to the foundation flexibility. In contrast, the
viscous energy absorbed by the flexible foundations is much greater than that absorbed by the
fixed base foundations. The rotation of the structural systems due to the foundation flexibility
increases the velocity response. The hysteretic energy dissipation in the 12-storey frame-walls
is even higher and results in absorbed energy per unit mass being much higher than that for the
18 storey frame-walls. Accordingly, the stiffer structures with fIXed base foundations are
considerably more damaged than that for relatively flexible structures with flexible foundations.
Examples of viscous and hysteretic energy absorbed by the 12 storey frames and frame-walls
with fixed bases and flexible foundations are shown in Figs.9A8, 9.49, 9.50 and 9.51
respectively. They show that the patterns of energy absorbtion are similar.
9.5.5 Member, Storey and Structural Damage Indices
Fig. 9.3 shows the damage index of the member can be calculated from either demand
versus capacity procedures or degradation procedures. The choice of the procedure depends on
the expected member damage modes, using either a single value, the cumulative value or the
combination of these damage parameters.
Results of these analyses indicate that the combination of a single damage parameter
and the cumulative damage parameters are likely to be the most appropriate choice.
Accordingly, the damage index proposed by PARK et al.(1984) as expressed in Eq.9.8 is used
244
0.3 , - - - - - - . - - - - -
B +''-----------t--''' ...'..-.... +-- .. ----.. -..+.. ----..--.. -C~-,--, .. ,,,--"\--,,,,,,,-,,-,,,,,1

K 0.25
(l)
-'""=
III
01
6 '"
E
III
Cl
iii
....!i
"
....2
IJ)
0.2
2+------,,----,,+----
I 12Stf. Frames,
O~J=~~~~~~~~~~~~~~~
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.15 +-----------+---------~~--------~
Damage Index Fixed Flex.
Fig. 9.52 Storey Damage Index Fig. 9.53 Str. Damage Index vs. Found Types
0.3 ,------------------------------
Storey
10+'''----,..---- ..+..,,---,.. '''"+C)---.--~J~I..-.. +....,,-,,......... -
0.25
8 .... ..r.S~q;;;u~a:;r::e---:-~:::'~~-1.. ...... -+-~"::::-(){5iNII
,J
Circle ..,
Ql
.E
Ql
Filled .,rn
6 Blank
~1IIIIlIIIIIIlII1IIIIIIIIIi!III. 0
.E 0.2
iii
....!i
4+,,,,,.. ,, ....,, .. i-,,..-.(r' .. t"'........,....nf....'... ,..,.... ;.... -..... ,,,......,., "
:J
J:>
IJ)
0.15
2+, .. , .. -,,+,'.,. 0 ....f, .... ,'''..'..''' ..l''''''''.. ~].
Square Frame
Circle FWalis
o .kJ~$::cf:-=-
0.05 0.1 0.15 0.2 0.25 0.3 0.1
Fig. 9.54 Frame & FrameWall Storey Damage Indices Fig. 9.55 Str. Damage Index vs. Found. Types
245
13 0.2 , - - - - - - - - - - -..-~.~-.-... --____,
16
14
12
0.15
'"
."
.E
10 <II
.,tn
E
8
0 '"
~
.a
6 "2
Storey Iii 0.1
0
0.05 0.1 0.15 0.2 0.05 + - - - - - - + - - - - - - - + - - - - - - - 1
Fig. 9.56 Storey Damage Indices Fig. 9.57 Structural Damage Indices vs. Found. Types
0.19
0.17
0.15
I<
'"
."
.E
III
'"
B 0
.
"
E
0.13
E
tI" 0.11
6 Frame
FWalis Iii"
Fixed
0.09
Flex.
Square Frame
0.07 Circle FWalis
o +~.---l.~-liI---t ---Cl'''''HIt---j----i
o 0.05 0.1 0.15 0.2 0.25
0.05
Damage Index.
Fixed Flex.
Fig. 9.58 Frame & Frame-Walls Storey Damage Indices Fig. 9.59 Structural Damage Indices vs. Found. Types
246
in this study. The damage index of the member is calculated by the summation of normalized
stmctural deformation and member dissipation energy. Instead of using the member
deformation, CARR & TABUCHI (1993) modified this parameter to use the member curvature.
The member curvature demand and member dissipation energy represent the single and
cumulative values respectively.
The storey damage index is based on the member damage indices in a particular storey.
As mentioned in Section 9.4, the weighting factors used in the calculation of the storey damage
indices depend on the failure mode mechanism. The strong column and weak beam failure
mode mechanism is selected in this analysis. Based on the results as presented in Chapter XII,
the member damage indices are strongly correlated with the member curvatures. Accordingly,
the weighting factors based on member curvatures are used in the calculation of storey damage
indices. The structural damage index is based on the storey damage indices. As will be
discussed in Chapter XII that the storey damage indices are also strongly correlated with the
curvature distribution over the building height. Hence, a similar weighting factor to that used
in the calculation of storey damage indices is used in the calculation of a structural damage
index.
Table 9.5 Damage Indices of the 12 and 18-storey Structures

Foundation Types
No. Types of the Structures and Remark
Damage Parameters Fixed Base Flex. Found.
1. The 12-storey Frame

a. Max. Member Damage Index 0.3340 0.2790 (- )
b. Max. Storey Damage Index 0.2929 0.2652 ( - )
c. Max. Structural Dm.1ndex 0.2377 0.2219 (- )
2. The 12-storey Frame-Walls
a. Max. Member Damage Index 0.3200 0.2670 (- )
b. Max. Storey Damage Index 0.2694 0.2314 -
( )
c. Max. Structural Dm.Index 0.1948 0.1700 ( - )
3. The 18-s1orey Frame
a. Max. Member Damage Index 0.2290 0.2070 ( - )
b. Max. Storey Damage Index 0.1798 0.1649 -
( )
c. Max. Structural Dm.lndex 0.1568 0.1317 ( - )
a. Max. Member Damage Index 0.2210 0.2530 (+)
b. Max. Storey Damage Index 0.1889 0.2177 (+)
c. Max. Structural Dm.Index 0.1167 0.1390 (+)
Remark: ( - ) : Decrease, (+) : Increase
247
Based on the laboratory tests discussed in Section 9.5 A. 7, the possible available
curvature ductility factor more than 30 may be achieved for a member with low level of axial
load. The maximum curvature ductility factors of 20 and 30 for columns and beams
respectively are used. PAULAY & PRIESTLEY (1992) stated that the available wall curvature
ductility factor depends on the required displacement ductility factor and the wall height to
width ratio. The value of wall height to width ratios of the building prototypes in this study
were 9.03 and 8.94 for the 12 and 18 storey frame-walls respectively. Accordingly, the
maximum wall curvature ductility factor of 16 was selected in this study.
The maximum member, storey and structural damage indices for the 12 and IS-storey
frames and frame-wall structures with fIxed bases and flexible foundations are given in Table
9.5. The distribution of storey damage indices for the 12-storey frame with fIxed base and
flexible foundations is shown in Fig.9.52 and the corresponding structural damage index is
shown in Fig. 9.53. The storey damage indices of frames with flexible foundations are smaller
than those with a fixed base in over all storeys and resulting a 6.6 % smaller structural damage
index. Fig.9.54 shows the distribution of storey damage index of the 12-storey frame and
frame-wall structures. As with the frame structures, the storey damage indices of the frame-
walls with flexible foundations are smaller than that for fixed base foundations resulting a 12.7
% smaller structural damage index.
As shown in Fig.9.54, the effect of walls reduce the storey damage indices over almost
all of the building height. This is because the presence of the walls reduces the beam curvature
demand. As the result, the damage indices of frame-wall structures for both types of
foundations are smaller than those for frames structures as shown in Fig. 9.55. The effect of
the walls reduces the structural damage indices by 18 % and 23.3 % for the fIXed and flexible
foundations respectively.
The storey damage indices for the I8-storey frame with flexible foundations are smaller
than those for the fixed base over the building height as shown in Fig.9.56 resulting in a 16
% smaller structural damage index as shown in Fig.9.57. Fig.9.58 shows the storey damage
index for the IS-storey frame and frame-walls respectively. The storey damage index of the
frame-walls with flexible foundations are greater in the lower and upper storeys. The hysteretic
energy absorbed by frame-walls with flexible foundations as given in Table 9.4 is smaller but
the beam curvature demands are greater than those for the fixed base foundations. As a result,
248
the structural damage index of frame-walls with flexible foundation is greater than that for the
fixed base case. This indicates that the beam curvature demands have greater effect on the
structural damage index than the hysteretic energy absorbtion. The rotation of the walls cause
increases in the damage indices at the column bases as shown in Fig. 9.54 and 9.58
respectively. This result confirms as reported in Ref. VIO where effects of the inertial
interaction in soil-structure interaction may increase, decrease or have no effect on the response
of structural system. The outcome depends on the characteristics of the response spectrum at
the fundamental period of the structural system.
The interpretation of inelastic building responses under seismic excitation have been
discussed. The response quantities of inelastic analyses are more complex than those of an
elastic analysis. The damage parameters were categorized into two groups, a single maximum
and cumulative values. Several approaches for the calculation of the member damage mode
have been proposed by researchers based on either demand versus capacity or degradation
procedures. The member, storey and structural damage indices were based on the damage
modes.
Effects of the soil-foundation flexibility on the inelastic response of the 12 and 18-
storey frame and wall-frame building structures have been investigated. The soil-foundation
impedances were calculated using the soil-foundation model discussed in Chapter V with the
soil shear modulus G = 685 kg/cm 2
The member, storey and structural damage indices were
those proposed by PARK & ANG (1985).
The maximum calculated soil-shear strain 'Y.. in all building prototypes is 1.74x10"" and
is less than 10-2 i.e the limit value for the visco-elastic soil mechanical model as discussed in
Section 4.3. This justifies the proposed visco-elastic soil-foundation mechanical model as
discussed in Chapter V. The soil shear modulus G = 0.70 Gmax used in the analysis was in
close agreement with the calculated soil shear strain.
During the loading history, the column footing foundations were a long way from the
tip lift-off limit. The rotations of the wall bases were more than 8 times than those of the
249
column bases. The wall footing foundations, in general, were in full contact with the supported
soil during the earthquake excitation.
Compared with flXed base foundations, the soil-foundation flexibility causes a greater
lengthening in frame-walls fundamental periods than was evident in the frame structures. The
soil-foundation flexibility decreases in structural stiffness of frame-wall structures was greater
than in the framed structures. The ratio of roof horizontal displacement to the sum of the inter-
storey drifts decreases as the structural fundamental period increases.
The base shear decreases as the flexibility of the foundations increases. The base shear
of the frame-walls decreases more than that in the frame structures. The rotation of the wall
base greatly reduces the wall first-storey moment, shear and plastic hinge rotations, but
increases greatly the first-storey moments, shears and the plastic hinge rotations in the frames.
The redistribution of wall and column forces due to foundation flexibility was more significant
in the more flexible structures.
The inter-storey drifts in all building prototypes were less than 1 % of the storey
height. The foundation flexibility increases the upper storey inter-storey drifts, especially for
more flexible structures. Note that the base rotation is added to the inter-storey drifts due to
deformation within the storey. The beam ductility demands and storey plastic hinge rotations
of buildings with flexible foundations decrease when compared with the fixed base foundations .
.In both types of foundation, the storey plastic hinge rotations tend to increase as the inter-storey
drift increases.
The foundation flexibility decreases the absorbed hysteretic energy and increases the
absorbed viscous energy more in frame-wall structures than in framed structures. The rate of
energy absorbtion depends on the characteristics of the earthquake input.
Except for the 1S-storey frame-walls, the maximum member, storey and structural
damage indices of buildings with flexible foundations were smaller than those for fixed base
foundations. The damage indices were more strongly correlated with the member curvatures
than the hysteretic energy dissipation.
Chapter X
Effects of Rigid Joint Lengths, NUlnber of Frames and
Hysteresis Rules on the Response of the 12 and 1S-storey
Frames and Frame-Wall Structures
10.1 INTRODUCTION
As discussed in Section 6.2 the interpretations of the results of inelastic analyses of

structures are more complex than that for elastic analyses. This is because a number of
uncertain parameters affect on structural responses. OTANI (1981) and OTANI (1984) discuss
the maximum response amplitudes and response histories of inelastic RC structures influenced
by various stiffness properties of the members. The different hysteresis rules are expected to
respond to an earthquake excitation in a dissimilar manner. SAlIDI (1980) also presented the
effect of the hysteretic models on the seismic response of RC frames. When compared with the
result of laboratory tests, it was found that the accuracy of the hysteresis rules in predicting the
inelastic member behaviour depends on the response amplitude.
The parameters involved in this chapter are the effects of the rigid joint zone lengths,
the number of frames in frame-wall structures, and different member hysteresis rules. The
effect of these parameters on the inelastic responses of the 12 and 18 storey frames and frame-
wall structures with fixed bases and flexible foundations will be discussed in the following
sections.
The calculation of soil-foundation impedances are based on constant coefficients as

discussed in Ref.H20 and T19, leads to the frequency independent response analysis. This
implies that the impedance coefficients are based on a particular frequency of excitation and
are assumed to be constant during the time integration. The excitation frequency w = 15.707
rad/sec corresponding to the frequency f = 2.5 Hertz is used in this study. This frequency is
in the range mentioned in Section 4.1 and discussed in Ref.L6. The member plastic hinge
rotations are based on a plastic hinge length 4 = 0.50 diU where d m is the depth of the
member.
250
251
10.2 EFFECTS OF THE RIGID JOINT ZONE LENGTHS
It is possible that beam column joints are a critical region in reinforced concrete frames
designed for inelastic response under seismic excitations. Experience from past earthquakes,
as reported in Refs.M30, S34 and S35, indicate that the damage of some buildings are caused
by the failure of the beam column joints. It was found that these failures were mainly because
of unanchored bars, lack of joint reinforcement and the storey stiffened by a masonry wall. The
sub-assemblage test results reported in Ref.P30 indicate that the beam column joints may
exhibit inelastic stress when the beam plastic hinges occur at the column face. Hence, the rigid
joint zone length may vary from zero for highly inelastic member to half of the column width
for fully elastic joints respectively.
As discussed in Section 6.7 the rigid joint zone length may significantly affect the
structural responses. The effect of the rigid joint zone lengths on the inelastic response of the
12-storey frames and the 18-storey frame-wall buildings with flXed base and flexible
foundations are investigated. The soil-foundation impedances were calculated based on the soil
shear modulus G = 420 kg/cm2 for the 12-storey frame structures and G = 685 kg/cm2 for the
18-storey structures. The rigid joint zone lengths ratio Z = 0.0, 0.5 and 1.0 correspond to the
beams with rigid zones equal to zero, one-forth and one-half of the column width respectively.
The modified Takeda hysteresis rules and the 1940 EI-Centro NSC earthquake are used.
10.2.1 Fundamental Period and Roof Horizontal Displacement
As shown in Table 10.1, increasing the rigid joint zone lengths cause a decrease in the
fundamental periods of free vibration of the 12-storey frames by 6.8 % and 6.2 % for the fixed
and flexible foundations respectively. The stiffening of the structures due to the rigid joint
length have been discussed in some referencesC16.07.01O. These shortened periods are even
greater than the lengthening of the natural periods due to foundation flexibility. The effects of
the rigid zone length on the response of framed structures is likely to be greater than the effects
of the different foundation types as discussed in Section 9.5.
The reduction of structural fundamental periods of the 1S-storey frame-walls as given

in Table 10.2 are 5.0 % and 5.6 % for fixed bases and flexible foundations respectively.
Compared with the lengthening of structural fundamental periods of the 18-storey frame-walls
due to flexible foundations, the reduction of these periods are much lower. This means that the
252
175
Storey
150
125 Square: left CoI.IZ=O.OJ

Circle : Int.Col.(Z=O.O)
Trlang. : left Col.lZ = 1.0)
Diamd. : Int.Col.IZ =1.0)
E 100
Filled : Ficlad
Blank : Flex.
75
50
Square Neg.En".
Circle Pos.En". 2
25 Filled Fixed
Blank Flex.
0 0.4 0.6 O.B 1.2 1.4

Z = 0.0 Z = 0.5 Z = 1.0 Column Moment Ratios {to Z = 0.501
Fig. 10.1 Base Shear VB. Rigid Joint lengths Fig. 10.2 Column Moment Ratios
12 ,------o-----,-----,~----~--__,
.. fix.
10
o Flex.
.,.
4
i ! -
~,-~-..-f~.~.--"". ...... 1.. _._._. ', ..... ,.l....... _._ .......... j ... _...._._....,....
1 !
! I l
I
o 5 10 15 20 25 o 5 10 15 20 25
Drift Ix 0.1 % HI. P.H.Rot.lx 0.001 Rad.1 StollaY P.H. Rotations Ix 0.001 Rod.\
Fig. 10.3 Inter-storey Drifts & Storey 1'. Hinge Rot. Fig. 10.4 Inter-storey Drifts & Storey P .Hinge Rot.
253
5 12
Triang. : Z = 0.0
Storey
Square: Z = 0.5
Circle Z = 1.0
10
4 Filled Fixed
Blank Flex.
..;
co B
a;
~
0 Squar~ : Z = 0.00
q 3 Circle : Z = 0.50
0
~
Triang. : Z = 1.00
!II 6
'"<>
.~
Filled fixed
...<> Blank Flex.
r<:
2
())
[J) 4
5:"
0:
o +-------~--------~------~--------- o 3 6 9 12
Left Col. Int.Col. Right Col. Curv. Duct. Demand IRB,End 1, Neg.Env.1
Fig. 10.5 Column's Base PI. Hinge Rotations Fig. 10.6 Beam Curv. Ductility Demands
300 ,---------------------------------, 60 ,------------------------------,
250 50
Total (File.)
200 40
E
~
t:.
o c:
q 0
o ~
Q.
~ 150 'iii 30
c: .~
<> 0
~
... ;..
o f!'
c::
i '"c:
L!J
a.: 100 20
Square Pos.Env.
Circle Neg.Env. Sqr.Filied : Vise .. Fix.
50
Filled
Blank
Fixed
Flex.
10 -- Ccl.FlIled : Vise., Flex.
Sqr.B1ank: HysL. Fix.

Ccl.Blank : Hyst .. Flex.
!
o +-------+-------+-------~----~ o +i------~,r-----_+,-------~----~
Z=O.OO Z=O.50 z= 1.00 Z=O.OO Z=0.50 Z=1.0
Fig. 10.7 Plastic Hinge Rotations Fig. 10.8 Viscous & Hyst. Energy Dissipation
254
effect of rigid joint zone lengths on the structural responses of frame-wall structures will be less
significant than the effects of the foundation types.
The effect of rigid joint zone length on the roof horizontal displacement of the 12-storey
frames is given in Table. 10. 1. In general, the effect of the rigid joint zone lengths on the roof
horizontal displacements are not clear. For the IS-storey frame-walls. the effect of the rigid
joint zone lengths increases the roof displacement in the negative envelope but the opposite
result was found in the positive envelope. The significance of the relationship between roof
horizontal displacement and structural damage index will be discussed later.
10.2.2 Base Shears and Member Forces
As given in Table 10.1 and shown in Fig. 10.1, the effects of the rigidjoint zone length
on the 12-storey frames increase the base shear with an average of 10.35 % and 3.50 % for
the fixed base and flexible foundations respectively. The shortening of the fundamental period
as discussed in the foregoing section results in higher base shear. The effects of the rigid joint
zone length on the total base shear of the 18-storey frame-walls are given in Table 10.2 and
shown in Fig. 10.10. Increase in the rigid joint lengths almost causes no affect in the total base
shear of the 18-storey frame-walls with fixed bases and only increases them by an average of
0.75 % for the flexible foundations. These results are smaller than those for the frame
structures implying that the effects of the rigid joint zone length on the base shear is more
significant in a frame than in a frame-wall structure.
The column moment ratios with respect to the rigid joint zone length Z = 0.50 for the
12-storey frames are presented in Fig. 10.2. These results show that the effects of rigid joint
zone length on the column moments is not obvious. As the frames become stiffer due to
increasing rigid joint zone lengths, the wall moments for the 12 and 18-storey frame-walls
decreased by an average of 3.06 % and 3.56 % respectively.
10.2~3 Inter-storey Drifts and Plastic Hinge Rotations
The effect of rigid joint lengths on the inter-storey drifts of the 12-storey frames as
shown in Figs. 10.3, is insignificant especially between Z = 0.5 and Z = 1. An increase in
storey drifts might be caused by the decrease in structural fundamental periods as discussed in
Section 10.2.1. Fig. 10.11 shows that the effect of rigid joint zone lengths on the inter-storey
255
1)0 4000
Tot. Base Shear

35
3000
'"
I\)
E
30
E
OJ
"
~
OJ ro
ii ~ 2000
is'"
(f)
<JJ
~
VI
'"'"
D
J: 25
....0
0
0:: Square Pos.Env.
Circle Neg.Env.
1000
~'"".
Fixed Filled : Fixed
20 Circle Flex. Blank : Flex.
Filled Pos.Env.
Blank Neg.Env.
15 +--------~------_4-------_+------~ o +-----i
Z = 0.0 Z = 0.5 Z = 1.0 Z = 0.0 Z = 0.5 Z = 1.0
Fig.10.9 Roof Displacements vs Rigid Joint lengths Fig. 10.10 Base Shears vs. Rigid Joint lengths
16 10
16
6 -r- ................i ........................ ~ .. ..,

14
Storey
12
...o 6 +........................ ~r:
10 .?5
~Cl
6 >-
2!
4 ~ .... -.... ......~~ ............ - ..-" .. ................t ..........................,
6
~
.'!l
Square Z = 0.0 .5
Circle Z 0.5=
4 Triang.
Diamd. :
Z = 1.0
Z =0.5.Fhe.
2 + . ....................+ ........................}.... .. I Fix.
2
o Flex.
0
0 5 10 15 20 o 5 10 15 20
Drifts Ix 0.1 % H), P.H.Rot.lx 0.001 Rad.) Storey P. Hinge Rot. he 0.001 Rad.1
Fig. 10.11 Inter-storey Drift & Storey P.Hinge Rot. Fig. 10.12 Inter-storey D.-itt vs. Storey P.H. Rot.
256
5 , ..... ~~~~~~~~~~~~-----, 18
16
Storey
4 14
'"0 12
OJ
tC
.-
0 3
0 10
0
!! Square Z = 0.00
c
.g B Circle Z = 0.50 l1li ................ 1
~ Traing. Z ~ 1.00
'"
0:::
2. Filled FiKed
'"'"<::
:E
6 ............... Blank
Flex.
a:
4 +............ .... ....I...... ......................
,fA-.............. .. ....
Square
Circle
Triang.
2.
Filled
Blank
0
o 0 2 4 6 B
Left Col. Int. Col. Right Col. Walls Curv. Due. Demands IRS.End 1. Neg.Env.1
Fig. 10.13 Column & Wall Bases P. Hinge Rotations Fig. 10.14 Beam Curvature Ductility Demands
250 ,-~--------~~~~~~~--~ 800
Square : Viscous
Circle Hyster.
Triang. Tot. En.
Filled Filled
Blank Flex.
200 600
-ci
OJ
D:
~ E
0 !::
~
.,"
0
.~
!!
Q.
<:: 150 'i;j 400
'".'"!! "
i5
rr:" >-
E'
(\)
1ll
C!
:E"
"
1.1.1
"-
100 200
Square Neg.Env.
Circle Pos.Env.
0 0 0
Filled Fixed
Blank FleK.
50 o +-~~~~------~~~~~+-~--~~
Z ~ 0.0 Z = 0.5 Z = 1.0 Z = 0.00 Z = 0.50 Z 1,00
Fig.10.15 Total P. Hinge Rot. vs. Rigid Joint Lengths Fig.1 0.16 Energy Dissipation vs. Rigid Joint Lengths
257
drifts of the 1S-storey frame-walls with flexible foundations tends to decrease in the upper and
lower storeys but increase in middle storeys. This is because the stiffer the frames the less the
interaction effect between the frames and the walls at those storeys.
Table 10.1 Response Quantities of the 12-storey Frames with Fixed Base and Flexible Foundations
Rigid Zone Length

No. Rem.
Z = 0.00 Z = 0.50 Z = 1.00
Parameters & Str. Systems
1. Fund. Periods (sec)
a. Fixed Bases 2.951 ( 0.245 N ) 2.753 ( 0.229 N ) 2.565 ( 0.214 N ) (-)
b. Flex.Foundations 3.076 ( 0.256 N ) 2.886 ( 0.241 N ) 2.707 ( 0.226 N ) (-)
2. Roof Hor. Displacement (m)
a. Fixed Bases
1. Pos.Envelope 0.1924 (0.43 % Ho 0.1839 (0.41 % Ho) 0.1852 (4.10 % Hb (.)
2. Neg.Envelope 0.2833 (0.63 % Ho 0.2991 (0.66 % lit,) 0.2939 (0.65 % HtJ (.)
b. Flex.Foundations
1. Pos.Envelope 0.1982 (0.44 % Ho 0.1897 (0.42 % lit,) 0.1893 (0.42 % Hb (.)
2. Neg.Envelope 0.2805 (0.62 % Ho 0.3022 (0.67 % Ho 0.3024 (0.67 % Hb (+)
3. Base Shear (Tonne) Fixed Flex. Fixed Flex. Fixed Flex.
1. Pos.Envelope 93.50 92.00 104.00 102.50 113.00 98.50 (+)(+)

2. Neg.Envelope 87.50 87.00 94.80 94.20 107.50 92.70 (+) (+)
4. Base P.H.Rot.(xO.OOI Rad)
a. Fixed Base
1. Left Column + 1.23, -1.05 +1.79, -1.23 +2.70, -1.10 (+)
2. Internal Column + 1.66, -1.23 +2.60, -1.83 +3.10, -1.31 (+)
3. Right Column +1.15, -1.08 +2.31, -1.18 +2.79, -1.38 (+)
b. Flexible Foundations
1. Left Column +1.07, -0.98 + 1.52, -1.01 +2.13, -1.17 (+)
2. Internal Column + 1.33, -1.19 +2.20, -1.25 +2.73, -1.25 (+)
3. Right Column +1.05, -1.04 + 1.49, -1.06 + 1.94, -1.11 (+)
5. ToLP.H.Rots. (xO.OOI Rad) Fixed Flex. Fixed Flex. Fixed Flex.
1. Pos.Envelope 156.140 147.940 159.704 160.099 167.133 168.370 (+) (+)

2. Neg.Envelope 174.250 164.200 188.244 181.722 213.496 210.294 (+)(+)
3. Total 330.390 312.140 347.948 341.821 380.629 378.664 (+)(+)
6. Energy Dissipation (Tm) Fixed Flex. Pixed Flex. Fixed Flex.
1. Viscous Energy 20.30 19.40 21.80 21.80 23.70 24.00 (+)(+)

2. Hyst. Energy 19.26 17.50 21.30 19.20 25.00 23.30 (+)(+)
3. Total Energy 39.56 36.90 43.10 41.00 48.70 47.30 (+)(+)
Remark: (+): Increase, (-) : Decrease, (.) : Not clear
The effect of rigid joint length on the storey plastic hinge rotations of the 12-storey
frames with flexible foundations seem more obvious when compared with the IS-storey frame-
walls. In general, the rigid joint zone lengths cause an increase in storey plastic hinge rotations
over the storey levels. These results confirm earlier findings about increasing base shear, as
discussed in the foregoing section. Increased storey plastic hinge rotations will strongly affect
the storey and stmctural damage indices as discussed in Section 10.2.6. Fig. 10.12 shows that
258
the plastic hinge rotations of the 18-storey frame-walls increase as the inter-storey drifts
increase and the inter-storey drifts are greater in the building on flexible foundations than that
in fixed bases. More scatter result is found in the 12-storey frames as shown in Fig. 10.4.
10.2.4 Column, WaH Base Plastic Hinge Rotations and Beam Curvature
Ductility Demands
The rigid joint zone lengths in the 12 storey frames increase the column base plastic
hinge rotations by an average of 48.8 % and 42.1 % for fixed bases and flexible foundations
respectively, as given in Table 10.1 or Fig. 10.5. This is because of an increase in base shear
as discussed in Section 10.2.2. Fig. 10.13 shows that the increase in column base plastic hinge
rotations of the 18-storey frame-walls is more strongly affected by the flexibility of the
foundations than by rigid joint zone lengths. Increased plastic hinge rotation of the column
bases due to rotation of the walls was also reported by GOODSIR (1985). This agrees with the
previous results that response of frame structures are more markedly affected by rigid joint
zone lengths than are frame-wall structures.
In the 12-storey frames, both foundation flexibility and rigid joint zone lengths have
a similar effect on the beam curvature ductility demands. Both of these parameters cause an
increase in beam curvature ductility demand as shown in Fig. 10.6. The rigidjoint zone lengths
increased the beam CUlvature ductility demands by an average of 14.06 % for flexible
foundations and but only 9.54 % for fixed base foundations respectively. This means the effect
of decreasing free vibration periods due to stiffening of the frames is more significant than the
effect of lengthening the free vibration periods of the frames due to the foundation flexibility.
The combinations of the column's base fixity and an increase in rigid joint zone lengths result
in higher beam curvature ductility demands. Foundation flexibility decreased the beam
curvature ductility demand by an average of 17.88 %, 15.21 % and 11.47 % for rigid joint
zone lengths of 0.0,0.5 and 1.0 of the one-half of the column width respectively. The effects
of foundation flexibility on the damage parameters are less significant in the stiffer structures.
As shown in Fig. 10.14, the beam curvature ductility demands over the storey levels
of the 18-storey frame-walls with flexible foundations increase as the rigid joint zone length
increases. In the case of fixed base foundations, the effect of rigid joint lengths decreases the
beam curvature ductility demands in the upper storeys but increases in the lower storeys. Rigid
joint zone lengths increase the beam curvature ductility demand by an average of 13.86 % and
259
10.84 % for the frame-walls on the flexible foundation and fixed bases respectively. These
results are similar with the case of the 12-storey frames discussed previously.
Table 10.2 Response Quantities of the lB-storey Frame-Walli with Fixed Base and Flexible Foundations
Rigid Zone Length

No. Parameters &. Str. Systems Rem.
Z = 0.00 Z = 0.50 Z = 1.00
( 7 Frames + 2 Walls )
a. Fixed Bases 3.196 ( 0.177 N) 3.041 ( 0.169 N ) 2.887 ( 0.160 N ) (-)
b. Flex.Foundations 3.318 ( 0.212 N ) 3.607 ( 0.200 N ) 3.401 ( 0.189 N ) (-)
a. Fixed Bases
1. Pos.Envelope 0.2576 (0.38 % 1ft, 0.2379 (0.35 % 1ft, 0.2580 (0.38 % Ift,) (.)
2. Neg.Envelope 0.2985 (0.44 % 1ft, 0.3067 (0.46 % Hb 0.3186 (O.4B % Ift,) (+)
b. Flex.Foundations
1. Pos.Envelope 0.3058 (0.46 % 1ft, 0.2945 (0.44 % 1ft, 0.2785 (0.42 % Ift,) (-)
2. Neg.Envelope 0.2773 (0041 % 1ft, 0.2987 (0.45 % 1ft, 0.3186 (0048 % Hb ) (+)
a. Fixed Bases
1. Frames + 673.0, - 865.0 + 670.0, - 838.0 + 581.0, - 809.0 (.)
2. Walls + 1648.0, -2294.0 + 1618.0, -2325.0 + 1685.0, -2352.0 (.)
3. Total Base Shear +2321.0, -3159.0 +2280.0, -3163.0 +2266.0, -3161.0 (.)
b. Flex.Foundations
1. Frames . +1163.0, -1175.0 +1257.0, -1301.0 +1367.0, -1434.0 (+)
2. Walls +1362.0, -1388.0 +1300.0, -132.3.0 + 1191.0, -1300.0 (.)
3. Total Base Shear +2525.0, -2563.0 +2557.0, -2601.0 +2558.0, -2734.0 ( +)
4. Maximum CoL Bases P.H.Rot. Fixed, Flex. Fixed, Flex. Fixed, Flex,
(x 0.001 Rad)
1. Left Column 0.000, 1.625 0.000, 2.060 0.000, 2.448 =,+)
2. Internal Column 0.000, 2.268 0.000, 2.620 0.000, 2.791 =,+)
3. Right Column 0.000, 1.130 0.000, 1.980 0.000, 1.942 =,+)
4. Walli 3.570, 0.000 3.030, 0.000 3.165, 0.000 +,=)
5. Tot.P.H.Rots.(xO.OOl Rad) Fixed Flex. Fixed Flex. Fixed Flex.
1. Pos.Envelope 115.330 152.170 126.080 163.700 157.110 179.210 (+) (+)

2. Neg.Envelope 126.590 139.170 135.160 168.690 149.910 198.670 (+) (+)
3. Total 241.920 291.340 261.240 332.390 307.020 377.880 (+) (+)
6. Energy Dissipation (Tm) Fixed Flex. Fixed Flex. Fixed Flex.
1. Viscous Energy 255.550 448.000 313.00 374.00 391.00 329.00 (+)(+)

2. Hyst. Energy 207.400 124.500 209.00 125.00 262.70 127.80 (+) (+)
3. Total Energy 406.950 573.400 522.00 499.00 653.70 456.80 (+) (+)
Remark: (+): Increase, (-) : Decrease, (=) : Equal: (.) : Not clear
10.2.5 The Accumulation of Plastic Hinge Rotations and Energy Dissipation
As shown in Figs. 10.7 and 10.15 the accumulation of plastic hinge rotations of the 12-
storey frames and the 18 storey frame-walls increase as the rigid joint zone length increase.
The total plastic hinge rotations of the 12-storey frames increased by an average of7.09 % and
10.34 % with fixed and flexible foundations respectively. However, the maximum plastic hinge
rotations of stmcture with fixed bases is still greater than those for flexible foundations.
260
In the IS-storey frame-walls, an increase of rigid joint zone lengths results in increased
of plastic hinge rotations by 12.75 % and 13.88 % for fixed bases and flexible foundation
respectively. The plastic hinge rotations of the IS-storey frame-walls with flexible foundations
are greater than for the fixed bases and result in a greater structural damage index such as
discussed in Section 9.5.5. The plastic hinge rotations of the IS-storey frame-walls as given
in Table 10.2 and Fig. 10.15 are smaller than those of the 12-storey frames given in Table 10.1
and Fig. 10.7. Thus, the 12 storey frames expected more damage that than do the IS-storey
frame-walls.
As shown in Table 10.1 and Fig. 10.8, an increase of rigid joint zone lengths of the
12-storey frames increase the total energy absorption by an average of 10.99 % for fixed bases
and 13.26 % for flexible foundations. The rotation of the flexible foundations results in smaller
hysteretic energy absorption than that for fixed base foundations. Fig. 10.16 also shows that
an increase of the rigid joint zone length of the 18-storey frame-walls results in increased of
hysteretic energy absorption for both types of foundation. The viscous energy absorption
increases for fixed bases but decreases for flexible foundations. The combination of foundation
types and the rigid joint zone lengths affect the overall stiffness of structures resulting in
different effects on the viscous energy absorption.
Compared with structures with fixed base foundations as given in Table 10.1, the total
plastic hinge rotations of structure with flexible foundations decreased by 5.5 %, 1.8 % and
0.5 % for rigid joint zone lengths of 0, 0.5 and 1.0 respectively. While the hysteretic energy
dissipation declined by 9.14 %, 9.86 % and 6.8 % respectively. Similar comparisons for the
18-storey frame-walls as shown in Table 10.2, give total plastic hinge rotations increased by
20.4 %, 27.2 % and 23.1 % whereas the hysteretic energy dissipation declined by 40 %, 41
% and 52 % respectively. As presented in a previous chapter, relatively stiff structures with
flexible foundations dissipate much less hysteretic energy than those with fixed base
foundations. Here, the single parameter energy-based concept for describing the structural
damage becomes less meaningfulL15
10.2,6 Storey and Structural Damage Indices
Fig. 10.17 shows that the storey damage indices over the storey levels of the 12-storey
frames with fixed and flexible foundations increase as the rigid joint zone lengths increase.
261
0.35 ~---------"--------,
10
0.3
Square Z = 0.00
8
Circle
Triang.
Z = 0.50
Z = 1.00
.,><
TI
E 0.25
OJ
Filled Fixed
----,. Blank Flex. '"
'"
6
.,
E
Cl
"
::J
l:i 0.2
4 2
Iii
2 +.. ..-}-......;..............+....U 0.15

Square Fixed
Circle Flex.
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.1

Damage Index Z = 0.00 Z = 0.50 Z = 1.00
Fig. 10.11 Storey Damage Indices Fig. 10.18 Structural D. Indices vs. Rigid Joint Length
0.3 ~---------------
10
0.25
Storey
S Square
Circle
Triang. '"
II)
"
. 0.2
Filled '"
tTl
6 ""Slank "E
OJ
a
-.;
...!:i 0.15
4
...2"
en
0.1
Square Fixed.
Circle Flex.
o 0.05 0.1 0.15 0.2 0.25 0.3

0.05
Damage Index
z= 0.00 Z = 0.50 Z 1.00
Fig. 10.19 Storey Damage Indices Fig. 10.20 Structural D. Indices vs. Rigid Joint Length
262
18 ...---~--.,..---.,----
16
o 0.05 0.1 0.15 0.2 0.25 0.3 o 0.02 0.04 0.06 0.08 0.1 0.12
Damage Indel(. Co efficient Lamda
Fig. 10.21 Max. Member Damage Index Fig.10.22 Coefficient lamda
18 0.2 T----------------,
16
Storey
14
12
~------.
>< 0.15
'"
't:I
!iii.
10 ..
01
IE
8 "
Cl
Square
e
~
Z 0.00
6
Circle
Triang.
Z = 0.50
Z = 1.00
g
(I) 0.1
Riled Fixed
4
Flex,
o
o 0.05 0.1 0.15 0.2 0.25 0.05
Damage Index z 0.00 Z = 0.50 Z = 1.00
Fig. 10.23 Storey Damage Indices Fig.1 0.24 Structural Damage Indices vs Rigid Zone Length
263
Increasing the rigid joint lengths leads to an increase in storey damage indices by an average
of8.52 % and 11.30 % and increased structural damage indices by an average of 7.3 % and
8.4 % for structures with fixed bases and flexible foundation respectively. These are the effect
of decreasing structural period of vibrations due to stiffening of the frames. Compared with the
fixed base structures the storey damage indices of frames with flexible foundations declined by
an average of 12.03 %, 7.73 % and 4.72 % for rigid joint zone lengths of Z = 0.0,0.50 and
1. 0 respectively. Fig. 10.18 shows that the 12-storey frames with flexible foundations suffer
less damage than those on fixed bases.
As shown in Figs. 10.19 and 10.20, increasing of rigid joint zone lengths increased the
storey damage indices by an average of 5.11 % and 15.43 % and caused an increase of
structural damage indices by 5.85 % and 14.11 % for 12-storey frame-walls with fixed bases
and flexible foundations respectively. The storey damage indices of structures with flexible
foundations are 12.71 % less than that for fixed base foundations. Comparing the structural
damage indices as shown in Figs. 10.18 and 10.20 implies that with the same earthquake
excitations the framed structures are more severely damaged than the frame-walled structures.
This is because the structural walls effectively control both the member forces and the
deformations .
The effect of rigid joint zone lengths on the member damage indices and weighting
factor A of the 18-storey frame-walls with fixed and flexible foundations are shown in Figs.
10.21 and 10.22 respectively. The distribution of member damage indices are similar to the
distribution of curvature ductility demand as shown in Fig. 10.14. As described in Section
12.6, for strong column and weak: beam failure mode mechanisms in buildings, the weighting
factors A are based on the distribution of member curvature rather than hysteretic energy. The
storey damage indices of the IS-storey frame-walls is shown in Fig. 10.23. As shown in Fig.
10.24, the structural damage indices for fixed base foundations increase as the rigid joint zone
lengths increase. In flexible foundation structures, although the storey damage indices in the
lower storeys increase the structural damage indices decrease as the rigid joint lengths increase.
This is because in the upper storeys where the members suffer more damage, the weighting
factors decrease as the rigid joint zone lengths increase.
Comparing Figs. 10.7 and 10.18 or Figs. 10.15 and 10.20 indicate that the higher total
plastic hinge rotations tend to show more damage. Figs. 10.19 and 10.23 show that the damage
indices in the upper stories of the 12-storey frame-walls tend to decrease but tend to increase
264
in the 18-storey frame-walls. The effect of higher modes seems more significant in the 18-
storey frame-walls than in the 12-storey frame-walls. This effect is also discussed in Ref.Z9.
Table 10.3 Damage Indices of the 12-storey Frame and Frame-Walls and the IS-storey Frame-Walls
Rigid Joint Zone Length

No. Types of Structure and
Damage Parameters Z = 0.00 Z = 0.50 Z = 1.00 Remark
1. The 12-storey Frames
n. Max. Member Damage Indices

1. Fixed Base Foundations 0.3420 0.3340 0.3680 (+)
2. Flexible Foundations 0.3040 0.2790 0.3250 (.)
b. Max. Storey Damage Indices
L Fixed Base Foundations 0.2893 0.2979 0.3109 (+)
c. Structural Damage Indices
2. Flexible Foundations 0.2076 0.2219 0.2443 (+)
a. Max. Member Damage Indices

3. The IS-storey Frame-Walls
a. Max. Member Damage Indices

1. Fixed Base Foundations 0.2310 0.2210 0.2190 (-)
1. Fixed Base Foundations 0.1859 0.1889 0.1857 (.)
2. Flexible Foundations 0.1506 0.1440 0.1371 (-)
Remark: (+) : Increase, (-) : Decrease, (=) : Equal: (.) : Not clear
10,3 EFFECT OF THE NUMBER OF FRAMES IN THE FRAME-WALL

STRUCTURES
Figs. 9.54 and 9.58 show that the structural walls control both the member
deformations and the forces in frame-wall structures. The damage indices of the frame-wall are
less than the framed structures in most storey levels. The effectiveness of the walls in
controlling the structural damage depend of the stiffness ratio between the walls and frames.
265
This ratio is usually indicated by the number of walls and frames within the building structures.
Ref.A6 rep0l1ed that ATC 3-06 recommends at least one wall for every four plane frames
although, in general, it is difficult to establish an optimum ratio.
The effects of combinations between 2-walls and 5-frames, 2-walls and 7-frames and
2-walls and 9 frames in the structural responses are investigated. These lateral load resistance
systems are conveniently denoted by SSI, SS2 and SS3 respectively. The soil-foundation
stiffnesses and damping coefficients were taken at the soil shear strain "Yes = 4xl()4 and were
based on soil shear modulus G = 420 kg/em2 and G = 685 kg/cm2 for the 12 and IS-storey
frame-walls respectively. The structure is excited by the 1940 El Centro N-S component ground
accelerations. The rigid joint zone length ratio Z = 0.5 and the modified Takeda flexural
hysteresis rules are used.
10.3.1 Fundamental Period ami Roof Horizontal Displacements
Each structure possesses 2 walls and only the number of frames is a variables. Tables
10.4 and 10.5 show that as the number of frames increases, the structural fundamental period
of the frame-wall structures with flexible foundations increased by an average of 0.49 % and
0.81 % for the 12 and 18 storey structures respectively. These period shifts are much less than
the effects of parameters discussed in the foregoing sections. As mentioned in Chapter IX, the
soil-foundation to column stiffness ratio is greater than that soil-foundation to wall stiffness
ratio. Hence, the relative soil-foundation stiffness on the Whole structure increases as the
number of frames increases, while the relative structural stiffness decreases with the number
of frames increases. The combination of these effects on the overall structural stiffness result
in only a slightly lengthening of the fundamental periods.
As the number of frames increases Table 10.4 and 10.5 show that the roof
displacements of the 12 and IS-store frame-wall structure with flexible foundations decrease
in both the positive and negative envelopes. More frames reduce the rotation of the wall bases
and result in smaller roof horizontal displacements. This is also affected by the shifting free
vibration periods of the frame-wall structures. As shown in the tables, the roof horizontal
displacements of structures on fixed bases shows a different trend when compared with the
structures on flexible foundations. The differences in the responses of the 12 and IS-storey
frame-walls on the flexible foundations and fixed bases will be discussed further in Sections
10.3.3 and 10.3.4.
266
750
:Square
Circle Wall
Pos.Env.
600
Neg.Env.
j Storey
._... -8..
450
t=
1;;
'"
.r:.
til
Square
OJ 'Circle
m"to Triang. 9F-2.W
300 Diamd. 7F-2W,Fx.
150
o -4500 3000 1500 o 1S00 3000 4500

2W-5F 2W-7F 2W-9F Wall Moment ITm)
Fig. 10.25 Base Shear vs. Number of Frames Fig. 10.26 Wall Moments
12~-------,------~~T.--~~---- 12 ~--""""""""""""-'---------~---------1
Storey
10
..
i'
s

;ft.
r-
/::'6
ei
~
5F2W
7F2.W ~
I:>
6
SF-2W >-
!ll
left Col. ...(;

~
Int. Col. l;; 4

Square: SF2W
Circle : 7F-2W
2. Trlang. : 9F-ZW
Dlamd. : 7F-2.W,Fx.
0.5 0.7S 1.25 1.5 S 10 15 20

Moment Ratie (Fle"L te Fix.). Storey P.H. Rot. (x 0.001 Rad.).
Fig. 10.27 Column Moment Ratios Fig. 10.28 Inter-storey Drift vs. Storey P.Hinge Rot.
267
6 12
Square SF-2W
Circle 7F-2W
Triang. 9F-2W
Diamd. : 7F-2W,Fx. 10
Filled FW18RZ
Blank FW12RZ
.., I!- 8
'"
0
Q
-
a:
<:)
~
0::
6
Q
..."
...'"
0
oc
'"
01
0:: 4
:f 2
Ii
2 4 6 8 10
Left Col. Int. Col. Right Col. Walls Curvature Ductility Demands
Fig. 10.29 Wall & Column's Base P.Hinge Rotations Fig. 10.30 Beam Curvature Ductility Demands
600 1000
Square Pas. Env. Square SF-2W

Circle Neg.Env. Circle 7F-2W
Triang. Tt. Fixed Triang. 9F2W
SOO
Diamd. n. Flex.
800 Filled Filed
Filled Fixed Blank Flex.
Blank Flex.
-c
'"
a:
...
0
0
ci
~
c 300
.2
400
E
!::.
.,"c-
~g
'OJ
.!!l
600
h
!0
0
>-
a: rn
OJ Cii 400
OJ
e w"
5: 200
0.:
200
100
o o
SF-2W 7F-2W 9F-2W SF-2W 7F2W 9F-2W
Fig. 10.31 P.Hinge Rotations vs. Number of Frames Fig. 10.32 Energy Dissipation vs. No. of Frames
268
10.3.2 Base Shears and Member Forces
The base shear resisted by each wall in the 12 and 18-storey frame-walls on both type
of foundations increases as the number of frames increases as shown in Tables 10.4 and 10.5.
The bases shear resisted by each frame in the 12-storey frame-walls on flexible foundations
decreases as the number of frames increases, but tends to increase in the 18-storey frame-walls.
In the case of frame-walls on fixed bases, the base shear resisted by each frame increases in
the 12-storey buildings, but tends to decrease in the 18-storey buildings. When the
superstructure stiffness decreases as the number of frames increases, the total base shear may
decrease. In contrast, the total base shear may increase when the relative soil-foundation
stiffness increases due to increasing the number of frames. As reported in many references, the
effect of higher modes on the structural responses are more significant in relatively tall
structures. The effect of these combinations on base shear, in general, is difficult to generalize.
As shown in the bracket in Table 10. Athe base shear resisted by each wall in the 12-
storey frame-walls on flexible foundations is 3.19, 3.39 and 3.99 times greater than base shear
resisted by each frames for S51, SS2 and SS3 respectively. With a similar structural systems,
a wall in the IS-storey frame-walls resist 3.20, 3.55 and 3041 times greater base shear than that
carried by a frame as shown in Table 10.5. Fig. 10040 shows that the base shear of a frame
in frame-wall structures on fixed bases and flexible foundations respectively is approximately
1.1 and 1.3 times greater than the base shear of a frame in frame structure on the flexible
foundations. Hence, the structural type factor S = 1.0 used for wall base shear in preliminary
design as taken in Section 3.3.2 is an underestimate. The total base shear in the whole frame-
wall buildings on fixed bases and flexible foundations respectively is around 2.5 and 2.0 times
greater than total base shear in the whole framed buildings on flexible foundations. This means
that if the frame-wall structures are assumed to be fixed at their base, the base shear design for
the foundations is 2.5 times greater than for frame's foundations, otherwise the frame-walls will
rock at their foundations.
The wall moments in the 12 and 18 storey frame-wall structures as shown in Fig. 10.26
and 10.35 tend to increase as the number of frames increases. This is because more frames
redistribute the flexural moment to the walls. The ratio of column moments in the structures
with :fIXed bases with respect to the flexible foundation are shown in Figs. 10.27 and 10.36
respectively. In the upper storeys, these ratios tend to decrease as the number of frames
increases for the 12 storey frame-walls and the opposite effect is found for the IS-storey frame-
269
walls. These results may be caused by the complex interaction between walls and frames, type
of foundations, number of frames and the effect of the higher modes. These ratios are always
greater than unity for the columns at the first storey, and similar results as discussed in Chapter
IX. This is the effect of the rotation of the wall bases.
Table 10.4 Resp. Quantities of the 12-storey Frame-Wans with Different Number of Frames
Lateral Load Resistance Systems.

No Parameters & Str.
Systems SS1 (2W + 5 P) SS2 (2 + 7 F) S53 (2 + 9 P) Rem

1. Fixed Bases 2.252 ( 0.18B N ) 2.325 ( 0.194 N ) 2.352 ( 0.196 N ) (+)
2. Flex. FoundatiOn! 2.756 ( 0.229 N ) 2.771 (0.231 N ) 2.783 ( 0.232 N ) (+)
2. Roof Hor. Displ. (m)
L Fixed Bases
a. Pos.Envelope 0.3088 (0.68 % H.,) 0.2982 (0.66 % H.,) 0.3002 (0.66 % H.,) (.)
b. Neg.Envelope 0.2333 (0.52 % H.,) 0.2716 (0.60 % H.,) 0.2722 (0.60 % Hv) (+)
1. Flex. Foundation
a. Pos.Envelope 0.2713 (0.600 % H.,) 0.2691 (0.596 % H.,) 0.2544 (0.56 % H.,) (-)
b. Neg.Envelope 0.3110 (0.689 % H.,) 0.3138 (0.695 % Hv) 0.3012 (0.66 % H b) (-)
a.Fixed Bases
1. Frames +521.7, -505.9 (104.3) +809.0, -780.0 (115.6) + 1043.5, -1141.9 (126.9)
2. Walls +1097.8, -1332.0 (666.0) +1075.0, -1334.0 (667.0) +1076.1, -1421.6 (710.8)
3. Tt.Shear +1619.5, -1837.9 + 1884.0, -2114.0 +2119.6, -2563.5
b. Flex.Found.
1. Frames +678.0, -657.0 (135.6) +912.0, -926.0 (132.3) + 1162.8, -1174.5 (130.5)
2. Walls +645.0, -866.0 (433.0) +588.0, -897.0 (448.5) +632.0, -1042.0 (521.0)
3. Tt.Shear +1323.0, -1523.0 +1500.0, -1823.0 +1794.8, -2216.5
4. Col.Bases Rot. (Flex)
(xO.OOl Rad.)
1. Left Column 3.507 2.896 2.428 (-)
2. Internal Column 3.578 2.773 2.388 (-)
3. Right Column 2.451 2.449 1.303 (-)
4. Walls 0.000 0.000 0.000 (-)
5. P.H.Rots.(xO.OOl Rad, Fixed Flex. Fixed Flex. Fixed Flex.
R. Pos .Envelope 163.06 188.25 188.07 180.05 189.61 156.89

b. Neg.Envelope 124.44 149.94 156.43 148.66 168.18 138.58
c. Total 285.50 338.19 344.50 328.71 357.79 295.47
6. Energy Dissip. (Tm) Fixed Flex. Fixed Flex. Fixed Flex.
a. Viscous Energ 297.81 314.80 306.00 355.00 350.36 402.90

b. Hyst. Energy 267.15 137.10 306.00 172.00 350.36 224.00
c. Total Energy 564.96 451.90 612.00 527.00 700.72 626.90
7. Base Curv.Duct.(Flex)
1. Left Column 2.661 2.277 1.842 (-)
2. Internal Column 2.492 2.064 1.664 (-)
3. Right Column 1.786 1.395 0.990 (-)
4. Wall 0.000 0.000 0.000 (=)
Remark: (+): Increase, (-): Decrease, (=): Equal: (.) : Not clear
270
50 ,----------------------------------, 1000 ,----------------------
40 noo
E
2-
... 30
""E
600
Q)
E~
"'"
ii '"
OJ
II)
is
""
(Jj
b
""'"
-
0 00
J: 20 400 Square Frame
c Circle Wall
0
a:
Filled Pos.Env.
Blank Neg.Env.
Square Pos.Env
Circle Neg.Env.
10 200
Filled Fixed
Blank Flex.
o +--------+--------~------~------__" o +-------+-------~------~----~
SF-2W 7F-2W SF-2W SF-2W 7F-2W SF-2W
Fig. 10.33 Roof Hor. Displ. vs. Number of Frames Fig. 10.34 Base ShearS vs. Number of Frames
16+-----------+---------1 l--+-------A
14 +------------+------------..
12 +------------.. -+
10
5F2W
7F-2W
SF-2W
Right Col.
Int. Col.
O~~~~~~~~~~~~~~~~~
-15000 -10000 -5000 o 5000 10000 15000 0.25 0.5 0.75 1.25 1.5 1.75
Wall Moment (Tm). Moment Ratio (Flex. to Fix.l
Fig. 10.35 Wall Moments Fig. 10.36 Column Moment Ratios

271
10 600 ~I~"-------------------------------,
o 500
Filled FiHlld
i' Blank flex.
"j
~
...ci
.!!
6 i -...".'''.'''.... ~~-f(" ..
III
a:
.::.
<:)
400
ci
~
c 300
'g
~
..
:!'" ZOO
a.:
2 --",","""C::::::--:-~;=;:--1:::.",i","'"'
Square 2W-5F .. ,,,,.. ,1
Circle 2W-7F
Trhmg. 2W-9F 100
,O!amd. 2W-7F.Fx.
o I) 10 15 20
Storey P.H.Rot.lx 0.001 Rad.l.
Sf-2W 7F5W 9F2W
Fig. 10.37 Inter-storey D,ifis vs. Storey !'.Hinge Rotations Fig. 10.38 P.H.Rotations vs. Number of Frames
1000 3.2
3
'Square Visco
Circle Hyat. 2.B
Trillng. n.En.
BOO 2.6
Filled Fixed
Blank Flail. i
II: 2.4
~ Total Boso Sheaf
><
J:: 2.2
E III
t:: .,
~
III
600
c 2
0 "
!XI
~
..
,9-
is
III
w 1.8
E
....f!
>- 1.6
E'
S
400 .If!
...
'"c: Ii 1.4
3j:.,
200
EIII
.t
1.2
~~
Frome Balle Shear
0.8 -t
0.6 .t
0 0.4
5F2W 7F2W 5F~2W 7F2W 9F-2W

9F-2W
Fig. 10.39 Energy Dissipations vs. Number of Frames Fig. 10.40 Bese Shear Ratio vs. Number of frames
272
Table 10.5 Resp Quantities of the 18-storey Frame-Walls with Different Number of Frames
Lateral Load Resistance Systems
N Parameters & Str. SSI (2 W + 5 F) SS2(2W+7F) SS3 (2 W + 9 F) lKem

0 Systems
1. Fixed Bases 2.945 ( 0.164 N ) 3.041 ( 0.169 N ) 3.109 (0.173 N) (+)
2. Flex.Foundations 3.573 (0.199 N ) 3.607 ( 0.200 N) 3.632 (0.202 N ) (+)
2. Roof Hor.Displ.(m)
1. Fixed Bases
a. Pos.Envelope 0.2248 (0.335 % ~) 0.2419 (0.361 % Hb) 0.2557 (0.381 % ~) (+)
b. Neg.Envelope 0.3171 (0.473 % HJ 0.3001 (0.447 % Hb) 0.2715 (0.405 % ~) (-)
2. Flex. Foundation
a. Pos.Envelope 0.2902 (0.432 % 1\) 0.2945 (0.439 % HJ 0.2900 (0.432 % Rt,) (-)
b. Neg.Envelope 0.3138 (0.468 % 1\) 0.2987 (0.445 % lIt,) 0.2323 (0.346 % Rt,) (-)
a. Fixed Bases
I. Frames +765.3, -485.1 (153.1) +637.5, -970.2 (138.6) + 1552.0, -1260,0 (140.0)
2. Walls + 1576.1, -2033.0 (1016.5) + 1557.9, -2294.8 (1147.4) +2021.0, -2664.0 (1332.0)
3. Tt.Shear +2341.4, -2518.1 +2195.4, -3265.0 +3573.0, -3924.0
b. Flex.Found.
1. Frames + 732.0, 814.5 (162.9) + 1257.0, -1301.0 (185.9) +1623.6, -1656.0 (184.0)
2. Walls +945.8, -1045.0 (522.5) + 1300.0, -1323.0 (661.5) +1503.6, -1255.0 (627.5)
3. Tt.Shear +1677.8, -1859.5 +2557.0, -2623.0 +3127.2, -2911.0
4. Max.P.RRot. (Flex)
(x 0.001 Rad.)
1. Left Column 0.000 2.060 1.943 (+)
2. Internal Column 1.497 2.620 3.391 (+)
3. Right Column 0.000 1.980 2.623 (+)
4. Walls 0.000 0.000 0.000 (=)
5. P.H.Rots.(O.OOIRad) Fixed Flex. Fixed Flex. Fixed Flex.
a. Pos.Envelope 129.44 117.110 132.92 163.700 167.62 164.600

b. Neg.Envelope 198.62 122.490 142.95 168.690 167.76 184.310
c. Total 328.06 239.600 275.87 332.390 335.38 348.910
6. Energy Dissip. (Tm) Fixed Flex. Fixed Flex. Fixed Flex.
a. Viscous Energy 204.38 217.160 306.67 385.00 350.36 394.850

b. Hyst. Energy 193.43 44.780 226.67 138.00 344.53 176.460
c. Total Energy 397.81 261.940 533.33 523.00 694.89 571.320
7. Column Bases
Curv.Ductility Dem.
1. Left Column 0.000 1.567 1.366 (.)
2. Internal Column 1.723 2.548 2.925 (+)
3. Right Column 0.000 1.182 1.845 (+)
4. Wall 0.000 0.000 0.000 (=)
10.3.3 Plastic Hinge Rotations and Beam Curvature Ductility Demands
As shown in Figs. 10.37 the variation between the maximum inter-storey drifts and the
storey plastic hinge rotations for the 18-storey frame-walls with flexible foundations is stronger
than that the for 12-storey frame-walls as shown in Figs. 10.28. Using only a single
273
excitation, it is difficult to generalize this variation. Inter-storey drifts of the frame-wall

buildings on the flexible foundations tend to be greater than those on fixed bases as shown in
Figs. 10.12 and 10.37. A more clear relationship between the inter-storey drifts and the plastic
hinge rotations as shown in Figs. 12.4 and 12.12. The effect of the number of frames on the
column base and wall base plastic hinge rotations is depicted in Fig. 10.29. These rotations
decrease as the number of frames increases for the i2-storey frame-walls. The rotation of the
walls may decrease when the number of frames increases and result in smaller plastic hinge
rotations. In the case of the IS-storey frame-walls, an opposite result was found. This is may
due to the complex combinations of soil-foundation. superstructure stiffness and effect of higher
modes as mentioned earlier.
The beam curvature ductility demand of the 12-storey frame-walls for fixed and flexible
foundations is shown in Fig. 10.30. For structures with flexible foundations, the effect of the
number of frames on the beam curvature ductility demands does not indicate a clear
relationship in the upper storeys but tends to increase in the lower storeys. In the case of
structures with fixed base foundations, the number of frames causes an increase in the beam
curvature ductility demands in most storeys for the 12-storey frame-walls and a decrease for
the 1S-storey frame-walls. For the is-storey frame-walls with flexible foundations, the
relationship between the number of frames and the beam curvature ductility demands is not
clear. The combination of effects as mentioned above causes a more complex structural
responses.
10.3.4 Total Plastic Hinge Rotations and Energy Dissipation
As discussed earlier the total plastic hinge rotations have a strong correlation with the
structural damage indices. This relationship will become more clear in Sections 12.7.7. Fig.
10.31 shows that the total plastic hinge rotations increase as the number of frames increases
for the 12-storey frame-walls with fixed base foundations but decreases for structures with
flexible foundations. In Fig. 10.38 the relationship is not clear for the IS-storey frame-walls
with fixed base foundations and increases for the flexible foundations. These complex
relationships may due to cross effects among relative soil-foundation, superstructure stiffness
and higher mode effects.
As shown in Figs. 10.32 and 10.39, the viscous and hysteresis energy dissipations
increase as the number of frames increases for both types of foundation. The viscous energy
274
dissipation of structures on flexible foundations are greater than that for fixed based
foundations. This is because the rotation of the structure due to the foundation flexibility. In
contrast, the hysteretic energy dissipations of structures with flexible foundations is much less
than that for fixed base structures as given in Table. 10.4. The 12-storey wall-frames with
flexible foundations dissipate 48.7 %, 43.8 % and 36.1 % hysteretic energy less than do for
the flXed base foundations for SSI, SS2 and SS3 respectively. A similar comparison for the
IS-storey wall-frames, the hysteretic energy dissipations decreased by 76.1 %,39.0 % and
48.8 % respectively. Considering the total plastic hinge rotation and the hysteretic energy
dissipation, as discussed in Section 10.2.5, it is clear that the use of the energy-based concept
as a single parameter in describing the damage indices of rocking structures becomes less
meaningful.
Figs. 10041 indicates that patterns storey damage indices ofthe 12-storey frame-walls
are similar to the curvature distributions over the building height as shown in Fig. 10.30. The
member damage index is therefore, strong correlated with the member curvature. The effect
of the number of frames on the storey damage indices does not show a consistent relationship.
This is similar to its effect on the curvature ductility demand as discussed in Section 10.3.3.
As shown in Table 10.6 and Fig. 10041, however, the maximum storey and structural damage
indices decrease as the number of frames increases for structures with flexible foundations.
This is because more frames reduce the rotation of the walls and the effects of higher modes
are insignificant. In the case of fixed base foundations, it shows that the maximum storey and
structural damage indices increase as the number of frames increases. As the number of frames
increase, the effectiveness of structural walls in controlling both member deformations and
member forces decreases. As a result, the structural damage indices decrease as the number
of frame increases for structures with flexible foundations but increases for structures with
fixed base foundations as shown in Fig. 10.44.
As given in Table 10.6 and Fig. 10.42, the maximum storey and structural damage
indices increase with the number of stories for the 18-storey frame-walls with flexible
foundations. These results are in opposition to the case of the 12-storey wall-frames. This is
because the effect of higher modes are more dominant than is the effectiveness of frames in
preventing the rotation of the structural walls. In the case of fixed base foundations, the
maximum storey and structural damage indices tend to decrease as the number of walls
275
18
16
10 +......l...... ...J\
14
Store V
8 + ....................;.......................+.......................:1'1\ 12
10
6 +............ .. +...... ...... ....i.... ...... Ui..iVi
11
6 Square SF-2W
5F-2W Circle 7F-2W
7f-2W Triang. Sf-2W
Sf-2W 4
Filled Fixed
2 fixed
Flex.
2
O~,~~~~~~~~~~~~~~~~~~
0.05 0.1 0.15 0.2 0.25 0.3 o 0.05 0.1 0.15 0.2 0.25 0.3
Damage Index Damage Index
Fig. 10.41 Storey Damage Indices Fig. 10.42 Storey Damage Indices
500 0.25 . , - - - - - - - - - - - - - - - - - - ,
0.2
400
"C
...'"
1:1: ...
0 '"
.5 0.15
0
c:i .
!5 '"'E"
c:: 300
'"c::.....
0
'"
0
0
11
iu
ill 0.1
'"
c:: '"
,I:;
(J)
:i:
a: Square fW12RZ
I~'"'"
200 FW12AZ Circle FW10RZ
Circle FW18AZ
0.05 Filled : Axed
Filled : Fixed Blank : Flex.
Blank : Flex.
100 0
SF-2W 7F .. 2W 9F-2W 5F2W 7F2W SF-2W
Fig. 10.43 Plastic Hinge Rot. vs. No. of Frames Fig.l0.44 Structural Damage Indices vs. No. of !'rames
276
increases. A more detailed study on the effect of the number of frames on the structural
response of frame-wall structures is required. Comparison between Figs. 10.43 and 10.44
shows that the structural damage indices have similar patterns to the total plastic hinge
rotations. This is further evidence that the structural damage indices are strongly correlated
with the member curvatures.
Table ]0.6 Damage Indices of the 12 and IS-storey Frame-Walls with Different Number of Frames
Lateral Load Resistance Systems

Damage Parameters 2W+5F (SSI) 2W+7F (SS2) 2W+9F (SS3) Remark
L The 12-storey Frame-Walls

1. Fixed Bases
a. Max. Member Damage Indi~ 0.3010 0.3200 0.3480 (+)
b. Max. Storey Damage Indices 0.2369 0.2694 0.2862 (+)
c. Structural Damage Indices 0.1669 0.1948 0.1981 (+)
2. Flex. Foundations
Max. Member Damage Indice~ 0.2570 0.2670 0.2420 (.)
b. Max. Storey Damage Indices 0.2330 0.2313 0.2111 (-)
c. Structural Damage Indices 0.1715 0.1700 0.1579 (-)
2. The IS-storey Frame-Walls

1. Fixed Bases
a. Max. Member Damage Indices 0.2550 0.2210 0.2360 (.)
b. Max. Storey Damage Indices 0.1896 0.1889 0.1840 (-)
c. Structural Damage Indices 0.1215 0.1167 0.1169 (.)
a. Max. Member Damage Indices 0.2080 0.2530 0.2710 (+)
10.4 EFFECTS OF DIFFERENT HYSTERESIS RULES IN FLEXURE
The effects of four different flexural hysteretic rules on the structural responses of the
12-storey frames and the 12 and 18-storey frame-walls subjected to the 1940 El Centro N-S
component earthquake are investigated. The hysteretic rules include the Bilinear, Clough,
Modified Takeda and the Q-Hyst rules. The bilinear factor of 0.06 and 0.03 respectively for
column bases and beams are used. The coefficient fJ for the unloading stiffness in Eq. 6.2 was
0.5 for the modified Takeda rules and 0.4 for Q-Hyst rules. These coefficients and the
corresponding reloading stiffnesses are used to control the fatness of the hysteretic rules.
277
1004.1 Base Shear
The foundation flexibility causes a decrease in the base shear by an average of 0.67 %,
7.62 % and 5.83 % for the 12-storey frames and the 12 and lS-storey frame-wans as shown
in Figs. 10.45, 10.54 and 10.66 or as given in Tables 10.7, 10.8 and 10.9 respectively for the
variation in the hysteresis rules. This is evidence that the effects of foundation flexibility on
the structural responses is more significant in relatively stiff structures than that for the more
flexible structures. The Clough, Modified Takeda and Q-Hyst models show a close agreement
in the structural responses. As shown in Figs. 10.45 and 10.54 the Bilinear rules gives a
significantly different total base shear.
Table 10.7 Response Quantities of the 12-storey Frames with Different Hysteresis Rules
Hysteresis Rules
No Response
Bilinear Clough Mod.Takeda Q-Hyst.
Quantities
1. Roof Hor. Displ.(m)
a. Fixed Bases +0.159, -0.264 +0.149, -0.269 +0.187, -0.299 +0.173, -0.278
b. Flex.Foundations +0.174, -0.275 +0.152, -0.274 +0.190, -0.301 +0.180, -0.289
1. Fixed Bases
n. Pos.Env. 116.90 103.20 104.00 101.50
b. Neg.Env. 94.90 87.10 94.80 92.80
a. Pos.Env. 115.80 102.50 102.50 102.50
b. Neg.Env. 94.90 86.60 94.20 92.20
3. Base Rot.(xO.001 Rad.) Fixed Flex. Fixed Flex. Fixed Flex. Fixed Flex.
u. Left Column 3.170 2.830 1.630 1.480 1.790 1.510 1.920 1.580
b. Int. Column 3.900 3.500 2.280 1.720 2.600 2.200 2.640 2.160
c. Right Column 3.140 2.600 1.660 1.290 2.130 1.490 2.500 1.459
4. Tt.P.H.Rot.(O.OOl Rad) Fixed Flex. Fixed Flex. Fixed Flex. Fixed Flex.
a. Pos.Env. 153.194, 154.236 134.152, 133.463 159.704, 160.099 152.510, 153.619
b. Neg.Env. 191.236, 191.564 177.085, 173.193 188.244, 181.722 184.155, 176.331
c. Total 344.430, 346.251 311.237, 306.656 347.948,341.821 336.665, 329.95(]
5. Energy Dissip. (Tm) Fixed Flex. Fixed Flex. Fixed Flex. Fixed Flex.
a. Viscous En. 24.90, 29.90 18.70, 19.20 21.60, 21.80 21.00, 20.50
b. Hyst En. 27.70, 25.00 28.30, 25.70 21.40, 19.20 24.50, 21.00
c. Total En. 52.60, 54.90 47.00, 44.90 43.00, 41.00 45.50, 41.50
6. Curv.DuctDemand Fixed Flex. Fixed Flex. Fixed Flex. Fixed Flex.
a. Left Column 2.640 2.356 1.357 1.225 1.491 1.259 1.597 1.318
b. Int. Column 2.585 2.320 1.507 1.268 1.723 1.457 1.747 1.431
c. Right Column 2.608 2.162 1.387 1.067 1.929 1.242 1.793 1.208
Remark: -
10.4.2 Member Forces
The storey shears of the structures with flexible foundations are normalized with respect
to that of the case fixed bases for both positive and negative envelopes. The storey shear of the
278
175 12~-------,-----UKU~r------'--------,
150
Storey
125 -
~
100
~
'"
III
..c
r.n
OJ
75
''""
m
50
Square Bllin.
Circle Clough
Square : Pos.Env. 2 Triang. Takeda
Circle : Neg.Env. Diamd. O-Hyst.
25 Filled : Fixed
Blank : Flex.
o +-----_+------~----~----_+----~ 0.5 0.75 1.25 1.5

BHin. Clough Tak. O-Hyst. Shear Ratio (Fle)(. to Fix .. Takeda J
Fig. 10.45 Base Shear vs. Fle)(. Hysteresis Rules Fig. 10.46 Storey Shear Ratios
12 12
FIx.
Storey 0 Flex.
10
10
I
Square: Takeda
3 . Circle : Bilin.
Triang. : Clough
Diamd : O-Hyst :c 8
'if'.
6 ci
~
, .t;:
;5 6
4 ......... _.....1.
o
o 5 10 15 20 o 5 10 15 20 25
Drift {)( 0.1 % HI, P.H. Rot.(x 0.001 Rad.l. Storey P. H. Rot. {x 0.001 Rad.l.
Fig. 10.47 Inter-storey Drift & Sorey P .Hinge Rot. Fig. 10.48 Inter-s'lorey Drift Vs. Storey P .Hinge Rot.
279
5 12
Square SHin
Circl .. Clough
Triang. Takeda
Diamd QHvst. 10
4
I.Filled , Fixed
Blank : Flex.
Square BlIin.
-t:.i B
l\'l Circle Clough
IX:
Triang . Takeda
....
0 3 Diamd. Q-Hyst.
0
0 Filled Fixed
~ 6 . Blank Flex .
..:Cl
IX:
(!)
en 2
l:
J: 4
0:
Storey
o
left Col. Int. Col. Right Col.
Curvature Ductility Demands
Fig. 10.49 Column's Base PI. Hinge Rotations Fig. 10.50 Beam Curvature Ductility Demands
275 80
~
250
Square
Circle
FiWed
Pos.Env.
Neg.Env.
, Fixed
70 1 ...
Square
Circle
Vise.En.
Hyst.En.
Triang. : Total En.
Blank : FleK. Filled Fixed

60 Blank Rex.
225
.,
-.;i
a::
0<:> E 50
!:.
0 200 c
.!! .~
~>~-
c: .e.
0
';l
., 40
.s
Q ~V~
..
is
a: 175 >-
III
!!'
III
rn c 30
.S: w
/~
::t
~
r
0:: 150
/
125
V 20
10
J Bilin. Clough Takeda QHyst.

0
BlIIn. Clough Takeda Q-Hyst.
Fig. 10.51 Plastic H.Rotations vs. Flex. Hvst. Rules Fig. 10.52 Energy Dissipations vs. Flex. Hyst. Rules
280
12-storey frames with flexible foundations tend be smaller than the fixed base case for all
hysteresis rules except for the Bilinear model as shown in Fig. 10.46. As shown in Figs. 10.55
and 10.67, the frame storey shears of the 12 and I8-storey frame-walls with flexible
foundations are greater than that for fixed base foundations for all hysteresis rules. This is
caused by reduced wall shear as shown in Figs. 10.56 and 10.68 respectively. The capacity of
structural walls to resist storey shears decreases when the wall footings rotate. Similar results
were reported by GOODSIR (1985).
Effects of higher modes and frame-wall interactions on the column and wall moments
of structuces with flexible foundations are more evident in the 18-storey frame-walls a shown
in Figs. 10.69 and 10.70 than they are in the 12-storey frame-walls shown in Figs. 10.57 and
10.58 respectively. As in Chapter IX, these effects cause an increase in column moments at
the fiTSt and the upper stoceys and increased wall moment in the middle storeys. It can be seen
that the Clough, Modified Takeda and Q-Hyst models give very similac responses.
Table 10.8 Response Quantities of the 12-storey Frame-Walls with Different Hysteresis Rules
Hysteresis Rules
No Response
Bilinear Clough Mod.Takeda Q-Hyst
Quantities
a. Fixed Bases +0.303, -0.248 +0.284, -0.251 +0.296, -0.268 +0.292, -0.257
b. Flex.Foundations +0.284, -0.304 +0.232, -0.302 +0.269, -0.314 +0.267, -0.312
2. Tot.Base Shear (ronne)
1. Fixed Bases
8, Frames +659.0, -753.0 +671.0, -718.0 +809.0, -780.0 +753.0, -776.0
b. Walls + 1092.0, -1346.0 + 1081.0, -1335.0 + 1075.0, -1334.0 + 1081.0, -1379.0
c. Total +1751.0 -2099.0 +1752.0, -2053.0 + 1884.0, -2114.0 + 1834.0, -2155.0
a. Frames +956.0, -941.0 +918.0, -905.0 +912.0, -926.0 +919.0, -927.0
b. Walls +697.0, -827.0 +618.0, -897.0 +588.0, -897.0 +600.0, -904.0
c. Total + 1653.0, -1768.0 +1536.0, -1802.0 +1500.0, -1823.0 +1519.0, -1831.0
3. P.H.Rot.(x 0.001 Rad.) Fixed Flex. Fixed Flex. Fixed Flex. Fixed Flex.
a. Left Column 0.000 2.780 0.000 2.626 1.080 2.855 0.000 2.816
b. Int. Column 1.463 3.108 0.000 2.717 1.507 2.940 1.512 2.808
c. Right Column 1.532 2.103 0.000 1.884 1.375 2.322 1.481 2.281
d. Wall 4.385 0.000 4.150 0.000 4.837 0.000 4.555 0.000
4. Tt.P.H.Rot,(0.001 Rad) Fixed Flex. Fixed Flex. Fixed Flex. Fixed Flex.
a. Pos.Env. 183.390, 178.670 166.620, 133.920 184.790, 147.640 178.120, 142.800
b. Neg.Env. 136.280,179.660 138.070, 170.140 152.180, 179.120 148.270,175.760
c. Total 319.670,358.330 304.690, 304.060 336.970, 326.760 326.390,318.560
a. Viscous En. 248.0, 344.0 240.0, 304.0 306.0, 355.0 278.0, 334.0
b. Hyst. En. 235.0, 193.0 313.0, 238.0 306.0, 172.0 313.0, 198.0
c. Total 483.0, 537.0 553.0, 542.0 612.0, 527.0 591.0, 532.0
Remark; -
281
40 3000
[G420, FW12RZ
LEI Centro NSC
35 2500
E
~
1'ru: 30 2000
E E
'"" ~
.!l
a. 'ru"
..c
is IJ'I
ru
III
~ m"
J: 25 1500
....0
0
n::
Square: Pos.Enll.
Square Pos.Enll. Circle Neg.EolI.
20 Circle Neg.Enll. 1000
Ailed Fixed
Filled Fixed Blank Flex.
Blank Flex.
15
Bllin. Clough Takeda QHyst. Bllin. Clough Tal<.
Fig. 10.53 Ruuf Hm. Displ. vs. Flex. Hysteresis Rules Fig. 10.54 Base Shears vs. FJe)(ural Hysteresis Rules
6 6
Square Bllin.
4 Circle 4
Triang. Bllin.
Dlamd. Clough
Triang. Takeda
Oiamd. QHyst.
2 2
0.5 0.75 1.25

Shear Ratio (Flex. to Fix. Takeda) Shear Ratio (Flex.to Fix. Takedal
Fig. 10.55 Storey Shear Ratio of the Frames Fig. 10.56 Storey Shear Ratio of the Walls
282
12~--------~----~~--TC~------,
10 10
Square
8 8 - Circle
Triang.
Diamd.
Filled
Ii 6 - Blank
Storey
Storay
4 4
Bilin.
2
Clough
Triang. Takeda
Dlamd. Q-Hyst.
O~~~~~-4~--~~~+-~~~~- O~~~~~~~_r~~~~~~~~~~~
0.5 0.75 1.25 0.25 0.5 0.75 1.25 1.5

Moment Ratio (FleK. to Fix. Takeda) Moment Ratio I Flex. to Fill.1
Fig. 10.57 Wall Moment Ratios (End 1) Fig.10.58 Column Moment Ratios (End lJ
12 ,-----------,----------,-----------
Floors
10 ------------------------UII.-----;f----------------------------------H:1\l---..---.
I
10 ...._._ ........._._. __._.--t.-.-.---..-.-.------ ...-.-l. . . . . . _. .~. _._._.~_ . _
..
I
8
8

'* 0
15 ci
0
Drift .!:!.
:t
is 6
4
8l1in.
Clough
TriBng. Q-Hy.t. o Flex.
: Takeda 4 +-....--....-.......... i..........------..-..--..---t1
2 Fill.
5 10 15 20 5 10 15 20
Drift Ix 0.1 % H), P.H. Rot. Ill. 0.001 Rad.) Storey P. H. Rotations Ix 0.001 Rad.)
Fig. 10.59 Inter-Storey Drift &. Storey P.H. Rot. (Flex.) Fig. 10.60 Inter-storey Drift liS. P.Hinge Rotations
283
The Bilinear model with no variation in unloading and reloading slopes gives different
structural responses as shown in Figs. 10.46, 10.55, 10.58, 10.67 and 10.69 respectively.
10.4.3 Inter-storey Drifts and Plastic Hinge Rotations
The hysteresis rules do not significantly affect the inter-storey drifts and storey plastic
hinge rotations of the 12~storey frames and the 12 and IS-storey frame-walls as shown in Figs.
10.47, 10.59 and 10.71 respectively. The maximum inter-storey drifts are less than 1 % of
storey height. In general, the effect of foundation flexibility is more significant than the effect
of different flexural hysteresis models. The variation of storey drift and plastic hinge rotations
of these structure are shown in Figs. 10.48, 10.60 and 10.72 respectively. Although the
relationship is defuse, in general, the storey plastic binge rotation increases as the storey drift
increases. Figs 11.60 and 10.72 show that greater inter-storey drifts of the frame-walls on the
flexible foundations than those on fixed bases is more obvious in the 18-storey frame-walls than
that for the 12-storey frame-walls. This is an evidence that the effect of inertial interaction is
more significant in the more slender structures.
10.4.4 Base Plastic Hinge Rotations and Beam Curvature Ductility Demands
As shown in Figs. 10.49, 10.61 and 10.73 and Tables 10.7, 10.8 and 10.9, the effects
of the hysteresis rules on the columns and wall base plastic hinge rotations are less significant
than effect of foundation flexibility. The effect of the foundation flexibility on the columns and
wall base moment-curvature relationships of the 12-storey frame-walls for the hysteresis rules
are shown in Figs. 10.77 through 10.84 respectively. The lateral forces in structures with fixed
base foundations are mainly resisted by the walls and result in high yield cycles in the wans
but many columns remain elastic. It is difficult to prevent the structural walls from showing
foundation rotation. As a result high yield cycles occur in the column bases while the structural
walls remain elastic.
The member curvature ductility demand of the 12-storey frames and frame-walls are
shown in Figs. 10.50 and 10.62 respectively. It is clear that the effects of foundation flexibility
is more significant than the choice of hysteresis rules. The beam curvature ductility demand
of stmcture with flexible foundations decreased by an average of 14.88 and 8.96 % for the 12-
storey frames and frame-walls respectively with the variation in hysteresis rules.
284
7
12
0420. FW12RZ Storey
EI Centro NSC
5
10
Square Bilin.
Circle Clough
Triang. Tllkeda
5 Diamd QHyst. Square Billn.
Circle Clough
1:1
u Triang. Taked"
ro Ailed Fh<ed
!I! Diamd. Q-Hyst.
....
0
Blank Flex.
0 4
ci Fillad
..
..!:!.
0
rr:
6 -Blank
:3
'"
t>l
.5
J: 4
ii:
2
o +-----__r-----e-----__ ----~~----~ 2 4 6 8 10
left Col. Int. Col. Right Col. Wall Curvature Ductility Demands
Fig.l0.51 Column &. Wall Bases P. Hinge Rotations Fig. 10.62 Beam Curvature Ductility Demands
250 ~------------------------------------ 700 ,--------------------------------,
600
200
500
a:..
-ei
E
t:.
;; 0::
400
~
0
ci
"I>.
~
..0
150 1ii
.!2
C
a:
III
,.. 300
'"
.!: ~
c
J: w
a::
200
100
Square: Vise. En.
Square : -Neg.Env. Circle : Hyst. En.
Circle : Pas. Enl/. 100 Triang. : Total En.
Filled Fixed
Blank : FleK. Filled FiKed
Blank FleK.
50
Bilin. Clough Takeda Q-Hyst. Bilin. Clough Takeda O-Hyst.
Fig. 10.63 PI. Hinge Rotations vs. Flex. Hyst. Hules Fig. 10.64 Energy Dissipation \IS. Flex. Hyst. Rules
285
40 - - - - - -.... -~
4000
3500
35
15 1000 +-----+ - - - t - - - + - - - - l - - - - - - i
BHin. Clough Takeda Q-Hyst. 8i1in Clough Takeda QHyst.
Fig. 10.65 Roof Hor. Displ. liS. Flex. Hyst. Rules Fig. 10.66 Total Base Shears liS Flex. Hyst. Rules
1B~--,----.~I~,c~--r----.----, 18
16 16
14+.. ....+.... 14
Storey
12
10 + .................,... _...... 10
8 8Hin. 8 8Hin.
Clough Clough
Takeda Takeda
6 "\1" ................. ,............ , QHyst. 6 QHyst.
4 + ....... ..f...... 4
0
0.5 0.75 1.25 1.5 1.75 2 0.5 0.75 1 1.25 1.5
Shea, Ratio (Flex. to Fix. Tak.l. Shear Ratio (Flex. to Fix. Tak.l.
Fig. 10.67 Storey Shear l'Iatios of the Frames Fig. 10.68 Storey Shear Ratios of the Walls
286
16 cI.-----'------,. . .1.."'<!:},,j-;JDm._: 13665. FW18RZ

Square: BlIln. 16 ..... EI C~ntro NSC
Circle : Clough
TOling.: Takeda
14 Oiamd : QHyst
R.Col.H
12 Blank Int.Col.1 +)
10
0.6 0.76 1.26 1.6 0.4 0.6 0.8 1.2 1.4

Moment Ratio IFle)!.. io Fix. Tak.) Moment R~lio (Flex. to Fix. Tok.1
Fig. 10.69 Column Moment Ratios Fig. 10.10 Wall Moment Ratios
10 .,----,---~,..-......--....,.-----,
16 + ...................+..... _...
14
12 ......_,
Bllin.
Clough
Ta.keda
QHvst.
41-
2+_....;
o 4 12 15 :w 2 6 10 14 18
Drifts i I( 0.1 % H I, St.P.H. Rot.lx 0.001 RIld.J Storey P.H. Rot. Ix 0.001 Had.1
Fig. 10.71 Interstorey Drifts and Storey P.Hlnga Rot. Fig. 10.72 Inter-storey Drifts vs. Storey P. Hinge Rot.
287
6 200
{';
0
5 !ill ~"J 0 5
-o:i
Square
Circle
Triang.
Bflin.
Clough
Takeda
175
.. ~ Ai)
C\1 Diamd Q-Hyst.

...
tr.
0
/I
Filled Fixed ." 0
{'; ~
0
Blank Flex. "'
0: 150 Cl
0
:;
.!:> ~ \~
..,
[.
I:
0 3 '"
....'" 50
0 U; 125
OC :r
III
OJ ..:
I: Square 8i1in.
2
:E Circle Clough
li: Triang. Takeda
100 ....... " ......... _..... Diamd. Q-Hyst. . ............ ,._._........
1 Filled FW18RZ
Blank FW12RZ
75
0 20 25 30 35
left Col. Int. Col. Right Col. Wall Displacement (em)
Fig. 10.73 Column and Wall Bases PI. Hinge Rotations Fig. 10.74 Roof Hor. Displacements vs. P.H. Rot.
Square Pos.Env.
Circle Neg.Env.
Filled Fixed
200 Blank Flex.
..
.,,;
0:
~
0
0
ci
3
.,c
0
....tI10
0:
OJ 150
en
.5
:r
0:
100
8i1in Clough Takeda Q-Hyst.
Fig. 10.75 Total P. Hinge Rot. vs Flex. Hyst. Rules Fig. 10.16 Energy Dissipations vs. Flex. Hyst. Rules
288
10.4.5 Total Plastic Hinge Rotations and Energy Dissipation
The variation of the total plastic hinge rotations with hysteresis rules of the 12-storey
frame-walls as shown in Fig. 10.63 and in general, are similar to the variation of roof
horizontal displacements as shown in Fig. 10.53. The foundation flexibility reduces the total
plastic hinge rotations except for the IS-storey frame-walls as shown in Fig. 10. 75. This
exceptional result is similar to that discussed in Chapter IX. The relationship between the roof
horizontal displacement and the total plastic hinge rotations is shown in Fig. 10.74. The plastic
hinge rotations tend to increase as the roof displacement increases through a large amount of
scatter is evident.
Table 10.9 Response Quantities of the is-storey Frame-Walls with Different Hysteresis Rules
Hysteresis Rules
No Response
Bilinear Clough Mod.Takeda Q-Hyst
Quantities
a. Fixed Bases +0.254, -0.301 +0.240, -0.304 +0.238, -0.307 +0.238, -0.308
b. FleX.Foundations +0.282, -0.271 +0.289, -0.285 +0.295, -0.299 +0.269, -0.289
1. Fixed Bases
a. Frames +639.0, - 810.0 +669.0, - 890.0 +670.0, - 838.0 + 728.0, - 912.0
b. Walls +1617.0, -2334.0 + 1581.0, -2298.0 + 1618.0, -2325.0 + 1580.0, -2279.0
c. Total +2256.0, -3144.0 +2250.0, -3188.0 +2288.0, -3163.0 +2308.0, -3191.0
a. Frames + 1268.0, -1313.0 +1235.0, -1301.0 + 1257.0, -1301.0 +1264.0, -1307.0
b. Walls +1352.0, -1397.0 + 1264.0, -1309.0 +1300.0, -1323.0 +1286.0, -1324.0
c. Total +2620.0, -2710.0 +2499.0, -2610.0 +2557.0, -2624.0 +2550.0, -2631.0
3. P.R.Rot,(O.OOl Rad.) Fixed Flex. Fixed Flex. Fixed Flex. Fixed Flex.
a. Left Column 0.000 2.040 0.000 1.900 0.000 2.060 0.000 1.990
b. Int. Column 0.000 2.480 0.000 2.360 0.000 2.620 0.000 2.570
c. Right Column 0.000 1.860 0.000 1.860 0.000 1.980 0.000 1.940
d. Wall 3.100 0.000 3.080 0.000 3.030 0.000 3.230 0.000
4.[rt.P.H.Rot(0.001 Rad) Fixed Flex. Fixed Flex. Fixed Flex. Fixed Flex.
a. Pos.Env. 128.400, 158.240 123.920, 146.480 126.080, 163.700 126.470, 159.860
b. Neg.Env. 133.710,170.110 131.520, 158.820 135.160, 168.690 133.470, 168.430
c. Total 262.110, 328.350 255.440, 305.300 261.240, 332.290 259.940, 328.290
a. Viscous En. 355.9, 333.40 317.9, 294.8 304.5, 385.0 313.4, 358.2
b. Hyst. En. 217.2, 106.40 237.2, 179.1 223.9, 138.0 223.9, 150.0
c. Total 573.1, 440.30 555.1, 473.9 528.4, 523.0 537.3, 508.2
Remark: -
The viscous energy dissipation of structures with flexible foundations is larger than
for those with fixed bases but the hysteretic energy dissipations are smaller. This is caused
by the rotation of the footing foundations due to the soil flexibility. The hysteretic energy
dissipated by the Clough model, in general, is higher than that other models. This is because
289
the unloading stiffness of Clough model is equal to the initial stiffness and results a relatively
fat hysteresis loop. Because of the effectiveness of this model in dissipating energy in small
amplitude cycles, this models even dissipates greater hysteretic energy dissipation than the
Bilinear model which only dissipates energy when the member yield stresses are reached.
This is another reason why the Bilinear model causes a different structural response when
compared with other the models.
Another interesting result is that the Q-Hyst model dissipates more hysteretic energy
than that does the modified Takeda Model. This is due to the fact that the reloading stiffness
of the Q-Hyst model is smaller than that for modified Takeda model so that in a relatively
small amplitude oscillation this model dissipates energy more effectively than that does the
modified Takeda model. Similar results were reported in Ref.06. As given in Tables. 10.7,
10.8 and 10.9 the Q-Hyst model showed a slightly smaller roof displacement than the
modified Takeda model. This result is in contrast to those reported in Ref.S29. In general,
however, the Q-Hyst and modified Takeda hysteretic models gave very similar structural
responses as was reported in Ref.SI4. Comparison between the total plastic hinge rotation
and hysteretic energy dissipation of the frame-wall structures on flexible foundations and the
rigid bases as shown in Tables 10.8 and 10.9 again shows that the energy concept in
describing the structural damage indices becomes meaningful as in the result in Sections
10.2.5 and 10.3.4.
The member, storey and structural damage indices of the 12-storey frames and
frame-walls are given in Table 10.10. As shown in Figs. 10.85 and 10.86 the Bilinear model
gives higher storey and structural damage indices when compared with the other models.
This is because this model gives higher total plastic hinge rotations than the other models as
shown in Fig. 10.51 and lower hysteretic energy dissipation when compared with the Clough
model as shown in Fig. 10.52. In this model the structure is stiffer because there is no
change in the member stiffness during the elastic part of the response. As a consequence,
this model gives greater total base shear and plastic hinge rotations than the other models.
On the other hand, the Clough model gives the smallest structural damage index because this
model results in the smallest plastic hinge rotations due to the highest hysteretic energy
dissipation. The Q-Hyst model causes a smaller total plastic hinge rotation and a higher
290
Moment (tm)
Moment (Tml
600
1000
400
500
200
Curv.lrad.(ml Curv.(flld./ml
0. 04 ~O.OO2 0.002 O. 04 -0. 04 -0.002 0.002 O. 04
Int. Column
Left Column Clough
Bilinear Fixed Base
Fixed Base
Fig. 10.71 Moment vs Curvature Relationship Fig. 10.78 Moment vs Curvature Relationship
Moment (tm) Moment (Tm)
800 8000
600 6000
400
Curv.lrad.fml
0.002 0.004 O. 06 -0. 02 0.02
St Walls
Q-Hyst.
Fixed Base,
291
eo ea-
Moment (Tm) Moment (Tm)
800
1500
600
1000
400
Curv.(rad.{m)
-0. 06 O. 09 -0. 06 0.006 O. 09
left Column Int. Column

Bilinear Clough
."Iex. Found.1 Flex. Found.
--------~1~oee_L-------------------~ L-~----~20ae_L-----------------~
BO
Moment (Tm) Moment (tm)
Right Column
Mod. Takeda BOO
Flex. Found. 6000
600
4000 -
400
2000
Curv.(rad./m)
-0. 09 -0.006 0.003 O. 06

-0. 004 -0.0002 0.0002 0.0004 0.0 06
- 000
-4000
-600 I Str. Walls
aHyst.
Flex.Found.
------------~Be&-~~ ------~
292
12 "II_Hr'r)_c:-- -~--~--~-- 0.3 ~------------------,
Square Bilin.
a Circle Clough
0.25
Trlang. Takeda ,.(
Diamd. Q-Hys!. ..,'"
.E
jFiIled : Fixed
.,
6 ",Jl:llank : FleK. '"
til
E
0"'
"iii
~
4
~ 0.2
2 ,
jAiled Fixed
Blank Flex.
o t:tl~~
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.15 +----t----t----+----+----i
Damage Index Bmn. Clough Tak. QHys!.
Fig. 10.85 Storey Damage Indices Fig. 10.86 Structural Damage Indices vs. Hyst. Rules
0.25 -r.----------------
Storey G420. FW12RZ

El Centro NSC
8 Square BUin.
Circle Clough >< 0.2
OJ
Trillng.
Oillmd.
Takeda
Q-Hyst.
".E
OJ
Cl
6 Filled
Blank
.,is
Q
1!
;l
'lJ
2
4 iii 0.15
.2 Filled Fixed
Blank Flax.
o
0.05 0.1 0.15 0.2 0.25 0.3 0.1
Damage IndeK BUin. Clough Tak. Q-Hys!.
Fig. 10.87 Storey Damage Indices Fig. 10.88 Structural Damage Indices \IS. Hyst. Rules
293
hysteretic energy dissipations when compared with the modified Takeda rules as shown in
the figures. As a result, the Q-Hyst model gives smaller structural damage than the modified
Takeda model. The foundation flexibility reduces the storey and structural damage indices
by an average of 6.89 and 6.05 % respectively.
Table 10.10 Damage Indices of the 12-storoy Frames IlIld Frame-Walls with Different Hysteretic Rules
Flexural Hysteretic Rules

No. Structural and Damage Remark
Bilinear Clough Mod.Tak. Q-Hyst.
Systems
l.Fixed Base Found.
11. Max.Member D.Index 0.3860 0.3310 0.3340 0.3400
b. Max.Storey D.Index 0.3049 0.2660 0.2929 0.2859
c. Struetural D.Index 0.2475 0.2197 0.2377 0.2284
2.F1ex.Foundations
11. Max.Member D.Index 0.3150 0.2750 0.2780 0.2990
c. Structural D.lndex 0.2346 0.2219 0.2219 0.2154
The 12-storey Frame-Walla
l.Fixed BaBe Found.
a. Max.Member D.Index 0.2960 0.2950 0.3200 0.3050
b. Max.Storey D.lndex 0.2590 0.2460 0.2694 0.2570
c. Structural D.Index 0.1871 0.1810 0.1947 0.1859
2.Flex.Foundations
a. Max.Member D.Index 0.2910 0.2780 0.2670 0.2760
c. Structural D.lndex 0.1909 0.1667 0.1700 0.1685
Remark: -
The effects of the hysteretic rules on the storey and structural damage indices are
shown in Figs. 10.87 and 1O.88respectively. The Clough model gives the smallest structural
damage index than the other models with the reasons mentioned above and shown in Figs.
10.63 and 10.64. The Q-Hyst model gives a smaller structural damage index than that modified
Takeda model for similar reasons. The anomalous performance of the Bilinear model on the
12-storey wall-frames with flexible foundations is shown in Figs. 10.63 and 10.88 respectively.
This model causes very high total plastic hinge rotations and results in higher structural damage
indices than for the case with fixed bases. Again, comparison between Fig. 10.86 and Fig.
10.88 indicate that the frame-walls tend to suffer less damage than frame structures.
The interpretation of the results of inelastic analysis of RC structures becomes more

complex due to a number of aspects affecting the structural responses. The effect of several
294
analysis aspects on the inelastic response ofRC structures including the rigid joint zone lengths,
number of frames in frame-walls and flexural hysteretic rules were investigated. As with
Chapter IX the investigations were conducted in several important damage parameters. The
damage parameters of structures with fixed and flexible foundations were compared.
The structural fundamental period of the frames and the frame-wall structures on the
flexible foundations and fixed bases decreases significantly as the rigid joint lengths increases.
A decrease in the period significantly affects the structural responses especially for the frame
structures. The shifting of the structural fundamental period is insignificant when the number
of frames in the frame-wall structures increases. Stiffening of the structural system due to an
increase in the rigid joint lengths cause an increase in roof horizontal displacements.
An increase in the rigid joint lengths gives a more significant increase in the base shear
on the frame structures than for frame-wall structures on both types of foundations. As with
previous results, rotation of the walls significantly increases in the first storey moment and
shear of the frame-planes. The total base shear in a whole frame-wall building may be several
times than that for a whole frame building. The structural type factor S = 1 for the
preliminary analysis of frame-wall structures is an underestimate. The type of member
hysteresis rules does not significantly affect the base shear except for the Bilinear rule.
The effect of the rigid joint lengths, number of frames and hysteresis cycles on the
inter-storey drift is insignificant. The maximum inter-story drifts (without P-delta and torsion
effects) is less than the limit value specified in NZS 4203: 1992. The effect of rigid joint
lengths, number of frames and hysteresis rules on the beam curvature ductility demands of
structures, in general, is less than the effect of the foundation types except for the frame
structures. The storey plastic hinge rotations tend to increase as the inter-storey drift increases.
The total plastic hinge rotation of the frame structures on flexible foundations is smaller
than that for the fixed bases as the rigid joint lengths are varied. The opposite result is found
in the 18-storey frame-wall structures. The effect of foundation flexibility on the total plastic
hinge rotation of the frames and the I2-storey frame-wall structures is less significant when
compared with that in the I8-storey frame-walls for all hysteresis rules. The number of frames
causes a complex effect on the total plastic hinge rotation.
295
The hysteretic energy dissipation of frame-wall structures with flexible foundations is
much less than those with fixed bases for all rigid joint lengths and hysteresis rules. The
opposite results are found in the viscous energy dissipations. The Clough and Q-Hyst hysteretic
models yielded more hysteretic energy absorption than the Bilinear and Modified Takeda rules
respectively. Considering the total plastic hinge rotation and the hysteretic energy dissipation
of the rocking structures, the single parameter energy-based concept for describing damage
indices of structures becomes less meaningful.
Both the foundation types and rigid joint lengths significantly affect on the storey and
structural damage indices of the frame structures. The storey damage indices had a similar
pattern with beam curvature ductility demands. The effect of the number of frames on these
damage indices were not clear. The Bilinear and Clough hysteretic models caused the biggest
and the smallest of structural damage indices respectively. In general, the Modified Takeda and
Q-Hyst models provided very similar structural responses, while some anomalous in structural
responses were found with the Bilinear model.
Chapter XI
Effects of Soil Shear Strains, Shear Moduli and Soil Models
on the Response of the 12 and lS-storey Frames and Frame-
Wall Structures
11.1 INTRODUCTION
As discussed in Chapter IV the soil shear modulus G is strain dependent. The variation
of soil shear modulus with soil shear strain is commonly expressed in a shear modulus
degradation curve G/G m"" as shown in Figs. 4.11, 4.12 or 4.13 respectively. Effects of the soil-
foundation impedances based on different levels of soil shear strains for the 12 and 18-storey
frame-walls are investigated. These soil-foundation impedances are based on the soil shear
modulus at shear strains 'Yu = 9x102 % , 'Y.. = 4xlO"2 % and 'Y" = 4xlO-J % corresponding
to shear modulus G = 0.50 Gmax , G = 0.70 Gmax and G = 0.95 Gmox respectively. These soil
shear strains are denoted by 0.5Gm, 0.7Gm and 0.95Gm as shown in the Fig. 11.1. The soil-
foundation impedances from soil shear strains of 0.0001 to 0.1 % are calculated with G = 420
kg/cm2 and G = 685 kg/cm2 for the 12 storey and 18-storey frame-walls respectively. These
soil shear strains cover the variation of soil shear strains from 0.95xlO-4 to 1. 74xlO-4 as
computed in Section 9.5.1.
In addition, effects of the soil shear modulus G of 125 kg/cm2, 260 kg/cm2, 540
kg/cm2, 685 kg/cm2 and 825 kg/cm2 on the structural responses of the 18-storey frame-walls
are also investigated. These soil shear moduli correspond to shear wave velocities of
approximately 85 mlsec, 120 m/sec, 150 m/sec, 170 m/sec, 190 mlsec and 210 mlsec
respectively. The first three soils may considerable soft soils and others may be classified as
intermediate soilsDJO,SJ1. The soil-foundation stiffnesses and damping coefficients are based on
G = 0.70 Gmw< corresponding to a soil shear strain 'Yso = 4xl0-4. The modified Takeda
hysteresis rule, the 1940 EI Centro N-S component earthquake and the rigid joint length Z =
0.50 are used.
The effect of the fictitious layer over half-space (FLHS) soil models on the responses
of the 12-storey frame-walls are compared with the effect of the fictitious layer over rigid
296
297
boundary (FLRB) and the Winkler soil-foundation models. The soil-shear modulus G = 420
kg/em2 is used.
11.2 EFFECTS OF SOIL SHEAR STRAINS AND SOIL MODULI
11.2.1 Fundamental! Period and Roof Horizontal Displacement
As presented in Tables 11.1 and 11.2, the fundamental periods of the 12 and 18-storey
frame-walls decreased by an average of 1.9 % and 4.2 % as the shear strains decrease. This
is because the lower the soil shear strains the stiffer the soils. These results are less than the
decrease in fundamental periods due to increasing the rigid joint zone lengths as discussed in
Section 10.2.1. This means that the effects of soil shear strains on the fundamental period of
structures are less significant than the effects of rigid joint lengths. As given in Table 11.3, a
decrease in the structural fundamental periods is more obvious in relatively soft soils and
almost linearly decreases for the intermediate soils. For example, an increase of soil shear
modulus from 125 kg/cm! to 260 kg/cm2 decreases the fundamental period by 9.1 %, while an
increase in the soil shear modulus from 685 kglcm 2 to 825 kglcm2 only decreases the structural
fundamental period by 1.7 %. Hence, effects of the soil shear moduli on the structural
responses may be significant in the relatively soft soils and may be insignificant in intermediate
soils. The roof horizontal displacement tends to increase as the soil strain decreases and the soil
shear modulus increase. This is due to a decrease in the structural fundamental period as
mentioned above.
U.2.2 Base Shears and Member Forces
In the 12-storey frame-walls, a decreases in the soil shear soil strains causes an increase
in the base shear by an average of 5.2 % as shown in Fig. 11.1. In the case of the 18-storey
frames, the base shear increases by an average of 3.84 % and is greater than 0.75 % increase
in the base shear due to an increase in the rigid joint lengths as discussed in Section 10.2.2.
As shown in Table 11.3 and Fig. 11.21, the increase-of soil shear moduli significantly
increases in the base shear in the soft soils as noted in Section 2.3.2.1 but the effect of the soil
shear moduli is almost insignificant in the intermediate soils. These results follow the decrease
in the structural fundamental period as discussed previously. Fig. 11.21 also shows that the
base shear of the frames decreases as the wall base shear increases due to an increase of the
soil shear modulus.
298
3000
Square Frames
Circle Walls
2500 Triang. n.B.Sheay
Filled Pos.Env.
Blank Neg.Env.
2000
ffi''"" 1500
.,
(j)
tU
ro 05Gm
07Gm
1000 .. 095Gm
07Gm,Fx.
500
-10000 -5000 o 5000 10000
05Gm 0.7Gm 095Gm Wall Moments I Tml. End 1.
Fig. 11.1 Base Shears vs. Soil Shear Stains Fig. 11.2 Wall Moments
12 "P-----L}{f\:-,--IIII:--t~k~
Storey
10
10
8 _. __.- ---.l..t.
1/
u
:r
y / '#.
... ,.
/~~ ci
~ ,.
6 -..-.... " j
~11 :ECl 3 .... _.........__ ....,..",.
IV! >-
Ii
f<:)
1;1
,..,
.:.
4 _1 "... '".~_ ~~
>< ..... ~ ..J

.5
Square: 05Gm
Circle : 07Gm
6
"Triang.: 095Gm
__ ~ .J.Fdled : St.PJ-l.Rot.
1:B!ank : St. Drift.
1.
5 10 15 20 25
Drift Ix 0.1 % HI. P.H.Rot.(x 0.001 Rad.) Storey P.H.Rot.(l< 0.001 Rad.!
Fi9.11.3 Inter-storey Drifts & Storey P .Hinge Rotations FIg.11.4 Inter-storey Drift liS. Storey P.H.Rotations
299
12 . -.......... - - - - L h - U : - - f r - - - - - , - - - - -
Storey
I)
Square 05Gm
Circle 0.7Gm
4- Triang. 095Gm
....o Diamd. 07Gm,Fx.
o
ci
.:5
.:: ;)
~
~
'li"P 4
J: 2
n:
G = 0.50 Gm
G = 0.7 OGm
G = 0.95 Gm
2 4 6 a
Left Col. Int.Col. Right Col. Walls Curvature Ductility Demands
Fig. 11.5 Wall Ii! Column's Base Rotations Fig. 11.6 Beam Curvature Ductility Demands
300 800
[square Vise.
leise,e Hyst.
250 ,Triang. Tt.En.
600
..; 200
III
a:
...
0
E
t:
0 C
ci
~
.,"
."
.9- 400
c 150
0
".,
." is
f!
0 ... III I11III
..
a: ~
'"
""
!:
C
UJ
~ 100
0::
200
50
o o +-------~ ............. -----h------~------~
05Gm 07Gm 095Gm 05Gm O.7Gm 095Gm
Fig. 11.7 P.Hinge Rotations \IS. Soil Shear Strains Fig. 11.8 Energy Dissipation vs. Soil Shear Strains
300
Table 11.1 Response Quantities of the 12-storey Frame-Walls with Different Soil Shear Strain
Soil Shear Strain 'Y..
Parameters & Str. Systems
( 7 Frames + 2 Walls) 'Y.. = 9xl0-4 'Y.. = 4xlO-4 'Y.. = 4xlO's Rem.

2.833 ( 0.236 N ) 2.771 (0.231 N) 2.728 ( 0.227 N ) (-)
1. Pos.Envelope 0.2541 (0.563 % Hu) 0.2691 (0.596 % Hb) 0.2781 (0.616 % Hb) (+)
2. Neg.Envelope 0.3090 (0.684 % H b) 0.3138 (0.695 % Hb) 0.3188 (0.706 % Hb) (+)
3. Base Shear (fonne)
1. Frames +935.0, -917.0 +912.0, -926.0 +913.0, -932.0 (+)
2. Wlllls +500.0, -761.0 +588.0, -897.0 +565.0, -925.0 (+)
3. Total Base Shear + 1435.0, -1678.0 +1500.0, -1823.0 +1478.0, -1857.0 (+)
4. Maximum Col.Bases
P.H.Rot.(x 0.001 Rad)
1. Left Column 2.993 2.855 2.721 (-)
2. Internal Column 2.948 2.940 2.796 (-)
3. Right Column 1.833 1.777 1.754 (-)
4. Walls 0.000 0.000 0.000 (=)
5. Tot.P.H.Rots.(xO.OOI Rad)
1. Pos.Envelope 126.070 144.170 155.899 (+)
2. Neg.Envelope 170.549 179.670 186.077 (+)
3. Total 296.619 323.840 341.976 (+)
6. Energy Dissipation (fm)
1. Viscous Energy 258.200 355.000 353.700 ( +)
2. Hyst. Energy 141.800 172.000 192.500 (+)
3. Total Energy 500.000 527.000 546.200 (+)
As shown in Figs. 11.2 and 11.10, the wall moments of the 12 and 18-storey frame-
walls increased by an average of 5.98 % and 5.94 % as the shear strains increase. These
results cannot be compared with the effect of joint rigid zone lengths because relative stiffness
of the frames and walls do not change with variation in the soil shear strains, but do alter with
the variation of the rigid joint zone lengths. The redistributions of moments to the middle
storeys due to the rotation of wall footings are more significant in the 18-storey frame-walls.
11.2.3 Inter-storey Drifts and Plastic Hinge Rotations
A decrease in the soil shear strain causes a signiticant increase in the inter-storey drifts
in the upper storeys of the 12 and 18 storey frame-walls as shown in Figs. 11.3 and 11.11
respectively. According to ZHU et al. (1992) these results occur mainly in relatively slender
stmctures subjected to a relatively high acceleration to velocity ratios where the effect of higher
modes becomes significant. Fig.11.n shows that the effect of the soil shear modulus on the
inter-storey drifts of the IS-storey frame-wall is less signiticant.
301
As shown in Fig. 11.3 the storey plastic hinge rotations of the 12-storey frames
increased by an average of 4.34 % as the soil shear strain decreases. This shows that in the
upper storeys the plastic hinge rotations have a strong relationship with the inter-storey drifts.
Similar relationships have been discussed in Ref.Z9. Fig. 11.22 shows that the storey plastic
hinge rotation of the 18-storey frame-walls tends to increase as the soil shear modulus
increases. It also shows that some storeys remain elastic as indicated by zero storey plastic
hinge rotations. The variation of storey plastic hinge rotation versus inter-storey drift of the 12-
storey frame-walls as shown in Fig. 11.4 and show greater scatter when compared with those
in Figs. 11.12. Fig. 11.23 shows that relatively soft soils causes greater inter-storey drifts, but
gives smaller storey plastic hinge rotations.
Table 11.2 Response Quantities of the IS-storey Frame-Walls with Different Soil Shear Strains
Soil Shear Strain 'Y..

No. Parameters & Str. Systems
( 7 Frames + 2 Walls ) 'Y.. = 9x1<r 'Y.. = 4x10-4 'Y.. = 4x10-s Rem.

3.868 ( 0.215 N ) 3.607 ( 0.200 N ) 3.552 ( 0.197 N ) (-)
1. Pos.Envelope 0.2920 (0.435 % ~) 0.2945 (0.439 % 1\) 0.2996 (0.447 % ~) (+)
2. Neg.Envelope 0.2767 (0.413 % ~) 0.2987 (0.445 % ~) 0.3027 (0.451 % ~) (+)
1. Frames +1272.0, -1299.0 +1257.0, -1301.0 +1240.0, -1287.0 (.)
2. Walls +1087.0, -1209.0 +1300.0, -1323.0 +1304.0, -1418.0 (+)
3. Total Base Shear +2359.0, -2508.0 +2557.0, -2624.0 +2544.0, -2705.0 (+)
4. Maximum Col. Bases
P.H.Rot.(x 0.001 Rad)
1. Left Column 1.426 2.896 1.563 (.)
2. Internal Column 2.703 2.773 2.903 (.)
3. Right Column 1.951 2.446 2.013 (.)
4. Walls 0.000 0.000 0.000 (=)
5. Tot.P.H.Rots.(xO.ool Rad)
1. Pos.Envelope 149.130 163.700 171.650 (+)
2. Neg.Envelope 144.880 168.690 183.240 (+)
3. Total 294.010 332.390 354.890 (+)
6. Energy Dissipation (Tm)
1. Viscous Energy 400.000 374.000 348.900 (+)
2. Hyst. Energy 101.900 125.000 132.460 (+)
3. Total Energy 501.900 499.000 481.360 (+)
11.2.4 Column, Wall Base Rotations and Beam Curvature Ductility Demands
Figs. 11.5, and 11.13 show that the plastic hinge rotations of column and wall bases
are not significantly affected by increasing the soil shear strain when compared with the effect
302
4000
:G685,FW18RZ
! Takeda, EcNS
3000
~
&i'" 2000 Square ,; 05Gm
Circle '! 07Gm
'"
In
Triang. : 095Gm
'"
III
1000
Square Frames
Circle Walls
Triang. n.B.Shear
Filled Pos.Env.
Blank Neg.Env.
0 -25000 -15000 -5000 5000 15000
05Gm 07Gm Q95Gm Wall's Moment, End 1.
Fig.11.9 Base Shears vs. Soil Shear Strains Fig. '11.10 Wall Moments
-
.. j;.
18 ..
12 ,-~~--,---~~--~~-- __~--~
1
16 ----~ .. Square
Circle
: 05Gm
: 07Gm
"
.......... 11>....
14 ..........
Triang. : 095Gm
Dlamd. : 07Gm,Fx.
1
r ,.p;'
10
Filled
Blank
07Gm,FJ(.
Flex
12
~
.
II
l#r
s:
'#
....
0 B
J~ r
~ 2:>
10
II>
.t:
d
8 >-
f:!
II E 6
Storey
l\\ '"0!-
6 J!!
f~
.E
4 1- ........
4
'~
2 v. . ....-. f-O'
/ !
!
o
o 2 4 6 8 10 o 5 10 15 20
Storey Drift (II 0.1 % H ) Storey P.H.Rotation (II 0.001 Rnd.)
Fig. 11.11 Inter-storey Drifts Fig. 11.12 Inter-storey Drifts vs. P.Hinge Rotations
303
18
16
Storey
4
14
...-:
to t 12
a:
...
!
0
,I
a 3 I
ci 10
!!
I
t:
.2 I
, 6
!
~
'"
c
'"c
OJ
"'-
2
I~
I 6
I 4 G = 0.7 Gm.Fx ....~ .. ~~~..i
.A
Square 05Gm
Circle 07Gm 2
I Triang.
Diamd.
09SGm
07Gm.Fx.
0
0 i

Left Col.

Int.Col. Right Col. Walls
~-- 0 :2 4
Beam Curvature Ductility Demand.
6 B
Fig. 11.13 Wall & Column's Base P Hinge Rotations Fig. 11.14 Beam Curvature Ductility Demands
300 aoo
Square Vise.
Circle Hyst.
260 Triang. Tt.En.
600 -'-
"a:'" 200
~
0
e k
0 !::
ci c
.2
..
3- ;1
c:
0
160
..
Cl.
'iii 400
~0 is
>-
.
a:
C)
.5
E'
'e"
oJ
:r 100
.,:
200
Riled : Pos.Env.
50
Blank : N"9. Env .
o o
05Gm 07Gm 095Gm 05Gm 07Gm 095Gm
Fig. 11.15 P.Hinge Rotations \IS. Soil Shear Strains Fig.11.16 Energy Dissipations vs. Soil Shear Strains
304
of the foundation flexibility. As with previous results, an increase in the plastic hinge rotations
of the column bases is more obvious in relatively soft soils as shown in Fig. 11.24. Figs 11.13
and 11.24 show that the column bases of the 18-storey frame-walls with fixed bases remain
elastic but response inelastically in the case of the flexible foundations. The opposite response
is found for the wall bases. The maximum plastic hinge rotation of 0.003 is much lower than
the rotation capacity as discussed in Section 95.4.5. Increased plastic hinge rotations, shear
forces and moments at the column bases, decreased wall moment and shear forces at the wall
bases due to the rotation of wall-footings are similar results as reported by GOODSIR (1985).
The maximum beam curvature ductility demand increases as the soil shear strain
decreases or the shear modulus increases. Increased base shear, as discussed in Section 11.2.2
cause an increase in beam curvature ductility demands. Figs 11.6 and 1l.14 show that the beam
curvature ductility demands of the 12 and 18-storey frame-walls increase by an average of
Table 11.3 Response Quantities of the 18-storey Frame-Walls with Different Soil Shear Moduli
Soil Shear Modulus G

No. [Parameters & SIr.Systems
( 7 Frames + 2 Walls) Rem.
G125 G260 G400 G540 G685 G825
4.344 3.918 3.765 3.694 3.607 3.545 (-)
1. Pos.Envelope 0.2703 0.2852 0.2872 0.2881 0.2848 0.2844 (.)
2. Neg.Envelope 0.2267 0.2280 0.2509 0.2608 0.2773 0.2902 (+)
a. Pos. Envelope
1. Frames +1326.08 +1391.3C +1391.30 +1302.00 +1268.00 +1213.00 (.)
2. Walls + 902.18 +1086.9<: +1184.80 + 1309.00 +1352.00 +1360.00 (+)
3. Total Base Shear +2228.26 +2478.2(' +2576.10 +2611.00 +2620.00 +2573.00 (.)
b. Negative Envelope
1. Frames -1354.48 -1388.06 -1392.50 -1345.00 -1313.00 -1279.00 (.)
2. Walls -1097.01 -1152.98 -1242.50 -1353.00 -1397.00 -1415.00 (+)
3. Total Base Shear -2451.49 -2541.04 -2635.00 -2698.00 -2710.00 -2694.00 (+)
4. Max.Bases Rot(O.OOI Rad)
1. Left Column 0.000 1.792 2.108 2.154 2.044 1.907 (.)
2. Internal Column 1.965 2.409 2.615 2.711 2.480 2.395 (.)
3. Right Column 1.658 1.870 2.032 2.151 1.860 1.655 (.)
4. Walls 0.000 0.000 0.000 0.000 0.000 0.000 (=)
5. Tot.P.H.Rot,(O.OOI Rad)
1. Pos.Envelope 129.290 143.230 140.860 167.142 167.523 170.202 (+)
2. Neg.Envelope 111.990 140.180 155.920 166.547 179.470 187.724 (+)
3. Total 241.280 283.410 296.780 333.689 346.993 357.926 (+)
l. Viscous Energy 492.700 410.580 363.140 341.400 335.000 333.300 (-)
2. Hyst. Energy 56.930 82.170 94.890 100.750 96.000 94.400 (-)
3. Total Energy 549.630 492.750 458.030 442.150 431.000 427.700 (-)
Remark: (+): Increase, (-) : Decrease, (==) : Equal: (.) : Not clear
305
7.01 % and 13.56 % respectively. In the case of the lS-storey frame-walls, this result is
slightly less than the 13.86 % increase in the beam curvature ductility demands due to the
increase in rigid joint zone lengths.
n Total Plastic Hinge Rotations, Energy Dissipation and Damage

Indices
As given in Tables 11.1, 11.2 and 11.3 and as shown in Figs. 11.7, 11.15 and 11.25
the total plastic hinge rotations of the 12 and 18 storey frame-walls increased by an average of
7.39 %, 9.90 % and 9.25 % respectively. This result is similar to that discussed in Chapter
IX where the total plastic hinge rotations increase with an increase in the overall stiffness of
the structure. In Figs. 11.8 and 11.16 the hysteretic energy dissipation increases while the
viscous energy dissipation decreases as the soil shear strains decreases. This is the effect of
increasing the overall stiffness of the structure. A decrease in the viscous energy dissipation
due to increasing the overall stiffness of the structure is more obviously as shown in Fig.
11.26. A decrease in the viscous energy dissipation is more significant in relatively soft soils.
This result confirms the discussion in Sections 10.2.1 and 10.2.2
The storey damage indices of the 12 and lS-storey frame-walls as shown in Figs.
11.17, 11.18 increase as the soil shear modulus increases. The storey damage indices increased
by an average of 8.17 % and 11.40 % and lead to an increase in the structural damage indices
by an average of7.76 %,4.85 % respectively. As with the previous results, the effects of soil
shear modulus on the maximum member and maximum storey damage indices are more
obvious in relatively soft soils as shown in Fig. 11.28. This figure also shows that in
intermediate soils the effect of variation of the soil shear modulus on the structural damage
index is almost insignificant. This also discussed in Ref. S36. As in previous results, the
higher total plastic hinge rotations results in higher structural damage indices. The effects of
the higher modes on the storey damage is more significant in higher structures such as the 18-
storey frame-walls than in the 12-storey frame-walls as shown in Figs. 11.17, 11.18 and 11.27
respectively. The absence of the effect of higher modes in the 12-storey frame-walls was also
mentioned in Ref. G9.
306
1B
16
Square G = 0.50Gn'l 10
Circle G = 0.70Gm
6
Triang. Square G = 0.5Gm
B U .. ..'l\~ .. ~~ ......,Circle 6 = 0.76m
Triang. 6 = 0.956
6
o 0.05 0.1 0.15 0.2 0.25 0.3 o 0.05 0.1 0.15 0.2 0.25
Damage Index. Damage Index.
500 0.28 -,-~~-~~~~-~-~-----
0.24
400
0.2
,;
a:"
.,><
."
~ .5
0
0 ., 0.16
"'"
0
E
-j 300 to
0
0
a: ]
:i 0.12
0 '"
tl
i;"
.~
n:'"
O.OB
Square : FW12RZ
200 Circle : FW18RZ
0.04
1 00 ~1~~~~-j-~ ---+-~---+I~-~~, o +-------+-------+--------+-----~
O.50Gm O.70Gm O.95Gm O.5Gm O.70Gm O.95Gm
Fig.11.19 P.Hinge Rotations vs. Soil Shear Strains Fig.11.20 Structural Damage Indices vs. Soil Shear Strains
307
3500 ..
.~ ~~-- ..............~~.--~~~~~~~--~. 18
16
3000 Storey
14
12
2500
E Frames
10
iii
Ql Walls
.<:
til
:WOO n.B.She"
.,'" 8
'"
rn Filled Pos,Enll .
Blank Neg.Enl/.
6
1500
1000
2
G260 G540 GB26 0

500 0 5 10 15 20
G125 G400 GG85 Drift {x 0.1% HI, P.H.Rot.(x 0.001 Rad.J.
Fig. 11.21 Base Shears vs. Soil Shear Moduli Fig. 11.22 Inter-storey Drifts 8. Storey P .Hinge Rot.
12 5
Square
Circle 6260
Triang. G400
Diamd. G540 Diamd. G540
10 4 G685
Plus G665 Plus
Cross G625 Cross G825
!Oiamd. GG8G,Fix
#. .,;
fIl
.... Ill:
0 8 (; 3
I
~ <:>
0
"
;t:!
't:
0 0
0&-+)0 ~
C e
)..
'"5
0 [DO M-1XN)X ...J9
0
ti 6 '4> 0
a:
o!. 2
...'" .,'"'"
.E
~ DElIDl>,
~
4~
X+X+ fl..
o 4 B 12 16 20
Storey P.H. ilot (x 0.001 !lad. left Col. Int.CoI. Right Col. Walls
Fig. 11.23 Inter-storey Drifts VIS. Storey P.H. Rotation Fig. 11.24 Column &. Wall Base P.H. Rotations
308
250 ,-------~~~~~~~~~~~__,
GOO , - - - - - - - - - - - - - - - - - - - - - - - - ,
500
200
400
"'"
...a:'"
150
~
'C
0 C
(;)
ci .,E'" 300
.!i C
~ >-
2'
n:: 100
i
a: 't:"
W
200
[Filled : Pos. Env. I
t
iBionk : Neg. Env.
501
if~
100
t G260 0540 G825 G260 G540 G625

I
o +---!----I---~--_+I .~-+---+--__1 0
G125 G400 G685 G125 G400 G6S5
Fig. 11.25 P .Hlngo Rotations vs. Soil Sheaf Moduli Fig. 11.26 Ener.9Y Dissipation vs. Soil Shear Moduli
0.4 , - - - - - - - - - - - - - - - - - - - - - - - - -
Square Member D.lndel(

0.35 Circle Storey D.lnda)!.
Triano. StNet. D.lndex
14
D.3
12
0.25
10
8
0125 ..
G260
i.
i
! t
~
E
0.2
6 .... L .............L .........jT:lang.

. ! Dl8md.
G400
G685
+-,--_...
I
0.15 t
Plus : G6B5 i
G825 !
t
Cross
4 +. -.:=tIpll_~_IIII+ . . . ,. . ,.
~.
01
..................t ..................+ .....:.........
o
o 0.05 0.1 0.15 0.2
i
0.25
Ii
0.3
00: ;.r__ +-~G_2t_6-0- _+- G- 5~I-O- - -1!- - G-8;2-S_-I
Storey DamBg" Index GUS G400 G68S
Fig. 11.27 Storey Damage Indices Fig. 11.28 Damage Indices vs. Soil Shear Moduli
309
Table 10.4 Damage Indices of the 12 and IS-storey Frame-Walls
Soil Shear Strain 'Y~

Damage Parameters 1'.. = 0.9xl~ 1'.. = 4xl0-4 1'.. = 4xlO-5 Remark
1. 12-storey Frame-Walls (faked a)
a. Max. Member Damage Indices 0.2280 0.2670 0.2840 (+)

2. IS-storey Frame-Walls (fakeda)
11. Max. Member Damage Indices 0.2240 0.2530 0.2380 (+)

GI25 G260 G400 G540 G685 G825

3. IS-storey Frame-Walls (Bilinear)
a. Max. Member Damage Indices 0.2050 0.2160 0.2480 0.2700 0.2780 0.2870 (+)
b. Max. Storey Damage Indices 0.1760 0.2076 0.2252 0.2303 0.2320 0.2595 (+)
c. Structural Damage Indices 0.1218 0.1457 0.1535 0.1559 0.1542 0.1507 (+)
Remark: (+): Increase, (-) : Decrease, (=) : Equal : (.) : Not clear
11.3 EFFECTS OF WINKLER FOUNDATION AND LAYER OVER RIGID

BOUNDARY SOIL MODELS
The response of the 12 storey frame-walls supported by different soil models when they
SUbjected to the 1966 Parkfield N65E component earthquake are compared. The dynamic
properties of the soils, modelling of the member hysteresis rules and the rigid joint length used
fOf the investigation are the same as mentioned in Section 11.1.
11.3.1 Winkler Foundation
As described in Chapter V, both the properties and geometry of the soil deposit may
be non-homogeneous. This may due to irregularity of the soil layer and the corresponding rigid
boundary. It is often assumed that both the soil layer surface and the rigid boundary surface
are perfectly horizontal planes. Two other soil models, a fictitious layer over a horizontal rigid
boundary (FLRB) and a soil model represented by a bed of independent springs or Winkler
model are used in comparison with the fictitious layer over the half-space soil model (FLHS),
described in Chapter V. The spring in the Winkler soil-foundation is modelled by the 4-node
beam member i.e one of the member model option in the RUAUMOKO computer program.
310
=c
001
0.005 -
r G420. WF12RZ
Takeda, Parkf.NE
~ oT
..,c:
0
!l0 -0.005
<C
[J)
Fictitious Layer over Half-Space
.g ( FLHS Model)
a -0.01
0
"-
Winkler Model
-0.015
-0.02
Fig. 11.29 Rotation of Wan Foundation Footings
0.015
0.01
Winkler Model
..
]
c: 0.005
'"
E
III
"'"
Ci. 0
is'" 4 8 12
iii
FLHS Model
'f" -0.005
>'"
-0.01
\f V .~, v' '..' v 'J ,,'
-0.015
Fig. 11.30 Vertical Displacement of Wall Footing Node
0.08
Winkler Model
0.04
fi
Ii
I'!
:
~
c:
'"E
"'"co
0. 0
.!:l
0
iii
"
'f
>'"
-0.04
-O.OB
Fig. 11.31 Vertical Displacements of Wall Foundation Footing Tips

311
Axial Deformation 1m}

,-
0 0.03
Storey
Circle : FLHS Model
/ -100
Triang.: Winkler Model
I)
Filled : Full Gray. Load
Blank : Half Gray. Load
E
.,
(.)
of -150
-200
-15 10 -5 o 5 10 15
Moment [ x 1000 Tm}
Fig. 11.32 Force Deformation Relationship of Soil Fig. 11.33 Wall Moments
12
Storey Storey
10 + .. .........................
~ 10+-..-.. ~.. ~-.. t~.A.~.. t....~~~.... .. ~--I
a+~ .... .... : .... ........

~ ~ .... ~n
6 +-...................................... ,.
Circle : FLHS Model Cilele : FLHS Model

Tiang. : Winkler Model Triang. : Winkler Model
2 2
: Full Grav. Load Filled : Full Grav.
o ... ~~~~~---I~~~~~-l~~ .....~~-1

5 10 15 20 0.2 0.4 0.6 0.0
Curvature Ductility Demands Member Damage Index
Fig. 11.34 Beam Curvature Ductility Demands Fig. 11.35 Maximum Member Damage Indices
312
0.01
G420, FW12RZ
Takeda, Pa,kf.NE
0.005 --
'"~ 0
!:
.g 2 6
gl
0
0::
III I ;
"0 t i
D \ ;
Z I,
-0.005 V
'"'"
'"
<0
; FLHS Model
V
Rctitious Laye, over Rigid Boundary (FLRB Model)

0.01 with Soil Laye, 30 m. depth
Soil Laye, 20 m.
Soil Laye, 10 m.
-0.015
Fig. 11.36 Rotation of the Wall Base Node for Difference Soil Models
0.05
0.04
0.03
0.02
!
C 0.01
'"
E
OJ
"
.!!l
c. 0
is'" 2 4
co
".,
'f -0.01
>
-0.02
-0.03
-0.04
-0.05
Fig. 11.37 Vertical Displacement of the Wall Footing Tips

313
~---ffiaoo&-,-------~-------~---,
6000
Rotation (radl Rotation (radl
0.01 0.2 -0 1 0.01 O. 2
-6000 FW12RZ -6000 FW12RZ

S.L. 10 m.
S.L. 20 m.
Parkt.N!:
Parkt.N!:
aoo L-----~el9-~--------------~----~
Fig. 11.38 Moment vs Wall Footing Rotation Fig. 11.39 Moment vs Wall Footing Rotation
00
8000
6000
...r:
"
E
0 6000
::E 4000 ...<::
"
E
0
::E 4000
Rotation (rad)
Rotation !rad I
0.01
-0 1 0.005 0.01 O. 15
-4000
6000 FW12RZ FW12RZ

S.L. 30 m. HalfSpace
Parkf.NE Parkf.N!:
Fig. 11.40 Moment vs Wall Footing Rotation Fig. 11.41 Moment vs Wall Footing Rotation
314
12~------ ---,----CB~r--,----------,
10 +....................i................n 'G.t.. .................... ......l
Storey Storey
Square: S.L. 10 m.
Circle : S.L. 20 m.
$ S.L . 30 m .......... + ......... .QH
<R)......t ..jTriang.
Diamd.: FLHS Mod.
II-
-15 10 5 o 5 10 15 5 10 15 20
Wall Moment Ix 1000 Tm) Curvature Ductility Demand
Fig. 11.42 Wall Moments Fig. 11.43 Beam Curvature Ductility Demands
950 12~----------,---~-- .. ----.-----------,
10+........ .. .. ........ +................CO..........+ ........ ........ 1
750 Storey
3~...... .... .... .. ........ , ....................,~~}.............. ..I
E
!::; Square: Visco En.
c
..,
0 Circle : Hyst. En.
III Triang. : Total En. 6 ................................,.. ..........,;NI.. ..........+ ...... .................... ..1
.r; 550
II>
.!!!
0
>.
~
OJ
C
w 4
350 Square S.L. 10 m.

Circle S.L. 20 m.
2
Triang. S.L. 30 m.
Diamd. FLHS Mod.
R.B.20m FLHS
150 +-~~~~~~~~~+-----~----~
0.2 0.4 0.6 0.8
R.B.10m R.B.30m Member Damage Index
Fig. 11.44 Energy Dissipations vs. Soil Models Fig. 11.45 Max. Member Damage Indices (Parkf.NE)
315
A comparison of the wall footing rotations based on the FLHS and Winkler models are
shown in Fig. 11.29. Only the rotation of the wall footing is compared because this footing is
sensitive to lift-off. The rotation is computed assuming one-half of the wall gravity load with
the foundation size being the same as for the case of a full wall gravity load. The Winkler
spring stiffnesses are based on the initial vertical displacement assumed to have the same
displacement as the FLHS modeL The pattern of footing rotations are similar but the Winkler
model results in a higher rotation than the FLHS model. This may be due to an incompatibility
of the rocking stiffness generated by the vertical forces in the springs in the Winkler model and
the rocking stiffness from the FLHS model. The absence of vertical dashpots in the Winkler
foundation, as described in Chapter VII, may also result in differences in the responses.
The vertical displacement of the wall base node (at the centre of the footing) of these
models is shown in Fig. 11.30. It shows that these models perform quite differently. The
Winkler foundation shows the footing lifting-off beyond the centre ofthe footing. On the other
hand in the FLHS model the centred vertical displacement must remains constant and the
footing lift-off occurs only at the tip of the footing, as shown in Fig. 11. 31. This is because
the vertical and horizontal soil-foundation stiffnesses in the FLHS model are assumed to remain
constant during the time integration as described in Section 13.4.
The maximum vertical load-deformation of the soil under the Winkler model is shown
in Fig. 11.32. The soil remains elastic although the footing undergoes a relatively high rotation.
This is because the static bearing capacity safety factor of the soil is twice that when the footing
foundation is subjected to the full wall gravity load. The wall moments, beam curvature
ductility demand and maximum member damage indices are shown in Figs. 11.33, 11.34 and
11.35 respectively. The effect on the wall footing, as modelled by the Winkler foundation, on
the structural responses, in general, does not seem significant.
11.3.2 Soil Layer over a Horizontal Rigid Boundary
Soil layers with depths of 10, 20 and 30 metres over a horizontal rigid boundary are
considered. The soil-foundation impedances are computed using formulae proposed by
ELSABEE et al. GS, The structural response based on this fictitious layer over a rigid boundary
(FLRB) is compared with the fictitious layer over a half-space (FLHS). The wall footings are
subjected to the full wall gravity loads.
316
0.01 ~----------------~~--------------------------------------------~------,
0.005
:s-
0 +-~~~~'t-"'"-""~>''--'-'+H-....-~'---II-\i+--'~~~
"
o 2
E
o 0.005
a:
01
.g FLRB Model
&00:: 1 Winkler Model
.0.02L----------------------------------------------------------------------
Fig. 11.46 Rotation of Wall Foundation Footings
0.006
0.003
.l:
Q)
0
E ll- S
.,u'"
.,
0. 0.003 Winkler Model
i5
FLRB Model
"'f"
OJ
-0.006
>
I'.
-0.009
-0.012
Fig. 11.47 Vertical Displacement of of Wall Footing Nodes
0.06
Winkler Model
0.04
0.02
...
l:
OJ
E
.."'"
0. 0
.!2
0
"
.2
t: -0.02
>'"
-0.04
FLRB Mod,,1
-0.06
Fig. 11.43 Vertical Displacements of Wall Foundation Footing Tips

317
400
Wall
3000
2000 -
1000 S3 Footing 107

105
Rotation (rail) 2.6 ::L6 ~
0.01 O. 2
j Gl,om"
I
12000
SFix.
// LFix.
~ Ground Surface Slab
--l-
!FW12RZ
-3000
Parkf.NE
iFLRB Model
LR:B. 10 m.
lFlex_ SCFlex.
Fig. ",49 Moment vs. Wall Footing Rotations Fig. 11. 50 Geometry of The Winkler and Type of Found.
25
25
Ax. Deformation Iml

Ax. Deformation 1m)
0 3 -0.02 0.01 0.01 O. 2

0.03 O. 6
-50
-75
100
- 1"
125
rW12RZ. Half iFW12RZ, Half

:Parkf.NE. R.B
iWinkler Model
Mb. 93, SF3.15
iParkf.NE, R.B
!Winkler Mod,,)
1Mb. 106, SF3.15
'" 1
175
Fig. 11.51 Force-Deform. Relationship of Soli Fig. 11.52 Force-Deform. Relationship (If Soil
318
25 ~ 25
Ax. Deformation 1m) Ax. Deformation 1m)
0 4 0.03 -0.02 0.01 0.2 0.03 0.02 0.01 0. 2
FW12RZ, Half
FW12RZ. Half
Parkf.NE. R.B
Parkf.N!!. R.B
Winkler Model
IWinkler Model
50 Mb. 108. SF3.15 50
Mh. 107, SF3.15
75 75
100 -100
E
E
., l:
.
"
(; 125 ~ -125
..
u.
<l
0<
<;(
160 -150
Fig. 11.53 Force,Deform. Relationship of Soil Fig. 11.54 Force-Deform. Relationship of Soil
I) +.""",.",,,,.,,,,,,,,,,.,,,,-,,,,,,,;
4 It
Square: Winkler. FLHS
Circle : Winkler. HRB10
Trian!!. : FLHS Model
2 2
8 12 16 20 0.2 0.4 0.6 O.B

Curvature Ductility Demands Member Damage Index
Fig. 11.56 Beam Curvature Ductility Demands Fig. 11.56 Maximum Member Damage Indices
319
Using these proposed formulae, the shallower the layer in the FLRB model the higher
the soil-foundation impedance, especially for the vertical vibration mode. The soil-foundation
radiation damping and rocking stiffnesses of this model are much less than those of the FLHS
model. The rotation of the wall footing using these soil models is shown in Fig. 11.36. It
shows that the shallower layer has larger footing rotations. This is because the larger vertical
stiffness results in a smaller initial vertical displacement of the footing, as shown in Fig. 11.37,
and the smaller rocking stiffness allows larger rotations. As shown in Fig. 11.37 the lift-off of
the wall footing tip is also more pronounced on the shallow soil layer than for the deeper
layers.
The wall base moment versus footing rotations of these soil models are shown in Fig.
11.38, 11. 39, 11.40 and 11.41 respectively. The decrease in rocking stiffness due to lift-off
of the footings is more obvious in the shallower layers. This is because greater footing lift-off
has occurred. The value of Mo. as given in Eg. 13.5 or shown in Fig. 13.39, increases as the
depth of the layer increases. This results from a higher value of q,o due to the larger initial
vertical displacement of the footing.
The structural responses including the wall moment, beam curvature ductility demands,
energy dissipation and maximum member damage indices of the 12-storey frame-walls
supported by these different soil models are shown in Figs. 11.42, 11.43, 11.44 and 11.45
respectively. As found in the previous section, the structural responses are not significantly
effected by the different soil models.
The effects of the static bearing capacity safety factor of the soil on the responses with
redesigned wall foundation footings are shown in Fig. 11.46 through Fig. 11.56 and will be
discussed in Section 13.5.
The effect of soil shear strains (SS), soil moduli (SM) and soil models on the seismic
response of the 12 and lS-storey frame-walls on the flexible foundation were investigated. A
decreased in soil-shear strain corresponds to an increase in the soil modulus. Investigations
were conducted on several important damage parameters including roof horizontal
320
displacements, member forces, total base shears, storey drifts, curvature ductility demands,
plastic hinge rotations, energy dissipations and damage indices.
Lengthening of the structural fundamental period is more obvious in relatively soft soils
than in intermediate soils. This result strongly affects the structural responses. Stiffening of the
structural system due to increased soil moduli cause an increase in roof horizontal
displacement.
The total base shear of frame-wall structures increases as the soil stiffness increases.
As with previous results an increase in the total base shear is more evident in relatively soft
soils than in intermediate soils. An increase in the soil stiffness also causes significant increase
in the wall moment.
The effect of increasing the soil stiffness on the inter-storey drifts were significant in
the upper storeys especially for the 18-storey frame-walls. The maximum inter-storey drifts was
less than the limit value specified in NZS 4203: 1992. An increase in the soil stiffness also gives
significant increase in the beam curvature ductility demands in both the 12 and IS-storey frame-
walls.
The storey and total plastic hinge rotations increase as the soil stiffness increases. Some
beams in the middle storeys of the I8-storey frame-walls remain elastic as indicated by zero
curvature ductility factors. Relatively high inter-storey drifts with small storey plastic hinge
rotations were found in the IS-storey frame-walls on a relatively soft soils. In general, the
storey plastic hinge rotation increases as the inter-storey drifts increase.
The hysteretic energy dissipation increases as the soil stiffness increases but viscous
energy dissipation decreases. Viscous energy dissipation is very sensitive in relatively soft soils.
The member, storey and structural damage indices tend to increase as the soil stiffness
increases.
The soil models, in general, do not significantly affect the structural responses. When
compared with the lumped spring-dashpot model, the Winkler soil-foundation model shows
very different veltical displacements at the centre of the wall bases.
Chapter XII
Effects of Earthquake Excitations on the Response of the
and IS-storey Frames and Frame-Wall Structures
]2,1 INTRODUCTION
The uncertainties associated with a proper selection of the design earthquake ground
motions for any building site have been recognized by engineers over past few decades. These
uncertainties are basically initiated by complex factors involving source mechanisms, source-
site transmission paths and local site conditions that affect the characteristics of strong motion
ground shaking as discussed in Ref. W31. The characteristics of earthquake strong motions are
usually quantified in terms of intensity. frequency content, duration, the number, size and
sequence of acceleration pulsesLlS W31 The 1985 Mexico earthquake showed that these
earthquake characteristics strongly affect the structural damage. A simple quantification of the
earthquake characteristics for design purposes, however, is still difficult to define.
The NZS 4203: 1984 clause C3.3 gives a simple guidance that for a class III building
on rigid or intermediate soils in seismic region A, the earthquake motions including 1940 EI
Centro N-S component may be used for design purposes. The NZS 4203:1992 Sections 4.10.1
and 4.10.2 state that when the numerical time history analysis of ultimate state limit building
response is used at least three different ealthquake records shall be employed. The minimum
length of earthquake input shall be taken as the greater of 15 seconds or 5 times the
fundamental period of the structure.
Attempts to describe the relationships between duration, frequency content, peak ground
acceleration and velocity was discussed by ZHU et al.(1988) and TSO et.al (1992). Records
from several ealthquakes indicate that at near the source the peak acceleration to ground
velocity (a/v) ratio is high, the record is of short duration, high frequency and impulsive. At
a large distance from the source, the a/v ratio is low, the record is of long duration, more
periodic and has a broad frequency content. Results of analytical investigations as reported by
ZHU et a1.(1992) indicate that a Iowa/v ratio of the ealthquake ground motions causes more
damage in high rise structures than motion with a high a/v ratio. TSO et.al (1993) reported that
the hysteretic energy dissipation for structures with relatively long fundamental periods is much
321
322
higher when subjected to earthquake ground motions with low acceleration to velocity ratios.
TSO et a1. (1993) discussed the characteristics of earthquake ground excitations and their affects
on the structural responses. It was demonstrated that the acceleration to velocity ratios strongly
affect both structural deformation and energy dissipation. Because of the uncertainties involved
in estimating the nature of ground motions that might occur in the future at a building site,
Ref.M25 suggested that several records should be used during the analyses of building
stmctures.
The structural fundamental periods of the building prototypes in this study are mostly
greater than 2 seconds. By considering the results above, several earthquakes with moderate
and low acceleration to velocity ratio are used. The earthquake ground motions and the
corresponding acceleration ratio are presented in Table 12.1.
Table 12.1 Acceleration to Velocity Ratio (a/v ratio)
No Earthquake Date Max. Ace. Max.VeL a/v ratio Remark

(g) (m/sec) g/m/sec
1. El Centro, N-S May 5, 1940 0.349 0.334 1.044 Md
2. Taft, N69W July 21, 1952 0.179 0.177 1.010 Md
3. EI Centro, E-W May 5, 1940 0.214 0.369 0.580 L
4. Bucharest, N-S March 4, 1977 0.206 0.751 0.272 L
5. Parkfield, N68E June 27, 1966 0.489 0.781 0.627 L
6. Tokachi Oki. N-S May 16, 1968 0.236 - - -
Remark TIO Z6
:1.2 g/mls > a/v> 0.8 g/mls : moderate, a/v < 0.8 g/mls : low
g: acceleration of gravity, Md : Moderate, L : low, - : not available
The available duration of the Bucharest earthquake record is only 16 seconds. A

duration of 20 seconds is used for the other earthquakes which is considerably greater than 5
times the structural fundamental period. The Parkf.1 and Taft.l as shown in Fig. 12.9
correspond to the 1966 Parkfield N65E and 1952 Taft N69W scaled to the maximum ground
acceleration of the 1940 El Centro NSC earthquake. The Parkf.2 is scaled according to spectral
acceleration of the 1940 El Centro NSC at the corresponding structural fundamental period.
According to Ref.TI1 the 1971 San Fernando earthquake recorded at Pacoima Dam seems to
be an upper bound for a possible seismic event in New Zealand, therefore, this earthquake is
not used in this study.
The soil-foundation stiffnesses and damping coeffIcients were taken at the soil shear
strains 1' .. :::: 4xlO-4 and were based on a soil shear modulus G = 420 kgfcm2 and G :::: 540
323
kg/cm2 for the 12 and I8-storey frame-walls respectively. Other soil dynamic properties were
similar to there used in Chapter IX. The rigid joint zone length Z = 0.5 and the modified
Takeda flexural hysteresis rule was used. The effects of different earthquake excitations on the
response quantities of the 12 and IS-storey frame-walls are investigated.
12.2 THE ROOF HORIZONTAL DISPLACEMENT AND BASE SHEAR
The roof horizontal displacement of the 12 and IS-storey frame-waIls with flexible
foundations are shown in Figs. 12.1 and 12.9 respectively. Comparing the maximum positive
and negative displacements, they show that some earthquakes cause a strong oscillation in only
one direction while others cause a similar magnitude of oscillation in both directions. This will
affect the hysteretic energy dissipation and will be discussed in Section 12.7. The variation of
the roof horizontal displacements under different earthquakes and their correlation with the total
plastic hinge rotations will be described latter. For structures with fixed base foundations the
roof displacements are given in Tables 12.2 and 12.3 respectively.
The variations of base shear resisted by frames and walls with different earthquake
excitations of the 12 and 18-storey frame-walls with flexible foundations are presented in Figs.
12.2 and 10.10 respectively. The base shear resisted by 2-WaIls and 7-frames are almost same
in some earthquakes but is very different in other earthquakes. This means that the base shear
resisted by one wall may be several times than that resisted by one-frame as is discussed in
Section 10.3.2. The variation of total base shear with different foundation types and earthquake
excitations are presented in Section 12.7.
Although the maximum ground acceleration of the 1940 EI Centro N-S is larger than
the E-W component or the 1977 Bucharest N-S components these latter two earthquakes cause
a larger horizontal displacement of the roof of the 12 and 18-storey frame-walls as shown in
. Tables 12.2 and 12.3. However, the total base shears from the latter two earthquakes are
smaller than that due to the 1940 EI Centro N-S earthquake. This is caused by the high rotation
of the base of the stmctural walls. The lower acceleration to velocity a/v ratio and the
frequency contents have a more significant effect on the response quantities of long period
structures as discussed in Ref.Z9.
324
80 3000
G420, FW12RZ Square: Frames

Modified Takedal Circle : Walls
! Trillny. : Tt.B.Sheaf
2500
Filled: Pos.Enl/. Filled Pos.Env.
60 Blank : Neg.Env. Blank Neg.Env.
2000
E E
E .
Ii 40
is" I
/\\ to
'"
.c:. 1500
.
til
/
0. I II>
0 I
l- ~
Q I
.~
\ / /
1000
20 \
~ 500
Taft Tok.Okl Prakf. Taft Tok.Okl Parkf.

0 a
40EcNS 40EcEW Such. 40EcNS 40EcEW OUch.
Fig. 12.1 Roof Horizontal Displ. vs. Earthq. Inputs Fig. 12.2 Base Shears \/s. Earthquake Inputs
Square: 40EcNS
Circle : Taft
Triang. : Tok.Okl
Diamd.: Such.
Cross : Parkf.
Plus : 40EcEW 25
Sec.FiIl.: EcNS,Fx.
X
20
X
/ :r X
r I
\" .....i...--a:::J.. +1I.O
..
tf.
c:i 15
!1
::
\
'1::
a
40EcNS
Taft
TolI.Old
Such.
Cross Parkf.
Plus 40EcEW
Sq.FlU. : EcNS,Fl(.
,16000 -10000 -5000 o 5000 10000 15000 o 20 40 60 aD

Moment ITml. Storey P.H.Rot.lx 0.001 RDd.l.
Fig. 12.3 Wall Moments Fig. 12.4 Inter-storey Drifts \/s. Storey P. Hinge Rot.
325
15 Square 40EcNS 12
Circle Taft
Triang. Tok.Okl
Diamd. Such.
Cross Parkl.
10
Plus 40EcEW
Sq.FiII.: EcNS,Fx.
.,
'0
/~\
a: 10
~
0
0
ci
~ 6
c:
..,
0
'';:;
0
n:
oJ)
4
'"c
:E 5
0:
o +-------~----_+------1_------~----_i o 5 10 15 20
left Col. Int. Col. Right Col. Walls Curvarture Ductility Demands
Fig. 12.5 Wall & Column's Base P.Hinge Rotation Fig. 12.6 Beam Curvature Ductility Demands
800 1000
Square Vise.
Circle Hyst.
~ill~d Pos.Env.
Triang. H.En.
Blank Neg.Env. BOO
600
.
..;
0:::
E
;; !:: 600
~ c::
0
~
..,a
c 400 '"
Co
.;;;
~ro .!!!
a
b >-
0:::
OJ
E'
OJ
400
'"r:::
J:
c:
w
a.:
200
200
Ec40NS Buch.
o +-----+----1_--+--1_---II--t---{ o
Taft Tok.Oki Parkl. Taft Tok.Oki Park!.
Fig. 12.7 Plastic Hinge Hotations vs. Emthq. Inputs Fig. 12.8 Energy Dissipations vs. Earthq. Inputs
326
Table 12.2 Response Quantities of the 12 Storey Frame-Walls with Different Earthquake Excitations
::
Earthquake Excitation
No. Response Rem.
40EcNS Taft,NW 40EcEW Tok.Oki NS Buch.NS Parkf.NE
Quantities
LFixed Bases
a. Pos. Env. 0.2982 0.1158 0.1920 0.3072 0.5059 0.6702
b. Neg. Env. 0.2716 0.1149 0.3293 0.3702 0.5981 0.2262
I.Flex.Foundations
a. Pos. Env. 0.2691 0.1633 0.2554 0.3425 0.5046 0.6600
b. Neg. Env. 0.3138 0.1157 0.4306 0.3210 0.6537 0.2499
2 2. Base Shear (Tonne) Fix. Plex. Fix. Flex. Fix. Flex. Fix. Flex. Fix. Flex. Fix. Flex.
1. Frames
a. Pos.Env. 809 912 282 478 833 928 912 902 877 913 837 946
b. Neg.Env. 780 926 185 573 784 823 799 888 896 948 1310 1254
2. Walls
a. Pos.Env. 1075 588 718 316 753 465 1026 809 760 609 1295 967
b. Neg.Env. 1334 897 722 340 753 542 750 370 538 242 1422 1187
3. Total Base Shear
a. Pos.Env. 1884150C 1000 794 1586 1193 1938 1711 16371522 2132 1913
b. Neg.Env. 21141823 907 913 1537 1365 1549 1258 1413 1190 27322441
3. Base Rot.(xO.OOI Rad.)
1. Fixed Bases
a. Left Column 1.378 * 0.000 2.684 4.351 13.020
b. Int. Column 1.907 * 1.525 3.000 4.732 12.212
c. Right Column 1.675 * 0.000 2.176 4.072 10.229
d. Wall
1. Flex.Foundations
4.837
"' 3.980 6.155 8.201 14.400
a. Left Colu mn 2.855 0.000 2.781 2.311 4.495 8.449

b. Int. Column 2.940 0.000 3.062 2.838 4.824 10.024
c. Right Column 2.330 0.000 1.836 2.221 4.301 9.425
d. Wall 0.000 0.000 0.000 0.000 1.972 7.665
4. Tot.P.H.Rot.(xO.001 Rad)
L Fixed Bases
a. Pos.Env.
b. Neg.Env.
188.807
156.430
"'
>I:
118.170
191.540
161.806
216.004
432.809
334.427
640.430
185.169
c. Total 345.237 * 309.710 377.810 767.23t:. 825.599
a. Pos.Env. 148.660 58.619 126.720 171.315 294.155 574.320
b. Neg.Env. 180.051 39.341 239.470 165.349 439.381 224.609
c. Total 328.711 98.020 366.190 336.664 733.536 798.929
1. Fixed Bases
a. Viscous En. 313.430
b. Hyst. En. 307.460
*>I: 205.220
195.900
432.830
402.990
395.520 302.840
537.310 518.650
c. Total En. 620.890 * 401.110 932.830 932.830 821.490
a. Viscous En. 355.000 62.100 266.680 360.000 382.100 574.320
b. Hyst.En. 172.000 8.700 154.800 200.000 385.060 224.609
c. Total En. 530.000 70.800 421.600 760.000 767.170 798.929
6. Col.Base Curv.Duct.Dmd.
(Flex. Foundations)
a. Left Column 2.166 0.000 1.999 1.754 3.278 6.439
b. Int. Column 2.219 0.000 2.187 2.079 3.447 7.540
c. Right Column 1.709 0.000 1.352 1.671 3.154 7.122
d. Wall 0.000 0.000 0.000 0.000 1.441 5.508
Remark: *: The results are not nrpco(>.n'rt.l
327
12.3 THE WALL MOMENTS, INTER-STOREY DRIFTS AND PLASTIC IDNGE

ROTATIONS
The variation of wall moments with different earthquake excitations of the 12 and 18-
storey frame-walls with flexible foundations are depicted in Figs. 12.3 and 12.11 respectively.
It shows that the effect of earthquakes on the wall moments do not show a consistent pattern.
For instance, the 1977 Bucharest N-S earthquake causes smaller wall moments in the 12 storey
frame-walls than does the 1940 El Centro N-S component although the former earthquake gives
higher response quantities for most other responses than the latter earthquake. These figures
show that earthquakes with low acceleration to velocity ratios cause a higher wall moment in
the lower storeys than in the middle storeys. This sensitivity of the lower storey responses due
to low frequency earthquake excitation was discussed in Ref. Z9. The larger earthquake, in
general, causes more redistribution of wall moments to the middle storeys.
The Parkfield N65E earthquake caused a maximum inter-storey drift of the 12-storey
frame-walls with flexible foundations of 2.28 % of the storey height. This drift close to the
limit value of 2.5 % (including P-delta and torsion effects) as suggested in NZS 4203: 1992
Section 2.5.4.5. The 1977 Bucharest NSC earthquake gave a maximum inter-storey drift of
1.44 % for the IS-storey frame-walls with flexible foundations. The variation between the
maximum inter-storey drifts and the plastic hinge rotations of the 12 and 18-storey frame-walls
in both direction of displacements are presented in Figs. 12.4 and 12.12 respectively. These
show that the storey plastic hinge rotations starts to develop at a particular inter-storey drift and
then increase with the inter-storey drift. The storey plastic hinge rotation tend to increase as
the inter-storey drifts increase.
When the structure strongly displaces in one direction, the total plastic hinge rotations
in the positive and negative envelopes will different from each other. As a result, a relatively
large maximum inter-storey drift will be related to different plastic hinge rotations as illustrated
in Figs. 12.4 and 12.12. For instance, the 1940 El Centro EWC, the 1977 Bucharest NSC and
the 1966 Parkfield N65E earthquakes cause a strong displacement to one side in the 12-storey
frame-walls as shown in Fig. 12.1 giving two different beam plastic hinge rotations for the
same storey drift as shown in Fig. 12.4. Similar results are found for the I8-storey frame-walls
and are shown in Fig. 12.12.
328
Table 12.3 Response Quantities of the 18 Storey Frame-Walls with Different Earthquake Excitations
No. Response Rem.
40EcNS Buch.NS 40EcEW PaM. 1 Parkf.2 Taft. 1
Quantities
1. Roof Hor. DiNpl.(m)
1.Fixed Bases
ll. Pos. Env. 0.2419 0.4807 0.2447 0.4822 0.5929 0.3202
b. Neg. Env. 0.3001 0.6933 0.4906 0.2278 0.2555 0.2885
1.Flex.Foundations
a. Pos. Env. 0.2954 0.3944 0.2731 0.4506 0.5647 0.3513
b. Neg. Env. 0.2937 0.6896 0.5669 0.3037 0.3386 0.3392
2. Base Shear (Tonne) Fix. Flex. IFix. Flex. Fix. Flex. Fix. Flex. Fix. Flex. Ipix. Flex
1. Frames
a. Pos.Env. 7101257 6501163 1000 1185 794 1250 1033 13Hi >I< 1347
b. Neg.Env. 963 1301 8881056 605 1209 12991198 13771299 lie 1377
2. Walls
15401300 1638 783 1076 752 1948 880 2000 1029 1468
a. Pos.Bnv. '>I:"
h. Neg.Env. 22951323 7761243 1466 752 1567 804 1926 956 1660
3. Total Base Shear
a. Pos.Bnv. 22502551 228819M: 20761937 27422130 30332345 >I: 2815
b. Neg.Env. 32582624 1664 2299 2071 1961 2866 2002 3303 2255 >I< 3037
3. Base Rot.(x 0.001 Rad.)
1. Fixed Bases
a. Left Column 0.000 0.000 0.000 2.450 3.620 0.000
h. lnt. Column 0.000 0.000 1.157 3.412 4.530 1.410
c. Right Column 0.000 0.000 0.000 3.200 4.290 0.000
d. Wall 3.030 3.581 3.745 5.380 5.930 4.050
1. Flex.Foundations
a. Left Column 2.060 2.220 2.740 2.414 2.249 2.459
h. Int. Column 2.620 2.614 3.023 2.882 3.790 3.700
c. Right Column 1.985 0.000 1.742 1.592 3.123 3.081
d. Wall 0.000 0.000 0.000 0.000 0.000 0.000
4. Tot.F.H.Rot.(xO.OOl Rad)
1. Fixed Bases
a. Pos.Env. 132.954 250.622 140.470 160.900 438.094 208.132
b. Neg.Env. 142.528 333.612 282.390 338.940 204.495 169.544
c. Total 275.482 584.23~ 422.860 499.840 642.589 377.675
a. Pos.Env. 172.880 166.89( 141.744 250.430 354.73C 220.410
b. Neg.Env. 167.100 381.96( 294.770 156.900 204.71C 201.560
c. Total 339.880 548.85( 436.514 407.330 559.44C 421.970
5. Energy DiNsipation (Tm)
1. Fixed Bases
a. Visoous En. 306.670 41O.44( 270.890 244.000 339.55(J 293.280
b. Hyst. En. 226.670 421.65( 266.420 333.610 500.000 266.420
c. Total En. 533.330 S32.09C 537.310 577.610 839.55C 559.700
a. Visoous En. 394.100 339.70< 289.200 279.900 402.98(] 543.300
b. Hyst.En. 134.300 180.90< 134.300 123.100 223.88(] 185.100
c. Total En. 528.400 520.60( 423.500 403.000 626.87(] 728.400
6. Col.Base Curv.Duct.Dmd.
(Flex.Found.)
a. Left Column 1.606 1.350 1.847 1.459 2.330 1.874
b. lnt. Column 2.530 2.012 2.531 2.196 3.053 3.198
c. Right Column 1.480 0.000 1.215 1.401 2.145 2.186
d. Wall 0.000 0.000 0.000 0.000 0.000 0.000
Remark: -
329
12.4 COLUMN, WALL BASE PLASTIC HINGE ROTATIONS AND BEAM

CURVATURE DUCTILITY DEMANDS
The variation of column and wall base plastic hinge rotations with different earthquakes
for the 12 and 18-storey frame-walls with flexible foundations are shown in Fig. 12.5 and
12.13 respectively. The maximum rotation is 0.0102 radians and is much less than the possible
maximum limit of 0.035 radians discussed in Section 9.5.4.5. It is also shows that the rotation
of the wall footings causes a significantly increase in the plastic hinge rotations in the column
bases but a great decrease in the wall base plastic hinge rotations for most earthquake
excitations. The comparison between the columns and wall base rotations of the structure with
fixed bases and flexible foundations will be discussed in Section 12.7.2.
As shown in Fig. 12.6, the effect of the foundation flexibility on the beam curvature
ductility demands of the 12-storey frame-walls induced by the El Centro N-S, Tokachi Oki,
Bucharest NSC and Parkfield earthquakes decrease by an average of 4.15 %, 5.12 %, 3.32 %
and 8.27 % respectively. In contrast, the 1940 El Centro E-W causes an increase of beam
curvature ductility demands by an average of 13.16 %. Increases in beam curvature ductility
demand due to foundation flexibility is found in the 18-storey frame-walls exited by the 1940
EI Centro N-S component as discussed in Chapter IX. This result is likely to be affected by
the spectral characteristics of the corresponding earthquake excitation at the fundamental period
of the structure as discussed in Ref. V 10.
The beam curvature ductility demands of the 18-storey frame-walls with flexible
foundations are shown in Fig. 12.14. Due to foundation flexibility, the beam curvature ductility
demands of structures subjected to the El Centro E-W, Bucharest, Parkf.l and Parkf.2
earthquakes decreased by an average of 13.95 %, 16.43 %,33.56 % and 23.66 % respectively.
In general, it is difficult to describe types of earthquake that always cause higher response
quantities in any type of structures. However, Figs. 12.6 and 12.14 show that the earthquakes
with a low value of the a/v ratio such as the 1940 EI Centro E-W, the 1977 Bucharest NSC
and the Parkfield N65E components tend to cause higher response quantities than those for
earthquakes with a moderate value of the a/v ratio such as the 1940 El Centro N-S and Taft
earthquakes.
330
80 4000
Square Frames
Circle Walls
3500 Triang. n.B.She!!f
RIled : POll.En".
Blank l Nell.En".
60 3000
E 2500
~
is.
1/1
E
i5 1;;
5
40
'"
.c 2000
!II
::c .,
.... III
~
"
0
I>:: 1500
20 1000
600
Buch. Parkf.1 Such. Parkf.1 Toft.1

0 0
40EcNS 40EcEW Parkf.2 1.I0EcNS 40EcEW Palkf.2
Fig. 12.9 Roof Hor. Displacements VS, Earthq. Inputs Fig. 12.10 Bass Shears vs. Earthquake Inputs
20 - , - - - - , . . - - - - ; - - - - - - , - - - - - - ,
Dillmd.: Taft
Tr.FIII. : Pelltt2.
Sq.Fill.: 40EcEW
25000 -15000 5000 5000 15000 25000 o 10 20 30 40

Moment (Tml St. P.H. Rot.lx 0.001 Rud.J.
Fig. 12.11 Wall Moments 12-Wallsl Fig. 12.12 intGr"storey Drifts vs. Storey P.Hinge Rot.
331
Square EC40NS
Circle Such.
Triang. Parkl1.
5 Cross Parkf2.
Diama. Taftl. 14 +..--........X~:--i,
Plus EC40EW
Filled EC40NS,Fx.
4 - 12 +--.--.....----.--.----.-X...--.. !--.~'SIII
.,;
OJ
rr. 10 + .... _)< ............f'
"""
0
0
ci 3
40EcEW
..'5
..:
8 x. --..--{\ Buch .
'"i"
Parkf.l
Parkf.2
0.: 40EcNS
2
Filled Fixed
4 ,...;11Z?jj~1l--..~BI~a~n;kll~~FI;e~x.II~
o o 5 10 15
Left Col. Int. Co!. Right Col. Walls Curvature Ductility Demands
Fig. 12.13 Wall & Column's Base P.Hinge Rotation Fig. 12.14 Beam Curvature Ductility Demands
500 1000
------------- P. Env
Square Vlsc.
Circle Hyst.
-Q--N.Env Triang. Tt.En.
400 800
"C
E
300 !:: 600
'"
0::
~
Cl
..,"
C
q '"
c.
Cl 'r;
..'5
.; i5"
>-
"
0: 2'
:r; 200
C'"
400
ll. W
100 200
Buch. Parkf.l Taft. 1 Such. Park!.1 Taft. 1

o +---~--t---+---t----tI__+__--+-___1 o ~---r---+--f---~-~
40EcNS 40EcEW Parkf.2 40EcNS 40EcEW Parkf.2
Fig. 12.15 Plastic Hinge Rotations vs. Earthq. Inputs Fig. 12.16 Energy Dissipations vs. Earthq. Inputs
332
Square 40EcNS
Corele 4{)EcEW
O.B Triang. Tok.Oki
Diamd. Buch.NS
Sq.FiII. Parkfield
'"
OJ
'tI
DIm = 12.899 X
S. 0.6
'"
"'"E
a'"
..,l; 0.4-
.,e
~
0.2
o
o 10 20 30 40 50 60 70 30
Member Curvature Ix 0.001 Rod./ml.
Fig. 12.17 Member Damage Index vs. Member Curvature
Square 40EcNS
Circle 40EcEW
0.8 Triang. Tok.Okl
Oiamd. Buch.NS II1II
l1li 0
Sq.FiIl. Parkfield
"'"
"CI
.5 O.S
(!)
.,'""E
0
.. 0
'"
.c 0.4
.,
E
::E D.
0.2
o
o 2 3 4 5
Member Hyst. Energy Dissipation ITm)
Fig. 12.1 B Member Damage Index vs. Mambaf Hyst. Energy Dissipation
333
12.5 TOTAL PLASTIC HINGE ROTATIONS AND ENERGY DISSIPATION
The variation of total plastic hinge rotations with respect to earthquake excitation of the
12-storey frame-walls with flexible foundations is shown in Fig. 12.7. Comparing Figs. 12.1
and 12.7 indicates that the variation of maximum floor displacements and total plastic hinge
rotations with respect to the earthquake excitations have a similar pattern. A similar pattern of
variation is also found in the case of the 1S-storey wall-frames by comparing Figs. 12.9 and
12.15. Hence, these response quantities are significantly correlated and equally important as
damage parameters in inelastic structural analyses. The roof horizontal displacements and total
plastic hinge rotations of the 12 and IS-storey frame-walls with fixed base foundations are
given in Tables 12.2 and 12.3 respectively. The relationship between the roof displacements
in terms ofthe overall drift of the structures and plastic hinge rotations is shown in Fig. 12.41.
The variation of the dissipated energy versus earthquake excitation for the 12-storey
frame-walls with flexible foundations is illustrated in Fig. 12.8. As mentioned in Section 12.2,
the magnitude and type of building displacements affects the hysteretic energy dissipation. A
relatively. large roof displacement combined with a relatively symmetric building oscillation
may give higher hysteretic energy dissipations. For example, Tokachi Oki earthquake gives a
relatively symmetric oscillation as shown in Fig. 12.1 and gives a higher hysteretic energy
dissipation than that induced by the 1940 El Centro E-W earthquake as shown in Fig. 12.8
although the maximum roof displacement caused by the former earthquake is less than that the
latter earthquake. Similar results are found in comparing the roof displacements and hysteretic
energy dissipations of the I8-storey frame-walls subjected to the Taft.l and Bucharest
earthquakes as shown in Figs. 12.9 and 12.16 respectively. Tables 12.2 and 12.3 show that the
hysteretic energy dissipation of rocking structures is much less than those with fixed bases
while the total plastic hinge rotation may decrease or increase with the variation of the
earthquake excitations.
12.6 MEMBER, STOREY AND STRUCTURAL DAMAGE INDICES
As discussed in Section 9.3 several approximate formulae for member damage indices
are based on the expected damage mode of the member and related to a single maximum
response, accumulation of response or combinations of these damage parameters. As mentioned
earlier the damage index proposed by PARK & ANG (1984) with (J = 0.05 is used in this
study_ PARK et a1.(1987) stated that to get a reasonable determination of the storey and
334
12 30
r_
!Squllre 40EeNS
Circls 40EcEW
Tria"g. Tok. Old
10 --Diamd. Buch.NS
-I Biy.Sq. : Parkfield
Fixed 60
: Flex.
~
B
'"
~t
IJ
'l'I
6
....!!!'"
til
Q
>-
Storey E!'
OJ
4- ..
.5
!:!.
1:
20
2
40ECEW Such.IIIS
o 0.2 0.4 0.6 0.8 o +---+---~-_4--_+---_+----~
Damage Index. 40EcNS Tok.Okl Parkf.
Fig. 12.1 () Storey Damage Indices fig. 12.20 Hyst. En. Dissipations VII. earthq. Inputs
1000 - , - - - - - - - - - - - - - - - - 0.65 -,------------------~
0.55
SOO
~
Ii
.,,;
~
'0
.5
... 0.45
600
....
g til
en
0 ~
.::s Q
III 0.35
c;
a:; l!::0
:i 40Cl ~
S ~//
it til
0.25
200
0.15
40EeEW Parkf.NE 40EcEW Parkf.NE

0.05
40EeNS Tok.Oki Buch.IIIS 40EeNS Tok.Okl Buch.IIIS
Fig.12.21 Plastic H. Rotations vs. EarthQ. Inputs Fig.12.22 Structural Damage Indices VII. Ellrthq. Inputs
335
18 ....--I'VJIIIIXII. 40EcNS 100

Tok.Oki
iTriang. : 40EcEW Takeda
Diamd. : Parkfield
Big Sq.: Buch.IIIS
BO
14
l~~~~~ fixed
: fie);.
E _.(J
t;
c
60
8.,
c.
'iii
.!!!
C
>-
.,:'
t:
w 40
...
III
:r>-
0
20
Tok.Okl Parkf.
o
o 0.1 0.2 0.3 0.4 0.5
40EclllS 40EcEW lluch.NS
Damage Index
Fig. 12.23 Storey Damage Indices Fig.12.24 Hyst. Energy Dissipations vs. Earthq. Inputs
1250 . - - - - - - - - - ...............~--------------- 0.5 , ..- - - - - - - - - - - - - - - - -
IFilled Fixed
~Iank Flex.
1000 0.4
-0 '"'"
"C
.E 0.3
750
...'"
IX:.
<:>
<II
.,'"
<:> E
.0 0"
~
..;
0
a:
e :::l
500 1) 0.2
:i
a: E
(/)
250 0.1
Tok.Oki Parkf. Tok.Oki Parkf.

o
40EcNS 40EcEW Buch.NS 40EcNS 40EcEW Buch.IIIS
Fig.12.25 Plastic Hinge Rotations vs. Earthq. Inputs Fig.12.26 Structural Damage Indices vs. Earthq. Inputs
336
300
10 +_t.()i
600
E
!::
.,
0::
*''""-
6 .,
'iii
a 400
>-
~
1 Square 40EcN-S '"

0::
UJ
4 r Circle Tok.Okl .:OJ
i Triang. 4OEcE-W >-
J::
i Diamd. Parkfield
!
._._...._..i.
Big S'Ir.: Buch.N-S 200
2
!Filled Find
l Blank
o Tok.Old Parllf.
o 0.1 0.2 0.3 0.4 0.6 0.6 0.7 o +---~--~----+---~---+--~
Damage Index. 40EcNS 40EcEW Buch.NS
Fig. 12.27 Storey Damage Indices Fig. 12.28 Hyst. En. Dissipations vs. Earthq, Inputs
1000 ,--------------------------------- 0.5 ,-------------------------------,
BOO 0.4-
'IJ
e: >(
<;
0 '"
'IJ
.5 0.3
0 600
e
>(
"
Ell
2 "
E
..," to
Cl
0
IX:
." ~
::I
'"
.5
J::
400 'U 0.2
2
c.: (jj
Ol
~
200 0.1
Tok.Oki Parkf. Tok.Okl Park I.

o +----+----+----+----+----+--~ o +-----+-----~--~~--~----_+----_o
40EcNS 40EcEW Such.fIIS 'WEeNS 40EcEW Such.
Fig. 12.29 Plastic Hinge Rot. Vii. Earthq. Inputs Fig. 12.30 Structural Damage Index vs. EarthQ. Input
337
1B _----c---[J-~ ___I\_-___c>_,_ ~~~-
16 16
14 + '"''""",'_''''''' -feY 14-.t.--,'"'''''--,-~",,,r~~:,,--,~~{)~''''''''~''''-'''''''"".. ,"
Storey
Storey
12 12 ...""---""""'",,,,,,+n
10 10
WEeN-S Square 40EcNS

lluch.N-S 6 ",,,,,: Circle Buch.NS
15
4OEcE-W Triang. 40EcEW
Parkfield 1 Diamd. Parkfield 1
Parkfield 2 1IIi+,_".. ,,"""",_~Sq.FiII. Parkfield 2
4- Taft 1 Cr.FiII. : Taft 1
o 0.1 0.2 0.3 0.4 o 0.1 0.2 0.3 0.4
Damage Index. Damage Index.
850 0.35 -~---------------~
0.3
650
.,
>< 0.25
'C
.,; .s
III
a:
... 01
t:I!
0
0 E'"
0"
ci 450 0.2
~
..:0 ro
a:: ...:;
::C
a.: ..."2
(J) 0.15
250
0.1
Buch.NS Parkf.1 Taft. 1 Buch.NS PaTkf.1 Taft 1

50 +--+--- -+--+1 ---t---t-----+I"'----i 0.05 I--+-- ,-+----I----+--l--~
40EcNS 40EcEW Parkf.2 40EcNS 40EcEW Parkf.2
Fig. 12.33 P .Hinge Rotations vs. Earthq. Inputs Fig. 12.34 Structural Damage Indices vs. Eathq. Inputs
338
structural damage indices weighting factors need to be used. KUNNATH et al.(1990) stated
that the weighting factor based on hysteretic energy in Eq. 9.10 is used for the weak column
and strong beam failure mechanism and a weighting factor for the strong column and weak
beam needs to be defined.
For this purpose the relationship between the member curvature and member damage
index and the variation of member hysteretic energy dissipation and member damage index are
plotted. These relationships are required to determine which parameter strongly affect on the
member damage index. The variations of member curvature versus member damage index of
the 12 storey frames with fixed bases is shown in Fig. 12.17, while the relationship between
the member hysteretic energy dissipation and member damage index is shown in Fig. 12. 18 .
Similar relationships for the 18-storey frames and frame-walls on the flexible foundations and
fixed bases are shown in Appendix C.
Table 12.4 Damage Indices of the 12 and 18-storey Frames wilh Different Earthquake Excitations
40EeNS 40EcEW Tok.Oki Parkf.NE Buch.NS
Systems
1.Fixed Base Found.
a. Max.Member D.Index 0.3680 0.4550 0.3800 0.7410 0.7920
b. Max.Storey D.lndex 0.3109 0.3742 0.3584 0.6347 0.6299
c. Structural D.lndcx 0.2585 0.2892 0.2933 0.4572 0.4923
2.Flex.Foundations
a. Max.Member D.lndex 0.3250 0.4510 0.3490 0.7280 0.7520
b. Max.Storey D.lndcx 0.2899 0.3707 0.3452 0.6105 0.6106
c. Structural D.lndcx 0.2443 0.2841 0.2798 0.4441 0.4751
1.Fixed Base Found.
b. Max.Storey D.lndex 0.2186 0.2689 0.2069 0.4370 0.4255
c. Structural D.lndex 0.1698 0.2076 0.1713 0.3212 0.3368
2.Flex.Foundations
b. Max.Storey D.Index 0.2105 0.2413 0.1928 0.4201 0.4334
c. Structural D.lndex 0.1538 0.1930 0.1476 0.3034 0.3185
Remark: -
These figures show that the relationships between hysteretic energy dissipation in the
member and member damage index are very scattered especially at large member damage
indices. On the other hand, the relationships between the member curvature and the member
damage index are very strong. Some anomalies in the relationships appear in the figures, and
are likely due to the low amplitude cyclic reversals that cause a relatively large member
hysteretic energy dissipation.
339
These results imply that the member damage indices are more strongly affected by the
member curvatures than the member hysteretic energy dissipation. These results are strongly
dependent on the selection of the member deterioration factor fJ. A relatively small (J is
associated with a ductile member where the damage is due to an excessive member
deformation, while the high fl corresponds to 11 brittle member where the damage is governed
by hysteretic energy dissipation. Hence, for a ductile RC building designed for strong column
and weak beam failure mechanism a weighting factor based on the member curvature is
suggested. The weighting factor Ai is written as,
(12.1)
where rpj is the curvature of member j or the total curvature of storey j and N is the number
of members or the number of storeys.
The storey damage indices of the 12 and 18-storey frames with rigid joint zone length
Z = 1.0 are shown in Figs. 12.19 and 12.23 respectively. It shows that an earthquake with low
acceleration to velocity, a/v, ratio as given in Table. 12.1 gives higher storey damage indices
in the lower stories than those in the upper stories. These results agree with the wall moments
as discussed in Section 12.3 and the results reported by ZHU et a1.(1992). The storey damage
indices of structures with flexible foundations are smaller than those of structures with fixed
bases in all storeys and for all earthquake excitations. Varying the earthquake excitation, the
storey damage indices decreased by the range of 2.79 % to 11.55 % with an average of 6.16
% for the 12-storey and 4.78 % to 14.33 % with an average of 8.39 % for the 18-storey
frames. In Fig. 12.19, the maximum storey damage index for the frames subjected to the
Parkfield earthquake is greater than that caused by the Bucharest earthquakes. However, the
Parkfield earthquake gives a lower structural damage index than the Bucharest earthquake
because the structural damage index is affected by the distribution of storey damage indices in
all storeys. Using 34 seconds of the earthquake duration, the Tokachi Oki earthquake provides
a greater hysteretic energy dissipation than 20 seconds of the 1940 El Centro NS earthquake.
The effect of earthquake duration will significantly affect the hysteretic energy dissipation in
the structure and was discussed in Ref. TIO and Z3.
As shown in Figs. 12.22 and 12.26 the structural damage indices of frame structures
with flexible foundations subjected to different earthquake excitations are smaller by an average
340
of 4.02 % for the 12 storey frames and 8.18 % for the 18-storey frames when compared with
those for fixed bases. Comparing Figs. 12.20, 12.21 and 12.22 for the 12-storey frames and
Figs. 12.24, 12.25 and 12.26 for the IS-storey frames indicate that the variations of structural
damage indices are more closely correlated with the total plastic hinge rotations than the
hysteretic energy dissipations.
Table 12.5 Damage Indices of the 12 and 18-storey Frame-Walls with Different Earthquake Excitations
Earthquake Excitations
40EcNS 40EcEW Tok.Oki Parkf.NE Buch.NS -
Systems
l.Fixed Base Found.
c. Structural D.Index 0.1948 0.1786 0.2169 0.4219 0.3632
2.Flex.Foundations
c. Structural D.Index 0.1700 0.2119 0.1817 0.3914 0.3608
Earthquake Excitations
No Structural and Damage Remark
Systems 40EcNS Buch.N~ 40EcEW Parkf.l Parkf.2 Taft. 1
The 18-storey Frame-Walls

l.Fixed Base Found.
a. Max.Member D.Index 0.2210 0.4680 0.3270 0.4080 0.4590 0.2960
b. Max.Storey D.Index 0.1889 0.3671 0.2489 0.3085 0.3567 0.2388
c. Structural D.Index 0.1167 0.2488 0.1709 0.2021 0.2415 0.1542
2.Flex.Foundations
a. Max.Member D.lndex 0.2420 0.4290 0.2950 0.2870 0.3720 0.2770
b. Max.Storey D.Index 0.2132 0.3333 0.2338 0.2344 0.3015 0.2493
c. Structural D.Index 0.1377 0.2258 0.1560 0.1490 0.1991 0.1619
Remark: -
The variation of the storey damage indices of the 12 and IS-storey frame-walls with
different earthquake excitations are presented in Figs. 12.27, 12.31 and 12.32 respectively. In
general, the storey damage indices decrease as the foundation flexibility increases. As shown
in Figs. 12.30 and Fig. 12.34 the structural damage indices of structures with flexible
foundations decreased by 0.9 % to 16.3 % for the 12-storey and 4.7 % to 26.3 % for the 18-
storey frame-walls. However, the 1940 EI Centro E-W component gave an increase in the
structural damage index by 22.6 % for the 12 storey frame-walls and the 1940 EI Centro N-S
and Taft. 1 earthquakes caused an increase in structural damage indices by an average of 11.25
%. As mentioned previously, the spectral characteristics of earthquakes significantly affect the
structural responses. The effect of acceleration to velocity ratio on the structural responses
cannot clearly be identified because the earthquakes are not uniformly normalized as carried
out in Ref.T12, T20, Z6, Z7 and Z9.
341
12.7 OVERALL RESPONSE QUANTITIES OF STRUCTURES
12.7.1 Base Shear
The base shear of the 12-storey frames and frame-walls with flexible foundations are
smaller than those for the flXEul base foundations as the earthquake excitations is varied as
shown in Figs. 12.35 and 12.36 respectively. These agree with the previous discussion and
confirm both the experimental and analytical investigations discussed in Chapter II. The base
shear decreased by an average of 1.82 % and 8.76 % for frames and frame-walls respectively.
This shows that the stiffer structures such as frame-walls are more affected by foundation
flexibility than are: frame structures. This result implies that the soil-structure interaction is
more significant for relatively stiff structures.
12.7.2 Plastic Hinge Rotation of Column and Wall Bases
The variation of column base and wall rotations as the earthquake excitations are varied
for both types of foundations ofthe 12 and 1S-storey frame-walls are shown in Figs. 12.37 and
12.38 respectively. In relatively small earthquakes, the column base plastic hinge rotations in
the frame-walls with flexible foundations are relatively high due to rotation of the wall footings.
Examples of the moment versus curvature relationships of columns and wall bases are shown
in Figs. 10.81, 10.82, 10.83 and 10.84 respectively. For fixed base structures the column
plastic hinge rotations are relatively small or even fully elastic because a large amount of the
flexural force is resisted by the walls. Examples of the moment curvature relationships of
columns and wall bases are shown in Fig. 10.77, 10.78, 10.79 and 10.80 respectively.
During a large earthquake, the column plastic hinge rotations of the frame-walls with
fixed base foundations are larger than those for flexible foundations. This is because the fixed
base columns have already undergone large inelastic deformations and the rotation of the
foundation footings as shown in Fig. 12.43 slightly reduce the plastic hinge rotation of the
column bases. However, at what earthquake level the rotation of the column bases of the
structure with fixed bases exceeds than that of the flexible foundations is difficult to defme.
Compared with the laboratory tests reported by WATSON and PARK (1994) the maximum
column base rotations of 0.0145 radians seen in Fig. 12.37 are nearly the ultimate rotation.
342
200 3000
Square Pas. En".

175 Circle Nag.En".
Square Pas. En". 2500
Circle Neg.En". Filled Fixed
Blank Flex.
150 Filled Fixed
Blank Flex.
2000
E 125 E
1;; .,
~
OJ
""
U)
"
.e
U)
OJ .,'"
'"'"
III
100
'"
fiI
1600
75
1000
50
TolI.OIII 40EcEW Parkf. Tok.Okl 40ECEW Porltf.NE

25 +---~--~----+---~--~----+---~ 500
Taft 40EcNS Bueh.NS Taft 40EcNS Huch.NS
Fig. 12.35 BaSil Shears liS. Earthq. Inputs Fig.13.36 Tot. Base Shears liS. Earthq. Inputs
20 B
Square Left.CoI. Square Left Col.

Circle Int. Col. Circle Int. Col.
Traing. Right Col. Trian!!. Right Col.
Dlamd. Wall o;amd. Wall
15 6 Filled Fixed
Filled Filled
Blank Flax. Blank FlaK.
,; ,;
a::'" '"
...
0
G420, FW12RZi
Mod. Takeda
a::
.-
0
0 0
d d
..
~
0
10 ~
,.;
0
a::
4
IX:
Q)
.,
'"
r:: '"
.5
J: J:
.,:
n:
5 2
Bueh Taft. Parkf.2

o +----+~~~--+_--~--~~--+=--~
'WEeNS Such. Parkf. 40EcNS 40EcEW Parkf.l
Fig. 12.37 BaSI:! P .Hinge Rot. vs. Earthq. Inputs Fig. 12.33 Base P.Hinge Rot. liS. Ellrthq. Inputs
343
20 -,----------~ ..... - - - - -
17.6
0.3
15
0.25
.
.,
B '"E
0
~ 0
.... 1! E
m 12.5
I>- '1J
At. D !f
/\
0.2
0 i '"
u"
0
iD 0
10
i
iD
...
0.15
... 7.5
0.1 5 ~--+---~-+----+---+---
F12 FW12 F18 FW1B F12 FW12 f18 fW18
Fig. 12.39 Fund. Period \IS. Struct, Prototypes Fig. 12.40 CoeH. Omega \IS. Struct. Prototypes
0.8
1000
: 12 St.Frm.
: 18 St.F,m.
0.7 : 12 St Fr-W 1II..t... 1
: 18 St.Fr-W
0.6 : Fixed
: Flex.
!(
..'"
"CI
11: 0.5
t>l
OJ
E
OJ
0 0.4
1'!
a.,
E 0.3
II)
.
0.2
0.1
2 0.2 0.6 1.4 1.B

Overall Drift ( % I
Fig. 12.41 P .Hinge Rotations vs. Overall Drifts Fig. 12.42 Structural Damage Index vs Overall Drift
344
12,7.3 Vertical Displacement of Footing Foundation
The footing foundation impedances are functions of the foundation geometlY.

Decreasing the soil contact due to footing foundation lift-offwill cause an reduction in the soil-
foundation impedances. Checking the soil-foundation status during the loading history is
necessary. The history of the vertical displacement of the outer tips of the footing of the 18-
storey frame-wall foundations subjected to the 1940 El Centro N-S component with the
differing soil shear modulus G is shown in Fig. 12.44. It shows that there are different values
of initial vertical displacements due to the difference of the vertical stiffnesses of the soils. The
smaller initial displacement due to the larger vertical stiffness of the soil may lead to foundation
tip lift-off as shown in the figure. The durations and the amounts of vertical tip lift-off are
relatively small compared with the earthquake duration and the effects on the structural
response are negligible.
The outer tip vertical displacement histories of the wall footings of 12 and 18-storey
frame-walls footings subjected to different earthquakes are shown in Figs. 12.45 and 12.46
respectively. The initial vertical displacements are equal because they used the same soil
vertical stiffnesses. The effect on the structural response of a relatively small foundation tip lift-
off when subjected to the 1977 Bucharest N-S component are also insignificant.
12.7.4 Fundamental Period of Structures
The ratio of fundamental period of free vibration of structures Tl and the number of
storeys N versus number of storeys for the 12 and 18-storey frames and frame-walls are shown
in Fig. 12.39. It shows that the ratio tends to decrease as the number of storey and the stiffness
of the superstructure increases. There is scatter because of involvement of the effects of
foundation flexibility, rigid j oint zone lengths, soil shear strains, shear moduli and the number
of frames. As with the results reported in experimental and analytical investigations discussed
in Chapter II, the flexibility of the foundations cause an increase in the fundamental period of
the structures.
The variation of structural fundamental periods with respect to building height Hb

including the results reported in Ref. DB and C17 are shown in Figs. 12.48, 12.49 and 12.50
respectively. The fundamental period T. of structures with flexible foundation according to Eq.
6-3 of the ATC 3-06 (1978) appears to overestimate when compared with results of this study
0.0008 ..,-----.....,.---------,-----.,----------,-------..., 0.01
0.000 6 + ..................................+ ........................i\..........-+..._.............._...............+.-.......- ..............- ........;........._,__..............1

0
0.0004
]
. 0.0002. .l............................ ...
!: 0.01
C
'0 '"
E
a: 'c"
u
.!E
~ -0.0002. +.................. -\---..1* is
..
C. 0.02
"a
..:l"'" -0.0004 +..................................;'" 'f" -0.03
>'" Sollel thin line : G = 540
-0.0005 +.......................................
: Parkfield IFW12lU1 Dot line G = 5135
0.04 +........__ .............. +.............._.._..I Soliel thick line: G = 825
.._......................,Dot line : Tok.Ok! IFW12lU1
-0.0008
long clash Parkf.1 IFW18AZ1
Solid thick line: Parkfr.2 IFW18AZ1
-0.001 .0.05
Ali). 12_43 RotatiOl'll of Column'. Footing
0.01 0.01
0 0
] ]
C -0.01
....,
!:
-0.01
'"
E E
III
"''""
C. -0.02 "'"
C. -0.02
Ci" i5"
c;; c;;
'f" 'f" 0.03
>" >'"
-0.04 -0.04
Dash line Parkf.NE

SoRd thick line: Bueh.fIIS
.0.05 0.05
Fig. 12.45 V,rtiC:lI 1Oi..,lacclmlnt of Wall Footing) Tlpilll

346
1.6 Ii -r----------- -------,

ATC 306 (1978)
1.5 II
1.4
1.3 4 ATC 306 (19781

1.2 II
1.1 d TI = 0.160*(H b )o.7S
,(
u:: 1
$l
0.9
'"
OJ
!;!;
0.11
0
'$:l
&! 0.7
".ig.. 0.6
Q..
0.5
0.4 -__ "i'H_H'~H __!_"_H,_.. _~IH'H'_"'H! ~
r-
0.3 ?:Q
0.2 +~H'H....t"_'-'''--''1'''''H'''H ..tH'''=~'1'H'~-~_-''''iH'H_'''H'~_'H'H ... j * Oillmd. {fixed! : Ref. C17
0.1 Cross lfixed!: Ref. B13
I}
.2 2.25 2.5 2.75 3 3.26 3.5 3.75 " o 20 60 80 100

fund. Period T1 Building Haight Hb (metree!
Fig. 12.47 Period Ratios vs. Fundamental Period Fig. 12.48 Fund. Period vs. Building Height
5.,-----------------,
ATC 3-06 (1978)
II
x
4 4 + //
ATC 3-06 11978! FW18

FW18/
II
/
FWl.2 FW12
t
Frame-Wa1ls (Fbu;d Bases) Frame-Walls IFlel(. Found.1
o 20 40 60 110 100 o 20 40 60 ao 100

Building Height Hb (metres) Building Height Hb (metres)
Fig. 12.49 Fund. Period vs. Building Height Fig. 12.50 Fund. Period vs. Building Height
347
especially for intermediate soils as shown in Figs 12.48 and 12.50, but relatively close for soft
soils as indicated by cross mark shown in Fig. 12.50. The RC structures as reported in
Ref.Cl7 seem stiffer than those in this analysis. The ratio of the fundamental period of the
structures on the flexible foundations and on the fIxed bases are shown in Fig. 12.47. Again,
the stlUctural fundamental period ratio suggested by ATC 3-06 (1978) overestimates the result
of this study. Fig. 12.47 also shows the structural fundamental periods normalized in terms
H bo.7s They show that these normalized value are almost constant in every type of bUilding.
Hence, the structural fundamental period of structures can be expressed in terms of 1\0.75 by
a coefficient. These fundamental periods Tl of inelastic RC structures can be expressed by the
relationships,
T = C HII.?S (12.2)
1 t f;
where Hb is the height of building in metres, C t is a coefficient depending on the types of

foundations and structural lateral load resistance system and Tl in seconds. The value of ~ =
0.145 and 0.130 for frames and frame-walls with fIxed base foundations respectively are
suggested. In the case of structure with flexible foundations the value of ~ = 0.160 and 0.155
are suggested for frames and frame-walls structures.
12.7.5 Coefficients of Wm
As discussed in Section 12.6 the member damage index is dependent on the member
energy dissipation and the member curvature. However, the result of this study indicates that
member damage indices are strongly correlated with the member curvature. For simplicity, the
member damage index can, therefore, be conveniently expressed in terms of member curvature
with the linear relationship as,
(12.3)
where DIm is the member damage index, Wm is a coefficient and cPm is the member curvature.
A linear regression on the relationships between the mem~ damage indices and
member curvature as shown in Fig. 12.17 and Figs. C.l, C.3, C.S, C.7, C.9, C.lt and C.12
have been made. Regressions cover 480, 760 and 912 member damage indices and earthquake
excitations for the 12 and 18-storey frames and wall-frames respectively. In all structural and
foundation systems, the regression lines almost exactly pass through the origin, meaning that
no member damage at zero member curvature. It will be proved latter that the member damage
348
index can be linearly correlated with the member curvature by the coefficient Wm These
coefficients and the corresponding coefficient correlations are presented in the following table.
Table 12.6 Coefficient Wm
Fixed Base Foundations Flexible Foundations

No. Str. Systems Average Wd
Coeff. Wm coer. VaT. CodY. Wm Coef.Var.
1. Frame Structures
a. 12-Storey 12Jl99 0.9956 12.916 0.9930

13.25
b. 1 B-storey 13.910 0.9527 13.702 0.9589
2. Frame-Walls
s. 12-storey 10.705 0.9527 11.100 0.9442

10.50
b. 18-storey 9.664 0.9294 10.620 0.9776
The table shows that the coefficients of correlation are nearly unity. Results of at-test
based on the statistical analysis concluded that the member damage indices are linearly
correlated with member curvatures. As in Section 12.6 using a relatively small value of {3 =
0.05 causes the damage indices to be strongly correlated with the member deformations. The
coefficient Wm for frame structures as given in the above table may give an upper bound value
because the member damage indices are based on a rigid joint length Z = 1.0
12.7.6 Overall Drift (OD) Wld P.Hinge Rotation Relationships
As discussed in Section 12.5, the roof displacements and total plastic hinge rotations
seem to have a strong relationship between them. Fig. 12.41 shows the variation between
overall drift and total plastic hinge rotation results from all building prototypes subjected to the
various earthquake excitations. Overall drift, aD, is defined as the ratio between the roof
horizontal displacement and height of the building Hb in percent ( % ). In general, the figure
shows that the total plastic hinge rotation increases significantly as the overall drift increases.
This relationship impl~s that both the roof horizontal displacement and the plastic hinge
rotations are equally important as damage parameters in the inelastic analysis of RC structures.
The relationship seems stronger in the case of frame-walls than for frame structures. This is
because the effect of higher modes on the top floor displacements are more significant in the
more flexible structures. The figure shows that the total plastic hinge rotation in radians for
each direction can be estimated by,
349
(12.4)
where PH~ is the total plastic hinge rotation in each direction, u\ is a coefficient as given in
Table 12.7 and OD is overall drift in percent ( % ).
The coefficients Wt as given in Table 12.7 indicate that with the same earthquake
excitations, frame structures exhibit greater total plastic hinge rotations than frame-walls.
Similarity, a structure with a fIXed base has greater plastic hinge rotations than structures with
flexible foundations. These agree with the results discussed in previous sections.
Table 12.7 Coefficient Wt and Ws
Fixed Base Foundations Flexible Foundations

No. Str. Systems Coeff. lilt Coeff.Var. Coeff. W. Coeff.Var. Remark
1. a. Frame-Walls (Z = 0.5) 0.362 "'" 0.36 0.8880 0.331 ".. 0.33 0.8987
b. Frames (Z = 1.0) 0,483 "" 0.48 0.8206 0.471 "'" 0.47 0.8693
No. Str. Systems Coeff. Iil. Coeff.Vae. Coeff. "'. Coeff.Var.
1. a. Frame-Walls (Z = 0.50) 0.2746 ,.. 0.275 0.9688 0.2478 .. 0.250 0.9460

b. Frames (Z = 1.0) 0.3784 "" 0.380 0.9663 0.3773 "" 0.375 0.9678
12.7.7 Overall Drifts and Structural Damage Index Relationships
As shown in Fig. 12.41, the plastic hinge rotations are strongly correlated with the
overall drift. As discussed in Section 12.6 the member damage indices are linearly correlated
with the member curvatures. The relationship between the structural damage indices and overall
drifts is shown in Fig. 12.42. These structural damage indices result from aU building
prototypes and earthquake excitations. It shows that, in general, the structural damage indices
are also linearly correlated with the overall drifts. Two different slopes appear in the figure.
The square marks correspond to full rigid joint lengths of the 12-storey frames. The effect of
this full rigid joint length on the damage indices of the 12-storey frames is very significant.
Using a linear regression analysis the structural damage index can be estimated by the formula,
(12.5)
where DIs is the structural damage index, w. is a coefficient as given in Table 10.17 and OD
is the overall drift in percent (%).
350
The regression line for the structural damage indices almost exactly pass through the
origin, meaning no structural damage is found at zero overall drift. As given in the table the
coefficient of variation of these relationships are relatively high. The result of a statistical
analysis concluded that the structural damage index is linearly correlated with the overall drift.
The plus and cross marks in the figure correspond to the 12-storey frames with the rigid joint
length Z = 0.0 and 0.5 respectively. The coefficient w. for these structures may be lower than
that given in Table 12.7. These results indicated that effects of rigid j oint length on the
structural response of relatively low frames are significant and need to be comprehensively
investigated.
The base shear of structures on flexible foundations and fixed bases varies with the
earthquake excitations. The base shear resisted by a walls may be several times than that
resisted by a frame. The decrease in the base shear due to the flexibility of the foundations is
more obvious in relatively stiff structures such the frame-walls. The foundation flexibility
redistributes the maximum wall moment from the lower storeys to the middle storeys in almost
all earthquake excitations.
The maximum calculated inter-story drifts may less than the limit value specified in
NZS 4203: 1992. The storey plastic hinge rotation tends to increase as the inter-storey drifts
increase. The beam curvature ductility demands of structure with flexible foundations, in
general, are smaller than those with fixed bases. The maximum value of beam curvature
ductility demands are 18.88 and 17.89 for fixed bases and flexible foundations respectively.
In ductile RC members these values can be easily provided.
The flexibility of the foundations may cause an increase or decrease in the total plastic
hinge rotations of the 12 and I8-storey frame-wall structures as the earthquake excitations are
varied. The total plastic hinge rotations, in general, were linearly correlated with the roof
horizontal displacement. The storey plastic hinge rotation tends to increase as the inter-storey
drifts increases.
The hysteretic energy dissipation of structures with flexible foundations is much less
than those with fixed bases especially for the frame-wall structures. The opposite results are
351
found for the viscous energy dissipation. The total hysteretic energy dissipations are affected
by the magnitude of the roof displacement and type of structural oscillation. Considering the
effect of foundation flexibility on the structural deformation and hysteretic energy dissipation
as with in Chapters IX and X, the use of the energy-based damage model as a single parameter
for describing the structural damage indices becomes less meaningful.
Using a relatively small value of {3 in the PARK & ANG (1984) damage indices for the
strong column-weak beam failure mechanism indicated that the member damage indices were
strongly correlated with the member curvatures. The storey damage showed a similar
correlation with the beam curvature ductility demands. The use of weighting factors based on
the member curvatures for the calculation of storey and structural damage indices was
suggested.
The member, storey and structural damage indices of structures with flexible
foundations, in general, were smaller than those with fixed bases except for a structure under
a particular earthquake as mentioned earlier. The frames suffered greater damage than the
frame-wall structures as the earthquake excitations are varied.
The total plastic hinge rotations and the structural damage indices were strongly
correlated with the overall structural drifts. The relationship for the frame-wall structures were
more strongly affected by the type of foundation than that for the frame structures. Hence, the
total plastic hinge rotations and top floor displacements were equally important in describing
the overall damage to the structures. The structural fundamental period of the inelastic response
of building structures can be approximated in terms of the building height. The approximate
formula for the structural fundamental period of structure on the flexible foundation suggested
by ATC 3-06 (1978) is an overestimate for structure on intermediate soils.
Chapter XIII
The Analysis and Design Recommendations for Rocking
Structures
13.1 INTRODUCTION
The effects of a rocking foundation on the inelastic structural response of frames and
wall-frame structures have been investigated and discussed in the preceding chapters. The
inelastic structural responses include the most common damage indicators such as the member
moments and shear forces, the roof horizontal displacements, the base shear forces, the inter-
storey drifts, the beam curvature ductility demands, the total plastic hinge rotations and damage
indices. The investigations also involved looking at the effects of rigid joint zone lengths, soil-
shear strains, soil shear moduli, the numbers of frames in the frame-wall buildings, the
different earthquake excitations and the hysteresis rules used for the inelastic members.
In structural analyses it is often assumed that the structures are fixed at their bases.
This assumption is not valid for a relatively stiff structure supported by a relatively soft soil.
The errors are compounded by the assumptions concerning the lateral constraint on the
foundation which is discussed in this chapter.
The effects of the interaction between walls and frames on the design of walls of
frame-wall structures still lacks attention. The current design procedures for structural walls
in frame-wall structures, as discussed in Ref.P18, is based on a free-standing wall. This model
is inadequate for frame-wall structures supported by a flexible soil-foundation system. In this
chapter, a new design procedure is proposed.
The foundation footing of the structural wall will rotate significantly during an
earthquake excitation. It is possible that this rotation may lead to lift-off of part of the footing
especially when of wall gravity load is smaiL This lift-off phenomenon decreases the soil-
352
353
foundation stiffness and affects the structural responses. Hysteresis rules for the soil-foundation
stiffnesses, taking into account lift-off, are developed and discussed in this chapter.
13.2 THE BUILDING'S LATERAL FIXITY
The equivalent static lateral force at each floor level, as specified by the building codes,
is a function of its floor height. NZS 4202: 1992 states that the height of the floor level is
measured from the level that provides lateral restraint Of lateral fixity of the building.
Unfortunately the location of this fixity level is not clear. It is common to assume that the
ground level is able to provide the lateral restraint of the building, accordingly, the floor level
height is measured from the ground surface. In the case of multistorey building frames
supported by surface foundation footings this assumption seems logical. However, some
building codes strongly recommend the use of embedded foundations rather than surface
foundations. The effect of variation of the location of the building lateral fixity, therefore,
needs to be investigated.
Four types of structure supported at different levels, namely the frames and frame-wall
buildings fixed at ground level, at footing level, supported by flexible soil-foundations and
flexible foundations with stub columns, denoted by SFix, LFix, LFlex and SCFlex respectively,
are investigated by comparing the building response quantities. These foundations are shown
in Fig. 11.50. The SCFlex is similar with LFlex but possesses additional beams and slabs
located at the ground surface and assumed to be in full contact with the ground surface. The
effect of SCFlex on the response quantities are only investigated for the l2-storey frames and
frame-walls respectively. The soil shear-moduli, earthquake excitations and other soil
properties, as used in the preceding chapter, are used here as wel1.
13.2.1 Frame Structures
The response quantities compared are the total base shears, the maximum beam
curvature ductility demands, the total plastic hinge rotations, the hysteretic energy dissipation
and the member, storey and structural damage indices. These response quantities are presented
in Table 13.1 and 13.2 respectively.
The base shear of the 12 and l8-storey structures subjected to different

eal1hquake excitations are shown in Fig. 13.1 and 13.2 respectively. Table 13.1 shows that the
354
base shears of the LFix and LFlex structures only differ by an average of 4.12 % and 2.24 %
for the 12 and IS-storey frames respectively. The base shear of the SCFlex is only 2.85 %
different from the SFix structures. Assuming that the SFix condition for the frames with
embedded foundations leads to an overestimation of the base shear by an average of 16.46 %
and 8.41 % for the 12 and 18-storey frames respectively. A shortening of the first storey height
causes a larger increase in the of structural base shear.
The beam curvature ductility demands of SFix, LFix, LFlex and SCFlex of
the 12 and IS-storey frames are shown in Fig. 13.5 and 13.6 respectively. These beam
curvature ductility demands are not very sensitive to the foundation types. In Table 13.1, the
beam curvature ductility demands of LFix are relatively closer to those of LFlex than the
SFix condition. The beam curvature ductility demands for the SCFlex only differs by an
average of 1.8 % from the SFix frames. Again, taking frames with embedded foundations
as SFix structures, gives an overestimation in the results.
In some earthquakes, the effects of foundation type on the total plastic hinge
rotations gives inconclusive results, see Table 13.1, Figs. 13.9 and 13.10 respectively. The
total plastic hinge rotation of the SCFlex and SFix frames only differs by an average of 2.22
%. Similar inconclusive results are found for the hysteretic energy dissipation for the 12 and
18-storey frames, as shown in Figs. 13.13 and 13.14 respectively. In the 12-storey frame
the hysteretic energy dissipation for the SCFlex case only differs by an average of 3.64 %.
In general, the LFix frames give results closer to the LFlex than SFix structures and the
SCFlex frames provide response quantities closer to the SFix frames.
The maximum member, storey and structural damage indices of the 12 and
18-storey frames are given in Table 13.2 or shown in Figs. 13.17 and 13.18 respectively.
The damage indices of the SFix and LFix structures are relatively close. In general, with
respect to LFlex structures, the SFix structures overestimate the damage indices when
compared with the LFix structures. Again, the SCFlex frame gives damage indices closer
to the SFix frame structures.
It is clear from these results that in all structural response quantities the LFix
structures provide results closer to the LFlex than those shown by the SFix structures. The
SCFlex gives results closer to the SFix frame structures. The differences in structural
Table 13.1 Damage Indicators and Damage Indices of the Frame and Frame-WaH Structures
Earthquake ExcilaliOnfi % Different
No. Structural Resp. Respect to
!llld Str. System 40EcNS Tok.Oki 40 Buc.h.NS LFI..e:!<.
l. Ba.B<) Shear (TOWle)

n.12 SI.Prnme
1. SFix. 127.17 121.74 12&.26 153.73 141.30 16.46
2. LPix. 10S.00 111.40 118.50 137.70 127.60 4.12
3. LFlex. 98.50 109.80 115:20 134.30 12&.40
4. SCFtex. 126.70 118.47 127.17 144.60 135.86 2.3S
h.IS St.Pmme
1. SPix. 1411.91 146.64 146.74 192.54 172.46 11.41
2. LPix. 142.39 135.44 13S.04 1110.59 165.12 2.24
3. LFlex.
c.12 St.P-W"lIs
139.61 132.10 137.10 174.60 161.20 -
1. SPix. 2384.33 1990.53 1963.76 3026.12 2051.44 32.21
2. LPix. 2114.00 1937.33 15116.27 2731.30 1636.20 15.32
3. LFlex.
4. SCFtex.
1823.00
1891.05
1514.50
1502.90
1392.70
1489.27
2440.30
2664.18
1521.71
1641.30
-8.30"
2. Curv .DucI.DellWl.ds
a.12St.Prome
1. SFiK. 11.16 11.06 13.05 22.10 23.62 11.S4
2. LPix. 10.38 10.47 12.81 21.15 22.31 6.44
3. LFlex.
4. SCFtex.
9.69
10.91
10.35
10.31
12.1I
13.31
19.35
22.13
20.33
22.86
-UUf
h.18 St.Fcnme
1. SPix. 7.64 7.33 9.73 15.51 15.60 10.44
2. LPix. 7.44 6.S0 9.67 15.07 14.65 6.18
3. LFlex.
c.12 SI. P-Wruln
6,88 6.68 8.52 14.09 14.56 -
1. SPix. 9.55 9.76 11.81 19.35 15.05 28.63
2. LPix. 8.04 9.55 9.10 18.88 14.94 18.91
3. LFlex. 6.49 7.48 10.35 17.89 15.21 -
4. SC'FIex. 7.92 7.89 9.58 18.96 14.82 5.04'
3. n.p1.R.Rol.(O.OO 1 Red)
11.12 SI.Frame
1. SPix. 41o.s2 442.73 394.81 771.86 870.53 10.22
2. LPiJ(. 401.93 446.05 399.49 748.59 800.93 7.58
3. LFlex. 34l.82 42.2.71 384.31 718.58 748.93 -
4. SCFtex. 393.00 423.92 389.96 751.15 1lO6.33 2.z:t
h.IS St.Prnme
1. SPix. 429.37 439.29 473.94 856.49 944.32 9.72
2. LPix. 417,94 409.51 <Ul9.22 835.13 952.16 7.72
3. LFlex.
c.12 SI.P-Walls
392.67 357.16 460.2S 784.97 906.33 -
I. SPill:. 312.40 390.75 282,95 821.57 799,96 9.30
2. LPix, 345,35 377.81 309.63 825.60 767.23 6.27
3. LFlex. 32S.71 336.66 365.72 798,93 733.54 -
4. SCFtex. 340.27 354.30 321.65 800.66 740.03 3.6d'
4. Ry81. Energy (Tin)
11.12 St.Prome
1. SPix. 28.89 51.20 25.17 46.51 54.44 16.70
2. LJTIX. 26.69 4<5.27 24.64 47.4<5 49.56 10.26
3. LFlex.
4, SCFtex.
23.71
25.40
40.68
43.79
22.58
23.71
43.31
45.26
46.52
49.27
-3.64'
h.lS St.Frrune
1. SPill:. 32.87 42.58 33.78 67.77 69.77 17.86
2. LPix. 31.20 35.95 34.94 64.92 67.96 11.08
3. LFlex. 27.80 27.57 33.53 61.76 65.63 -
c.12 St.F-Walla
1. SPix. 244.44 422.22 211.11 525.93 551.U5 34.33
2. LFix. 307.46 402.96 195.90 518.65 537.31 29.30
3. LFlex. 172.00 200.00 154.50 426.69 385.01 -
4. SCFtex. 227.58 252.94 166.67 459.85 423.23 20.90'
.-
nemar\c; : with respect to SPill: I.\.lll LPix for the framea wul frll.llllrwalls reII]JCCtiveiy.
356
180 210
SFh(.
lAx. Square : SFIJI.
lFlex. Clmle : lFllI.
SCFleK. Trlllng. : lFlex.
160 190
.~
140 170
E E
~
OJ liJ
"''""
U'J '"
.0::
IJ)
~~
.,"
'"''""
I:Q
120 &l 160
100
.-7 If 130
Tok.Olcl
GO 110 +----4--+---1----+---;---1
40EcNS 40EcEW Ouch.NS WEcNS 40EcEW Huch.IIIS
fig. 13.1 Base Shears ve. Earthquake Inputs Fig. Base Shears vs. Earthquake Inputs
3150 1,==========:;------, 5000 - , - - - - - - - - - - - - - - - - - ,

Square SAx.
Clrele LFlx.
Triang. lFlax.
Olamd. SCFlex. 4500
3250
4000
2150
E E 3500
::;
'" 2250
;j
III
..'"
1150
2600
1250
L
:WOO
l.Iuch.NS Parkf.1 Tnft.1

_T_O_k-t.O_k_I_--t-_p_<l_rk-tf_.N_E_--t-_--I
150 1500 -;---j---r--+---4-r-I---t-----l
WEeNS 40EcEW Such.IIIS 40EcNS 40EcEW Porkf.2
Fig. 13.3 Base Shears vs. Earthquake Inputs Fig. 13.4 Base Shears vs. Earthquake Inputs
357
30 20
Square SFix. F12 Square SFill. IFHlI

,_ _I
Circle LFlx. Circle LFill.
Triang. LFiex. 18 Triang. lFlell.
Dlamd. SCFlex.
25
16
II)
.,
"0
'".,
0::
20
0::
'"
E
14
E III
0'" C
~ ~ 12
$.., ~
:l
0'" 0
~ '15
c: 10
'"
(.l
:J
(.l
6
10
Tok.Oki Parkf.NE Tok.Okl PBrkf.NE

5 t -I 4
40EcNS 40EcEW Buch.NS 40EeNS 40EcEW Bueh.NS
Fig.13.5 Max. Curv. Duct. Demands vs. Earthq. Inputs fig.13.6 Max. Curv. Duct. Demands vs. Earthq. Inputs
24,----------------, 20
Square SFix. Square SFlx. [F-W}!ffiZj

Circle LFlx. Circle LFix.
Trlang. LFI"x. Triang. LFlell.
Oiamd. SCFI"lI.
20
16
'"
'C
E:
I/)
't>
E:
'"Q)
E
16 III
E
0 "
0
~ ~ 12
~:J ~:J
0 0
c:
:J
12
~
:l
(.l U
B
B
Tok.Oki Parkf.NE Buch.NS Parkf.1 Taft.1
40EeNS 40EeEW Bueh.NS IWEeNS 40EcEW Parkf.2
Fig.13.7 Max. Curv. Duct. Demands vs. Earthq. Inputs Fig.13.3 Max. Curv. Duct. Demands vs. Earthq. Inputs
358
response between LFix and LFlex and between SCFlex and SFix in the frame structures are
relatively small. Accordingly, the. assumption of the structural lateral fixity at the level of
ground surface and foundation footing for stub column frame structures and frame structures
supported by embedded foundations respectively, are still acceptable.
13.2.2 Frrune-Wall Structures
The base shear of the 12 and 18-storey frame-walls are shown in Figs. 13.3 and
11.4 respectively . Unlike the frame structures, the effect of the foundation types referred to as
SFix, LFix, LFlex and SCFlex on the structural responses are relatively significant. With
respect to the LFlex structures, the LFix structures show an average increase in the base shear
by 15.2 % and 21.6 % for the 12 and IS-storey frame-wall structures respectively. The
assumption of the SFix foundation causes an average increase in the base shear by 32.21 % and
40.96 % for the 12 and IS-storey frame-walls respectively. It shows that relatively stiff
structures with a small first storey height show a marked increase in the base shear. With
respect to LFix, the base shear of 12-storey frame wall SCFlex structures shows an average
different of 8.3 %.
The beam curvature ductility demands of the 12 and I8-storey frame-walls are
shown in Figs. 13.7 and 13.8 respectively. The effects of the foundation type is not significant
except for the 18-storey frame-walls subjected to Parkf.1 and Parkf. 2 ground accelerations. The
maximum curvature ductility demands are 19.35 and 14.17 for the 12 and IS-storey structures
respectively. These maximum beam curvature ductility demands are still within the ductility
capacity of the bearns. As with the previous results and there shown in Table 13.1 the SCFlex
of the frames and frame-wall buildings provide beam curvature ductility demands closer to
those of the SFix frames and LFix frame-walls respectively.
The total plastic hinge rotation of the 12 and I8-storey frame-walls are shown
in Figs. 13.11 and 13.12 respectively. With respect to the LFlex structures, the LFix structutes
show an average increase in the total plastic hinge rotation of 6.27 % and 14.65 %, whereas
a 9.3 % and 21.97 % differences respectively are found for the SFix structures. The total
plastic hinge rotation of the SCFlex give an average difference of 3.6 % when compared with
the LFix frame-wall structures. It shows that, in general, the SFix structures suffer greater
damage than the LFix or LFlex structures.
359
The hysteretic energy dissipation ofthe 12 and IS-storey frame-walls are shown
in Figs. 13.15 and 13.16 respectively. The hysteretic energy dissipation for the LFlex and
SCFlex structures is much less than the SFix or LFix structures for all earthquake excitations.
The difference in the hysteretic energy dissipation between these types of foundations as shown
in Figs. 13.15 and 13.16 is almost independent of the earthquake excitations. The hysteretic
energy demand of the 1S-storey frame-walls with fixed bases is more than double that for those
with flexible soil-foundations, while the maximum period shift due to foundation flexibility is
only 20 % as discussed in Section 12.7.4.
Table 13.2 Damage Indicators of Lhe Frames and Frame-Wall Structures
Earthquake Excitations % Increase

Structural Resp. Respect to
No. and Str. System 40EcNS Buch.NS 40EcEW Park.f.1 Park.f.2 LFLex.
1. 18-St. Frame-Walls
a.Base Shear
1. SFix. 3399.25 2605.06 2380.43 3492.53 4130.55 40.96
2. LFix. 3257.45 2368.23 2070.88 2865.70 3302.17 21.60
3. LFlex. 2624.30 2299.20 1960.90 2130.40 2344.10 -
b.Curv.Duct.Dmd.
L SFix. 6.02 13.36 9.37 12.68 14.17 27.17
2. LFix. 6.50 13.15 9.21 11.69 13.13 20.80
3. LFlex. 6.79 12.12 8.46 8.23 10.59 -
c.P.H. Rotation
1. SFix. 301.83 637.10 413.69 512.86 694.26 21.97
2. LFix. 275.48 584.23 422.81 499.84 642.59 14.65
3. LFlex. 339.88 548.85 436.51 407.33 559.44 -
d.Hyst.En.Dissp.
1. SFix. 238.80 462.96 255.23 361.48 551.85 130.74
2. LFix. 226.67 421.65 266.42 333.61 500.00 116.94
3. LFlex. 142.53 180.90 134.30 123.10 -223.80 -
e.Struct.D.Ind.
1. SFix. 0.1137 0.2455 0.1655 0.2221 0.2631 15.60
2. LFix. 0.1167 0.2478 0.1709 0.2121 0.2415 13.56
3. LFlex. 0.1377 0.2258 0.1560 0.1490 0.1991 -
Struct. System 40EcNS Tok.Oki 40EcEW Parkf.NE Buch.NSC
2. Struct. D.Index
a.12 St.Frame
1. SFix. 0.2673 0.2940 0.2845 0.4658 0.5111 12.27
2. LFix. 0.2585 0.2933 0.2892 0.4572 0.4923 6.86
3. LFlex. 0.2443 0.2789 0.2841 0.4441 0.4751 -
b.18 St.Frame
1. SFix. 0.1760 0.1832 0.2067 0.3327 0.3454 5.49
2. LFix. 0.1698 0.1713 0.2076 0.3213 0.3368 3.84
3. LFlex. 0.1538 0.1476 0.1930 0.3034 0.3185 -
c.12 St.F-Walls
1. SFix. 0.1857 0.2151 0.1692 0.4188 0.3752 9.76
2. LFix. 0.1948 0.2169 0.1786 0.4219 0.3632 9.26
3. LFlex. 0.1700 0.1817 0.2119 0.3914 0.3608 -
Remark: -
360
1000 1000
Squate SFh"
Circle LF1K.
Tdnng, LFlex,
Diamd. SCFI",. Jill
..
'ti
II:
;;
0
000
f 'ti
111
a:
....
0
0
lIOO
(;;
!(
i
a:
(J)
600
f d
!(
..:c
a::
600
m '"
Dl
5:'"
Ji
:I:
0,; 0,;
400 400 IF1S1
Tok.Old Parkt.NE Tok.Okl Parkf.NE

2.00 +---+---1---+---1---+---1
40EcNS 40EcEW Buc".NS 40EcNS 4{)EcEW Buch.NS
Fig. 13.9 P .Hinoe Rotations \IS. Earthq. Inputs Fig. 13.10 P.Hinge Rotations \IS. Earthq. Inputs
1000 900
800
!Square : SAx.
ICircle : LAx.
Triang.: LFlex.
Diamd. : SeFlax.
I~
700 -+-
,
Square
Circle
Triang.
SFllI.
LFix.
LFlex.
FW18RZ
.
'ti 'ti
'"
a::
-
a:
0
0
ci
600
(;
0
ci
~ .!! 500
i
a: Z
., a:.
.,
t:l 400
C 0>
:r: C
:f
0,; 0,;
300
200
Tok.Okl Parkf.NE Bueh.NS ParkU Taft. 1

o 100 +---\.-...--t---t-----+--t---t---
40EcNS 40EcEW Buch.NS 40EcNS 40EcEW Parkf.2
Fig. 13.11 P .Hinge Rotations liS. Earthq. Inputs Fig. 13.12 P .Hinge Rotations liS. Earthq. Inputs
361
70 ,--------------------------------, 90 - - - - -.... _ _.. ..... _ - - - - - ,
Square SFix. Square SFi,..

Circle LFix, Circle lFlx.
Trlong. LFlex. Tri8ng. LFlex.
60 Diamd.: SCFlex.
70
E 50 E
t: t:.
I:
~" '..,",
"
.!i!-
V>
.,
,Q.
,!2
V>
0 50
i'5 40
...~ >-
E'
'"
<:
UJ
'<:"
UJ
..:
I/)
.,...
.;
>- 30 :r
J:
30
20
Tok.Oki Parkf.NE Tok.Oki Parkf.NE

10 10
40EcNS 40EcEW Buch.NS 40EcNS 40EcEW Buch.NS
fig.13.13 Hyst. Energy Dissipations vs. Earthq. Inputs fig. 13.14 Hyst. Energy Dissipations vs. Earthq. Inputs
600 600
,-""...
Square SFix. !Square SFiK.
Circle lFlx. Circle lFlx.
Triang. lFlex. Triang. LFlex.
Diamd. SCFlex.
E 400 E 400
!:: !::
c c
0
.~ ;
.'" .8-
Q.
'u; II>
0 is
... ...E'
;;'"
c:
UJ
..:V>
>- 200
..""
UJ
VI
:>- 200
:I: :::r:::
Tok.Oki Parkf.NE Buch.NS Parkf.l Taft. 1

o
Fig.13.15 Hyst. Energy Dissipations vs. Earthq. Inputs Fig.13.16 Hyst. Energy Dissipations vs. Earthq. Inputs
362
0.6
Square : Mh. D. Ind. Square: Mb.D. Ind.

Circle : St. D. Ind. Circle : St. D. Ind.
Trslng. : Str. D. Ind. Toang. : Str. D. Ind.
0.5
Thick line : SFill. 0.3 Thick linG : SFlx.
Filled lFiK. Filled lFb{.
Blank : lFlex Blank : lFlex.
0.4
0.6
..,"
OJ
"'"
"G
.E .E
<l> 0.3
'"'" '"
til
E '"
E
0'"
({]
t.::I
0.4
0.2
[~ 0.2
11'121
0.1
Tok.Oki Parkf.NE Tek.Oki Parkf.NE

0 0
40EcNS 40EcEW Buch.NS 40EcNS 40EcEW Buch.NS
Fig. 13.17 Daml'lge Indices vs. Earthq. Inputs Fig. 13.18 Damage Indices vs. Earthq. Inputs
Square: Mb.D.lnd.
0.8 0.6,--------ICircle : St. D.lnd.II!I-_ _ _ _ _---,
Square Mb.D.lnd. Triang. : Str.D.lnd.
Circle St. D.lnd.
Thick line : SFix.
Traing. Str.D.lnd.
Filled : lFlx.
Thick line SFlx. 0.6 Blank : lFlell.
Filled lFix.
Brank lFlex.
0.6
0.4
><
""'"
.E
~ 0.4 0.3
to
E
III
Cl
0.2
0.2
0.1
IFW18Rzi
Tek.Ok! Pnrl<f.NE Buch.NS Park!. 1 Taft.1

o +---1-----+ --+----II---___i__ o +--_+--r--~--__I----+_-_+--~
Fig. 13.19 Damage Indices vs. Earthq. Inputs Fig. 13.20 Damage Indices vs. Earthq. Inputs
363
This is evidence that the soil-foundation system may dissipate a considerable amount of energy;
an phenomenon that does not exist for the fixed base structures.
The maximum member, storey and structural damage indices of the SFix, LFix and
LFlex foundations for the 12 and is-storey frame-walls are shown in Figs. 13.19 and 13.20
respectively. In general, the structures with fixed base foundations, either SFix or LFix, suffer
more damage than the flexible soil-foundation structures. The figures show no significant
difference of the damage indices between the SFix and LFix structures when compared with
the LFlex structures.
The effect of foundation type on the structural response quantities of frame-wall

structures are more significant than for the more flexible frame buildings. There is a little
difference between LFix and LFlex structures for most response quantities such as beam
curvature ductility demands, total plastic hinge rotations, hysteretic energy dissipation and
damage indices. However, a significant difference is found between the SFix and the LFlex
foundation. In most response quantities, the SCFlex gives closer results to the LFix than the
others. It is difficult to approximate the location of the lateral fixity for frame-walls when
compared with frame structures. In the case of the frame-wall structures supported by
embedded flexible foundation footings the LFix conditions provide a slightly better prediction
of the structural response quantities than does the SFix conditions. For frame-waH structures
with stub columns (SCFlex) the assumption of the structural lateral fixity at the level of the
foundation footing (LFix) gives better prediction on the response quantities than if the fixity
is assumed to be at the ground surface.
13.3 THE WALL DESIGN MOMENT ENVELOPES OF FRAME-WALL

STRUCTURES
In Ref.P18 it was stated that the design moment envelopes for the walls of frame-wall
structures were similar with the design moment envelopes for a free standing cantilever wall.
The only difference was in providing a slightly higher flexural resistance in the upper stories
to overcome the possibility of reversed moment predicted by an elastic analysisP1B
As shown in Figs. 12.3 and 12.11, the distribution of the wall bending moments up the
height of frame-wall structures with flexible foundations are significantly different from those
from an equivalent static analysis of a free standing cantilever wall. The wall moments
364
-2.5 -2 -1.5 -1 -0.5 0 0.5 1.5 2. 2.5 -2 -1.5 -1 0.5 o 0.5 1.5
Normalized of Wall Moments! Normlillzed of Wall Moments
Fig. 13.21 Normalized of Wall Moments Fig. 13.22 NOfmalized of Wall Moments
-<-&tmct.
Walls
I> '" Pos.

o - Neg.
3.95 Gee. 4.00 Bee. 4.80 8Ge.

Fig. 13.23 Wall Plastic Hinges of the 12-storey Frame-Wall"
62
J". .J.,
4.50 seG. 4.95 sec. 6.00 sec.

Fig. 13.24 Wall Plastic Hinges of the HI-storey Frame.Walis
365
,-----------ae,ee_,---------------~ ,---~~ee-,---------------------~
Moment (Tm) Moment (Tm)
6000 15000
4000
Curv.(Rad/m)
-0. 01 0.0005 O. 01 -0. 005 O. 01
-6000 FW12RZ -15000 FW18RZ

Mmb. No 43 Mmb. No 2
End 2, Parkf.NE End 2, Parkf.NE
~--------~~e'e-~--------------~ ~--~~ee_~--------------------~
Fig. 13.25 Moment III Curv. Relationship Fig. 13.26 Moment & Curv. Relationship
kMi
/1':
/ ...
/
113 Hb .'
/// /'/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
Le /
/
Hb
~m-+- ____-,~i____ Ref. B27
/
!~---Ref. P18
i
,/
Lec = (0.21 f) Le Flex.Found.
I "'I "" l . Co"."''''

-+ +-----: ---~
I~ t-o.~~:
Lp
M; - - -......-
f = frequancy (Hz) ---+---+-t

M_i
k depends on min. reinforcement 1.2 Mi +
Fig. 13.27 Design Proposed of Wall Moment Envelopes of the Frame-Wall Structures
366
normalized with respect to wall base maximum moments computed from an equivalent static
analysis, for the 12 and IS-storey frame-walls are shown in Figs. 13.21 and 13.22 respectively.
It shows that the normalized moments are much higher than the design envelopes especially at
the middle height storeys.
A number of analyses of the inelastic response of the walls in the 12 and IS-storey
frame-walls have been carried out. Walls with barbel cross sectionll8 were designed using the
approximate formulae as discussed in Section 3.5.3 with the wall height to length ratios are
9.03 and 8.94 for the 12 and IS-storey frame-walls respectively. According to Ref.P18 the
slenderness of these walls are considered as moderate. The effective wall moment of inertia lwe
as suggested in Ref. P 18 was used. The design assumed that the walls carried the full wall
gravity load. The results from the Parkfield N-E component earthquake indicated that some
walls yielded in the middle storeys as shown in Figs. 13.23 and 13.24 respectively. The
examples of wall hysteretic flexural responses of the 12 and 18-storey frame-walls are shown
in Fig. 13.25 and 13.26 respectively and show significant plastic yielding occurring in the wall.
Similar results were discussed in Ref.B27.
To protect these structural walls from yielding, a proposed wall moment envelope for
frame-wall structures with fixed bases and flexible soil-foundations were developed and are
shown in Fig. 13.27. The dotted and dashed lines are the design envelopes presented by Refs.
PI8 and B27 respectively. It shows that the proposed design envelope for frame-wall structures
supported by flexible foundations is smal1er in the lower stories and higher in the upper stories
when compared with the design envelopes in Ref. B27.
A number of analyses of for the 12 and 18-storey frame-walls designed to the proposed
wall moment envelopes subjected to the Parkfield N-E and the Bucharest N-S component
earthquakes have been conducted. Results indicate that no wall plastic hinges occur on the
middle height storeys of either buildings. The examples of the moment-curvature relationships
at the wall base of the 12 and IS-storey frame-walls with fixed and flexible foundations
subjected to Parkfield N-E earthquake are shown in Figs. 13.28, 13.29 13.30 and 13.31
respectively. Fig. 13.31 shows that the base walls remain elastic. The maximum wall
curvatures are still much less than 0.03 rad/m i.e the available maximum wall curvature from
the laboratory tests reported in Ref.G9.
367
r--------------4eaea_.-----------~ ,-----------------~8~OOe_,_--------_,
Moment (Tm' Momant ITm)

BODO
6000
Curl/.lrad/m,
-0. 06 -0. 04 -0.003 -0.002 0.001 O. 02
FWF12 FW12RZ
Takeda Takeda
Parkf.NE Parkf.NE
L-------------~1~OOOa_L-----------~
Fig. 13.28 Moment vs Curv. Relationship of Wall Base Fig. 13.29 Moment vs Curv. Relationship of Wall Base
Moment (Tm'
Moment (Tm,
20000
10000
5000
Curl/.lrad/ml
-0. 04 -0.0002 0.0002 0.0 04
-20000 !FW18RZ
Takeda Takeda
Parkf.NE Parkf.NE
Fig. 13.30 Moment vs Curv. Relationship of Walt Base Fig. 13.31 Moment vs Curv. Relationship of Wall Base
368
0.009
FW12RZ
Takeda
0.006
Parkf.NE
0.003
0
.
;::
.2
...
III
0
II:
OJ
"g
-0.003
0 Filled Full Grav. Load
z Blank Half Grav. Load
'"
III
'"
III
-0.006
-0.009
-0.012
----Half
Hi -10 -5 o 5 10 15
0.015
Wall Moment III 1000 Tml
Fig. 13.32 Wall Footing Rotations flg.13.33 Max. Wall Momonts 12Wallsl
Momenl
43 6000
Struct.
Walls
I
., .. .... .... .J..
3.95 sec. " : Pos.

o Neg.
!-.o!Iiil!--- Struct.
Walls -SOOO
fFW12HF
Mb. No.43
lEnd 1
4.75 sec.
Fig. 13.34 Wall Plastic Hinges Fig. 13.35 Wall Moment lit Cur\!, Relationship
369
The effect of the wall gravity load on the inelastic response was also investigated. The
structural wall was designed using the present design envelopes and with one-half of the wall
gravity load. The effect of the wall gravity load on the wall footing rotation of the 12-storey
frame-walls subjected to Parkfield N-E component is shown in Fig. 13.32. Full and Half in the
figure correspond to the structural walls with full and one-half of the applied gravity load
respectively. It shows that using one-half of wall gravity load, the wall footing rotation is
almost double that associated with the full gravity load. An increase in the waH footing rotation
causes an increase in the wall moment especially in the middle height storeys as shown in Fig.
13.33. This is similar to the effect of flexible soil-foundation on the wall moment discussed in
the preceding Chapter. It shows in Fig. 11.34 that some plastic hinging occurs in the middle
storeys when the structural wall was designed using the present design moment envelopes. Fig.
13 .35 shows a moment curvature relationship where a significant plastic deformation has
developed. No plastic hinge was found on the upper stories when the structural wall was
designed to the proposed moment design envelopes.
13.4 THE HYSTERESIS RULES FOR SOIL-FOUNDATION IMPEDANCES
Fig. 13.32 shows that the effect of the gravity load on the wall base node rotation is
very significant. The higher wall footing rotation results in a significant lift-off of the footing
foundation of the 12 and 18-storey frame-walls as shown in Figs. 13.36 and 13.37 respectively.
These figures correspond to the vertical displacement of wall footing tip where the walls carry
one-half of the gravity load. This loading condition may occur when the structural walls are
located as the outer plane-frames in the structures. It shows that the effects of the gravity load
on the lift-off of the wall footings are very significant. The high frequency impact effects do
not appear such as in the case of rigid bases or in the results of laboratory tests of the two
spring model discussed in Ref.C7 and H8.
An increase of the footing rotation may cause tip lift-off in one side and deeply
penetrate the soil on the other side. This will cause an increase in the soil stress and decrease
the length of contact between the soil and footing. The soil-foundation impedance may then
decrease because the soil foundation impedances depend on the contact geometry between the
soil and footing. This effect can be referred to as a geometric non-linearitj8. A decrease in the
soil-foundation impedance will affect the structural responses.
370
o.os
0.04
'\
1\ ... Pa,kf.NE
!\
0.02
!
:g
....c:
III
E
"'""
"5.
0
.!!!
0
t::OJ
>
-0.02
-0.04
G420, FW12RZ
Mod. Takeda
0.06
fig. 13.36 Vertical Dispacements of Wall Footing Tips
0.02
Buch.NSC
f\ Pa'kf.NE
0.01
0
:g
...c
"
E
'"" 0.01
"
Q.
.!!!
c
e..
>
0.02
-0.03
G540.fW18RZ
Mod. Takeda
-0.04
fig. 13.31 Vertical Dispacements of Wall Footing Tips

371
The geometry of lift-off of the footing is shown in Fig. 13.38. The 1;0 and p are the
angles of rotation when the footing starts to lift-off and the soil starts to yield respectively.
Assuming that the static contact pressure and deformation of soil follows an elasto-plastic
relationship, according to Ref.E24 the footing rotation p can be computed by,
(13.1)
where,
(13.2)
where SF is the static soil bearing capacity safety factor, Wo is the initial vertical displacement
of the foundation and B is the total length of the footing.
Eq. 13.1 shows that when SF is equal 2 the soil yield stress and the footing lift-off will
commence simultaneously, whereas when SF is greater than 2 soil yield will occur after the
outer tip of the footing lifts-off. From the static equilibrium between the applied moment and
vertical load, the footing length contact when the soil commences to yield is,
11' = [~:] (13.3)
The corresponding vertical displacement at the centre of the footing is,
(13.4)
Taking into account the effect of the foundation lift-off on the structural responses, the
analytical solution of the lift-off equations of motion as carried out in Ref.C3, C4, C7, S9 and
510 or as discussed in Chapter II can be used. In terms of an engineering application, this
approach is not simple because the equations are complicated and highly nonlinear. Ref. J8
used an embedment coefficient and effective width ratio of the footing for taking into account
the effect of geometric non-linearity. The approximate soil foundation impedance formulae need
to be built-in in the structural computer program. In general, this approach may not suitable
in most inelastic structural analysis programs such as RUAUMOKO or DRAIN-2D. A more
practical approach, therefore, is needed. For this purpose a hysteresis rule for the soil-
372
Ground Surface c
wo M Inltialllertlcal displacement at the

centre of the footing foundation
wtp = tip vertical displacement in which

o Mi
the soil start to vield
B = Foundation Lenllth
B' Contact Length
Fig. 13.38 Lift-off Geometry of Foundation footing Fig. 13.39 Moment Vs. Found. Rotation Hvst. Rulas
IFW12420 I
2 Ivart. Mode:
.., 2.
c 1.6
OJ
E
0
:i
~0
Z
Norm. Rotation Norm. Rotation
-2 2. 3 .2, -1 2. 3
-1
-2
---- ------.......--.-. Theory
----Model
Fig. 13.40 Norm. of Soli-Found. Rocking Hyst. Rules Fig. 13.41 Norm. of Soil-Found, Vert. Mode Hyst. Rulas
373
foundation impedance was developed using a simple computer program. The proposed design
soil-foundation impedance hysteresis rule is shown in Fig. 13.39.
When the soil and the footing remains in full contact i.e before footing lift-off, the soil-
foundation impedances remain constant as indicated by line O-A-B. M" and <p" correspond to
lift-off moment and footing rotation respectively. During lifting-off of an embedded footing,
it is difficult to predict the effect of soil-sidewall contacts on the soil-foundation impedances.
These contact area may less on the lift-off side but increase on the other side. For practical
purpose these contact area are assumed to be the same as when the footing remains in full
contact with the soil. When the footing rotation is greater than <p" the footing starts to lift-off
and the soil-foundation impedances decrease as indicated by lines A-C or B-D respectively. The
relationship between the applied moment and rotation in these regions can be expressed as,
(13.5)
M = M {A ..l _ (A -1) }
t " t II
11.
and,
~I [!:l (13.6)
where Mi is the moment corresponding to footing rotation <Pi> M" = K. 4>" is the moment at
initial lift-off and 1(.. is the rotational soil-foundation stiffness.
Examples of the normalized moment and normalized rotation relationships for wall
foundations of the 12-storey frame-walls are shown in Fig. 13.40. These relationships are
normalized with respect to initial moment M" and initial lift-off rotation 4>". The dotted line in
the figure represents the theoretical normalized value from soil-foundation impedances and the
solid-line indicates the proposed model. The normalized soil-foundation rocking hysteresis rules
for the proposed model are shown in Fig. 13.40 and are based on Eq. 13.5 with A, p and q
equal to 4, 0.04 and 0.27 respectively. The figure shows that the proposed model fits the
theoretical hysteresis rules welL
There is no unique relationships between the applied axial load and the decrease of soil-
foundation vertical stiffness. A similar relationship is also found in a horizontal vibration mode.
This is because a decrease in the vertical soil-foundation stiffness is caused not by the variation
of the veltical load but due to a decrease in the footing contact length when the footing lifts-off.
374
However, to show the hysteresis rules for the soil-foundation vertical stiffness, the fictitious
normalized axial load is plotted with the rotation of the footing foundation as shown in Fig.
13.41. The hysteresis rule of the model is based on Eq. 13.5 with A, p and q equal to 4,0.222
and 0.05 respectively. The normalized hysteresis rules for the rocking and vertical modes for
a wall foundation of an 18-storey frame-wall is shown in figs. 13.42 and 13.43 respectively.
These soil-foundation hysteresis models are based on A, p, and q equal to 4, 0.04 and 0.255
for the rocking mode and 4,0.216, and 0.06 for the vertical vibration modes respectively. The
theoretical hysteresis rules, in general, are well fitted by the proposed models. They show that
the decrease in soil-foundation impedances due to the footing lift-off are more significant in the
rocking mode and relatively less significant in the vertical and horizontal vibration modes.
Several analyses using the proposed hysteresis model for the frame-wall structures have
been conducted. The results are compared with the theoretical hysteresis rules on the 12-storey
frame-walls subjected to the Parkfield N-E and the Bucharest N-S component earthquakes are
shown in Figs. 13.44 and 13.45 respectively. The solid lines corresponds to the output
hysteresis value computed from the RUAUMOKO program using the proposed model and the
dotted lines represent the theoretical hysteresis rules. The proposed hysteresis models perform
velY well.
The behaviour of this model for the i8-storey frame-walls excited by Parkfield N-E and
Bucharest N-S component earthquake are shown in Figs. 13.46 and 13.47 respectively. The
dotted-line represents the moment-rotation relationship from theory and the solid line represents
the relationship computed from the RUAUMOKO computer program based on the proposed
hysteresis model. It shows that the proposed hysteresis model for the moment-rotation of the
soil-foundation fits the theory moment-rotation relationship quite well.
Effects of the wall gravity load on the beam curvature ductility demand and maximum
member damage index are shown in Figs. 13.48 and 13.49 respectively. As with the previous
results, the distribution of member damage indices over the floor levels is similar to the
distribution of the member curvature ductility demand. A decrease in the wall gravity load
causes an increase in the beam curvature ductility demand and member damage indices.
375
-- ..
..
I: 1.5
~
<>
:ill
0
e
z
Norm. Rotation
.f!, 3 -2 -1 1 Z :3 <I- 4
.,
-1.6
-2
Fig. 13.42 Norm. of Soil-Found. Rocking Hyst. Rules Fig. 13.43 Norm. of Soil-Found. Vert. Mode Hyst. Rules
10
!FW12420 FW12420
Rock. Mode iRock. Mode
Parkf.NE 7!iOO IEluch.NSC 7500
5000
0. 16 -0.01 0.01 O. 15 -0. 15 0.01 O. 15
7500
Fig. 13.44 Theory vs Computed Rocking Hyst. Rules Fig. 13.45 Theory vs Computed Flocking Hyst. i~ules
376
G540. FW18RZ G540.FW1BRZ

Rocking Mode Moment (Tm) Rocking Mode i Moment ITm)
Parkfield NEe Bucharest NSC
10000 10000 -
-0. 15 -0.01 0.01 O. 15 -0. 15 0.01 0.01 O. 15
-10000 - 1 - , - - - - - - - -
10000
- - - - Compo
..... -.-................ Theory
Fig. 13.46 Theory VIS. Computod Racking Hyst. Rules Fig. 13.47 Theory vs Computed Rocking Hyst. Rules
12 12~----------r---~~---.----------.
10 10 ........................--......j ....................... -<jII <,..}t ..-_-I
a a.----......+.. ..-....... +.............-~.o
6 6
4 ................... -.-.-.-O~j ..............- ..............-_-+- ......-..1 4~.. ---(Jr_----+ .. ----.... j.. .. --1
Filled: Full Wall Grav. Load Filled: Full Wall Grav. Land
Blank: Half Wall Grav. Load Blank: Half Wall Gray. Load
2 ....---....0 ..-..... - ......---+.......--..--..1 2 Q .. _ . + ...............................+ ............_... _.-.............
O~~~~~-+--~~~~~r-~~~~~
5 10 15 20 0.25 0.4 0.55 0.7

Curvature Ductility Member Damage Indo!!
Fig. 13.48 Beam Curv. Ductility Demands Fig. 13.49 Max, Member Damage Index
377
25 25
Ax. Deform. 1m! Ax. Deform. 1m!
-0.02 O. 2 -0 4 -0.02 2
IFW12RZ, Half FW12RZ, Half
Parkf.NE, RB10 Parkf.NE, RB10
, Winkler Model Winkler Model
Mb. 10B. SF4.56 Mb. 108. SF4.31
15 75
-100 100
E
E
'"~ .
u
;;
-125
o
u..
u..
~
.. -150
-175
20
Fig. 13.50 Force - Deform. Relationship of Soil Fig. 13.51 Force - Deform. Relationship of Soil
225 ,-------,-------,-------,-------,
25 Square: S.L. 10 m.
Circle : S.L 20 m.
Al(. Deform. (m) Triang_: S.L 30 m.
200 FLHS Modelllll.. ",+_,,..,1
-0 4 0.02 Filled : Yield Force

Blank ; Compo Force
FW12RZ. Half
Parkf.NE. RB10
Winkler Model
Mb. 108. SF3.57 -50
-75
100
E
.
u
;;
-125
u..
,(
-150
-175 -::-
100 ~~,~--~~~~~~~~~--~~
zoe. J 2.75 3.25 3.75

SSC Safety Factor
4.25 4.75
Fig. 13.52 Force - Deform. Relationship of Soil Fig. 13.53 Force liS. SBe Safety Factor of Soil
378
13.5 STATIC BEARING CAPACITY SAFETY FACTOR OF SOILS
Before discussing the effects of the static bearing capacity safety factor of soils on the
soil stress using a variety of soil models, it is necessary to discuss the response and the soil
stress of the redesigned wall footing carries one-half of the gravity load. As shown in Fig.
11.32, the maximum soil stress stiII remains elastic because the static bearing capacity safety
factor (SF) of the soil is relatively high. Fig. 11.37 shows that the soil layer with a depth of
lO metres causes the highest footing lift-off. The comparison of footing rotation between the
Winkler and FLRB soil model with a layer of 10 metres, where the wall carries one-half of
gravity load and the footing is redesigned with a SF of 3.154, is shown in Fig. 11.46. The
geometry of the Winkler model is shown in Fig. 11.50. The Winkler model shows a higher
footing rotation although. the difference in the maximum footing rotation is less than to the
FLHS model as shown in Fig. 11.50. The maximum footing rotations of the Winkler models
in these figures are nearly the same, but the FLRB soil model shows a higher rotation than the
FLHS model. This is due to the FLRB model possessing a relatively smaller rocking stiffness
than the FLHS model.
The vertical displacement of the wall base node and the corresponding vertical
displacement of the wall footing tips are shown in Figs. 11.47 and 11.48 respectively. The
Winkler model depicted in Fig. 11.47 shows that, after the earthquake induced oscillation
vanishes the veltical displacement of the wall base node does not return to the initial
displacement. The reason is due to the inelastic deformation of part of the soil foundation. The
wall moment and footing rotation relationships of a wall supported by the FLRB is shown in
Fig. 11.49. The rocking stiffness reduces as the tip of the wall footing lifts-off, as depicted in
Fig. 11.48.
The variation between force and deformation of the soil represented by springs 93, 106,
107 and 108 are shown in Figs. 11.51, 11.52,11.53 and 11.54 respectively. They show that,
with a static bearing capacity safety factor of the soil, SF == 3.154, the soil beneath the tip of
the footing foundation, as indicated by springs 107 and 108 or shown in Figs. 11.53 and 11.54,
show significant yielding. The comparison of the beam curvature ductility demand and
maximum member damage indices of the 12-storey wall frames supported by these soil models
are shown in Figs. 11.55 and 11.56. As with the previous results, the effects of the different
soil models on the structural responses does not appear significant.
379
The effect of the static bearing capacity safety factors of the soil on the soil stresses
using different soil models are investigated. The wall footing foundations are redesigned with
as the length to width ratio varies between 1.166 to 1.365 corresponding to a SF of the soil of
4.56,4.31,4.10, 3.57 and 3.109 respectively. The analyses involves the FLHS and FLRB soil
models with the depth of the soil layer of 10, 20 and 30 metres respectively. The force-
deformation relationships of the soil corresponding to the safety factor 4.56, 4,31, 3.57 and
3.109 for the FLRB with the depth of the soil layer of 10 metres are depicted in Figs. 13.50,
13.51 and 13.52 respectively. The variation between the axial forces versus static bearing
capacity safety factor of soil for difference soil models is shown in Fig. 13.53.
It shows that the required static bearing capacity safety factor where the soil
commences to yield varies with the soil models and depth of the soil layers. The required static
bearing capacity safety factor is higher in shallower soil layers than that in the deeper layers.
This is because the initial vertical displacement of the wall footing in shallower soil layers is
less than that in deeper layers. On the other hand, as shown in Fig. 11.37, the wall footing lift-
off is more pronounced in shallow soil layers than in deep layers. As a result, with the same
static bearing capacity safety factor, the vertical forces are greater in the FLRB with shallower
soil layers than that for deeper layers.
13.6 THE SOIL-FOUNDATION MODEL
The soil deposit may be non-homogenous in both the soil properties and soil geometry.
The soil layer arid rigid boundary are commonly assumed to have horizontal planes. The
properties of soil layers are often transformed to be an equivalent properties in a homogenous
elastic half-spaceHlI.S19. The effects of the soil stiffnesses and soil models on the damage
parameters of the 12 frame-walls supported by partially embedded foundation footings have
been discussed in Chapter XI. It was found that the choice of the soil models does not
significantly affect the structural responses when compared with the effects of the stiffnesses
of the soft soils. The effects of the depth of the soil layer on the structural responses are also
insignificant.
Based on the above results, the effects of the soil geometry is less impOltant than the
effects of the soil properties. Hence. in the soil-structure interaction analysis of building
structures the use of equivalent soil properties of the soil layers is justified. The soil stresses
found immediately beneath of the foundation footings controls the choice of the soil model.
380
Investigations into the analysis and design recommendations for rocking structures have
been presented. The responses of frames with LFix foundations were close to the LFlex
foundations. The response quantities of frames with SCFlex foundations and SFix foundations
were very close. For the frame structures supported by partially embedded foundations and
frame structures supported by SCFlex foundations, the approximation of the building lateral
fixity at the level of foundation footing and at the ground surface were suggested. The
structural responses of the wall-frame structures supported by SFix and LFix foundations were
significantly different when compared with LFLex foundations. If the frame-wall buildings
supported by SCFlex foundations the building lateral fixity at the level of foundation footing
is suggested. The energy dissipated by the soil-foundation systems was very significant.
A new wall moment design envelope for wall in wall-frame structures with flexible
foundations was proposed. Several analyses using the new wall moment design envelope with
full and one-half of gravity load were carried out. It was found that no wall plastic hinges
occurred in the upper storeys.
A significant foundation footing lift-off was found when the structural wall carried only
one-half of the gravity load. Taking into account the effect of the decrease in the soil-
foundation impedance on the structural response, hysteresis rules for the soil-foundation
impedances have been developed. A number of analyses to examine the behaviour of the
proposed hysteresis models were carried out. It was found that the hysteresis models agreed
well with the theoretical one. A decrease in the wall gravity load causes a slightly increase in
the structural responses.
To avoid yielding of the soil beneath the footing foundation tips, the location of the wall
in the interior plane-frames is suggested. If structural walls were located in the outer plane-
frames of the building, in general, the static bearing capacity safety factor of soil SF should
be taken exceed than 4.0.
The effect of soil properties is more significant than the effect of soil geometry. The
used of equivalent soil properties of the soil layers in the choice of soil model is justified. The
soil stress immediately beneath the footing and the foundation embedment are two important
aspects in the choice of the soil-foundation model.
Chapter XIV
Conclusions and Recommendations for Further Research.
14.1 GENERAL
Research work on the rocking of multi-storey buildings has been carried out. The
research investigated the effect of parameters on the inelastic responses of the 12 and 18-storey
frames and frame-wall structures. The main parameter was the effects of soil-foundation
flexibility in comparison with the fixed base foundations. The other effects were the rigid joint
lengths, soil-shear strains, soil moduli, number of frames to walls, earthquake excitations and
flexural hysteresis rules for the members. The damage indicators from the inelastic response
quantities included the member, storey and structural damage indices. These quantities include
the fundamental periods, roof horizontal displacements, member forces, inter-storey drifts,
curvature ductility demands, plastic hinge rotations and energy dissipations.
The present wall moment design envelopes for frame-wall structures and the location
of the building'S lateral fixity were reviewed. In addition, the effect of the wall gravity loads
and soil models on the structural responses were examined. Based on the results of the
investigations, provisions for the analysis and design of rocking structures were recommended.
Each chapter, in general, has own summary and conclusions.
14.2 CONCLUSIONS
The major results and findings of this research are as follows :
1. The flexibility of the foundations shifts the fundamental period of the structures as reflected
in earlier findings Cl .C4 ;m,P6. These period shifts are slightly affected by the intermediate soil
stiffnesses and are significantly affected by soft soil stiffness. A greater period shift is
found for relatively stiff structures such as frame-walls than in the flexible structures such
as the frame structures. This confirms the results as presented in Ref.R3. The
381
382
approximate formulae for the fundamental natural period proposed by ATC 3-06 (1978)
overestimate the fundamental period of the structures. This shift in the period affects the
structural responses.
2. The shift of the structural fundamental periods reduces the base shear. A decrease in the
base shear is negligible for framed structures but is more obvious in the relatively' stiff
structures such frame-wall structures. With respect to the flexible foundations, the
assumption of base fixity of the frame-wall structures at the ground level overestimates the
total base shear. Rotation of the structural walls decreases markedly the wall base shear and
at the same time increases the frame base shear. More appropriate detailing in shear
reinforcement of the column bases of the frames is necessary. Frame structures with stub
columns (SCFlex) showed very similar base shear to the frames fixed at the ground level
(SFix). The total base shear of the frame-wall structures are more than double the base
shear resisted by the equivalent frame structures.
3. Rotation of the wall bases increases the column base moments quite markedly. This rotation
redistributes the maximum wall moment from the base to the middle height storeys. This
phenomenon shows an inadequacy of the present wall moment design envelopes. The
approximate formula of wall height to width ratio and new wall moment design envelopes
show an improved wall performance.
4. The flexibility of the foundations increases the inter-storey drifts in the upper storeys and
decreases it in the lower storeys of framed structures. In the frame-wall structure, this
flexibility tends to increase the inter-storey drifts in the lower and upper storeys and
decreases it in the middle storeys. The maximum inter-storey drifts might less than the limit
value specified by the NZS 4203: 1984 and NZS 4203: 1992 loading codes.
5. The flexibility of the foundations increases the columns base plastic hinge rotations. A
significant increase in the rotations is found in the column bases of frame-wall structures
due to larger shears and moments. The rotation of the wall foundation footings decreases
the plastic hinge rotations of the wall bases. The inter-storey drifts tend to increase as the
storey plastic hinge rotation increase.
6. The flexibility of foundations, in general, decreases the beam curvature ductility demands.
In the frame-wall structures, the decrease in these demands are more marked due to the
383
presence of structural walls than the decrease of the ductility demands due to the flexibility
of the foundations, especially in relatively low rise bUildings. In the higher buildings i.e
the IS-storey frame-walls, the effects of the higher modes are more significant.
7. The flexibility of the foundations may decrease or increas~ the total plastic hinge rotations
in framed or frame-wall structures. This rotations strongly correlate with the overall drifts
of the framed structures and the frame-wall structures. The structures with flexible
foundations, in general, show less damage than those with fixed bases.
8. The viscous and hysteresis energy dissipations of the superstructures are strongly affected
by the flexibility of the foundations. This flexibility tends to increase significantly the
viscous energy dissipation due the increase of velocities resulted from the rocking of the
foundations. On the other hand, the hysteretic energy dissipation of the superstructures
decreases dramatically, especially for the relatively stiff frame-wall structures. In higher
buildings such as the lS-storey frame-walls on the flexible foundations, the inertial
interaction is mo~ obvious and the energy dissipated by the soil-foundation system
becomes more significant. Considering the effects of the foundation flexibiliJ:y on the
structural deformation and the hysteretic energy dissipation, the use of the energy-based
damage model as a single parameter as the damage index of rocking structures becomes
less meaningful.
9. The rigid joint length significantly affects the responses of frame structures. Increasing the
rigid joint lengths decreases the structural fundamental period and increases most structural
responses, particularly in framed structures. In the l8-storey frame-walls with flexible
foundations, however, increasing the rigid joint lengths causes a decrease in the viscous
energy dissipation.
10. The magnitude of the soil-shear strains significantly affect some structural responses such
the beam curvature ductility demands, the total plastic hinge rotations and storey damage
indices. The structural responses are less affected by soils with intermediate stiffness but
are strongly affected by the soils with soft stiffnesses.
11. Increasing the number of frames keeping the number of walls constant slightly increases
the fundamental period of the structure and tends to decrease the roof horizontal
displacements. Increasing the number of frames increases the flexibility of the
384
superstructure but stiffens the soil-foundation impedances for the reasons discussed in
Chapter IX. The effects of increasing the number of frames on some structural responses
of the frame-wall structures are velY complex and the results cannot be generalized.
12. The effects of the acceleration to velocity, a/v, ratio on the structural responses are more
obvious. The earthquakes with a relatively high peak ground velocity or Iowa/v ratio, such
as the 1977 Bucharest N-S component tends to cause more severe damage than the 1966
Parkfield N-E component ealthquake even though the peak ground acceleration of the
former is less than one-half of the latter ealthquake's peak ground acceleration. It is found
that the effect of the foundation flexibility on the structural responses, in general, depends
on the spectral characteristics of the applied earthquake excitation in the region of the
structural fundamental period. This result agrees with the conclusions discussed in
Ref. V 10.
13. The structural responses, in general, are not significantly affected by different flexural
hysteresis rules for the members. In some cases, the Bilinear rules shows a different
response when compared with the others. The Q-Hyst dissipates more hysteretic energy
than does the modified Takeda rules, especially for low amplitude cyclic responses. Due
to a higher unloading stiffness, the Clough rule dissipates more hysteretic energy and
results in smaller damage indices.
14. It was found that two damage indicators, the structural deformation and hysteretic energy
dissipations, are dominant in the case of rocking structures with a strong column-weak
beam failure mechanism. For ductile members with low values of the deterioration index
{3, the corresponding member damage index is strongly correlated with the member
deformation rather than with the hysteretic energy dissipation. A weighting factor for the
storey and structural damage indices based on the member deformation is suggested. It was
found in the structural prototypes under consideration that the flexibility of the
foundations, in general, reduces the member, storey and structural damage indices.
15. A wall footing with a small gravity load has a greater tendency to lift-off. The proposed
foundation footing hysteretic rules well simulate the soil-foundation impedances before and
after footing lift-off. The footing lift-off significantly affects the soil stress but does not
significantly affect the structural responses. The static bearing capacity safety factor SF of
385
the soil higher than 4.0 is needed when the walls are located in outer plane-frames in the
building.
16 If the frame building structures are supported by LFlex and SCFlex partially embedded
foundation footings the approximation to the building lateral fIXity at the ground level and
the level of foundation footings respectively is suggested. If the frame-wall buildings are
supported by SCFlex partially embedded foundation footings the approximation that the
building lateral fixity is at the level of the foundation footings is still justified.
17 The type of the soil model does not significantly affect the structural responses. The
properties of soils more significantly affect the structural responses than does the geometry
of the soils. If the variation of soil properties in the layered soils are not greatly different
the use of equivalent soil properties is justified. The effect of the soil stresses and the effect
of foundation footing embedment are two important considerations in the selection of the
soil-foundation model.
18 The dynamic responses of the building structures supported by flexible foundations are
different from those of structures on fixed bases. This difference is significant for stiffer
structures such as frame-wall buildings. Considering the effects of the soil-structure
interactions by introducing the dynamic foundation factor as discussed in Refs. E8 and H30
may not give accurate predictions of the inelastic responses of the rocking stiff structures.
The effects of the foundation flexibility on the inelastic responses of rocking stiff structures
should be employed directly in the analyses.
14.3 RECOMMENDAnON FOR FURTHER RESEARCH
Some major results and findings on the effects of foundation flexibility on the inelastic
responses of frame and frame-wall structures have been highlighted. However, to more
accurately estimate both of strength and deformation demand and supply in the aseismic design
of reinforced concrete structures, further intensive research, both in the laboratory and in
analytical investigation is necessary.
386
14.3.1 Laboratory Work
1. Even though the concrete joint is expected to response elastically, laboratory tests
indicate that it is difficult to prevent inelastic deformation in the beam plastic hinge
spreading through into the joint. Laboratory work to investigate a more accurate
evaluation of the equivalent rigid joint length for the concrete joint is important. This
is because there is no effective rule for estimating this length and the effect of the rigid
joint length on the inelastic responses of frame structures is significant.
2. The soil-foundation compliancies are basically derived from the design of machine
foundations where a steady state harmonic loading is applied to the foundation footing.
Laboratory tests on footings with non-harmonic transient loadings through the base is
likely to more closely represent the earthquake loading. The effects of these two
different loading systems on the soil-foundation impedances is still uncertain.
14.3.2 Analytical Work
1. The soil shear strain in this study is assumed to be constant throughout the duration of
the excitation which implies a constant soil shear modulus and linear soil-foundation
impedances. In reality, however, the soil shear strain is time dependent. Considering
a time dependent effect on the soil shear modulus will provide more realistic analyses.
2. The dynamic soil-foundation impedances used in this study are also assumed to possess
constant coefficients, meaning that the analyses are frequency independent. As is
discussed in many references, the dynamic effects are frequency dependent and result
in the soil-foundation impedances being not only time but also frequency dependent.
A more precise analyses should take into account these variations. Work is also
required to determine how these effects can be used in non-linear time-domain analyses
where modal methods cannot be used.
3. Using frames and walls designed to the codes, there is probably an optimum
combination of frames and walls. At the present time there is no clear guidance on this
appropriate combination. Analytical investigations into this problem lead to interesting
results when using different combinations of material strengths, as the changes in
strength and stiffness are not linearly correlated.
387
4. The use of different types of the walls such as coupled-walls or box-type walls and
more refined wall modelling will give more insight into the behaviour of rocking frame-
wall structures. A more comprehensive investigation may also be achieved by
considering the flexibility of the floor slab and using a more suitable modelling of walls
such as a filament or spring element models.
5. The foundations of stmctural walls often use a group of pile foundations especially for
high-rise buildings located on a relatively soft soils. The inelastic behaviour of the
rocking frame-wall structures supported by pile foundations needs investigation. The
problems include the dynamic impedances of a soil-pile group foundation for different
pile bearing systems, the lift-offbehaviour and the effect of kinematic interaction in the
soil-structure interaction analysis as the piles are deeply embedded in the soil medium.
6. The member damage index used in this study assumes that the concrete members are
very ductile so that the deterioration index fJ is relatively smalL A study of the member
damage using different levels of the deterioration index becomes more significant when
combined with different rigid joint lengths and different material strengths.
7. The selection of the method of analysis of the member damage index depends mainly
on the typical damage mode observed under the dynamic loading. The appropriate
choice of this mode will be achieved by conducting a more comprehensive study into
the behaviour and characteristics of earthquake ground motions on the variety of
structural configurations.
388
References
Abbreviation Lists :
ASCE : American Society of Civil Engineer

NZSEE: New Zealand National Society for Ea11hquake Engineering
WCBB World Conference on Earthquake Engineering
ACI American Concrete Institute
EESD Ealthquake Engineering and Structural Dynamics
SDEE Soil Dynamics and Ealthquake Engineering
[AJ] ABDEL-HADY, HERRIN M, Characteristic of Soil Asphalt as A Rate of Process,

Journal of the Highway Division ASCE, Vo1.92, HW1, pp. 49-69, March 1966
[A2] ANDRIONO T, Seismic Resistant Design of Multistorey Structures, Ph.D Thesis

Department of Civil Engineering University of Canterbury, 1989.
[A3] ANDRIONO T, CARR A.J, Reduction and Distribution of Lateral Seismic Inertia
Forces on Base Isolated Multistorey Structures, Bulletin of the NZSEE), Vo1.24, No.3,
pp. 225-237, September 1991
[A4J ANDRIONO T, CARR AJ, A Simplified Earthquake Resistant Design Methodfor Base
Isolated Multistorey Structures, Bulletin of the NZSEE, Vo1.24, No.3, pp. 238-250,
September 1991
[A5] ANANDAKRISHNANM, KRISHNASWAMY N.R, Response of Embedded Footing

to Vertical Vibrations, Journal of the Soil Mechanic and Foundation Division ASCE,
Vo1.99, SMlO, pp. 863-883, October 1973
[A6] AKTAN A.E, BERTERO B.B, Conceptual Seismic Design of Frame Wall Structures,
Journal of the Structural Engineering, ASCE VoI.1lO, No.ll, pp. 277S-2797, 1984
[A7] AKTAN A.E, BERTERO V. V, Seismic Response ofRC Frame Wall Structures, Journal
of the Structural Engineering, ASCE, Vol.llO, No.S, pp. 1803-1821, 1984
[AS] ANDERSON J.C, TOWNSEND W.H, Models for RC Frames with Degrading
Stiffness, Journal of the Structural Division, ASCE, VoLl03, ST12, pp. 2361-2371,
1977
[A9] AOYOMA H, Outline of Earthquake Provisions in the Recently Revised Japanese

Building Code, Bulletin of the NZSEE VoI.14, No.2, pp. 63-13-, 1981
[A10] ARCHER J.S, Consistent Mass Matrix for Distributed mass Systems, Journal of the
Stmctural Division, ASCE, Vo1.89, ST4, pp. 161-177, 1963
[All] ANDERSON D.G, RICHART F.E, Effects of Straining on Shear Modulus of Clays,
Journal of the Geotechnical Engineering Division, Vo1.102, GT9, pp.975-987, 1976
389
[A12] ARAKAWA T, Finite GroW1.d Strains Induce During Earthquake for Application to
Seismic Design of Underground Structures, Fifth Int. Conference on Numerical Methods
in Geomechanics, Nagoya, pp. 1513-1518, Apri11985
[A 13] AYDINOGLU M. N, CAKIROGLU A, Dynamic Interaction Between Soil and a Group

of Buildings , Proceeding of the Sixth WCEE, New Delhi, Vol.U, pp.1S96-1600, 1979
[A14] AHMAD S, GAZETAS G, Torsional Stiffness of Arbitrary Shaped Embedded

Foundations, Joumal of the Geotechnical Engineering ASCE, VoLl18, No.8 1992, pp.
68-1185, 1992
[A15] ARNOLD C, REITHERMAN R, Building COI'ifiguration and Seismic Design, John

Wiley & SOIlS, New York, 1982
[A16J ABDEL-FATAH B, WIGHT 1.K, Study of Moving Beams Plastic Hinging Zone for
Earthquake Resistant Design, ACI Structural Joumal, pp. 31-39, Jan-Feb. 1987
[A17J AL-HADDAD M.S, WIGHT J.K, Relocating Beam Plastic Hinging Zone for
Earthquake Resistant Design of RC Buildings, ACI Structural Journal, pp. 123-133,
March-April 1988
[AlS] ANTES H, ESTORFF O.V, Dynamic Response Analysis of Rigid Foundations and of
Elastic Structures by Bowulary Element Procedures, SDEE, Vol.8, pp. 68-73,
1989
[A19] ANG A.H.S, State of The An Report: Probabilistic Seismic Sqfety and Damage
Assessment ofStroctures, Proceeding ofthe Ninth WCEE, Tokyo-Kyoto, Vol. VIII, pp.
717-728, 1988
[A20] AHMAD S, GAZETAS G, Torsional Radiation Damping of Arbitrary Shaped

EmbeddedFoundations, Journal ofthe Geotechnical Engineering ASCE, Vol. 118, No.8,
pp. 1168-1185, 1992
[A21] AMBRASEYS N.N, ELNASHAI A.S, BRODERICK B.M, SALAMA AJ, SOLIMAN
M.M, The Erzincan Turkey EanJuJuake 13 March 1992, EERC, No. 92, Dep. of Civil
Engineering, Imperial College of Science, Technology & Medicine, December 1992
[Bl] BEREDUGO Y 0, NOVAK M, Coupled Horizontal and Rocking Vibration of

Embedded Footings, Canadian Geotechnical Journal, Vo1.9, pp. 477-497, 1972
[B2] BIGGS J.M, Introduction to Structural Dynamics. Mc Graw-HiU Publishing Company,

New York, 1964
[B3] BOWLES J.E, Foundation Analysis and Design, Me Graw-Hill Publishing Company,
New York, 1988
[B4] BALENDRA T, LEE S.L, Simplified Dynamic Model for Soil Structure Interaction
Problems, Proceeding the Ninth WCEE, San Francisco, Vol.VIII. pp. 926-936, 1984
390
[BS] BATHE K.J, WILSON E.L, Stability and Accuracy Analysis of Direct Integration
Methods, EESD, Vol.1, pp. 283-291, 1973
[B61 BEHPOOR L, GHAHRAMAMI A, Correlation of SPT to Strength and Modulus of

Elasticity os Cohesive Soils, Proceeding of the Twelfth Int. Conference on Soil
Mechanics and Foundation Engineering, Rio De Janeiro, YoU, pp. 175-178, 1989
[B7] BERTERO V. V , State of the Art Report .' Ductility Based Structural D esig n, Proceed ing
of the Ninth WCEE, Tokyo-Kyoto, Vol. VIII, pp. 673-685, 1988
[B81 BERTERO V.V, Seismic Behaviour ofRC Wall Structures, Proceeding of the Seventh
WCEE, Istambul Turkey, Vo1.6, pp.323-330, 1980
[B9] BROOKER E.W, IRELAND H.O, Earth Pressures at Rest Related to Stress History,
Canadian Geotechnical Journal, Vol.II, pp. 1-15, 1965
[BIO] BALDI G, JAMIOLKOWSKI M, LO PRESTI D.C.F, MANFREDINI G, RIX G.J,

Italian Expen'ence in Assessing Shear Wave Velocity from CPT and SPT, Proceeding
ofthe Twelfth International Conference on Soil Mechanics and Foundation Engineering,
Rio De Janeiro, ( Discussion Session ), pp. 157-167, 1989
[B 11] BERRILL J.B, DAVIS R. 0, McCAHON I.F, Christchurch Seismic Hazard Pilot Study,
Bulletin of the NZSEE, Vol. 26, No.1, pp. 14-27, 1993
[B12] BANON H, BIGGS J.N, IRVINE H.M, Seismic Damage in Reinforced Concrete
Frames, Journal of Structural Division, ASCE, ST9, pp. 1713-1729, 1981
[B13] BANON H, VENEZIANO D, Seismic Safety of Reiriforced Concrete Members and

Structures, EESD, Vo1.10, pp. 179-192, 1982
[B14] BARKAN D.D, Dynamic Bases and Foundations, McGraw Hill Book Company Inc,
New York, 1962
[B15] BARKAN D.D, ILYICHEV V.A, Dynamic of Bases and Foundations, Proceeding of
the Ninth International Conference on Soil Mechanics and Foundation Engineering,
Tokyo, Vo1.2, pp. 630-637, 1977
[B 16] BROKKEN S. T, BERTERO V. V, Studies on Ejlects of Infills in Seismic Resistant RC

Constructions, Report No.UBC/EERC-81112, Dep. Of Civil Engineering, University
of California, Berkeley, Cali fornia, 1981
[BI7] BERTERO V.V, BERTERO R.D, Formulation of a Conceptual Seismic Codes,

Proceeding of The TOM PAULAY Symposium; Recent Development in Lateral Force
Transfer in Buildings, ACI, EERC and UCSD, September 1993.
[B 18J BERTERO V. V, Evaluation of Response Reduction Factors Recommended By ATC and

SEADC, Proceeding of the Third National Conference on Earthquake Engineering,
South Carolina, pp. 1663-1673, 1986
[B 191 BERTERO R. D, BERTERO V. V, Conceptual Earthquake Resistant Design

MethoMlogy, Report No. UCB/EERC-92116, Dept. of Civil Engineering,University
of California, Berkeley California, 1992
391
[B20] BERTERO V. V, BENDlMERAD F. M, SHAH H.C, Fundamental Period ofReinforced

Concrete Moment Resisting Frames, Report No.87, The John A. Blume Earthquake
Engineering Centre, Dept. of Civil Engineering, Stanford University, 1988
[B2l] BERTERO V.V, BROKKEN S, Infills in Seismic Resistant Buildings, Journal of the
Structural Engineering Division, ASCE, Vol. 109, No.6, pp. 1337-1361, 1983
[B22] BERTERO V. V, Lessons Learned from Recent Earthquakes and Research

Implementations for EtJrthquake Resistant Design of Building Structures, Earthquake
Spectra, Vol.2, NoA, pp.825-856, 1986
[B23] GRANDORIG, BENEDETII D, On the Choice of the Acceptable Seismic Risk, EESD,
Vol.2, pp.3-9, 1973
lB24] BARTLETT P.E, Foundation Rocking on a Clay Soil, Report No.154, School of
Engineering, Dept. of Civil Eng., University of Auckland. 1976
[B25] BERTERO V.V, UANG C.M, Issues and Future Directions in the Use of An Energy
Approach for Seismic Resistant Design of Structures, Nonlinear Seismic Analysis &
Design of RC Buildings, Edited by Fajfar & Krawinkler. Elseiver Applied Science,
1992
[B26] BENDIMERAD F.M, SHAH H.C, HOSKINS T, Extension of tlhe Study on :

FwuJamental Period of RC Moment Resisting Frame Structures, Report No. 96, The
John Blume Earthquake Engineering Research, Dept. of Civil Engineering. Stanford
University, 1991
[B27] BACHMANN H, UNDE P, Dynamic Ductility Demand and Capacity Design of

Earthquake Resistant RC Walls, Proceeding of the "Tom Paulay Symposium", ACI-
EERC-UCSD, pp. 115-139, September 1993
[B28] BENUSKA L, Loma Prieta Earthquake, Reconnaissance Report. Earthquake Spectra,

Vo1.6, Chapter V, pp 127-149, 1990
[B29] BOOTH E, Concrete Structures in Earthquake Regions: Design and Analysis, Longman
Scientific & Technical, 1994
[B30] BERRILL J, DAVIS R.O, Engineering Seisnwlogy Notes, Dept. of Civil Engineering,
University of Canterbury NZ (Unpublished)
[Cl] CASTELLANI A. Foundation Compliance Effects on Eal1hquake Response Spectra,

Journal ofthe Soil Mechanics and Foundation Division, ASCE, Vo1.96, SM4, pp. 1335-
1355, July 1970
[C2] CHAPRA S.C, CANALE R.P, Introduction to Computing for Engineers, Mc Graw-
Hill Book Company, New York, 1986
[C3] CHIK SIM YIM, A.K CHOPRA, Effects of Trallsient Foundation Uplift on Earthquake
Response of Structures, Report No. UCB/EERC-83/09, College of Engineering
University of California Berkeley, 1983
392
[C4] CHIK SIM YIM, A.K CHOPRA, Earthquake Response of Structures with Partial Uplift
on Winkler Foundation, EESD, VoI.12, pp. 263-281, 1984
[C5] CHRISTENSEN R.W, WU T.W, Analysis of Gay Deformations A Rate Process,

Journal of the Soil Mechanics and Foundation Division ASCE, Vo1.90, SM6, pp. 125-
157, November 1964
[C6] CHOPRA A.K , GUTIERREZ J.A, Earthquake Response Analysis of Multistorey

Buildings Including Foundation Interaction, EESD, Vo1.3, pp. 65-77, 1974
[C7] CHOPRA A.K, SOLOMON C.S YIM, Simplified Earthquake Analysis of Structures
with Foundation Uplift, Journal of the Structural Engineering, ASCE, Vol.ll1, No.4,
pp. 906-930, April 1985
[C7] CLOUGH R.W, HUCKELBRIDGE A.A, Preliminary Experimental Study of Seismic

Uplift of a Steel Frame, Report No. UCB/EERC-77/22, College of Engineering
University of California Berkeley, August 1977.
[C8] CLOUGH R.W, PENZIEN J, Dynamics of Structures, 2nd Edition, Mc Graw-Hill,

1992
[C9] CONNOR I.J, BREBBIA C.A, Finite Element Technique for Fluid Flow, Newnes-
Butterworths, London, 1976
[ClO] COOK R, Concept and Applications of Finite Element Analysis, John Wiley & Sons,
New York, 1981
[C1l] CAUCHEY T.K, O'KELLY M.E.J, Classical Normal Modes in Damped Linear
Dynamics Systems, Journal of the Applied Mechanics, pp. 583-588, September 1965
[C12] CARD AN B, Concrete Shear Walls Combined with Rigid Frames in Multistorey
Buildings Subjected to Lateral Loads, ACI Journal No.3, pp. 299-316, September 1961
[C13] CARR A.J, MOSS P.J, The Effect of Large Displacements of the Earthquake Response
of Tall Structures, Proceeding of Seventh WCEE, Istambul, Vol. VIII, pp. 113-116,
1980
[C14] CARR A.l, Ruaumoko, Computer Program Library, Dept. of Civil Engineering,
University of Canterbury, 1993
[C1S] CARR A.J, MOSS P.J, PARDOEN G.C, Imperial County Service Buildings: Elastic
& Inelastic Response Analyses, Research RepOit No. 79/15, Dept. Of Civil Engineering,
University of Canterbury, Christchurch, New Zealand, 1979
[C 16] CARR AJ, Structural Dynamics Notes, Dept. of Civil Engineering, University of
.Canterbury (Unpublished).
[C17] CARR A.J, TABUCHI M, The Structural Ductility and The Damage Index for
Reiriforced Concrete Structure Under Seismic Actions, Proceeding of the Second
European Conference on Structural Dynamics, Norway, VoU, pp.169-176, 1993
393
[CI8] CARTER B.H.P, The Seismic Behaviour of RC Frame Shear Wall Structures, ME
Thesis Report, Dept. Of Civil Engineering, University of Canterbury, 1980
[CI9] CHEUNG P.C, Seismic Design Of Reinforced Concrete Beams Column Joints with
Floor Slab, Ph.D Thesis, Dept. of Civil Engineering, University of Canterbury, New
Zealand, 1991
[C20] CHOPRA A.K, KAN C, Effects of Stijfness Degradation on a Ductility Requirement

for Multistorey Buildings, EESD Vol.2, pp. 35-45, 1973
[C21] CLOUGH R. W, WILSON E.L, KING LP, Large Capacity Multistorey Frame Analysis
Programs, Journal of the Structural Division, Vo1.89, ST4, pp. 179-204, 1963
[C22] CLOUGH R.W, MOJTAHEDI S, Earthquake Response Analysis Considering Non

Proportional Damping, EESD VolA, pp. 489-496, 1976
[C23] CRESPELANI T, GHINELLI A, VANNUCHI G, An Evaluation of the Dynamic Shear

Modulus of a Cohesive Soils, Proceeding of the Twelfth International Conference on Soil
Mechanics & Foundation Engineering, Rio De Janeiro, VoUlI, pp. 1935-1940, 1989
[C24] CHOPRA A.K, HEJAL R, Earthquake Response of Asymmetric Frame Buildings,

Proceeding of the Ninth WCEE, Tokyo-Kyoto, Vol.V, pp.31-36, 1988
[C25] CLOUGH R. W, BENUSKA K.L, WILSON E.L, Inelastic Earthquake Response of Tall
Buildings, Proceeding of the Third WCEE, New Zealand, Vol.lI, pp.68-89, 1965
[C26] CHAI Y.R, PRIESTLEY M.J.N, SEIBLE F, Seismic Retrofit of Circular Bridge
Columnfor Enhanced Flexural Response, AC] Structural Journal, pp. 573-584, Sept.-
Oct., 1991
[C27] CHEN W. W, CHATERJEE M, UNEMORI A, Comparison of Analysis Method for

Seismic Soil-Structure Interaction, Proceeding of the Seventh WCEE, ]stambul, Turkey,
Vol.V, pp.141-148, 1980 .fl.
[C28] CAUGHEY T.K, Classical Nonnal Modes in Damped Linear Dynamic System, Journal
of Applied Mechanics, June, pp.269-271, 1960
[C29] COSENCA E, MANFREDI G, RAMASCO R, The use of Damage Functional in

Eartlujuake Engineering: A Comparison Between Different Methods, EESD, Vo1.22,
pp. 855-868, 1993
[D 1] D AWE D. J , Matrix and Finite Element Displacement Analysis of Structures, Clavendon

Press, London, 1984
[D2] DESAI C.S, Elementary Finite Element Method ,Prentice Halt Inc, New Jersey, 1979.
[D3] DOBRY R.D, GAZETAS G, Dynamic Response of Arbitrarily Shaped Foundations,

Journal of Geotechnical Engineering, Vo1.112, No.2, pp. 109-135, Febmary 1986
[D4] DOBRY R.D, GAZETAS G, STOKOE K.H, Dynamic Response of Arbitrarily Shaped
Fowzdation : Experimental and Verification, Journal of the Geotechnical Engineering
ASCE, Vol. 112, No.2, pp. 136-154, February 1986
394
[DS] DUNCAN J.M, YANGC.Y, Nonlinear Analysis of Stress and Strain in Soils, Journal
of the Soil Mechanics and Foundation Division,ASCE, Vol.96, SMS, pp. 1629-1653,
September 1970
[D6] DASB.M, Fundamentals oj Soil Dynamics, Elseiver, New York, 1983
[D7] DELINIC K, Dynamic Response of Non-Proportionally Damped Mechanical Systems,

Proceeding of the Eighth WCEE, San Francisco, pp. 379-384, 1984
[D8] DUNCAN P .E, TAYLOR R.E, A Note on the Dynamic Analysis of Non-Proportionally
Damped Systems, EESD Vo1.7, pp. 99-105, 1979
[D9] DOWRICK D.1, Eanhquake Resistant Designfor Engineers and Architects, John Wiley
& Sons, 1987
[D 10] DOWRICK D.J, Magnitude Reassessment of New Zealand Earthquakes, EESD VoL 20,
pp. 577-596, 1991
[D11] DOWRICK D.l, Attenuations of Modified MercaW Intensity in New Zealand, EESD,
Vol.21, pp. 181-196, 1992
[D12] DOWRICK D.l, SRITHARAN S, Attenuation of Peak Accelerations in Some Recent

New Zealand Eanhquakes, Bulletin of the NZSEE, VoJ.26, pp.3-13, 1993
[D 13] DORECHO A. T, Seismic Design of Reinforced Concrete Structures, Chapter IX of The

Seismic Design Handbook, Edited by Farzard Naeim, Van Nostrand Reinhold, New
York, 1989
[D 14] De BARROS F. C.P , LUCO J .E, Discrete Models for Vertical Vibrations of Surface and
Embedded Foundations, EESD, Vol. 19, pp. 289-303, 19
[El] EDMOND F.D, CARR A.l, GOLDSMITH P.R, NORTH P.J, WOOD J.H,
PRESTON R.L, Seismic Design of Bridges , Bulletin NZSEE , Vol. 13, No.3, pp. 248-
261, September 19.~O
[E2] EL HIFNAWY L, NOVAK M, Uplift in Seismic Response of Pile Supported Building,

EESD, Vol. 14, pp.573-593, 1986
[E3] EVISON R.J, Rocking Foundation , ME Thesis Report, Department of Civil

Engineering, University of Canterbury, 1977
[E4] ERDIK M, GULKAN P, Assessment of the Soil-Structure Interaction Effects on

Prefabricated Structures, Proceeding of the Eighth WCEE, San Francisco, VoL VIII ,
pp. 913-920, 1984
[E5] ELTON D.I, RAY R.P, Resilient Moduli of Soils, ASeE Geotechnical Special
Publications, No.24, 1989
[E6] ENRIQUE DEL VALLE e, Earthquake Damage to Non-Structural Element, Proceeding

of the Eighth WeEE, San Francisco, Vol. V, pp. 1127-1134, 1984
395
[E7] ESTORFF O.V, KAUSEL E, Coupling of Boundary Element and Finite Elementfor
Soil Structure Interaction Problems, EESD, Vol. IS, pp. 1065-1075, 1989
[E8] ELHMADI K, HEIDEBRECHT A, A Proposed Dynamic Foundation Factor for the

National Building Codes of Canada, Canadian Journal of Civi Engineering, VoL 18,
pp. 974-984, 1991
[F 1.] F AlF AR P, VIDIC T, FISCHINGER M, Seismic Demand in Medium and Long Period
Structures, EESD Vo1.18, pp. 1133-1144, 1989
[F2] FANG H.Y, Foundation Engineering Handbook, Van Nostrand Reinhold, New York,
1991
[F3] FU C.S, YU Z.D, Analysis ofDynamic Behaviour of Structures Including the Effect of
an Adjacent Buildings, Proceeding of the Eighth WCEE San Francisco, Vol.VIII, pp.
833-840, 1984
[F4] FOTOPOULO M, KOTSANOPOULOS P, GAZETAS G, TASSOULAS J .L, Rocking

Damping of Arbitrary Shaped Embedded Foundations, Journal of the Geotechnical
Engineering ASCE, Vol. 115, No.4, pp. 473-490, 1989
[FS] FILLIPOU F.C, ISSA A, Non-linear Analysis of Reinforced Concrete Frames Under
Cyclic Load Reversals, Report No. UCB/EERC-88/12, Dept. of Civil Engineering,
University of California at Berkeley, California, 1988
[F6] FRASER D.J, Conceptual Design and Preliminary Analysis of Structures, Pitman
Publishing Inc, Massachusetts, 1981
[F7] FUKUSHIMA Y, TANAKA T, A New Attenuation Relation for Peak Horizontal

Acceleration of Strong Earthquake Ground Motions in Japan, Bulletin of the
Seismological Society of America, Vol.80, No.4, pp.757-783, 1990
[F8] FINTEL M, GHOSH S.K, Application of Inelastic Response History Analysis in the
Seismic of a 35 Storey Frame-WaU Structures, EESD, VoL9, pp.543-556, 1981
[F9] FAJFAR P, VIDIC T, FISHINGER M, On Energy Demand and Supply in SDOF

Systems, Nonlinear Seismic Analysis & Design of RC Buildings, Edited by Fajfar &
Krawinkler, Elseiver Applied Science, 1992.
[G1.] GAZETAS G, DOBRY R.D, TASSOULAS J.L, Vertical Response of Arbitrarily

Shaped Embedded Foundations , Journal of the Geotechnical Engineering ASCE,
Vo1.111, No.6, pp. 750-771, June 1985
[G2] GAZETAS G, TASSOULAS J.L, Horizontal Stiffness of Arbitrarily Shaped Embedded

Foundation, Journal of the Geotechnical Engineering ASCE, VoU13, NO.5, pp. 440-
475, May 1987
[G3] GRIFFITHS D.V, SMITH LM, Numerical Methods for Engineers A Programming
Approach, Blackwell Scientific Publications, Oxford London, 1991
396
[G4] GAZETAS G, ROESSET J.M, Vertical Vibration Rf Machine Foundation, Journal of

the Geotechnical Engineering Division ASCE, Vo1.105, GT12, pp. 1435-1454,
December 1979
[G5] GAZETAS G, Analysis ofMachine Foundation Vibrations: State ofArt, SDEE), VoL2,
No.1, pp. 2-42, 1983
[G6] GAZETAS G, Formulas and Charts for Impedance of Surface and Embedded
Foundations, Journal of Geotechnical Engineering, VoLlI7, No.9, pp. 1363-1381, 1991
[G7] GOODSIR W.l, The Response of Coupled Shear Walls and Frames, ME Thesis Rep01t
Dept. of Civil Engineering, University of Canterbury, 1982
[GB] GOODSIR W.J, PAULAY T, CARR A.J, A Study of the Inelastic Seismic Response
ofRC Coupled Frame-Wall Structures, Bulletin of the NZSEE VoI.16, No.3, pp. 185-
200, 1983
[G9] GOODSIR W.J, The Design of Coupled Frame-Wall Structures for Seismic Actions,
Ph.D Thesis, Dept. of Civil Engineering, University of Canterbury New Zealand, 1985
[GlO] GUTIERRES J.A, CHOPRA A.K, Evaluation of Methods for Earthquake Analysis of
Soil Structure Interaction, Proceeding of the Sixth WCEE, New Delhi, pp. 1459-1447,
1977
[Gll] GOTO N, Studies on Dynamic Elastic Modulus of Soils, Proceeding Asian Regional
Conference on Soil Mechanics and Foundation Engineering, Israel, pp. 272-275, 1967
[GI2] GAZETAS G, Foundation Vibration: Chapter 15, Foundation Engineering Handbook

Edited by FANG H.Y, Von Nostrand Reinhold, 1991
[Gl3] GAZETAS G, TASSOULAS J.L, Horizontal Damping of Arbitrary Embedded

Fowldations, Journal of the Geotechnical Engineering ASCE, Vol.113, No.5, pp. 458-
475, 1987
[G14] GAZETASG, DAKOULASP, PAPAGEORGIOUP,Local-SoilandSource Mechanism

Effects in the Kalamata (Greece) Earthquake, EESD, Vo1.19, pp. 431-456, 1990
[GI5] GAZETAS G, HATSIKONSTANTINOUE, Elastic Formulaefor Lateral Displacement

and Rotation ofArbitrary Shaped Embedded Foundations, Geotechnique, Vol. XXXVIII,
pp. 439-444
[G 16] GUNTURI S.K. V, SHAH H.C, Building Specific Damage Estimation, Proceeding of
the Tenth WCEE, Madrid, Spain, pp. 6000-6006, 1992
[G17] GUTIERREZ J.A, CHOPRA A.K, A Substructure Methodfor Earthquake Analysis of

Structures Including Structure Soil-Interaction, EESD, Vo1.6, pp.51-69, 1978
[GI8] GUTIERREZ J.A, A Substructure Methodfor earthquake Analysis of Soil-Structure

Interaction, Report no. UCB/EERC-76/9, DepLof Civil Engineering, University of
California at Berkeley, California, 1976
397
[GI9] GUPTA S, PENZlEN J, YEH C.S, LIN T.W, Three Dimensional Hybrid Modelling
of Soil-Structure Interaction, EE5D, VoLlO, pp. 69-87, 1982
[G20] GUANGQIANH, LIAN W, YICUN Z, GOUYING D, A New Procedurefor A seismic

Design of Multistorey Shear Type Buildings Base upon Deformation Checking,
Proceeding The Eighth WCEE, Vol. 6, pp. 483-490, 1984
[G21] GAZETAS G, TASSOULAS T, DOBRY R, O'ROURKE M.J, Elastic Settlement of

Arbitrary Shaped Foundation Embedded in Half-Space, Geotechnique Vol. XXXV,
No.3, pp. 339-346, 1985
[HI] HADJIAN A.H, Soil Structure Interaction - An Engineering Evaluation, Journal of

Nuclear Engineering & Design, Vo1.38, pp. 267-272, 1976
[H2] HALL Jr J.L. KISSENPFENNING J.F, Special Topic on Soil Structure Interaction,
Journal of the Nuclear Engineering & Design, Vol.38, pp. 273-287, 1976
[H3] HARDIN B.O, BLACK W.L, Vibration Modulus of Normally ConsolidaJed Clay ,
Journal of the Soil Mechanics and Foundation Division ASCE, Vo1.94, 5M2, pp. 353-
369, March 1968
[H4] HARDIN B.O, BLACK W.L, Vibration Modulus of Normally ConsolidaJed Clay,
Closure of Discussion, Journal of Soil Mechanics and Foundation Division, ASCE, pp.
1531-1537, November 1969
[HS] HARDIN B.O, DRNEVICH V.p. Shear Modulus and Damping in Soils: Measurement
and Parameter Effects, Journal of the Soil Mechanics and Foundation Division ASCE,
Vol.98, SM6, pp. 603-623, June 1972
[H6] HARDIN B.O, DRNEVICH V.P, Shear Modulus and Damping in Soils: Design
Equations and Curves, Journal of the Soil Mechanics and Foundation Division ASCE,
Vol.98, SM7, pp. 667-692, July 1972
[H7] HATZIKONSTANTINOU TASSOULAS J.L. GAZETAS G, KOTSANOPOULOS

K, FOTOPOULOU M, Rocking Stiffness ofArbitrarily Shaped Emhedded Foundations,
Journal of the Geotechnical Engineering ASCE, Vol.1l5, No.4, pp. 457-491, April
1989
[H8] HUCKELBRIDGE A.A, CLOUGH R.W, Seismic Response of Uplifting Building

Frame, Journal of the Structural Division ASCE, Vol. 104, STS, pp. 1211-1229, August
1979
[H9] HOUSNER G. W , The Behaviour ofInverted Pendulum Structures During Earthquakes,

Bulletin of Seismological Society of America, VoL53, No.2, pp. 403-417, February
1963
[H1O] HUMPRlES W.K, WAHL H.E, Stress History Effects on Dynamics Modulus of Clay,
389, March 1968
398
[HI 1] HADJIAN A.H, ELLISON B, Equivalent Propertiesfor Layered Media, SDEE VolA,
NoA, pp. 203-209, 1985
[HI2] HARA A, OHTA T, NIWA M, TANAKA S, BANNO T, Shear Modulus and Shear
Strength of Cohesive Soils, Soil and Foundations, Vo1.l4, No.3, pp. 1-11, 1974
[H13] HARA A, KIYOTA Y, Dynamic Shear Tests of Soils for seismic Analysis, Proceeding
of the Ninth International Conference on Soil Mechanics and Foundation Engineering,
Tokyo, pp. 247-250, 1977
[H14] HERRERA I, BIELAK J, Soil-Structure Interaction as a Diffraction Problem,

Proceeding of the Sixth WCEE, New Delhi, pp. 1467-1472, 1977
[HIS] HILBER H.M, HUGHES T.l.R, TAYLOR R.L, Improved Numerical Dissipationfor
Time Integration Algorithms in Structural Dynamics, EESD Vo1.5, pp. 283-292, 1977
[HI6] HUMAR J.L, Design foe Seismic Torsion Forces, Canadian Journal of Civil
Engineering, VoU1, pp. 148-162, 1984
[H17] HADJIAN A.H, LUCO J.E, On the Importance of Layering on the Impedance
Functions, Proceeding of the Sixth WCEE, New Delhi, pp. 1675-1681, 1977
[H18] HANNA A, GHALY A, Effect of Ko and Overconsolidation on Uplift Capacity, Journal

of Geotechnical Engineering ASCE, Vol.11S, pp.1449-1469, 1992
[H19] HABBIBULAH A. ETABS; Three Dimensional Analysis of Building system; User

Manual, Berkeley, California, 1989
[H20] HADJIAN A.H, LUCO J.E, TSAr N.C, Soil-Structure Interaction Continuum or Finite
Elements ?, Nuclear Engineering & Design, VoUl, pp. 151-167, 1974
[H21] HAKUNO M, State of-the Art Report: Seismic Uplift and Sliding of Structures from
Supporting Ground, Proceeding of the Ninth WCEE, Tokyo-Kyoto, Vol. VIII, pp.245-
252, 1988
[H22] HUMAR J.L, Dynamics of Structures, Prentice Hall, New Jersey, 1990
[H23] HILBER H.M , Analysis & Design of Numerical Integration Methods in Structural
Dynamics, Report No. UCB/EERC-76129, Dept. of Civil Eng., University of
California, Berkeley, California, 1976
[H24] HILBER H.M, HUGHES J .R, Collocations, Dissipation an Overshoot for Time
Integration Schemes in Structural Dynamics, EESD, Vol. 6, pp. 99-117, 1978
[H25] HOUSNER G. W, Limit Design of Structures to Resist Earthquakes, Proceeding of First

WCEE, Berkeley, California, pp.5.1-5.13, 1956
[H26] HOUSNER G.W, Earthquake Research Needs for Nuclear Power Plants, Journal of
Power Division ASCE, Vo1.97, POI, pp. 77-91, 1971
[H27] HOUSNER G. W, Measures of Severity of Earthquake Ground Shaking, Proceeding of

US National Conference on Earthquake Engineering, Michigan, pp.25-33, 1975
399
[H28] HOPKINS D.C, CLARK W.D, MATUSCHKA T, SINCLAIR J.C, The Phillipines
Earthquakes of July 1990, Bulletin of the NZSEE, Vol. 24, pp. 3-96, 1991
[H29] HADJIAN A.H, Foundation-Soil Interaction: The State of Confusion, International

Symposium on Soil-Structure Interaction, Roorkee, India, pp. 181-185, 1977
[H30] HOSNI S, HEIDEBRECHT A, Implications of Soil-Structure Interaction Effects on the

NBCC 1990 Base Shear Provisionfor High Rise Buildings, Canadian Journal of Civil
Engineering, Vol. 21, pp. 428-438, 1990
[11] IDRISS LM, DOBRY R, SIGH R.D, Nonlinear Behaviour of Soft Clay During Cyclic
Loading, Journal of the Geotechnical Engineering Division, ASCE, Vol. 104, GT12, pp.
1427-1447, 1978
[12] lMAI T, P and S-Wave Velocities of the Ground in Japan, Proceeding of the Ninth
International Conference on Soil Mechanics and Foundation Engineering, Tokyo, Vo1.2,
pp. 257-260, 1977
[I3] IMAI T, TONOUCHI K, Correlation of N Value with S-Wave Velocity and Shear
Modulus, Proceeding of the Second European Symposium on Penetration Testing,
Amsterdam, pp. 67-72, 1982
[14] IMAI T, YOKOTA K, Relationships between N Value and Dynamic Soil Properties,
Proceeding of the Second European Symposium on Penetration Testing, Amsterdam,
pp. 73-78, 1982
[I5] IMBEAULT F.A, NIELSEN N.N, Effect of Degrading Stiffness on the Response of
Multistorey Frames Subjected to Earthquakes, Proceeding of the Fifth WCEE, Rome,
Vo1.2, pp. 1756-1765, 1974
[16] ISHIHARA K, Evaluation of Soil Propertiesfor Use in Earthquake Response Analysis,

International Symposium on Numerical Models in Geomechanics, Zurich, pp. 237-259,
1982
[17] ISHIHARA K, Factors Effecting Dynamics Properties ofSoils, Proceeding of the Fourth
Asian Regional Conference on Soil Mechanics and Foundation Engineering, Bangkok,
pp.S7-100, 1971
[18] !TOH T, Damped Vibration Mode Superposition Method for Dynamic Response
Analysis, EESD, VoL2, pp. 47-57, 1973
[Jl] JAMES M.L, SMITH G.M, WOLFORD J.C, Applied Numerical Methodfor Digital
Computation with Fortran and CSMP, Harper & Row Publisher, New York, 1977
[J2] JOHNSON G.R, CHRISTIANO P, EPSTEIN H.I, Stiffness Coefficientfor Embedded

Footing, Journal of the Geotechnical Engineering Division ASCE, Vol.101, GTS,
pp. 789-801, August 1975
[13] JUMIKIS A.R, Soil Mechanics, Van Nostrand Company Inc., New York, 1966
[J4J JOHNSON G.R, EPSTEIN H.I, Backfill Effects on Circular Foundation Stiffness,
Joumal of the Geotechnical Engineering Division, VoL 103, GT8, pp. 899-903, 1977
400
[J5] JOHNSON G.R, EPSTEIN H.I, CHRISTIANO Stiffness Coefficients for Layered
Medium, Journal of the Structural Division, Vol.l00, pp. 1537-1542, 1974
[J6] JURY RD, Seismic Load Demands on Columns ofRC Multistorey Frames, ME Thesis
Report, Dep. of Civil engineering, University of Canterbury, 1978
[J7] JENNING P.C, BIELAK J, Dynamics if Building Soil Interaction, Bulletin of the
Seismological Society of America, Vo1.63, No.1, pp. 948, 1973
[J8] llANGUO X, DANMIN W, TIEMING F, A Modified Lumped Parameter Parametric

Modelfor Non-linear Soil-Structure Interaction Analysis, Vol. 12, pp. 273-282,
1993
[Kl] KAUSEL E, ROESSET J.M, Dynamic Stiffness of Circular Foundations, Journal of

the Engineering Mechanics Division ASCE, Vol. 10 1, EM6, pp. 771-785, December
1975
[K2] KIM, NOVAK M, Dynamic Properties of Some Cohesive Soils of Ontario, Canadian
Geotechnical Journal, Vol. 18, pp. 371-389, 1981
[K3] KOMAMURA F, HUANG RJ, New Rheological Modelfor Soil Behaviour, Journal
of the Geotechnical Engineering ASCE, Vol. 100, GT7, pp. 807-823, July 1974
[K4] KAWAKAMI H, TASAKI S, Soil-Structure Interaction ofRigid Structures Considering

Through Soil Coupling between Adjacent Structures, Proceeding of the Ninth WCEE,
Tokyo-Kyoto, VoLIII, pp. 1479-1489, 1988
[K5] KIDDER RL, Reduction of Structural Frequency Equations, AIAA Journal 1972
[K6] KHAN F.R, SBAROUNIS J.A, Interaction of Shear Walls and Frames, Journal of the
Structural Division, ASCE, Vol.90, ST3, pp. 285-335, 1964
[K7] KRINITZCKY E.L, GOULD J.P, EDINGER PH, Fundamentals of Earthquake

Resistant Construction, John Wiley, New York, 1993
[K8] KOKUSHO T, YOSHIDA Y, ESASHI Y, Dynamic Properties of Soft Clay for Wide
Strain Range, Soils and Foundations, Vo1.22, No.4, pp. 1-18, 1982
[K9] KOKUSHO T, Cyclic Triaxial Test ofDynamic Soil Propertiesfor Wide Strain Range,
Soils and Foundations, Vol.20, No.2, pp. 45-60, 1980
[KlO] KOBORI T, MINAI R, KUSAKABE K, Dynamic Cross-Interaction Between Two

Adjacent Foundations, Proceeding of the Sixth WCEE, New Delhi, Vol.II, pp. 1484-
1489, 1979
[K11] KAN C.L, CHOPRA A.K, Effect of Torsional Coupling on Earthquake Forces in
Buildings, Journal of the Structural Division, ASCE, Vol. 103, ST4, pp. 805-819,1977
[KI2] KLINGER R.E, BERTERO V.V, Earthquake Resistant of lrifilled Frames, Journal of
the Structural Engineering Division, ASCE, Vol. 104, ST6, pp. 973-989, 1978
401
[K13] KUNNATH S.K, REIN HORN A.M, PARK Y.J, Analytical Model ofInelastic Response
ofRC Structures, Journal of the Structural Engineering Division, ASCE, VoL 116, No.4,
pp. 996-1018, 1990
[K14] KUNNATH S.K, REIN HORN A.M, lnelllstic Three Dimensional Response Analysis
ofRC Building Frames, National Centre for Earthquake Engineering Research Report
NCEERC-89-0011, 1989
[K15] KAUSELE, TASSOULAS J.L, Transmitting Bountiaries: A Closed Form Comparison,

Bulletin of SeismologiCal Society of America, Vol.71, No.1, pp.143-159, 1981
[K16] KAUSEL E, Local Transmitting Boundaries, Journal of the Engineering Mechanics,

Vo1.114, No.6, pp.1011-1027, 1988
[K17] KAUSEL E. Transmitting Boundaries, Proceeding of Second International Conference

on Recent Advanced in Geotechnical Earthquake Engineering and Soil Dynamics,
Missouri, Vo1.U, pp. 2121, 1991
[K18] KAUSELE, ROES SET J.M, WASS G,Dynamic Analysis ofFooting in Layered Media,
Journal of Engineering Mechanics Division, ASCE, Vol.101, EMS, pp. 679-693, 1975
[K19] KARABALIS D.L, BESKOS D.E, Dynamic Response of Three Dimensional Rigid
SUiface Foundation by Time Domain Boundary Element Metlwd, EESD, Vol.12, pp.
73-93, 1984
[K20] KARABALIS D.L, BESKOS D.E, Dynamic Response of Three Dimensional Flexible
Foundation by Time Domain BEM & FEM, EESD, Vol.4, pp. 91-101, 1985
[K21] KUWAMURA H, GALAMBOS T.V, Earthquake Load for Structural Reliability,

Journal of Structural Engineering Division ASCE, Vol. 115, No.6, pp. 1446-1462, 1989
[K22] KANDA J, INOUE T, Role of Modelling Second Moment Reliability Evaluation for
Seismic Safety ofReactor Buildings, Engineering Structure, Vol. 16, No.7, pp. 507-517,
1994 .
[Ll] LARKIN T.J, Propagation of Seismic Wave Through Nonlinear Media, Ph.D Thesis
Research Report No. 144 Department of Civil Engineering, University of Auckland,
1976
[L2] LEE I.K, Soil Mechanics " Selected Topics, Butterworths, Sydney, pp.504-515, 1968
[L3] LIN M.L, NI S.H, WRIGHT S.G, STOKOE K.H, Characteristic of Material Damping
in Soil, Proceeding of Ninth World Conference on Earthquake Engineering, Tokyo-
Kyoto Japan, pp.IH-5 - III-lO, August 1988
[LA] LUCO J.E, WESTMANN R.A, Dynamic Response of Circular Footing, Journal ofthe
Engineering Mechanics Division ASCE, Vo1.97, EM5, pp. 1381-1395, October 1971
[LS] LYSMER J, RICHART F.E, Dynamics Response of Footing to Vertical Loading,

Journal of the Soil Mechanics & Foundation Division ASCE, Vo1.92, SM1, pp. 65-91,
January 1966
402
[L6] LA Y S.S.P, Statistical Characterization of Strong Motions Using Power Spectral

Density Function, Bulletin of the Seismological Society of America, Vol. 72, No.1, pp.
259-274, 1982
[L7] LARKIN T.J, CHAN S. Y, The Dynamic Response of Volcanic Soils, Sixth Australia-
New Zealand Conference on Geomechanics, Christchurch, pp. 342-347, 1992
[L8] LUCO J .E, Impedance Functionsfor a Rigid Foundation on a Layered Medium, Nuclear
Engineering, Vo1.29, pp. 205-217, December 1974
[L9] LUCO J.E, Vibrations of a Rigid Foundations on a Layered Medium, Nuclear

Engineering, VoL36, pp. 325-340, March 1976
[L1O] LIANG Z, LEE G.C, Damping of Structures, Part I : Theory of Complex Damping.
Report NCEER 91-0004, State University of New York, Buffalo, 1991
[L11] LAMAR S, FORTOUL C, Brick Masonry Effect in Vibration of Frames, Proceeding

of the Fourth WCEE, Chile, VoI.III, pp. 91-98, 1969
[L12] LYSMER J, KUHLEMEYER R.L, Finite Dynamic Modelfor Infinite Media, Journal
of the Engineering Mechanics Division, ASCE, Vol. 95, EM4, pp. 859-877, 1969
[L13] LYSMER J, Analytical Procedures in Soil Dynamics, Report No. UCB/EERC-78/29,

Dept. of Civil Engineering University of California, Berkeley, California, 1978
[L14] LEGER P, DUSSAULT S, Seismic Energy Dissipation in MDOF Structures, Journal

of the StmcturaI Engineering, ASCE, VoLlI8, No.6, pp. 1251-1269, 1991
[L15] LIN J, MAHIN S.A, Effects of Inelastic Behaviour on the Analysis & Design of
Earthquake Resistant Structures, Report No. UBC/EERC-85108, Dept. of Civil
Engineering, University of California, Berkeley, 1985
[Ml] MAJOR A, Dynamics in Civil Engineering: Fundamental Soil Dynamics,

lmtrumentation and Tolerances, Akademia Kiado Bucharest, 1980
[M2] MEAD T.L ,Dynamic Soil Model, M.E Thesis Report. School of Engineering, Dept.
of Civil Engineering, University of Auckland, 1979
[M3] MEEK J.W, Effect on Foundation TIpping on Dynamic RefJponse , Journal of the
Structural Division ASCE, Vol.1Ol, ST7, pp. 1297-1311, 1975
[M4] MEEK J.W, Dynamic Response of TIpping Core Building, EESD, pp. 437-454, 1978
[M5] MEEK J.W, VELETSOS A.S, Simple Models for Foundations in Lateral & Rocking
Motion, Proceeding Fifth WCEE, Rome, pp. 2610-2613, 1974
[M6] MITCHELL J.K, SEED H.B, PADUANA J, The Creep Deformation and Strength
Characteristics of Soil under the Action of Sustain Stress , Report No. TE 65-8,
Department of Civil Engineering, University of California, Berkeley, 1965
403
[M7] MACKY T.A, SAADA A.S, Dynamic ofAnisotropic Clays Under Large Strain, Journal
of the Geotechnical Engineering, ASCE, VoL 110, No.4, pp. 487~504, 1984
[M8] MAHIN S.A, BERTERO V.V, An Evaluation of Inelastic Seismic Design Spectra,
Journal of the Structural Division, ASCE, VoLl07, S1'9, pp. 1771-1795, 1981
[M9] MATASOVIC N, VUCETIC M, A Pore Pressure Modelfor Cyclic Straining of Clay,

Soils and Foundations, Vo1.32, No.3, pp. 156-173, 1992
[MlO] MAU S.T, Structure Identification Using Classical and Nonclassical Nonnal Modes,
Proceeding of the Eighth WCEE, San Francisco, Vol. VIII, pp. 403-410, 1984
[Mll] MAYNE P.W, KULHAWY F.H, Ko-OCR Relationships in Soils, Journal of the
Geotechnical Engineering Division, ASCE, Vol. 108, GT6, pp. 851-872, 1982
[M12] McGREGOR J.G, MAJUMDAR S.N.G, NIKHED R.P, ADAMS P.P, Approximate
Inelastic of Shear Wall-Frame Structures, Journal of the Structural Engineering
Division, A5CE, Vol.98, 5T11, pp. 2351-2361, 1972
[M13] MIDORIKAWAS, Site Effects on Strong Motion records on the 1985 Chile
Earthquake and their Nonlinear Behaviour, Proceeding of the Tenth WCEE, Madrid,
pp. 1031-1037, 1992
[MI4] MUTO K, Aseismic Design and Analysis of Buildings, Maruzen Company Ltd, Tokyo,
1m '
[MI5] MEYER C, ROUFAEIL M.S, ARZOUMANIDIS S.G, Analysis of Damage Concrete
Framesfor Cyclic Loads, EESD, VoLl1, pp. 207-228, 1983
[M16] MESKO URIS K, KRATZIG W.B, Seismic Damage Assessment of Buildings,

Proceeding of the International Conference on Earthquake Resistant Construction and
Design, Berlin, pp. 427-441, June 1989
[M17] MITA A, LUCO J.E, Impedance Functions and Input Motions for Embedded Square
Foundations, Journal of the Geotechnical Engineering ASCE, Vol. 115, No.4, pp. 491-
503, 1989
[MI8] MITA A, LUCO J .E, Dynamic Response of a Square Foundation Embedded in a Elastic
Half-Space, 5DEE, Vol.8, No.2, pp. 54-67, 1989
[M19] MEYERHOF G.G, Safety Factor in Soil Mechanics, Canadian Geotechnical Journal,
Vol. 7, pp. 350-355, 1970
[M20] MEYER C, Inelastic Dynamic Analysis of Tall Buildings, EESD, Vo1.2, pp.325-342,
1974
[M21] MOSS P.J, CARR A.J, Aspects of the Analysis of Frame-Panel Interaction, Bulletin of
NZSEE, Vol. 1, No.1, pp. 124-144, 1974
[M22] MOSS P.J, CARR A.J, The Effects of Large Displacement on the Earthquake Response
of Tall Frame Structures, Bulletin of the NZSEE, Vol. 13, No.4, pp. 317-328, 1980
404
[M23] McCABE S.L, HALL W.J, Assessment of Seismic Structural Damage, Journal of the
Structural Division, ASCE, VoLlI5, No.9, pp. 2166-2183, 1989
[M24] McCABE S.L, Damage of Reserve Capacity Evaluation of Structures Subjected to

Strong Earthquake Ground Motions, Proceeding of the Tenth WCEE, Madrid, Spain,
pp.3653-3658, 1992
[M25] MAHIN S, BERTERO V. V, An Evaluation of Some Methods for Predicting Seismic

Behaviour of RC Buildings, Report No. UCB/EERC-7515, Dept. of Civil Engineering
University of California, Berkeley, California, 1975
[M26] MAHIN S, BERTERO V. V, Non-linear Seismic Response of Coupled Wall System,

Journal of the Structural Engineering Division, ASCE, Vo1.102, ST9, pp.1759-1780,
1976
[M27] MURAKAMI H, SHIOYA S, YAMADA R, LUCO J.E, Transmitting Boundariesfor

the Time Harmonic Elastodynamic on Infinite Domain, International Journal for
Numerical Methods in Engineering, VoU7, pp.1697-1716, 1981
[M28] HAHIN S, LIN J, Construction of Inelastic Response SpectraforSDOF Systems, Report

No. UCB/EERC-83/17, Dept. of Civil Eng., University of California, Berkeley, 1983
[M29] MONTGOMERY C.J, Influence of P-Delta EjJects on Seismic Codes, Canadian Journal
of Civil Engineering, VoLB, pp. 31-43, 1981
[M30] MORAN D, FERVER G, STRAITA J, WYLLIE J r, Engineering Aspects of the Lima,

Peru Earthquake of 3 October 1974, EERI, 1975
[M31] MOINFAR A.A, NADERZADEH A, The Manjil, Iran Earthquake of 20 June 1990,
Bulletin of the NZSEE, Vol. 23, No.4, pp. 254-283, 1990
[M32] MEYER C, An Energy-Based Damage Model for Inelastic Dynamic Analysis of

Reinforced Concrete Frames, Elsevier Applied Science, 1992, pp. 115-124
[NI] NOVAK M, Prediction of Footing Foundations, Journal of the Soil Mechanics and
Foundation Division ASCE, Vo1.96, SM3, pp. 837-861, 1970
[N2] NOVAK M, BEREDUGO Y.O, Vertical Vibration of Embedded Footing, Journal of

the Soil Mechanics and Foundation Division ASCE, Vo1.98, SMI2, pp. 1291-1310,
December 1972
[N3] NOVAK M, Dynamics Stiffness and Damping of Piles, Journal of the Canadian
Geotechnical Eng., No.ll, pp. 574-598, 1974
[N4] NOVAK M, EL-HIFNAWY I, Effect of Foundation Flexibility on Dynamic Behaviour

of Buildings, Proceeding of the Eighth WCEE, San Francisco, pp. 721-728, 1984
[N5] NAKAI S, FUKUWO N, HASEGAWA M, Approximate Three-Dimensional Analysis

of Embedded Structures, Proceeding of the Eighth WCEE, San Francisco, pp. 689-696,
1984
405
[N6] NEWMARK N.M, A Method of Computation for Structural Dynamics, Journal of the
Engineering Mechanics Division, ASCE, Vo1.85, EM3, pp. 67-94, 1959
[N7] NOV AK M, HIFNAWY L.E, Effect of Soil-Structure Interaction on Damping of

Structures, EESD, VoLlI, pp. 595-621, 1983
[N8] NOV AK N, Effect of Soil on Structural Response to Wind and Earthquakes, EESD
Vo1.3, pp. 79-96, 1974
[N9] NOVAK N, Foundation and Soil-Structure Interaction, Proceeding of the Sixth WCEE,
New Delhi, Vol.II, pp. 1421-1437, 1977
[01] o 'HALLORAN M, The Earthquake Stability ofEarth Structure, Ph.D Thesis Research
Report No.414 Department of Civil Engineering, University of Auckland, 1986
[02] OHSAK! Y, rwASAKI R, On Dynamic Shear Moduli and Poisson Ratios of Soil
Deposits, Soils and Foundations Vo1.13, No.4, pp. 61-73, 1973
[03] OSAWA Y, KITAGAWA Y, IRIE Y, Evaluation of Various Parameters on Response

Analysis of Earthquake Motions Including Soil Building System, Proceeding of the
WCEE, New Delhi, pp. 1461-1446, 1977
[04] O'ROURKE M.J, CASTRO G, HOSSAIN I, Horizontal Shear Strain Due To Seismic
Waves, Journal of the Geotechnical Engineering, Vol. 110, No.9, pp. 1173-1187, 1984
[05] OTANI S, Non-linear Dynamic Analysis of RC Building Structures, Canadian Journal

of Civil Engineering, Vol.7, pp. 333-344, 1980
[06] OTANI S, Hysteresis Models of RC for Earthquake Response Analysis, Journal of the
Faculty of Engineering The University of Tokyo, Vol.36, No.2, pp. 407-441, 1981
[07] OTANI S, Inelastic Analysis ofRC Frame Structures, Journal of the Structural Division,
ASCE, Vol.100, ST7, pp. 1433-1449, 1974
[08] O'REILLY M.P, BROWN S.P, Cyclic Loading of Soils; from Theory to Design, Van
Nostrand Reinhold, New York, 1991
[09] OHTA Y, GOTO N, Empirical Shear Wave Velocity Equations in tenn of

Characteristi c Soil Indexes , Earthquake Engineering & Structural Dynamics, Vol. 6, pp.
167-187, 1978
[010] OTANI S, Structural Dynamic Notes, Dept. of Civil Engineering University of

Canterbury, 1994 ( Unpublished).
[011] OTANI S, CHEUNG V.W.T, LAI S.S, Behaviour and Analytical Models ofReinforced
Concrete Columns Under Biaxial Earthquake Loads, Proceeding of Third Canadian
Conference on Earthquake Engineering, VoUI, pp. 1141-1167, 1979
[012] OTANI S, Hysteretic Models of Reinforced Concretefor Earthquake Response Analysis,

Proceeding of the Eighth WCEE, San Francisco, VoI.IV. pp. 551-558, 1984
406
[PI] PARMELEE R.A, Building Foundation Interaction Effects, Journal of the Engineering
Mechanics Division ASCE, Vo1.93, EM2, pp. 131-153, April 1967
[P2] PAZ M , Structural Dynamics,' Theory and Computation, Van Nostrand Reinhold
Company, New York, 1991
[P3] PRAKASH S , Soil Dynamics, Mc Graw-Hill Book Company, New York, 1981
[P4] PRIESTLEY M.J.N, EVISON R.J, CARR A.J ,Seismic Response ojStructures Free
to Rock on their Foundation, Bulletin of the NZSEE, Vol. 11 ,No.3, pp. 141-150, 1978
[P5] PSYCHARIS LN, JENNING P.C, Rocking oj Slender Rigid Bodies Allowed to Uplift,
EESD, VoU1, pp. 57-76, 1983
[P6] PSYCHARIS LN, Dynamic oj Flexible System with Partial Uplift, EESD, Vol. 11 , pp.
501-521, 1983
[P7] PSYCHARIS LN, Dynamic Behaviour oj Rocking Structures Allowed to Uplift, Ph.D
Thesis Research Report No. EERL 81-02, California Institute of Technology, Pasadena
California, 1982
[P8] PAIS A, KAUSEL E, Approximate Formulas jor Dynamic Stiffness oj Rigid

Foundations, SDEE, Vol.7, No.4, pp. 213-227, 1988
[P9] PARK R, State the Art Report: Ductility Evaluation from Laboratory and Analytical
Testing, Proceeding of the Tenth WCEE, Tokyo-Kyoto, Vol.VIII, pp. 606-616, 1988
[P10] PAULAY T, SANTHAKUMAR A.R, Ductile Behaviour oj Coupled Shear Walls,

Journal of Structural Division, ASCE, Vol. 102, STl, pp. 93-108, 1976
[PIll PAULAYT, GOODSIRW.J, TheDuctilityojStructural Walls, NZSEE, Vol. 18, No.3,

pp. 250-269, 1985
[P12] PAULAY T, GOODSIR W.J, The Capacity Design oj Reinforced Concrete Hybrid
Structures jor Multistorey Buildings, Bulletin ofthe NZSEE, Vol. 19 , No.1, pp. 1-17,
1986
[P13] PAULAY T, Seismic Design in RC: The State ojArt in New Zealo.nd, NZSEE, Vol.21,
No.3, pp. 208-232, 1988
[P14] PAULAYT, An Application oj Capacity Design Philosophy to Gravity Load Dominated

Ductile RC Frames, Bulletin of the NZSEE, Vol.lI, No.1, pp. 50-61, 1984
[PI5] PAULAY T, Seismic DeSign in RC : The State ojthe Art in New Zealand, Proceeding
the Ninth WCEE, Tokyo-Kyoto, Vol. VIII , pp. 687-692, 1988
[P16] PAULAY T, Moment Redistribution in Continuous Beams oj Earthquake Resistant

Multistorey RC Frames, Bulletin of the NZSEE, Vol. 10, No.2, pp. 85-94, 1977
[Pl7] PAULAYT, Column Evaluation ojActions, Bulletin of the NZSEE, Vol. 10, No.2, pp.
85-94, 1977
407
[PIS] PAULAY T, PRIESTLEY M.J.N, Seismic Design of Reinforced Concrete & Masonry
Buildings, John Wiley & Sons, Inc., 1992
[PI9] PENDER M.J, Earthquake Soil-Structure Interaction, Spring, and Dashpot Models and
Real Soil Behaviour, Bulletin of the NZSEE, Vol. 16, No.4, pp. 320-330, 1983
[P20] PENDER M.J, DUSKE G.C, PEPLOE R.J, Cyclic Undrained Stiffness of Stiff Clay
and Volcanic Ash, Sixth Australian-New Zealand Conference on Geomechanics,
Christchurch,pp. 363-363, 1992
[P21] PENZIEN J, BERTERO V.V, ATALAY B, Inelastic Cyclic BehaviourofRC Flexural

Members, Proceeding of the Review Meeting US-Japan Corporative Research in
Ealthquake Engineering, pp. 52-74, 1975
[P22] PRIESTLEY M.J.N, PARK R, POTANGAROA R.T, Ductility of Spirally-Confined

Concrete Columns, Journal of the Structural Division, ASCE, Vol. 107, STl, pp.181-
202, 1981
[P23] PRAKASH S, PURl V.K, Foundation for Machines: Analysis and Design, John Wiley
& Sons, New York, 1988
[PM] PRIESTLEY M.J.N, CALVI G.M, Toward a Capacity Design Assessment Procedure
for Reinforced Concrete Frames, Technical Report No.2, New Zealand Concrete
Society, 1992
[P25] PENZIEN J, Dynamic Response of Elasto Plastic Frames, Journal of the Structural
Division ASCE, Vol.86, S1'7, pp. 81-94, 1960
[P26] PENZIEN J, Elasto-Plastic Response of Idealized Multistorey Structures Subjected to

a Strong Motion Earthquakes, Proceeding of Second WeEB, Tokyo, Vol.lI, pp.739-
760, 1960
[P27] PARK Y.J, ANG A.H.S, Mechanistic Seismic Damage Model for RC, Journal of
Structural Engineering Division ASCE, Vol.11l, No.4, pp. 722-739, 1985
[P28] PARK Y.J, ANG A.H.S, WEN Y.K, Seismic Damage Analytic of RC Buildings,
Journal ofthe Structural Engineering Division ASCE, Vol.U1, No.4, pp.740-757, 1985
[P29] PARK Y.J, REINHORN A.M, KUNNATH S.K, !DARC: Inelastic Damage Analysis
of Reinforced Concrete Frame-Shear Wall Structures, Technical Report NCEER 87-
0008, University of New York at Buffalo, 1987
[P30] PARKR,PAULAY T, Joints in Reinforced Concrete Frames Designedfor Earthquake

Resistance, Research Report No.84-9, Dept. of Civil Eng., University of Canterbury
New Zealand, 1987
[P31] PRIESTLEY M.J.N, PARK R. Strength and Ductility of Concrete Bridge Columns
Under Seismic Loadings, ACI Structural Journal, pp. 61-79, Jan.-Feb. 1987
[P32] PARK R, PAULAY T, Reinforced Concrete Structure, John Wiley & Sons, 1975
408
[P33] PAULAY T, The Design of Reinforced Concrete Ductile Shear Walls for Earthquake
Resistance, Research Report No.81-1, Dept. of Civil Eng., University of Canterbury
NZ, 1981
[P34] PARMELEE R.A, PEREMAN D.S, LEE S.L, Seismic Response of Multistorey
Structures on Flexible Foundations, Bulletin of the Seismological Society of America,
Vo1.59, No.3, pp. 1061-1070, 1969
[P35] PRIESTLEY M.1.N, Myths and Fallacies in earthquake Engineering: Conflicts between
Design and Analysis, Proceeding of the TOM PAULAY SYMPOSIUM: Recent
Development in lateral Force Transfer in Buildings, ACr, EERC, UCSD, California,
September 1993
[P36] POWELL G, ALLAHABADI R, Seismic Damage Prediction by Detenninistic Methods;

Concept and Procedures, EESD Vol. 16, pp. 719-734, 1988
[P37] PARK R, PRIESTLEY M.l.N, GILL W.D, Ductility of Square Coriflned Concrete
Columns, Journal of the Structural Division ASCE, Vol. 108, pp.929-950, 1982
[P38] PAULAY T, The Design of Ductile Reinforced Concrete Structural Walls for
Eanhquake Resistance, Earthquake Spectra Vo!.2, pp.783-828, 1986
[P39] PAULAY T, UZUMERI S.M, A Critical Review of The seismic Design Provision for
Ductile Shear Walls of the Canadian Code and Commentary, Canadian Journal of Civil
Engineering, No.2, pp.592-601, 1975
[R1] RAY R.P, WOODS R.D, Modulus and Damping due to Uniform and Variable Cyclic
Loading, Journal ofthe Geotechnical Engineering ASCE, Vo1.114, No.8, pp. 861-871,
August 1988
[R2] RICHART F.E, WHITMAN R. V, Comparison of Footing Vibration Test with Theory,
193, Janua'ry 1967
[R3] RAMEY G.E, YOO C.H, MOORE R.K, BUSH T.D, STALLING J.M, Lumped
Parameter Modelling and Dynamic Response Evaluation of Soil-Structure Systems,
Proceeding of the Eighth WCEE, San Francisco, pp. 977-983, 1984
[R4] ROUFAIEL M.S.L, MEYER C, Analytical Modeling of Hysteretic Behaviour of RC

Frames, Journal of Structural Engineering ASCE, Vol 113, No 3, pp. 777-784, March
1987
[R5] RICHART Jr F.E, HALL Jr, WOODS R.D, Vibration of Soils & Foundations,
Prentice Hall Inc., New Jersey, 1970
[R6] ROESSET J.M, WHITMAN R.V, DOBRY R, Modal Analysis for Structure with
Foundation Interaction, Journal of the Structural Division, ASCE, ST3, pp. 399-416,
1972
[R7] ROESSET J.M, TASSOULASJ.L,Dynamic Interaction between A djacent Foundations,

EESD, Vo1.15, pp. 323-343, 1987
409
[RS] ROSENBLUETH E, HOLTZ I, Elastic Analysis of Shear walls in Tall Buildings, ACI
Journal, No.12, pp. 1209-1222, June 1960
[R9] RICHART J.E, Soil Structure Interactions, Proceeding of the Ninth ICSMFE, Tokyo,
Vol.2, pp.637-641, 1977
[RIO] ROESSET J, Soil Structure Interaction : Theoretical Aspects, Proceeding of Second

International Conference on Recent Advance in Geotechnical Earthquake engineering
& Soil Dynamics, Missouri, VoLIII, pp. 2113-2114, 1991
[R11] ROESSET J, ETIOUNEY M.M, Transmitting Boundaries: A Comparison, International

Journal for Numerical & Analytical Methods in Geomechanics, VoL 1, pp. 151-176,
1977
[RI2] ROMANUEL C, KUNDU T, A Hybrid Modelling of Soil Structure Interaction

Problems for Deeply Embedded Structures in a Multi-layered Media, EESD, Vo1.22,
pp. 557-571, 1993
[S1] SEED H.B, LYSMER J, HWANG R, Soil Structures Interaction Analysesfor Seismic
Response, Journal of the Geotechnical Engineering Division ASCE, Vol.101, GT5, pp.
439-457, May 1975
[S2] SEED H.B, Landslides During Earthquake Due To Liquefaction, Journal of the Soil
Mechanics & Foundation Division ASCE, Vol. 94, SM5, pp. 1055-1121, September
1968
[S3] SEED H.B, IDRISS LM, Influence of Soil Condition on Ground Motions During
Earthquake, Journal of the Soil Mechanics & Foundation Division ASCE, Vo1.95, SM1,
pp. 99-137, January 1969
[S4] SEED H.B, MURTIN G.R, The Seismic Coefficient in Earth Dam Design, Journal of
the Soil Mechanics & Foundation Division ASCE Vo1.92, SM3, pp. 25-58, May 1966
[S5] SEED H.B, WONG R. T, IDruSS I.M, TOKIMATSU K, Moduli and Damping Factors
for Dynamic Analyses of Cohesion less Soils, Report no. UCB/EERC 84-14, College of
Engineering University of California Berkeley, September 1984
[56] SHERIF M.A, ISHIBASHI I, Dynamic Shear Moduli for Dry Sand, Journal of the
Geotechnical Engineering Division ASCE, Vol.l02, GTll, pp. 1171-1181, 1976
[S7] SHAMBU P.D, KAMESWARA RAO N.S.V, Dynamic of Rectangular Footing by

Finite Elements, Journal of the Geotechnical Engineering ASCE, Vol. 104, GT5, pp.
621-637, May 1978
[S8] SINGH A, MITCHEL J.K, General Stress Strain-Time Function for Soils, Journal of
the Soil Mechanic & Foundation Engineering ASCE Vo1.94, SM1, pp. 21-46, January
1968
[S9] SOLOMON C.S-YIM, CHOPRA A.K, Dynamics of Structures on 1Wo Spring

Foundation Allow to Uplift, Journal of the Engineering Mechanics ASCE, Vol.llO,
No.7, pp' 1124-1147, July 1984
410
[510] SOLOMON C.S-YIM, CHOPRA A.K ,Simplified Earthquake Analysis of Multisto rey
Structures with Foundation Uplift , Journal of the Structural Engineering ASCE,
Vol.111, No.12, pp. 2708-2731, December 1985
[Sl1] STOKOEK.H, RICHARTF.E, Dynamic Response of Embedded Machine Foundations,

Journal of the Geotechnical Engineering Division ASCE, Vol.l00, GT4, pp. 427-447,
April 1974
[512] SUKUE L , Rheological Aspects of Soil Mechanics , Wiley-International Science,

London, 1969 .
[513] SUN 1.1, GOLESARKHI R, SEED H.B, Dynamic Moduli and Damping Factors for
Cohesive Soils, Report No. UCB/EERC-88/15, College of Engineering University of
California Berkeley, August 1988
[514] SAJIDI M, Hysteresis Models for Reinforced Concrete, Journal of the Structural
Division, ASCE, Vol. 108, ST5, pp. 1077-1086, 1982
[SI5] SAJIDI M, SOZEN M.A, Simple and Complex Models for Nonlinear Seismic
Response, of Reinforced Concrete, University of Illinois Research Report, 1979
[SI6] SHINOZUKA M, KAMEDA H, KOIKE T, Ground Strain Estimationfor Seismic Risk

Analysis, Journal of the Engineering Mechanics, ASCE, Vol.l09, No.1, pp. 175-191,
1983
[S 17] SINGH M. P, Seismic Response by SRSS for Nonproportional Damping, Journal of the
Engineering Mechanics, ASCE, Vo1.106, EM6, pp. 1405-1419, 1980
[518] SOTIROV P, Damping of a Soil Structure System, Proceeding of the Seventh WCEE,
Istambul, Vo1.5, pp. 97-104, 1980
[SI9] SRIDARAN A, GANDHI N.S.V.V.S.J, SURESH 5, Stiffness Coefficients of Layered

Soil Systems, Journal of the Geotechnical Engineering, Vo1.l16, No.4, pp. 604-624,
1990
[S20J SING M.P, GHAFORY-ASHTIANY M, Modal Time History Analysis of Non

Classically Damped Structures for Seismic Actions, EESD, Vo1.l4, pp. 133-146, 1986
[S21] SHAMES H, Introduction to Solid Mechanics, Prentice Hall, 1989
[S22] SOGIMURA Y, Seismic Strain Induced in the Ground During Earthquakes, Research
Report No: UBC/EERC 77/14, Dept. of Civil Engineering University of California at
Berkeley, 1977
[S23] SAATCIOGHU M, DERECHO A.T, CARLEY W.G, Modelling Hysteretic Behaviour

of Coupled Walls for Dynamic Analysis, EESD, Vol. 11, pp. 711-726, 1983
[S24] SCOTT R.F, Foundation Analysis, Prentice Hall Inc, New York, 1981
[525] STUART J.G, Interference Between Foundations with Special Reference to Surface
Footings in Sand, Geotechnique Vol. VII, pp. 15-23, 1962
411
[S26] SIEFFERT J.G, CEVAER F, Handbook of Impedance Functions, OvestEditions, 1991
[S27] SCHEULLER W, High Rise Building Structures, John Wiley & Sons, New York, 1977
[S28] STRATTA J.L, FELDMAN J, Interaction of Infill Walls and Concrete Frames During
Earthquakes, Bulletin of the Seismological Society of America, VoL61, No.3, pp. 602-
612, 1971
[S29] SAIlDI M, Influence of Hysteretic Models on Calculated Seismic Response of RC

Frames, Proceeding of Seventh WeEE, Istambul Turkey, VoL5, pp.423-430, 1980
[S30] STEPENS J.E, YAO J.P.T, Damage Assessment Using Response Measurement, Journal
of the Structural engineering Division, VoLI13, No.87, pp.787-801, 1987
[531] SEED RB, DICKENSON S.E, MOK C.M, Site Effects on Strong Shaking and Seismic
Risk: Recent Development and Their Impact on Seismic Design Code and Practice,
Structural Conggres XII ASCE, Vol. I, pp. 573-578, April 1994
[532] SPYRAKOS C.C, BESKOS D.E, Dynamic Response of Flexible Strip Foundations by
Boundary Elements and Finite Elements, SDEE, Vot5, pp.84-96, 1986
[S33] SEED H.B, UGAS C, LYSMER J, Site Dependent Spectrafor Earthquake Resistant
Design, Bulletin of the Seismological Society of America, Vo1.66, No.1, pp.221-243,
1976
[534] STRATTA J.L, Wyllie L.A, Reconnaissance Report, Friuli, Italy Earthquake 1976,
EERI, 1979
[535] STRATTA, J.L, ASCALANTE L.E, KRINITZSKY E.L, MORELLI U,

Reconnaissance Report: Earthquake in Campania-Bassilicata, Italy November 23 1980,
EERI & NR Council, 1981
[S36] SIVARAKUMARANK.S, BALENDRA T, Seismic Ana/ysis ofAsymmetric Multistorey

Buildings Including Foundation Interaction and P-DeltaEffects, Engineering structures,
Vol.16, pp.609-624, 1994
[Tl] TANIGUCHI E, OGASAW ARA H, Effects of Shear Strain on the Dynamic

Deformation Coefficients of Clays, Proceeding of the Fifth Japan Earthquake
Engineering Symposium, pp. 705-712, November 1978
[TI] TAWFIQ K.S, AGGOUR M.S, AL-SANAD H.A, Dynamic Properties of Cohesive
Soilsfrom Impulse Testing, Proceeding of the Ninth WCEE, Tokyo-Kyoto, Vol.lII, pp.
11-16, August 1988
[T3] TAYLOR P.W, The Properties of Soils Under Dynamic Stress Condition, with
Applications to Design of Foundations in Seismic Areas, Ph.D Thesis, Department of
Civil Engineering, University of Auckland, 1971
[T4] TERZAGHI K, PECK R.B, Soil Mechanics in Engineering Practice, Jolm Wiley &
Sons Inc, New York, 1967
412
[T5] THOMSON W. T , Theory of Vibration with Applications, George Allen & Unwin,
London, 1981
[T6] TRUITT M.M, Soil Mechanics & Technology, Prentice Hall Inc, 1983.
[TI] TUMA 1.J, ABDEL-HADY M, Engineering Soil Mechanics , Prentice Hall inc,
Englewood Cliff, New Jersey, pp. 47-71, 1973
[T3] TOKIDA K, AIZA W A K, TAMURA K, Attenuation Characteristics of Ground Strains

Induced During Earthquakes, Proceeding of Second International Conference on Recent
Advances in Geotechnical Engineering & Soil Dynamics, St.Louis, pp. 1221-1227, 1991
[T9] TOKIDA K, AIZAWA K, TAMURA K ,Attenuation of Ground Strains During

Earthquake Base on Dense Array, Eighth Japan Earthquake Engineering Symposium,
Tokyo, pp. 313-318, 1990
[TlO] TRAILL-NASH R.W, Modal Methods in the Dynamics of Systems with Non Classical
Damping, EESD, Vo1.9, pp. 153-169, 1981
[Tll] TSAr N.C, Modal Damping for Soil Structure Interaction, Journal of the Engineering
Mechanics Division, ASCE, Vo1.100, EM2, pr. 323-341, 1974
[TI2] TSO W.K, ZHU T.J, HEIDERBRECHT A.C, Seismic Energy Demand on RC Moment
Resistant Frames, EESD, Vol. 22, pp. 533-545, 1993
[TI3] TRIERS G.R, SEED H.B, Cyclic Stress Strain Characteristic of Clay, Journal of the
Soil Mechanics and Foundation Division, SM2, Vo1.94, pp. 555-569, 1968
[TI4] TIMOSHENKO S, GOODIER1.N, Theory of Elasticity , McGraw Hill Book Company,

New York 1951
[TI5] TSO W.K, DEMPSEY K.M, Seismic Torsional Provisions for Dynamic Eccentricity,
EESD, Vol.8, pp.275-289, 1980
[TI6] TSO W.K, MENG V, Torsional Provisions in Building Codes, Canadian Journal of
Civil Engineering, Vol.9, pp.38-46, 1932
[TI7] TAKEDA T, SOZEN M.A, NIELSEN N.N, Reinforced Concrete Response to

Simulated Earthquakes, Journal of the Structural Engineering Division ASCE, Vol.96,
ST12, pp.2557-2573, 1970
[TI8] TARANATH B.S, Structural Analysis and Design of Tall Buildings, McGraw Hill Book
Company, New York, 1988
[TI9] TSAI N.C, NIEHOFF D, SWATTA M, HADJIAN A.H, The Used of Frequency
Independent Soil Structure Interaction Parameters, Nuclear Engineering & Design,
VoL31, pp. 168-183, 1976
[T20] TSO W.K, ZHU T.I, HEIDEBRECHT A.C, Engineering Application of Ground motion
A/V Ratio, SDEE 11, pp. 133-144, 1992
413
[T21] THOMSON E.D, CARR A.J, MOSS P.J, P-Delta Effects in the Seismic Response of
Ductile RC Frames, Proceeding of the Second Pacific Conference on Earthquake
Engineering, VoU, pp.49-60, 1991
[Ul] DANG C.M, BERTERO V.V, Evaluation of Seismic Energy in Structures, EESD,
Vol. 19, pp. 77-90, 1990
[V1] VAN KOTEN H, Structural Damping , Heron Journal Department of Civil

Engineering of the Delft University of Technology, VoL22, pp. 5-69, 1977
[V2] VELETSOS A.S, VERBIC B, Vibration of Viscoelastic Foundations, EESD, Vol.2,

PI'. 87-102, 1973
[V3] VELETSOS A.S, VERBIC B, Basic Response FWlctionsfor Elastic Foundation, Journal
of the Engineering Mechanics Division ASCE, Vol. 100, EM2, pp. 189-202, April 1974
[V4] VELETSOS A.S, WEI Y.T, Lateral and Rocking Vibration of Footings, Journal of the
Soil Mechanics & Foundation Division ASCE, VoL97, SM9, pp. 1227-1248, September
1971
[V5] VUCETIC M, Soil Properties and Seismic Response on Earthquake Engineering,

Proceeding of Tenth WCEE, Madrid, pp. 1199-1204, 1992
[V6] VUCETIC M, DOBRY R, Effect of Soil Plasticity on Cyclic Response, Journal of the
Geotechnical Engineering, ASCE, Vol. 117, No.1, pp. 89-107, 1991
[V7] VELETSOS A.S, VENTURA C.E, Modal Analysis of Non Classically Damped Linear
Systems, EESD, Vol 14, pp. 217-243, 1986
[V8] VESIC A.B, Bending qf beams Resting on Isotropic Elastic Solid, Journal of the
Engineering Mechanics Division ASCE, Vol.87, EM2, pp. 35-53, 1961
[V9] VELETSOS A.S, NEWMARK N .M, Effect of Inelastic Behaviour on the Response of
Simple Systems to Earthquake Motions, Proceeding of Second WCEE, Tokyo, Vol.U,
pp. 895-912, 1960
[VIO] VELETSOS A.S, Dynamic of Soil Structure Interaction Revisited, Fourth US National
Conference on Earthquake Engineering, California, VoU, pp.85-88, 1990
[Vll] VELETSOS A.S, TANG Y, Detenninistic Assessment of Effects of Ground Motion

Incoherence, Journal of Engineering Mechanics, Vol. 116, No.5, pp.l109-1124, 1990
[VI2] VELETSOS A.S, PRASAD A.M, Seismic Interaction of Structures and Soils,' Stochastic
Approach, Journal of the Structural Engineering, ASCE, No.4, pp. 935-952, 1989
[Wl] WEAVER W.J.R, GERE J.M, Matrix Analysis of Framed Structures, Van Nostrand
Reinhold, New York, 1980
[W2] WHITMAN R.V, RICHART F.E, Design Procedures for Dynamically Loaded
FOWldations , Journal of the Soil Mechanics and Foundation Division ASCE, Vo1.93,
SM3, pp. 169-193, November 1967
414
[W4] WHITMAN R.V, RICHART F.E, Design Procedures for DYfUlmically Loaded
Foundations, Journal of the Soil Mechanics & Foundation Division ASCE, Vo1.93,
SM6, pp. 169-193, November 1967
[W5] WHITMAN R.V, HOLT R.J, MURPHY V.J, DOBRY R, POBLETE M,THIERS
G.R ,Discussion of the Vibration of Nonnally Consolidated Clay (by. Hardin B.O,
Black W.L), Journal of the Geotechnical Engineering and Foundation Division ASCE,
pp. 656-662, March 1969
[W6] WHITMAN R.V, PROTONOTARIOS J.N, Case Study of Dynamics Soil Structure
Interactions, Journal of the Soil Mechanics and Foundation Division ASCE, Vol.99,
SMll, pp. 997-1009, November 1973
[W7] WILSON E.L, PENZIEN J, Evaluation of OrthogofUlI Damping Matrices ,

International Journal of Numerical Methods in Engineering, VolA, pp. 5-10, 1972
[W8] WINTERKORN H.F, HSAI YANG FANG, Foundation Engineering Handbook, Van
Nostrand Reinhold Company, New York, 1975
[W9] WOLF J.P, SOMAINI , Approximate Dynamic Model of Embedded Fowuiation in

Time Domain, EESD, Vol. 13, pp. 683-703, 1986
[W1O] WOLF J.P, Soil Structure Interaction AfUllysis in Time Domain , Prentice Hall
Englewood Cliffs, New Jersey, 1988
[WIl] WU T.H, Soil Mechanics, Allyn and Bacon Inc, Boston 1976, pp.85-103
[W12] WARBURTON G.B, A Simple Model to Illustrate the Significance of Foundation

Flexibility in Soil-Structure Interaction Problems, Proceeding ofthe Eighth WCEE, San
Francisco, Vol.VIII, pp. 729-736, 1984
[W13] WARBURTON G.B, SONI S.R, Errors in Response Calculations for Non Classically
Damped Structures, EESD, Vol.5, pp. 365-376, 1977
[W14] WESTERMO B.D, WONG H.L, On the Fundamental Differences of Three Basic Soil-
Structure Interaction Models, Proceeding of the WCEE. New Delhi, pp. 1455-1460,
1977
[W15J WILSON E.L, FARHOOMAD I, BATHE K.J, Non-linear DYfUlmic AfUllysis of

Complex Structures, EESD, Vo1.1, pp. 241-252, 1973
[WI6] WEST I.M, STUART J.G. Oblique Loading Resulting from Interference Between
Surface Footings on Sand, Proceeding ofthe ICSMFE, Montreal, Vol II, pp. 214-217,
1965
[WI7] WILSON E.L, DOVEY H.H, Three Dimensional Analysis ofBuilding Structures-TABS,
Research Report No. EERCI72/8, Dept. of Civil Engineering, University of California,
Berkeley, 1972
[W13] WINOKUR A, GLUCKJ, Lateral Loads in Asymmetric Multistorey Structures, Journal

of the Structural Division, ASCE, Vo1.94, STI. pp. 645-656, 1967
415
[WI9] WOLF J.P, Consistent Lumped Parameter Models for unbounded Soils .. Physical
Representation, EESD, Vo1.20, pp. 11-32, 1991
[W20] WYNHOVEN J.H, ADAMS P.P, Analysis of Three Dimensional Structures, Journal
of the Structural Division, ASCE, ST!, pp. 233-247, 1972
[W21] WANG M.L, SHAH S.P, Reinforced Concrete Hysteretic Model based on the Dalnage
Concept, EESD, VoL 15, pp.993-1003, 1987
[W22] WONG H.L, LUCO J .E, Table of Impedance Functions for Square Foundations on
Layered Media, SDEE, VolA, pp.64-81, 1985
[W23] WONG H.L, LUCO J.E, Dynamic response of Rigid Foundations of Arbitrary Shape,
EESD, VolA, pp. 579-587, 1976
[W24] WONG H.L, LUCO J.E, Table of Impedance Functions and Input Motions for
Rectangular Foundations, Report No. CE 78-15, Dept. of Civil Engineering, University
of Southern California, 1975
[W25] WEN R.K, NATARAYAN P.S, Inelastic seismic Behaviour of Frame-Wall Systems,
Proceeding of the Fifth WCEE, Rome, VoU, pp.1343-1352. 1973
[W26] WEN R.K, JANSSEN J.G, Dynamic Analysis of Elasto-Inelastic Frames, Proceeding
of Third WCEE, New Zealand, VoUI, pp.714-729, 1965
[W27] WHITE W, VALLIAPP AN S, LEE I. K, Unified Boundary for Finite Dynamic Model,
Journal of the Engineering Mechanics Divisions ASCE, Vo1.103, EMS, pp.949-964,
1977
[W28] WAYNE W.H, CHATTERJEE M, UNEMORY A, Comparison of Analysis Methods

for seismic Soil-Structure Interactions, Proceeding of the Seventh WCEE, Istambul
Turkey, VoL5, pp.141-148, 1980
[W29] WATSON S, ZAHN F.A, PARK R, Corifining Reinforcementfor Concrete Columns,

Journal of Structural Engineering ASCE, Vo1.120, No.6, pp.1798-1824, 1994
[W30] WATSON S, PARK R, Simulated Seismic Load Test on Reinforced Concrete Columns.
Journal of Structural Engineering ASCE, Vol. 120, No.6, pp. 1825-1849, 1994
[W31] WERNER S.D, Engineering Characteristics of Earthquake Ground Motion, Nuclear

Engineering and Design, Vol. 36, pp. 367-395, 1976
[W32] WYLLIE Jr L.A, FILSON l.R, Annenia Earthquake Reconnaissance Report, Chapter
& and 7, Earthquake Spectra, pp. 70-100, august, 1989
[Xl] Canadian Geotechnical Society, Canadian Foundation Manual, Bi Tech Publisher L H,

Vancouver, 1982
[X2] Standards Association of New Zealand (SANZ), New Zealand Standard; Code and
Practice for the design of Concrete Structure NZS 3101: 1982
416
[X3] Standards Association of New Zealand (SANZ), Commentary on the Design of Concrete
Structure, NZS 3101:1982
[X4] SANZ, New Zealand Standard Code of Practice for General Structural Design and
Design Loadings for Buildings NZS 4203: 1984
[X5] SANZ, New Zealand Standard Code of Practice for General Structural Design and
Design Loadings for Buildings NZS 4203: 1992
[Yl] YOSHINORI NH, Experimental Half-Space Dynamic Stiffness , Journal of the

Geotechnical Engineering ASCE, Vol.l13, No.ll, pp. 1359-1373, November 1987
[Y2] YOSHIHIRO SUGIMURA, Seismic Strain induced in the Ground During Earthquake,
Report no. UCB/EERC-77114, College of Engineering University of California
Berkeley, 1977
[21] ZEEVAERT L , Foundation Engineering Jor D(fficult Subsoil Conditions , Van

Nostrand Reinhold Company, New York, 1973
[Z2] ZEN K, UMEHERA D, HAMADA K, Laboratory Test and in-Situ Seismic Survey on
Vibratory Shear Modulus oj Clayey Soils with Various Plasticities, Proceeding of the
Fifth Japan Earthquake Engineering Symposium, pp. 721-728, November 1978
[Z3] ZAHRAH T.F, HALL W.J, Earthquake li:nergy Absorbtion in SDOF Structures, Journal
of Structural Engineering, ASCE, Vol.1lO, No.8, pp. 1757-1772, 1984
[Z4] ZINN R, STANGENBERG F, ImportanceoJSoilDamping Idealization in Soil-Structure

Interactions, Proceeding of the Tenth WCEE, Tokyo, Vo1.III, pp. 1543-1548, 1988
[Z5] ZHAO X.Q, Seismic Soil-Structure Interaction, Ph.D Thesis, Dept. of Civil
Engineering, University of Canterbury, 1989
[Z6] ZHU TJ, TSO W.K, HEIDBRECHT A.C, Effect oj Peak Ground A/V Ratio on
Structural Damage, Journal of Structural Engineering ASCE, Vo1.114, No.5, pp. 1019-
1037, 1988
[Z7] ZHU T.], HEDBRECHT A.C, TSO W.K, Effect oj Peak Ground Acceleration to
Velocity Ratio on Ductility Del'nand oJInelastic Systems, EESD, Vol. 16, pp.63-79, 1988
[Z8J ZAHN F.A, PARK R, PRIESTLEY MJ.N, CHAPMAN H.E, DevelopmentoJDesign

Procedures Jor the Flexural Strength and Ductility oj RC Bridge Columns, Bulletin of
the NZSEE, Vol. 19, pp.200-212, 1986
[Z9] ZHU T.J, TSO W.K, HEIDEBRECHT A.C, Seismic Performance oj RC Ductile
Moment Resisting Frame Building Located in Different Seismic Regions, Canadian
Journal of Civil Engineering, pp. 688-710, 1992
A.I
Appendix A,I :
Rewriting Eq. 4.22.a,
(A.1.1)
P==ky+fld'y
tit
(A. 1.2)
The first derivative of Eq. A.I.I may written in the form,
(A.I.3)
Eq. A.1.2 is a first order non-homogenous linear differential equation and it is

convenient to express it in terms of,
y + OCt) y = R(t) (A.I.4)
Solution of Eq. A.I.4 can be obtained by the integral factor,
(A,1.S)
ef
Q(t)dt
Z =
Solution of this differential equation of may be determined by multiplication of Eq.

A.1.4 by Eq. A.I.5, then integrating and yields,
y =e -fQ(t)dt [ fRet) eQ(t)dtdt + c] (A. 1.6)
Eq. A.1.6 is the general solution of Eq. A.l.4. Values of the displacement y depend
on the function Q(t) and R(t) and the initial conditions. If Lbe values of Q(t) = kl1'/ and R(t) =
P/1'/ are taken as constants, with initial condition at t = 0, Y = 0 and P = Po , then Eq. A.1.6
may be rewritten as,
(A.1. 7)
A.2
Appendix A.2 :
Rewriting Eq. 4.25,
dy _ ldP p
+ - (A.2.1)
dt k dt
or,
dP + !p = kdy
dt 11 dt (A.2.2)
In the case of P(t) = Po is a constant, integrating of Eq. A.2.1 with respect to time
t leads to,
(A.2.3)
If an external load is applied to the system, beginning at time t = 0, there is no strain
contribution from the dashpot, only the spring gives the initial strain, and the value of C) in
Eq. A.2.3 is equal to zero, therefore,
(A.2.4)
Eq. A.2.4 indicates that if the spring stiffness k and viscosity 11 are assumed to be
constant, the strain or displacement will decrease more rapidly at the beginning of the time
period under consideration.
Similarly to Eq. A.I.3, Eq. A.2.2 may be written as the following first order
differential equation,
(A.2.S)
P+ Q(t) P = R(t)
Eq. A.2.5 is similar to Eq. A.l.4, therefore the solution of Eq. A.2.5 yields,
A.3
(A.2,6)
As in the Voigt-Kelvin model, variation of stress with time ofEg. A.2.6 depend on the
functions Q(t) , R(t) and dy/dt. Supposing Q(t), R(t) and the strain or displacement yare
constants, with the initial condition at t =0 ,P = Po and dy/dt := 0, then Eg. A.2.6Ieads to,
(- .! t) (A.2.7)
p = P" e \1
Appendix A.3 :
Rewriting Eq. 4.32.a,
dy (A,3.1)
.- + (k 1) Y = 1. P
tli 1
Eq. A.3.1 is similar to Eq. A.1.2, and the solution of Eq. 4.39 is similar to the
previously solutions, therefore,
(A.3.2)
The relationship between strain or displacement y and time t from Eq. A.3.2 depends
upon 1\. If A is assumed to be a constant and with the initial condition that at t = 0 there is no
strain contribution of the dashpot element, Eq. A.3.2 yields,
(A.3.3)
--- l1li0
.01
(01
: Fie !
~o
!Elct. Col. i 11116) ~
e 0 .. 0 1
(9 014 0 14

III 0 0

Ii! 0 12 /.0 12
0 I 0
0 10 .0 10
0 0 0
: Md+1I Blank: Md+e
Mrd Mn!
0
: l1li1
0 lit 0 : lVIl
0 ill
0 0 0
II 0 It 0
0

t
l1li 0 0
.
iii
. 0 ill 0
, 0
0 0
III
0 ill
)- 0

~. 0
0
150 -100 50 0 60 100 150 -150 -100 60 0 SO 100 150

Beam Moment tTml ea..m Moment ITml
Fig. B. i R3diatribution of Beam Momente Fig. B.2 Redistribl.l1ion of 8eem Moments

C.1
Square 40EcNS
Circle 40EcEW
0.8 Trlan 9 Tok.Oki
Diamd. Buch.NS
Sq.FiII. Parkfield
><
III
'tI
.E 0.6
OJ
Cl
ro
E
iii
t:l
a; 0.4
J:l
E
III
::iE
0.2
o
o 10 20 30 40 50 60 70 so
Member Curvature (x 0.001 Rad.lm).
Fig. C.1 Member Damage Index vs. Member Curvatature
Square 40EcNS
Circle 40EcEW
0.8 ..... .. Trian 9
~
Tok.Oki
Diamd. Buch.NS
0
Sq.FIII. Parkfield
>< 000
III
"C
.5
0.6
!lJ
C1
<II
E
III
t:l
0:;
.c 0.4-
E
'"
::iE
0.2
o
o 1 2 3 4 5
Hyst. Energy Dissipation (Tm)
Fig. C.2 Member Damage Index vs. Member Hyst. Energy Dissipation
C.2
0.8
Square 40EcNS
Circle Tok.Oki
0.6 .......... Triang. 40EcEVV . .....f...... .... ....... -.+...............................................-~ .......... ~
Diamd. Parkfield
.,
X
"I.':l
Sq.FiII. Buch.NS
.E
'Il>
OJ
OJ
DIm := 13.911 X tjJ
E 0.4
OJ
0
.
III
.a
E
III
o oo
::iE o I!III I!III
0.2
o
o 10 20 30 40 50
Member Curvature tjJ Ix 0.001 Rad./m)
Fig. C.3 Member Damage Index vs. Member Curvature
0.8
Square WEeNS
Circle Tok.Okl
0.6 Triang. 40EcEW
Diamd. Parkfield
Sq.FilI. Bui:h.NS
.,
X
't:J
.E
OJ
Cl
lI'I
E 0.4
lI'I
0
...
III
.J:l
E
III
::iE
0.2
o
o 2 3 4
Hyst. Energy Dissipation (Tm)
C.3
0.7
Square 40EcN-S
0.6 Tok~ Oki I11III...............+ ............................................. !.........................-.................-.+..........7"--.~ ...............-.." .. .
Circle
Triang. 40EcE-W
Diamd. Parkfield
0.5 Filled : Buch.N-SII+i..~~T<.. +11
><
/l)
"C
.= DIm =13.702 X
0>
OJ 0.4
<0
E
l\'J
Q
ill 0.3
.0
E
OJ
::;;:
0.2
0.1 . . =...............................+....................... .. .. + .. ..IF18.Takeda
0 10 20 30 40 50
Member Curvature {x 0.001 Rad./ml
0.7
Square 40EcN-S
0.6 Circle Tok. Oki II1II.............................. -+ ...............................................................,.....................1
Triang. 40EcE-W
Diamd. Parkfield
0.5 ............ ... Filled : Buch.NS II.. ,.."'."'Yx' 0-...""-1...... ....1
)(
II>
1:1 II
-=
II>
18
tl) 0.4
I
E
I
CI
:u
J:l
0.3
E
~
til e 0
0.2
0.1
0
0 1 2 3 4
Member Hyst. Energy Dissipation (Tm)
Fig. C.G Member Damage Index vs. Member Hyst. Energy Dissipation
C.4
0.8
Square 40EcN-S
0.7
Circle Tok.Oki
Triang. 40EcE-W 0
Diamd. Parkfield
0.6
Filled : Buch.N-S
)(
w
"'C
.s 0.5
w
01
ro
E
OJ 0.4
0
b
W
.0
E
w 0.3
:2:
0.2
0.1
0
0 10 20 30 40 50 60 70 80
Member Curvature tjJ Ix 0.001 Rad.lml.
0.8
Square 40EcN-S
0.7 Circle Tok.Oki.
Triang. 40EcE-W 0
Diamd. Parkfield 0
0.6 Filled : Buch. N-S
,(
QJ
'tl
.E 0.5
QJ
0>
III
E 0.4
c'"
Q;
..c
E 0.3
QJ
:2:
0.2
0.1
odg
Ii
0
0 5 10 15 20 25 30
Member Hyst. Energy Dissipation(Tm)
C.5
0.8
0.7
0.6
><
III
't1
..E 0.5
(\)
I:Il
III
E 0.4
<!J
Ci
t;
.0
E
(J)
0.3
:iE DIm := n.10 X
0.2
0.1
o-
0 10 20 30 40 50 60
Member Curvature (x 0.001 Rad.lml.
0.8
0.7
0.6
><
<!l
"..E 0.5
III
rn
ru
E
III 0.4
0
@
..c
E 0.3
OJ
::E
0.2
0.1
0
0 5 10 15 20 25 30
Member Hyst. Energy Dissipation (Tm)
Fig. C.10 Member Damage Index vs. Hyst. Energy Dissipation

C.6
0.6
0.5 .......... - Square : 40EcNS

Circle : Buch.NS
Trlang. 40EcEW
Diamd. : Parkf.1
Sq.FiII. : Parkf.2
X ...-... CLFil!.
(j)
"C
0.4 - : Taft 1
.5
(j)
(j)
I
E
t\J 0.3
Cl
lil
.!J
E
(j)
:2: 0.2
0.1
o
o 10 20 30 40 50
Member Curvature t. x 0.001 Rad./m).
Fig. C.l1 Member Damage Index vs. Member Curvature
0.5
0.4
)(
ill
"C
.5
0.3
OJ
CI
m
E
m
C
.
.2: 0.2
E
OJ
:E
0.1
o
o 10 20 30 40 50
Member Curvature I Ix 0.001 Rad.im).
Fig. C.12 Member Damage Index V5. Member Curvature

D.I
Appendix D:
Columns and Beam Properties of the 12-storey Frame

Columns Beams
Size (m) My Outer Ends (tm) My Int. Ends (!m) Reinforc.
Storey Storey Size (m) D28
No Ext. Col. Int. Col. Pas. Neg. Pos. Neg. top/bottom
1 I 0.65xO.65 0.75xO.75 1 OAOxO.80 66040 -58.05 58.14 -73.07 4/4
2 2-4 0.60xO.60 0.70xO.70 2-4 OAOxO.80 66.40 -58.05 58.14 -73.07 4/4
3 5-7 0.60xO.60 0.70xO.70 5-8 oAOxO. 80 65.52 -58.65 58.74 -71.90 3/4
4 8 - 10 0.60xO.60 0.70xO.70 9 - 12 0.30xO.80 52.37 -45048 45.62 -58.82 3/3
5 11 - 12 0.60xO.60 0.70xO.70
Column Bases: Outer column: Pyc = -984.86 t, PB = -742045 t, MB = 76.98 tm, MlB = 104.67 tm,
M1B = 93045 tm, Mo = 42.81 tm, Pyt = 150.65 t
Int. Column : The internal column is modelled as beam element (the axial load is nearly constant) :
AXI = AX2 = -838.30 t, P~" = -1532.97 t, My = 177.51 tm, Pyt = 338.96 t.
Columns and Beam Properties of the IS-storey Frame

Columns Beams
Size (m) My Outer Ends (tm) My Int. Ends (tm) Reinforc.
Storey Storey Size (m) D28
No Ext. Col. Int. Col. Pos. Neg. Pos. Neg. top/bottom
I. 1 0.80xO.80 0.95xO.95 I OAOxO.80 79.45 -71.46 71.51 -86.03 5/5
2. 2 0.80xO.80 0.95xO.95 2-8 0.50xO.90 91.13 -80.88 81.19 -97.43 5/5
I 3. 3-4 0.80xO.80 0.95xO.95 9 - 12 0.50xO.80 91.13 -80.88 81.19 -97.43 5/5
4. 5-6 0.80xO.80 0.90xO.90 13 - 15 OAOxO.90 74.97 -66.18 66.27 -82.39 4/4
5. 7-9 0.80xO.80 0.90xO.90 16 - 18 0.30xO.80 64.99 -45.74 45.84 -71.39 3/4
6. 10 - 12 0.70xO.70 0.80xO.80
7. 13 - 15 0.70xO.70 0.80xO.80
8. 16 - 18 0.70xO.70 0.80xO.80
Column Bases: Outer column: Pyc = -1455.83 t, PB = -1133.50 t, MB = 123046 tm, MlB = 176.96 tm,
M2B = 154.86 tm, Mo = 54.60 tm, P yt = 150.65 t
Int. Column : The internal column is dodelled as beam element (the axial load is nearly constant) :
AXl = AX2 = -1314.86 t, Pyc = -2234.20 t, My = 301.29 tm, Pyt = 286.85 t.
Note: The beam yield moment My includes the contribution of the yield slab reinforcement at the flange region.
The column and walls of the upper-storeys are assumed remain elastic
D.2
Column, Beam and Wall Properties of the I2-storey Frame-Wall Buildings

Column Beams Wans Ow = 5 m)
Outer Ends
My (tm) St. bw bi At Aw
1-----,---......, St Size (m) Pos. Neg.
m. m' m2
1.020
1.020
1.080
5 .975
Column Bases: Outer column: Pyc = -1002.36 t, Po = -746.03 t. Me = 84.82 tm, MIs == 114.82 tm,
M2B = 103.62 tm, M., = 52.98 tm, Pyt 188.32 t
Int. Column : The internal column is modelled as beam element (the axial load is nearly constant)
AXl = AX2 = -838.30 t, Pyc -1700.79 t, My = 217.03 tm, Pyl 338.96 t.
Cant. Walls Wall Base : The structural wall is modelled as beam element (the axial load is nearly constant)
AXI = AX2 = -1220.0 t, Pyc == -5145.56 t, = 3709.81 tm, P YI 737.10 t.
Note: bw = web thickness, bf = boundary element length, Aw web area, Ab boundary element area
Column, Beam and Wall Properties of the IS-storey Frame-Wall Buildings

Column
St.
No 1-----,-----1 St.
Column Bases: Outer co[umn: Pvc = -1473.32 t, Po = -1136.41 t, Mo = 133.60 tm, Mm 189.96 tm,
M2B = 167.83 tm, M., = 67.61 tm, Pyt = 188.32 t
Int. Column : The internal column is modelled as beam element (the axial load is nearly constant)
AX! = AX2 = -1314.86 t, Pye = -2575.31 t, My 296.32 tm, Pyt 301.29 t.
Cant Walls Wall Base : The structural wall is modelled as beam element (the axial load is nearly constant)
AXI = AX2 = -2198.77 t, Pyc = -8132.79 t. 1\IIy 9076.30 un, PYI 922.47 L
Note: bw web thickness, b f = boundary element length, Aw = web area, Af boundary element area
D.3
CoLumns Properties of the 12-storey

Column Properties
0.200
Note: A = Cross Section Area, AS = Shear Area, I Moment Of fuertia, END = Length of Rigid Block
Beam Properties of the 12-storey Frame

Initial Cond., Properties of the Left Bay Beams and External Nodal Loads and Lumped Weight of Exterior Joint
0 Storey VI =V2 (t) 1=M2 (tm) FY (t) A (m2) AS (m") I (m4)
1. 14.57 35.310
2. 2 14.57 25.740 34.630
3. 3 14.57 25.740
4. 4 14.57 0.175
5. 14.57 0.175 34.16
14.57 0.175 34.16
33.94
33.72
33.51
13.51 24.100 33.51
13.51 24.100 33.31
13.51 24.10 23.64
Note: V Fixed-End Shear, M Fixed End Moment, FY = Joint External Static Load,
WY Lumped Nodal Weight
DA
Beam Properties of the 12-storey Frame

Initial Cond., Properties of the Right Bay Beams and Intemal Nodal Loads and Lumped Weight of Interior Joint
No Storey VI =V2 (t) Ml=M2 (tm) FY (t) A (m2) AS (m2) I (m4) ENDI (m) END2 (m) WY (t)
1. 1 14.57 25.740 59.570 0.2570 0.2570 0.00884 0.188 0.163 59.57

2. 2 14.57 25.740 58.210 0.2570 0.2570 0.00884 0.175 0.150 58.21
3. 3 14.57 25.740 58.210 0.2570 0.2570 0.00884 0.175 0.150 58.21
4. 4 14.57 25.740 57.890 0.2570 0.2570 0.00884 0.175 0.150 57.89
5. 5 14.57 25.740 57.570 0.2570 0.2570 0.00884 0.175 0.150 57.57
6. 6 14.57 25.740 57.570 0.2570 0.2570 0.00884 0.175 0.150 57.57
7. 7 14.57 25.740 57.280 0.2570 0.2570 O.O~ 0.175 0.150 57.28
8. 8 14.57 25.740 56.980 0.2570 0.2570 0.00884 0.175 0.150 56.98
9. 9 14.57 25.740 56.980 0.1927 0.1927 0.00661 0.175 0.150 56.98
10. 10 13.51 24.100 56.710 0.1927 0.1927 0.00663 0.175 0.150 56.71
11. 11 13.51 24.100 56.440 0.1927 0.1927 0.00663 0.175 0.150 56.44
12 12 13.51 24.10 39.380 0.1927 0.1927 0.00661 0.175 0.150 39.38
Columns Properties of the I8-storey Frame

Column Properties
External Column Internal Column
Storey
No A (ml) AS (m2
) I (m4) ENOl (m) IEND2 (m) A (m2
) AS (m 2
) I (m4) ENOl (m) END2 (m)
1 1 0.5632 0.5632 0.02389 0.175 0.200 0.8392 0.8392 0.054300 0.175 0.200
2 2 0.6400 0.6400 003414 0.200 0.225 0.9025 0.9025 0.067870 0.225 0.225
3 3-4 0.6400 0.6400 0.03414 0.225 0.225 0.9025 0.9025 0.067870 0.225 0.225
4 5-6 0.6400 0.6400 0.03414 0.225 0.225 0.8100 0.8100 0.054680 0.225 0.225
5 7-9 0.6400 0.6400 0.03414 0.225 0.225 0.8100 0.8100 0.054680 0.225 0.225
6 10 - 12 0.4900 0.4900 0.02000 0.225 0.225 0.6400 0.6400 0.034130 0.225 0.225
13 - 15 0.4900 0.4900 p.02000 0.213 0.213 0.6400 0.6400 0.034130 0.213 0.213
16 - 18 0.4900 0.4900 p.02000 0.213 0.213 0.6400 0.6400 0.034130 0.213 0.213
0.5
Beam Properties of the I8-storey Frame (Left Bay, External Joint)

Storey V1=V2 (t) M1=M2 (tm) FY (t) A (mZ) AS (m2) ~ (m4) ~ND1 (m) fEND2 (m) WY (t)
:No
1. 1 14.57 25.740 37.450 0.2464 0.2464 0.00770 0.200 0.238 37.45
2. 2 15.85 27.700 36.260 0.3465 0.3465 0.01321 0.200 0.225 36.26
3. 3 15.85 27.700 36.260 0.3465 0.3465 0.01321 0.200 0.225 36.26
4. 4 15.85 27.700 36.260 0.3465 0.3465 0.01321 0.200 0.225 36.26
5. 5 . 15.85 27.700 36.260 0.3465 0.3465 0.01321 0.200 0.225 36.26
6. 6 15.85 27.700 35.740 0.3465 0.3465 0.01321 0.200 0.225 35.74
7. 7 15.85 27.700 35.740 0.3465 0.3465 0.01321 0.200 0.225 35.74
8. 8 15.85 27.700 35.740 0.3465 0.3465 0.01321 0.200 0.225 35.74
I 9. 9 15.85 27.700 35.160 0.3465 0.3465 0.01321 0.175 0.200 35.16
10 10 15.85 27.700 34.950 0.3465 0.3465 0.01321 0.175 0.200 34.95
1 .
11. 11 15.85 27.700 34.950 0.3465 0.3465 0.01321 0.175 0.200 34.95
12. 12 15.85 27.700 34.950 0.3465 0.3465 0.01321 0.175 0.200 34.95
13. 13 13.25 24.120 34.950 0.2772 0.2772 0.011056 0.175 0.200 34.95
14. 14 13.25 24.120 34.950 0.2772 0.2772 0.011056 0.175 0.200 34.95
15. 15 13.25 24.120 34.820 0.2772 0.2772 0.011056 0.175 0.200 34.82
16. 16 13.25 24.120 34.690 0.1848 0.1848 0.005824 0.175 0.200 34.69
17. 17 13.25 24.120 34.690 0.1848 0.1848 0.005824 0.175 0.200 34.69
18. 18 13.25 24.120 24.320 0.1848 0.1848 0.005824 0.175 0.200 24.32
Beam Properties of the 18-storey Frame (Right Bay, Internal Joint)

Initial Cond., Properties of the Right Bay Beams and Internal Nodal Loads and Lumped Weight of Internal Joint
No
Storey V1=V2 (t) Ml=M2 (tm) FY (t) !A (mZ) fAs (m2) (m4) END! (m) IEND2 (m) WY (t)
1. 1 14.57 25.740 62.490 0.2464 0.2464 0.00770 0.238 0.200 62.49

2. 2 15.85. 27.700 60.380 0.3465 0.3465 0.01321 0.225 0.200 60.38
3. 3 15.85 27.700 60.380 0.3465 0.3465 0.01321 0.225 0.200 60.38
4. 4 15.85 27.700 60.380 0.3465 0.3465 0.01321 0.225 0.200 60.38
5. 5 15.85 27.700 60.380 0.3465 0.3465 0.01321 0.225 0.200 60.38
6. 6 15.85 27.700 60.000 0.3465 0.3465 0.01321 0.225 0.200 60.00
7. 7 15.85 27.700 60.000 0.3465 0.3465 0.01321 0.225 0.200 60.00
8. 8 15.85 27.700 60.000 0.3465 0.3465 0.01321 0.225 0.200 60.00
9. 9 15.85 27.700 59.610 0.3465 0.3465 0.01321 0.200 0.175 59.61
10. 10 15.85 27.700 59.610 0.3465 0.3465 0.01321 0.200 0.175 59.61
11. 11 15.85 27.700 59.610 0.3465 0.3465 0.01321 0.200 0.175 59.61
12. 12 15.85 27.700 59.250 0.3465 0.3465 0.01321 0.200 0.175 59.25
13. 13 13.25 24.120 58.890 0.2772 0.2772 0.011056 0.200 0.175 58.89
14. 14 13.25 24.120 58.890 0.2772 0.2772 0.011056 0.200 0.175 58.89
15. 15 13.25 24.120 58.550 0.2772 0.2772 0.011056 0.200 0.175 58.55
16. 16 13.25 24.120 58.210 0.1848 0.1848 0.005824 0.200 0.175 58.21
1.7. 17 13.25 24.120 58.210 0.1848 0.1848 0.005824 0.200 0.175 58.21
18. 18 13.25 24.120 40.260 0.1848 0.1848 0.005824 0.200 0.175 40.26
D.6
Columns Properties of the 12-storey Frame-Wall

Column Properties
Storey
No iA (ml) AS (ml) I (m4) ENDI (m) iEND2 (m) A (m2) AS (ml) I (m4) ENDI (m) IEND2 (m)
1 I 0.3929 0.3929 ~.01190 0.188 0.200 0.5231 0.5231 0.021094 0.188 0.200
2 ~-4 0.3600 0.3600 ~.01080 0.200 0.200 0.4900 0.4900 0.020080 0.200 0.200
3 ~-7 0.3600 0.3600 p.Ol080 0.200 0.200 0.4900 0.4900 0.020080 0.200 0.200
4 8 - 10 0.3600 0.3600 p.01080 0.200 0.200 0.4900 0.4900 0.020080 0.200 0.200
5 11 - 12 0.3600 0.3600 ~.01080 0.200 0.200 0.4900 0.4900 0.020080 0.200 0.200
Structural Wall Properties of the 12-storey Frame-Wall
Properties and Initial Conditions of Structural Wall
No Storey A (ml) AS (ml) 1 (m4) fENDI (m) ~ND2 (m) AXI = AX2 (t)
1 1 1.6583 1.6583 3.17225 0.000 0.000 1220.0000
2 2-4 1.8200 1.8200 4.55335 0.000 0.000 1133.5000
3 5-7 1.7800 1.7800 ~.43073 0.000 0.000 892.8200
4 8 - 10 1.5250 1.5250 3.97252 0.000 0.000 587.0000
5 11 - 12 1.5250 1.5250 3.97252 0.000 0.000 281.0000
Beam Properties of the 12-storey Frame-Wall (Left Bay, External Joint)

No Storey Vl=V2 (t) Ml=M2 (tm) FY (t) A (ml) AS (ml) I (m4) ENDI (m) fEND2 (m) Wy
(t)
I. 1 14.57 25.740 35.310 0.2570 0.2570 0.00884 0.163 0.188 35.31
2. 2 14.57 25.740 34.630 0.2570 0.2570 0.00884 0.150 0.175 34.63
3. 3 14.57 25.740 34.630 0.2570 0.2570 0.00884 0.150 0.175 34.63
4. 4 14.57 25.740 34.390 0.2570 0.2570 0.00884 0.150 0.175 34.39
5. 5 14.57 25.740 . 34.160 0.2570 0.2570 0.00884 0.150 0.175 34.16
6. 6 14.57 25.740 34.160 0.2570 0.2570 0.00884 0.150 0.175 34.16
7. 7 14.57 25.740 33.940 0.2570 0.2570 0.00884 0.150 0.175 33.94
8. 8 14.57 25.740 33.720 0.2570 0.2570 0.00884 0.150 0.175 33.72
9. 9 14.57 25.740 33.720 0.1927 0.1927 0.00663 0.150 0.175 33.51
10. 10 13.51 24.100 33.510 0.1927 0.1927 0.00663 0.150 0.175 33.51
11. 11 13.51 24.100 33.310 0.1927 0.1927 0.00663 0.150 0.175 33.31
12 12 13.51 24.10 23.640 0.1927 0.1927 0.00663 0.150 0.175 23.64
D.7
Beam Properties of the I2-storey Frame-Wall (Right Bay, Internal Joint)

Initial Cond . Properties of the Right Bay Beams and Internal Nodal Loads and Lumped Weighl of Interior Joint
No
Storey Vl=V2 (t) !Ml=M2 (tm) FY (I) A (m2) AS (m2) I (m4) END! (m) iEND2 (m) WY (t)
1. 1 14.57 25.740 59.570 0.2570 0.2570 0.00884 0.188 0.163 59.57
2. 2 14.57 25.740 58.210 0.2570 0.2570 0.00884 0.175 0.150 58.21
3. 3 14.57 25.740 58.210 0.2570 0.2570 0.00884 0.175 0.150 58.21
4. 4 14.57 25.740 57.890 0.2570 0.2570 0.00884 0.175 0.150 57.89
5. 5 14.57 25.740 57.570 0.2570 0.2570 0.00884 0.175 0.150 57.57
6. 6 14.57 25.740 57.570 0.2570 0.2570 0.00884 0.175 0.150 57.57
7. 7 14.57 25.740 57.280 0.2570 0.2570 0.00884 0.175 0.150 57.28
8. 8 14.57 25.740 56.980 0.2570 0.2570 0.00884 0.175 0.150 56.98
9. 9 14.57 25.740 56.980 0.1927 0.1927 0.0066~ 0.175 0.150 56.98
10. 10 13.51 24.100 56.710 0.1927 0.1927 0.00663 0.175 0.150 56.71
11. 11 13.51 24.100 56.440 0.1927 0.1927 0.0066 0.175 0.150 56.44
12 12 13.51 24.10 39.380 0.1927 0.1927 0.0066 0.175 0.150 39.38
Colurrms Properties of the 18-storey Frame-Wall

I
Column Properties
Storey
No
2
A (m ) ~S (m2) I (m4) IENDI (m) END2 (m) fA (m2) AS (m2) I (m4 ) ENDI (m) IEND2 (m)
1 1 0.5632 0.5632 p.02389 0.175 0.200 0.8392 0.8392 0.054300 0.175 0.200
2 2 0.6400 0.6400 p.03414 0.200 0.225 0.9025 0.9025 0.067870 0.225 0.225
3 3-4 0.6400 0.6400 p.03414 0.225 0.225 0.9025 0.9025 0.067870 0.225 0.225
4 5-6 0.6400 0.6400 p.03414 0.225 0.225 0.9025 0.9025 0.067870 0.225 0.225
5 7-9 0.6400 0.6400 0.03414 0.225 0.225 0.8100 0.8100 0.054680 0.225 0.225
6 10 - 12 0.6400 0.6400 p.03414 0.225 0.225 0.8100 0.8100 0.054680 0.225 0.225
13 - 15 0.6400 0.6400 0.03414 0.225 0.225 0.6400 0.6400 0.034130 0.213 0.213
16 - 18 0.6400 0.6400 p.03414 0.200 0.200 0.6400 0.6400 0.034130 0.213 0.213
Structural Wall Properties of the I8-storey Frame-Wall
Properties and Initial Conditions of Structural Wall
No Storey A (m 2) AS (m2) I (m4) IENDI (m) END2 (m) AXI = AX2 (t)
1 1-2 2.7048 2.7048 9.65043 0.000 0.000 2198.7700
2 3 3.3600 3.3600 18.5229 0.000 0.000 2005.0050
3 4-7 3.3600 3.3600 18.5229 0.000 0.000 1811.8250
4 8 - 11 3.3600 3.3600 18.5229 0.000 0.000 1325.6100
5 12 - 15 3.3600 3.3600 18.5229 0.000 0.000 835.3700
6 16 - 18 3.3600 3.3600 18.5229 0.000 0.000 345.0500
D.8
Beam Properties of the 18-storey Frame-Wall (Left Bay, External Joint)
Initial Cond., Properties of the Left Bay Beams and External Nodal Loads and Lumped Weight of Exterior loint
Storey Vl=V2 (t) !Ml=M2 (tm) FY (t) IA (m2) lAs (m2) I (m4) ENDI (m) END2 (m) WY (t)
iNa
l. 1 15.85 27.700 37.990 0.2464 0.2464 0.00770 0.200 0.238 37.99
2. 2 15.85 27.700 36.260 0.3465 0.3465 0.01321 0.200 0.225 36.26
3. 3 15.85 27.700 36.260 0.3465 0.3465 0.01321 0.200 0.225 36.26
4. 4 15.85 27.700 36.260 0.3465 0.3465 0.01321 0.200 0.225 36.26
5. 5 I 15.85 27.700 36.260 0.3465 0.3465 0.01321 0.200 0.225 36.26
6. 6 15.85 27.700 36.370 0.3465 0.3465 0.01321 0.200 0.225 36.37
7. 7 15.85 27.700 36.010 0.3465 0.3465 0.01321 0.200 0.225 36.01
8. 8 15.85 27.700 36.010 0.3465 0.3465 0.01321 0.200 0.225 36.01
9. 9 15.85 27.700 35.800 0.3465 0.3465 0.01321 0.175 0.200 35.80
10. 10 15.85 27.700 35.520 0.3465 0.3465 0.01321 0.175 0.200 35.52
11. 11 15.85 27.700 35.520 0.3465 0.3465 0.01321 0.175 0.200 35.52
12. 12 15.85 27.700 35.100 0.3465 0.3465 0.01321 0.175 0.200 35.10
13. 13 13.25 24.120 34.690 0.2772 0.2772 0,Ol1056 0.175 0.200 34.69
14. 14 13.25 24.120 34.690 0.2772 0.2772 0,Ol1056 0.175 0.200 34.69
15. 15 13.25 24.120 34.390 0.2772 0.2772 0.011056 0.175 0.200 34.39
16. 16 13.25 24.120 33.700 0.1848 0.1848 0.005824 0.175 0.200 33.70
17. 17 13.25 24.120 33.700 0.1848 0.1848 0.005824 0.175 0.200 33.70
18. 18 13.25 24.120 23.840 0.1848 0.1848' 0.005824 0.175 0.200 23.84
Beam Properties of the 1S-storey Frame-Wall (Right Bay, Internal Joint)

Initial Cond., Properties of the Right Bay Beams and Internal Nodal Loads and Lumped Weight of Internal loint
Storey Vl=V2 (t) Ml=M2 (tm) FY (t) A (m 2) AS (m2) ~ (m4) IENDI (m) END2 (m) WY (t)
No
1. 1 15.85 27.700 63.840 0.2464 0.2464 0.00770 0.238 0.200 63.84
2. 2 15.85 27.700 61.180 0.3465 0.3465 0.01321 0.225 0.200 61.18
3. 3 15.85 27.700 61.180 0.3465 0.3465 0.01321 0.225 0.200 61.18
4. 4 15.85 27.700 . 6U80 0.3465 0.3465 0.01321 0.225 0.200 61.18
5. 5 15.85 27.700 61.180 0.3465 0.3465 0.01321 0.225 0.200 61.18
6. 6 15.85 27.700 60.400 0.3465 0.3465 0.01321 0.225 0.200 60.40
7. 7 15.85 27.700 59.610 0.3465 0.3465 0.01321 0.225 0.200 59.61
8. 8 15.85 27.700 59.610 0.3465 0.3465 0.01321 0.225 0.200 59.61
9. 9 15.85 27.700 58.910 0.3465 0.3465 0.01321 0.200 0.175 58.91
10. 10 15.85 27.700 58.210 0.3465 0.3465 0.01321 0.200 0.175 58.21
11. 11 15.85 27.700 58.210 0.3465 0.3465 0.01321 0.200 0.175 58.21
12. 12 15.85 27.700 57.600 0.3465 0.3465 0.01321 0.200 0.175 57.60
13. 13 13.25 24.120 56.980 0.2772 0.2772 0.011056 0.200 0.175 56.98
14. 14 13.25 24.120 56.980 0.2772 0.2772 0.011056 0.200 0.175 56.98
15. 15 13.25 24.120 56.700 0.2772 0.2772 0.011056 0.200 0.175 56.70
16. 16 13.25 24.120 56.420 0.1848 0.184S 0.005824 0.200 0.175 56.42
17. 17 13.25 24.120 56.420 0.1848 0.1848 0.005824 0.200 0.175 56.42
18. IS 13.25 24.120 39.360 0.1848 0.1848 0.005824 0.200 0.175 39.36
Equivalent Spring and Dashpot (at G = 0.70 Gm.J
I Shear Modulus G
I Foundations
. Equivalent Spring and Dashpot Coefficient
I Vert.Spring Hor.Spring Rock.Spring Vert-Damp. Hor-Damp. Rock.Damp.

I
i
KV (tim) KH (tim) KR (tID) CV (tsec!m) CH (tsec!m) CR (tsec!m)
2
1. G 420 kg!cm External Column 31477.610 37530.740 182682.600 1620.733 1387.326 ! 2792.093
(For 12-storey Internal Column 42044.000 45535.970 459574.700 2681.123 2082.435 7564.134
Frame and Frame-
Walls) Structural Wall 44267.120 46983.100 560212.800 2933.645 2238.338 9377.984
, ,
!
2. G 685 kg/cm2 External Column 59073.900 68312.140 459475.700 2766.440 2323.288 6726.401
Internal Column 80607.500 84931.380 1210161.00 4735.830 .836 18901.77 !

18-st.Frame
and Frame-Walls,
Chapter IX) Structural WaH 85373.670 88468.250 1461434.00 5241.020 3922.350 23177.260
I
.I
!
il
3. G = 540 External Column 46522.320 53817.950 360592.000 2398.426 1990.151 5455.221
Internal Column 63458.190 66894.700 947534.800 4023.070 2999.781 13895.55

!I
(For 18-stFrame I
and Frame-WaUs,
.~
Structural Wall 71726.790 72930.970 1429531.00 24370.38
Properties of Winkler Foundation ( 12-storey Frame-Walls, Section 11.3 )
Member Type Number

No Parameters
37 40 41 42 43
l. Spring Stiffness 3752.7 3752.757 6254.594 6254.594 6254.594
Bilinear Factor 0.05 0.05 0.05 0.05
Neg.Spring Yield -271.887 -271.887 -453.146 -453.146

Widodo Thesis

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Widodo Thesis

Uploaded by

Copyright:

Available Formats

A thesis

submitted in partial ..........,..........,..,......

DEPARTMENT OF CIVIL ENGINEERING

Analytical investigations using the same methods in more realistic applications of

Methods to overcome the disadvantages need to be developed. A new simple wall

Finally, I wish to express my deep gratitude to my parents for their prayers, to my

1.2 Objectives of the Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2.2 Experimental Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2.2.1 Clough and Huckelbridge (1977) . . . . . . . . . . . . . . . . . . . . . . . 7

2.2.2 Priestley, Evison, Carr (1978) . . . . . . . . . . . . . . . . . . . . . . . . 9

2.3 Analytical Investigations

2.3.2.1 Flexible Elastic SDOF Structure on the Winkler Foundation .. 19

2.3.2.2 Flexible Elastic SDOF Structure on a Two Spring-Dashpot

2.3.2.3 Flexible Elastic MDOF Structure on a Two Spring-Dashpot

2.3.3 Psycharis (1982) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

2.3.4 EI-Hifnawy and Novak (1986) . . . . . . . . . . . . . . . . . . . . . . . . 33

2.4 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

3.5.3 Design of Structural Walls. . . . . . . . . . . . . . . . . . . . . . . . . . 59

6.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 122

7.1 Introduction ... . . . . . . . . . . . . . 144

7.2.2 Dynamic Response of Soil-Foundation System

8.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 185

8.2.1 Lumped Parameter Method. . . . . . . . . . . . . . . . . . . . . . . . .. 185

9.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 207

10.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . " 250

10.3.4 Total Plastic Hinge Rotations and Energy Dissipation. . . . . . . .. 273

11.1 Introduction.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 296

12.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 321

13.1 Introduction . . . . . . . . . . . . . . . . 352

14.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 381

f = coefficient for second moment of inertia of flanged section

lw = length of structural wall

In" modulus of vertical subgrade reaction

r" = rigid joint zone reduction factor

Ym horizontal rocking displacement at the centre of mass of model

displacement of dashpot element

Cd,i = dynamic damping coefficient corresponds to vibration mode i

Cwrx2 = rocking damping coefficient induced by shear wave at sidewalls

E" = kinetic energy

Go = soil shear modulus at first cycle of cyclic loading

I"" = second moment of inertia of footing with respect to x-axis

lwg - gross second moment inertia of wall

~ = equivalent soil-foundation stiffness

~ = post elastic stiffness

Kdy,ernb = dynamic horizontal stiffness of embedded footing in y-directioll

K,,,,emb = static horizontal stiffness of embedded footing in x-direction

Ksy,emb = static horizontal stiffness of embedded footing in y-direction

K"I = static stiffness corresponds to vibration mode i

Kuxtsur = surface static rocking stiffness with respect to x-axis

K'IY"ur -- su rface static rocking stiffness with respect to y-axis

Ksx.lliur = static horizontal stiffness of surface fou ndation in x-direction

Kunld = reloading stiffness

KRR = rocking soil-foundation impedance

KHR - horizontal-rocking soil-foundation impedance

~ = length of beam plastic hinges

Lr = live load reduction factor

Lu minimum uniformly distributed live load

Me = critical overturning moment

M d +e moment due to dead and earthquake load

Mo moment amplitude of footing foundation

Moo moment after redistribution

N = number of blows in SPT soil test

R.n :::: moment Reduction factor

Sl>S2 = stiffness parameter of embedded foundation