See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/257943495

Seismic Risk Assessment of Unreinforced Brick Masonry Buildings System of

Northern Pakistan

Thesis · January 2009

CITATIONS READS

39 3,849

1 author:

Mohammad Javed
University of Engineering and Technology, Peshawar
44 PUBLICATIONS   324 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

DISASTER MITIGATION THROUGH CONSERVATION View project

All content following this page was uploaded by Mohammad Javed on 03 June 2014.

The user has requested enhancement of the downloaded file.

Seismic Risk Assessment of Unreinforced Brick
Masonry Buildings System of Northern Pakistan

by

MOHAMMAD JAVED

Department of Civil Engineering

Submitted in partial fulfillment
of the requirements for the degree of Doctor of Philosophy

Department of Civil Engineering,

N-W.F.P. University of Engineering and Technology,
Peshawar, Pakistan, 2009

© Mohammad Javed, 2009

 ABSTRACT

This research work was aimed at assessing the seismic risk of unreinforced brick masonry buildings’
system of Northern Pakistan, constructed in stone dust mortar. To accomplish this, four series of
unreinforced brick masonry piers constructed in stone dust mortar were tested in the in-plane direction
using quasi-static method of testing. Each pier series comprised of three piers with identical properties
and thus a total of twelve piers were tested. Aspect ratio and pre-compression were kept as the main
variables. Various properties such as displacement ductility factors, ultimate drift ratios, coefficient of
equivalent viscous damping, stiffness degradation and modulus of rigidity were determined using the
experimental data from quasi-static cyclic tests on the piers. The effect of pre-compression on the
coefficient of equivalent viscous damping and stiffness degradation were studied. Similarly, the effect
of drift ratio on the coefficient of equivalent viscous damping was also studied. Various performance
levels for unreinforced brick masonry piers, in relation to drift ratios, were also recommended.
Based on the results of experimental work, a methodology was proposed for lateral strength
assessment of unreinforced brick masonry buildings. The developed methodology produced
satisfactory results when compared with the results of full-scale unreinforced masonry (URM)
buildings tested at University of Pavia, Italy [MKC 95] and Georgia Institute of Technology, USA [
Yi 04]. Although the proposed methodology was used for brick masonry buildings, it can be applied to
other types of masonry (such as stone and concrete block masonry constructed in cement: sand mortar,
lime mortar, etc.) if the properties required to quantify the seismic performance (e.g., displacement
ductility factors and ultimate drift ratios of masonry piers, etc.) are experimentally known. Various
performance levels for unreinforced brick masonry buildings were also recommended in relation to
drift ratios.
Finally, seismic capacities of thirty-one buildings were evaluated. The buildings’ stock
consisted of seventeen single-story and fourteen double-story buildings. The buildings were selected
keeping in view their common typological use in urban areas of Northern Pakistan. Fragility curves,
showing the probabilities of reaching or exceeding various performance levels at various levels of
ground shaking, were drawn for various performance limit states. It was found by studying the
fragility curves that the probability of occurrence for various performance levels of single- and double-
story buildings do not differ significantly. It was also concluded that unreinforced brick masonry, if
properly constructed, can be safely used in localities placed in seismic zone 2b [BCP 07] and below.

ii
 ACKNOWLEDGEMENTS
All credit goes to the Almighty Allah who gave the author the ability to complete this challenging job.
The author expresses deep gratitude to his research supervisor Prof. Dr. Akhtar Naeem Khan. There
were considerable difficulties, partly due to his personal reasons, during this research project.
However, he always remained a source of inspiration for author. The author always got support from
him whether it was in the form of technical guidance or moral support. A number of times, the author
met him with many uncertainties regarding this research project but after the meeting, he felt himself
full of inspiration. In simple words, this project would never have been accomplished without his
guidance and support.

A lot of credit also goes to the author’s co-supervisor, Professor Dr. Guido Magenes, University of
Pavia, Italy. In 2005, the author had the opportunity to stay for six months in Pavia, Italy, in order to
study various courses related to earthquake engineering at ROSE school. During that stay, the author
had a deep interaction with him. He was always kind and willing to guide the author with regard to his
research work. On many occasions, he spared a lot of time to discuss various issues related to author’s
research work. He also provided author a lot of guidance in resolving various issues related to his
experimental work carried out at Institute of Engineering Mechanics (I.E.M.), Harbin, China. It will
not be wrong to say that the existing quality of this research project would not have been achieved
without his periodic technical guidance.

The author highly appreciates the guidance provided by his thesis evaluator, Prof. Dr. Miha
Tomaževic, especially during the initial stages of this research work. He was kind enough to invite him
to Ljubljana, Slovenia when he was studying in ROSE school, Italy. He hosted him during his stay in
Ljubljana and resolved a number of issues related to his research work. The author was very impressed
by his approach of resolving complicated issues in a simplified way, which places him among the top
researchers in the research field of seismic behavior of masonry structures.
It is the author’s honor to acknowledge the efforts of his second thesis evaluator, Prof. Zia Razzaq,
Old Dominion University, USA. He, in addition to providing valuable suggestions for improving the
quality of work, did his ample effort to transform this document into a standard dissertation.

The author is also grateful to Prof. Dr. Junwu Dai at I.E.M., Harbin, China who managed to arrange
his experimental work at I.E.M. and provided him every type of facility during his stay in Harbin,
China. Thanks are also due to Mr. Feng, graduate student at I.E.M., who not only worked as translator
during the author’s stay in Harbin, China, but also facilitated him whenever he needed his help.

It would not be out of place to thank all of my colleagues and support staff at the Department of
Civil Engineering, NWFP UET, who very diligently supported my research endeavors and provided
me every type of supported that I needed.
Finally, the author greatly acknowledges the NWFP University of Engineering and Technology, for
granting him four years of study leave and the Higher Education Commission (HEC) for providing
funds for his research project.

iii
 SYMBOLS

Roman upper case

Abm =Area of brick -mortar joint between the two bricks in a triplet used in the triplet

tests

Ac =Area of brick surface in contact with water used in the IRA tests

Af = Factor by which the ground motion amplifies at the top a of SDOF system.

Ap = Cross sectional area of a pier subjected to vertical loads= lp.t

An = Net area of the prism used in diagonal tension test

Ca =A coefficient used to calculate seismic demand on an elastic SDOF system using

Building Code of Pakistan [BCP 07]

Cv = A coefficient used to calculate seismic demand on an elastic SDOF system using

Building Code of Pakistan [BCP 07]

Ed = Energy dissipated by a pier in one complete cycle during a quasi-static cyclic test

Ei = Energy required in one cycle to impose a target lateral force/displacement on a pier

during quasi-static cyclic test

Em = Modulus of elasticity of masonry perpendicular to the bed joints

Emx = Modulus of elasticity of masonry parallel to the bed joints

Fop = Inertial force produced in a wall per unit height per unit length in the out-of-plane

direction

Fop,cr = Inertial force in a wall per unit height, for a unit length, required to produce

flexural tensile cracking in the out-of-plane direction

Gm = Modulus of Rigidity of masonry

Gm,cr =Modulus of rigidity of masonry piers at the formation of first significant crack

iv
H = Total height of a building

Hbt = Horizontal reaction at the bottom of a wall, for a unit length, in the out-of-plane

direction

Ht = Horizontal reaction at the top of a wall, for a unit length, in the out-of-plane direction

Iop = Moment of inertia of a pier, per unit length, in the out-of-plane direction

Ip = Moment of inertia of a pier in the in-plane direction

Isp = Moment of inertia of a spandrel

Isp1 = Moment of inertia of spandrel, connected to the left side of pier at the top end

Isp2 = Moment of inertia of spandrel, connected to the right side of pier at the top end

Isp3 = Moment of inertia of spandrel, connected to the left side of pier at the bottom end

Isp4 = Moment of inertia of spandrel, connected to the right side of pier at the bottom end

Mop,cr = Mid-height moment, for a unit length, causing flexural tensile cracking in a wall

the out-of-plane direction

K = Lateral secant stiffness of pier corresponding to a lateral displacement d.

= A parameter that quantifies relative stiffness of spandrels connected to the top and

bottom ends of a pier

Kb = Lateral elastic stiffness of a building

KE = Equivalent lateral elastic stiffness of SDOF system corresponding to the building,

KE = Kb

Kp = Lateral elastic stiffness of URM pier determined from the idealized bi-linear lateral

force displacement curve of a pier shown in fig 3.52

Kep = Lateral effective stiffness of an URM pier = 0.75Kp

Kop = Slope of aop/g vs ∆op diagram shown in Figure 4.9

Kpf = Flexural stiffness of pier as defined by Muto [PP 92]

v
Kspf = Flexural stiffness of a spandrel

Kspf1 = Flexural stiffness of a spandrel connected to the left side of pier at the top end

Kspf2 = Flexural stiffness of a spandrel connected to the right side of pier at the top end

Kspf3 = Flexural stiffness of a spandrel connected to the left side of pier at the bottom end

Kspf4 = Flexural stiffness of a spandrel connected to the right side of pier at the bottom end

Kw = Lateral elastic stiffness of an URM wall

M = Greater of the bending moments acting at the top and bottom end of a pier in the in-

plane direction

Mbt = Bending Moment acting at the bottom end of a pier in the in-plane direction

Mfl,cr = Bending Moment causing flexural tensile cracking in the bed joints of a unit

length of pier in the in-plane direction

Mmid = Mid-span bending Moment

Mop = Mid-height moment per unit length in a wall in the out-of-plane direction

Mop,cr = Mid-height moment causing flexural tensile cracking in a unit length of wall in

the out-of-plane direction

Mt = Bending moment acting at the top end of a pier in the in-plane direction

(MVu)t = Bending moment acting at the top of pier in the in-plane direction caused by a

lateral Vu

N = Vertical reaction at the top of a wall, in the out-of-plane direction, for a unit length

of the wall

ND = Vertical force applied on prism during diagonal tension test

(ND)u = Vertical force causing failure of prism in diagonal tension test

P = Total vertical load acting on a pier = σo.lp.t

R = Response modification factor of a building

vi
Rc = Vertical reaction at the mid-height of a wall, for a unit length, in the out-of-plane

direction = N+W/2

R∆ = Response modification factor of a building corresponding to a lateral displacement,

∆, at the top of building

S = Section modulus

Sa = Spectral acceleration in SDOF system corresponding to the representative building,

in the in-plane direction

Sa,op = Spectral acceleration in SDOF system corresponding to the representative building

, in the out-of-plane direction

Sd = Spectral displacement in SDOF system corresponding to the representative building,

in the in-plane direction

(Sd)∆ = Spectral displacement in SDOF system, corresponding to a lateral displacement,

∆, at the top of the building, in the in-plane direction

Ss = Average value of shear stresses produced along the vertical diagonal of prism used in

diagonal tension test

(Ss)max = Average shear stress in the prism along the vertical diagonal in a diagonal

tension tests corresponding to ultimate vertical load, (ND)u, assuming linear elastic

behavior

T = Natural time period of the building

To =A controlling period used in design response spectrum as recommended by Building

Code of Pakistan [BCP 07]

T1 = Natural time period of the building corresponding to the fundamental mode of

vibration

vii
Ts = A controlling period used in design response spectrum as recommended by Building

Code of Pakistan [BCP 07]

Tonne = 2204.6 lb

V= Lateral force acting on a pier/wall

Vfl,cr = Lateral force causing flexural tensile cracking in the bed joints of a pier

Vb = Base shear in the building

Vbe = Base shear in an elastic system corresponding to the representative building

(Vbe)∆= Base shear in an elastic system corresponding to a lateral displacement, ∆, at the

top of the representative building

Vd = Lateral force required to cause diagonal tension shear failure in an URM pier

Vdcr = Lateral force corresponding to the formation of first diagonal tension crack in the

URM pier

Vest = Estimated lateral strength of an URM pier

Vfcr = Lateral force at the formation of first significant crack in the URM pier

Vmax = Maximum lateral resistance of an URM pier determined from experimental work

Vmin = Minimum Lateral force along the degrading portion of lateral force displacement

envelope curve of an URM pier

Vr = Lateral force required to cause rocking failure in an URM pier

Vsf = Lateral strength of an URM pier determined from analytical equations

Vsl= Lateral force required to cause sliding shear failure in an URM pier

Vyb = Base shear in the building at the start of ‘yielding’

Vmb = Maximum lateral resistance of the building

Vus = Shear force producing failure in a triplet during the triplet test

viii
Vu = Lateral force in the bi-linear idealization of experimental force-displacement curve

of an URM pier = 0.9 Vmax

Vub = Base shear in the building at ultimate conditions

Vyb = Base shear causing ‘yielding’ in the building, determined from capacity curve

W = weight of the unit length of a wall

Ww = Weight of wet brick sample

Wd = Weight of oven dry brick sample

Roman lower case

ac= Depth of uniform compression stress block at the crushing of masonry under the

combined action of axial and/or flexural stresses = 0.85cc

ae,op = Acceleration required in the linear elastic system, corresponding to an URM wall,

to cause failure in the out-of-plane direction.

ag,op = Ground acceleration corresponding to ae,op

aop = Acceleration induced in a wall per unit height per unit length, in the out-of-plane

direction

aop,cr = Acceleration in a wall per unit height per unit length corresponding to the

formation of flexural tensile cracking in the out-of-plane direction in the URM wall

b = Shear correction factor, indicating distribution of shear stresses along the pier’s cross-

section, used for calculating diagonal tensile strength of URM pier

bb = Width of a brick

c = Cohesion ( also called bond-shear strength) between mortar and brick units

ix
cc = Depth of compression zone under the combined action of axial and/or flexural

stresses

c/c = Centre to centre

ddcr = Lateral displacement (w.r.t pier’s bottom) at the top of an URM pier at the

formation of first diagonal tension crack

d = Any lateral displacement (w.r.t pier’s bottom) at the top of masonry pier

dy = Lateral displacement (w.r.t pier’s bottom) at the top of an URM pier at the verge of

'yielding'

dywe = Lateral displacement at the top of a wall element at the verge of ‘yielding’

dfcr = Lateral displacement (w.r.t pier’s bottom) at the top of an URM pier at the

formation of first significant crack

du = Lateral displacement (w.r.t pier’s bottom) at the top of an URM pier corresponding to

a lateral force of 0.8 Vu along degrading curve or at the end of test, whichever occurs

first

duwe = Lateral displacement at the top of a wall element at the ultimate displacement, du,

in the corresponding pier

dywe = Lateral displacement at the top of a wall element at the initiation of ‘yielding’ in

the corresponding pier

dVmax = Lateral displacement (w.r.t pier’s bottom) at the top of an URM pier

corresponding to the stage when the piers attains its maximum resistance

f = Natural frequency of the building

f1 = Natural frequency of the building corresponding to fundamental mode of vibration

fb = Compressive strength of a brick unit perpendicular to bed joint

fbt = Flexural tensile strength of a brick

x
fm = Compressive strength of masonry

fmo = Compressive strength of mortar

ft= Bond strength of masonry in tension

ftu = Principal tensile strength of the masonry determined from diagonal tension tests

gl= Gage length

h= Story height

ho = Distance of lateral force from bottom of the pier during quasi-static cyclic test

hc = Height of the centroid of a top floor’s wall w.r.t the floor level

hi = Height of i-th floor level

hp = Height of a masonry pier

hpb = Distance of pier’s bottom from floor level

hsp = Depth of spandrel

hw = Over all height of a wall

hwe = Height of a wall element

lp = Length of a pier

lprism = Length of a prism used in the diagonal tension test

lw = Length of a wall. Normally, lw= lp

lsp = Length of a spandrel

lsp1 = Length of a spandrel connected to the left side of the pier at the top end

lsp2 = Length of a spandrel connected to the right side of the pier at the top end

lsp3 = Length of a spandrel connected to the left side of the pier at the bottom end

lsp4 = Length of a spandrel connected to the right side of the pier at the bottom end

mw = Mass of a wall per unit height per unit length

mi = Total mass lumped at i-th floor level

xi
mE = Equivalent mass in a SDOF system

r = Coefficient of correlation

r2 = Coefficient of determination

t = Thickness of a masonry pier/wall

tb = Thickness of a brick

tprism= Thickness of a prism used in the diagonal tension test

w = weight of a wall per unit height per unit length

x= Eccentricity of ‘Rc’ from the centroid of a wall in the out-of-plane direction

xmax= Maximum eccentricity of ‘Rc’ from the centroid of a wall in the out-of-plane

direction

Greek upper case

Γ = Modal participation factor

∆ = Lateral displacement at the top of a building

∆be = Lateral displacement corresponding to Vbe

(∆be)∆= Lateral displacement corresponding to (Vbe)∆

∆e = Mid-height deflection in the unit length of a wall caused by ae,op

∆op = Mid-height deflection in the unit length of the wall in the out-of-plane direction

∆op,cr = Mid-height deflection in the unit length of the wall corresponding to the flexural

tensile cracking in the out-of-plane direction

∆yb = Lateral displacement at the top of a building at the verge of ‘yielding’

∆ub.= Lateral displacement at the top of building at collapse prevention performance level

∆uw = Lateral displacement at the top of a wall at the ultimate conditions in the

corresponding pier

xii
∆yw = Lateral displacement at the top of a wall the at the initiation of ‘yielding’ in the

corresponding pier

∆V = Vertical shortening in the gage length of a prism during the diagonal tension test

∆H = Horizontal elongation in the gage length of a prism during the diagonal tension test

Greek lower case

α1 = A parameter, used by Muto [PP 92], to quantify the stiffness of spandrels connected

to the top end of pier relative to the spandrels connected to the bottom end of pier

αv = Shear ratio of a pier= M/Vlp

γs = shear strain corresponding to a diagonal shear stress Ss

δ = Drift ratio in a pier/wall

δdcr = Drift ratio in a pier (w.r.t pier’s bottom) at the formation of first diagonal tension

crack = ddcr/hp

δu= Drift ratio in a pier (w.r.t pier’s bottom) corresponding to du = du/ hp

δVmax = Drift ratio in a pier (w.r.t pier’s bottom) at a distress state when the pier gains its

maximum resistance = dVmax/ hp

δy= Drift ratio in a pier/wall element/ wall at the verge of yielding

εc,op = strain corresponding to σc,op

εop,cr = strain corresponding to σop,cr

η = (Distance of inflection point in a pier w.r.t its bottom) / lp = ηo + η1

ηo = A factor, used in Muto method, for determining location of inflection point in a pier

η1 = A correction factor, used in Muto method, for determining location of inflection

point in a pier

xiii
µ = Coefficient of friction of masonry

µb = Displacement ductility factor of an URM building

µp = Displacement ductility factor of an URM pier

µwe = Displacement ductility factor of a wall element

µw = Displacement ductility factor of a wall

µx = Mean value of population used in the statistical analysis

(µb)∆ = Displacement ductility factor of an URM building corresponding to a lateral

displacement, ∆, at the top of building

ξeq = Coefficient of equivalent viscous damping

σo = Pre-compression on a wall/pier

σc,op = Maximum compressive stress at the mid-height of an URM wall in the out-of-

plane direction

σf = Maximum flexural tensile stress under the action of bending moment, M

σn = Pre-compression applied on the triplet during a triplet test

σop,cr = Maximum compressive stress, at the mid-height of an URM wall, at the formation

of the flexural tensile crack in the wall in the out-of-plane direction.

σt = σf - σo

σx = Standard deviation of population used in the statistical analysis

τd = Diagonal tensile shear strength of URM piers

τu = Shear strength of masonry triplets determined from triplet test

φi = Fundamental mode normalized displacement at i-th floor level

φop =Curvature corresponding to Mop

φop,cr = Curvature corresponding to Mop,cr

ψ′ = Boundary condition factor equal to the greater of η and 1-η

xiv
 ABBREVIATIONS

A.R. = Aspect ratio of pier = hp/lp

ASTM = American Society for testing materials

C.O.V. = Coefficient of variation

CP = Collapse prevention

CS = cement: sand

CSK = Cement:sand:khaka

D.P.C. = damp proof course

F.F.L. = Floor finish level

IO = Immediate occupancy

IRA = Initial rate of Absorption

LS = Life safety

LVDT = Linear variable Displacement Transducer

N.A. = Neutral axis

NLD = Negative loading direction

N.S.L. = Natural surface level

O = Operational

P.C.C. = Plain cement concrete

PLD = Positive loading direction

PsD = Pseudo dynamic

R.C.C. = Reinforced cement concrete

URM = Unreinforced masonry

URBM = Unreinforced brick masonry

xv
 Dedicated to

my late parents, wife,

and

All those, who devoted their lives for the welfare of human being

xvi
 Table of Contents

ABSTRACT ..................................................................................................................................... ii
ACKNOWLEDGEMENTS ............................................................................................................ iii
SYMBOLS ...................................................................................................................................... iv
Table of Contents ......................................................................................................................... xvii
List of Figures .............................................................................................................................. xxii
List of Tables ............................................................................................................................. xxviii
Chapter 1 INTRODUCTION ........................................................................................................... 1
1.1 PROBLEM BACKGROUND AND DEFINITION ............................................................... 1
1.2 OBJECTIVES ........................................................................................................................ 1
1.3 OUTLINE OF THE DISSERTATION .................................................................................. 2
Chapter 2 LITERATURE REVIEW AND THEORITICAL BACKGROUND .............................. 4
2.1 INTRODUCTION .................................................................................................................. 4
2.2 RESPONSE OF MASONRY BUILDINGS SUBJECTED TO SEISMIC EXCITATIONS . 4
2.3 STRUCTURAL DAMAGE IN MASONRY BUILDINGS ................................................... 5
2.4 IN-PLANE DAMAGE ........................................................................................................... 5
2.4.1 Shear cracks ..................................................................................................................... 5
2.4.2 Crack damage at openings ............................................................................................... 6
2.4.3 Pier flexural-rocking failure ............................................................................................ 6
2.5 OUT-OF-PLANE AND LOCAL DAMAGE MECHANISMS ............................................. 6
2.5.1 Lateral thrust from roofs .................................................................................................. 6
2.5.2 Wedge- type separation at the top of wall junctions due to lateral thrust from the
supporting roof truss ................................................................................................................. 6
2.5.3 Lack of connection between the orthogonal walls .......................................................... 7
2.5.4 Damage at wall-to-roof connections................................................................................ 7
2.5.5 Out-of-plane overturning of gables ................................................................................. 7
2.5.6 Diffusion of stone masonry external veneer .................................................................... 7
2.6 DAMAGE TO MASONRY NON-STRUCTURAL ELEMENTS ........................................ 7
2.6.1 In-plane failure of infill walls .......................................................................................... 7
2.6.2 Out-of-plane failures of infill walls ................................................................................. 8
2.6.3 Failure of the boundary and parapet walls....................................................................... 8
2.7 SIMPLIFIED FORMULATIONS FOR PREDICTING IN-PLANE FAILURE MODES IN
URM WALLS .............................................................................................................................. 8

xvii
 2.7.1 Rocking/toe crushing failure ........................................................................................... 8
2.7.2 Sliding shear failure ......................................................................................................... 9
2.7.3 Diagonal tension shear failure ......................................................................................... 9
2.8 LABORATORY TESTING TECHNIQUES RELATED TO EARTHQUAKE
ENGINEERING ......................................................................................................................... 10
2.8.1 Quasi-static tests ............................................................................................................ 11
2.8.2 Pseudo-dynamic (PsD) tests .......................................................................................... 11
2.8.3 Shaking table tests ......................................................................................................... 12
2.8.4 Shock table dynamic test ............................................................................................... 13
2.9 BASIS FOR ADOPTED EXPERIMENTAL APPROACH ................................................ 14
2.10 SEISMIC VULNERABILITY ASSESSMENT OF BUILDINGS ................................... 15
2.10.1 Observed vulnerability ................................................................................................ 15
2.10.2 Vulnerability functions based on expert opinions ....................................................... 16
2.10.3 Analytical approach based on simple models.............................................................. 16
2.10.4 Score assignment ......................................................................................................... 17
2.10.5 Detailed analysis procedures ....................................................................................... 17
2.11 SELECTION OF METHOD FOR ASSESSING VULNERABILITY OF URM
BUILDINGS OF NORTHERN PAKISTAN ............................................................................. 18
Chapter 3 PERFORMANCE OF UNREINFORCED BRICK MASONRY (URBM) PIERS
UNDER IN-PLANE QUASI-STATIC LATERAL CYCLIC LOADINGS .................................. 39
3.1 INTRODUCTION ................................................................................................................ 39
3.2 SELECTION OF MATERIAL USED IN EXPERIMENTAL WORK ............................... 39
3.2.1 Bricks............................................................................................................................. 39
3.2.2 Mortar ............................................................................................................................ 40
3.3 MATERIAL TESTS............................................................................................................. 40
3.3.1 Brick tests ...................................................................................................................... 41
3.3.2 Mortar test ..................................................................................................................... 42
3.3.3 Masonry assemblage tests ............................................................................................. 42
3.4 IN-PLANE QUSAI-STATIC CYCLIC TESTS ON UNREINFORCED BRICK
MASONRY PIERS .................................................................................................................... 45
3.4.1 Geometry of piers .......................................................................................................... 45
3.4.2 Pre-compression on piers .............................................................................................. 46
3.4.3 Testing setup.................................................................................................................. 46
3.5 CRACKING PATTERN AND FAILURE MODES IN THE TESTED PIERS .................. 48
3.5.1 PI series ......................................................................................................................... 48
xviii
 3.5.2 PII series ........................................................................................................................ 48
3.5.3 PIII series ....................................................................................................................... 49
3.5.4 PIV series ...................................................................................................................... 49
3.6 LATERAL LOAD- DISPLACEMENT DIAGRAMS ......................................................... 50
3.7 DIAGONAL TENSILE SHEAR STRENGTH OF PIERS .................................................. 51
3.8 EVALUATION OF DISPLACEMENT BASED PROPERTIES FROM QUASI-STATIC
CYCLIC TESTS......................................................................................................................... 51
3.8.1 Displacement ductility factor of URBM piers ............................................................... 51
3.8.2 Modulus of rigidity of the masonry ............................................................................... 52
3.8.3 Energy dissipation ......................................................................................................... 53
3.8.4 Stiffness degradation ..................................................................................................... 54
3.9 DRIFT RATIOS CORRESPONDING TO VARIOUS PERFORMANCE LEVELS FOR
TESTED PIERS ......................................................................................................................... 55
3.10 RESULTS, CONCLUSIONS AND RECOMMENDATIONS .......................................... 56
Chapter 4 METHODOLOGY FOR EVALUATING LATERAL STRENGTH OF URM
BUILDING SYSTEMS ................................................................................................................ 117
4.1 INTRODUCTION .............................................................................................................. 117
4.2 TERMINOLOGY ............................................................................................................... 118
4.3 CONSTRUCTION OF CAPACITY CURVES ................................................................. 118
4.4 VARIOUS PERFORMANCE LEVELS PROPOSED FOR UNREINFORCED BRICK
MASONRY BUILDINGS........................................................................................................ 119
4.4.1 Operational level (O) ................................................................................................... 120
4.4.2 Immediate occupancy (IO) .......................................................................................... 120
4.4.3 Life safety (LS)............................................................................................................ 120
4.4.4 Collapse prevention (CP) ............................................................................................ 120
4.5 EVALUATION OF GROUND ACCELERATION CORRESPONDING TO VARIOUS
PERFORMANCE LEVELS ..................................................................................................... 120
4.6 DETAILS OF PROCEDURE FOR CONSTRUCTING THE CAPACITY CURVE FOR
URM BUILDING..................................................................................................................... 121
4.6.1 Structural model .......................................................................................................... 121
4.6.2 Material properties and geometrical properties of structural members ....................... 121
4.6.3 Determination of inflection points in piers .................................................................. 121
4.6.4 Calculation of gravity loads on piers ........................................................................... 123
4.6.5 Calculation of the flexural tensile cracking strength and shear strength of piers ........ 124
4.6.6 Bi-linear idealization of the lateral force-displacement curves of piers ...................... 126
xix
 4.6.7 Bi-linear idealization of the lateral force-displacement curves of the corresponding
walls...................................................................................................................................... 127
4.6.8 Construction of the capacity curve of URM building ................................................. 129
4.7 EVALUATION OF THE GROUND ACCELERATION CORRESPONDING TO THE
VARIOUS PERFORMANCE LEVELS IN A BUILDING..................................................... 130
4.7.1 Base shear and lateral displacements in the elastic MDOF system corresponding to the
actual building ...................................................................................................................... 130
where Kb is the lateral stiffness of the building as determined from Equation 4.1d............. 131
4.7.2 Lateral displacement in the elastic SDOF system corresponding to the actual building
.............................................................................................................................................. 131
4.7.3 Acceleration in the elastic SDOF system, (Sa)∆, corresponding to the lateral
displacement, (Sd)∆ ............................................................................................................... 131
4.7.4 Ground acceleration, (ag)∆ inducing spectral acceleration, (Sa)∆ in the elastic SDOF
system ................................................................................................................................... 132
4.8 RESONSE OF AN URM BUILDING IN THE OUT-OF-PLANE DIRECTION ............. 132
4.8.1 Mid-height curvature at out-of-plane cracking of wall................................................ 134
4.8.2 Mid-height deflection and associated acceleration at the onset of out-of-plane flexural
tensile cracking of wall ......................................................................................................... 135
4.8.3 Acceleration and corresponding mid-height deflections for various levels of strains
after the out-of-plane flexural tensile cracking of the wall .................................................. 136
4.8.4 Acceleration in the corresponding linear elastic system.............................................. 137
4.8.5 Evaluation of ground acceleration producing ae,op ....................................................... 137
4.9 COMPARISON OF THE PROPOSED METHOD WITH THE EXPERIMENTAL
RESULTS OF FULL-SCALE TESTED MODEL BUILDINGS ............................................ 138
4.9.1 University of Pavia building ........................................................................................ 138
4.9.2 Georgia Institute of Technology building ................................................................... 139
4.10 STEP BY STEP PROCEDURE FOR ASSESSMENT OF SEISMIC CAPACITY FOR
URM BUILDING..................................................................................................................... 140
4.11 CONCLUSIONS AND RECOMMENDATIONS ........................................................... 141
Chapter 5 SEISMIC RISK ASSESSMENT OF THE URBM BUILDING SYSTEMS OF
NORTHERN PAKISTAN............................................................................................................ 158
5.1 INTRODUCTION .............................................................................................................. 158
5.2 ESTABLISHING THE MAGNITUDES OF THE MATERIAL PROPERTIES OF THE
MASONRY USED IN PESHAWAR ...................................................................................... 158
5.2.1 Compressive strength of masonry, fm .......................................................................... 158
xx
 5.2.2 Principal tensile strength of masonry, ftu ..................................................................... 159
5.2.3 Bond strength of masonry in tension, ft ....................................................................... 159
5.2.4 Cohesion or bond shear strength, c, and coefficient of friction of masonry, ............ 159
5.2.5 Modulus of elasticity of the masonry, Em .................................................................... 159
5.2.6 Modulus of rigidity of the masonry ............................................................................. 160
5.3 STRUCTURAL MODELING............................................................................................ 160
5.4 DESIGN RESPONSE SPECTRUM .................................................................................. 160
5.5 SEISMIC CAPACITY EVALUATION OF EXAMPLE BUILDING .............................. 161
5.6 SEISMIC RISK ASSESSMENT OF REPRESENTATIVE BUILDINGS ........................ 170
5.7 CONCLUSIONS ................................................................................................................ 172
Chapter 6 SUMMARY, CONCLUSIONS AND RECOMMENDATIONS ................................ 190
6.1 SUMMARY ....................................................................................................................... 190
6.2 CONCLUSIONS AND RECOMMENDATIONS ............................................................. 190
6.2.1 Experimental work ...................................................................................................... 191
6.2.2 Proposed methodology ................................................................................................ 191
6.2.3 Seismic risk assessment of URBM building stock ...................................................... 192
6.3 RECOMMENDED FUTURE WORK ............................................................................... 194
VITA AUCTORIS ....................................................................................................................... 200

xxi
 List of Figures

Figure 2.1 Response of a masonry building subjected to the ground shaking. (Figure adopted from
http://www.iitk.ac.in/nicee/EQTips) .............................................................................................. 19
Figure 2.2 Influence of ties and diaphragms on the seismic behavior of a masonry building
(Tomaževic [To 99]): ..................................................................................................................... 19
Figure 2.3 Various types of failures observed in unreinforced masonry buildings under seismic
excitations. ...................................................................................................................................... 20
Figure 2.4 Typical diagonal tension shear cracks in an URM pier ................................................ 20
Figure 2.5 Typical cracks in URM buildings originating from the corners of openings .............. 21
Figure 2.6 Rocking/toe crushing failure in URM piers ................................................................ 22
Figure 2.7 Out-of-plane failure of URM walls due to the improper floor- wall connection,
aggravated by additional thrust from the roof trusses .................................................................... 23
Figure 2.8 Wedge type separation of masonry below roof purlins at the buildings’ corners ........ 24
Figure 2.9 Separation of orthogonal walls made of (a) stone masonry (b) brick masonry ....... 25
Figure 2.10 Diffusion of low quality masonry in the top courses of wall due to the local
interaction with the heavy roof ....................................................................................................... 26
Figure 2.11 Typical overturning of a gable in the out-of-plane direction ...................................... 26
Figure 2.12 Diffusion of external veneer in the double-leaf stone masonry walls due to the
absence of connection between the masonry courses across the transverse direction.................... 27
Figure 2.13 Failure of infill masonry walls due to the diagonal compressive forces transferred
from beam-column joint. ................................................................................................................ 28
Figure 2.14 Overturning of free- standing masonry walls in the out-of plane direction. .............. 29
Figure 2.15 (a) Typical overturning observed in boundary walls. Overturning of a slender
parapet wall was prevented, as shown in figure (b), due to proper bracing across the length. ...... 30
Figure 2.16 (a) Rocking in an URM pier (b) Possible toe-crushing after rocking. ...................... 31
Figure 2.17 Sliding shear failure in squat URM piers due to sliding of bed joints. ..................... 32
Figure 2.18 Diagonal tension shear failures in URM piers constructed with: (a) Strong
brick-weak mortar (b) Strong mortar-weak brick........................................................................ 33
Figure 2.19 Testing facility at National Building Institute, Ljubljana, Slovenia, employed to
conduct quasi-static cyclic tests...................................................................................................... 34
Figure 2.20 Various types of loading histories employed in quasi-static cyclic tests ................... 34
Figure 2.21 Diagram illustrating general concept of PsD testing technique. ........................ 35
Figure 2.22 Diagram illustrating concept of ‘ Sub-structuring’ in PsD testing ............................. 35

xxii
Figure 2.23 Shaking table installed at the Earthquake Simulation laboratory, Ministry of
Construction, Tsukuba-shi Ibaraki-ken, Japan ............................................................................... 36
Figure 2.24 (a) Shock Table facility (b) Signature of shock due to collision ................................. 36
Figure 2.25 Failure modes assumed by D’Ayala et al [DSOP 97] in URM buildings ................. 37
Figure 3.1 Typical masonry prism used for determining compressive strength of masonry......... 58
Figure 3.2 Typical arrangement of displacement dials gages for determining vertical
deformations in a masonry prism. .................................................................................................. 58
Figure 3.3 Testing setup used for conducting diagonal tension tests (a) Front view of prism
with LVDT installed for measuring horizontal elongation (b) Rear view of prism with
LVDT for determining vertical shortening..................................................................................... 59
Figure 3.4 Testing arrangement for carrying out triplet tests ........................................................ 60
Figure 3.5 Results of regression analysis for determining cohesion and coefficient of friction of
masonry from triplet tests ............................................................................................................... 61
Figure 3.6 Testing setup used for conducting in-plane quasi-static cyclic tests on URBM piers .. 62
Figure 3.7 Details of testing setup used for conducting quasi-static cyclic tests on URBM piers 63
Figure 3.8 Typical lateral displacement history imposed on the piers’ top end during quasi-static
cyclic tests ...................................................................................................................................... 64
Figure 3.9 Pier PIa after diagonal tension cracking....................................................................... 64
Figure 3.10 Multiple inclined cracks along each diagonal of pier PIb .......................................... 65
Figure 3.11 Diagonal tension cracks in pier PIc............................................................................ 65
Figure 3.12 Pier PIc at end of testing (Also note vertical cracks in the middle portion).............. 66
Figure 3.13 Spalling of masonry due to crushing and in pier PIc during the final stages of testing
........................................................................................................................................................ 66
Figure 3.14 Seperation of the diagonally cracked ‘wedge’ due to the crushing of masonry at the
bottom of pier PIc ........................................................................................................................... 67
Figure 3.15 Vertical splitting of masonry across the head joints on the right end of pier PIIa ...... 67
Figure 3.16 Excessive compressive stresses in the masonry at the bottom of pier PIIc caused
crushing .......................................................................................................................................... 68
Figure 3.17 Extensive sliding of damaged ‘wedge’, on the right end of pier PIIa, resulted in
terminating the test in spite of small degradation of lateral strength .............................................. 68
Figure 3.18 A closer view of pier PIIa. Extensive sliding of diagonally cracked wedge caused its
separation from rest of the pier. ...................................................................................................... 69
Figure 3.19 Separation of diagonally cracked ‘wedge’ on the left end of pier PIIb at the end of
testing ............................................................................................................................................. 69
Figure 3.20 Excessive crushing at the bottom end of pier PIIb resulted in ................................... 70
xxiii
Figure 3.21 A view of pier PIIc at the end of testing .................................................................... 70
Figure 3.22 Crushing of brick at the bottom of pier PIIc .............................................................. 71
Figure 3.23 Multiple inclined cracks along each loading direction in pier PIIIa. ......................... 71
Figure 3.24 Vertical cracks along both the ends of pier PIIIb ....................................................... 72
Figure 3.25 Vertical splitting of masonry and crushing of bricks in pier PIIIa. ............................ 72
Figure 3.26 Crushing of bricks at the top end of pier PIIIb........................................................... 73
Figure 3.27 Horizontal and vertical splitting of masonry on right end of pier PIIIa ..................... 73
Figure 3.28 Vertical splitting and crushing of masonry on the top and ........................................ 74
Figure 3.29 Pier PIVc at the end of test under a pre-compression of 61 psi ................................. 74
Figure 3.30 Branching of vertical crack in pier PIVc..................................................................... 75
Figure 3.31 Pier PIVc after testing under a pre-compression of 93 psi ........................................ 75
Figure 3.32 Vertical gap between top surface of pier PIVc and concrete beam at the top end load
........................................................................................................................................................ 76
Figure 3.33 Splitting of masonry in pier PIVc due to the concentration of................................... 76
Figure 3.34 Pier PIVa just before failure of epoxy bond between foundation .............................. 77
Figure 3.35 Sliding of bed joint in pier PIVa at third course from bottom .................................. 77
Figure 3.36 Pier PIVa after testing. Note damage to brick units and vertical cracks at the right end
........................................................................................................................................................ 78
Figure 3.37 Vertical splitting of masonry across all the head joints of pier PIVa ......................... 78
Figure 3.38 Pier PIVb just after the formation of diagonal tension cracks ................................... 79
Figure 3.39 Bed joint sliding in pier PIVb at fourth layer from top end of pier. Note
sliding in other courses and formation of multiple inclined cracks. ............................................... 79
Figure 3.40 Separation of diagonally cracked ‘wedge’ due to the formation of multiple inclined
cracks on the right end of pier PIVb. .............................................................................................. 80
Figure 3.41 Right end view of pier PIVb , .................................................................................... 80
Figure 3.42 Lateral load – displacement diagrams for pier PIc (a) Hysteresis loops (b) Envelope
of first cycle of loading................................................................................................................... 81
Figure 3.43 Lateral load – displacement diagrams for pier PIIa (a) Hysteresis loops (b) Envelope
of first cycle of loading................................................................................................................... 82
Figure 3.44 Lateral load – displacement diagrams for pier PIIb (a) Hysteresis loops (b)
Envelope of first cycle of loading .................................................................................................. 83
Figure 3.45 Lateral load – displacement diagrams for pier PIIc (a) Hysteresis loops (b) Envelope
of first cycle of loading................................................................................................................... 84
Figure 3.46 Lateral load – displacement diagrams for pier PIIIa (a) Hysteresis loops (b)
Envelope of first cycle of loading .................................................................................................. 85
xxiv
Figure 3.47 Lateral load – displacement diagrams for pier PIIIb (a) Hysteresis loops (b)
Envelope of first cycle of loading .................................................................................................. 86
Figure 3.48 Lateral load – displacement diagrams for pier PIVa tested at pre-compression of 93
psi (a) Hysteresis loops (b) Envelope of first cycle of loading..................................................... 87
Figure 3.49 (a) Hysteresis loop for pier PIVa tested at a pre-compression of 61 psi ................... 88
Figure 3.50 Lateral load – displacement diagrams for pier PIVb tested at pre-compression of 93
psi (a) Hysteresis loops (b) Envelope of first cycle of loading..................................................... 89
Figure 3.51 Comparison of diagonal tensile shear strength of piers obtained from experimental
work with the diagonal tensile shear strength of piers determined from Turnsek equation. .......... 90
Figure 3.52 Bi-linear idealization of URM piers based on equal energy principal. Vu = 0.9 Vmax
for shear failure [MC 97] ................................................................................................................ 90
Figure 3.53 Bi-linear idealized curves for pier PIc. p =10.69 and 8.54 for positive and negative
loading directions, respectively. ..................................................................................................... 91
Figure 3.54 Bi-linear idealized curves for pier PIIa. p =4.86 and 11.03 for positive and negative
loading directions, respectively. ..................................................................................................... 91
Figure 3.55 Bi-linear idealized curves for pier PIIb. p =5.16 and 9.09 for positive and negative
loading directions, respectively. ..................................................................................................... 92
Figure 3.56 Bi-linear idealized curves for pier PIIc. p =7.68 and 12.62 for positive and negative
loading directions, respectively. ..................................................................................................... 92
Figure 3.57 Bi-linear idealized curves for pier PIIIa. p =4.59 and 11.16 for positive and
negative loading directions, respectively........................................................................................ 93
Figure 3.58 Bi-linear idealized curves for pier PIIIb. p =7.03 and 1.21 for positive and negative
loading directions, respectively. ..................................................................................................... 93
Figure 3.59 Bi-linear idealized curves for pier PIVa. p = 8.71 and 7.24 for positive and negative
loading directions, respectively. ..................................................................................................... 94
Figure 3.60 Bi-linear idealized curves for pier PIVb. p =18.66 and 32.46 for positive and
negative loading directions, respectively........................................................................................ 94
Figure 3.61 Effect of pre-compression on displacement ductility of tested piers ........................ 95
Figure 3.62 Effect of Pre-compression on ultimate drift ratio of tested piers ............................... 95
Figure 3.63 Energy determination in one cycle of loading [Va 04] a. Dissipated energy b.
Input energy .................................................................................................................................... 96
Figure 3.64 Influence of the drift ratio on the coefficient of equivalent viscous damping (PI
series).............................................................................................................................................. 97
Figure 3.65 Influence of the drift ratio on the coefficient of equivalent viscous damping (PII
series).............................................................................................................................................. 97
xxv
Figure 3.66 Influence of the drift ratio on the coefficient of equivalent viscous damping (PIII
series).............................................................................................................................................. 98
Figure 3.67 Influence of the drift ratio on the coefficient of equivalent viscous damping (PIV
series).............................................................................................................................................. 98
Figure 3.68 Influence of the drift ratio on the coefficient of equivalent viscous damping (All
piers) ............................................................................................................................................... 99
Figure 3.69 Influence of the pre-compression on the coefficient of equivalent viscous damping.
σo = 61 psi and 93 psi for Piers belonging to PII and PIII series, respectively ............................ 100
Figure 3.70 Degradation of stiffness with the increase in drift ratio for PI series piers ............ 101
Figure 3.71 Degradation of stiffness with the increase in drift ratio for PII series piers ........... 101
Figure 3.72 Degradation of stiffness with the increase in drift ratio for PIII series piers .......... 102
Figure 3.73 Degradation of stiffness with the increase in drift ratio for PIV series piers .......... 102
Figure 3.74 Degradation of stiffness with the increase in drift ratio for all piers...................... 103
Figure 3.75 Influence of the pre-compression on the stiffness degradation. σo = 61 psi and 93 psi
for Piers belonging to PII and PIII series, respectively ................................................................ 104
Figure 4.1 Terminology. figure adopted from research work of Lang [La 02] with some
modification.................................................................................................................................. 142
Figure 4.2 Fictitious example building with rigid diaphragm ...................................................... 143
Figure 4.3 capacity curve of example fictitious building, shown in figure 4.2, along x direction
...................................................................................................................................................... 144
Figure 4.4 Elastic design response spectrum recommended by Building Code of Pakistan [BCP
07]................................................................................................................................................. 144
Figure 4.5 Free-body and bending moment diagram of a masonry pier (restrained on the top and
bottom end) under the action of lateral force V............................................................................ 145
Figure 4.6 (a) Front elevation of example building as shown in figure 4.2b, (b) Variation of lateral
displacements in wall # 2 at the verge of ‘yielding’ in the corresponding pier at lowest story (c)
Variation of lateral displacements in wall # 2 at ultimate conditions in the corresponding pier at
lowest story................................................................................................................................... 146
Figure 4.7 (a) Free-body diagram of a unit length of a face-loaded wall ..................................... 147
Figure 4.8 A typical graph between acceleration and corresponding mid-span deflection for an
URM wall loaded in the out-of-plane direction............................................................................ 148
Figure 4.9 Determination of acceleration corresponding to the failure in the linear elastic system
corresponding to a non-linear system using equal energy principle............................................. 148
Figure 4.10 Variation of response acceleration with height [PP 92] ..................................... 149

xxvi
Figure 4.11 (a) Plan and (b) Elevation of the model building tested at University of Pavia, Italy.
All the mentioned dimensions are in meters. The arrows show points of application and direction
of lateral forces [MKC 95]. .......................................................................................................... 150
Figure 4.12 Displacement history imposed on the second floor of the model building model
tested at the University of Pavia [MKC 95] ................................................................................. 151
Figure 4.13 Comparison of results from proposed methodology with the experimental results of
model building tested at the University of Pavia, Italy ................................................................ 151
Figure 4.14 Plan of the model building tested at the Georgia Institute of Technology, USA [Yi
04]................................................................................................................................................. 152
Figure 4.15 Elevation views of building [Yi 04] shown in figure 4.14: (a)
Walls A & B (b) Wall 1 ......................................................................................................... 153
Figure 4.16 Typical displacement history applied during the test on the model building tested at
the Georgia Institute of Technology, USA [Yi 04] ...................................................................... 154
Figure 4.17 Comparison of results from proposed methodology with the experimental results of
model building tested at the Georgia Institute of Technology, USA ........................................... 155
Figure 5.1 Cross section of a typical foundation, used in Pakistan, to support up a brick masonry
building up to double-story height ............................................................................................... 174
Figure 5.2 Ground and first floor plans of example building ....................................................... 175
Figure 5.3 Front elevation of example building shown in Figure 5.2 ........................................ 176
Figure 5.4 Capacity curve of example building, along with capacity curves of component walls,
in the shorter direction .................................................................................................................. 177
Figure 5.5 Capacity curve of example building, along with capacity curves of component walls,
in the longer direction,.................................................................................................................. 177
Figure 5.6 Variation of mid height deflection in wall No. 7 with corresponding acceleration, aop/
g .................................................................................................................................................... 178
Figure 5.7 Acceleration-mid height displacement curves for wall No.7 (nonlinear system) and
equivalent linear elastic system, based on principle of equal energy ........................................... 178
Figure 5.8 Fragility curves for single-story buildings for various performance levels .............. 179
Figure 5.9 Fragility curves for double-story buildings for various performance levels ............. 179
Figure 5.10 Fragility curves for all buildings for various performance levels houses ............. 180

xxvii
 List of Tables

Table 2.1 Various methods of evaluating seismic vulnerability of buildings’ stock [La 02] ......... 38
Table 3.1 Proportions of stone dust aggregates used in mortar .................................................... 105
Table 3.2 7-day compressive strength of trial mortar cubes........................................................ 105
Table 3.3 IRA tests on brick units ............................................................................................... 106
Table 3.4 Compressive strength of brick units (psi) ..................................................................... 106
Table 3.5 Flexural tensile strength of brick units ......................................................................... 107
Table 3.6 Compressive strength of CSK 1:4:4 mortar ................................................................. 108
Table 3.7 Compressive strength of brick masonry prisms ........................................................... 109
Table 3.8 Modulus of elasticity of brick masonry prisms ............................................................ 109
Table 3.9 Principal tensile strengths of masonry prisms .............................................................. 110
Table 3.10 Modulus of rigidity of masonry prisms determined from diagonal tension tests ....... 110
Table 3.11 Results of triplets tests ................................................................................................ 111
Table 3.12 Characteristics of pier specimens ............................................................................... 111
Table 3.13 Typical forces and displacements applied on piers during cyclic tests ...................... 112
Table 3.14 Lateral Forces in masonry piers at various levels of distress during testing .............. 113
Table 3.15 Displacement ductility factors for tested pier specimens ........................................... 114
Table 3.16 Modulus of rigidity of masonry determined from quasi-static cyclic tests on piers .. 115
Table 3.17 Lateral forces and corresponding drifts at various stages of loading during the test . 116
Table 4.1 Building performance levels recommended by FEMA 356 [FEMA 356] guidelines .. 156
Table 4.2 Values of ηo for multi story frames with j stories [PP92]............................................ 156
Table 4.3 Values of correction coefficient η1 for different beam stiffnesses [PP 92] ................ 157
Table 5.1 Geometrical properties of structural members of the example building along shorter side
...................................................................................................................................................... 181
Table 5.2 Inflection points in the piers of the example building along shorter side.................... 182
Table 5.3 Failure modes and lateral strengths of the piers of the example building along shorter
side ............................................................................................................................................... 182
Table 5.4 ‘Yield’ displacements, ultimate displacements and displacement ductility factors of
the piers, wall elements and walls of the example building along shorter side ............................ 183
Table 5.5 Lateral displacements corresponding to various performance levels in the example
building along shorter side ........................................................................................................... 183
Table 5.6 Base shear and corresponding lateral top displacement in the equivalent elastic MDOF
system for various performance levels of the example building along shorter side ..................... 184

xxviii
Table 5.7 Weight of slab supported by various structural walls in lb .......................................... 184
Table 5.8 Determination of parameters required to calculate Γ, mE and hE for the example building
...................................................................................................................................................... 185
Table 5.9 Horizontal ground acceleration corresponding to various performance levels of the
example building along shorter side ............................................................................................. 185
Table 5.10 Geometry, failure modes and displacement based characteristics of the walls along the
longer side of example building ................................................................................................... 185
Table 5.11 Horizontal ground acceleration corresponding to various performance levels along the
longer side of example building ................................................................................................... 186
Table 5.12 Comparison of Horizontal ground accelerations corresponding to various performance
levels of the building in both the orthogonal directions of the example building ........................ 186
Table 5.13 Values of aop/g and ∆op for various levels of compressive strains at the mid height for
the wall No. 7 in the out-of-plane direction.................................................................................. 187
Table 5.14 ag/g values corresponding to various performance levels in single- story
buildings ...................................................................................................................................... 187
Table 5.15 ag/g values corresponding to various performance levels in the double- story
buildings ...................................................................................................................................... 188
Table 5.16 Average values of ground acceleration corresponding to the various performance
levels for various categories of buildings’ stock ......................................................................... 189
Table 5.17 Effect of seismic zoning on the ground acceleration corresponding to the various
performance levels in the example building ................................................................................. 189
.

xxix
 Chapter 1
INTRODUCTION

1.1 PROBLEM BACKGROUND AND DEFINITION

An earthquake jolted the Kashmir and Northern Areas of Pakistan on October 8, 2005. It was one of
the deadliest earthquakes (with Mw = 7.6) according to the number of fatalities in the history of indo-
Pakistan subcontinent. More than 70,000 people died; 80,000 left injured, 2.5 million people were
displaced and 3.3 to 3.5 million people were affected. The total economic loss of the earthquake was
estimated to be US$5.2 billion. According to the Asian Development Bank and the World Bank
4,55,000 buildings were partially or fully damaged. Most of the damaged and collapsed buildings were
constructed with unreinforced stone and brick masonry. However, the performance of unreinforced
brick masonry (URBM) was found to be much better than unreinforced stone masonry. Further details
on the performance of masonry structures during this earthquake are discussed by Javed et al [JKM
08].

The disaster demanded for taking the appropriate measures in order to minimize losses against
future seismic activities. However, in order to reduce potential damage to structures, a comprehensive
seismic risk assessment, a combination of seismic hazard and seismic vulnerability, is needed at least
for major towns. The seismic hazard depends upon geology of the area and, therefore, cannot be
controlled. However, seismic vulnerability of structures against future seismic activities can be
reduced by implementing the remedial measures, once the existing seismic vulnerability of buildings
located in the area under interest is known. It is, therefore, necessary to assess the seismic risk of
existing building stock of Northern Pakistan. This dissertation is primarily focused on seismic risk
evaluation of URBM buildings of Northern Pakistan.

1.2 OBJECTIVES
The primary purpose of this research work is to assess the seismic risk of URBM buildings of
Northern Pakistan constructed in stone dust mortar. However, seismic risk assessment of buildings is
not possible without having quantitative information regarding the seismic performance of buildings
at component level. In this regard, walls, being the most critical parts of a masonry building, require
more attention. Although, significant experimental work has been done in the past on structural
elements made with concrete and steel leading to the well-defined procedures regarding the mechanics
of concrete and steel structures, no such clear understanding exists about masonry structures. This is
mainly due to the reason that most of the experimental research on material side is done in developed
countries where structures are built by using concrete and/or steel.

1
 The research work, although not so extensive as in the case of concrete and steel, on masonry
structures has led to development of some fundamental analytical tools for predicting lateral strength,
and performance based design concepts of URM structures [FEMA 356]. However, it is a known fact
that one of the most important factors that significantly influence the behavior of masonry is mortar. A
change in either the constituents and/or proportions of the cementitious material and fine aggregate
induces appreciable changes in the overall behavior of masonry systems. It was, therefore, required to
study the behavior of URBM using stone dust mortar. The mortar contains stone dust as fine
aggregate, which is obtained from the crushing industry plants producing coarse aggregates. The stone
dust is locally known as ‘khaka’. The name seems to be derived from the word khak, which means
clay.

1.3 OUTLINE OF THE DISSERTATION

The dissertation is organized in six chapters.

Chapter 2 starts with general discussion on the response of masonry buildings subjected to seismic
excitations. Detailed discussion is also done on various possible local and global failure modes in
unreinforced masonry (URM) buildings as well as simplified procedures for predicting the failure
mode and the associated strength. Various techniques employed for simulating seismic forces in
experimental work on masonry structures also discussed. A brief discussion is also made on
methodologies for determining seismic vulnerability of buildings

Chapter 3 covers the details of in-plane quasi-static cyclic tests carried out on URBM piers
constructed in stone dust mortar.

Various mechanical properties such as compressive strength, modulus of elasticity and modulus of
rigidity are determined for brick masonry constructed in stone dust mortar. Similarly, properties of
masonry piers related to seismic performance such as displacement ductility factors, coefficient of
equivalent viscous damping and stiffness degradation are also evaluated. Drift ratios are also
recommended for URBM piers constructed in stone dust mortar for various performance levels such as
operational, immediate occupancy, life safety and collapse prevention in piers.

A methodology for seismic assessment of URM building is recommended in Chapter 4. The

recommended method estimates the lateral strength and ultimate displacements of URM buildings
with reasonable accuracy when compared with the results of experimtal work on full scale buildings
[MKC 95 and Yi 04].

In Chapter 5, material properties of brick masonry constructed in stone dust mortar, based on the
experimental work carried out as a part of this work and other works carried out on the stone dust
brick masonry [AliQ 04, AliM 06, Ban 06 and Badr 06], are recommended for Peshawar. Seismic
assessment of a double-story URBM example building is also carried out. Finally, seismic risk
2
assessment is carried out for 31 representative buildings. The buildings’ stock consisted of 17 single-
story and 14 double-story buildings. Fragility curves, based on statistical analysis of the result of
seismic assessment of buildings, are drawn for various performance levels.

Chapter 6 presents the conclusions and recommendations based on the work explained in Chapter 3
through 5. Finally, recommendations for future research are outlined.

3
 Chapter 2
LITERATURE REVIEW AND THEORITICAL BACKGROUND

2.1 INTRODUCTION
In this chapter, theoretical background, related to the material presented in rest of the document, is
briefly discussed. The chapter starts with brief discussion on response of masonry buildings subjected
to seismic excitations. Various global and local failure mechanisms in URM buildings are discussed in
detail in Sections 2.3 through 2.6. Simplified criteria for estimating lateral strength of URM piers are
presented in Section 2.7.

Various experimental techniques employed for simulating earthquake effects on structural

elements/structures as well the quasi-static testing technique selected for conducting experimental
work in this study are discussed in Sections 2.8 and 2.9, respectively.

Finally, various methodologies for assessing the seismic vulnerability of buildings and the selection
of a particular method for determining vulnerability assessment of URBM buildings’ stock for various
cities of Northern Pakistan are discussed in Sections 2.10 and 2.11, respectively.

2.2 RESPONSE OF MASONRY BUILDINGS SUBJECTED TO SEISMIC EXCITATIONS

Masonry buildings differ from other types of buildings by the arrangement and relative stiffness of
lateral load resisting elements. In case of masonry buildings, these structural elements are, normally,
interconnected orthogonally to each other with relatively flexible diaphragms.

The resisting mechanism of a masonry building is summarized in Figure 2.1. As can be seen from
this figure, during a seismic activity ground forces from foundations are transferred to the in-plane
walls, which are the stiffest components of the building. The in-plane walls, if properly connected to
diaphragm, transfer these forces to diaphragm, which ultimately transfer the forces to the attached
walls in the out-of-plane direction. The diaphragm acts as a deep beam, simply- supported at the ends.
The diaphragm deflects, under the action of transmitted inertial forces, in an amount which depends
upon the in- plane stiffness of the diaphragm. In case of a wooden diaphragm, the flexibility of the
diaphragm can result in excessive deflection, causing damage to the walls connected to the diaphragm
in the out-of-plane direction.

Separation of the walls transverse to the in-plane walls can also occur in the absence of proper
connection between the diaphragm and the connecting walls. In such a case, the out-of-plane walls
vibrate independently, thus increasing the vulnerability of the out-of-plane bending failure. The
behavior of masonry buildings, under lateral forces, with various types of floor systems and the floor-
wall connections can be seen in Figure 2.2.

4
2.3 STRUCTURAL DAMAGE IN MASONRY BUILDINGS
Structural damage, based on observation and on correlation with past experience, can be classified
according to the possible inferred causes.

The in-plane wall failure modes are usually related to an overall global response. Such type of
failure can be distinguished from local collapse mechanism, which can occur due to intrinsic
vulnerability of the material or unsuitable structural details.

Structures with proper connections between the orthogonal walls as well as between the walls and
the floors can exploit the in-plane resistance of walls, allowing the building to resist the seismic action
as a whole, and the damage associated to this kind of response is generally related to the in-plane
response of the masonry walls. Depending on the geometry and position of the walls and on the
distribution of the openings, these damage tend to be located in specific portions of masonry such as
masonry piers and spandrel beams.

In the presence of low-quality construction and inadequate structural details, the response of the
masonry building tends to be governed by local phenomena and damage mechanisms. In poor quality
masonry walls, made of two leaves of irregular stones and no transverse connection offered by the
through elements, a typical collapse is the out-of-plane crumbling of the external veneer. Similarly,
absence of good connections between floors and walls or roofs and walls, and absence of the out-of-
plane restraints such as the ties or the ring beams, absence of proper quoins, give rise to out-of-plane
overturning of single walls.

In the following sections, several examples of damage observed in masonry structures are presented
and subdivided according to the different failure modes. Detailed discussion on the topic can be found
in the recently published paper of the author [JKM 08]. Some of the typical damage in URM buildings
is summarized in Figure 2.3.

2.4 IN-PLANE DAMAGE

2.4.1 Shear cracks

In-plane diagonal cracks and X-diagonal cracks, as shown in Figure 2.4, result from lateral forces
acting in the plane of the walls. Depending on the level of drift demand, the damage can be moderate
and easily repairable, or severe to the extent that the buildings are usually unfit for further use.
However, the consequences of such failures to the residents using the buildings are significantly less
serious than in the case of the out-of-plane wall overturning.

5
2.4.2 Crack damage at openings
Figure 2.5 shows an example of crack damage at openings. These can be a result of both the in-plane
shear forces and the out-of plane flexure of the wall, and are initiated from concentration of the
stresses produced at the corners of the openings. They are not particularly serious unless the relative
displacement across the cracks is large, in which case instability of the section of the wall above the
opening becomes an issue.

2.4.3 Pier flexural-rocking failure

Pier flexural-rocking failure cracks, as shown in Figure 2.6, are produced in case of slender piers
(portion of the wall between two openings). Failure initiates with large flexural cracks developing at
the bottom and the top of the pier. As the displacement increases, the pier deforms as a ‘rigid body’
rotating about the compressed toe.

Rocking may occur in piers having relatively higher aspect ratios ( i.e., height-to-length ratio) with
lower magnitudes of compressive stresses acting over piers.

2.5 OUT-OF-PLANE AND LOCAL DAMAGE MECHANISMS

2.5.1 Lateral thrust from roofs

Out-of-plane lateral thrust at roof level due to inclined roofs, added to inertial forces, can become a
significant cause of collapse of masonry structures. A number of cases were observed in 2005 Kashmir
earthquake [JKM 08] where masonry buildings supporting sloping roofs were either collapsed or
badly damaged while nearby structures having the same type of structural layout and construction
materials but non-thrusting roof diaphragms, experienced less damage.

In addition to the roof lateral thrust, another observed reason for out-of-plane failures, shown in
Figure 2.7, is the lack of connection between the walls and the supporting roof. A number of such
types of failures were observed in buildings, during 2005 Kashmir earthquake [JKM 08], where
wooden roof trusses were just resting on the walls, thus providing no out-of-plane restraint. The
supporting wall(s) thus failed in the out-of-plane bending as the building was unable to develop ‘box
action’ against the lateral vibrations induced by the earthquake.

2.5.2 Wedge- type separation at the top of wall junctions due to lateral thrust from the
supporting roof truss
Such types of localized failures normally occur in the masonry walls supporting roofs inclined in the
both horizontal directions. Masonry gets separated, as shown in Figure 2.8, in the form of wedges
below the roof level due to thrust from roof purlins, added to the inertial forces. Such failure
mechanisms are mostly observed in case of openings close to the corner.

6
2.5.3 Lack of connection between the orthogonal walls
Such type of failure results from out-of-plane vibration of the walls. As the wall bends outward, the
intersecting perpendicular walls offer a restraint that relies on tensile strength of masonry. The
orthogonal walls are separated, as shown in Figure 2.9, when the outward force exceeds the tensile
strength of masonry. Intersection of the orthogonal walls is also characterized by the shear stresses due
to the flange-web action that can further facilitate cracking.

2.5.4 Damage at wall-to-roof connections

Damage at wall-to-roof connections occurs as a result of the forces transmitted between the walls and
roof, mostly due to out-of plane horizontal excitation of the walls but also due to vertical ground
accelerations, which could be very high in localities very close to the fault rupture. The top of the wall
may slip out from underneath the roof and/or crush under the dynamic loading. Such types of failure
particularly occur in the case of poor quality masonry (rubble stone) and relatively heavy and rigid
roofs, not connected properly to the supporting walls. Also, in the case of light roofs, the friction under
the roof bearing may not be sufficient to avoid slippage. Such type of damage is shown in Figure 2.10

2.5.5 Out-of-plane overturning of gables

Gables, due to an absence of vertical load and lateral restraint essentially behave like cantilever walls.
The like hood of such types of failure increases with the increase in the height of the centroid of gables
above the ground level. Example of such failure is shown in Figure 2.11

2.5.6 Diffusion of stone masonry external veneer

Diffusion of stone masonry veneer can occur in double-leaf stone masonry walls. In most cases, no
through stones are provided to properly connect the leaves across the thickness. Masonry stones get
separated from each other when inertial forces in the out-of-plane directions, due to seismic
excitations, exceed the shear strength of masonry. Examples of such types of failures are shown in
Figure 2.12.

2.6 DAMAGE TO MASONRY NON-STRUCTURAL ELEMENTS

2.6.1 In-plane failure of infill walls

In-plane failure of infill walls occurs when the infill walls either have a small thickness or made of
poor quality masonry. In some cases, poorly detailed columns also fail in the proximity of the beam-
column joints due to transfer of horizontal forces from the infill walls to the adjacent columns. Such
types of failures are shown in Figure 2.13.

7
2.6.2 Out-of-plane failures of infill walls
The acceleration in a structure increases with an increase in height. If not anchored properly, the
nonstructural components can overturn in an out- of- plane direction as shown in Figure 2.14.

2.6.3 Failure of the boundary and parapet walls

The boundary and parapet walls are more susceptible to overturning due to the cantilever type
boundary conditions. Such walls easily overturn if not braced properly. A lot of casualties due to such
type of failure occurred in the 2005Kashmir earthquake. Failures of such type are shown in Figure
2.15.

2.7 SIMPLIFIED FORMULATIONS FOR PREDICTING IN-PLANE FAILURE

MODES IN URM WALLS
Masonry is a non-homogeneous and anisotropic composite structural material, consisting of masonry
units and mortar. Therefore, its behavior is complex. The accurate prediction of lateral load carrying
capacity of the URM walls is difficult because of the complex brick-block mortar interaction.

In addition to sophisticated failure theories for predicting failure modes for given states of stresses
[HD 81, MM 82, Pa 82 and Ga 89], there exist analytical models which predict lateral strength of
URM piers, corresponding to an in-plane failure mode, with reasonable accuracy and are discussed in
the following sub-sections.

2.7.1 Rocking/toe crushing failure

As already discussed, rocking failure initiates with large flexural cracks developing at the bottom and
the top of a pier. With an increase in the lateral displacement, the pier deforms as a ‘rigid body’ by
rotating about the compressed toe. On the other hand, toe crushing is defined as a compressive failure
of masonry occurring at the toe of the pier. Toe crushing is typically observed after rocking
deformations.

Rocking and toe crushing may occur in piers having relatively higher aspect ratios with medium to
high magnitudes of compressive stresses acting over piers. Both types of failures are shown in Figure
2.16. Toe crushing is relatively brittle mode of failure if no rocking occurs before failure.

Lateral force corresponding to this failure mode can be determined with the following equation,
which can be derived [MC 97] by using the moment equilibrium of the pier:

l .t σ σ l .t σ σ
V . 1 . 1 .
h 2 k. f α 2 0.85f

where:

8
 Vr = Shear force corresponding to toe crushing/rocking failure;
lp and t = Length and thickness of the pier, respectively;
σo = Vertical pressure acting on the top of the pier;
fm = Compressive strength of the masonry;
ho = Effective height of the pier (height of zero moment). In case of cantilever pier ho= height of
pier, hp;
αv = Shear ratio of the pier = M/Vlp = ho/lp = ψ′ hp/lp where ψ′ is a boundary condition factor =1.0
for cantilever pier and 0.5 for piers with both ends fixed; and
k = Equivalent stress block factor which can be taken as 0.85.

2.7.2 Sliding shear failure

Such type of failure occurs in piers having low aspect ratios, lower compressive stresses acting over
piers and/or poor bond between brick and mortar. A typical sliding shear failure is shown in Figure
2.17. This type of failure is quite ductile and a considerable amount of energy is dissipated without
significant damage due to the sliding of wall over separated mortar joint. Normally, Coulomb type
criterion [MC 97] is used to predict this type of failure mode
τ c µ. σ .

where τu = ultimate shear strength, c = cohesive strength of masonry, µ = coefficient of friction

between bricks and mortar, and σ n = normal compressive stress.
Magenes and Calvi [MC 97] extended the Coulomb criteria to determine the shear force
corresponding to sliding shear failure, Vsl, in URM piers and derived the following equation:

1.5c µ. σ
V c. α l .t .
1 3
σ
P P
where V
6α 2α

where P = total vertical load acting on a pier. The other various right hand side terms have been
defined in Section 2.7.1

2.7.3 Diagonal tension shear failure

Diagonal tension cracks are approximately sloped at 45o and their path depends upon the relative
strength of bricks and mortar. In the case of strong bricks and weak mortar, the cracks pass through
head and bead joints and are “stair stepped” as shown in Figure 2.18a. In the case of weak bricks and
strong mortar, they can pass through the bricks as shown in Figure 2.18b. Usually a very small load is

9
resisted after the formation of the diagonal cracks. Diagonal tension failure is a brittle failure mode
leading to the sudden stiffness and strength degradation. However, if out-of-plane failure is prevented,
the damaged wall is usually able to prevent global failure of structure. Such type of failure occurs in
piers having low to medium aspect ratios and medium to high compressive stresses acting over the
walls. The shear strength corresponding to this failure mode is determined by using the equation
proposed by Turnšek and Sheppard [TS 80]. The equation is based on the assumption that diagonal
cracking takes place when the maximum principal stress at the center of the wall, due to the
combination of the horizontal and vertical loads, exceeds the tensile strength of masonry. The
horizontal shear force in an URM pier corresponding to the opening of shear cracks, Vd, recommended
by Turnšek and Shepherd [TS 80] is as follows:

f σ
V 1 l .t .
b f

where:
Vd = Lateral force corresponding to diagonal tension failure;
ftu = Principal tensile strength of masonry;
b = Shear correction factor defining the distribution of horizontal shear stresses along the section;
b = 1 for hp/lp ≤ 1.0, b = hp/lp for 1 < hp/lp < 1.5, and b = 1.5 for hp/lp ≥ 1.5 [BT 84]; and
σo = Vertical pressure acting on the top of pier.

2.8 LABORATORY TESTING TECHNIQUES RELATED TO EARTHQUAKE

ENGINEERING
Although the capacity and response of a structure/structural element can be predicted by using
analytical methods, the confidence that can be placed in the results is limited to a great extent by the
uncertainties associated with the simplified modeling processes and nonlinear behavior of material.
Due to these reasons, experimental testing remains to be the most reliable way of evaluating the
inelastic behavior of structural systems and to devise structural details to improve their seismic
performance.
Various methods for simulating earthquake effects on structures are described in this section.
Selection of a particular testing technique depends both on the aims of testing and available budget.
There exist basically two types of tests, which are conducted for improving an understanding of the
structural behavior. One type of test, known as a ‘proof test’, is conducted for the purpose of verifying
the global response of a complete structure. A second type of test, known as a ‘basic test’, is generally
carried out on a structural element or sub-assemblage.

10
 Basic tests are conducted for understanding the progress of damage evolution. Such tests are
also carried out to obtain the information needed to model the behavior of the element in a computer
program. On the other hand, proof tests are generally needed to verify the adequacy of a complete
structure, or of a particular construction method or of a design method. Various types of testing
techniques employed for carrying out basic and proof tests are discussed in the following sub-sections.

2.8.1 Quasi-static tests

The quasi- static cyclic tests can be used to conduct both basic and proof tests. In quasi-static
tests, loads and/ or displacements are applied at slow rates. Normally, such types of tests are carried
out to study structural performance of structures and structural members such as the rate of
propagation of cracks, hierarchy of collapse and associated levels of damage, etc. A typical setup for
quasi- static testing installed at the National Building Institute in Ljubljana, Slovenia is shown in
Figure 2.19.
Quasi-static tests are performed by imposing predefined displacement or force histories on the
testing specimen. Different types of displacement histories employed for conducted quasi-static tests
are shown in Figure 2.20.
The slow loading rate during the test has the advantage of providing an insight regarding the
behavior of a structure/structural member in the post-yielding regime. However, the associated
disadvantage is that the effects of acceleration-dependent inertial forces and velocity-dependent
damping forces are eliminated, which can be significant for some structural types.
Similarly, by using the jacks and actuators the external actions are ‘lumped’ on the structure.
These actions try to simulate the inertial forces that are developed due to mass on the structure.
Consequently, the technique is not useful for structures possessing distributed mass such as hydraulic
structures. However, many structures can be adequately modeled with lumped masses in a discretised
manner and therefore are not affected by this restriction.

2.8.2 Pseudo-dynamic (PsD) tests

The concept of pseudo-dynamic test (also called computer-actuator on-line test or hybrid test)
was originated in Japan by Takanashi [Ta 75]. Since then, many researchers [TN 87, MS 89]
developed the concept and verified the suitability of this method.
Quasi-static tests are simple, relatively inexpensive, and do not require very special type of
apparatus. However, the displacement history has to be defined before the test, which is the main
limitation of this testing technique. Similarly, the applied cyclic displacement history may not cover
the range of displacements, which the structure would undergo under dynamic action.
The basic concept of pseudo-dynamic test is that the dynamic response is computed using the
experimental result in each time step. During the analysis process, the computer calculates the
structural response (displacement) in a time step. Inertial and damping forces, required during the
11
analysis process, for the solution of the equations of motion are modeled analytically. The computer,
after calculating structural displacement at a specific time step, electronically provides this result to the
actuator system. In the experimental process, actuator control system imposes the calculated
displacement and then measures and returns the restoring force, R(t), to the computer. With the
measured data, the computer can calculate the response in the next time step. With this feedback
procedure, the nonlinear inelastic dynamic response can be obtained without shaking table test
devices. The flow of this feedback is shown in Figure 2.21.
The method has difficulty in idealizing infinite degrees of freedom as a few degrees of
freedom. However, it enables a dynamic test with a static test device, which attracts many researchers.
The process automatically accounts for the hysteretic damping, due to inelastic deformation and
damage to the structural materials, which is usually the major source of energy dissipation. Inertia
forces are not experimentally produced and are modeled numerically. This eliminates conducting the
test on a real time-scale, and allows very large models of structures to be tested with only a relatively
modest hydraulic power requirement.
Another major advantage of this technique over the quasi-static testing is the use of a special
procedure of pseudo dynamic test known “sub-structuring”. Taking advantage of this technique,
researchers can test only a part of the structure. The rest of the structure can be analytically modeled
on a host computer. e.g., one can model a bridge deck analytically on a host computer and can carry
out the test on the bridge piers by pseudo dynamic testing facilities (see Figure 2.22). This saves
substantial amount of cost and time related to testing work.
A major drawback of PsD testing technique is that a lot of time is consumed to conduct the
experiment. This is mainly due to the time required by computer hardware to solve the equations of
motion for determining the displacements and by the control system to execute the calculated
displacements. Simulation of an earthquake, lasting 20 seconds, takes hours of time with this
technique. As a result, the response determined by this test of the structures that are sensitive to
loading rate (such as masonry) becomes questionable.
Another drawback is, that due to lumped mass idealization, the testing method is not adequate
for structures with distributed mass, e.g., hydraulic structures. To test such types of structures, a
refined spatial discretization would be required resulting in a large number of actuators. This reduces
the effectiveness of testing for such structures. The applicability of this method also relies on the
appropriate assignment of damping properties. It has been observed [SM 85] that the use of constant
damping matrix based on the elastic properties of the system resulted in unpredictable results.

2.8.3 Shaking table tests

Shaking tables are composed of a rigid platform, which can be moved along one or more axes,
by a number of hydraulic actuators, to produce a required base motion. Shaking tables can be
classified according to the number of degrees of freedom along which the motion can be produced.
12
The most sophisticated shaking tables can reproduce a motion defined by six degrees of freedom such
as shown in Figure 2.23.
Shaking tables are the best means to induce any required base motion in a model structure. It
is the closest mean of simulating the earthquake loads on a model structure. If the model structure is
properly scaled, actual conditions in term of actions and displacement in a prototype structure for a
specific earthquake history can be reproduced.
One of the major advantages of shaking table tests is the elimination of the effect of strain
rate, which is always a major source of concern in quasi-static and pseudo-dynamic testing. Similarly,
problems such as rate and pattern of applied displacement history also get eliminated, variation in
which can some times cause drastic changes in the results in a quasi-static test.
The capacity of shaking tables is often reduced due to hydraulic power limitations. This
constitutes the major limitation in most of the available shaking tables. Consequently, the weight of a
model may be limited according to scale laws due to the capacity of shaking table. Scaling may result
in poor representation of the behavior of materials and connections.
In case of reinforced concrete structures, scaling may result in changes of the mechanical
properties of the rebars, and the bond mechanism would be distorted. In case of masonry structures, it
would be impossible to control the change in mechanical properties of masonry by scaling the size of
bricks and thickness of mortar. Increase in bond strength is reported as a major problem in case of
scaled models of masonry buildings. The scaling phenomena change the cracking pattern and can
result in unrealistic failure mechanisms. Even in the case of steel structures, scaling may result in
substantially different behavior of bolted or welded connections.
The use of shaking table is an accurate experimental technique when applied to small or light
specimens, while it suffers severe limitations whenever the structure has to be scaled to be
accommodated by the table. Another problem, which is encountered in shaking tables, especially with
a single degree of freedom, is to control movement in a non-desirable direction. Such sort of problem
can cause excessive error in the result of in-plane testing of walls due to movement in transverse
direction. A slight eccentricity in the transverse direction can cause out-of-plane bending, causing a
drastic change in results, especially in URM walls.

2.8.4 Shock table dynamic test

In the absence of a shaking table, some laboratories have developed simple simulators for dynamic
testing of structures. One such kind of a facility, named as the shock table, is installed at the University
of Roorkee, India and was developed by Keightly [Ke 72]. The shock table facility was basically
developed for conducting dynamic tests on a low grade masonry house extended up to the weight of
20 ton at a considerably low cost. The arrangement, as shown in Figure 2.24, comprises of:
(i) a track, (ii) a shock table, (iii) dead Load striking wagons, and (iv) winch mechanism for
pull wagons.
13
 Ten helical coil compression springs are mounted around pipe pieces and welded on each end
of the platform to help moderate the impact. The loaded wagons are placed on the track on both sides
of the shock table. The loaded wagon gives impact through springs when allowed to roll down the
gentle incline. The shocking table, due to transferred momentum, collides with the other dead load
wagon, which remains temporarily at rest. A single shock from the end wagon imparts a half -sine
pulse to the central wagon. When another wagon is used to take the reaction, it imparts another half-
sine pulse from the rebound. In this way, one impact of the end wagon can produce a series of half-
sine pulses.
The shock table motion is basically an impulse type of motion with characteristics such as low
duration, high base acceleration and high frequency content against the actual ground motion. It is
difficult to extrapolate structural behavior under real earthquake motion from the shock table tests. The
facility is suitable only for studying the relative merits of different resistance measures in structural
models and for conducting feasibility studies on new concepts for earthquake resistance.

2.9 BASIS FOR ADOPTED EXPERIMENTAL APPROACH

The two major decisions with regard to the experimental work performed as a part of this study
and discussed in Chapter 3, were:
1. To conduct tests on piers (structural elements) instead of building (structure).
2. To employ a quasi-static testing technique.
The Reasons for the first decision were that a project for seismic risk assessment of masonry
buildings needs to be focused on testing of structural components in order to understand the
following fundamental issues:
1a. The types of failure mechanisms of prevailing interests for seismic behavior.
1b. Main parameters (material properties, geometry and boundary conditions that influence the
type of failure mechanism).
1c. The influence of the type of failure mechanism on the cyclic behavior of walls.
The quasi-static testing was adopted due to the following reasons:
2a. The seismic risk assessment of a building stock, as discussed in Chapter 5, is based on a
performance- based concept for which one must have an understanding of the hierarchy of the
failure of structural systems. Such information would not have been possible with shaking
table tests.
2b. Several researchers [ PA 90 ] have pointed out that masonry is very sensitive to rate effect and
masonry elements and structures undergo more damage due to a propagation of cracks if the
applied force or displacement is held constant under quasi-static testing. For example, Paulson
and Abrams [PA 90] tested two identical three-storey buildings constructed with reinforced
hollow concrete blocks. One of these buildings was tested on a shaking table and the

14
 displacements at the top floor were recorded. A second model was tested under a quasi-static
loading environment with displacement history recorded from shaking table tests was applied
to the floor through an actuator. After the test, the damage to both the buildings was
compared. It was observed that the model tested under quasi-static loads was severely
damaged as compared to the other model.

2.10 SEISMIC VULNERABILITY ASSESSMENT OF BUILDINGS

Experimental testing may be adequate to determine the seismic performance of a single building.
However, it is not feasible to determine the seismic performance of a building stock, located in a city,
by experimentally testing their representative models. Therefore, seismic performance of a building
stock located in a city is evaluated by using a technique known as seismic risk assessment.
Seismic risk assessment can be defined as a combination of seismic hazard and vulnerability.
Seismic hazard depends upon geology of the area under consideration and is, therefore, site specific.
Seismic vulnerability depends upon:
1. The material of which the structure is made.
2. The mechanical properties of construction materials.
3. The geometry and layout of a building and its structural components.
4. The detailing of structural components as well as floor-wall connection details.
There are various approaches regarding assessment of seismic vulnerability of a building stock.
Detailed discussion on this topic is given by Lang [La 02]. According to Lang, various methods for
seismic vulnerability can be classified as given in Table 2.1.
These methods are briefly discussed in the following sub-sections.

2.10.1 Observed vulnerability

This method of assessment is based on the statistics of the buildings’ damage from the past
earthquakes. The method is specifically suitable for poor quality non-engineered construction whose
resistance is difficult to calculate by analytical or numerical methods.
First such type of a study was made by Whitman [WRH 74]. He gathered the data of
approximately 1600 buildings damaged due to the San Fernando earthquake of 9th February, 1971.
Based on the gathered data, he carried out a statistical analysis and developed damage probability
matrices (DPM). These DPM indicated the probability of occurrence of various distress levels at
various intensities of earthquake. Other examples of seismic vulnerability assessment, based on
observed vulnerability method also exist [CS 92, PS 89].
The method has the disadvantage of requiring a substantial amount of data inventory.
Additionally, the analysis of the data, for the area under consideration, cannot be extended to other
towns and cities. Similarly, the method does not possess the ability to calculate reduction in
vulnerability of buildings as a result of retrofitting or strengthening of buildings.
15
2.10.2 Vulnerability functions based on expert opinions
The first such kind of seismic vulnerability assessment was carried out by Applied Technology
Council (ATC) under the sponsorship of the Federal Emergency Management Agency (FEMA) of
USA. The results of the study were summarized in a report (ATC 13). The ATC -13 report developed
the DPM for 78 earthquake engineering facilities, out of which 40 belonged to buildings. These DPM
were developed by asking 58 experts to estimate expected percentage of damage that a specific
structural type would undergo under the action of various intensities of earthquake. Each expert was
asked to fill a comprehensive questionnaire by utilizing their best knowledge.
Since then a number of modified and improved versions of such type of a methodology have
been proposed [KNKH 97, PCM 00, FKLG 01]. However, the drawback of the methodology lies in
the subjective opinion. Each expert, depending upon his knowledge and engineering judgment has his
own opinion. In addition, the results obtained for the target area cannot be extended to other towns and
cities.
Another drawback of the observed vulnerability and expert opinion methods of vulnerability
assessment is that these methods cannot be applied to areas with no past earthquake experience.

2.10.3 Analytical approach based on simple models

Vulnerability assessment methods of observed vulnerability or expert opinions are applicable to areas
with some past earthquake experience. For other areas, analytical methods for determining the seismic
vulnerability are needed.
Analytical methods using simple building models are based on a small number of input
parameters. Consequently, these have the advantage of analyzing a large number of buildings in a
relatively short period of time. The efficiency of a method increases if the input parameters required
for analyzing the buildings can properly capture the overall seismic behavior of the buildings.
One such method is developed by D’Ayala et al [DSOP 97] for earthquake loss estimation of
historical towns of Europe. The method has been used for earthquake loss estimation of a case study in
Alfama district, Portugal [ DSOP 97] and villages hit by the Umbria-Marche earthquakes in 1997 [SD
99].
Two types of collapse mechanisms, namely, in-plane and out-of-plane, are considered. The
potential collapse mechanism and the corresponding capacity are determined by the geometry and
boundary conditions, usually based on visual observations. The in-plane and out-of-plane collapse
mechanisms are further divided into various subclasses as shown in Figure 2.25. Each collapse
mechanism is related to a damage grade recommended by European Macroseismic Scale 98 [EMS
98].
Another such approach, applicable to both masonry and R.C. structures, was developed
by Calvi [Ca 99] and was applied to Catania City [FPCB 99]. The method only considers in-plane

16
failure modes and ignores the possibility of out-of-plane failure of wall. The model is based on
only few number of parameters such period during which the building was constructed,
construction material and number of stories in the building.
The analysis of the building, by this method, is carried out in order to determine the
displacement capacity and the displacement corresponding to various defined limit states.
Whereas, the displacement demand is determined by using displacement spectrum.
The structural model of each type of building (i.e masonry or R.C) is defined by two
parameters, namely, equivalent period of vibration and displacement demand reduction factor
(determined by using energy dissipation characteristic of building structure)
The methods of analytical model based on simple models have the advantage of analyzing a large
number of building stock in a relatively short time. However, the assumptions made for
simplification purpose affect the accuracy of results.

2.10.4 Score assignment

This method is aimed at identifying a seismically hazardous building situated in a target area.
Structural deficiencies, based on correlation between observed damage and structural characteristics,
are identified. Based on different structural deficiencies, scores are then assigned by experts.
The method serves as a first phase for screening seismically hazardous buildings in an area.
Buildings with very low scores are then investigated in detail. First such type of methodology was
proposed by the Applied Technology Council [ATC 14]. The methodology was aimed at developing a
procedure for evaluating the seismic performance of existing building in order to identify hazardous
buildings that can impose risk to human lives. Since then, a number of improved versions have been
proposed [FEMA 154, FEMA 155, FEMA 178, FEMA 310].
The advantage of this method over vulnerability assessment methods based on observations
and expert opinion is that vulnerability assessment of the buildings in a target area can be updated after
the retrofitting or strengthening of buildings.

2.10.5 Detailed analysis procedures

The buildings identified as hazardous from the method of score assignment need further detailed
investigation for a final decision. The method used for this purpose can be divided into two classes,
namely, static and dynamic procedures. Each method can be further sub-divided in two methods,
namely, linear and non-linear methods. This makes a total of four methods as follows:
1. Linear static procedure.
2. Non-linear static procedure.
3. Linear dynamic procedure.
4. Non-linear dynamic procedure.

17
However, these methods are time-consuming and not suitable for an earthquake scenario project in
which a lot of buildings need to be investigated.

2.11 SELECTION OF METHOD FOR ASSESSING VULNERABILITY OF URM

BUILDINGS OF NORTHERN PAKISTAN
The merits and de-merits of all the methods for determining seismic vulnerability are briefly discussed
in Section 2.10. As already discussed in Section 2.10.2, observed vulnerability and expert opinion
methods of vulnerability assessment are applicable to only those areas, which have some past
earthquake experience. Therefore, the applicability of these methods, for most other cities of Northern
Pakistan becomes questionable.
The analytical approaches based on simple models are simple and time-saving, however,
ignoring important parameters regarding geometrical properties and building details affect the
accuracy of results.
Score assignments are quite time-consuming. A lot of labor work is required for the detailed
survey of the buildings whose vulnerability is required to be determined. Additionally, the detailed
inside survey of a residential building for rating structural deficiencies is not feasible in view of the
cultural setup of Northern Pakistan.
The linear analysis procedures, although relatively simple, were not considered suitable as
these methods are strictly applicable to linear elastic systems. The masonry buildings, even
unreinforced, are distressed well into nonlinear inelastic range before failure.
The nonlinear dynamic analysis procedures involve high computational cost and have limited
applicability and, therefore, are not feasible when the vulnerability of a large number of buildings
needs to be assessed.
Keeping in view the merits and de-merits of all methods of vulnerability assessment of
buildings, a method combining a non-linear static analysis procedure and an analytical method based
on simple models is adopted in the present study. The method, in general, has the approach similar to
the one being proposed by Lang [La 02]. However, as discussed in Chapter 4, the proposed evaluation
method involves a relatively simple way of determining failure modes in URM piers. Additionally, the
method places much emphasis on ultimate drift ratios instead of the displacement ductility factor of
URM piers for which the existing literature shows more scatter [MC 97]. The method, which is
discussed in detail in Chapter 4, allows the seismic vulnerability assessment of a large number of
buildings without ignoring important parameters influencing the seismic vulnerability of URM
buildings.

18
 Figure 2.1 Response of a masonry building subjected to the ground shaking.
(Figure adopted from http://www.iitk.ac.in/nicee/EQTips)

Figure 2.2 Influence of ties and diaphragms on the seismic behavior of a masonry building
(Tomaževic [To 99]):
(a) building with wooden floors and without ties; (b) building with wooden floors and tied
walls; (c) building with rigid floors and tie-beams

19
 In-plane shearr Cracks

Separatioon of orthogonall walls

Out-of- plane ovverturning of gaable

Cracks at opeening

o plane vibratioon of load bearingg walls

Out- of-

Diagonal crack
k at corner

Figure
F 2.3 Varrious types off failures obseerved in unreeinforced massonry buildin
ngs under seissmic
excitations.

Figure 2.4 Typical diagon

nal tension sh
hear cracks in
n an URM pieer

20
 (a)

(b)

Figure 2.5 Typical cracks in URM buildings originating from the corners of openings

21
 (a)

(b)

Figure 2.6 Rocking/toe crushing failure in URM piers

22
 (a)

(b)

(b)

Figure 2.7 Out-of-plane failure of URM walls due to the improper floor- wall
connection, aggravated by additional thrust from the roof trusses

23
 (a)

(b)

Figure 2.8 Wedge type separation of masonry below roof purlins at the buildings’ corners

24
 (a)

(b

Figure 2.9 Separation of orthogonal walls made of (a) stone masonry (b) brick masonry

25
Figure 2.10 Diffusion
D of low
l quality mmasonry in thee top courses of wall due too the
local interacttion with the heavy roof

Figuree 2.11 Typicaal overturningg of a gable in

n the out-of-p
plane direction
n

26
 (a)

(b)

Figure 2.12 Diffusion of external veneer in the double-leaf stone masonry walls due to the absence
of connection between the masonry courses across the transverse direction

27
 (a)

(b)

Figure 2.13 Failure of infill masonry walls due to the diagonal compressive
forces transferred from beam-column joint.
Note fracture of column in the vicinity of beam- column joint in figure (b)

28
 (a)

(b)

Figure 2.14 Overturning of free- standing masonry walls in the out-of

plane direction.

29
 (a)

(b)

Figure 2.15 (a) Typical overturning observed in boundary walls. Overturning of a slender parapet
wall was prevented, as shown in figure (b), due to proper bracing across the length.

30
 (a)

(b)

Figure 2.16 (a) Rocking in an URM pier (b) Possible toe-crushing after rocking.

31
 (a)

(b)

Figure 2.17 Sliding shear failure in squat URM piers due to sliding of bed joints.

32
 (a) (b)
Figure 2.18 Diagonal tension shear failures in URM piers constructed with:
(a) Strong brick-weak mortar (b) Strong mortar-weak brick.

33
Figure 2.19 Testing facility at National Building Institute, Ljubljana, Slovenia, employed
to conduct quasi-static cyclic tests

(a) (b) (c)

Figure 2.20 Various types of loading histories employed in quasi-static cyclic tests

(a) Variable Amplitude (b) Constant Amplitude (c) Mixed Amplitudes

34
Figure 2.21 Diagram illustrating general concept of PsD testing technique.

Figure 2.22 Diagram illustrating concept of ‘ Sub-structuring’ in PsD testing

35
 Figure 2.23 Shaking table installed at the Earthquake Simulation laboratory, Ministry
of Construction, Tsukuba-shi Ibaraki-ken, Japan

(a)

(b)
Figure 2.24 Shaking table installed at the Earthquake Simulation laboratory, Ministry
of Construction, Tsukuba-shi Ibaraki-ken, Japan

36
 (a)

(b)

Figure 2.25 Failure modes assumed by D’Ayala et al [DSOP 97] in URM buildings
(a) In-plane failure modes (b) Out-of-plane failure modes

37
Table 2.1 Various methods of evaluating seismic vulnerability of buildings’ stock [La 02]

Increasing computational effort

Expenditure

Application Building stock Individual building

Simple Detailed
Method Observed Expert Score
analytical analysis
Vulnerability opinion assignments
models procedure

38
 Chapter 3
PERFORMANCE OF UNREINFORCED BRICK MASONRY (URBM) PIERS
UNDER IN-PLANE QUASI-STATIC LATERAL CYCLIC LOADINGS

3.1 INTRODUCTION
The aim of this experimental work was to study the performance of brick masonry shear walls,
constructed in locally used ‘khaka’ (i.e., stone dust) mortar, under the action of in-plane lateral quasi-
static cyclic loading.
As part of this study, twelve piers were tested. Aspect ratio and pre-compression levels were
varied to observe their effect on the behavior of the piers. However, due to loss of data of four pier,
because of problems with the data acquisition system, lateral force-displacement hysteresis diagram of
eight piers are presented along with the detailed discussion on the failure modes and cracking pattern.
Comparison of the experimental results of the tested piers with Turnšek Equation [TS 80] for
determining diagonal tensile shear strength of URM piers is also made.
In addition, displacement ductility factors, moduli of elasticity and rigidity, energy dissipation
characteristics and stiffness degradation of the tested piers are also discussed in a quantifiable manner.
Similarly, the effect of pre-compression level on displacement ductility, energy dissipation
characteristics and stiffness degradation is also studied.
Finally, drift based performance levels are recommended for brick masonry piers constructed in stone
dust mortar.

3.2 SELECTION OF MATERIAL USED IN EXPERIMENTAL WORK

The research work was aimed at evaluating the seismic performance of URBM piers, constructed in
stone dust mortar, being commonly used in Northern Pakistan. Unfortunately, testing facilities
required for this purpose were not available at the national level. The experimental work was,
therefore, carried out using the laboratory facilities of the Institute of Engineering Mechanics (IEM) at
Harbin, China. Consequently, there was no choice other than to use local material available in Harbin
city. However, best efforts were made to simulate the mechanical properties of brick masonry and its
constituents being used in Northern Pakistan. As a bench mark, properties of bricks masonry and its
constituents used by Ali [AliQ 04] in his PhD work were set as standard for this experimental study.
An outline is given in the following sub-sections for the criteria used in the selection of
materials for the present study.

3.2.1 Bricks

The size of solid fired clay bricks used in Harbin were similar to those being used in Northern
Pakistan. The bricks available in Harbin had no frog, had an average size of 9.29 x 4.33 x 2.09 in. The
39
ones used in Northern Pakistan, contain a frog, and have an average size of 8.75 x 4.25 x 2.75 in.
However, the compressive strength of bricks in Harbin was high as compared to that of the bricks used
in Northern Pakistan. The compressive strength of solid fired clay brick used in Northern Pakistan
occasionally exceeds 3000 psi [AliQ 04]. The average compressive strength of solid fired clay bricks
collected from various sources of Harbin city was found to have an average compressive strength of
4000 psi.
Bricks used in this experimental study had an average compressive strength of 3200 psi and
were obtained from a demolished building in Harbin city. The use of old bricks was considered to be
more realistic on two counts: it simulated good quality brick strength available in Pakistan, and it was
a closer representation of a ‘real life’ or existing brick masonry building stock as discussed in Chapter
4.

3.2.2 Mortar

It is a common practice in Northern Pakistan to use cement: sand: khaka (CSK) mortar in a ratio of
1:4:4 by weight. It was, therefore, decided to use the same mortar in this experimental work. Khaka
(i.e., stone dust), obtained from a stone crushing industry of Harbin, and was sieved by using sieves
Nos. 4, 8, 16, 30,50 and 100. The sieved stone dust was mixed according to the proportions of stone
dust used in the work of Ali [AliQ 04]. The details of mixed proportions are provided in Table 3.1.
Another benchmark was to achieve a mortar having a compressive strength of 900 psi at 28
days maturity as used by Ali [AliQ 04]. For this purpose, a CSK mortar with a proportion of 1:4:4 was
prepared with water/cement (w/c) ratios of 1.1, 1.4 and 1.7. Mortar cubes with each side equal to 2.75
in. were prepared and tested at the age of 7 days. Based on the assumption that in 7 days the mortar
gains approximately 70% of its 28-day compressive strength, a w/c ratio of 1.1 was found suitable to
achieve a target compressive strength of 900 psi at the age of 28 days. The 7-day strength values for
the mortar cubes are provided in Table 3.2. The mean and percent coefficient of variation (% C.O.V.)
are also given in this table. However, while casting masonry piers for quasi-static cyclic tests, the
mortar with a w/c ratio of 1.1 was found to be less workable and thus the w/c ratio was increased to
1.2.
The entire experimental work was carried out in two phases. In the first phase, samples were
cast and tested to determine the physical and mechanical properties of masonry and its constituents. In
the second phase, shear-compression quasi-static cyclic tests, commonly known as quasi-static cyclic
tests were carried out on masonry piers. The procedure and the results of all these tests are discussed
in Sections 3.3 and 3.4.

3.3 MATERIAL TESTS

Tests were conducted on masonry and its constituents to determine the following physical and
mechanical properties:
40
Brick tests:
a. Initial rate of absorption, IRA
b. Compressive strength, fb
c. Flexural tensile strength, fbt

Mortar test:
Compressive strength, fmo

Masonry assemblage tests:

a. Compressive strength, fm
b. Modulus of elasticity, Em
c. Principal tensile strength, ftu
d. Modulus of rigidity, Gm
e. Cohesion, c, and coefficient of friction, µ

3.3.1 Brick tests

3.3.1.1 Initial rate of absorption, IRA

The IRA test was carried out in accordance with Section 10 of ASTM C – 67. The IRA was calculated
as:
IRA (gm/min/30in2) = 30 (Ww - Wd)/Ac
where Wd = weight of oven dry brick in grams ; Ww = weight of wet brick in grams, after placing in a
tray of water for one minute duration, as per ASTM C – 67 specifications; and Ac = area of brick
surface in contact with water. The results of IRA tests are provided in Table 3.3.

3.3.1.2 Compressive strength, fb

The compressive strength test was carried out according to the specification mentioned in Section 7 of
ASTM C – 67. Specimens were capped on both the sides with a rich paste of cement and sand. The
specimens were then cured for 7 days and were then tested in uni-axial compression. The compressive
load on the brick specimens was applied until failure, in a direction perpendicular to the 9.29 x 4.33in.
side. Compressive strengths of the tested specimens are provided in Table 3.4.

3.3.1.3 Flexural tensile strength, fbt

The flexural tensile strength test was carried out according to the specification mentioned in Section 6
of ASTM C-67. Specimens were tested in flexure using simply-supported end conditions with a span
of 7.95 in. A vertical concentrated force, applied at the middle, was slowly and gradually increased

41
until failure. The flexural tensile strength of the bricks were calculated by using the well known
flexure formula [PB 98] as follows:
M
f .
S
where Mmid = mid-span bending moment in the brick, and Sb = section modulus of the brick section =
bbtb2/6 where bb and tb are the width and thickness of the brick, respectively. The flexural tensile
strength of the tested bricks are provided in Table 3.5.

3.3.2 Mortar test

3.3.2.1 Compressive strength, fmo

In order to determine compressive strength of mortar used in masonry piers, tests were carried out on
mortar cubes as per ASTM C-109 specifications. The size of mortar cube was 2.75 x 2.75 x 2.75 in .
Two mortar samples were taken while casting each masonry pier. CSK mortar used in masonry piers
was having a ratio of 1:4:4 by weight, with a w/c ratio of 1.2. Results of the compressive strength of
the mortar cubes are provided in Table 3.6

3.3.3 Masonry assemblage tests

3.3.3.1 Compressive strength, fm

Compression tests on masonry prisms were carried out in accordance with ASTM E-447
standards. The prisms, prepared in CSK 1:4:4 mortar, were 22.6 in. high with an average cross
sectional area of 9.25x 9.25 in. Nine prisms were cast using the English bond with an average mortar
thickness of 3/8 in. English bond was specifically used due to its common use in the brick masonry
buildings of Northern Pakistan. This type of bonding arrangement was also used in the quasi-static
cyclic tests on brick masonry piers, which will be discussed in section 3.4.
In order to achieve a leveled surface for uniform distribution of vertical stresses during the
tests, prisms were capped on top and bottom ends by using a rich cement sand mortar. A typical
prism used in compression test is shown in Figure 3.1. All prisms were tested after moist curing for at
least 28 days. Prisms were tested under a uni-axial compression loading till failure. Compressive
strength of a masonry prism, fm, was calculated as follows:
Peak vertical load
f .
Cross sectional area

Results of these tests are provided in Table 3.7

42
3.3.3.2 Modulus of elasticity, Em
Modulus of elasticity of the masonry was determined by using the stress-strain data of the masonry
prisms used to determine the compressive strength of masonry. ASTM E-447 specifications were
followed to determine the modulus of elasticity of masonry.
Two steel rods, each with flat strips welded at the ends, were inserted in the holes that were
drilled at selected points along the gage length, as shown in fig. 3.2. The vertical displacements at the
ends of steel rods were found using the displacement dial gages. Difference in the displacements of the
two dial gages provided the shortening in the gage length. Due to limitations of the testing
arrangement, the displacements were not recorded after peak compressive force. However, the
recorded data was sufficient enough to determine modulus of elasticity of the masonry prisms.
Modulus of elasticity of the masonry, Em, was determined by using the following relation [PB 98]:
∆ Longitudinal stress
E .
∆ Longitudinal strain
Where ∆longitudinal stress and ∆longitudinal strains were calculated by using the ASTM E-447
standards as follows:
f f
∆ longitudinal stress – ; and
3 20
f f
∆ Longitudinal strain Difference of strains correponding to and , respectvely.
3 20
Calculated values of the modulus of elasticity of the tested masonry prisms are provided in Table 3.8

3.3.3.3 Principal tensile strength, ftu

In order to determine the principal tensile strength, the diagonal tension tests were carried out
in accordance with ASTM E 519 – 02 specifications. Four square prisms, each with a length of 29.5 in.
and thickness of 9.45 in. were tested using the testing arrangement shown in Figure 3.3.
The vertical load applied at the top end of the prism produced shear stresses in the prism along
its vertical diagonal plane. These stresses, using the equation recommended by ASTM E 519 – 02
specifications, were calculated as follows:
0.707 ND
Ss .
A
where Ss = Average value of shear stress produced along the vertical diagonal under the action of
vertical load, ND, and An = net area of the prism. For solid masonry An, as defined by ASTM E 519 –
02 specifications, is the average length of two sides multiplied by the thickness of prism.
Unfortunately, ASTM specifications do not provide any information to determine the
principal tensile strength of masonry. Therefore, the equation recommended by the RILEM
specifications [RILEM LUM B6], was used as follows:

43
 0.5 ND
ftu .
A
where ftu = Principal tensile strength of the masonry prism, and, (ND)u = peak value of the
vertical load applied on the masonry prism during the test.
Principal tensile strength of the masonry prisms, calculated by the Equation 3.4b, are provided in
Table 3.9. The average value of ftu/fm was found to be 24.5/658.1 = 0.037 (i.e., 3.7%) with a % C.O.V.
of 16.7.

3.3.3.4 Modulus of rigidity, Gm

Modulus of rigidity of the masonry was calculated as per ASTM E 519 – 02 standards by
using load-displacement data of prisms that were tested to determine the principal tensile strength of
the masonry. The vertical load applied at the top the end of masonry prism during diagonal tension
tests produced elongation and shortening along the horizontal and vertical diagonals, respectively.
These deformations were recorded by using linear variable displacement transducers (LVDTs) shown
in Figure 3.3 a and b, respectively.
Shear stresses along the vertical diagonal plane of the prisms, for various magnitudes
of vertical loads, were calculated by using the Equation 3.4a. The corresponding shear strains were
calculated by using ASTM E 519 – 02 specifications as follows:
∆V ∆H
γ .
gl
where γs = shear strain corresponding to a shear stress Ss, gl= gage length , ∆V = shortening
in vertical gage length, and ∆H = elongation in horizontal gage length.
Once shear stresses and corresponding shear strains for various magnitudes of vertical loads are
determined, the modulus of rigidity can be determined by using standard the mechanics of solids
approach [PB 98] as follows:
∆ Shear stress
G .
∆ Shear strain
Unfortunately, ASTM standard specifies no values of stresses (like fm/3 and fm/20 for use in
Equation 3.3) for determining the modulus of rigidity of masonry. Consequently, there was no way
other than to adopt some rational procedure for this purpose. For the purpose of this study, the ∆ shear
stress and ∆shear strains were determined as:
S S
∆ Shear stress – ; and
2 20
S S
∆ Shear strain Difference of strains correponding to and , respectively.
2 20
where (Ss)max = Shear stress corresponding to peak vertical load calculated from eq. 3.4a.
The modulus of rigidities for all the four masonry specimens, calculated by Equation 3.6 are
given in Table 3.10. The average value Gm/Em was found to be 7.68 (with a C.O.V. of 153 %) which
44
was too high as compared to an upper limit of 0.4 reported in the literature [To 99] and is, therefore,
not recommendable for practical application.
Fortunately, another method for determining modulus of rigidity of masonry also exists. The
method makes use of lateral force-displacement data of the quasi-static cyclic tests on masonry piers.
The modulus of rigidity of masonry determined from that method, discussed in Section 3.8.2, provided
satisfactory results.

3.3.3.5 Cohesion, c, and coefficient of friction, µ

There exists no ASTM standards to conduct tests on masonry triplet for determining c and µ.
Therefore, in order to determine the shear strength of the masonry, which depends upon c and µ, EN-
1052-3 specifications were followed to carry out triplets tests. Testing arrangement for conducting
these tests is shown in Figure 3.4.
Eleven samples were tested in order to determine cohesion and coefficient of friction of the masonry
triplets. Three specimens were tested without pre-compression. Where, three specimens each were
tested under the pre-compression force of 1 and 2 tonne. One of the three specimens, planned cast to
test under the pre-compression force of 3 tonne, got damaged during the handling and, therefore, only
two specimens were tested under the pre-compression force of 3 tonne.
Shear strength of a triplet, τu, was determined [PB 98] as follows:
V
τ .
2A
where Vus = shear force required to produce shear sliding in a triplet, and Abm = area of brick-mortar
joint between the two bricks. For a triplet, total area of brick-mortar joints = 2 Abm
Linear regression analysis were carried out in order to determine c and µ, using the test data
summarized in Table 3.11. The regression analysis, graphically shown in Figure 3.5, resulted in
values of c = 56.84 psi and µ = 0.59.

3.4 IN-PLANE QUSAI-STATIC CYCLIC TESTS ON UNREINFORCED BRICK

MASONRY PIERS
One the most important objectives of the experimental work was to study the seismic behavior of the
solid fired clay brick masonry shear piers constructed in cement, sand, and khaka (CSK) mortar. To
accomplish this, shear-compression quasi-static cyclic tests (commonly known as quasi-static cyclic
tests) were carried out on four series of piers. The details of testing are given in the following sub-
sections.

3.4.1 Geometry of piers

Piers with three different geometries were prepared for testing as a part of the experimental work.
These piers had the aspect ratios (i.e., height-to-length ratio) of 0.66, 0.93 and 1.22. Three piers each
45
were cast with aspect ratios of 0.66 and 1.22, and six piers were cast with aspect ratio of 0.93. Keeping
in view the capacity of the actuator used for applying lateral force, the lengths of each pier was limited
to 4.46 ft.
All piers were constructed using English bond, which is commonly used in Northern Pakistan. The
thickness of mortar used in Northern Pakistan is approximately 0.5 in. with the bricks having an
average thickness of 2.75in. Since the bricks used in the piers had a thickness of 2.1 in. instead of 2.75
in, therefore, the thickness of mortar used in the piers cast for the experimental work was accordingly
scaled to 3/8 in.
Piers with the aspect ratios of 0.66 and 0.93 were cast to simulate the geometry of typical
walls in a room, with a door opening at the end. Where, the piers with an aspect ratio of 1.22 were cast
to simulate the geometry of a typical slender pier between two openings.

3.4.2 Pre-compression on piers

Pre-compression levels on the piers ranged from 61 to 103 psi. These values of pressures were selected
to simulate the vertical load on piers supporting two to three stories. A total of four series of piers,
with three piers in each series, were tested.
Based on the aspect ratio and pre-compression level, piers were divided into four categories,
namely PI, PII, PIII and PIV. Each pier in a given series was designated with a lower case alphabet at
the end of the category designation. e.g., The first pier of the PI category was designated as PIa.
Similarly, the second and the third pier of same category were designated as PIb and PIc, respectively.
The details of the dimensions, aspect ratios, and pre-compression levels are given in Table 3.12.

3.4.3 Testing setup

Quasi-static cyclic tests were carried out for three different levels of pre-compression (σo/fm=0.091,
0.138 and 0.153 corresponding to σo= 61 psi, 93 psi and 103 psi, respectively) by using the testing
setup shown in Figure 3.6. The piers were fixed at the top and bottom ends through stiff reinforced
concrete beams, which in turn were connected to the testing frame. The vertical pre-compression was
developed through two hydraulic jacks operated by an electric motor. A set of steel rollers were placed
between the jacks and the top concrete beam to allow lateral displacement of the pier. Details of
testing setup are shown in Figures 3.7a through e.
The lateral seismic forces were simulated by step wise linearly increasing static lateral
displacements. The typical lateral displacement history is shown in Figure 3.8. The testing process, as
shown in Figure 3.8, consisted of two phases. The first phase of testing was carried out in the force
controlled testing environment with a total of four linearly increasing force cycles with the each cycle
being repeated three times. Objective of force controlled testing phase was to have a well defined
lateral load-displacement curve before diagonal tension shear cracking. The lateral strengths of piers
before testing were estimated by using Equations 2.1, 2.3 and 2.4. For the purpose of safety, the
46
estimated lateral strength of a piers, Vest, was taken as 75% of the lateral strength of the pier by using
afore mentioned equations.
Each test was started by applying a lateral force of Vest/4. A total of three cycles were repeated
for this level of force. Each cycle was started by applying the lateral force towards right side, causing a
‘push’ in the pier. This loading direction was termed as the positive loading direction (PDL). After
applying the target force, the pressure in the actuator was gradually decreased to zero and a lateral
force with the same magnitude was applied in the reverse direction, causing a ‘pull’ in the pier and
was termed the negative loading direction (NLD). The lateral force was then increased to a magnitude
of Vest/2 and the above mentioned procedure was again repeated for three cycles. Similarly, three cycle
each were repeated for lateral forces having the magnitudes of ¾ Vest and Vest. The frequency of
loading was kept 1/60 Hz during this phase of testing.
The second phase of testing was carried out in the displacement controlled testing
environment. The frequency of displacement during this testing phase varied from 1/60 Hz for
displacement cycles with lower amplitudes to 1/120 Hz for displacement cycles with higher
amplitudes. This phase of testing was started by imposing a lateral displacement slightly greater than
the maximum displacement recorded at the end of force controlled testing phase (multiple of 0.01 in.).
Three cycles of the displacement were completed. The next target displacement was imposed by
increasing the previous displacement by 0.01 in. The target displacement was again repeated three
times and so on. A typical displacement history followed during the quasi-static cyclic testing is given
in Table 3.13.
Since the displacement history was applied manually, occasionally the displacement increased
suddenly. Variation in the pre-compression also occurred, especially after the diagonal tension
cracking in a pier, due to the vertical component of pier’s displacement as a result of the increase in
the imposed lateral displacement. However, change in pre-compression, which was occasionally more
than 4-5%, was instantly controlled.
Deformations in the piers were recorded by means of the LVDTs shown in Figure 3.6.
LVDT1 was the most important of all the displacement transducers and was used to measure (or
impose) the lateral displacement at the top end of a pier. LVDT2 was used to monitor possible
slippage of a pier with respect to the R.C. beam anchored to the testing floor. LVDT3 was used to
check the possibility of rocking, especially in slender piers. LVDT4 and5 were used to record the
diagonal deformations in a pier.
Data of the three pier specimens, namely PIa, PIb and PIIIc, could not be not recorded due to
technical problems with the data acquisition system. Similarly, due to vertical stress concentration at
the ends, as a result of uneven upper horizontal surface of the pier, data of pier PIVc did not reflect the
typical hysteretic behavior and was, therefore, discarded. Hence, the results of eight, out of total
twelve tested piers, will be discussed in the rest of the dissertation. It is important to note that the loss

47
of data of four piers did not significantly affect the purpose of experimental work, which was aimed at
determining the average values of properties related to the seismic performance of the URBM piers.

3.5 CRACKING PATTERN AND FAILURE MODES IN THE TESTED PIERS

Two failure modes, namely diagonal tension and bed joint sliding accompanied by diagonal tensile
cracking, were observed while testing the piers. In the case of piers with high to medium aspect ratios
(PI through PIII), diagonal tension was the predominant failure mode. In the case of squat piers (PIV
series) bed joint sliding was observed after formation of diagonal tension cracks. Detailed discussion
on the observed behavior of piers is given in following sub-section:

3.5.1 PI series

Three piers, each having an aspect ratio of 1.22, were tested under a pre-compression level of 103 psi.
Due to technical problems with the data acquisition system, testing of pier PIa, shown in Figure 3.9,
was stopped after formation of diagonal crack. Consequently, pier PIa was not tested to failure.
In case of pier PIb, more than one inclined cracks were developed in each direction of lateral loading,
as shown in Figure 3.10. Unfortunately, data of this pier could not be saved due to shut down of
computing system during the final stages of testing.
In case of pier PIc, as shown in Figure 3.11, one diagonal crack was developed along each
direction of lateral loading. As the test progresses, vertical cracks also developed in the middle portion
of pier, as shown in Figure 3.12.
In the case of piers PIb and PIc, diagonal cracks extended towards corners with the increase in
lateral displacement thus leading to the crushing of masonry as a result of insufficient area to resist
compressive stresses. A typical example of masonry crushing is shown in Figure 3.13. Crushing of the
masonry at the bottom produced separation of the diagonally cracked part of pier PIc, as shown in
Figure 3.14.

3.5.2 PII series

Piers in this series were having an aspect ratio of 0.93 and tested under a pre-compression level of 61
psi. In all the tested piers, one of the diagonally cracked wedge shaped portion of pier tended to
separate from rest of the pier with the increase in applied lateral displacement, as shown in Figures
3.18 and 3.19. The relatively lower vertical compression level (σo /fm = 0.09) may be one of the reason
for such type of behavior. Splitting of the masonry at the ends , as shown in Figure 3.15, and crushing
of masonry at the bottom, as shown in Figure 3.16 and 3.22, was observed in piers, causing extensive
damage. Consequently, in spite of little reduction in lateral force after peak resistance and insignificant
increase in lateral displacements, test was stopped in case of all the three piers due to the danger of
falling the diagonally cracked wedges of piers, as shown in Figures 3.15, 3.18 and 3.20. However,
other parts of the each pier were found intact.
48
3.5.3 PIII series
Piers in this series were having the same aspect ratio of 0.93. However, the pre-compression level was
increased to 93 psi. The number of inclined cracks developed in the piers were relatively more as
compared to the piers belonging to PII series, as shown in Figures 3.23 and 3.24. Unlike piers
belonging to series PII, the damage in the piers was not 'localized' (i.e., excessive separation of only
diagonally cracked wedges) and the damage was distributed through out the each pier, as shown in
Figures 3.23 and 3.24. Vertical cracks were also observed in case of pier PIIIb, as shown in Figure
3.24. Crushing of bricks at the corners and splitting of masonry in the piers was relative more
extensive as compared to the piers belonging to series PII (see Figures 3.25 through 3.28). One of the
reason for such type of behavior may be relatively higher pre-compression level.

3.5.4 PIV series

Piers in this series were having the lowest aspect ratio of 0.66 with a pre-compression level of 93 psi.
Three piers were tested in this series. Pier PIVc was tested first of all , followed by piers PIVa and
PIVb.
Pier PIVc reflected quite surprising behavior, which was not compatible with the rest of all
the tested piers. The pier was initially tested at a pre-compression of 61 psi. During the force
controlled phase, a vertical crack produced at the bottom during the first cycle at an applied lateral
force of 4.5 tonne. The corresponding displacement was found to be 0.02 in. In the remaining two
cycles, the crack propagated in the upward direction until it reached the top surface of the pier,
dividing the pier in almost two parts, as shown in Figure 3.29. However, no additional crack
developed till an imposed lateral displacement of 0.12 in. ( corresponding drift ratio = 0.33%). Test
was stopped at this stage and re-started after increasing the pre-compression level to 93 psi. At a
lateral displacement of 0.14 in., another vertical crack, branching from the existing vertical crack,
developed and instantaneously reached the top surface of pier, as shown in Figure 3.30. First diagonal
crack developed at a drift as high as 0.24 in., with a corresponding drift ratio equal to 0.66%. Test was
stopped at a lateral displacement of 0.36 in.
A careful study after the test revealed that the upper surface of the pier was quite uneven,
causing a significant gap between the top surface of the pier and the connected R.C. beam, as shown in
Figure 3.32. This gap caused uneven distribution of vertical pressure at the top of the pier. Magnitude
of the vertical pressure was more at the contact area, especially at the right side of pier, causing
splitting of the masonry in that area, as shown in Figure 3.33.
Pier PIVa was initially tested under a pre-compression of 61 psi. The pier underwent diagonal
tension cracking in both direction of loading, as shown in Figure 3.34, at a lateral displacement of 0.06
in. However, after the formation of diagonal tension cracking, epoxy resin, which was used to adhere
the foundation beam of the pier to R.C. beam anchored to the testing floor, failed in shear and the pier

49
slipped at that joint. Testing of pier PIVa was re-started under an increased vertical pressure of 93 psi.
Sliding of bed joint occurred at a lateral drift of 0.08 in. in the third course above the bottom on the left
end, as shown in Figure 3.35. Vertical cracks developed on the right side, along with the splitting of
masonry on the right end, as shown in Figures 3.36 and 3.37. Relatively higher drift was achieved in
the pier before failure. Consequently, pier underwent severe damage, as shown in Figures 3.36 and
3.37.
Pier PIVb was tested under a vertical pressure of 93 psi from the very start. Surprisingly, the
pier underwent diagonal tensile cracking at a drift ratio as low as 0.043 % (almost half of drift ratio for
other piers). Bed joint sliding occurred after the formation of diagonal tension cracks, as shown in
Figure 3.39. A number of inclined cracks were also developed on the right side of the pier, as shown in
the Figure 3.39, causing severe damage on the right side of the pier, as shown in Figures 3.40 and
3.41. The test was stopped at a stage when the diagonally cracked wedges were about to fall due to
extensive damage, as shown in Figures 3.40 and 3.41.
Results of piers PIVa and PIVb , due to difference in behavior as compared to all other piers
were not included in the statistical analysis for determining various parameters quantifying the
performance of piers.

3.6 LATERAL LOAD- DISPLACEMENT DIAGRAMS

Lateral load-displacement relationships are normally presented by two types of diagrams as follows:
1. Hysteresis loops, showing lateral load - displacement relationship at various stages of
loading during a test.
2. Envelope of first cycle of each imposed force/displacement cycle using hysteresis loops.
Initially drawn hysteresis loops of the tested piers, using recorded data of data acquisition system,
were not smooth. Since the frequency of applied loading was very low, therefore, data was ‘filtered’
by using ‘low pass filtering’ technique. The resulting hysteresis loops became reasonably smooth at a
‘cut off frequency’ of 0.1 Hz. However, the peak values of lateral forces were reduced by around 10%.
The data was then filtered by ‘moving average filtering’ technique. By using this technique, the
resulting loops became not only smooth but the peak values were also affected only by 3-5%.
Results of all tested piers are graphically shown in Figures 3.42 through 3.50. It may be noted
that the point shown by triangle corresponds to the formation of first significant crack, defined as the
first crack, which was wide enough to be visually seen during the testing. The exact value of lateral
force and displacement at the formation of the first significant crack was determined by a significant
decrease in the slope of lateral load-displacement curve.
Similarly, point shown by circle corresponds to the formation of first diagonal tension crack,
determined by using the afore mentioned criteria. Graphs having only one circle represents the pier in
which the diagonal tension crack was also the first significant crack.

50
 Lateral forces at various distress levels, namely first significant cracking, diagonal tension
cracking and maximum resistance, are summarized in Table 3.14. It is worth mentioning that the
lateral strength of piers was found to be approximately same in both directions of loading except piers
PIIa and PIIb as shown in Figures 3.43 and 3.44, respectively.

3.7 DIAGONAL TENSILE SHEAR STRENGTH OF PIERS

Diagonal tensile shear strength of an URM masonry pier is calculated by using Turnšek Equation [TS
80] given as follows:

f σ
τ 1 .
b f

where b = shear correction factor which accounts for the distribution of shear stresses along the pier’s
length. b = 1 for hp/lp ≤ 1.0, b = hp/lp for 1.0 < hp/lp < 1.5 and b = 1.5 for hp/lp ≥ 1.5 [BT 84]. Similarly,
the principal tensile strength of masonry prisms, ftu , is determined from the Equation 3.4b.
The experimental values of the diagonal tensile shear strengths of the tested piers were
compared with the diagonal tensile shear strengths of piers determined from Equation 3.8, as shown in
Figure 3.51. Figure 3.51 shows that Turnšek Equation estimates the diagonal tensile shear strength of
the URM piers with reasonable accuracy as the ratio of the diagonal tensile shear strength of piers
determined from the experiments to the diagonal tensile shear strength determined by Turnšek
Equation is close to 1.0 for most of the tested piers.

3.8 EVALUATION OF DISPLACEMENT BASED PROPERTIES FROM QUASI-

STATIC CYCLIC TESTS

3.8.1 Displacement ductility factor of URBM piers

For estimating ductility piers in a quantitative manner, a bi-linear idealization of the lateral force-
displacement curve recommended by Magenes et al [MC 97], as shown in the Figure 3.52, was used.
The displacement ductility factor for an URM pier, µp, using the recommended criteria, can be
calculated by using the relation:
d
µ .
d
where:
dy = Lateral displacement at the top of a pier at the start of 'yielding ' = Vu / Kp in which Vu = 0.9 Vmax
(Vmax = the maximum resistance of the pier in the lateral direction); Kp = elastic stiffness of the pier =
0.75Vu/ d0.75Vu (d0.75Vu = the displacement corresponding to 0.75Vu); and
du = Lateral displacement at the top of the pier when the lateral force Vu is decreased by 20% along the
degrading curve, as shown in Figure 3.52

51
 Idealized bi-linear curves for tested piers, determined from afore mentioned procedure, are
shown in Figures 3.53 through 3.60. The experimental lateral force-displacement data given in these
figures, correspond to the first cycle of loading. It is worth mentioning that in case of all the piers,
except pier PIVa, the test was stopped due to extensive damage before the lateral force degraded to
0.8Vu. Consequently, calculation of displacement ductility factor was controlled by excessive damage
of piers instead of Equation 3.9. This can also be seen from the results of tested piers given in Table
3.15 where Vmin/ Vu > 0.8 for all the piers, except PIVa.
Displacement ductility factors of the tested piers are given in Table 3.15. The average value of
µp for tested piers was found to be 6.31 with a C.O.V. of 26.3%. It is important to note that for the
purpose of conservativeness, lower of the two values of µp for each direction of loading of a tested pier
were considered in statistical analysis. Additionally, data of pier series PIV was not considered in
statistical analysis due to their different behavior as compared to the other piers.
The influence of pre-compression level on displacement ductility factor was also
investigated. It is reported by Vasconcelos [Va 04] that the displacement ductility factors for URM
piers decreases with the increase in pre-compression level. In order to further investigate this
observation, regression analysis values was carried out for the displacement ductility factors of piers
(lower of two values for each loading direction) belonging to PII and PIII series and corresponding
pre-compression, as shown in Figure 3.61. However, no correlation was found between displacement
ductility factors and pre-compression levels as indicated by the coefficient of determination, r2, which
was as low as 0.0012. However, relatively good correlation was found between drift ratio at ultimate
limit state and pre-compression level (r2 = 0.497), as shown in Figure 3.62. However, these statistical
findings correspond to very limited data. Further tests in this regard will aid to a better understanding
of the phenomena.

3.8.2 Modulus of rigidity of the masonry

As discussed in Section 3.3.3.4, the modulus of rigidities of the masonry calculated from the diagonal
tension tests of masonry prism were not realistic. Another approach for determining the modulus of
rigidity of masonry is based on the results of quasi-static cyclic tests of masonry piers. Using this
approach, modulus of rigidity of masonry, Gm, is calculated by considering elastic stiffness of the
idealized bi-linear lateral force-displacement curve of the pier, as shown in Figure 3.52, with the
following relation [PB 98]:
V .h MV . h 1.2V . h
d .
3E . I 2E . I G .A
where (MVu)t is the moment at the top end of the pier corresponding to Vu. (MVu)t is taken as positive
as long as the pier bend in double curvature and vice versa.
Values of Gm for tested piers, using Equation 3.10, were determined by using an average
value of Em = 188358 psi as given in Table 3.8. The average values of Gm/Em and Gm,, as given in
52
Table 3.16, were found to be 0.56 and 116762 psi, respectively, with a C.O.V. of 62.3%. One of the
reason for such high variation is the result of pier PIVb which, as given in Table 3.16, having very
elastic high stiffness. The average value of Gm /Em for pier PIVb was found to be 1.36, which is very
high as compared to the range of 0.1 - 0.4 reported in literature [To 99]. However, C.O.V. was
significantly reduced by excluding the data of pier PIVb. By doing so, average values of Gm/Em and
Gm were reduced to 0.45 and 93047 psi, respectively, with a C.O.V. of 23.1%.
Modulus of rigidities of the tested masonry piers at the formation of first significant crack,
Gm,cr, were also determined for the purpose of comparison with Gm. The average values of Gm,cr /Em
and Gm,cr, by considering the lateral force-displacement values at the formation of first significant
crack, were found to be 0.42 and 87761 psi, respectively, with a C.O.V. of 38.8%. Again, by excluding
the data of pier PIVb, these values were reduced to 0.37 and 76814 psi, respectively, with a C.O.V. of
19.8%
Correlation between Gm and Gm,cr was also investigated. This was carried out by using the
average values of Gm of all the tested piers in both the directions of loadings. The mean value of
Gm,cr for all tested piers was found to be 0.81Gm with a C.O.V. of 23.8%

3.8.3 Energy dissipation

It was found that C.O.V. for drift ratios of tested piers, excluding PIV series, corresponding to
diagonal tensile cracking was 36 % when displacement was determined from visual observation.
However, C.O.V. reduced to 8.8 % when displacement corresponding to diagonal tensile cracking
was determined from energy dissipation characteristics of piers. Therefore, in this study, lateral forces
at first significant cracking and diagonal tensile shear cracking were determined from a careful study
of the magnitudes of the ratios of the dissipated energy to the input energy values for each
displacement cycle. For this purpose, coefficient of equivalent viscous damping, ξeq, was calculated
for each cycle of imposed force/displacement by using the following equation[Ch 01]:
E
ξ .
2πE
where Ei = energy required in one cycle to impose a target lateral force/displacement in the pier and is
determined by computing the area of force-displacement diagram, both in positive and negative
loading directions, as shown in Figure 3.63a; and
Ed = Energy dissipated in one cycle of imposed lateral force/displacement and is determined by
computing the areas under the positive and negative branches of force displacement hysteresis loops,
as shown in fig 3.63b.
Whenever a crack develop in a pier, the dissipated energy, Ed, is increased due to permanent
deformation in the pier indicated by an increase in the slope of unloading lateral force-displacement
curve. The increase in Ed is accompanied by an increase in the
ξ as ξ α E as indicated by Equation 3.11 . For the purpose of determining the lateral force
53
causing significant crack, ξ was calculated for first cycle of every imposed displacement. The first
cycle with a sudden increase in ξ indicated that significant crack have developed. Exact value of
cracking load and corresponding displacement was determined from a significant decrease in the slope
of lateral force-displacement diagram of the pier. The point at which the slope of lateral-force
displacement curve decrease significantly, counter checked with visual observation, provided the force
and displacement corresponding to significant/ diagonal tension shear cracking.
Variation in the coefficient of equivalent viscous damping for first cycle of imposed force/
displacement against drift ratios, for various tested piers, are shown in Figure 3.68. It can be concluded
from Figures 3.64 through 3.68, that the coefficient of equivalent viscous damping for most of the
piers, increased at a higher rate when the drift ratio was increased from 0.05% (average drift ratio
corresponding to the formation of flexural- tensile cracking in the bed joints of piers) to 0.1% (average
drift ratio corresponding to the initiation of diagonal tensile shear cracking). ξeq increased at a
comparatively lower rate when the drift ratio increased from 0.1% to 0.2% (average value of drift
ratio corresponding to maximum lateral resistance). The coefficient of equivalent viscous damping,
except piers belonging to PII series, became constant for drift ratios greater than 0.2 %, as shown in
Figures 3.64 through 3.68
Coefficient of equivalent viscous damping, ξeq, of slender piers (PIc) was found to be
relatively lower as compared to other piers, as shown in Figure 3.68. One of the reason for this
behavior may be possible rocking.
Similarly, variation in ξeq for pier PIVb, as shown in Figures 3.67 and 3.68, was relatively higher as
compared to other piers. One of the reason for this behavior may be severe damage due to the
formation of multiple inclined cracks on the right side of pier, as shown in Figure 3.39. These cracks,
later on, joined main diagonal crack and enhanced the severity of damage in corresponding diagonal
wedge which was about to fall during final stages of testing, as shown in Figure 3.40
The effect of pre-compression on ξeq was also studied and graphically shown in Figure 3.69. It
can be concluded, as shown in Figure 3.69, that an increase in the pre-compression is accompanied by
a decrease in the coefficient of equivalent viscous when the drift ratio exceed 0.2% (average value of
the drift ratio corresponding to lateral strength of piers). In the other words, increase in the pre-
compression reduces the rate of damage after the piers attains their maximum resistance.

3.8.4 Stiffness degradation

It is a well known phenomenon that stiffness of any member/structure decreases with an increase in
the deformations beyond the elastic limit. Degradation of stiffness with the increase in drift ratio for
the tested piers was studied by plotting graphs in a dimensionless form, with K/Kp on y-axis and drift
ratio, δ, on x-axis. Where, K represent lateral stiffness of a pier at a lateral displacement d, and Kp
represent elastic stiffness of pier determined from the idealized bi-linear lateral force displacement

54
curve of the pier by Magenes and Calvi criterion [MC 97], as shown in Figure 3.52. Results of K/Kp
vs δ in % are graphically represented in Figures 3.70 through 3.74.
It was expected that pier PIc, being slender of all piers, will undergo less stiffness degradation
with increase in drift ratio due to possible rocking effect. This was found to be true for negative
loading direction (NLD), Figure 3.74b. However, surprisingly, stiffness degradation for the same pier
was found to be more as compared to other piers for positive loading direction (PLD), Figure 3.74a.
Similarly, stiffness degradation was found to be maximum for pier PIVb and minimum for PIVa, as
shown in Figures 3.74a and b. Severe damage in pier PIVb due to formation of multiple inclined
cracks may be one of the reason for such behavior.
To study the influence of pre-compression on stiffness degradation, δ vs. K/Kp curves for
piers belonging to PII and PIII series were plotted in Figure 3.75. It was found that stiffness
degradation decrease with an increase in pre-compression level. This was found to be true for negative
loading direction, however, this effect was not significant in case of positive loading direction.

3.9 DRIFT RATIOS CORRESPONDING TO VARIOUS PERFORMANCE LEVELS

FOR TESTED PIERS
Behavior of members/ structures based on drifts is the backbone of performance-based concepts. An
example of which is FEMA 356 [FEMA 356] which recommends drift based performance levels for
URM buildings. A criteria for performance based evaluation of URM piers, failing in shear, based on
results given in Table 3.17 is recommended as a part of this study. Table 3.17 summarizes the drift
ratios that can be related, at least for special type of brick masonry piers constructed in stone dust
mortar, with various limit states. It may be noted that data of pier series PIV was not included in
statistical analysis due to different behavior as compared to the other piers.
Normally, four performance levels are defined for performance based evaluation, namely
operational (O), immediate occupancy (IO), life safety (LS) and collapse prevention (CP). Based on
the damage described in their definitions they can be related to various levels of distress. e.g., minimal
damage that can occur in an URM piers during a seismic activity is the flexural tensile cracking of bed
joints. These cracks are some times too narrow to be visible after the seismic event. Drift ratio
corresponding to flexural cracking of bed joint was related to the operational level (O). IO
performance level can be related to a drift ratio corresponding to the formation of first diagonal tension
crack. It was observed in the testing work that the diagonal crack just after its formation is too narrow
to be seen, as shown in Figures 3.9, 3.11and 3.38, and maintain the integrity of the pier. Average value
of the drift ratio, for piers (excluding PIV series) at the formation of diagonal tensile cracking was
found to be 0.088% with a C.O.V. of 8.8 %.

55
 Similarly, LS performance level can be related to a drift ratio when a pier just achieves its
maximum resistance. Average value of the drift ratio at maximum resistance (excluding the data of
pier series PIV) was found to be 0.22% with a C.O.V. of 35.1%.
Similarly, it was observed during the testing that in spite of extensive damage at the end of
test, the piers were still able to support vertical load and did not collapse( see Figures 3.17, 3.28 and
3.32). Therefore, drift ratio at a stage when either the shear resistance dropped to 80% of Vu or the test
was stopped due to extensive damage (which ever occurred first) can be established as a basis for CP
performance level. Mean value of the drift ratio at this level (excluding the data of pier series PIV) was
found to be 0.46% with a C.O.V. of 26.8%.
Due to the scatter in the mean values of drift ratios corresponding to IO, LS and CP
performance levels, it is recommended, for the purpose of design/assessment of URBM buildings
constructed in solid fired clay bricks using stone dust mortar, to adopt mean minus standard deviation
values for various performance levels. For IO, LS and CP performance levels these values after
rounding were determined be 0.08%, 0.15% and 0.35%, respectively.

3.10 RESULTS, CONCLUSIONS AND RECOMMENDATIONS

Experimental work was carried out in order to determine mechanical properties of masonry and its
constituent used in Northern Pakistan. In-plane quasi-static cyclic tests were also conducted to study
the various properties, related to seismic performance, of URBM pier
Results obtained from experimental work along with the conclusions and recommendations are given
as follows:
1. Average value of compressive strength of masonry, fm, was found to be 658 psi. Similarly,
ratio of principal tensile strength to the compressive strength of masonry, ftu/fm was found to
be 3.72%, which is with in the range of 1-8%, as reported in literature [To 99].
2. Ratio of modulus of elasticity of masonry to the compressive strength of masonry, Em/fm, was
found to be 294
3. Modulus of rigidity of masonry, Gm, determined from diagonal tension tests on prisms was
found un-realistic. The average value of Gm/Em from these tests was found to be 7.68 which is
too much high as compared to a range of 0.1-0.4 reported in literature [To 99].
4. Values of Gm/Em from quasi-static cyclic tests on masonry piers were found to be satisfactory.
Average value of Gm/Em determined from these tests was 0.45.
5. Modulus of rigidity of masonry piers at the formation of first significant crack, Gm,cr , were
also determined for the purpose of comparison with Gm. The average values of Gm,cr /Em was
found to be 0.37
6. Two failure modes were observed during testing the piers. Piers corresponding to PI to PIII
series, with aspect ratios of 0.94 and 1.22, failed in diagonal tension. Where as piers

56
 corresponding to series PIV, which were squatter as compared to other piers with an aspect
ratio of 0.66, failed in diagonal tension accompanied by sliding of bed joint/joints.
7. It was observed, by comparing the cracking patterns of piers corresponding to PII and PIII
series, that the number of cracks in the piers PII series were less but more wider than that of
PII series.
8. Turnšek Equation [TS 80] was found to predict the diagonal tensile shear strength of URM
piers strength with reasonable accuracy in case of most of the piers, provided the principal
tensile strength of masonry is determined by using RILEM Equation [RILEM LUM B6].
9. Tests were stopped in all piers, except PIVa, without any significant degradation of lateral
force after the piers gained their maximum resistance. Consequently the displacement
ductility factors were determined from the ultimate lateral displacement at the end of test,
instead of Magenes and Calvi criteria of lateral displacement corresponding to 0.8Vu
10. The average value of µp for tested piers was found to be 6.31. For the purpose of
conservativeness, the lower of two values of µp for each pier were considered for determining
µp.
11. No correlation was found between displacement ductility factors and pre-compression levels
for tested piers (r2 = 0.0012). However, relatively better correlation was found between drift
ratio at ultimate limit state and pre-compression level (r2 = 0.497). However, these statistical
findings correspond to very limited data.
12. By comparing the results of piers of PII and PIII series, it is concluded that an increase in the
pre-compression cause reduction in coefficient of equivalent viscous damping after the piers
attains their maximum resistance.
13. By comparing the results of piers of PII and PIII series, it is concluded that increase in pre-
compression has no effect on stiffness degradation in piers with the increase in drift.
14. It is proposed , for the purpose of design/assessment of URM piers constructed in solid fired
clay bricks using stone dust mortar, to use drift ratios of 0.08%, 0.15% and 0.35%, for
Immediate Occupancy, Life Safety and Collapse Prevention performance levels in piers,
respectively. The Operational performance level for the pier can be related to drift ratio at the
formation of flexural tensile cracking in bed joint of pier

57
Figure 3.1 Typical masonry prism used for determining compressive strength of masonry

Figure 3.2 Typical arrangement of displacement dials gages for determining

vertical deformations in a masonry prism.

58
 (a)

(b)

Figure 3.3 Testing setup used for conducting diagonal tension tests (a) Front view of prism
with LVDT installed for measuring horizontal elongation (b) Rear view of prism with
LVDT for determining vertical shortening.

59
 (a)

(b)
Figure 3.4 Testing arrangement for carrying out triplet tests
(a) Specimen without pre-compression
(b) Specimen with pre-compression applied horizontally

60
 200

150
Shear strength, τ u- psi

100
τu = 0.588σn +56.84
2
τu = 0.588σn + 56.84
r = 0.827 r2 = 0.827
50

0
0 50 100 150 200 250
Pre-compression, σn- psi

Figure 3.5 Results of regression analysis for determining cohesion and coefficient of friction
of masonry from triplet tests
61
 Steel Beam
Providing
Reation to
Jack Forces

Steel Beam Steel Beam

LVDT-1

R.C Beam

Masonry wall

Actuator
LVDT-5
LVDT-4

LVDT-3 LVDT-2

R.C Beam anchored to test floor

Figure 3.6 Testing setup used for conducting in-plane quasi-static cyclic tests on URBM piers

62
 (a) (b)

(c)
(d)

Figure 3.7 Details of testing setup used for conducting quasi-static cyclic tests on URBM piers
(a) Foundation beam of the pier connected to the bottom R.C. Beam of testing setup via
epoxy resin
(b) R.C. beam anchored to the test floor
(c) Piers connected to the top R.C. beam of the testing setup via epoxy resin
(d) Actuator, connected to the testing setup by vertical steel beam for applying lateral
force on the pier

63
 0.4
Force Controlled cycles
Displacement controlled cycles
0.3

0.2
Displacement (in.)

0.1

0
0 500 1000 1500 2000 2500 3000 3500 4000 4500

-0.1

-0.2

-0.3

-0.4
Time (sec)

Figure 3.8 Typical lateral displacement history imposed

on the piers’ top end during quasi-static cyclic tests

Figure 3.9 Pier PIa after diagonal tension cracking

64
Figure 3.10 Multiple inclined cracks along each diagonal of pier PIb

Figure 3.11 Diagonal tension cracks in pier PIc

65
 Figure 3.12 Pier PIc at end of testing (Also note vertical cracks in the middle portion).

Figure 3.13 Spalling of masonry due to crushing and in pier PIc during the final stages of testing

66
 Figure 3.14 Seperation of the diagonally cracked ‘wedge’ due to
the crushing of masonry at the bottom of pier PIc

Figure 3.15 Vertical splitting of masonry across the head joints on the right end of pier PIIa

67
 Figure 3.16 Excessive compressive stresses in the
masonry at the bottom of pier PIIc caused crushing

Figure 3.17 Extensive sliding of damaged ‘wedge’, on the right end of pier PIIa, resulted in
terminating the test in spite of small degradation of lateral strength

68
Figure 3.18 A closer view of pier PIIa. Extensive sliding of diagonally cracked
wedge caused its separation from rest of the pier.

Figure 3.19 Separation of diagonally cracked ‘wedge’

on the left end of pier PIIb at the end of testing

69
Figure 3.20 Excessive crushing at the bottom end of pier PIIb resulted in
separation of diagonally cracked ‘wedge’ from the rest of pier.

Figure 3.21 A view of pier PIIc at the end of testing

70
 Figure 3.22 Crushing of brick at the bottom of pier

Figure 3.23 Multiple inclined cracks along each loading direction in pier PIIIa.

71
 Figure 3.24 Vertical cracks along both the ends of pier PIIIb

Figure 3.25 Vertical splitting of masonry and crushing of bricks in pier PIIIa.

72
 Figure 3.26 Crushing of bricks at the top end of pier PIIIb

Figure 3.27 Horizontal and vertical splitting of masonry on right end of pier PIIIa

73
 Figure 3.28 Vertical splitting and crushing of masonry on the top and
bottom corners in pier PIIIa

Figure 3.29 Pier PIVc at the end of test under a pre-compression of 61 psi

74
 Figure 3.30 Branching of vertical crack in pier PIVc

Figure 3.31 Pier PIVc after testing under a pre-compression of 93 psi

75
Figure 3.32 Vertical gap between top surface of pier PIVc and concrete beam at the top end

Figure 3.33 Splitting of masonry in pier PIVc due to the concentration of

vertical stresses at the right end

76
Figure 3.34 Pier PIVa just before failure of epoxy bond between foundation
beam of wall and R.C. beam anchored to the testing floor

Figure 3.35 Sliding of bed joint in pier PIVa at third course from bottom

77
 Figure 3.36 Pier PIVa after testing. Note damage to brick units and
vertical cracks at the right end

Figure 3.37 Vertical splitting of masonry across all the head joints of pier PIVa

78
 Figure 3.38 Pier PIVb just after the formation of diagonal tension cracks

Figure 3.39 Bed joint sliding in pier PIVb at fourth layer from top end of pier.
Note sliding in other courses and formation of multiple inclined cracks.

79
Figure 3.40 Separation of diagonally cracked ‘wedge’ due to the formation of
multiple inclined cracks on the right end of pier PIVb.

Figure 3.41 Right end view of pier PIVb ,

80
 30

20

Force (kip) 10

0
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5

-10

-20

-30

Displacement (in.)
(a)
30
Positive loading direction
Negative loading direction

20
Force (kip)

10

0
0.00 0.10 0.20 0.30 0.40 0.50

Displacement (in.)

(b)

Figure 3.42 Lateral load – displacement diagrams for pier PIc (a) Hysteresis
loops (b) Envelope of first cycle of loading

81
 40

30

20

10
Force (kip)

0
-0.3 -0.2 -0.1 0 0.1 0.2 0.3

-10

-20

-30

-40

Displacement (in.)
(a)
40
Positive loading direction
Negative loading direction

30
Force (kip)

20

10

0
0.00 0.10 0.20 0.30

Displacement (in.)

(b)
Figure 3.43 Lateral load – displacement diagrams for pier PIIa (a) Hysteresis
loops (b) Envelope of first cycle of loading

82
 30

20

10
Force (kip)

0
-0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2

-10

-20

-30
Displacement (in.)
(a)
30
Positive loading direction
Negative loading direction

20
Force (kip)

10

0
0.00 0.05 0.10 0.15 0.20

Displacement (in.)

(b)

Figure 3.44 Lateral load – displacement diagrams for pier PIIb (a) Hysteresis
loops (b) Envelope of first cycle of loading

83
 30

20

10
Force (kip)

0
-0.3 -0.2 -0.1 0 0.1 0.2 0.3

-10

-20

-30

Displacement (in.)

(a)
30
Positive loading direction

Negative loading

20
Force (kip)

10

0
0.00 0.05 0.10 0.15 0.20 0.25

Displacement (in.)
(b)
Figure 3.45 Lateral load – displacement diagrams for pier PIIc (a) Hysteresis
loops (b) Envelope of first cycle of loading

84
 30

20

10
Force (kip)

0
-0.3 -0.2 -0.1 0 0.1 0.2 0.3

-10

-20

-30

Displacement (in.)

(a)
30
Positive loading direction
Negative loading direction

20
Force (kip)

10

0
0.00 0.10 0.20 0.30

Displacement (in.)
(b)
Figure 3.46 Lateral load – displacement diagrams for pier PIIIa (a) Hysteresis
loops (b) Envelope of first cycle of loading

85
 30

20

10
Force (kip)

0
-0.3 -0.2 -0.1 0 0.1 0.2 0.3

-10

-20

-30
Displacement (in.)

(a)
30

20
Force (kips)

10

Positive loading direction

Negative loading direction
0
0.0 0.1 0.2 0.3

Displacement (in.)

(b)

Figure 3.47 Lateral load – displacement diagrams for pier PIIIb (a) Hysteresis
loops (b) Envelope of first cycle of loading

86
 40

30

20

10
Force (kip)

0
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
-10

-20
(a)
-30

-40
Displacement (in.)
(a)

40
Positive loading direction
Negative loading direction

30
Force (kip)

20

10

0
0.0 0.1 0.2 0.3 0.4 0.5

Displacement (in.)

(b)

Figure 3.48 Lateral load – displacement diagrams for pier PIVa tested at pre-
compression of 93 psi (a) Hysteresis loops (b) Envelope of first cycle of loading

87
 40

30

Force (Kips) 20

10

0
-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1
-10

-20

-30

-40
Displacement (inches)

(a)
40

30
Force (Kips)

20

10 Positive loading direction at 61 psi

Negative loading direction at 61 psi
Positive loading direction at 93 psi
Negative loading direction at 93 psi
0
0 0.1 0.2 0.3 0.4 0.5

Displacement (inches)

(b)

Figure 3.49 (a) Hysteresis loop for pier PIVa tested at a pre-compression of 61 psi
(b) Envelopes of first cycle of loading for pier PIVa tested at 61 and 93 psi

88
 40

30

20

10
Force (kip)

0
-0.3 -0.2 -0.1 0 0.1 0.2 0.3
-10

-20

-30

-40

Displacement (in.)
(a)
40
Positive loading direction
Negative loading direction

30
Force (kip)

20

10

0
0.0 0.1 0.2 0.3

Displacement (in.)

(b)
Figure 3.50 Lateral load – displacement diagrams for pier PIVb tested at pre-
compression of 93 psi (a) Hysteresis loops (b) Envelope of first cycle of loading
89
 2.0

1.62
VExperimental / VTurnšek Equation
1.5

1.21 1.25

1.06 0.95
0.97 0.99
1.0
0.90

0.5

0.0
PIc PIIa PIIb PIIc PIIIa PIIIb PIVa PIVb
Pier designation

Figure 3.51 Comparison of diagonal tensile shear strength of piers obtained from experimental
work with the diagonal tensile shear strength of piers determined from Turnsek equation.

Figure 3.52 Bi-linear idealization of URM piers based on equal energy principal.
Vu = 0.9 Vmax for shear failure [MC 97]

90
 30

Positive loading direction

Negative loading direction

20
Force (kip)

10

0
0.0 0.1 0.2 0.3 0.4 0.5
Displacement (in.)

Figure 3.53 Bi-linear idealized curves for pier PIc. p =10.69 and 8.54 for positive
and negative loading directions, respectively.

40
Positive loading direction
Negative loading direction

30
Force (kip)

20

10

0
0.0 0.1 0.2 0.3

Displacement (in.)

Figure 3.54 Bi-linear idealized curves for pier PIIa. p =4.86 and 11.03 for
positive and negative loading directions, respectively.

91
 30
Positive loading direction
Negative loading direction

20
Force (kip)

10

0
0.00 0.05 0.10 0.15 0.20

Displacement (in.)

Figure 3.55 Bi-linear idealized curves for pier PIIb. p =5.16 and 9.09 for positive
and negative loading directions, respectively.

30
Positive loading direction
Negative loading direction

20
Force (kips)

10

0
0.0 0.1 0.2 0.3
Displacement (in.)

Figure 3.56 Bi-linear idealized curves for pier PIIc. p =7.68 and 12.62 for
positive and negative loading directions, respectively.

92
 30

20
Force (kip)

10

Positive loading direction

Negative loading direction
0
0.0 0.1 0.2 0.3
Displacement (in.)

Figure 3.57 Bi-linear idealized curves for pier PIIIa. p =4.59 and 11.16 for
positive and negative loading directions, respectively.

30

20
Force (kip)

10

0
0.00 0.10 0.20 0.30
Displacement (in.)

Figure 3.58 Bi-linear idealized curves for pier PIIIb. p =7.03 and 1.21 for
positive and negative loading directions, respectively.

93
 40
Positive loading direction
Negative loading direction

30
Force (kip)

20

10

0
0.0 0.1 0.2 0.3 0.4 0.5
Displacement (in.)

Figure 3.59 Bi-linear idealized curves for pier PIVa. p = 8.71 and 7.24 for
positive and negative loading directions, respectively.

40

30
Force (kip)

20

10

Positive loading direction

Negative loading direction
0
0.0 0.1 0.2 0.3

Displacement (in.)

Figure 3.60 Bi-linear idealized curves for pier PIVb. p =18.66 and 32.46 for
positive and negative loading directions, respectively.

94
 9

Displacement ductility factor, µp

3 µp = -0.0028σo + 6.0716
2
r = 0.0012

0
0 20 40 60 80 100
Pre-compression,σo (psi)

Figure 3.61 Effect of pre-compression on displacement ductility of tested piers

1
Utimate drift ratio, δu (percent)

0.75

0.5

0.25

δu = 0.0037σo + 0.161
r2 = 0.497
0
0 20 40 60 80 100

Pre-compression, σo (psi)

Figure 3.62 Effect of Pre-compression on ultimate drift ratio of tested piers

95
 Force
gy ed
er at
Positve Input

En issip
D
Energy

Negative Input
Energy

Displacement
Displacement

Figure 3.63 Energy determination in one cycle of loading [Va 04] a. Dissipated energy b. Input energy

96
 16
PIc

12

ξeq (percent)

0
0.0 0.1 0.2 0.3 0.4 0.5 0.6

Drift ratio, δ (percent)

Figure 3.64 Influence of the drift ratio on the coefficient of equivalent viscous damping (PI series)

16
PIIa
PIIb
PIIc
12
ξeq (percent)

0
0.0 0.1 0.2 0.3 0.4 0.5 0.6

Drift ratio, δ (percent)

Figure 3.65 Influence of the drift ratio on the coefficient of equivalent viscous damping (PII series)

97
 16
PIIIa
PIIIb

12

ξeq (percent)

0
0.0 0.1 0.2 0.3 0.4 0.5 0.6

Drift ratio, δ (percent)

Figure 3.66 Influence of the drift ratio on the coefficient of equivalent viscous damping (PIII series)

16
PIVa
PIVb

12
ξeq (percent)

0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Drift ratio, δ (percent)

Figure 3.67 Influence of the drift ratio on the coefficient of equivalent viscous damping (PIV series)

98
 16

12
ξeq (percent)

8 PIc
PIIa
PIIb
PIIc
4
PIIIa
PIIIb
PIVa
PIVb
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2

Drift ratio, δ (percent)

Figure 3.68 Influence of the drift ratio on the coefficient of equivalent viscous damping (All piers)

99
 16

ξeq (percent) 12

PIIa
PIIb
4
PIIc
PIIIa
PIIIb
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6

Drift ratio, δ (percent)

Figure 3.69 Influence of the pre-compression on the coefficient of equivalent viscous damping.
σo = 61 psi and 93 psi for Piers belonging to PII and PIII series, respectively

100
 3.0

PIc (PLD) PIc (NLD)

2.5

2.0

K/Kp
1.5

`
1.0

0.5

0.0
-0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8
Drift ratio,δ (percent)

Figure 3.70 Degradation of stiffness with the increase in drift ratio for PI series piers

2.5
PIIa (PLD) PIIa (NLD)

2.0 PIIb (PLD) PIIb (NLD)

PIIc (PLD) PIIc (NLD)

K/Kp

1.5

1.0 `

0.5

0.0
-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6
Drift ratio,δ (percent)

Figure 3.71 Degradation of stiffness with the increase in drift ratio for PII series piers

101
 3.0

PIIIa (PLD) PIIIa (NLD)

2.5

PIIIb (PLD) PIIIb (NLD)

2.0

K/Kp
1.5

1.0

0.5

0.0
-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6
Drift ratio,δ (percent)

Figure 3.72 Degradation of stiffness with the increase in drift ratio for PIII series piers

3.0

PIVa (PLD) PIVa (NLD)

2.5

PIVb (PLD) PIVb (NLD)

2.0
K/Kp

1.5

`
1.0

0.5

0.0
-1.2 -0.9 -0.6 -0.3 0.0 0.3 0.6 0.9 1.2
Drift ratio,δ (percent)

Figure 3.73 Degradation of stiffness with the increase in drift ratio for PIV series piers

102
 3.0
PIc (PLD) PIc (NLD)

PIIa (PLD) PIIa (NLD) 2.5

PIIb (PLD) PIIb (NLD)

2.0
PIIc (PLD) PIIc (NLD)

K/Kp
PIIIa (PLD) PIIIa (NLD)
1.5
PIIIb (PLD) PIIIb (NLD)
`

PIVa (PLD) PIVa (NLD) 1.0

PIVb (PLD) PIVb (NLD)

0.5

0.0
12 09 06 03 00 03 06 09 12
Figure 3.74 Degradation of stiffness with the increase in drift ratio for all piers

103
 3.0
PIIa (PLD) PIIa (NLD)

2.5 PIIb (PLD) PIIb (NLD)

PIIc (PLD) PIIc (NLD)

2.0
PIIIa (PLD) PIIIa (NLD)

K/Kp
PIIIb (PLD) PIIIb (NLD)
1.5

1.0

0.5

0.0
-1.2 -0.9 -0.6 -0.3 0.0 0.3 0.6 0.9 1.2

Figure 3.75 Influence of the pre-compression on the stiffness degradation. σo = 61 psi and 93 psi for
Piers belonging to PII and PIII series, respectively

104
 Table 3.1 Proportions of stone dust aggregates used in mortar

Sieve No. % Weight retained on sieve

4 0

8 22.36

16 22.18

30 20.82

50 27.64

100 7.00

Table 3.2 7-day compressive strength of trial mortar cubes

Compressive strength (psi)

Sample No. w/c =1.1 w/c= 1.4 w/c =1.7

1 645 512 154

2 606 435 148

3 625 338 148

4 637 477 133

5 622 557 128

6 651 486 168

Mean 631 468 147

% C.O.V. 2.6 16.1 9.9

105
 Table 3.3 IRA tests on brick units

Sample Wet weight of sample Dry weight of sample Area of brick IRA
No. (gm) (gm) (in2) (gm/min/30 in2)
1 2400 2325 41.32 54.5

2 2500 2375 42.61 88.0

3 2400 2320 40.92 58.7

4 2450 2350 42.78 70.1

5 2450 2350 39.41 76.1

Mean 69.5

% C.O.V. 19.4

Table 3.4 Compressive strength of brick units (psi)

Length Width Area Maximum Load Strength

S. No.
(in.) (in.) (in2) (Tonne) (psi)
1 9.33 4.45 41.51 62.8 3336

2 9.29 4.41 40.97 52.0 2799

3 9.37 4.49 42.05 62.0 3251

4 9.25 4.37 40.43 85.4 4657

5 9.37 4.25 39.84 64.3 3559

6 9.29 4.33 40.24 45.0 2466

7 9.29 4.37 40.60 42.8 2324

8 9.21 4.25 39.17 47.5 2674

9 9.37 4.49 42.05 63.0 3303

10 9.29 4.37 40.60 45.2 2455

11 9.21 4.33 39.90 80.5 4449

12 9.25 4.25 39.34 55.7 3122

Mean 3200

% C.O.V. 23.4

106
 Table 3.5 Flexural tensile strength of brick units

Ultimate vertical Mid-span Flexural tensile

Sample Thickness Width
load moment strength
No. (in.) (in.)
(Tonne) (in.lb) (psi)
1 0.255 1117.1 2.087 4.409 349.1

2 0.224 983.0 2.047 4.567 308.1

3 0.347 1519.2 2.008 4.409 512.8

4 0.235 1027.7 2.087 4.449 318.3

5 0.255 1117.1 2.087 4.409 349.1

6 0.357 1563.9 2.087 4.449 484.4

7 0.224 983.0 2.087 4.370 310.0

8 0.326 1429.8 2.047 4.370 468.4

Mean 387.5

% C.O.V. 22.2

107
 Table 3.6 Compressive strength of CSK 1:4:4 mortar

Sample no. Ultimate load Compressive strength Age at testing

(Tonne) (psi) (days)
1 2.50 726 30

2 2.20 639 30

3 2.55 740 30

4 2.30 668 30

5 2.30 668 30

6 2.50 726 30

7 2.20 639 28

8 2.05 595 28

9 2.35 682 28

10 2.25 653 28

11 2.40 697 28

12 2.05 595 28

13 2.55 740 26

14 2.75 798 26

15 2.6 755 26

16 2.65 769 26

17 2.40 697 26

18 2.60 755 26

19 2.75 798 25

20 2.55 740 25

21 2.85 827 25

22 3.15 915 25

23 2.50 726 25

Mean 719

% C.O.V. 10.3

108
 Table 3.7 Compressive strength of brick masonry prisms

Sample Height Ultimate Load Area Compressive strength

No. (in.) (Tonne) (in2) (psi)
1 22.44 30.00 85.60 772.8

2 22.63 26.00 86.32 664.2

3 22.83 21.37 89.28 527.8

4 23.03 28.40 86.69 722.4

5 22.64 22.08 87.06 559.3

6 22.64 24.68 87.43 622.4

7 22.44 24.33 87.80 611.0

8 22.64 25.39 87.80 637.6

9 22.64 31.80 87.06 805.4

Mean 658.1

% C.O.V. 14.2

Table 3.8 Modulus of elasticity of brick masonry prisms

Sample Compressive strength, fm Modulus of elasticity, Em

No. (psi) (psi) Em/ fm
1 772.8 157330 203.6

2 664.2 199130 299.8

3 527.8 209290 396.5

4 722.4 172860 239.3

5 559.3 205810 368.0

6 622.4 228450 367.0

7 611.0 127720 209.0

8 637.6 221770 347.8

9 805.4 172860 214.6

Mean 188358 294

% C.O.V. 17.5 26.6

109
 Table 3.9 Principal tensile strengths of masonry prisms
Sample length Thickness Ultimate load Tensile strength
No. (in.) (in) (Tonne) (psi)

1 29.50 9.45 7.44 29.42

2 29.50 9.45 5.34 21.12

3 29.50 9.45 5.35 21.16

4 29.50 9.45 6.65 26.30

Mean 24.5

% C.O.V. 16.7

Table 3.10 Modulus of rigidity of masonry prisms

determined from diagonal tension tests

Sample No. Gm (psi) Gm/Em

1 531795 2.82

2 65162 0.35

3 439576 2.33

4 4749153 25.21

Mean 7.68

% C.O.V. 153

110
 Table 3.11 Results of triplets tests

Area of brick Normal Ultimate shear Pre-compression, Shear strength, τu

Sample
mortar joint, Abm force, N force, Vus σn = N/Abm = Vus/(2.Abm)
No.
(in2) (Tonne) (Tonne) (psi) (psi)
1 31.68 0 1.42 0 49.25

2 31.68 0 1.20 0 41.71

3 31.68 0 1.56 0 54.26

4 31.68 1 3.13 69.61 108.62

5 31.68 1 3.05 69.61 105.98

6 31.68 1 2.46 69.61 85.54

7 31.68 2 5.46 139.22 189.85

8 31.68 2 4.20 139.22 146.06

9 31.68 2 4.11 139.22 142.72

10 31.68 3 4.78 208.83 166.08

11 31.68 3 4.30 208.83 149.46

Table 3.12 Characteristics of pier specimens

Relative
Distance of lateral
Pier Height, Length, Thickness, Aspect ratio Pre-compression,
force from the pier’s pre-compression
Series hp (ft) lp (ft) t (in.) = hp/lp σo (psi)
bottom, ho (ft)
= σo/ fm

PI 5.45 4.46 9.29 1.22 3.07 103 0.153

PII 4.20 4.46 9.29 0.94 2.30 61 0.091

PII 4.20 4.46 9.29 0.94 2.30 93 0.138

PIV 2.95 4.46 9.29 0.66 1.69 93 0.138

111
 Table 3.13 Typical forces and displacements applied on piers during cyclic tests

Cycle No. Target Force Remarks

V
1
4
V
2
2
3V Force-controlled phase of testing.
3 Vest = estimated lateral strength of pier
4
4 V

Target Displacement, in.

5 0.02

6 0. 03

7 0. 04

8 0. 06

9 0. 08

10 0.10
Displacement-controlled phase of testing
11 0.12

12 0.14

13 0.16

14 0.20

15 0.24

16 0.28

17 0.32

18 0.36

19 0.40

20 0.44

21 0.48

22 0.52

23 0.56

24 0.60

112
 Table 3.14 Lateral Forces in masonry piers at various levels of distress during testing

Vfcr Vdcr Vmax Vmin

Loading
Pier designation Failure Mode
Direction
(kip) (kip) (kip) (kip)
+ 20.71 23.61 26.17 21.05
PIc Diagonal Tension
- 24.37 17.84

+ 14.16 30.37 28.78

PIIa Diagonal Tension
- 12.37 17.81 14.98

+ 23.16 25.18 20.47

PIIb Diagonal Tension
- 17.94 20.82 17.75

+ 18.84 18.84 21.94 19.47

PIIc Diagonal Tension
- 19.04 17.17

+ 25.38 25.38 28.69 22.98

PIIIa Diagonal Tension
- 27.81 20.89

+ 20.91 21.54 27.00 25.21

PIIIb Diagonal Tension
- 23.72 22.06

PIVa at + 25.45 25.45

σo = 61 psi - Diagonal Tension followed by Bed Joint
Sliding
PIVa re-tested at + 36.29 19.71
σo = 93 psi - 39.03 24.94

+ 29.72 29.72 37.12 29.82

PIVb Diagonal Tension followed by Bed Joint
- 34.02 25.28 Sliding

113
 Table 3.15 Displacement ductility factors for tested pier specimens

Pier Vu dy du u Kp
Loading
designat µp
direction
ion (kip) (in.) (in.) (%) (kip/in.)

+ 23.56 0.89* 0.033 0.348 0.53 10.69 714

PIc - 21.96 0.81* 0.048 0.412 0.63 8.54 458

+ 27.34 1.05* 0.029 0.139 0.28 4.86 943

PIIa
- 16.03 0.93* 0.022 0.246 0.49 11.03 729

+ 22.66 0.90* 0.030 0.157 0.31 5.16 755

PIIb
- 18.75 0.95* 0.017 0.157 0.31 9.09 1103

+ 19.74 0.99* 0.030 0.234 0.46 7.68 658

PIIc
- 17.13 1.00* 0.019 0.232 0.46 12.62 902

+ 25.81 0.89* 0.035 0.159 0.32 4.59 737

PIIIa
- 25.04 0.83* 0.026 0.293 0.58 11.16 963

+ 24.60 1.02* 0.041 0.284 0.56 7.03 600

PIIIb
- 21.36 1.03* 0.027 0.279 0.55 10.21 791
**
+ 32.67 0.60 0.047 0.370 1.05 8.71 695
PIVa
**
- 35.34 0.71 0.042 0.274 0.77 7.24 841

+ 33.90 0.88* 0.014 0.255 0.72 18.66 2421

PIVb
- 30.67 0.82* 0.007 0.239 0.68 32.46 4381

*
Test was stopped before lateral force reduced to 0.8 Vu

**
Results of pier re-tested at a pre-compression level of 93 psi

114
 Table 3.16 Modulus of rigidity of masonry determined from quasi-static cyclic tests on
piers

Elastic section Cracked section

Pier designation Loading

Direction Gm (psi) Gm,cr (psi)
Gm/Em Gm,cr/Em

+ 0.55 113508 0.50 104404

PIc
- 0.34 71310

+ 0.56 115488
PIIa
- 0.42 86786 0.38 79047

+ 0.43 89933
PIIb
- 0.63 130898 0.43 88804

+ 0.38 78358 0.31 63515

PIIc
- 0.54 112740

+ 0.43 90040 0.34 70159

PIIIa
- 0.56 115353

+ 0.36 73532 0.30 62886

PIIIb
- 0.46 94584

PIVa at + 0.33 68881

σo = 61 psi - 0.50

PIVa re-tested at + 0.28 58672

σo = 93 psi - 0.34 71450

+ 1.02 210940 0.79

PIVb 164390
- 1.71 354591

115
Table 3.17 Lateral forces and corresponding drifts at various stages of loading during the test

Diagonal cracking Maximum resistance

Ultimate limit state
limit state limit state
Loading
Pier designation ddcr (in) % δdcr dVmax (in) % δVmax du (in) % δu
Direction
+ 0.058 0.090 0.196 0.30 0.348 0.53
PIc
- 0.170 0.26 0.412 0.63

+ 0.048 0.054 0.096 0.19 0.139 0.28

PIIa
- 0.121 0.24 0.246 0.49

+ 0.047 0.103 0.070 0.14 0.157 0.31

PIIb
- 0.065 0.13 0.157 0.31

+ 0.036 0.087 0.191 0.38 0.234 0.46

PIIc
- 0.091 0.18 0.232 0.46

+ 0.044 0.097 0.070 0.14 0.159 0.32

PIIIa
- 0.096 0.19 0.293 0.58

+ 0.046 0.163 0.136 0.27 0.284 0.56

PIIIb
- 0.086 0.17 0.279 0.55
PIVa at + 0.030 0.103
σo= 61 psi -
PIVa re-tested + 0.096 0.27 0.370 1.05
at σo= 93 psi - 0.067 0.19 0.274 0.77

+ 0.015 0.043 0.184 0.52 0.255 0.72

PIVb
- 0.156 0.44 0.239 0.68

116
 Chapter 4
METHODOLOGY FOR EVALUATING LATERAL STRENGTH OF URM
BUILDING SYSTEMS

4.1 INTRODUCTION
A static pushover method for carrying out non-linear analysis of URM buildings having rigid
diaphragm, with emphasis on brick masonry, is proposed in this chapter. A procedure for constructing
capacity curves, a graphical representation of the base shear acting on the building against
corresponding lateral displacement at the top of the building (also known as pushover curve), is
recommended on the basis of the assumption that the masonry buildings predominantly vibrate in the
fundamental mode when subjected to seismic excitations. This assumption has also been verified by a
number of shaking table tests [TW 94, TK 97, LP 94, TBW 04]. These tests showed that masonry
buildings, irrespective of the type of masonry (i.e., URM, confined or reinforced), under seismic
excitations vibrate in the fundamental mode with a dominant shear behavior.
The proposed methodology is also based on the assumption, supported by numerous
experimental [TW 94, TK 97, LP 94, TBW 04] and field observations [To 99] that the lowest story in
masonry buildings suffer highest damage as compared to the upper stories when subjected to lateral
loading. Consequently, piers in the lowest story are most critical.
Results of quasi-static cyclic tests on piers, described in Chapter 3, are also used for establishing
various performance levels of URM buildings.
Finally, results of proposed methodology are compared with full-scale URM buildings tested at the
University of Pavia, Italy [MKC 95] and Georgia Institute of Technology, USA [Yi 04].
It is worth mentioning to state that the methodology proposed in this chapter, in addition to the
assumptions mentioned in the first two paragraphs, is based on the following assumptions:
a) Torsion in the buildings is neglected.
b) Possible pounding with the surrounding buildings is ignored.
c) Amplification of displacements in the out-of-plane direction in the exterior walls is ignored by
assuming rigid diaphragm.
d) Contribution of R.C. columns is neglected.
e) Variation in axial forces on the piers’ top due to increase in lateral displacements is not
considered.
f) The ultimate drift ratios for piers used in the methodology is determined by quasi-static cyclic
tests on URBM constructed in stone dust mortar as discussed in chapter 3. However, the
proposed methodology can be applied to other types of masonry (such as stone and concrete
block masonry constructed in cement: sand mortar, lime mortar, etc.) provided the properties

117
 required to quantify the seismic performance (e.g., displacement ductility factors and ultimate
drift ratios of masonry piers, etc.) for such type of masonry are experimentally determined.

4.2 TERMINOLOGY
Some terms will be repeatedly used hereafter and are given in Figure 4.1 and defined as follows:
1. A wall is defined as a structural element of the building with a length, lw, and a height, hw. In
case where all the walls of a building are vertically projected to the same height, the height of
the wall, hw, is equal to the total height of the building, H.
2. A wall element is defined as any part of a wall of length, lw, and height, hwe.
3. A pier is defined as a wall element of length, lw, and with a height equal to the height of the
adjacent opening, hp. The opening can be a window or a door.
4. A spandrel is a horizontal structural element, which exists between two openings in the
vertical direction.

4.3 CONSTRUCTION OF CAPACITY CURVES

In order to construct the capacity curve of a building, the lateral force-displacement curve for each pier
is drawn using bi-linear idealization recommended by Magenes and Calvi [MC 97]. The lateral
displacement at the top of a wall at the verge of ‘yielding’ and ultimate conditions in the
corresponding pier is determined using a procedure discussed in detail in Sub-section 4.6.6.
It is assumed that the walls are connected to rigid diaphragm . This assumption implies that
walls, connected to the diaphragm are laterally displaced by the same magnitude when subjected to
lateral forces. Hence, lateral force- displacement curve of a building can be obtained by superimposing
lateral force-displacement curves of all the wall connected to the rigid diaphragm.
The construction of a capacity curve for a masonry building is illustrated briefly with the
example of a fictitious building with a rigid diaphragm, shown in Figure 4.2. The base shear acting on
the building with the corresponding lateral displacement at the top of the building, ∆, can be
determined by summing the lateral forces in all the walls as follows:
V ∆ V V V V .
which can also be expressed as:
K .∆ K .∆ K .∆ K .∆ K ∆ .
where:
(Vb)∆ = Base shear in the building corresponding to a lateral displacement at the top of building, ∆,
provided displacement in all the piers are with in ‘linear elastic’ range;
V1, V2, V3 and V4 = Lateral forces acting on walls 1, 2, 3 and 4,respectively;
Kb = Lateral stiffness of the building;
Kw1, Kw2, Kw3 and Kw4 = Lateral stiffnesses of walls 1, 2, 3 and 4, respectively, and;

118
∆1, ∆2, ∆3 and ∆4 = Lateral displacements at the top of walls 1, 2, 3 and 4, respectively.
Since a rigid diaphragm ensures equal lateral displacement at the top of all walls, therefore, ∆1
= ∆2= ∆3 = ∆4 = ∆ and Equation 4.1b can be re-written as:
K .∆ K .∆ K .∆ K .∆ K .∆ .
or as:
K K K K K ∑K .
The capacity curve of the example building is shown in Figure 4.3. From Figure 4.3, it is evident that
there are four parameters that defines the capacity curve of a building, namely, Vyb, ∆yb,Vub and ∆ub,
where Vyb = base shear in the building at the verge of ‘yielding’; and
∆yb = lateral displacement at the top of the building corresponding to Vyb.
Vyb and ∆yb are found from the intersection of the elastic portion of the capacity curve with portion of
the curve representing zero, or almost zero, stiffness of the building, as shown in Figure 4.3.
Mathematically, ∆yb can be determined as:
V V
Δ .
K ∑K
The quantity Vub represents the base shear in the building at ultimate conditions, and ∆ub is the lateral
displacement at the top of building corresponding to Vub. It may, however, be noted that the bi-linear
idealized curves of walls, shown in Figure 4.3, cannot be constructed without carrying out bi-linear
idealization of the piers, which in addition to lateral strength of piers also depend upon the geometry
and material properties of piers and connected spandrels.

4.4 VARIOUS PERFORMANCE LEVELS PROPOSED FOR UNREINFORCED

BRICK MASONRY BUILDINGS
It is a common practice in earthquake engineering to indicate distress levels in a structure by
performance levels. For the purpose of this study performance levels, as recommended by FEMA 356
guidelines [FEMA 356] for building system were adopted as reproduced in Table 4.1. However, the
drift ratios corresponding to various performance levels of the URM buildings, as recommended by
FEMA 356, are not adopted in this study and the performance levels are based on the results of tested
piers discussed in Chapter 3. One of the main reasons for this decision is that drift ratios recommended
by FEMA 356 are based on the available literature which is mainly comprised of test results on
cantilever piers. Such type of cantilever piers undergoes high drifts before collapse and do not reflect
the actual behavior of piers in masonry buildings, which bend in double curvature when subjected to
lateral loads.
In this study, various building performance levels were defined as follows:

119
4.4.1 Operational level (O)
For the purpose of this study, operational performance level for an URM building is related to a
distress level when flexural tensile cracks develop in the bed joint of at least half of the piers that exist
in the building. This criterion is selected keeping in view the fact that flexural tensile cracking in the
bed joints of the first pier normally occurs at very low level of ground accelerations and the building
under very light damage and do not need any repair after the seismic activity.

4.4.2 Immediate occupancy (IO)

The IO performance limit state is related to the stage when the first shear crack, which develops at a
drift ratio of 0.08% in the corresponding pier, produce in the building.
In case of a building in which no pier undergoes shear cracking, the IO performance limit
state is related to a distress level when the first pier in the building is displaced to a drift ratio of
0.16%. The larger value of the drift ratio is selected since in case of a pier undergoing rocking failure,
lesser damage in the building is produced as compared to the damage due to shear cracking of a pier.

4.4.3 Life safety (LS)

LS performance level is related to a distress state when the building at the top is displaced to a lateral
displacement equal to the average of lateral top displacement at IO and collapse prevention (CP)
performance levels.

4.4.4 Collapse prevention (CP)

CP limit state for URM building is related to the distress state when the lateral resistance of a building
drops by 20% i.e., Vub = 0.8Vmb. Here, Vub= base shear in the building at collapse prevention limit
state, and Vmb = maximum resistance of the building determined from the capacity curve.

4.5 EVALUATION OF GROUND ACCELERATION CORRESPONDING TO

VARIOUS PERFORMANCE LEVELS
Ground accelerations corresponding to various performance levels cannot be evaluated without using a
design response spectrum. For the purpose of this study, evaluation of ground acceleration was carried
out using UBC-97 elastic design response spectrum recently adopted by “Building Code of Pakistan,
seismic provisions” [BCP 07] shown in Figure 4.4. Use of this design response spectrum is discussed
in Section 4.7.
Determination of seismic demand from a design response spectrum requires three parameters,
namely, damping ratio (ξ), soil type, and the time period of the structure (T). Once these parameters
are known, ground acceleration corresponding to a specific performance level can be evaluated from
the design response spectrum.
120
Further details regarding evaluation of ground acceleration corresponding to various performance
levels are provided in Section 4.7

4.6 DETAILS OF PROCEDURE FOR CONSTRUCTING THE CAPACITY CURVE

FOR URM BUILDING.
A procedure for developing capacity curves of masonry buildings was briefly discussed in Section 4.3.
However, a considerable amount of calculation work is involved in constructing a capacity curve for a
masonry building. Various steps involved in the construction of a capacity curve for an URM building
are discussed in the following sub-sections.

4.6.1 Structural model

The very first step in the analysis of a structural system is its structural idealization. An URM building
is not a combination of beams and columns in a way as a R.C. or steel building is. However,
experience over decades had shown [To 99, PP 92] that the analysis of a masonry building by
idealizing it as a framed structure results in satisfactory results. For the purpose of this work, an
approximate method of structural analysis proposed by Muto [PP 92] was adopted. The method, used
due its simplicity, is summarized in Section 4.6.3.
Additionally, it is assumed that the building is uniform about its centre of rigidity so that
torsional effects can be neglected. Similarly, it is assumed that vertical stresses at the top of pier
remains unchanged during the entire history of lateral displacement.

4.6.2 Material properties and geometrical properties of structural members

The following material properties are required for assessing the capacity of URM structures in the
lateral direction:
i. Compressive strength of masonry, fm;
ii. Principal tensile strength of masonry, ftu;
iii. Bond strength of masonry in tension, ft;
iv. Cohesion between brick units and mortar, c, and coefficient of friction of masonry, µ;
v. Modulus of elasticity of masonry perpendicular to the bed joints, Em;
vi. Modulus of elasticity of masonry parallel to the bed joints, Emx; and
vii. Modulus of rigidity of masonry, Gm.

4.6.3 Determination of inflection points in piers

The inflection point in a pier shifts from its original position, when a building is subjected to lateral
forces. Muto [PP 92] proposed a method for determining the points of inflection in the columns of a
framed structure when subjected to lateral loading. The method involves three parameters:
i) Number of stories in the building, j, and story level of pier, i;
121
 ii) Coupling effect of the spandrels at the top and bottom of piers, quantified by a
parameter α1; and
iii) Relative stiffness of the spandrels connected to the top and bottom of pier, quantified
by a parameter .
Various steps involved in locating the inflection point in a pier with Muto’s method, are
given in sub-sections 4.6.3.1 through 4.6.3.3

4.6.3.1 Stiffnesses of pier and connected spandrels

Based on Muto’s method [PP 92], the flexural stiffnesses of piers and spandrels are calculated as
follows:
Flexural stiffness of a pier, K 2E . I /h .
Flexural stiffness of the spandrel connected to the left end of pier at the top is:
E .I
K .
l

where Isp1, lsp1= Moment of inertia and length of spandrel, connected to the left side of pier at the top
end, respectively.
Similarly, flexural stiffness of spandrel connected to the right side of pier at the top end is:
E .I
K .
l

Flexural stiffness of spandrel connected to the left side of pier at the bottom end is:
E .I
K .
l

Flexural stiffness of spandrel connected to the right side of pier at the bottom end is:
E .I
K .
l
Normally, lsp3 = lsp1 and lsp4 = lsp2.

4.6.3.2 Relative stiffness of pier and spandrels

The flexural stiffness of spandrels relative to the connected pier is:

K K K K
K .
2K
The stiffness of the top spandrels relative to the bottom spandrels is:
K K
α .
K K

122
For a pier at the ground floor level, portion of the wall below the pier projecting on either sides
(including the foundation) acts as the bottom spandrels.
At should be noted that numerical error may sometimes be encountered while calculating α1,
e.g., when there are no spandrels on top and bottom end of a wall (i.e., a wall with no opening). Using
stiffness of zero for such spandrels in Equation 4.6 will result in a mathematically indefinite value.
To avoid this problem, a small stiffness (say, 0.001) is assigned to the non-existing spandrel.

4.6.3.3 Location of the inflection point in a pier

Consider the free- body diagram of a pier, as shown in Figure 4.5, subjected to lateral loading. Let ηlp
be the distance of the inflection point from the bottom of the pier. According to Muto’s method:
η η η .

where ηo is an inflection point factor, which depends on , number of stories (j) in the
building and story level (i) of the pier. The values of ηo for various values of i and j [PP 92] are
provided in Table 4.2.
Similarly, η1 is a correction factor that takes into account the effect of the difference in
flexural stiffness of the top and bottom spandrels, on the location of inflection point in the pier. The
value of η1, as recommended by Muto, is determined from Table 4.3, by considering both K and α1.
Muto suggest that if α1 > 1.0, value of 1/ α1 shall be taken as α1. However, in such a case, the value of
η1 is taken as negative of the value determined from Table 4.3.
For the purpose of conservativeness in this work, η was taken as 1.0 for the cantilever walls
even if the calculation indicated a value less than 1.0.

4.6.4 Calculation of gravity loads on piers

Two approximate methods, namely, tributary area method and coefficient method (previously
recommended by ACI code [ACI 318-63]), are used in the practice for the purpose of calculating
gravity loads transferred by the two way-slabs to the supporting walls. Since the ACI method takes
into account the effect of boundary conditions of the slabs, therefore, it was preferred over the
tributary area method in the present study.
In a single-story building, once the load supported by the slab to the transferring walls is
calculated, the weight of the spandrels below the roof, if any, is added to find the pressure (σo) at the
top of the pier. In multi story buildings, the gravity load due to other roofs and walls above the ground
floor slab is, of course, also added.
It is generally assumed in the practice that the vertical component of ground acceleration is
approximately equal to the 2/3 times the peak value of the horizontal component of the ground
acceleration [PP 92]. This assumption was also used while calculating the vertical pressures at top of
the masonry piers. e.g., if the peak horizontal ground acceleration determined from design response
123
spectrum was found to be 0.3g, then the vertical pressure on top of the pier was reduced, for the
purpose of safety, by an amount equal to 2/3(0.3g) = 0.2g. The peak horizontal ground acceleration,
which depends upon the zone (in which the building is located) and soil conditions, can be calculated
from the design response spectrum recommended by Building code of Pakistan [BCP 07] as shown in
Figure 4.4.

4.6.5 Calculation of the flexural tensile cracking strength and shear strength of piers
Double-curvature in the pier, as shown in Figure 4.5, under the action of a lateral force ‘divide’ it into
‘two’ parts. One portion of the pier is subjected to positive bending moment while rest of portion is
subjected to the negative bending moment. Since the shear force, V, has same magnitude for ‘both’
parts, the failure will occur in the portion of the pier with the greater end moment. Therefore, for the
purpose of calculating forces corresponding to flexural tensile cracking and shear strength of piers, the
greater of η and (1-η) values is used.
For the purpose of brevity, the notation ψ′ will be used for the greater of η and (1- η) values
while evaluating flexural tensile cracking force or shear strength such that:
ψ′ = η if η ≥ 0.5 (4.8a)
ψ′ = 1- η if η < 0.5 (4.8b)

4.6.5.1 Shear force corresponding to the flexural tensile cracking in bed joints of the pier, Vfl,cr

Bed joint in a masonry pier crack in tension when the maximum tensile stress, σt, under the combined
actions of axial and flexural stresses, becomes equal to the tensile strength of the bond between bricks
and mortar, ft. σt can be expressed as follows:
σ σ σ .
where σo = stress in a pier due to axial load; and σf = maximum flexural tensile stress in the pier under
the action of bending moment, M, which can be calculated [ PB 98] as:
l
M 6M
2
σ .
t. l t. l
12

At flexural tensile cracking, σ f and M Mf , Vf , . ψ h

By inserting the values in Equation 4.10a:

6V . ψ . h
σ .
t. l

By substituting the values of σt and σf in Equation 4.9a and then re-arranging, following expression
can be obtained:

124
 σ f l .t
Vf , .
6ψ . h
Similarly, lateral forces corresponding to various failure modes in the piers can be calculated by
equations given in following sub-sections.

4.6.5.2 Shear force corresponding to the rocking /toe crushing failure, Vr

Lateral force corresponding to rocking/toe crushing failure, Vr, can be calculated [MC 97] as:

l .t σ σ
V . 1 .
α 2 0.85f
M V. ψ . h ψ .h
where α Shear ratio
V. l V. l l

4.6.5.3 Shear force corresponding to the sliding shear failure, Vsl

Lateral force corresponding to sliding shear failure, Vsl, can be calculated as:

1.5c µ. σ
V c. α l .t .
1 3
σ
P P
where V
6α 2α
The above-mentioned equation, proposed by Magenes and Calvi [MC 97], was derived by considering
the equilibrium conditions of a monotonically loaded pier. It was assumed that masonry cannot resist
the tensile stresses.
Under the action of lateral cyclic forces, masonry undergoes flexure tensile cracking on both
the sides when σt ft. Consequently, cohesion between bricks and mortar is no more available. By
inserting c= 0 in Equation 4.12a, following expression is obtained:
V µ. l . t .

4.6.5.4 Shear force corresponding to the diagonal tension shear failure, Vd

Lateral force corresponding to diagonal tension shear failure, Vd, can be calculated by using the
equation proposed by Turnšek and Sheppard [TS 80] as follows:

f σ
V 1 l .t .
b f

125
Minimum of the three values obtained from Equations 4.11, 4.12b and 4.13 determines the shear force
corresponding to failure, Vsf.

4.6.6 Bi-linear idealization of the lateral force-displacement curves of piers

Bi-linear idealization of the lateral force-displacement curve of an URM pier requires three
parameters, namely, ‘yield’ force, Vu, corresponding displacement, dy, and ultimate displacement, du,
as shown in Figure 3.52.
‘Yield’ force, Vu, as recommended by Magenes and Calvi [MC 97] can be calculated as:
V 0.9V .
However, determination of ‘yield’ displacement, dy , involve lateral stiffness of the pier which can
determined by taking into account the effect of both flexural and shear stiffness of the pier. For the
pier shown in Figure 4.5, displacement at the top end, d, can be determined [PB 98] as follows:
V. h M .h 1.2V. h
d .
3E . I 2E . I G .A

where:
Mt = Moment at the top end of pier = V(hp-ηhp);
Ap= Cross-sectional area of the pier = t.lp; and
Ip = Moment of inertia of the pier in the in-plane direction = t.lp3/12
By substituting the values of Mt, Ap and Ip, Equation 4.15a can be re-written as follows:
d 12 h 3η 1 1 E l
.
V t. E l 6 10 G h

or as:

t. E l
12 h
K .
3η 1 1 E l
6 10 G h
V
where K Lateral elastic stiffness of pier
d
The bi-linear idealization of the lateral force-displacement curve of a masonry pier by
Magenes and Calvi criterion [MC 97] involve the use of lateral elastic stiffness of the pier. Where
lateral elastic stiffness of the piers is calculated by determining the slope of a line joining origin to a
point on the curve having an ordinate of 0.75Vu. During the current experimental work, it was
observed that the piers did not suffer any observed damage at a lateral force equal to 0.75Vu. It was,
therefore, decided to use the lateral stiffness corresponding to the formation of a significant crack
which was determined from visual observation and a careful study of hysteresis loops.

126
 From the statistical analysis of experimental data discussed in Chapter 3, lateral stiffness at
the formation of first significant crack (termed as effective stiffness, Kep) was determined to be 0.82Kp
with a standard deviation of 0.072Kp. Using a mean minus standard deviation value results in:
K 0.82K 0.072K 0.75K .
By substituting the value of Kp from Equation 4.15c, Equation 4.16a can be re-written as:

t. E l
0.75
12 h
K .
3η 1 1 E l
6 10 G h

Having found the effective stiffness, ‘yield’ displacement in the pier, dy, can be determined as:
V
d .
K
Similarly, corresponding drift ratio, δy, can be calculated as:
d
δ .
h
Based on experimental work discussed in Chapter 3, ultimate drift ratio for a pier, δu, failing in shear,
is taken as 0.35%. Where ultimate drift ratio for a pier, δu, failing in rocking was decided to be taken
as 0.8 % [AM 04]. i.e.,
d 0.0035h for a pier failing in shear .
d 0.008h for a pier failing in rocking .

4.6.7 Bi-linear idealization of the lateral force-displacement curves of the corresponding walls

Bi-linear idealization of the lateral force-displacement curve of the wall corresponding to a pier is
carried out by keeping in view the fact, that under the action of lateral forces, lowest story piers in a
masonry building are significantly damaged as compared to the upper story piers.
Consider the front wall of the example building shown in the Figure 4.2, reproduced in the
Figure 4.6a. The variation of lateral displacement along the height of the wall No.2 at the verge of
‘yielding’ and ultimate conditions are shown in Figures 4.6b and c, respectively. The lateral
displacements vary linearly along the wall’s height before the ‘yielding’ start in the corresponding
pier. Once the pier starts ‘yielding’ in the lateral direction, drift in the pier start increasing without any
increase in the lateral force, till the ultimate limit state is reached. Since, the lateral force remains
constant, drift ratio in other parts of the corresponding wall (above and below the pier) remains
unchanged.
Following relation can be established by considering Figure 4.6c:
∆ d ∆ d .
where:
127
∆uw = Lateral displacement at the top of the wall at ultimate conditions in the corresponding pier;
∆yw = Lateral displacement at the top of the wall at the start of ‘yielding’ in the corresponding pier;
dywe = Lateral displacement at the top of the wall element at the start of ‘yielding’ in the corresponding
pier; and
duwe = Lateral displacement at the top of the wall element at ultimate conditions in the corresponding
pier.
∆yw, dywe, and duwe are calculated as follows:
∆ δ .h .
Where hw = Height of the wall.
Equation 4.20a can also be expressed as:
d
∆ .h .
h
Lateral displacement at the top of a wall element at the start of ‘yielding’ in the corresponding pier,
dywe, can be calculated as follows:
d δ .h .
where hwe = Height of a wall element corresponding to the pier.
Equation 4.21a can also be expressed as:
d
d h .
h
or:
d
d h h .
h
where hpb = Vertical distance between pier’s bottom and floor level.
Similarly, lateral displacement at the top of a wall element at the ultimate conditions in the
corresponding pier, duwe, can be calculated as:
d δ .h 0.0035 h for shear failure .
d δ .h 0.0080 h for rocking failure .

Having found ∆yw, dywe and duwe from Equations 4.20, 4.21 and 4.22, respectively, ∆uw can
easily be determined from Equation 4.19
The ultimate displacements in a wall, in addition to Equation 4.19, can also be determined by
relating it to the displacement ductility factor of corresponding wall element. This can be
accomplished by dividing both sides of Equation 4.19 by ∆yw as follows:
∆ d d
1 .
∆ ∆ ∆
∆
where Displacement ductility factor of the wall, µ .
∆
128
From equation 4.20 a, we have Δ δ .h
The above-mentioned equation can also be expressed as:
d
Δ .h .
h
Inserting the values of ∆uw/∆yw = µw and ∆yw in Equation 4.23a and re-arranging results in:
h d
µ 1 1 .
h d
d
By substituting with µ , Equation 4.23b can be expressed as:
d
h
µ 1 µ 1 .
h
Using equation 4.23c, ∆uw (which is equal to w.∆yw) can be expressed as:
h
∆ 1 µ 1 ∆ .
h
where hwe = hpb + hp
For a pier starting from floor level, hpb = 0. For such a condition hwe = hp, duwe = du, and dywe = dy and
Equation 4.23d can be expressed as:
h d
∆ 1 1 ∆ .
h d
or:
h
∆ 1 µ 1 ∆ .
h
Once Vu, ∆yw and ∆uw are determined, the idealized bi-linear lateral force-displacement curve for the
wall can easily be drawn.

4.6.8 Construction of the capacity curve of URM building

As discussed in Section 4.3, the capacity curve of a masonry building drawn on the basis of rigid
diaphragm assumption implies that the displacements at all the points on the diaphragm are equal. In
other words, the walls connected to rigid diaphragm are equally displaced under the action of lateral
forces. Hence, the capacity curve of a building with rigid diaphragm can be constructed by simply
superimposing the lateral-force displacement curves of the individual structural walls of the building.
Once the capacity curves is drawn, displacement ductility factor of the building, µb, can be
calculated as:
∆
µ .
∆
where:
∆yb = Lateral displacement at the top of the building at the verge of ‘yielding’; and
∆ub = Lateral displacement at the top of the building when Vyb drops by 20%.
129
In a building where all the walls are vertically projected to the same height, ∆yb = ∆yw and ∆ub = ∆uw
Having found the magnitude of µb, response modification factor of the building, R, can be
calculated by any appropriate method. There exist several approaches for determining the value of R
[MB 94,VFF 94,CG 99], however, for the purpose of simplicity, a widely used and effective method
proposed by Newmark et al [VN 60, VNC 65] was used in this study. According to this approach, R,
can be calculated by using the Equation 4.25 as long as 33 Hz f1 2 Hz . where f1 = frequency of
elastic SDOF system in the fundamental mode of vibration
R 2µ 1 .

4.7 EVALUATION OF THE GROUND ACCELERATION CORRESPONDING TO

THE VARIOUS PERFORMANCE LEVELS IN A BUILDING.
Following steps are involved in evaluating the ground acceleration corresponding to a specific
performance level in a building:

4.7.1 Base shear and lateral displacements in the elastic MDOF system corresponding to the
actual building
Ground acceleration corresponding to a lateral displacement at the top of an inelastic system can be
evaluated by using elastic design response spectrum provided the inelastic system is transformed to
equivalent elastic SDOF system.
Let ∆ be the lateral displacement at the top of building corresponding to a specific performance level.
By definition [Ch 01]:
V ∆
R∆ .
V
This can be re-written as follows:
V ∆ R∆. V .
where Vyb = base shear in the building at the initiation of ‘yielding’, (Vbe)∆ = base shear in the
equivalent elastic system at a lateral displacement, ∆, in the corresponding inelastic system (i.e. actual
building), and R∆ = response modification factor corresponding to the lateral displacement ∆
A value of Δ Δ indicates that the buidling is still in the linear elastic limit and R 1
A value of Δ Δ indicates that the buidling has been distressed into inlelastic limit. In such a
case R can be calculated as follows:
R∆ 2 µ ∆ 1 .
where:
∆
µ ∆ .
∆

130
Once (Vbe)∆ is known by substituting the value of R∆ in Equation 4.26b, determined from Equation
4.27a, the corresponding lateral displacement, (∆be)∆, can be determined as follows:
V ∆
∆ ∆ .
K

where Kb is the lateral stiffness of the building as determined from Equation 4.1d.

4.7.2 Lateral displacement in the elastic SDOF system corresponding to the actual building

Having found the lateral displacement in the elastic MDOF system from Equation 4.28, lateral
displacement in the corresponding elastic SDOF in the fundamental mode, (Sd)∆, is determined from
the relation [Ch 01] as follows:
S ∆ ∆ ∆⁄ Γ .
where Г is the modal participation factor [Ch 01] which can be calculated as follows:
∑ m.
Г .
∑ m.
where mi is lumped mass at the i-th floor level, is the fundamental mode normalized
displacement at the i-th floor level, and j indicates the total number of floor levels.
Normaized displacement at top floor level, 1.
It has already been discussed that the masonry buildings under seismic excitations vibrate in
the fundamental mode with a predominant shear behavior. A shear behavior can more appropriately be
represented by a linearly increasing lateral displacement curve along the height. Therefore, for the
purpose of this study, the inverted triangular mode shape, with the normalized displacement at the top
floor, 1, was adopted.

4.7.3 Acceleration in the elastic SDOF system, (Sa)∆, corresponding to the lateral displacement,
(Sd)∆
Acceleration at the top of equivalent elastic SDOF, in the fundamental mode, can be
determined by using the relation [Ch 01]:
S ∆ S ∆. 2πf .
Relation 4.31a holds good as long as f 0.2 Hz. where f1 is calculated [Ch 01] as follows:

1 KE
f .
2π mE

where KE and mE are the equivalent stiffness and equivalent mass of the idealized as SDOF system
corresponding to the actual building, respectively.
Stiffness of the building, Kb, determined from equation 4.1d can be used as the equivalent stiffness of
SDOF system, KE. where equivalent mass and height of the SDOF can be calculated [Ch 01] as:

131
mE m. .

∑ h .m .
hE .
∑ m.
where h is the height of the i th floor

4.7.4 Ground acceleration, (ag)∆ inducing spectral acceleration, (Sa)∆ in the elastic SDOF system
In this study UBC-97 design spectrum, recently adopted by the Building Code of Pakistan [BCP 07],
shown in Figure 4.4, was used to calculate ground acceleration inducing spectral acceleration in an
elastic SDOF system [Ch 01] as follows:
S ∆
a ∆ .
A
where Af is the factor by which ground acceleration amplifies at the top of a SDOF system.
It is shown in Figure 4.4, that spectral acceleration linearly amplify from 1 to 2.5 times the
ground acceleration when the time period of elastic SDOF system increase from 0 to To. The
amplification factor, however, remains constant (i.e., 2.5) for SDOF systems with To ≤ T ≤ Ts .where
To and Ts are controlling periods and are calculated [BCP 07] as:
C
T .
2.5C
T 0.2 T .
where Cv and Ca are the seismic coefficients [BCP 07], which, in addition to soil conditions, also
depend upon the one of the five zones, namely, Zone 1, 2a, 2b, 3 and 4, in which the building is
located.

4.8 RESONSE OF AN URM BUILDING IN THE OUT-OF-PLANE DIRECTION

Previous discussion in this chapter was focused on the behavior of URM masonry walls when
subjected to actions in the in- plane direction. Depending upon the boundary conditions of an URM
wall, floor to wall connection, material of diaphragm, and the story height, an URM wall can also fail
in the out-of-plane direction, even it might be safe under the action of in-plane forces. Therefore,
response of the URM walls under the action of out-of-plane inertial forces is an important issue and is
discussed in the following paragraphs.
Priestley [Pr 85] developed a mechanics based method for evaluating the capacity of an URM
wall in the out-of-plane direction. The method is based on the fact that an URM wall do not collapse
just after the formation of flexural tensile crack and undergoes large amount of displacement before
the collapse. Collapse of the wall occur due to the instability when the wall is displaced to an extent

132
such that the resultant compressive force acting on cracked section acts well away from the line of
action of action of axial loads (i.e., axial gravity force due to overlying slab and self weight of wall).
Consider a face- loaded wall, shown in Figure 4.7a. It is assumed that diaphragms connected
at the ends are rigid enough so that the floor accelerations do not significantly differ at the ends. It can,
therefore, be assumed that the wall is loaded with a uniform inertial force, Fop, along the height due to
seismic excitations. Additionally, it is assumed the floors move in-phase and act as hinged supports at
the floor-to-wall connections.
Consider the free body diagram for the unit length of wall after the out-of-plane flexural
tensile cracking, shown in Figure 4.7a. By developing the moment equilibrium equation about the
centroid at the bottom of wall, it can be derived that:
h W ∆
H F . . .
2 h 2
where h= height of wall between two floors (i.e., story height), Ht = horizontal reaction at the top of
wall, W = weight of the wall, ∆op = deflection of the wall at the mid-height, and Fop = inertial force per
unit height (Fop = mw.aop in which mw = mass per unit height of the wall).
Consider free body diagram of the upper half of the wall as shown in Figure 4.7b. Considering the
moment equilibrium about the centroid at the mid height of the wall and simplifying the equation
results in the following expression:
h ∆
R .x F . W. N. ∆ .
8 2
where Rc=N+W/2 in which N= gravity load acting at the top end of the wall, and x = eccentricity of
‘Rc’ from with respect to the centroid of wall.
Equation 4.37 can be re-arranged and expressed as:
F 8⁄h R .x ∆ W ⁄2 N .
W
By replacing N with R , Equation 4.38b can expressed as:
2
F 8R ⁄h x ∆ .
Similarly:
F 8R
a x ∆ .
m m .h
where mw = Mass of the wall per unit height per unit length
it is worth mentioning that while calculation N and W, the gravitational acceleration, g, is reduced by
an amount equal to 2/3Cag so as to take into account the adverse effect of the seismic excitations in the
vertical direction.
The values of aop for various increasing values of ∆op are determined using Equation 4.39,
until the wall become unstable at the maximum possible value of ∆op = xmax = t/2-ac/2 [Pr 85]. where

133
ac is the depth of uniform compressive stress block at failure, ac = Rc/(0.85fm.t). A typical plot of aop/g
against ∆op result in a curve, shown in Figure 4.8
The spectral acceleration required to produce failure in the corresponding linear elastic SDOF, ae,op, is
determined by using equal energy approach. This is carried out by equating areas under the curve of
acceleration-displacement curve of in-elastic system with the corresponding linear elastic system [Pr
85], shown in Figure 4.9, as follows:
Area under the triangle, A Area under the curve, A
or:
1 a ,
.Δ . A .
2 g
where ∆e = Displacement in the equivalent elastic system produced by ae,op
Equation 4.40a can be expressed as:
2g A
a , .
Δ
Similarly, slope of a-∆ diagram, Kop, as determined from Figure 4.9 is:
a ,
g
K .
Δ
Equation 4.41a can be expressed as:
a ,
Δ .
g. K
By substituting the value of ∆e in Equation 4.40a:
1 a ,
. A .
2K g
Equation 4.40c can be expressed as:
a ,
2A . K .
g
Stepwise procedure for determining ∆e,op and corresponding ground acceleration ,ag,op, is explained in
following sub-sections.

4.8.1 Mid-height curvature at out-of-plane cracking of wall

Consider a face-loaded wall of unit length. By considering both the flexural and axial stresses in the
out-of-plane direction of the wall, and ignoring the flexural tensile strength at brick-mortar interface, it
can be derived [PB 98] that:
R .t
M , .
6
where Mop,cr = Mid-height moment at the onset of out-of-plane flexural tensile cracking of the wall
Similarly, it can also be derived by basic mechanics of solids [PB 98] that:

134
 σ ,
R t .
2
or:
2R
σ , .
t
Similarly [ PB 98]:
σ ,
, .
E
By substituting the value of op,cr from Equation 4.43b, Equation 4.44a can be expressed as:
2R
, .
t. E
where corresponding curvature at the on-set of flexural tensile cracking in the out-of-plane direction
can be calculated [ PB 98] as:

, 2R
φ , .
t t. E
where σop,cr = Maximum compressive stress at mid-height of wall corresponding to Mop,cr, , =
strain corresponding to σop,cr, and φop,cr = curvature at mid-height section caused by Mop,cr.

4.8.2 Mid-height deflection and associated acceleration at the onset of out-of-plane flexural
tensile cracking of wall
Mid-height deflection at the onset of out-of-plane flexural tensile cracking can be calculated [PB 98]
as:
5F , . h
∆ , .
384E . I
where Iop = Moment of inertia of wall cross-section for a unit length in the out-of-plane direction, and
Fop,cr= inertial force, per unit height per unit length of the wall, required to produce flexural tensile
cracking in the out-of-plane direction.
The moment required to produce out-of-plane flexural tensile cracking in a length of wall, Mop,cr , by
assuming hinged supports at the ends, can be calculated as:

F , .h
M , .
8
By equating Equations 4.42 and 4.47a:
4R . t
F , .
3h
Inserting the value of Fop,cr in Equation 4.46a results in:
20R t. h
∆ , .
1152E . I
Similarly, acceleration corresponding to Fop,cr is calculated as:
135
 F ,
a , .
m
By substituting the value of Fop,cr from Equation 4.48, Equation 4.49a can expressed as:
4R . t
a , .
3m . h

4.8.3 Acceleration and corresponding mid-height deflections for various levels of strains after
the out-of-plane flexural tensile cracking of the wall

1. Assume a mid-height compressive strain, , , such that , ,

2. Compute the depth of compression zone, cc

After the flexural tensile cracking of masonry at the mid-height of the wall in the out-of plane
direction, Equation 4.43a can be expressed as:
σ,
R .c .
2
or:
2R 2R
c .
σ, , .E

Similarly:
, , .E
φ .
c 2R
3. Compute deflection
It is assumed that curvature and deflection vary in same ratio along height after
flexural tensile cracking of wall in the out-of-plane direction, i.e.:
∆ φ
.
∆ , φ ,

or:
φ
∆ ∆ , .
φ ,

4. Compute acceleration
Acceleration is determined by Equation 4.39 and is reproduced below:
8R
a x ∆ .
m .h
where:
t c
x for linear elastic stresses; and
2 3
t a R
x x at the crushing of masonry in compression, where a
2 2 0.85f

136
 Procedure described in Section 4.8.3 is revised for various levels of increasing compressive
strains until the wall fail. Plotting acceleration and corresponding mid-height deflection data result in a
curve shown in Figure 4.8.

4.8.4 Acceleration in the corresponding linear elastic system

Acceleration required to produce failure in the corresponding linear elastic system, ae,op, is determined
by using equal energy principle. This is done by equating areas under the curve of acceleration-
displacement curve of actual inelastic system and corresponding linear elastic system, shown in Figure
4.9. The procedure has already been discussed in section 4.8 and Equation 4.40d was derived which is
reproduced below
a ,
2A . K .
g

4.8.5 Evaluation of ground acceleration producing ae,op

Acceleration in the building in the out-of-plane direction is determined on basis of rigid diaphragm
assumption. Once the floors are assumed rigid, no amplification of accelerations in the face-loaded
walls occur. Consequently, in the diaphragm plane, acceleration at any point in a wall in the out-of-
plane direction is equal to the acceleration in the orthogonally connected walls loaded in the in-plane
direction.
An estimate of the storey accelerations from the spectral acceleration of the equivalent SDOF is shown
in Figure 4.10 [PP 92]. Spectral acceleration in equivalent SDOF system in out-of-plane direction,
Sa,op, by assuming a linearly varying mode shape, can be determined from Figure 4.10 as:
a ,
S, . hE .
h
where hc = Distance of the centroid of the wall shown in the Figure 4.7a above the ground. The
Equation 4.53a is valid as long as hE hc. The corresponding ground acceleration, ag,op, is determined
[Ch 01] as:
S ,
a , .
A
However, in the case hE hC, following relation, using Figure 4.10, can be established:
a , a S , a
.
h hE
Using the relation, Sa,op = Af. ag,op, Equation 4.53b can be expressed as:
a , hE
a , .
A 1 h hE
Where Af for determining ag,op, using either Equation 4.54a or 4.54b, is calculated by using design
response spectrum recommended by Building code of Pakistan [BCP 07], as shown in Figure 4.4. In

137
order to calculate time period of the building in the out-of-plane direction, kE for the building is
determined in the orthogonal direction (i.e., considering walls loaded in the in-plane direction)

4.9 COMPARISON OF THE PROPOSED METHOD WITH THE EXPERIMENTAL

RESULTS OF FULL-SCALE TESTED MODEL BUILDINGS
Authenticity of any numerical or analytical method can only be validated by comparing its result with
that of some experimental work. For this purpose, results from analytical procedure for lateral strength
assessment of URM buildings, explained in previous sections were compared with the full scale URM
buildings tested at university of Pavia, Italy [MKC 95] and Georgia Institute of Technology, USA [Yi
04]. The details of these experimental works as well as comparison of the results of test data with
results of proposed method are discussed in following sub-sections:

4.9.1 University of Pavia building

A full-scale building was tested at the university of Pavia [MKC 95] under quasi-static cyclic loading.
The geometric details of building are shown in Figure 4.11 (dimension shown are in meters). Solid
fired clay bricks were used in the building with a 1:3 (hydraulic lime: sand) mortar. The mechanical
properties of material were as follows:
i. Compressive strength of the masonry = 900 psi (6.2 MPa);
ii. Compressive strength of the mortar = 290-435 psi (2-3 MPa);
iii. Initial bond shear strength =33.4 psi (0.23 MPa);
iv. Equivalent internal shear coefficient = 0.57;
v. Modulus of elasticity of the masonry =214.6 ksi (1480 MPa); and
vi. Modulus of rigidity of the masonry =52.2 ksi (360 MPa)
In the absence of experimental value, ftu was assumed to be equal to 0.035fm in the current study.
This decision was based on the discussion made is Section 5.2.2.
The walls were constructed with solid fired clay brick masonry with a uniform thickness of
9.84 in. (250mm). The wall D (termed as ‘Door wall’), shown in building’s plan in Figure 4.11a, was
disconnected from the orthogonal walls. Where as wall B, termed as ‘Window wall’, was connected
to orthogonal walls with an interlocking brick pattern around the corners.
The floors consisted of isolated steel beams (I-section) having a depth of 5.51 in. (140 mm) to
simulate a very flexible floor system. Both the horizontal and vertical loads were applied through these
beams. Concrete blocks were used to simulate the gravity load. The total gravity load at the first and
second floor were 55.84 kip (248.4 kN) and 53.23 kip (236.8 kN), respectively, representing a
distributed load of 209 psf (10 kPa). These loads resulted in a uniformly distributed vertical stresses of
58-72 psi (0.4-0.5 MPa) on the ground floor piers.

138
 The building was tested in a direction parallel to ‘Door wall’ by simultaneously applied four
concentrated lateral forces on ‘Door’ and ‘Window’ walls. Two lateral forces were applied on each
wall, one each at first and second floor level. Typical displacement history applied during the testing
work is shown in Figure 4.12.
The results of proposed methodology with the ‘Door wall’ are compared in Figure 4.13.
Comparison with ‘Window wall’ was not made due to the presence of large flange action as a result
of the walls connected in the orthogonal direction .
By comparing the results, as shown in Figure 4.13, it can be observed that the proposed
method determines a relatively lower initial stiffness of the wall as compared to the experimental
results. Although, the proposed method underestimated the ultimate displacement of wall, however,
predicted the lateral strength of wall with reasonable accuracy

4.9.2 Georgia Institute of Technology building

A full-scale two-story URBM building was tested at the Georgia Institute of Technology, USA [Yi
04], under quasi-static loading. The plan of the tested building is shown in Figure 4.14.
One of the main objectives of the experimental work was to investigate the behavior of
existing brick masonry buildings stock of mid America, mostly constructed in 1950s using strong
brick units and weak mortar.
Walls in the building were having two different thicknesses. Walls 1 and wall 2, shown in
Figure 4.14, were two wyhtes with a thickness of 8 in. Where as walls A and B, were three wyhtes,
with a thickness of 12 in.
The walls were constructed using the stretcher bond with one header course after every sixth
stretcher course. Two different types of brick units were used in the building. Solid fired clay brick
units, having a compressive strength 6230 psi, were used in the lower story of building. Where as,
upper story was constructed with fired clay brick units, having a circular hole along stretcher, with a
compressive strength of 5285 psi. Mortar used in the entire work was having a ratio 0.5:2:9
(cemet:sand:lime) with a compressive strength of 41 psi. Other material properties were as under:
i. Compressive strength of the masonry (solid brick) = 1458 psi;
ii. Compressive strength the masonry (hollow brick) = 593 psi;
iii. Initial bond shear strength =60 psi;
iv. Equivalent internal shear coefficient = 1.1;
v. Shear sliding coefficient for the cracked bed joint =1.0; and
vi. Modulus of elasticity of the masonry =1168 ksi
In the absence of experimental values, Gm and ftu were assumed to be 0.25Em (as used in the building
tested in University of Pavia) and 0.035fm , respectively.
The walls were connected to each by proper toothing at the corners. Walls A and B were kept
identical. However, wall 1 was specially designed to allow a larger flange at the end of wall B.
139
 The roof and first floor were constructed with timber. The roof diaphragm was supported on
2x10 wooden joists spanning 12ft at a c/c spacing of 16 in. The joists were laterally supported along
full length by the full depth wooden blocks spaced at 4ft c/c. In addition, a wooden stud was provided
to support the central portion of building along north side. These studs were supported on walls A and
B.
The diaphragms were connected to the walls with the threaded shear ties and special type of
tension ties. No additional weight was added to the first and roof diaphragms. Consequently, the
gravity loads supported by the ground floor piers were quite low. A linear elastic finite element
analysis estimated that the pressure at the top of ground floor piers, except one pier, varied from 20-30
psi.
The entire test was conducted in a displacement controlled testing environment. A typical
displacement loading, shown in Figure 4.16, was followed during the entire testing. The walls 1 and 2
were loaded first, simultaneously, with the help of actuators attached on the walls at the first and top
floors. The testing continued until fully developed failure mechanism was observed. After then, walls
A and B were loaded and tested in the same manner
The results of proposed methodology with the walls A and 1 are compared in Figures 4.17a
and b, respectively.
By comparing the the analytical and experimental results of wall A, it can be observed, that
the proposed method determines the initial stiffness of wall with reasonable accuracy. The proposed
analytical method also well estimated the lateral strength and ultimate displacement of wall A when
compared with the experimental results.
The proposed method, when compared with the experimental results of wall 1, underestimated
the initial stiffness of wall 1. The method well underestimated the lateral strength of wall 1 when
compared with experimental results. The ultimate dipslacements estimated by analytical method were
found to be relatively higher as compared with that of experimental results. However, it is worth
mentioning that the building was not tested till collapse so the experimetal displacements of the walls
were not truly ‘ultiamate’.

4.10 STEP BY STEP PROCEDURE FOR ASSESSMENT OF SEISMIC CAPACITY

FOR URM BUILDING
The propsed method for seismic assessmen of URM buildings can be summarized in a step wise
manner as follows:
Step 1: Establish material properties.
Step 2: Determine geometrical properties of the piers and spandrels.
Step 3: Determine inflection point in the piers.

140
Step 4: Determine flexural tensile strength and lateral strength of piers by taking into account the
effect of vertical ground acceleration.
Step 5: Draw the idealized bi-linear lateral force-displacement curves for the walls.
Step 6: Draw the capacity curves for the building and identify various performance levels.
Step 7: Determine the base shear and displacement in corresponding elastic system for various
performance levels.
Step 8: Determine the modal participation factor, equivalent mass and equivalent height for elastic
SDOF system for the building.
Step 9: Determine horizontal ground accelerations corresponding to various performance levels.
Step 10: Determine building capacity in the out-of-plane direction

4.11 CONCLUSIONS AND RECOMMENDATIONS

By comparing the results of the proposed method with the experimental results of full-scale buildings
[MKC 95 and Yi 04], it is concluded that the proposed methodogoly conservatively estimates the
lateral strength and ultimate lateral displacements of URM buildings. The method also well estimate
the elastic stiffness of walls, used for determining the natural period of vibration of the equivalent
SDOF for the URM building. The proposed method is explained, with the help of detailed
calcualtions, by taking an actual two-story buliding as an example, in Section 5.5.
It is, however, recommended that the future work shall be carried out inorder to incoporate, in the
proposed methodolgy, the effect of torsion in the buildings and amplification of displacements in the
walls in the out-of-plane direction due to the flexible diapragm.

141
 hwe hp

hpb

Figure 4.1 Terminology. figure adopted from research

work of Lang [La 02] with some modification

142
Figure 4.2 Fictitious example building with rigid diaphragm
(a) Plan (b) Front elevation (c) Rear elevation

143
 WALL 1
WALL 2
WALL 3
WALL 4
(∆yb, Vyb) WALL 5

Base shear,Vb (kips)

(∆ub, Vub)

Lateral displacement at top floor, ∆ (in.)

Figure 4.3 capacity curve of example fictitious building, shown in figure 4.2, along x direction

Figure 4.4 Elastic design response spectrum recommended by Building Code of Pakistan

144
Figure 4.5 Free-body and bending moment diagram of a masonry pier (restrained on
the top and bottom end) under the action of lateral force V

145
Figure 4.6 (a) Front elevation of example building as shown in figure 4.2b, (b) Variation of lateral displacements in wall # 2 at the verge of ‘yielding’ in the
corresponding pier at lowest story (c) Variation of lateral displacements in wall # 2 at ultimate conditions in the corresponding pier at lowest story

146
 (a) (b)

Figure 4.7 (a) Free-body diagram of a unit length of a face-loaded wall

(b) Free-body diagram of the upper half of the wall shown in (a)

147
aop / g

∆op

Figure 4.8 A typical graph between acceleration and corresponding mid-span

deflection for an URM wall loaded in the out-of-plane direction

∆e

148
Figure 4.10 Variation of response acceleration with height [PP 92]

149
 (a)

(b)

Figure 4.11 (a) Plan and (b) Elevation of the model building tested at University of Pavia, Italy. All the
mentioned dimensions are in meters. The arrows show points of application and direction of lateral forces
[MKC 95].

150
 Figure 4.12 Displacement history imposed on the second floor of the model
building model tested at the University of Pavia [MKC 95]

40

30
Base shear (kips)

20

10
Results obtained from proposed methodology
Eperimental results under positive loading direction
Eperimental results under negative loading direction
0
0 0.2 0.4 0.6 0.8 1
Lateral displacement at the top floor (inches)

Figure 4.13 Comparison of results from proposed methodology with the experimental
results of model building tested at the University of Pavia, Italy

151
Figure 4.14 Plan of the model building tested at the Georgia Institute of Technology, USA [Yi

152
 (a)

(b)

Figure 4.15 Elevation views of building [Yi 04] shown in figure 4.14:
(a) Walls A & B (b) Wall 1
(Number in the parenthesis indicates number of masonry units)

153
Figure 4.16 Typical displacement history applied during the test on the model
building tested at the Georgia Institute of Technology, USA [Yi 04]

154
 50

Base shear (kips) 40

30

20

10
Results obtained from proposed methodology
Eperimental results under positive loading direction
Eperimental results under negative loading direction
0
0 0.1 0.2 0.3 0.4 0.5 0.6

Lateral displacement at top floor (inches)

(a)

70
Results obtained from proposed methodology
Eperimental results under positive loading direction
60
Eperimental results under negative loading direction

50
Base shear (kips)

40

30

20

10

0
0 0.1 0.2 0.3 0.4

Lateral displacement at the the top floor (inches)

(b)
Figure 4.17 Comparison of results from proposed methodology with the experimental
results of model building tested at the Georgia Institute of Technology, USA
(a) Wall A ; (b) Wall 1
155
Table 4.1 Building performance levels recommended by FEMA 356 [FEMA 356] guidelines
Performanc Overall Damage Description
e level
No permanent drift. Structure substantially retains original strength and
Operational
Very Light stiffness. Minor cracking of facades, partitions, and ceilings as well as
(O)
structural elements. All systems important to normal operation are functional.
Immediate No permanent drift. Structure substantially retains original strength and
occupancy Light stiffness. Minor cracking of facades, partitions, and ceilings as well as
(IO) structural elements.
Some residual strength and stiffness left in all stories. Gravity-load bearing
Life safety elements function. No out-of plane failure of walls or tipping of parapets.
Moderate
(LS) Some permanent drift. Damage to partitions. Building may be beyond
economical repair.
Collapse Little residual stiffness and strength, but load bearing columns and walls
prevention Severe function. Large permanent drifts. Some exits blocked. In-fills and un-braced
(CP) parapets failed or at incipient failure. Building is near collapse.

Table 4.2 Values of ηo for multi story frames with j stories [PP92]

j i 0.1 0.2 0.3 0.4 0.6 0.8 1.0 3.0 5.0

1 1 0.80 0.75 0.70 0.65 0.60 0.50 0.55 0.55 0.55
2 2 0.50 0.45 0.40 0.40 0.40 0.40 0.45 0.45 0.50
1 1.00 0.85 0.75 0.70 0.65 0.65 0.60 0.55 0.55
3 3 0.25 0.25 0.25 0.30 0.35 0.35 0.40 0.45 0.50
2 0.60 0.50 0.50 0.50 0.45 0.45 0.45 0.50 0.50
1 1.15 0.90 0.80 0.75 0.70 0.65 0.65 0.55 0.55
4 4 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.45
3 0.35 0.35 0.35 0.40 0.40 0.45 0.45 0.50 0.50
2 0.70 0.60 0.55 0.50 0.50 0.50 0.50 0.50 0.50
1 1.20 0.95 0.85 0.80 0.70 0.70 0.65 0.55 0.55

156
Table 4.3 Values of correction coefficient η1 for different beam stiffnesses [PP 92]

α1 0.1 0.2 0.3 0.4 0.6 0.8 1.0 3.0 5.0

0.4 0.55 0.40 0.30 0.25 0.20 0.15 0.15 0.05 0.05

0.5 0.45 0.30 0.20 0.20 0.15 0.10 0.10 0.05 0.05

0.6 0.30 0.20 0.15 0.15 0.10 0.10 0.05 0.05 0.00

0.7 0.20 0.15 0.10 0.10 0.05 0.05 0.05 0.00 0.00

0.8 0.15 0.10 0.05 0.05 0.05 0.05 0.00 0.00 0.00

0.9 0.05 0.05 0.05 0.05 0.00 0.00 0.00 0.00 0.00

157
 Chapter 5
SEISMIC RISK ASSESSMENT OF THE URBM BUILDING SYSTEMS
OF NORTHERN PAKISTAN

5.1 INTRODUCTION
In this chapter, material properties of brick masonry constructed in stone dust mortar, being commonly
used in Peshawar are discussed. These properties were determined by extensive laboratory testing
carried out as a part of present study as discussed in Chapter 3. The experimental data of the other
researchers of UET, Peshawar [AliQ 04, AliM 06, Ban 06 and Badr 06] was also used for this
purpose. After thoroughly studying the experimental data of these research works [AliQ 04, AliM 06,
Ban 06 and Badr 06] carried out on CSK and CS mortars, it was concluded that the mechanical
properties for brick masonry constructed in CSK 1:4:4 mortar and brick masonry constructed in CS
1:8 mortar were having almost same magnitudes, and, therefore, it was decided to use the mechanical
properties for brick masonry constructed in CSK 1:4:4 mortar for brick masonry constructed in CS 1:8
mortar. For the purpose of conservativeness, the magnitudes of various properties of the brick masonry
obtained from laboratory testing were also lowered. This decision enabled to apply the mechanical
properties of brick masonry to the brick masonry buildings’ stock located in other parts of Northern
Pakistan.
For the purpose of illustrating the methodology developed in Chapter 4, seismic assessment of
a double-story URBM example building was carried out, both in the in-plane and the out-of-plane
directions and explained in a stepwise manner.
Finally, results of the seismic assessment of 31 buildings are presented. The buildings’ stock consisted
of 17 single-story and 14 double-story buildings. The typologies of the buildings were selected by
keeping in view their common use in Peshawar and other urban areas of Northern Pakistan. Fragility
curves, indicating the probabilities of occurrence of various performance levels for various levels of
ground shaking, are presented and discussed.

5.2 ESTABLISHING THE MAGNITUDES OF THE MATERIAL PROPERTIES OF

THE MASONRY USED IN PESHAWAR

5.2.1 Compressive strength of masonry, fm

Based on experimental work carried out as a part of this research work and research work carried by
Ali [AliQ 04] and Ali [AliM 06], it was concluded that brick masonry used in Peshawar and
constructed with CSK 1:4:4 mortar has an average compressive strength of 650 psi. However, these
results belonged to specimens, which were constructed in laboratory under very favorable conditions.
Such conditions are seldom achieved in the practice. e.g., depending upon the source of bricks, there

158
can be significant variation in compressive strength of bricks for buildings constructed at same
location with in the city. Similarly, it is very common in non-engineered construction to use mortar
even for 2 to 3 hours after mixing, thus causing significant reduction in the compressive strength of
mortar and consequently the masonry product. Satisfactory curing of masonry is seldom accomplished.
Keeping in view these facts, a compressive strength of 500 psi was established for the URBM
buildings’ stock of Peshawar (approximately 75% of compressive strength achieved in laboratory).

5.2.2 Principal tensile strength of masonry, ftu

Based on results of diagonal tension tests on masonry prism discussed in Chapter 3, principal tensile
strength of masonry was taken as 3.5% of the compressive strength of masonry (i.e., 17.5 psi) for
URBM buildings’ stock of Peshawar.

5.2.3 Bond strength of masonry in tension, ft

Ali [AliQ 04], based on the results of experimental study carried work by as a part of his PhD research
work, concluded that average bond strength of masonry in tension (constructed in CSK 1:4:4) used in
Peshawar is 20.7 psi. However, for the purpose of conservativeness, bond strength of the masonry in
tension was ignored for the seismic capacity assessment of URM buildings’ stock of Peshawar.

5.2.4 Cohesion or bond shear strength, c, and coefficient of friction of masonry,

Under lateral cyclic loading, bed joints get cracked under the action of induced flexural tensile stresses
in both the directions, and consequently, cohesion between bricks and mortar is no longer available.
Therefore, cohesion of the masonry was ignored for evaluating seismic capacities of URBM buildings’
stock of Peshawar. The coefficient of friction for URBM, was taken as 0.55, keeping in view the
results of current experimental work and those carried out by Ali [AliM 06] on URBM constructed in
stone dust mortar.

5.2.5 Modulus of elasticity of the masonry, Em

Modulus of elasticity of masonry, in a direction perpendicular to bed joint, was found to be 294fm from
the experimental work carried out as a part of this work. Research work carried out by Ali [AliQ 04]
and Ali [AliM 06] on brick masonry constructed in CSK 1:4:4 mortar indicated that Em = 390fm and
244fm, respectively. Keeping in view the results of the current experimental work and those of other
researchers on CSK mortar [AliQ 04 and AliM 06], a value of Em = 350fm was established for URBM
buildings’ stock of Peshawar. However, no information exists in literature regarding modulus of
elasticity of masonry in a direction parallel to the bed joints, Emx. Limited information is available,
however, on the relation between compressive strengths of masonry in the directions perpendicular
and parallel to the bed joints. Lang mentioned [La 02] that for clay brick masonry constructed in lime
mortar, f 0.3 to 0.5 times f .

159
where fmx = Compressive strength of masonry parallel to the bed joints, and, fm = compressive strength
of the masonry perpendicular to the bed joints.
Tests results reproduced by Drysdael et al [DHB 99] indicate that for clay brick masonry
f 0.6 f . Thus if it is assumed that f 0.5 f and the elastic modulus factor for masonry, km,
is same in both the directions, then a relation between Em and Emx can be established as follows:
E .f . 0.5f
0.5 .
E .f .f
Therefore, the magnitude of Emx used in this work was decided to be taken as equal to 0.5Em.

5.2.6 Modulus of rigidity of the masonry

Banori [Ban 06] concluded from the results of diagonal tension tests on the masonry prisms that
modulus of rigidity of the brick masonry constructed in stone dust mortar, Gm = 0.1Em. On the other
hand, based on the results of the quasi-static cyclic tests on the masonry piers, as discussed in Section
3.8.2, the average value of the modulus of rigidity of the brick masonry constructed in stone dust
mortar, Gm, was found to be 0.45Em. Due to significant difference in the results, an average value of
Gm = 0.20Em was decided to be used for the seismic assessment of the URBM buildings of Peshawar.

5.3 STRUCTURAL MODELING

In Pakistan, structural walls have a thickness ≥ 9 in. Walls with a thickness of 4.5 in. are normally
provided for partitioning purpose. However, 4.5 in. walls are constructed after the slabs are cast and
consequently, these do not support gravity loads. Hence, for the purpose of seismic assessment of
URBM buildings, only those walls were considered as structural walls, which had a thickness ≥ 9 in.
Similarly, walls orthogonal to the structural walls produce flange action when subjected to
lateral loading and thus increase the lateral resistance of the structural walls. However, the flange
action may be poor (or even absent), even in relatively new buildings, due to improper toothing,
foundation settlement (especially in external walls) and minor damage due to past earthquakes.
Hence, for the purpose of seismic capacity evaluation of the buildings’ stock of Peshawar, it was
decided to ignore the flange action. Additionally, the height of masonry buildings in Peshawar and
other parts of Northern Pakistan occasionally exceed two stories and a typical foundation section,
shown in Figure 5.1, is followed in construction practice. This section was, therefore, also used for
determining stiffness of spandrels connected to the bottom of ground floor piers in URBM buildings.

5.4 DESIGN RESPONSE SPECTRUM

As discussed in Section 4.7.4, UBC-97 design response spectrum [UBC 97] as recently adopted by the
Building Code of Pakistan [BCP 07], shown in Figure 4.4, was used to determine the horizontal
ground acceleration, (ag)∆, causing a lateral displacement, ∆, at the top of the building. (ag)∆ was
determined by using Equation 4.34 as follows:
160
 S ∆
a .
∆ A
where:
(Sa)∆ = Spectral acceleration in the elastic SDOF system corresponding to the lateral top displacement,
∆, in the actual building; and
Af = Factor by which the ground motions amplify at the top of the equivalent elastic SDOF system.
It is shown in the Figure 4.4 that spectral acceleration linearly amplify 1 to 2.5 times the
ground acceleration when the time period of the elastic SDOF system increase from 0 to To. The
amplification of ground acceleration, however, remains constant (i.e., 2.5) for elastic SDOF systems
with To ≤ T ≤ Ts . Where To and Ts are controlling time periods, as recommended by Building Code of
Pakistan [ BCP 07], and are calculated by using Equations 4.35a and 4.35b as follows:
C
T .
2.5C
T 0.2 T .
where as Cv and Ca are seismic coefficients, which, in addition to soil type, also depend upon the zone
in which a building is located [ BCP 07]. Soil conditions in Peshawar vary from loose soil in ‘City’
area to very dense soil in Hayatabad (a satellite township). However, no detailed geological
information exists regarding variation in soil conditions from place to place within the Peshawar. In
the absence of geological information regarding soil conditions, Building Code of Pakistan [BCP 07]
recommends to assume soil type SD, which was, therefore, adopted for seismic capacity assessment of
URBM buildings’ stock of Peshawar. Peshawar is placed in Zone 2b [BCP 07], for which Ca and Cv
for soil type SD are 0.28 and 0.4, respectively.

5.5 SEISMIC CAPACITY EVALUATION OF EXAMPLE BUILDING

A double-story small residential building, shown in Figure 5.2 and 5.2, was considered as a case study
for illustrating the methodology proposed in Chapter 4. The detailed calculations carried out to
evaluate the seismic capacity of the building are provided, in a step wise manner, as follows:
Step 1: Establishing magnitudes of the material properties
Following mechanical properties of the masonry, as discussed in Section 5.2, were established for
evaluating the seismic capacity of the example building
i. Compressive strength of the masonry, fm = 500 psi;
ii. Principal tensile strength of the masonry, ftu = 17.5 psi;
iii. Bond strength of masonry in tension, ft = 0;
iv. Cohesion, c = 0;
v. Coefficient of friction of masonry, µ = 0.55;
vi. Modulus of elasticity of the masonry perpendicular to the bed joints, Em = (350)(500) =175 ksi;
vii. Modulus of elasticity of the masonry parallel to the bed joints, Emx = (0.5)(175) = 87.5 ksi; and
161
viii. Modulus of rigidity of the masonry, Gm = 0.2(175) = 35 ksi

Step 2: Geometrical properties of the piers and spandrels

According to architectural details provided in Figures 5.2 and 5.3, the geometrical properties of
structural elements are given in Table 5.1 for the shorter direction of the building. The overall depths
of spandrels connected to the bottom of the ground floor piers were determined by adding a foundation
depth of 4.5 ft. below floor level.

Step 3: Determination of inflection point in the piers

Flexural stiffnesses of piers and spandrels were determined by using the Muto method [PP 92]. The
foundation was assumed to have a uniform width of 9 in. throughout its depth. This was assumed for
the purpose of simplification in calculations. However, the magnitudes of flexural stiffness of the
bottom spandrels were not significantly changed as a result of this assumption. The flexural stiffness
of each individual pier, Kpf, and spandrel, Kspf, were calculated using Equations 4.3 and 4.4 as follows:
2E . I
K .
l
E .I
K .
l

t. l t. h
where I and I
12 12
After calculating K and α1 for each pier using Equations 4.5 and 4.6, ηo and η1 were determined by
using Table 4.2 and 4.3, respectively. Similarly, ψ‫׳‬, was determined by using the criteria set forth in
Equation 4.8a and 4.8b. The calculated values of afore mentioned parameters are given in Table 5.2

Step 4: Lateral strength of the piers

Calculation of lateral force corresponding to flexural tensile cracking force, Vfl,cr, and the lateral force
corresponding to the failure of a pier, Vsf, require the determination of pressure acting at the top end of
the pier due to gravity loads. Following values were used for this purpose for calculating the gravity
loads:
i. Weight density of plain concrete, γconc. = 144 lb/ft3;
ii. Weight density of R.C.C, γRCC. = 150 lb/ft3;
iii. Weight density of brick masonry, γm. = 120 lb/ft3;
iv. Weight density of mud, γmud. = 110 lb/ft3;
v. Weight of 6 in. thick floor slab = 75 lb/ft2;
vi. Weight of 1.5 in. thick floor finishing on floors (other than the top floor) = 18 lb/ft2;
vii. Weight of 4.5 in. thick mud insulation on top floor = 45 lb/ft2; and
viii. Weight of 1.5 in. thick brick tiles insulation on top floor = 16.25 lb/ft2
162
 ix. Live load on floors (other than the top floor) = 15 lb/ft2
In order to take into account the effect of vertical ground acceleration, the vertical pressure acting on a
pier was reduced by subtracting 2/3Ca.σo, as already discussed in Section 4.6.4. Similarly, Ca for soil
type SD was taken equal to 0.28 for Peshawar (zone 2b) as per recommendations of the Building Code
of Pakistan.
Equations 4.9b, 4.11, 4.12b and 4.13 were then used to determine values Vfl,cr, Vr,Vsl and Vd,
respectively.
σ f l .t
Vf , .
6ψ . h
l .t σ σ
V . 1 .
α 2 0.85f
V µ. l . t .

f σ
V 1 l .t .
b f

Lateral force corresponding to the failure of a pier, Vsf, was taken as the minimum of Vr,Vsl and Vd.
The calculated values of above mentioned parameters are given in Table 5.3

Step 5: Bi-linear idealization of the walls corresponding to the piers

In order to construct the idealized bi-linear curves for the corresponding walls, Vu, was determined by
the relation Vu = 0.9Vsf. Similarly, Kep dy, and δy were determined by using Equations 4.16b, 4.17a
and 4.17b, respectively as follows:

t. E l
0.75
12 h
K .
3η 1 1 E l
6 10 G h
V
d .
K
d
δ .
h
du was calculated by the criteria set forth in Equations 4.18a and 4.18b. where, µp, for each pier was
then calculated by using Equation 3.9 as follows :
d
µ .
d
Similarly, dywe, duwe and ∆yw were determined by inserting values of hpb, hp, hwe and hw (given in Table
5.1) and dy (calculated from Equation 4.17a) in Equations 4.21b , 4.22(a and b) and 4.20b, respectively
as follows:

163
 d
d h .
h
d δ .h 0.0035 h for shear failure .
d δ .h 0.0080 h for rocking failure .
d
∆ .h .
h
∆uw was, then, calculated by inserting ∆yw and duwe/dywe = µwe in Equation 4.23d.
h
∆ 1 µ 1 ∆ .
h
Calculated values of the above mentioned parameters are given in Table 5.4

Step 6: Capacity curve of the building and identification of various performance levels
Using the values of Vu, ∆yw and ∆uw, capacity curve for all the nine walls along the shorter direction of
the building were constructed as shown in Figure 5.4. Based on the rigid diaphragm assumption, the
capacity curves of all the walls were superimposed to get the capacity curve of the building along its
shorter direction.
Lateral displacements at the top of the building for O, IO, LS and CP performance levels were then
determined using the criteria set forth in Section 4.4. Results of all the calculations are provided in
Table 5.5. According to criteria set forth in Section 4.4, drifts at the top of the building for various
performance levels, given in Table 5.5, were calculated as follows:
i. As discussed in Section 4.4, O performance level for an URM building can be related to the
lateral displacement at the top of building the when the flexural tensile cracks develop in the
bed joint of at least half of the piers in the building. Wall No. 4 was the fifth (of nine walls)
which underwent flexural tensile cracking and fulfilled this criterion. Hence, lateral
displacement at the top of wall No. 4 at flexural tensile cracking was taken as the lateral
displacement corresponding to O performance level in the building, ∆O.
ii. No wall in the building underwent shear cracking. Therefore, minimum lateral displacement
at the top of a wall in which the component pier attained a drift ratio of 0.16% was taken as
lateral displacement corresponding to IO performance level in the building , ∆IO. Wall No. 10
fulfilled this criterion.
iii. CP performance level was related to the distress state when the building lateral resistance
dropped by 20%. This was determined from capacity curve of building shown in Figure 5.4.
The lateral displacement at the top of the buildings at this performance level, ∆CP, was found
to be 0.69 in.
iv. LS performance level was related to a distress state when the lateral displacement at the top
of the building reach a value equal to the average of the lateral top displacement at IO and CP
performance levels.
164
 0.13 0.69
i. e. , ΔLS 0.41 in.
2

Step 7: Base shear and displacement in the corresponding elastic system for various
performance levels
Lateral displacements at the top of building for various performance levels determined in previous
step, were used to find base shear and displacement in the corresponding linear elastic MDOF system.
This was carried out as follows:
Displacement ductility of building, (µb)∆, corresponding to a lateral top displacement, ∆, was
determined by using Equation 4.27b.
∆
µ ∆ .
∆
where value of ∆yb, as used in Equation 4.27b, was calculated by using Equation 4.2 as:
V V
Δ .
K ∑K
where, Vyb, determined from the capacity curve as shown in Figure 5.4, was found to be 32.8
kip. Similarly elastic stiffness of the building, Kb, determined by summing stiffnesses of the
component walls given in Table 5.4 was found to be 242 k/in.
After determining (µb)∆, using equation 4.27b, the corresponding values of R, Vbe and ∆be were
calculated by using Equations 4.27a, 4.26b and 4.28, respectively as follows:
R∆ 2 µ ∆ 1 .
V ∆ R∆. V .
V ∆
∆ ∆ .
K
The calculated values of above mentioned parameters are given in Table 5.6

Step 8: Modal participation factor, Γ, equivalent mass, mE, and equivalent height, hE, for the
equivalent elastic SDOF system
In order to determine Γ, mE and hE in the elastic SDOF system corresponding to the building ( MDOF
system), lumped masses at each story level were calculated by summing the weight of slab supported
by each wall and are given in Table 5.7. Based on the assumption of inverted triangular mode shape,
Γ, mE and hE [Ch 01], given in table 5.8, were determined by using Equations 4.30, 4.32 and 4.33,
respectively, as follows:
∑ m.
Г .
∑ m.

165
mE m. .

∑ h .m .
hE .
∑ m.
Following values were obtained as a result of these calculations:
Г 1.17, mE 3446 slug, and hE 18.52 ft.

Step 9: Ground acceleration corresponding to the various performance levels

Ground acceleration in the elastic SDOF system, (ag)∆, causing displacement, ∆, in the corresponding
building can be determined once the magnitudes of the spectral displacement, (Sd)∆, and spectral
acceleration, (Sa)∆, are known. (Sd)∆ is determined by dividing the displacement in the corresponding
elastic MDOF system, (∆be)∆, by the modal participation factor, Γ. (Sd)∆ and (Sa)∆ were determined
by using Equations 4.29 and 4.31a as follows:
S ∆ ∆ ∆⁄ Γ .
S ∆ S ∆. 2πf .
where:

1 KE 1 K
f .
2π mE 2π mE

f1 was found to be 4.41 Hz. The corresponding period of vibration determined by using the
relation T1 = 1/f1 was found 0.23 s. Amplification factor, Af, was determined from design response
spectrum [BCP 07] using T1 and the controlling time periods (i.e., TS and To), which were calculated
by using the values of seismic coefficients Ca and Cv
For a building located in zone 2b and constructed on soil type SD, Ca and Cv are 0.28 and 0.4,
respectively [ BCP 07]. Consequently:
T C ⁄ 2.5C 0.57 s and T 0.2 T 0.11 s
Since To < T1 < Ts , therefore Af, as determined from the design response spectrum shown in Figure
4.4 is equal to 2.5, and:
S ∆ S ∆
a ∆
A 2.5
The calculated values of above mentioned parameters are given in Table 5.9

Step 10: In-plane capacity of the example building in the longer direction
The calculations carried out, in previous nine steps of the example to determine ground acceleration
corresponding to various performance levels, belonged to the shorter direction of the example
building. However, it is not possible to predict the direction of earthquake. Moreover, the seismic

166
waves are normally skewed to the building, thus inducing seismic forces in both the direction of a
building. Therefore, seismic capacity assessment of a building need to be carried out for both
directions of the building.
Data required for the ground acceleration evaluation corresponding to various performance
levels of the building in longer direction are given in Table 5.10. It may be noted that only important
parameters are provided for the purpose of comparison with the shorter side of building in order to
decide the more vulnerable direction of the building.
The capacity curve of the building along the longer direction is shown in Figure 5.5. The
displacement at CP performance level determined from Figure 5.5 was found to be 0.47 in. Where as
the displacement at the LS performance limit state, taken as average of displacements corresponding to
IO and CP performance levels, was found to be 0.305 in.
Other characteristics of the building in longer direction were found to be:
i. Vyb = 79.2 kip;
ii. KE = Kb = 734 kip/in.;
iii. f1 = 8.04 Hz; and
iv. T1 = 0.12 s.
Magnitude of the other parameters required to determine horizontal ground acceleration corresponding
to various performance levels are given in Table 5.11.

Step 11: Comparison of the horizontal ground accelerations corresponding to the various
performance levels of the building in both the orthogonal directions
Results of ground accelerations, for both the directions of example buildings, corresponding to various
performance levels are given in Table 5.12. By comparing the results given in Table 5.12, the
magnitudes of ag/g for shorter side of building were found to be lower as compared to the longer
direction side of the building for all the performance levels. Hence, results of the shorter direction
govern the seismic vulnerability of building.
Step 12 Seismic capacity of the building in the out-of-plane direction
Since the out-of-plane bending capacity of the masonry walls are sensitive to height (Mu,op α 1/h2) and
vertical load (Mu,op α Rc). Therefore, a wall with the lowest vertical load was selected for checking the
seismic capacity of building in the out-of-plane direction. In case of the example building, wall No. 7,
located along longer direction with the minimum vertical load, N= 301 lb/ft, fulfilled this criteria. The
seismic capacity of the wall No.7 in the out-of-plane direction is determined as follows:
Step 12.1 Mid-height curvature and deflection at the verge of flexural tensile cracking of the wall
No. 7 in out-of-plane direction
Following data was used to determine mid-height curvature and deflection at the onset of out-of-plane
flexural tensile cracking of the wall No. 7:

167
 i. Load at the top of the wall per unit length, N = total load/ length of wall = 7070/23.5 =
301 lb/ft;
ii. Height of the wall, h = 10.5 ft;
iii. Height of the wall's centroid above floor level, hc = 11+10.5/2 = 16.25 ft;
iv. Equivalent height of the wall, hE = 18.52 ft;
v. Thickness of the wall, t = 0.75 ft; and
vi. Mass of the wall per unit height for a unit length, mw = weight of the wall per unit height
per unit length/g = [(0.75)(1)(1)(120)]/32.2 = 2.8 slugs.
Mid-height curvature and deflection at the onset of flexural tensile cracking in the wall in the out-of-
plane direction were determined by using Equations 4.45 and 4.46b as follows:
2R
φ , .
t. E
20R t. h
∆ , .
1152E . I
where Rc = Vertical load per unit length at mid height of the wall by taking into account the effect of
vertical ground acceleration and was calculated as:
2 W
R 1 C .g N .
3 2
Similarly:
R .t
M , .
6
2R
σ , .
t
2R
, .
t. E
Calculation by the above-mentioned equations resulted in the following values:
i. Rc = 629 lb;
ii. Mop,cr = 78.6 ft.lb;
iii. σop,cr = 12 lb/in2;
iv. , = 6.66 x10-5 ft/ft ;
v. φop,cr = 8.88 x10-5 rad/ft; and
vi. ∆op,cr = 0.0122 in.

Step 12.2: Mid-height deflection, ∆op, and associated acceleration, aop, for various levels of
compressive strains, ,

According to the criteria set forth in Section 4.8.3, mid height deflection, ∆op, and corresponding
acceleration, aop, were determined for various magnitudes of compressive strains, , , by using
Equations 4.52b and 4.39, respectively, as follows:
168
 φ
∆ ∆ , .
φ ,

8R
a x ∆ .
m .h
where:
,
φ .
c
in which:
2R
c .
, .E

and:
t c
x for linear elastic stresses; and
2 3
t a R
x x at the crushing of masonry in compression, where a
2 2 0.85f
Initially, values of ∆op and aop for wall No.7 were determined for , 2 , as shown in first
column of Table 5.13. Values of strains were then stepwise increased to a limit such that the
corresponding stresses were within linear elastic range (i.e., fc,op ≤ fm/3).
t a R
Instability occured in the wall at Δ x where a
2 2 0.85f . t
aop/g and ∆op for various levels of compressive strains in masonry were calculated and given in Table
5.13. These results were used to draw the graph between aop/g and ∆op as shown in Figure 5.6

Step 12.3: Acceleration in the corresponding linear elastic system

Acceleration in the corresponding linear elastic system, ae,op, was determined by using Equation 4.40d
based on equal energy principle as follows:
a ,
2A . K .
g
where A2 = area under the curve shown in Figure 5.7 = 0.028 ft, and Kop= slope of a-∆ diagram =
a
g 0.0629
61.7 1/ft
Δ 0.00102
Inserting values of A2 and Kop in Equation 4.40d results in ae,op/g = 1.87
ae,op/g is the acceleration acting at the mid-height of wall No.7 at top floor i.e., hc = 16.5 ft. Since hE
=18.52 > hC, spectral acceleration in the corresponding elastic SDOF system with an equivalent height,
hE is calculated by using Equation 4.53a as:
S , a , hE
. 2.13.
g g h
Assumption of rigid diaphragm implies that the lateral displacement in a wall in the out-of-
169
plane direction is equal to the lateral displacement in the orthogonally connected intersecting walls in
the in-plane direction. Therefore, time period of the building in the shorter direction (a direction
perpendicular to the side along which wall No.7 exist) is used to determine Af. In the other words,
value of Af equal to 2.5, as calculated in step No. 9, applies in this case. Corresponding ground
acceleration is, therefore, equal to ag,op/g = 2.13/2.5 = 0.85 (or ag,op =0.85g) which is greater than
ground acceleration of 0.296g corresponding to collapse prevention level in the in-plane direction
along the shorter side of the building. Hence the example building is safe against out-of-plane collapse
of walls.
Based on the criteria set forth in step No. 12, it was determined that a 9 in. thick wall at the
top floor, restrained on both ends by rigid floor, with a minimal axial load of zero at the top of wall in
a three story building (maximum height of the masonry buildings in Northern Pakistan) with equal
mass distribution on each floor, will collapse at a ground acceleration of approximately 0.55g.
Similarly, it was determined that a 4.5 in. thick wall at the top floor, under afore mentioned
conditions, will collapse at ground acceleration of 0.16g and 0.13g for two and three story building,
respectively. The vulnerability is further increased if the walls are free standing cantilevers. It is,
therefore, recommended not to construct 4.5 in. URM walls on the second (and above) floor level in a
building.

5.6 SEISMIC RISK ASSESSMENT OF REPRESENTATIVE BUILDINGS

A total of 31 buildings were selected for carrying out seismic risk assessment of URBM buildings’
stock of Peshawar. The buildings’ stock consisted of 17 single-story and 14 double-story buildings.
The buildings were selected keeping in view their common typological use in Peshawar and other
urban areas of Northern Pakistan. For the purpose of comparison, single and double story buildings
were divided in to two separate groups. Results of ground acceleration corresponding to various
performance levels for single and double-story buildings are given in Tables 5.14 and 5.15,
respectively. Where, statistical analysis of the data of various performance levels of the buildings is
summarized in Table 5.16.
It was concluded from the results summarized in Table 5.16 that difference in the average values of
ag/g corresponding to various performance levels, except for operational performance level, is
insignificant. Hence the results of single and double-story buildings can be combined to represent the
behavior of all buildings.
Due to variation in the results with in the same building class, it is more convenient to express
the vulnerability of a building class using a probabilistic distribution instead of a single deterministic
vulnerability curve. The variation in results can be represented by a normal distribution, which is
defined by the mean and the standard deviation of population. The cumulative distribution for each
damage grade or performance level in the literature is called a fragility curve [KNKH 97, HLH 97].

170
The mean, µx, and the standard deviation, σx, of the population are estimated using the mean and the
standard deviation of the sample. In the other words, building fragility curves are defined as functions
that describe the probability of a building belonging to a certain building class of reaching or
exceeding a particular damage grade (i.e., performance level) given a deterministic estimate of the
ground acceleration

a PL
P PL PL \ .
g

Using the Equation 5.3, the fragility curves of 17 single-story , 14 double-story and all 31
buildings were drawn and shown in Figures 5.8, 5.9 and 5.10, respectively. As already discussed, the
results of ag/g corresponding to various performance levels for single and double-story building do not
differ considerably. It is, therefore, recommended that fragility curves for all 31 buildings, shown in
Figure 5.10, can be used both for single as well as double story buildings.
In order to study the variation in the magnitudes of seismic capacities of URBM buildings
with the increase in vertical ground acceleration, seismic assessments of randomly selected buildings
were also carried out by selecting design response spectrums of zone 2a, 3 and 4. However, no
considerable variations in the results were found. The results of ground acceleration for various
performance levels of example building, shown in Figure 5.2, are summarized in Table 5.17. By
comparing the results given in Table 5.17, it is concluded the buildings’ stock, located in any seismic
zone, will reach a specific performance level at almost same magnitude of ground acceleration.
It is important to note that the URBM buildings in Peshawar and other areas of Northern
Pakistan are constructed either in CSK 1:4:4 or CS 1:8 mortar. It was concluded, by comparing the
results of experimental works [AliQ 04, AliM 06, Ban 06 and Badr 06] with the current experimental
work that the mechanical properties of URBM constructed in CSK 1:4:4 mortar do not significantly
differ from those of URBM constructed in CS 1:8 mortar. Similarly, quality of the URBM as well as
building typology in urban areas of Northern Pakistan do not also vary significantly. Based on these
facts, it was, therefore, concluded that the results of seismic risk assessment of URBM buildings of
Peshawar, constructed in stone dust mortar, can be applied to the URBM buildings (constructed in
CSK and CS mortar) located in other urban areas of Northern Pakistan.
Thus from the fragility curve shown in Figure 5.10, it can be concluded that for URBM buildings
constructed in Northern Pakistan on soil type SD:
1. At a ground acceleration of 0.12g, PGA corresponding to Zone 1 (no significant town in
Northern Pakistan), the whole buildings’ stock will suffer insignificant damage (O
performance level). there is a probability of 0.4 that buildings’ stock will suffer minor damage

171
 (IO performance level). Such buildings can be re-occupied for use but will need to be repaired
after accommodating. However, no building will be distressed to LS performance level.
2. At a ground acceleration of 0.22g, PGA corresponding to Zone 2a (D.I. Khan), all the
buildings will be distressed beyond IO performance level. However, there is a probability of
0.4 that buildings’ stock will suffer damage corresponding to LS performance level. Buildings
suffering such damages can be occupied after repair work. Similarly, there is a very low
probability of 0.12 that buildings’ stock will distress to CP performance level.
3. At a ground acceleration of 0.28g, PGA corresponding to Zone 2b (Islamabad, Mardan,
Peshawar, Swabi, Haripur, Kohat, Bannu, Tank, Miran shah) , there is very high probability
of 0.8 that the URBM building stock will be distressed to LS performance level. However,
there is a probability of 0.4 that will buildings’ stock will suffer damage corresponding to CP
performance level. Buildings undergoing such damage cannot be re-used for accommodation
and need to be reconstructed after demolishing.
4. At a ground acceleration of 0.36g, PGA corresponding to Zone 3 (Abbotababad, Manshera,
Mingora, Dir, Gilgit, Dasu), almost whole URBM buildings’ stock will suffer damage
corresponding to LS performance level. Similarly, there is a high probability of 0.75, that the
URBM buildings’ stock will distress to CP performance level.
5. At a ground acceleration of 0.44g, PGA corresponding to Zone 4 (Balakot, Bagh,
Muzaffarabad, Chitral), there is a probability as high as 0.92 that the URBM buildings’ stock
will distress to CP performance level.

5.7 CONCLUSIONS
1. It was found that a 9 in. thick wall at the top floor in a two story building, restrained on both
ends by rigid floor and supporting minimal gravity load will collapse in the out-of-plane
direction at a ground shaking of 0.65g. Where, the maximum value of ag in the in-plane
direction of all the double story buildings was found to be 0.401g (Building P17-5m, Table
5.15). Hence, it is concluded the double story buildings are safe in the out-of-plane direction.
2. It was found that 4.5 in. thick walls at the top floor in a three story and two story building,
restrained on both ends by rigid floor and supporting minimal gravity load will collapse in the
out-of-plane direction at ground shaking of 0.13g and 0.16g, respectively. These values are
well below the ground acceleration triggering the collapse of building in the in-plane
direction. It is, therefore, recommended not to construct 4.5 in. walls on the second (and
above) floor levels.
3. It is concluded that a horizontal ground acceleration of 0.12g, PGA corresponding to Zone 1
(no significant town in Northern Pakistan) for soil SD, will cause very light damage (damage
corresponding to O performance level) in the whole buildings’ stock. For this level of ground

172
 shaking, there is a probability of 0.4 that the buildings’ stock will suffer minor damage
(damage corresponding to IO performance level). Such buildings can be re-occupied for use
but will need to be repaired after accommodating.
4. It is concluded that an earthquake having a horizontal ground acceleration of 0.22g, PGA
corresponding to Zone 2a (D.I. Khan) for soil SD, all the buildings will be distressed beyond
IO performance level. However, there is a probability of 0.4 that buildings’ stock will suffer
damage corresponding to LS performance level. Buildings suffering such damages can be
occupied after repair work. Similarly, there is a very low probability of 0.12 that buildings’
stock will distress to CP performance level.
5. At a ground acceleration of 0.28g, PGA corresponding to Zone 2b (Islamabad, Mardan,
Peshawar, Swabi, Haripur, Kohat, Bannu, Tank, Miran shah) for soil SD, there is a high
probability of 0.8 that the URBM building stock will be distressed to LS performance level.
However, there is a probability of 0.4 that will buildings’ stock will suffer heavy damage (
damage corresponding to CP performance level). Buildings undergoing such damage cannot
be re-used for accommodation and need to be reconstructed.
6. At a ground acceleration of 0.36g, PGA corresponding to Zone 3 (Abbotababad, Manshera,
Mingora, Dir, Gilgit, Dasu) for soil SD, almost whole URBM buildings’ stock will suffer
damage corresponding to LS performance level. Similarly, there is a high probability of 0.75,
that the URBM buildings’ stock will distress to CP performance level.
7. At a ground acceleration of 0.44g, PGA corresponding to Zone 4 (Balakot, Bagh,
Muzaffarabad, Chitral) for soil SD, there is a probability as high as 0.92 that the URBM
buildings’ stock will distress to CP performance level.
8. It is recommended, based on the conclusions mentioned in 3 to 7, that the URBM buildings
constructed with 9 in. thick wall and having a height up to two-stories can be safely used in
locations placed in zone 2b and below.
9. It is recommended, based on the conclusions mentioned in 3 to 7, that the URBM buildings
constructed with 9 in. thick wall, irrespective of the story height, shall not be constructed in
any location placed in zone 3 and above.
10. It is concluded that the seismic capacity of a brick masonry building is not significantly
influenced by the soil conditions. However, the soil conditions plays a vital role on the
seismic demand on a building. A loose soil results in the increase in the seismic demand on
the building and thus increases the probability of damage corresponding to a certain
performance level. In order to carry out accurate seismic risk assessment, it is, therefore,
recommended to carry out a detailed microzonation, at least for the densely populated cities of
the Northern Pakistan.

173
Figure 5.1 Cross section of a typical foundation, used in Pakistan, to support
up a brick masonry building up to double-story height

174
 4'-3" 4'-3" 3'-3" 4'-3" 4'-3" 3'-3"

4 5 6

STORE STORE
10'

10'
16
BED ROOM BED ROOM
9'

8
7
23'-6"

23'-6"
BATH
11' 10'-3"
10
LOUNGE LOUNGE BATH
9
32'

32'

3'
4'-3" 3'
3'-6"

11 13
12
3 DRAWING &
DINING KITCHEN
9'

BED ROOM TERRACE

3'-6"

15 14

6'-5" 4'-6"
11' VERANDAH
10'-3"
2 1

3'-3"
3'-3" 3'-3" 3'-3"

LAWN

GROUND FLOOR PLAN FIRST FLOOR PLAN

Figure 5.2 Ground and first floor plans of example building

175
1'
6"
11'
6"

3'

4'-6"
11'-6"

6'

5'-6"
2'

F.F.L
1'-6"

1'-6"

N.S.L

FRONT ELEVATION

Figure 5.3 Front elevation of example building shown in Figure 5.2

176
 40 Operational Level
Immediate Occupancy Level
Life Safety Level
(∆yb,Vyb) Collapse Prevention Level

30
Base shear (kips)

20

10

0
0 0.2 0.4 0.6 0.8 1 1.2
Lateral displacement at the top floor (in.)

Figure 5.4 Capacity curve of example building, along with capacity curves of component
walls, in the shorter direction

100
Operational Level
Immediate Occupancy Level
Life Safety Level
80 Collapse Prevention Level
Base shear (kips)

60

40

20

0
0 0.2 0.4 0.6 0.8 1
Lateral displacement at the top floor (in.)

Figure 5.5 Capacity curve of example building, along with capacity curves of
component walls, in the longer direction,
177
 0.2

0.15

aop / g

0.1

0.05

0
0 0.1 0.2 0.3 0.4
∆op (inches)

Figure 5.6 Variation of mid height deflection in wall No. 7 with corresponding acceleration, aop/ g

178
 1.0

0.8
Probability of occurence

0.6

0.4

Operational Performance Level

0.2
Immediate occupancy Performance Level
Life safety performance Level
Collapse prevention performance Level
0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Ground Acceleration, ag/g

Figure 5.8 Fragility curves for single-story buildings for various performance levels

1.0

0.8
Probability of occurence

0.6

0.4

Operational Performance Level

0.2
Immediate occupancy Performance Level
Life safety Performance Level
Collapse prevention performance Level
0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Ground Acceleration, ag/g

Figure 5.9 Fragility curves for double-story buildings for various performance levels

179
 1.0

0.9

0.8
Probability of occurence

0.7

0.6

0.5

0.4

0.3

0.2
Operational Performance Level
Immediate occupancy performance Level
0.1 Life safety Performance Level
Collapse prevention performance Level
0.0
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
Ground Acceleration, ag/g

Figure 5.10 Fragility curves for all buildings for various performance levels
180
 Table 5.1 Geometrical properties of structural members of the example building along shorter side

Wall No. 1 2 4 5 6 9 10 11 15

Thickness of pier, t , ft 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75

Length of pier, lp , ft 3.25 3.25 4.25 4.25 3.25 11.00 3.00 4.25 6.50

Height of pier, hp , ft 6.00 6.00 6.00 6.00 6.00 10.50 4.50 10.50 10.50

Distance of pier’s bottom from floor 2.00 2.00 2.00 2.00 2.00 0.00 4.00 0.00 0.00
level, hpb , ft
Height of wall element corresponding to 8.00 8.00 8.00 8.00 8.00 10.50 9.50 10.50 10.50
the pier, hwe , ft = hp+hpb
Over all height of the corresponding 21.50 21.50 21.50 21.50 21.50 21.50 10.50 21.50 10.50
wall, hw , ft
Length of spandrel connected to the left 6.00 0.00 0.00 4.00 2.50 0.00 0.00 3.00 0.00
side of pier, lsp1 , ft
Length of spandrel connected to the 0.00 6.00 4.00 3.00 2.00 0.00 2.00 0.00 0.00
right side of pier, lsp2 , ft
Depth of spandrel connected to the top 5.50 0.00 0.00 5.50 3.00 0.00 0.00 4.00 0.00
end on left side of pier, hsp1, ft
Depth of spandrel connected to the top 0.00 5.50 5.50 3.00 5.50 0.00 2.50 1.50 0.00
end on right side of pier, hsp2 , ft
Depth of spandrel connected to the 6.50 4.50 4.50 6.50 6.50 4.50 4.50 4.50 4.50
bottom end on left side of pier, hsp3 , ft
Depth of spandrel connected to the 4.50 6.50 6.50 4.50 6.50 4.50 8.50 4.50 4.50
bottom end on right side of pier, hsp4 , ft

181
 Table 5.2 Inflection points in the piers of the example building along shorter side

Wall No. 1 2 4 5 6 9 10 11 15

Kpf , kip.in. 216267 216267 483623 483623 216267 4791600 226800 414534 988650

Kspf1 , kip.in. 262041 0.001 0.001 393061 102060 0.001 0.001 201600 0.001

Kspf2 , kip.in. 0.001 262041 393061 85050 786122 0.001 73828 0.001 0.001

Kspf3 , kip.in. 432534 0.001 0.001 648802 1038083 0.001 0.001 287044 0.001

Kspf4 , kip.in. 0.001 432534 648802 287044 1297603 0.001 2901741 0.001 0.001

1.61 1.61 1.08 1.46 7.45 0.00 6.56 0.59 0.00

α1 0.61 0.61 0.61 0.51 0.38 1.00 0.03 0.70 1.00

η0 0.58 0.58 0.6 0.58 0.55 1 0.55 0.66 1

η1 0.05 0.05 0.07 0.05 0.05 0 0.05 0.07 0

η = η0+ η1 0.63 0.63 0.67 0.63 0.6 1 0.6 0.73 1

ψ‫׳‬ 0.63 0.63 0.67 0.63 0.6 1 0.6 0.73 1

Table 5.3 Failure modes and lateral strengths of the piers of the example building along shorter side

Wall No. 1 2 4 5 6 9 10 11 15

σo , ksi 0.025 0.022 0.017 0.019 0.021 0.017 0.017 0.016 0.006

Vfl,cr , kip 1.25 1.10 1.41 1.68 1.08 3.48 1.02 1.04 0.44

Vr , kip 3.54 3.14 4.03 4.80 3.09 10.07 2.94 2.99 1.31

Vsl , kip 4.80 4.23 4.40 4.92 3.96 10.95 3.03 4.10 2.37

Vd , kip 6.37 6.14 8.05 8.28 6.04 29.09 5.31 7.44 9.52

Failure Rocking Rocking Rocking Rocking Rocking Rocking Rocking Rocking Rocking
mode
Vsf , kip 3.54 3.14 4.03 4.80 3.09 10.07 2.94 2.99 1.31

182
 Table 5.4 ‘Yield’ displacements, ultimate displacements and displacement ductility
factors of the piers, wall elements and walls of the example building along shorter side
Wall No. 1 2 4 5 6 9 10 11 15

Vu , kip 3.19 2.83 3.63 4.32 2.78 9.07 2.64 2.69 1.18

Kep , kip/in. 53.0 53.0 83.5 87.6 55.9 128.3 82.0 57.6 44.5

Kw, kip/in 14.8 14.9 23.6 24.6 15.5 62.4 35.3 18.6 43.7

dy , in. 0.060 0.053 0.043 0.049 0.050 0.071 0.032 0.047 0.027

du , in. 0.576 0.576 0.576 0.576 0.576 1.008 0.432 0.672 1.008

dywe , in. 0.080 0.071 0.057 0.065 0.067 0.071 0.060 0.047 0.027

duwe , in. 0.596 0.594 0.590 0.592 0.593 1.008 0.460 0.672 1.008

∆yw , in. 0.215 0.190 0.154 0.176 0.179 0.145 0.075 0.144 0.027

µp 9.6 10.9 13.4 11.8 11.5 14.2 13.5 14.3 37.3

µwe 7.5 8.4 10.3 9.1 8.9 14.2 7.6 14.3 37.3

µw 3.4 3.8 4.5 4.0 3.9 7.5 6.4 5.3 37.3

∆uw , in. 0.731 0.712 0.687 0.702 0.706 1.083 0.475 0.770 1.008

Table 5.5 Lateral displacements corresponding to various performance levels in the example
building along shorter side

Wall No. 1 2 4 5 6 9 10 11 15

Failure mode Rocking Rocking Rocking Rocking Rocking Rocking Rocking Rocking Rocking

∆ O, in. 0.085 0.074 0.060 0.069 0.069 0.054 0.029 0.052 0.010

∆IO, in. 0.27 0.25 0.23 0.24 0.24 0.27 0.13 0.27 0.20

∆ CP, in. 0.69

∆LS, in. 0.41

183
Table 5.6 Base shear and corresponding lateral top displacement in the equivalent elastic
MDOF system for various performance levels of the example building along shorter side

Operational Immediate Collapse

performance level Life Safety
Occupancy Prevention
∆ , in. 0.06 0.13 0.41 0.69

(µb)∆ Elastic Range 1.01 3.28 5.47

R Elastic Range 1.01 2.36 3.15

Vbe , kip Elastic Range 32.6 76.3 101.9

∆be , in. 0.06 0.129 0.301 0.402

Table 5.7 Weight of slab supported by various structural walls in lb

Wall No. Weight due to first floor slab Weight due to second floor slab

1 1760 1870
2 680 1870
3 13680 14390
4 1310 1600
5 1510 2020
6 560 1600
7 5160 7070
8 6200 11460
9 5470 9130
10 2120 1180
11 2540 1365
12 8170 9300
13 1920 3080
14 1310 0
15 5290 0
16 7520 12420

65200 78355
∑
2025 2433
∑ , slugs
where ∑ = Total weight of floor slab at a particular floor level, and ∑ = Total
mass of the floor slab corresponding to ∑

184
 Table 5.8 Determination of parameters required to calculate Γ, mE and hE for
the example building
m, slugs h, ft φ m.φ m.φ2 h.m.φ

First floor 2025 10.75 0.5 1012.5 506.25 10884

Second floor 2433 21.75 1.0 2433 2433 52918

Table 5.9 Horizontal ground acceleration corresponding to various performance

levels of the example building along shorter side

Performance level O IO LS CP

∆be , in. 0.060 0.129 0.301 0.402

Sd -in. 0.051 0.110 0.257 0.343

Sa , ft/s2 3.800 8.100 18.900 25.200

ag/g 0.047 0.101 0.235 0.313

Table 5.10 Geometry, failure modes and displacement based characteristics of the walls
along the longer side of example building
Wall No. 3 7 8 12 13 14 16
Thickness of pier, t , ft 0.75 0.75 0.75 0.75 0.75 0.75 0.75

Length of pier lp , ft 32.00 23.50 9.00 9.00 3.50 3.50 10.00

Height of pier, hp , ft 10.50 10.50 7.00 9.00 6.00 6.00 9.00

Failure mode Sliding Sliding Sliding Sliding Rocking Rocking Rocking

Vu, lb 24.01 15.72 12.2 13.51 5.39 1.01 7.62

µp 10.3 11.0 5.7 4.9 7.9 38.4 12.5

µwe 10.3 11.0 5.7 4.9 6.2 29.1 12.5

µw 5.5 5.9 2.5 2.6 4.9 22.4 5.8

∆O , in. 0.090 0.062 0.128 0.115 0.051 0.010 0.065

∆IO ,in. 0.15 0.14 0.18 0.19 0.17 0.13 0.27

∆CP , in. 0.47
∆LS ,in. 0.305

185
 Table 5.11 Horizontal ground acceleration corresponding to various performance
levels along the longer side of example building

Performance level O IO LS CP
∆ , in. 0.065 0.140 0.305 0.470

µ∆ Elastic Range 1.31 2.85 4.39

R Elastic Range 1.27 2.17 2.79

Vbe , kip Elastic Range 100.71 171.72 220.98

∆be , in. 0.065 0.136 0.232 0.298

Sd ,in. 0.055 0.116 0.198 0.254

Sa –ft/s2 11.8 25.0 42.600 54.7

ag/g 0.147 0.311 0.529 0.680

Table 5.12 Comparison of Horizontal ground accelerations corresponding to various

performance levels of the building in both the orthogonal directions of the example building

performance level O IO LS CP

ag/g
0.047 0.101 0.235 0.313
(shorter direction)
ag/g
0.147 0.311 0.529 0.680
(longer direction)

186
Table 5.13 Values of aop/g and ∆op for various levels of compressive strains at the mid height for the
wall No. 7 in the out-of-plane direction

, , , , , ,

6.7x10-5 1.3x10-4 2.0x10-4 2.7x10- 4 4.0x10-4 6.7x10-4

fc,op lb/in2 12 23 35 47 70 117

cc, ft 0.75 0.38 0.25 0.19 0.13 0.08

φop, rad/ft 0.0001 0.0004 0.0008 0.0014 0.0032 0.0089

∆op ,ft 0.00102 0.004 0.009 0.016 0.037 0.102

aop / g 0.0629 0.125 0.143 0.150 0.141 0.126

Table 5.14 ag/g values corresponding to various performance levels in single- story buildings

Building’s designation O IO LS CP
P1-5m-ss 0.047 0.160 0.277 0.357
P3-5m-ss 0.024 0.142 0.234 0.297
P5-5m-ss 0.053 0.154 0.270 0.348
P13-5m-ss 0.032 0.139 0.343 0.463
P17-5m-ss 0.037 0.135 0.320 0.430
P19-5m-ss 0.026 0.097 0.168 0.216
P21-5m-ss 0.022 0.098 0.195 0.258
P31-5m 0.025 0.080 0.220 0.303
P32-5m 0.021 0.159 0.280 0.361
P7-10m-ss 0.041 0.137 0.242 0.312
P15-10m-ss 0.035 0.169 0.350 0.466
P19-10m 0.050 0.076 0.185 0.250
P21-10m 0.021 0.109 0.266 0.363
P23-10m 0.032 0.114 0.239 0.316
P25-10m 0.039 0.084 0.196 0.265
P29-10m-ss 0.019 0.091 0.216 0.291
P21-20m-ss 0.039 0.175 0.248 0.329
Mean 0.033 0.125 0.250 0.331
Standard Deviation 0.011 0.033 0.053 0.072
% C.O.V. 32.4 26.6 21.3 21.6

187
 Table 5.15 ag/g values corresponding to various performance levels in the double- story buildings

Building’s designation O IO LS CP
P1-5m 0.055 0.124 0.194 0.247
P3-5m 0.036 0.101 0.178 0.229
P5-5m 0.065 0.147 0.200 0.240
P13-5m 0.047 0.101 0.235 0.313
P17-5m 0.072 0.094 0.289 0.401
P19-5m 0.043 0.119 0.196 0.250
P21-5m 0.039 0.112 0.143 0.168
P7-10m 0.061 0.119 0.235 0.311
P9-10m 0.071 0.091 0.188 0.251
P13-10m 0.093 0.164 0.243 0.301
P15-10m 0.045 0.114 0.154 0.281
P25-10m 0.089 0.166 0.294 0.384
P29-10m 0.029 0.092 0.171 0.224
P21-20m 0.031 0.122 0.193 0.243
Mean 0.055 0.119 0.208 0.275
Standard Deviation 0.021 0.025 0.046 0.063
% C.O.V. 37.0 20.7 21.9 23.0

188
 Table 5.16 Average values of ground acceleration corresponding to the various performance
levels for various categories of buildings’ stock
Single story buildings Double story buildings All buildings

Performance Mean Mean Mean

Std. dev % C.O.V. Std. dev % C.O.V. Std. dev % C.O.V.
level ag/g ag/g ag/g

O 0.033 0.011 32.4 0.055 0.020 37.0 0.043 0.019 44.6

IO 0.125 0.033 26.6 0.119 0.025 20.7 0.122 0.029 24.0

LS 0.250 0.053 21.3 0.208 0.046 21.9 0.231 0.053 23.1

CP 0.331 0.072 21.6 0.275 0.063 23.0 0.305 0.073 23.8

Table 5.17 Effect of seismic zoning on the ground acceleration corresponding

to the various performance levels in the example building
ag/g
Performance level Zone 2a Zone 2b Zone 3 Zone 4

Operational (O) 0.050 0.047 0.045 0.041

Immediate Occupancy (IO) 0.102 0.101 0.099 0.096

Life safety (LS) 0.150 0.147 0.144 0.140

Collapse Prevention (CP) 0.324 0.316 0.304 0.296

189
 Chapter 6
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS

6.1 SUMMARY
This research was aimed at the seismic risk assessment of URBM buildings of Northern Pakistan
constructed in stone dust mortar. Such type of mortar is widely used, not only in Peshawar but in most
of Northern Pakistan region. To achieve this objective, the in-plane, quasi-static cyclic tests were
carried out on URBM piers constructed in stone dust mortar. Various properties of masonry required
for seismic performance evaluation of URBM buildings, such as fm, Em, Gm, µ∆, ξeq, etc, were
determined from experimental work performed as a part of this work.
Based on the results of quasi-static cyclic tests on piers, a mechanics based methodology is proposed
for lateral strength evaluation of URBM buildings. The method can be applied to any type of masonry
(brick, stone or concrete block) provided the properties of the masonry required for seismic
performance evaluation are experimentally known.
The methodology was then used to evaluate the ground acceleration corresponding to various
performance levels for 31 existing URBM buildings. The selected buildings comprised of 17 single-
story and 14 double-story buildings.
It was found by comparing the results of experimental work carried out as a part of this study with the
experimental works of other researchers [AliQ 04, AliM 06, Ban 06 and Badr 06] that the mechanical
properties of the brick masonry constructed CSK 1:4:4 mortar do not significantly differ from the
brick masonry constructed in CS 1:8 mortar. Further more, it was found that the typology of the
URBM buildings in urban areas of Northern Pakistan do not differ significantly when compared with
the 31 brick masonry buildings selected from city of Peshawar. It was, therefore, decided to extend
and apply the results of seismic risk assessment of 31 representative buildings to the other urban areas
of Northern Pakistan. Fragility curves were drawn for 17 single-story, 14 double-story, and combined
31 buildings. By comparing the results, it was found that single and double-story buildings were
almost having same performance against various levels of ground shaking. It was also found that
URBM buildings, if properly constructed, can be safely used in localities placed in zone 2b and below.
However, use of URBM in other zones was found to be unsafe.

6.2 CONCLUSIONS AND RECOMMENDATIONS

The conclusions and recommendations presented in this section are further classified in three sub-
sections, namely, experimental work, proposed methodology and seismic risk assessment of URM
building stock.

190
6.2.1 Experimental work
1. It is concluded that the Turnšek Equation [TS 80] estimates the diagonal tensile shear strength
of URM piers with reasonable accuracy, provided the principal tensile strength of masonry is
determined by the RILEM Equation [RILEM LUM B6].
2. A significant variation in the displacement ductility factors of the tested piers was found for
both direction of loading. It is, therefore, proposed that the displacement ductility factor of a
pier, µp, shall be taken as the minimum of the two values determined for each direction of
loading instead of conventional approach of averaging the µp for both direction of loading.
3. A displacement ductility factor of 6.0 is recommended for URBM piers, constructed in stone
dust mortar and failing in diagonal tension shear failure.
4. Based on the results of the tested piers, it is concluded that there exist no correlation between
displacement ductility factor and pre-compression level as indicated by a coefficient of
determination, r2 = 0.0012.
5. By comparing the results of piers of PII and PIII series, it is concluded that an increase in the
pre-compression cause reduction in coefficient of equivalent viscous damping after the piers
attains their maximum resistance.
6. By comparing the results of piers belonging to PII and PIII series, it is concluded that stiffness
degradation in piers with the increase in drift is not influenced by the magnitude of pre-
compression.
7. For performance based design/assessment of URBM piers constructed in stone dust mortar,
drift ratios of 0.08%, 0.15% and 0.35% are proposed for IO, LS and CP performance levels,
respectively. The O level for a pier can be related to the drift ratio at the formation of first
flexural tensile crack in the bed joints.

6.2.2 Proposed methodology

1. It is proposed that the O performance level in an URBM building shall be related to a lateral
displacement at the top of the building corresponding to the formation of flexural tensile
cracks in the bed joint of the 50% of the total piers in the building.
2. It is proposed that the IO performance level in an URBM shall be related to a lateral
displacement at the top of the building corresponding to the formation of first shear crack in
the building. For a pier failing in shear, it is proposed the first shear crack in the pier initiates
when the pier is displaced to the drift ratio of 0.08%. In case of a building in which no pier
undergoes shear cracking, the IO performance level is related to a distress state when a first
pier in the building is displaced to a drift ratio of 0.16%.
3. It is proposed that the LS performance level in an URBM building shall be related to the
lateral displacement at the top of the building , ∆LS, such that ∆LS = (∆IO + ∆CP)/2. where ∆IO

191
 and ∆CP are the lateral displacement at the top of the building corresponding to IO and CP
performance levels.
4. It is proposed that the CP performance level shall be related to the lateral displacement at the
top of the building at a distress state when lateral resistance of the URBM building drops by
20%.
5. In order to take into account the effect of vertical ground acceleration, it is proposed that the
pressure on the top surface of piers due to the gravity load shall be reduced by an amount
equal to 2/3Caσo.
6. By comparing the results of propsoed methodology with the experimtal results of full scale
model buildings [MKC 95, Yi 04], it is concluded that the proposed methodogoly
conservatively estimate the lateral strength and ultimate displacements of URBM buildings.

6.2.3 Seismic risk assessment of URBM building stock

1. It was found that a 9 in. thick wall at the top floor in a two story building, restrained on both
ends by rigid floor and supporting minimal gravity load will collapse in the out-of-plane
direction at a ground shaking of 0.65g. Where, the maximum value of ag in the in-plane
direction of all the double story buildings was found to be 0.401g (Building P17-5m, Table
5.15). Hence, it is concluded the double story buildings are safe in the out-of-plane direction.
2. It was found that 4.5 in. thick walls at the top floor in a three story and two story building,
restrained on both ends by rigid floor and supporting minimal gravity load will collapse in the
out-of-plane direction at ground shaking of 0.13g and 0.16g, respectively. These values are
well below the ground acceleration triggering the collapse of building in the in-plane
direction. It is, therefore, recommended not to construct 4.5 in. walls on the second (and
above) floor levels.
3. It is concluded that the seismic capacity of a brick masonry building is not significantly
influenced by the soil conditions. The seismic demand, however, increases with a decrease in
the stiffness of soil.
4. It is concluded that a horizontal ground acceleration of 0.12g, PGA corresponding to Zone 1
(no significant town in Northern Pakistan) for soil SD, will cause very light damage (damage
corresponding to O performance level) in the whole buildings’ stock. For this level of ground
shaking, there is a probability of 0.4 that the buildings’ stock will suffer minor damage
(damage corresponding to IO performance level). Such buildings can be re-occupied for use
but will need to be repaired after accommodation.
5. It is concluded that an earthquake having a horizontal ground acceleration of 0.22g, PGA
corresponding to Zone 2a (D.I. Khan) for soil SD, all the buildings will be distressed beyond
IO performance level. However, there is a probability of 0.4 that buildings’ stock will suffer

192
 damage corresponding to LS performance level. Buildings suffering such damages can only
be occupied after repair work. Similarly, there is a very low probability of 0.12 that
buildings’ stock will distress to CP performance level.
6. At a ground acceleration of 0.28g, PGA corresponding to Zone 2b (Islamabad, Mardan,
Peshawar, Swabi, Haripur, Kohat, Bannu, Tank, Miran shah)for soil SD, there is a high
probability of 0.8 that the URBM building stock will be distressed to LS performance level.
However, there is a probability of 0.4 that will buildings’ stock will suffer heavy damage (
damage corresponding to CP performance level). Buildings undergoing such damage cannot
be re-used for accommodation and need to be reconstructed.
7. At a ground acceleration of 0.36g, PGA corresponding to Zone 3 (Abbotababad, Manshera,
Mingora, Dir, Gilgit, Dasu) for soil SD, almost whole URBM buildings’ stock will suffer
damage corresponding to LS performance level. Similarly, there is a high probability of 0.75,
that the URBM buildings’ stock will distress to CP performance level.
8. At a ground acceleration of 0.44g, PGA corresponding to Zone 4 (Balakot, Bagh,
Muzaffarabad, Chitral) for soil SD, there is a probability as high as 0.92 that the URBM
buildings’ stock will distress to CP performance level.
9. It is recommended that the URBM buildings constructed with 9 in. thick wall and having a
height up to two-stories can be safely used in locations placed in zone 2b and below.
10. URBM buildings constructed with 9 in. thick wall, irrespective of the story height, are not
recommended be constructed in any location placed in zone 3 and above.
11. In order to carry out an accurate seismic risk assessment, it is, recommended to carry out a
microzonation, at least for the densely populated cities of the Northern Pakistan.
12. A compressive strength of 500 psi (approximately equal to 75% of the compressive strength
achieved in the laboratory) is proposed for the URBM buildings’ stock of Northern Pakistan
constructed either in CSK 1:4:4 or CS 1:8 mortar.
13. The modulus of elasticity of the brick masonry perpendicular to the bed joints, Em, is
recommended to be taken as 350 times the compressive strength of masonry for the URBM
buildings’ stock of Northern Pakistan constructed either in CSK 1:4:4 or CS 1:8 mortar.
14. The modulus of elasticity of the brick masonry in a direction parallel to the bed joints, Emx, is
proposed to be taken as half of the modulus of elasticity in the direction perpendicular to bed
joints. i.e., Emx = Em/2
15. The modulus of rigidity of masonry, Gm, shall be taken as 20% of Em for the URBM
buildings’ stock of Northern Pakistan constructed either in CSK 1:4:4 or CS 1:8 mortar.

193
6.3 RECOMMENDED FUTURE WORK
Future work is required for the refinement of the proposed methodology for seismic assessment of
URBM buildings. The following work is, therefore, recommended to be carried out in the future
1. The proposed methodology does not take into account the effect of torsion on the lateral
strength assessment of URM buildings. A future study is, therefore, required to incorporate
the torsional effect on the lateral strength of masonry buildings.
2. The proposed method is valid for rigid diaphragms in which the out-of-plane displacement in
the connected face loaded walls is constant throughout its length when subjected to lateral
forces. However, in the case of flexible diaphragms, the displacements in the out-of- plane
directions are amplified in the face loaded walls under the action of lateral forces. These
displacement are maximum at the centre of diaphragm and the connected walls in the out-of-
plane direction. Future work is needed to incorporate the effect of diaphragm flexibility on the
seismic capacity of URM buildings in the out-of-plane direction.
3. The drift ratios recommended for various performance levels in URBM piers were based on a
limited number of cyclic tests. Further tests may result in refining the drift limits
corresponding to various performance levels. However, it is recommended to conduct tests on
piers under the double-curvature conditions, instead of cantilever piers that fail at higher drift
levels.
4. The Turnšek equation does not take into account the affect of boundary conditions on
diagonal tensile shear strength of URM piers. It is the author’s pint of view that flexure-shear
interaction influence the diagonal tensile shear strength of URM piers. It is, therefore,
recommended that the effect of boundary condition on the diagonal tensile shear strength of
URM piers shall be investigated by carrying out a series of tests.
5. The sizes of piers were relatively small due to the limited capacity of the actuator used in the
experimental work. The piers having aspect ratios of 0.66 and 0.93 approximately represented
1:2.5 model of the full size walls. Testing of large piers, with same aspect ratios, will help in
studying the size effect on the result of piers.
6. The material properties, especially compressive strength of masonry, used to evaluate
rocking/toe crushing strength of piers were determined by testing the masonry prisms under
monotonically increasing vertical loading. It is recommended that in future experimental
works, tests shall be carried out to investigate the effect of the vertical cyclic loading on the
compressive strength and modulus of elasticity of the masonry prisms.
7. In-plane quasi-static cyclic tests shall be carried out in URBM piers constructed in locally
used cement, sand mortar as well as stone and concrete block masonry.
8. The bricks used in the experimental work were of relatively good quality. It has been
observed during the laboratory testing of bricks that compressive strength of bricks vary from

194
1000 -3000 psi in Peshawar. A series of quasi-static cyclic tests on brick masonry piers by
varying the compressive strength of bricks are recommended to study the effect of fm on
displacement based properties of piers, such as displacement ductility factors, ultimate drifts
and energy dissipation characteristics.

195
 Bibliography

[ACI 318-63] “ACI standard building code requirements for reinforced concrete”, American Concrete
Institute, Michigan, 1963

[AliM 06] Ali M.: “To study the compressive strength properties of brick masonry works used in
Pakistan”. M.Sc. thesis, NWFP University of Engineering and Technology, Peshawar, Pakistan, 2006.

[AM 04] Abrams D.P and Magenes G.: “Seismic design and assessment of masonry structures”.
Lecture notes on the course titled Seismic design and assessment of masonry structures, ROSE
school ,Pavia, 2004.

[ATC 13] Applied Technology Council “Earthquake damage evaluation data for California”. ATC-13,
Redwood City, California, 1985.

[ATC 14] Applied Technology Council “Evaluating the seismic resistance of existing buildings”.
ATC-14, Redwood City, California, 1987.

[AliQ 04] Ali Q.: “Seismic performance study of brick masonry building systems in Peshawar
region”. Ph.D. thesis, NWFP University of Engineering and Technology, Peshawar, Pakistan, 2004

[Ban 06] Banori F.A.S.: “To study the modulus of rigidity of local brick masonry system”. M.Sc.
thesis, NWFP University of Engineering and Technology, Peshawar, Pakistan, 2006

[Badr 06] Badrashi Y.I.: “Development of non-linear shear strength constitutive material model for
local brick masonry”. M.Sc. thesis, NWFP University of Engineering and Technology, Peshawar,
Pakistan, 2006.
[BCP 07] Building Code of Pakistan, Seismic Provision, SP-2007, Ministry of Housing and works,
Government of Islamic republic of Pakistan, 2007.

[BT 84] Benedetti D. and Tomazevic M.: ‘Sulla verifica sismica di costruzioni in muratura (on the
seismic assessment of masonry structures)’,Ingegneria Sismica, Vol. I, No. 0, 1984, pp. 9–16 (in
Italian).

[Ca 99] Calvi G.M.: “A displacement-based approach for vulnerability evaluation of classes of
buildings”. Journal of Earthquake Engineering, Vol. 3, No. 3, 1999.

[CG 99] Chopra A.K, Goel R.K: “Capacity-demand-diagram methods based on inelastic design
Spectrum”. Earthquake Spectra, Vol. 15, No. 4, 1999.

[Ch 01] Chopra A.K.: “ Dynamics of structures- Theory and application to earthquake engineering”
second edition, Prentice Hall Publishers, New Jersey, 2001.

[CS 92] Coburn A. and Spence R.: “Earthquake Protection”. John Wiley & Sons, Chichester, 1992.

[DHB 99] Drysdale R.G., Hamid A.A. and Baker L.R.: “Masonry structures- Behavior and design”,
The Masonry Society, Colarado,99

[DSOP 97] D’Ayala D., Spence R., Oliveira C. and Pomonis A.: “Earthquake loss estimation for
Europe’s historic town centres”. Earthquake Spectra, Vol. 13, No. 4, 1997.

196
[EMS 98] European Macroseismic Scale 1998, Centre Européen de Géodynamique et de
Séismologie, Luxembourg, 1998.

[FEMA 154] “Rapid visual screening of buildings for potential seismic hazards: A handbook”.
FEMA 154, Washington, 1988.

[FEMA 155] “Rapid visual screening of buildings for potential seismic hazards: supporting
documentation”. FEMA 155, Washington, 1988.

[FEMA 178] NEHRP “Guidelines for the seismic rehabilitation of buildings”. FEMA 178,
Washington, 1992.

[FEMA 310] “Handbook for the seismic evaluation of existing buidings – a prestandard”.
ASCE/FEMA 310, Washington, 1998.

[FEMA 356] “Prestandard and commentary for the seismic rehabilitation of buildings”.
ASCE/FEMA 272(356), Washington, 2000.

[FKLG 01] Fäh D., Kind F., Lang K. and Giardini D.: “Earthquake scenarios for the city of Basel”.
Soil Dynamics and Earthquake Engineering, Vol. 21, pp. 405- 413, 2001.

[FPCB 99] Faccioli E., Pessina V., Calvi G.M. and Borzi B.: “A study on damage scenarios for
residential buildings in Catania city”. Journal of Seismology, Vol. 3, No. 3, 1999.

[Ga 89] Ganz H.R.: “Failure criteria for masonry”, proceedings of 5th Canadian Masonry
Symposium, Vancouver, Canada, 1989.

[HD 81] Hamid A.A., and Drysdale R.G., "Proposed failure criteria for concrete block masonry
under biaxial stresses," ASCE Journal of Structural Engineering, Vol. 107, No. 8, pp. 1675-1687,
1981.

[JKM 08] Javed M., Khan A.N. and Magenes G.: “Performance of masonry structures during
earthquake -2005 in Kashmir”. Mehran University Research Journal of Engineering & Technology,
Vol. 27, No. 3, pp. 271-282, 2008

[Ke 72] Keightley W.O.: “Report on Indo-U.S. Sub commission o Education and culture”,
Department of Earthquake Engineering, University of Roorkee, 1972.

[KNKH 97] Kircher C.A., Nassar A.A., Kustu O. and Homes W.T.: “Development of building
damage functions for earthquake loss estimation”. Earthquake Spectra, Vol. 13, No. 4, 1997.

[La 02] Lang K.: “Seismic vulnerability of existing buildings”. PhD thesis, Institute of Structural
Engineering Swiss Federal Institute of Technology, Zurich, Switzerland, 2002.

[LP 94] Limogelli, M.P., and Pezzolli, P.: “Analysis of seismic performance of masonry buildings
excited by a shaking table”. European Earthquake Engineering, Vol. 7, No.2, pp 18-30, 1994

[Mag 08] Recently imposed limits, personally communicated by Magenes.G. Associate professor,
University of Pavia,Italy

[MB 94] Miranda E., Bertero V.V.: “Evaluation of strength reduction factors for earthquake resistant
design”. Earthquake Spectra, Vol. 10, No. 2, 1994.

197
[MC 97] Magenes G. and Calvi G. M.: “In-plane seismic response of brick masonry walls”.
Earthquake Engineering and Structural Dynamics, Vol. 26, pp.1091-1112, 1997.

[MKC 95] Magenes G., Kingsley G. R. and Calvi G. M.: “Seismic testing of a full-scale, two-story
masonry building: test procedure and measured experimental response”. Università degli Studi di
Pavia, 1995.

[MM 82] Mann W. and Muller H. “Failure of shear –stressed masonry- an enlarged theory , tests and
application to shear walls”. Proceedings of British ceramic society, Vol. 30, pp. 223-235, 1982.

[MS 89] Mahin S.A., and Shing P.B.: “Pseudo-dynamic test method-Current status and future
directions”, ASCE Journal of Structural Engineering, Vol. 115, No. 8, pp. 2113-2128, 1989

[Pa 82] Page A.W.: "An experimental investigation of the biaxial strength of brick masonry"
Proceeding of 6th International Brick Masonry Conference, Rome, Italy, 1982.

[Pr 85] Priestley M.J.N.: “Seismic behavior of unreinforced masonry walls”, Bulletin of the New
Zealand National Society for Earthquake Engineering, Vol.18, No.2, 1985

[PA 90] Paulson T.J. and Abrams D.P.: “Correlation between static and dynamic response of model
masonry buildings”, Earthquake Spectra, Vol. 6, No. 3, pp573-591, 1990.

[PB 98] Popov.E.P., and Balan T.A..: “Engineering mechanics of solids”. Second edition, Prentice
Hall publishers, New Jersey, 1998

[PCM 00] Pujades L.G., Canas J.A., Mena U.and Espinoza F., Alfaro A., Caselles J.: “Seismic risk
evaluation in Barcelona, Spain”. Proceedings of the Twelfth World Conference on Earthquake
Engineering, Auckland, New Zealand, 2000.

[PP 92] Paulay T. and Priestley M.J.N.: “Seismic design of reinforced concrete and masonry
buildings”. John Wiley & Sons, New York, 1992.

[PS 89] Porro B. and Schraft A.: “Investigation of insured earthquake damage”. Natural Hazard, Vol
2, pp. 173-184, 1989.

[RILEM LUMB6] RILEM. LUMB6 “Diagonal tensile strength tests of small wall specimens”. Tech
Rep, RILEM, 1994.

[SD 99] Spence R. and D’Ayala D.: “Damage assessment and analysis of the 1997 umbria-marche
earthquakes”. Structural Engineering International, Vol. 9, No. 3, 1999.

[SM 85] Shing P. B. and Mahin S.A.: “Computational aspects of a seismic performance test method
using on-line computer control” Earthquake Engineering and Structural Dynamics, Vol 13, pp. 507-
526, 1985

[Ta 75] Takanashi K. “Nonlinear earthquake response analysis of structures by a computer-actuator

on-line system”, Bulletin of Earthquake Resistant Structure Research Center, No. 8, University of
Tokyo, 1975.

[TBW 04] Tomaževic, M., Bosiljkov, V. and Weiss, P.: “Structural Behavior Factor For Masonry
Structures”, Proceedings of the13th World Conference on Earthquake Engineering,
Vancouver, Canada, 2004.
198
[TK 97] Tomaževic, M. and Klemenc, I.: “Verification of seismic resistance of confined
masonry buildings”. Journal of the Earthquake Engineering and Structural Dynamics, Vol.
26, No.10, pp. 1073-1088, 1997

[TN 87] Takanashi K. and Nakashima M. “Japanese activities on on-line testing”, ASCE Journal of
Engineering mechanics, Vol. 113, No. 7, pp. 1014-1031, 1987.

[To 99] Tomaževic, M.: “Earthquake-resistant design of masonry buildings”, Imperial College
Press, London, 1999.

[TS 80] Turnšek V. and Sheppard P.: “The shear and flexural resistance of masonry walls”.
Proceedings of International Research Conference on Earthquake Engineering, Skopje , 1980.

[TW 94] Tomaževic, M. and Weiss P: “Seismic behavior of plain and reinforced masonry
buildings”. ASCE Journal of Structural Engineering, Vol. 120, No.2, pp. 323-338, 1994.

[Va 04] Vasconcelos G.F.M.: “Experimental investigations on the mechanics of stone masonry:
Characterization of granites and behavior of ancient masonry shear walls”, PhD thesis, Universidade
do Minho, Minho, Portugal , 2004

[VFF 94] Vidic T., Fajfar P., Fischinger M.: “Consistent inelastic design spectra: Strength and
displacement”. Earthquake Engineering and Structural Dynamics, Vol. 23, pp. 507-521, 1994.

[VN 60] Veletsos A.S., Newmark N.M.: “Effect of inelastic behaviour on the response of simple
systems to earthquake motions”. Proceedings of the Second World Conference on Earthquake
Engineering, Vol. 2, pp. 895-912, Tokyo, Japan, 1960.

[VNC 65] Veletsos A.S., Newmark N.M., Chelapati C.V.: “Deformation spectra for elastic and
elastoplastic systems subjected to gound shock and earthquake motions”. Proceedings of the Third
World Conference on Earthquake Engineering, Vol. 2, pp. 663-680, Wellington, New Zealand, 1965.

[WRH 74] Whitman R.V., Reed J.W. and Hong S.T.: “Earthquake damage probability matrices”.
Proceedings of the fifth World Conference on Earthquake Engineering, pp. 2531, Rome, 1974.

[Yi 04] Yi T.: “Experimental investigation and numerical simulation of an unreinforced masonry
structure with flexible diaphragms”. PhD thesis, Georgia Institute of Technology , Georgia, U.S.A,
2004.

199
 VITA AUCTORIS

Engr. Mohammad Javed was born in 1969 in Mardan, NWFP, Pakistan. In 1987, he completed his
college education at Nisar Shaheed College, Risalpur. He pursued his studies in Civil Engineering
from 1988 to 1992, and, received his B.Sc. (Civil Engineering) degree from NWFP University of
Engineering and Technology, Peshawar. In 1996, he completed his M.Sc. in structural engineering
from the same department. He commenced his doctoral studies from the same department under the
Higher Education Commission Indigenous PhD 300 scholarship scheme in 2003.

200

View publication stats