Performance Behavior of Confined Brick Masonry …prr.hec.gov.pk/jspui/bitstream/123456789/550/1/358S.pdfHorizontal bed joint reinforcement is recommended to connect the tie columns
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Performance Behavior of Confined Brick Masonry Buildings under Seismic Demand
By
AMJAD NASEER
DISSERTATION Submitted in partial fulfillment of the requirements for the degree of Doctor of
performed generally well. However, some failures were observed due to inadequate wall
strength, poor quality of construction, inadequate number and arrangement of confining
element (figure 2.13) (EERI, 1999 & 2006).
Figure 2.13 Damage confined masonry building (1999 Tehuacan earthquake) due to non-
confinement in window (EERI, 1999)
On January 13, 2001, an earthquake of magnitude Mw=7.7 occurred in the Republic of El
Salvador in Central America. The earthquake resulted in 844 people dead and 75,000
buildings completely damaged. The use of ‘confined masonry’ walls is more popular for the
construction of residential buildings. In this system, cast-in-place, slender R/C columns are
presented at most of the extreme edges and intersections of the walls. In addition, cast-in-
place R/C collar beams are also provided. Buildings up to three and more stories are used.
The large number of existing confined masonry buildings were not severely damaged during
the earthquake, except in few cases of shear failure and separation of block masonry wall and
vertical confining column (Yoshimura, K., and Kuroki, M., 2001). It was reported by
25
Ascheim et al (2006) that ‘mixto’ confined masonry construction was used in the post-
earthquake rehabilitation following the 2001 earthquakes.
It was observed during March 31, 1983 Popayan, Colombia earthquake that adding
horizontal beams to unreinforced masonry reduced the seismic damage, while the addition of
tie column eliminate damage (Schultz, A. 1994). In the 1999 El Quindio, Colombia
earthquake (Mw 6.2), shear cracks and out of plane failure because of the inadequate
connection between wall and confining elements was observed (EERI, 2000).
The 1985 Llolleo, Chile earthquake (Mw = 7.8) caused the collapse of 66,000 buildings and
damaged upto 127,000. Out of 84,000 housing units, 13,500 were of confined masonry. The
buildings were from 3 to 5 storeys in height (Sventlana Brzev 2007). Confined masonry in
Chile has generally performed well during earthquakes; however, elimination of tie column
for economic purposes lead to the extensive inclined cracking and horizontal cracks between
floor slabs and masonry wall were observed during the 1985 Chile earthquake (Schultz, A.
1994). In the figure 2.14 below complete wall panel collapsed, after first developing diagonal
shear cracks. The tie columns can also be seen damaged.
Figure 2.14 Damaged confined masonry during 1985 Llolleo, Chile earthquake (Moroni et
al., 2003)
Damage to short residential confined masonry was observed due to structural configuration
and omission of tie members during 1990 Pomasqui, Ecuador earthquake (Schultz, A. 1994).
26
2.5 CODES /GUIDELINES RECOMMENDATIONS
The use of confined masonry has been practiced in Latin America, Asia and Europe. In most
of these countries the specification for confined masonry is part of their code or country
guidelines. The specifications are developed after experiences during the past earthquakes
and extensive experimentation. It has been observed that the confined masonry improves
both ductility and seismic resistance of the structure. In the subsequent sections, the
specifications of different codes and guidelines are discussed.
2.5.1 Specifications of Eurocode 6 & 8
According to the requirements of Eurocodes, no contribution of vertical confinement to
vertical and lateral resistance of the structure should be taken into account in the calculation
(C1.4.9, EC 6). Eurocode 6: Design of masonry structures gives some basic rules for the
confined masonry as discussed below; however, some additional requirements have been
specified in Eurocode 8: Design provisions for earthquake resistance of structures.
2.5.1.1 Construction Technique
According to Eurocode reinforced concrete or reinforced masonry vertical (tie column) and
horizontal (bond beam) confining element should be provided to the masonry wall so that
they act together during lateral action. Concrete for confining elements should be cast after
the construction of masonry wall. The confining elements should be provided at the
following locations:
• At all free edges of the structural walls,
• At the walls intersection,
• Tie columns should be placed at a maximum spacing of 13 ft (4 m) and,
• At both sides of opening having an area of more than 16 sft (1.5 m2)
• Tie beams should be provided at every floor level and at a vertical spacing of 4 m.
27
2.5.1.2 Geometric requirements in the confining masonry and area of reinforcement
The confining element should have cross-sectional area of 31 in2 (0.02 m2), with minimum
dimension of 6 inch (150 mm) in plan of the wall. In the case of double leaf wall, the
confining element should be equal to the wall thickness. The minimum wall thickness is the
one which give robust wall or determined from calculations.
The longitudinal reinforcement should be 0.8 % of the cross-sectional area of the confining
element and should not be less than 0.31 in2 (200 mm2). The stirrup should not be less than #
2 (6 mm) bar and should be provided at least at 12 inch (300 mm).
Horizontal reinforcement not less than # 2 (6 mm) diameter bars or equivalent and spaced at
12 inch (300 mm) should be provided and anchored in concrete column and in the mortar
joints.
According to additional requirements in Eurocode 8 (section 9.5.3), the minimum area of
reinforcement is 0.5 in2 (300 mm2) or 1% of the cross-sectional area of the confining
element. The stirrup should be provided at 6 inch (150 mm) center to center. The bar
diameter mentioned in Eurocode 8 is 5 mm which is slightly less #2 bar. The bars should be
spliced at length of 60 times diameter of bar.
The minimum thickness of the wall should be 9.5 inch (240 mm). The minimum effective
height to thickness ratio of the wall should be 15 and length of wall to clear height of the
opening (adjacent to the wall) should be 0.3.
2.5.1.3 Material Strength
The minimum compressive strength of masonry mortar of 730 psi (5 MPa) should be used
for confined masonry buildings.
2.5.1.4 Number of stories and wall density ratio
Eurocode 8 recommends minimum number of stories depending on the seismicity of the area
and wall density ratio. Table 2.7 gives the number of stories corresponding to minimum wall
density ratio and the maximum ground acceleration. In the table, k is a corrective factor
based on minimum unit strength of 725 psi (5 MPa) for confined masonry. Where k = 1+(lav-
28
2)/4≤2 for buildings having 70% of the shear walls under consideration are longer than 6.5 ft
(2 m), however, for all other cases k = 1. In the expression lav is average wall length.
Table 2.7 Recommended allowable number of stories above ground and minimum area of
shear walls for simple buildings (EC 8)
Acceleration at
site ≤ 0.07 k.g ≤ 0.10 k.g ≤ 0.15 k.g ≤ 0.20 k.g
No of Stories Minimum sum of cross-sections areas of horizontal shear walls in each direction,
as percentage of the total floor area per storey
2 2.0 % 2.5 % 3.0 % 3.5 %
3 2.0 % 3.0 % 4.0 % N/A
4 4.0 % 5.0 % N/A N/A
5 6.0 % N/A N/A N/A
2.5.2 Additional Recommendations
However, detailed and through discussion on the earthquake resistance of masonry buildings
has been given, some additional recommendations regarding confined masonry have been
provided in addition to Eurocode specifications by (Tomazevic, M., 99). The information has
been based on actual observation of masonry during earthquake and experimental research
work carried out by the author.
Because of the lack of experimental work on the behavior of confined masonry, the amount
of reinforcement in confining element is determined on empirical basis. Table 2.8 gives
number of plain bars depending on the number of stories and seismic demand.
29
Table 2.8 Typical reinforcement of vertical confining element
No of Stories
Seismic Demand
Low
(PGA = 0.1 g)
Moderate
(0.1 g < PGA < 0.2 g)
High
(0.2 g < PGA < 0.4 g)
2 1-2 4-8 mm bars 4-10 mm bars (4#3) 4-12 mm bars (4#4)
4 1-2 4-8 mm bars 4-10 mm bars (4#3) 4-12 mm bars (4#4)
4 2-4 4-8 mm bars 4-10 mm bars (4#3) 4-12 mm bars (4#4)
6 1-2 4-10 mm bars (4#3) 4-12 mm bars (4#4) 4-14 mm bars (4#4)
6 3-4 4-8 mm bars 4-10 mm bars (4#3) 4-12 mm bars (4#4)
6 5-6 4-8 mm bars 4-10 mm bars (4#3) 4-12 mm bars (4#4)
In order to fully utilize the resistance and energy dissipation capacity of masonry, the
masonry walls should be horizontally reinforced. Specially shaped ladder-type or truss type
reinforcement should be used in horizontal mortar joint provided at vertical spacing of 24
inch (600 mm). The horizontal reinforcement should be anchored in the tie column. In such
case the EC 8 requirements for connecting tie column and masonry wall should be waived
(Tomazevic, M., 99).
The seismic resistance of the masonry buildings should be verified, however, the number of
stories and height has been recommended in table 2.9 on the basis of masonry materials, wall
density ratio and configuration of building.
30
Table 2.9 Recommended maximum building height and number of stories (n)
Design ground acceleration < 0.2 g 0.2-0.3 g ≥ 0.3 g
Unreinforced
Masonry
Height, ft (m) 39.4 (12.0) 29.5 (9.0) 20.0 (6.0)
No of stories, n 4 3 2
Confined
Masonry
Height, ft (m) 59.0 (18.0) 49.0 (15.0) 39.5 (12.0)
No of stories, n 6 5 4
Reinforced
Masonry
Height, ft (m) 79.0 (24.0) 69.0 (21.0) 59.0 (18.0)
No of stories, n 8 7 6
Concrete of minimum compressive strength of 2,200 psi/15.2 MPa (C15) is recommended
for confined masonry elements.
It is also emphasized that the configuration of the building should be simple and regular. That
is the load bearing wall should be symmetrically distributed and should not change their
position and shape along the height of the building. A detailed description is presented in
(Tomazevic 1999) about the configuration or architectural requirements.
2.5.3 Confined Masonry Guidelines
Different confined masonry guidelines are available including City University guidelines
(Virdi and Raskkoff), Catholic University of Peru (PUCP and SENCICO, 2005), the
International Association of Earthquake Engineers (IAEE) (IAEE and NICEE, 2004), and
UNESCO (Ghaidan, 2002). The guidelines are in agreement on many points. However, there
are a few points where the guidelines have some differences. Some of the guidelines have
detailed approach and discuss comprehensively all the points, while the others are very
specific. A detailed discussion on the different guidelines and item by item comparison are
given in (UBC EERI Student Chapter, 2008 report). It has been tried to evaluate the
guidelines in the light of current research on confined masonry. It was concluded that Peru
guidelines could be used. However, recommendations were provided for the improvements
in the Peru guidelines.
31
Earthquake resistant confined masonry reports (Svetlana Brzev, 2007) discusses the different
requirements of confined masonry building construction. The report discusses different
factors which affect the seismic resistance of confined masonry. Confined masonry
performance during earthquakes is presented. Chile: NCh2123.97, Mexico: Mexico City
Building Code (NTC-M 2004), Eurocode 6 & 8 and Iranian code of practice for seismic
resistant design of building (Standard 28000) are summarized. Finally, the architectural and
construction guidelines are presented. It has been concluded that confined masonry generally
performed well; however, good quality materials and simple architectural design should be
used.
2.6 THEORY OF STRUCTURAL MODEL
A structural model is defined as any structural element or assembly of structural elements
built to reduced scale which is to be tested, and for which laws of similitude must be
employed to interpret test results (Harris.G and Sabnis 1999). Any structural model must be
fabricated, loaded, tested and results interpreted according the laws of similitude
requirements. Dimensional analysis is used to determine the similitude requirements of
physical quantities of model and prototype structures. The following relationship should be
satisfied for the different physical quantities involved in the structural modeling:
p m FQ Q x S= (2.3)
Where Qp is any physical quantity required, Qm is the physical quantity measured on the
model and SF is the scale factor corresponding to the physical quantity. The different types
of models used in the experimental analysis of masonry buildings depend on the materials
used for the fabrication of the models (Tomazevic. M and Velechovshky. T 1992)
1. In the case of complete model similarity special model materials are used for the
manufacturing of model. In such cases stresses are scaled to the geometric scale.
However, strain remains the same in the prototype and model (figure 2.15 (a)).
Specific weight, poisson’s ratio and damping are also same for model and prototype.
32
2. When prototype materials are used for the construction of model, the similarity is
called simple model similarity. In such case stress and strain are similar in both
prototype and model as shown in figure 2.15 (b).
Figure 2.15 Stress-strain relationship of model and prototype materials in the case of
complete model similarity
The physical quantities involved in the dynamic testing of masonry structure with
theoretically obtained scale factor fulfilling simple and complete model similitude
requirements are given in table 2.10.
StressStress
Strain
33
Table 2.10 Similitude requirements in dynamic testing
Quantity General equation Scale Factor
Complete model Simple model
Length (L) SL = Lp/Lm SL SL
Strain (ε) Sε = εp/εm 1 1
Strength (f) Sf = fp/fm SL 1
Stress (σ) Sσ = fp/fm SL 1
Young's modulus (E) SE = Sσ/Sε SL 1
Sp. Weight Sr = rp/rm 1 1
Force (F) SF = SL2Sf SL
3 SL2
Time (t) St = SL√(SrSε/Sf) √SL SL
Frequency (Ω) SΩ = 1/St 1/√SL 1/SL
Displacement (d) Sd = SLSε SL SL
Velocity (ν) Sv = Sε√SfSr √SL 1
Acceleration (a) Sa = Sf/SLSr 1 1/SL
The scale factor of 7 is upper limit for complete model in modeling brick or block masonry
and 4 for stone-masonry buildings (Tomazevic 1992). It is important in the dynamic testing
of model that dynamic behavior as well as failure mechanism is correctly simulated. Similar
distribution of masses and stiffness is required in the prototype and model building for
fulfilling similarity of dynamic behavior. And similar working stress level in the load bearing
walls of the prototype and model should be achieved for the similarity in failure mechanism.
Both requirements of dynamic behavior and working stress are fulfilled in following
complete model similitude requirements. However, special arrangements are required for the
mass distribution and working stress level, in case of simple model requirements.
2.7 EXPERIMENTAL TESTING
In this section experimental testing on shake table and cyclic testing of confined masonry
walls are briefly presented.
34
2.7.1 Shaking Table Test on 24 Simple Masonry Building (D.Benedetti, 1998)
A total of 14 shaking table tests were carried out in the ISMES (Seriate, BG, Italy) and LEE
(Laboratory for Earthquake Engineering, NTUA, Athens, Greece). The tests were carried out
on 1:2 scaled models. Actual prototypes materials were used in the model buildings. The
acceleration and stress scales were, respectively, Sa = 1 and Sσ = 1. As same materials were
used for models and real buildings, additional masses were applied to the model to respect
the scale.
Bricks and stones were used for the construction of model buildings. Intentially, low strength
mortar and low quality workmanship were used to simulate the field situation. The models
were subjected to base excitation in three translation directions. The vertical component was
70 % of the horizontal component. After damaging the models, different techniques were
used to repair and strengthen it.
It was concluded on the basis of analysis of the test results that lateral resistance increases for
both brick and stone masonry model after strengthening. Total collapse could be avoided by
using horizontal ties.
The values of the reduction factor ‘q’ are in general slightly higher for stone masonry than
for brick masonry. ‘q’ value for both the cases is 1.5 times higher than the code
recommended values. Strengthening increased such values in some instances by a factor up
to 1.8.
2.7.2 Verification of Seismic Resistance of Confined Masonry Buildings (Tomzevic, M., 1997)
Two models of a typical three story confined masonry buildings have been tested on shaking
table at Slovenian National Building and Civil Engineering Institute (ZAG) in Ljubljana,
Slovenia. The models were scaled at 1:5, conforming to the requirements of EC-8 for simple
buildings in plan. They were subjected to a series of simulated seismic ground motions with
increased intensity of shaking.
The structural system consisted of perimetral and internal walls in both orthogonal directions
dividing the plan into four rectangles. All the perimeteral walls and an internal wall in shorter
35
direction were pierced with window and door opening. Wall to floor area was 5% in both
orthogonal directions.
Diagonal cracks were observed in all stories at maximum resistance state. Crushing of
concrete and rupture of reinforcement of tie-column was also observed at the ultimate state.
It could be concluded from the tests results that prototype buildings would withstand, with
moderate damage to all walls, strong earthquakes with peak ground acceleration of 0.8g and
withstand without collapse a sever earthquake of PGA more than 1.3g.
The value of behavior factors ‘q’ was 2.91 and 2.47 for M1 and M2 models, respectively. It
could be concluded that confined masonry buildings possess more energy dissipation
capacity than code proposed values; however, taking into account the drift limitation of story
the codal values seem reasonable. Comparing ‘q’ values from previous research on un-
reinforced and reinforced masonry buildings of similar configuration and size, EC8
underestimate their energy dissipation capacity. However, further experimental and
analytical research is needed on response modification factor.
2.7.3 Shaking Table Tests of Small-Scale Model of Masonry Building: Advantages and Disadvantages (Tomzevic, M., 2000)
In this article, experiences of reduced scale shaking table test in context of advantages and
disadvantages have been discussed. It is emphasized that significant development has been
made in the numerical techniques and numerical models; however, experimental testing of
structures and sub-assemblages is needed. Seismic behavior of buildings could be determined
by cyclic test of walls, variability in the masonry need to test complete structural system.
Similitude in the dynamic behavior as well as failure mechanism of prototype and model is
considered important factor in modeling. Distribution of masses and stiffness in the prototype
and models need to be simulated. However, the failure mechanism requires similar working
stress level that is, working stresses in load bearing walls and compressive strength of
masonry of prototype and model. Although all the structural details are not precisely
modeled, the global seismic behavior of prototype building could be accurately simulated if
the behavior of model wallets is similar to prototype. Mechanical properties such as
36
compressive strength, tensile strength, ductility and energy dissipation, are simulated in
testing walls.
It is concluded that reliable information as regards the global seismic behavior and failure
mechanism can be obtained by testing small scale models of buildings on simple earthquake
simulator, although neither the physical models nor the seismic ground motion are modeled
in great detail. Special model materials for complete or prototype materials for simple model
can be used for the construction of the models. Scale factor 7 and 4 has been found as the
practical upper limit for modeling brick and block masonry, and stone masonry respectively.
In the case of complete model the requirements of dynamic behavior and failure mechanism
is automatically fulfilled. However, special arrangement is need in the case of simple models
where prototype materials are used.
2.7.4 Seismic Behavior of a Three-Story Half Scale Confined Masonry Structure (Bartolome, A.S., et al., 1992)
A three story confined masonry building modeled at 1:2.5 scale has been tested on shake
table. The vibration properties, strength of materials and axial stress of the model were kept
similar to those of the actual buildings. May 31, 1970 earthquake record was used for the
dynamic test. Each dynamic test run was preceded by free vibration test, consisting of four
pulses of small amplitude.
Flexural crack at the wall base was first developed at 0.52g. It resulted in the vertical
reinforcement to yield. At ultimate test run (0.85g) shear failure in both first story walls
occurred. The horizontal reinforcement ruptured during the last test, showing that it
effectively worked under dynamic conditions. The time period and coefficient of damping
obtained by free vibration were smaller than the pulse test and the analytically obtained
values. Both time period and damping increased with the increased in intensity of excitation.
Damping was 4% at 0.52g and increased to 7% at the final test run when PGA was 0.85g.
Stiffness of the walls decreased in the subsequent test run.
It was concluded that shear failure could occur during strong excitation even the structure
was predicted to be failed in flexure. Therefore, the design process of a confined masonry
building should include the possibility of a shear type of failure to avoid structural collapse.
37
2.7.5 Seismic Response Pattern for URM Buildings (Abram 2000)
Two reduced scale, unreinforced masonry buildings (URM) were tested on shake table in
Newmark Laboratory, University of Illinois at Urbana Champaign, USA to highlight selected
aspects of dynamic response that help confirm or deny present engineering practices for
seismic evaluation of URM buildings. The models were constructed of clay masonry units on
three-eights scale. Type O mortar was used in a two-wythe, running bond pattern. For the
first test structure perforations in each of the two parallel shear walls were chosen so that
lateral stiffness and strength of the two wall elements were similar. For the second test, the
size and placement of perforations were varied to result in dissimilar stiffness and strengths
for two walls.
Each test structure was subjected to scaled motions measured during the 1985 Nahanni
earthquake. The time scale of the recorded earthquake motion was compressed by a factor of
1.6, which was equal to the square root of the length scale of 2.5.
It is concluded that substantial nonlinear behavior, which was largely attributed to rocking
behavior of the first-story piers was observed. The models remain stable at first story drift as
large as 0.9%. Waveforms of base shear were in phase with deflection histories suggesting a
predominant first mode response. The elastic base shear, determined from measured base
motions and measured mode shapes, was as much as 4.3 times the measured base shear
maxima and as much as 7.7 times the estimated story shear rocking capacity. Measured
lateral drifts were significantly less than anticipated elastic displacements.
2.7.6 Experimental Study on Earthquake Resistant Design of Confined Masonry Structures (Ishibashi, K., et al., 1992)
This paper presents results obtained from testing of three full scale confined masonry
specimens. Each model consisted of two wall units made with clay bricks. The first
specimen, practically lacked the flexural coupling, walls were only connected through high
strength Dywidag bars that transferred the lateral force between the walls. In the second and
third specimens, walls were linked by a cast in place reinforced concrete tie beams and slab.
Parapet wall is also provided in the third specimen. Figure 2.16 illustrates the three wall
specimens and reinforcement scheme in the confining elements. The specimens were tested
38
first applying load cycles with maximum shears equal to 5 ton, 10 ton and 18 ton which
caused first crack in the masonry panel. Then displacement cycles were applied up to 0.012
drift. The specimens were designed and constructed following the requirements of the
Mexico City Building Code.
Figure 2.16 Cyclic load test specimens and reinforcement details
It has been concluded on the basis of previous research and on this research program that
masonry strength depends on the strength of brick units, and is less depended on the mortar
characteristics. The vertical load increases the shear capacity and stiffness. However, large
vertical forces reduce the available ductility of the structures. The masonry wall confinement
improves the energy dissipation characteristics and deformation capacity. The form of the
opening clearly affected the final crack pattern. However, the mode of failure for all
specimens was governed by shear deformations in the masonry panels. The type of opening
affected the initial stiffness of the specimens, the stiffness decay was similar and follow a
parabolic curve.
39
2.7.7 Cyclic Loading Tests of Confined Masonry Wall Elements for Structural Design Development of Apartment Houses in the Third World (Hiroto Kato et al 1992)
A static cyclic test on half scale confined masonry walls of three-four story apartment house,
were carried out. However, effects of axial and shear reinforcement on the load carrying
capacity, drift ratio and damage pattern were studied in this paper, the overall objectives of
the research project were to analyze damage patterns of masonry structures, to examine
improvement methods for minimizing earthquake damage, and to prepare guidelines for
confined masonry. Several static cyclic loading and shaking table tests were planned to
clarify basic characteristics of confined masonry structure.
Four confined masonry walls were constructed with different axial and shear ratio. Specimen
A with rich axial and shear reinforcement, specimen B with rich axial and poor shear
reinforcement, specimen C with poor axial and rich shear reinforcement and specimen D is
of poor axial and shear reinforcement.
It was concluded that axial reinforcement in column can improve load carrying capacity of
the confined masonry walls. The ductility of walls can be improved with the increase in shear
reinforcement as it effectively bind the unreinforced walls and avoid brittle failure. The
subjects in seismic design are to find better combinations of the ratio of axial reinforcement
and that of shear reinforcement.
40
CHAPTER 3 EXPERIMENTAL PROGRAM: MASONRY MATERIALS AND
MASONRY ASSEMBLAGE
3.1 INTRODUCTION
In order to evaluate the behavior of reduced scale typical buildings under seismic demand,
simulation of masonry materials and assemblage is essential. Extensive experimental work
has been carried out in this study for the simulation of masonry materials and masonry
assemblage. Basically, the experimental work of this study has been divided into three
phases. In the first phase mechanical properties of prototype masonry materials and
assemblages have been determined. In the second phase, simulation study has been done for
masonry materials and assemblages. The model masonry assemblages have been tested in
compression, diagonal compression and constant vertical compression and lateral cycle load.
And in the last phase of experimental study, reduced scale models have been fabricated and
tested on unidirectional earthquake simulator (shake table). The first two phases are the part
of this chapter, while the third phase has been discussed in Chapter 4 and 5. This chapter is
organized into three main sections.
Section 3.2. In this section the mechanical properties of prototype masonry materials and
masonry assemblage have been discussed. The following tests were carried out on the
masonry units and assemblage.
Masonry Unit:
I. Compressive strength II. Water absorption and density
Masonry Assemblage:
I. Compressive strength test and Modulus of Elasticity II. Diagonal Tensile (shear) Strength Test and Modulus of Rigidity
III. Cyclic Test
Section 3.3. The mechanical properties of model masonry materials and masonry assemblage
are presented in this section as follows:
41
Model Masonry Unit:
I. Compressive Strength and Density
Model Masonry Assemblage:
I. Compressive Strength and Modulus of Elasticity, II. Diagonal Tensile (shear) Strength Test and Modulus of Rigidity
III. Cyclic Test
Section 3.4. In the third and last section of this chapter, the mechanical properties of
prototype and model are compared and discussed. The actual scale factor from this study is
also compared with the true scale factor as per complete model similitude requirements.
3.2 PROTOTYPE MASONRY TEST
3.2.1 Masonry Unit (Solid Burnt Clay Brick)
Masonry unit samples were randomly collected from the local brick kiln. As described above
compressive strength and water absorption tests have been conducted on the masonry unit.
Although water absorption is not required for simulation purposes, the test has been carried
out generally to determine the physical properties of locally available brick.
3.2.1.1 Compressive Strength of Unit
Compressive strength of brick masonry units (figure 3.1) were carried out in the 200 tonne
Universal Testing Machine (UTM-200) in accordance with ASTM C 67. According to the
standard, length of test specimen for compressive strength should be equal to one half the full
length of the unit ± 1 inch. However, in this study compressive strength of full unit was
determined. Load was applied perpendicular to the bed face (length x width). The brick was
capped on both sides with gypsum 24 hours prior to testing. The compressive strength of
masonry unit is given in Table 3.1
42
Figure 3.1 Compressive strength test of solid brick masonry unit
Table 3.1 Compressive strength of masonry unit (solid burnt clay brick)
S. No Length
in (mm)
Width
in (mm)
Height
in (mm)
Crushing
Load (t)
Compressive Strength
psi (MPa)
1 8.5 (216)
4.19 (106)
2.75 (70) 27.90 1728.0
(11.9)
2 8.5 (216)
4.25 (108)
2.85 (72) 31.10 1897.4
(13.1)
3 8.63 (219)
4.25 (108)
2.9 (74) 46.10 2771.8
(19.1)
4 8.44 (214)
4.13 (105)
2.9 (74) 42.20 2672.3
(18.4)
5 8.5 (216)
4.13 (105)
2.7 (69) 39.90 2508.0
(17.3)
6 8.5 (216)
4.19 (106)
2.75 (70) 46.0 2848.0
(19.6)
7 8.44 (214)
4.25 (108)
2.75 (70) 53.80 3307.0
(22.8)
8 8.69 (221)
4.25 (108)
2.75 (70) 25.00 1492.0
(10.3)
9 8.56 (218)
4.2 (107)
2.8 (71) 39.40 2415.0
(16.7)
10 8.5 (216)
4.15 (105)
2.75 (70) 27.90 1743.0
(12.0)
Average 8.53 (216)
4.2 (107)
2.79 (71) 37.93 2338.0
(16.0) Cov (%) 1% 1% 3% 25% 25%
According to Eurocode-8 (EC 8), the minimum acceptable normalized compressive strength
of a masonry unit, normal to the bed face, is 725 psi (5.0 MPa). However, according to
43
recently developed seismic Building Code of Pakistan (SBC-07), the minimum compressive
strength of solid burnt brick is 1196 psi (8.25 MPa). Test data of masonry unit presented in
section 2.2.1.1 and test conducted in this research work reveals that the compressive strength
of unit complies with the EC 8 and PBC 07.
3.2.1.2 Water Absorption
Water absorption was determined for the masonry units in accordance with ASTM C 67. The
test data is given in Table 3.2.
Table 3.2 Water absorption of burnt clay brick
S.No Dry Wt
lb (N)
Sat. Wt
lb (N) Absorption (%)
1 5.76 (25.6)
7.02 (31.2) 21.88
2 6.15 (27.4)
7.2 (33.4) 17.07
3 5.82 (26.0)
7.02 (31.2) 20.62
4 6.09 (27.1)
7.27 (32.3) 19.38
5 6.21 (27.6)
7.31 (32.5) 17.71
Average 6.1 (27.1)
7.16 (32.0) 19.33
Cov 10%
The absorption of the masonry unit is very important property. Brick masonry made with
highly absorptive unit would have much less shear and tensile bond strength than those of
masonry made with less absorptive brick units [Calvi, G. M., et al, 1996]. In fact, brick
absorb water from mortar, leaving small amounts of water for hydration of cement which in
turn makes the mortar weak in both tension and compression and consequently resulted in
weakening of the masonry.
Although absorption of masonry unit is not considered for simulation, however, it was
determined for prototype unit to know the general trend of locally available units.
3.2.2 Masonry Mortar
Cement-Sand-Khaka (stone dust) mortar is selected for this study. The constituent materials
were mixed in proportion of one part cement to four parts each khaka and sand by volume.
44
Water to cement ratio was kept 1.6. Sand from Nizampur and khaka from Peshawar region
were used. The proportion of mortar selected is representative of field conditions. Specimens
of mortar were cured for 7 days in water and then were kept in moist room until testing day.
The 28 days compressive strength of the mortar, given in table 3.3, is determined on 2" cube.
The mortar specimens were sampled and tested according to ASTM C 109.
Table 3.3 Compressive strength of masonry mortar
S.No Area
in2 (mm2)
Load
(t)
Compressive Strength
psi (MPa)
1 4 (2581) 1.73 953.2
(6.6)
2 4 (2581) 2.11 1162.6
(8.0)
3 4 (2581) 1.55 854.0
(5.9)
4 4 (2581) 1.70 936.7
(6.5)
5 4 (2581) 1.85 1019.3
(7.0)
6 4 (2581) 1.92 1057.9
(7.3)
Average 4
(2581) 1.81 997.3 (6.9)
Cov 11%
The mean compressive strength of mortar collected from field (section 2.2.2) and used in this
study is well above the minimum compressive strength required by EC 8 and PBC 07.
According to EC 8, the masonry mortar used for un-reinforced and confined masonry should
not be less than 725 psi (5.0 MPa). And according to PBC 07 minimum compressive strength
of mortar in seismic zone 2, 3 and 4 should be 595 psi (4.1 MPa) and not greater than 75 %
of the compressive strength of masonry unit.
3.2.3 Prototype Masonry Assemblage
The mechanical properties of the constituent materials of masonry have been reported in the
preceding section. The prototype masonry has been tested for compression, diagonal tension
and cyclic behavior. The mechanical properties from these tests are to be used for simulation
purposes. The scope of the test is discussed in the subsequent sections.
45
Same type of mortar has been used for all prototype specimens. The proportion of the mortar
has been finalized to represent the field condition. One part of cement with four part each of
Khaka (stone dust) and sand have been mixed keeping water to cement ratio of 1.6. Sizes of
specimens are different for compression, tension and cyclic test.
3.2.3.1 Compression Strength and Modulus of Elasticity
The compressive strength and modulus of elasticity are the important parameters in the axial
and seismic resistance of masonry. These mechanical properties have been determined by
testing 15¾ x 9 x 19 inch (400 x 229 x 480 mm) thickness x width x height brick prism
under uni-axial compression. The mortar joint is 3/8 to ½ inch (10-12 mm) in thickness. The
bricks are laid in running bond. The prisms have been constructed by expert local mason.
The wetting of bricks and mixing of materials for mortar have been done by the mason and a
helper in such a way so as to be representative of field conditions. Samples of mortar have
been taken for the quality control.
The masonry specimens were stored and cured in the lab. The curing was started after 48
hours of specimens preparation. The specimens were wet cured for minimum of 7 days and
kept in the moist room for the rest of the time before testing.
The specimens were capped with gypsum on both ends 24 hours prior to testing. The
specimens were tested in 200 tonne Universal Testing Machine (UTM) in the Material
Testing laboratory, N-W.F.P UET, Peshawar as shown in figure 3.2. 3/4" (19 mm) thick steel
plate was used at the top between upper platen of UTM and the specimen.
Separate load cell was used for load readings. The load cell and four displacement gages
were connected with the UCAM-70 Data logger, data acquisition system. Two of the four
gages were connected on the top of the specimen to get the overall displacement of the
specimen and two (one on each face) were connected on the mid height. The dimensions and
instrumentations are shown in figure 3.3.
The compression load was applied at an average rate of 0.4 tone/sec.
46
Figure 3.2 Compression test of prototype masonry
Figure 3.3 Dimensions and instrumentations of prototype specimen for compression test.
The compressive strength of masonry prism is calculated by dividing maximum load over the
plan area of the prism. ASTM C1314 standard requires multiplying the masonry prism
strength by correction factor. The compressive strength values in the table below are un-
corrected values. The modulus of elasticity has been determined as specified in the ASTM
C1314, that is, secant modulus of elasticity between 1/20th and 1/3rd of the maximum
compressive stress of the prism.
47
Table 3.4 gives compressive strength and modulus of elasticity of brick masonry prism and
figure 3.4 illustrates the typical stress strain curve.
Table 3.4 Compressive strength and modulus of elasticity of prototype brick masonry
S.No
Designation
Compression Strength
psi (MPa)
Modulus of Elasticity "E"
ksi (GPa)
Maximum Average COV
(%) E Avg. E COV (%)
1 CE-1 705.4 (4.8)
839.63 (5.8) 24%
211.9 (1.46)
288.0 (2.0) 23%
2 CE-2 980.9 (6.7)
340.0 (2.3)
3 CE 3 1043.5 (7.2)
312.0 (2.2)
4 CE-4 628.7 (4.3)
218.0 (1.5)
The compressive strength of masonry from the equation proposed by Miha Tomazevic (Miha
Tomazevic 1999) is as follows:
0.65 0.25 ( )k b mf Kf f MPa= (3.1)
= 5.39 MPa or 781 psi
Where:
fb = Normalized compressive strength of unit in MPa
fm = Compressive strength of mortar in MPa
k is a constant and its value depends on the classification of masonry unit. In this case k is
0.5. By comparing compressive strength from experiment and equation, good correlation is
obtained. However, the equation is an approximate approach and could not be preferred over
experimentally determined value. The value of E is 342 times the compressive strength of
masonry, which lies between the range specified in (Miha Tomazevic 99).
48
y = 2E+09x3 - 5E+07x2 + 389478xR2 = 0.9942
0.0
100.0
200.0
300.0
400.0
500.0
600.0
700.0
800.0
900.0
1000.0
1100.0
0 0.002 0.004 0.006 0.008 0.01 0.012
Strain (in/in)
Stre
ss (p
si)
Figure 3.4 Typical stress strain curve of prototype brick masonry in compression
3.2.3.2 Diagonal Tension (Shear) in Masonry
In order to determine the tensile strength (shear strength) of masonry five square specimens
of nominal size 28¾ x 28¾ x 9 inch (730 x 730 x 229 mm) have been prepared and tested in
accordance with ASTM C519-02. Although, ASTM standards specify 48 x 48 inch
specimen, smaller specimen have been constructed to better handle them. The specimens
were constructed in English bond with joints thickness of 3/8 to ½ inch (10-12 mm). Figure
3.5 shows the diagonal specimen during testing.
The dimensions and instrumentations are illustrated in figure 3.6. Four gages have been used
to measure vertical shortening and horizontal extensions. Two gages are connected in a way
to measure vertical shortening and the other two horizontal extensions. The specimen is
positioned in standard loading shoes on the top and bottom corners. Vertical compression
load is applied through 50 tonne actuator in a manner that the specimen failed in 3-4 minutes.
The load and displacement readings of a specimen (PTG-5) could not be recorded because of
data logger not in a recording mode.
49
Figure 3.5 Diagonal compression (shear) test
Figure 3.6 Dimensions and instrumentations of diagonal tension test
The shear strength is calculated by the following formula:
0.707sPSA
= (3.2)
Where:
P = applied load, lb,
A = net area = (W+H)t/2
W = width of specimen, inch,
50
H = height of specimen, inch,
t = thickness of specimen, inch
The shear strain is calculated as follows:
( )V Hg
γ∆ + ∆
= (3.3)
where:
γ = shear strain,
∆V = vertical shortening, inch,
∆H = horizontal extension, inch.
The modulus of rigidity, G is calculated as follows:
sSGγ
= (3.4)
as ASTM E519 does not specify any range of stresses for the determination of modulus of
rigidity. However, 1/20th and 1/3rd of the maximum shear stresses are taken over which
stresses are assumed linear. The shear strength and modulus of rigidity is given in table 3.5.
Table 3.5 Tensile strength and modulus of rigidity of prototype brick masonry
S.No Designation
Shear Strength
psi (MPa)
Modulus of Rigidity "G"
ksi (GPa)
Maximum Average Cov (%) G Average Cov
(%)
1 PTG-1 41.7 (0.29)
51.3 (0.35) 14%
30.6 (0.21)
28.1 (0.20) 37%
2 PTG-2 49.7 (0.34)
15.6 (0.11)
3 PTG-3 55.9 (0.39)
26.0 (0.18)
4 PTG-4 57.8 (0.40)
40.4 (0.28)
The shear strength is about 6% of the compressive strength. Shear modulus is 10% of the
elastic modulus which lies in the experimentally evaluated range of 6% to 25% suggested by
Tomazevic M., 1999. According to EC 6 specification the shear modulus, G is 40% of the
elastic modulus is much higher then the experimental results.
51
3.2.3.3 Cyclic Test of Prototype Masonry Walls
Five 36 inch square specimens were prepared in English bond for cyclic test. The mortar
joint was ¾ to ½ inch (10-12 mm) thick. Reinforced concrete beams, 9" in thickness, were
provided at the top and bottom of the masonry wall. The specimens were wet cured for 7-10
days and was kept in moist room until testing. Mortar samples were collected during
construction of the masonry specimens for quality control. The mortar samples were tested at
the same day of cyclic test. The specimens were white washed two days prior to testing for
the visibility of cracks during testing.
The dimensions and instrumentations are illustrated in figure 3.7. The specimens were tested
in straining frame, structural laboratory. Three load cells were used for applying vertical
compression and horizontal cyclic loads. A 50 tonne capacity actuator was used for applying
constant vertical compression load. Vertical load was applied on 3" (76 mm) thick steel plate
placed on 1.5" (38 mm) diameter steel rollers. The steel rollers in turn are rolling on a highly
polished 3/8" (10 mm) thick steel plate during the horizontal displacement of the specimen.
Two 25 tonne capacity actuators were used to apply horizontal cyclic load, each one applying
half cycle. Two LVDTs were connected on either side of the specimen to measure top
displacement. Two LVDTS were also connected at the bottom of the specimen to measure
shear sliding if any. A LVDT is also used to measure the slippage of the bottom concrete
beam over the steel girder (figure 3.8).
The bottom beam of the specimen is connected with a stiff steel girder with the help of four
bolts on each side.
All the instruments are connected with UCAM70 data logger for data acquisition. A list of
equipment and their specification is given in Appendix A.
52
Figure 3.7 Dimensions and instrumentations of prototype cyclic load test
Figure 3.8 Cyclic test setup
7 tonne of vertical compression load was applied on the specimen. The vertical compression
load is equivalent to 47 psi (0.32 MPa) compression stresses which is representing vertical
stresses in the pier in two story typical Pakistani masonry building (5 marlas/1361 sft typical
houses). The test was displacement control. Displacement history is illustrated in figure 3.9.
53
-14.0-12.0-10.0-8.0-6.0-4.0-2.00.02.04.06.08.0
10.012.014.0
Dis
plac
emen
t (m
m)
Figure 3.9 Displacement history for prototype cyclic test.
Figure 3.10 shows typical hysteresis loop of the cyclic test. The load-displacement envelop
(figure 3.11) was plotted by maximum load and its corresponding displacement in the first
cycle of each increment. Rocking was predominant failure mode of all specimens. One
specimen was damaged during transportation and positioning. Bilinear idealized curve is also
Table 3.10 Compressive strength of lime based-masonry mortar
No Speciman
Designation
Mix
proportion
Water
content*
(ml)
Compressive Strength, psi (MPa)
7 days
Strength 28 days Strength
90 days
Strength
Avg.
Strength
Avg.
Strength
Cov
(%) Avg. Strength
1 LS 13/F 1:03:00 450 59.5 (0.41)
75.3 (0.52) 11.2 134.0
(0.92)
2 LSP133 1:03:03 450 94.3 (0.65)
119.4 (0.82) 25.4 113.9
(0.79)
3 LSSr133 1:03:03 650 108.8 (0.75)
137.75 (0.95) 8.0 126.7
(0.87)
4 LSSr11515 1:1.5:1.5 650 94.3 (0.65)
119.4 (0.82) 21.8 132.2
(0.91) *The water content is required for 3.90 lb (1765 g) of dry materials mixed in the proportion given in column 3 Cement lime sand mortar (CLS115) has been selected for its 28 days compressive strength
matching prototype mortar strength.
3.3.3 Micro Concrete
For the slab and confining elements concrete, different cement-sand and cement-lime-sand
mixes have been prepared. Ordinary Portland cement and hydrated lime has been used. River
sand of grain size from 0-2 mm have been used for the micro concrete. 7-days and 28-days
compressive strength data of micro concrete is presented in table 3.11.
Cement sand mortar (CS16) has been selected for floor concrete and cement-lime-sand
(CLS115) has been used for confining element. The selected mix proportion represents the
lower bound value of compressive strength of concrete as discussed in section 2.2.
63
Table 3.11 Compressive strength of micro-concrete
S.No Designation Water
content (ml)
7 days Compressive
Strength, psi (MPa)
28 days Compressive Strength
psi (MPa)
Average Cov (%) Average Cov (%)
1 CS15 550 ml 391.7 (2.7) 3.14 870.0 (6) 8.98
2 CS16 - 287.6 (1.98) 6.45 388.2 (2.68) -
3 CS110 550 ml 79.5 (0.55) 18.36 185.9 (1.28) 13.34
4 CLS114 650 ml 575.3 (3.97) 6.45 1117.8 (7.71) 14.0
5 CLS115 - 212.8 (1.47) 7.61 367.2 (2.53) -
6 CLS116 650 ml 219.8 (1.52) 15.2 322.7 (2.23) 26.45
7 CLS1110 600 ml 142.6 (0.98) 8.63 386.0 (2.66) 3.63
8 CLS1112 550 ml 123.9 (0.85) 16.58 266.6 (1.84) 27.35
9 CLS128 700 ml 108.7 (0.75) 14.7 161.4 (1.11) 8.69
10 CLS1210 600 ml 114.6 (0.79) 4.67 261.9 (1.81) 15.46
11 CLS1212 600 ml 98.2 (0.68) 15.56 173.0 (1.19) 45.87
12 CLS1215 550 ml 74.8 (0.52) 10.82 184.7 (1.27) 8.77
13 CLS148 900 ml 72.5 (0.5) 14.78 81.8 (0.56) 44.0
*The water content is required for 6.10 lb (2771 g) of dry materials mixed in the proportion given in column 2
3.3.4 Reinforcing Bar
3 mm diameter aluminum wire has been used as model reinforcing bar for confining
elements. The tensile strength of the wire is 29,000 psi (200 MPa). For the model concrete
floor, 3 mm diameter galvanized wire was used.
3.3.5 Model Masonry
In order to fulfill the simulation requirements, model masonry walls have been tested in
compression, diagonal tension and cyclic test. Modulus of elasticity, modulus of rigidity and
energy dissipation capacity has been determined for the extrapolation purposes.
3.3.5.1 Compressive Strength and Modulus of Elasticity
Model walls from all the three types of model bricks were constructed. The walls had the
same bond pattern as was followed for the prototype masonry walls. Horizontal and vertical
64
joint thickness was 2.5-3.0 mm. The walls have been built in CLS 115 mortar. No moist
curing has been done for the walls. Three walls have been built for each type of brick unit.
The walls have been instrumented with four deflection gages. Two gages were used to
measure vertical deformation. The other two gages were connected in the way to measure
deformation of overall depth of the wall. Both sides of the model walls were capped with
gypsum 24 hours prior to testing. The walls have been tested in the universal testing
machine. Deformation and load data has been acquired by data logger. The dimensions and
instrumentations of the walls for compression strength is illustrated in figure 3.16. The
compression strength and modulus of elasticity of the model wall is given in table 3.12.
Figure 3.17-3.19 shows typical stress strain curve for the three different types of bricks
prism.
Figure 3.16 Dimensions and instrumentations of model walls for compression.
65
Table 3.12 Compressive strength and modulus of elasticity of model masonry
S.No
Designation
Compression Strength
psi (MPa)
Modulus of
Elasticity "E", ksi (GPa)
fc Avg. fc Cov
(%) E Avg. E Cov (%)
1 CE 2.1/111-1 322.63 278.6
(1.92) 26%
50.86 46.28
(0.32) 29% 2 CE 2.1/111-2 316.97 56.68
3 CE 2.1/111-3 196.13 31.31
4 CE 2.1/211-1 701.36 692.9
(4.78) 4%
137.57 96.54
(0.66) 37% 5 CE 2.1/211-2 663.60 74.26
6 CE 2.1/211-3 713.77 77.78
7 CE 1.9/311-1 654.89 707.2
(4.88) 10%
70.62 56.15
(0.39) 22% 8 CE 1.9/311-2 677.11 48.11
9 CE 1.9/311-3 789.50 49.72
It could be seen from table 3.12 that compressive strength of larger sized bricks (that are
brick #2 and 3) are more than the actual model brick. The masonry constructed from brick #2
and 3, resulted in high elastic modulus than the masonry constructed of brick #1. In order to
compare compressive strength and modulus of elasticity of model and prototype good
correlation has been obtained.
Specimen designation is explained below:
CE 2 / 1 1 1 -1
Aspect ratio
Type of brick Masory mortar
Compression and Modulus of Elasticity
Mortar for Brick No. of specimen
66
y = 6E+07x3 - 4E+06x2 + 63813xR2 = 0.9838
0.0
50.0
100.0
150.0
200.0
250.0
300.0
350.0
400.0
0 0.005 0.01 0.015 0.02 0.025 0.03
Strain (in/in)
Stre
ss (p
si)
Figure 3.17 Typical stress-strain curve for model brick No.1
y = 6E+07x3 - 4E+06x2 + 63813xR2 = 0.9838
0.0
100.0
200.0
300.0
400.0
500.0
600.0
700.0
800.0
900.0
0 0.005 0.01 0.015 0.02 0.025
Strain (in/in)
Stre
ss (p
si)
Figure 3.18 Typical stress-strain curve for model brick No.2
67
y = 6E+07x3 - 4E+06x2 + 63813xR2 = 0.9838
-100.0
0.0
100.0
200.0
300.0
400.0
500.0
600.0
700.0
800.0
0 0.005 0.01 0.015 0.02 0.025 0.03
Strain (in/in)
Stre
ss (p
si)
Figure 3.19 Typical stress-strain curve for model brick No.3
3.3.5.2 Tensile Strength and Modulus of Rigidity
Model masonry square walls of size 6.81 x 6.81 x 2.2 (173 x 173 x 56 mm) were constructed
and tested in diagonal compression complying ASTM C 1314. Three walls, each from the
three types of bricks, were tested at the age of 28 days. The walls were constructed in English
bond with 2.5-3 mm joint thickness (figure 3.20). The modeled walls were air cured in the
laboratory at temperature 20-25 C°.
68
Figure 3.20 Model wall for tensile (shear) strength
Four displacement transducers were used to capture horizontal elongation and vertical
shortenings. The two horizontal and two vertical transducers were connected with a steel
plate as shown in figure 3.21. Steel shoes were placed on top and bottom of the specimen.
The specimens were properly caped with gypsum. The walls dimensions are illustrated in
figure 3.22.
Figure 3.21 Model wall during diagonal compression (shear) test
69
Figure 3.22 Dimensions of Model walls
The walls were tested in 200 tonne UTM. Additional load cell, connected with the UCAM 70
data logger, were used to capture load values. The displacement gages were connected with
the data logger. Results of diagonal tension test are presented in Table 3.13. Shear stress to
shear strain is plotted in figure 3.23-3.25.
70
Table 3.13 Tensile (shear) strength and modulus of rigidity of model masonry
The base shear coefficient and story rotation angle are ploted in figure 5.18. The data upto
test run 60 could only be plotted. Data for higher test run were not measured as the
instruments were romoved from the model. The model could have resisted story forces
more than the measured corresponding to test run, however, it is assumed that the maximu
resistance is observed during test run 60 on the basis of damage propogation. The cracks
144
developed during test run 30 were opened up and propagated in to the confining element and
sheared the mico-concrete.
0
0.5
1
0.0% 0.5% 1.0% 1.5%
Story Drit Rotation (%)
Bas
e Sh
ear
Coe
ffici
ent
Figure 5.18 Story resistance envelop
The first story rotation angle corresponding to significant cracks is 0.28 % and the
corresponding base shear coefficient is 0.447. It is worth mentioning that the model has been
tested untill the in-plane walls practically collapsed or severely damaged. The maximum
ground motion (PGA) recorded at this stage was 3.65 g. However, the ultimate limit states
could not be characterised by the experiemental resistance envolpe, available visual
information and experiemental resistance envelope of single story model of this research, and
test carried out on confined masonry buildings at other institutes could be utilised to define
the limit states.
The ratios of base shear coefficient (BSC) and first story rotation angle at maximum
resistance and ultimate state to damage limit state has been determined for different confined
masonry models and is summerized in table 5.11. The number of story and wall density ratio
of each model is also provided. It could be seen from the table that no specific coorelation
has been obtained in BSC and drift ratio and wall density or number of storires.
145
Table 5.11 Base shear and first story ratio at maximum and ultimate state
Model
Designation
No of
story
of model
building
Wall Density
ratio (%)
Maximum
Resistance/Damage Limit
Ultimate state/Damage
Limit
Reference
BSC
First Story
Drift,
Ф
BSCFirst Story Drift,
Ф (%)
M1/1d 2 3.5 1.38 5.2 0.6 10.37 Tomazevic, M., 2004
M1/1c 2 3.5 1 1 0.433 15 Tomazevic, M., 2004
M1 3 5.0 1.57 1.36 0.533 43 Tomazevic, M., 1996
M2 3 5.6 2.04 4.64 1.057 67.9 Tomazevic, M., 1996
M1 1 - 1.1 1.86 0.62 5.1 Alcocer, S.M., 2004
M3 3 4.1 1.1 1.85 0.91 6.73 Alcocer, S.M., 2004
Single story 1 5.64 1.53 4.72 1.06 6.28 This research
The scatter of the data of first story rotation angle at ultimate limit is large as it depends on
the structural system, type of materials used (fired clay brick, concrete block etc) and
structural configuration of the building used.
The following BSC and story drift ratios are expected at ultimate state for the double story
model studied in this research work:
Ratio of Base Shear Coefficient at Ulitmate and Damage limite state = 0.75
Ratio of First Story Drift at Ulitmate and Damage limite state = 5.5
The base shear coefficient and first story drift ratio at the ultimate state comes out to be 0.336
and 1.43 % respectively. A similar attempt has also been made by (Tomazevic, M., 2007) to
correlate the observed damage with displacement capacity and limit states. The ranges of
values for story rotation assigned to characteristic limit states are given as under:
Crack limit, Øcr = 0.2 % to 0.4 %,
Maximum Resistance, ØRmax = 0.3 % to 0.6 % and,
Limite state at collapse, Øcoll = 2.0 % to 4.0 %.
The base shear coefficient and story rotation angle for the three limit states for double story
model are given in table 5.12. The maximum ground acceleration (pga) measured during the
146
last test run was 3.65 g which resulted in the disintegation of ground floor in-plane walls.
However, in order to prevent the disintegraton of wall at the ultimate state, a lower value of
PGA is conisdered. The PGA measured at the late test run is reduced by 30%. The hysteretic
envelop is plotted in figure 5.18.
Table 5.12 Base shear and first story ratio at maximum and ultimate state
Description of limit state Base shear coefficient Story rotation angle
(%)
Maximum Ground
Acceleration (g)
Elastic Limit (R30) 0.448 0.26% 0.636
Maximum Resistance (R60) 0.680 0.70% 1.153
Ultimate State 0.336 1.43% 2.55
By comparing the values of rotation angle at the three limit states, it could be concluded that
experimentally obtained rotation angle at damage limit state is within the range assessed by
(Tomazevic, M., 2007). However, the rotation angle at ultimate state is less than the assessed,
the value is close to the upper bound at the maximum limit state.
0
0.2
0.4
0.6
0.8
0.00% 0.25% 0.50% 0.75% 1.00% 1.25% 1.50%
Story Drift Rotation (%)
Bas
e Sh
ear
Coe
ffic
ient
Figure 5.19 Story resistance envelop
147
5.3.9 Response Modification Factor of Double Story Model
The base shear coefficient and story drift ratio has been idealized by bilinear curve as shown
in figure 5.19. 20% degradation from the maximum resistance limit is considered in the
idealization of the curve. The different parameters of the idealized curve are given below:
du = Ultimate displacement of the idealized bilinear curve = 0.36 inch (9.14 mm)
de = displacement at the elastic limit of the idealized bilinear curve = 0.11 inch (2.72 mm)
Global ductility ratio, µu = 3.4 and,
Response modification factor, R from equation 5.5 = 2.41
The response modification factor evaluated for double story building on the basis of global
ductility factor seems small as suggested by other authors. This could be partly because of
the reasons that full response upto collapse were not measured and maximum resistnace was
conservatively assumed at lower value. Secondly, the response accelerations corresponding
to displacement in channel # 9, considered for the evaluation of base shear, are small as
compared the maximum acceleration in each test run.
0
0.2
0.4
0.6
0.8
0.00% 0.25% 0.50% 0.75% 1.00% 1.25% 1.50%
Story Drift Rotation (%)
Bas
e Sh
ear
Coe
ffic
ient
Figure 5.20 Hysteretic envelope and idealised bilinear relation
5.4 SEISMIC RESISTANCE OF PROTOTYPE BUILDING
The results obtained during shaking table test of the single and double story models are
extrapolated to prototype building by taking into consideration the laws of model similarity.
_____ Hystersis Envelope
------- Idealised Curve
148
The maximum shake table motion during characteristics limit states should be converted to
maximum ground motion which would cause these limit state in the real building. Similarly,
the properties of model observed during the shaking table such as frequency and damping
during each test run, and base shear and drift at the three limit states should be scaled to the
prototype building.
However, properties of masonry materials and masses over each floor are adequately
simulated; the mechanical properties of masonry assemblage are not correctly modeled.
Actual relationship developed as a result of experimentally obtained properties of prototype
and model assemblage should be used to extrapolate model result to prototype.
As discussed earlier, shear was predominant mode of failure in both the models, scale factors
for each physical quantity is calculated as a result of scale factor corresponding to shear
strength. Table 5.13 gives the actual scale factors calculated keeping the geometric scale
factor of 4 and strength factor of 1.6. The scale factors for acceleration and time comes out to
be 0.4 and 3.16 respectively.
Table 5.13 Modeling scale factors to extrapolate model test results to prototype building
Physical Quantity Relationship Scale factor
True Model This Study
Length (L) SL = Lp/LM 4 4
Strength (f) Sf = fp/fm =SL 4 1.6
Strain (ε) Sε = εp/εm 1 1
Sp. Mass (γ) Sγ=γp/γM 1 1
Displacement (d) Sd = SL 4 4
Force (F) SF = SL2 Sf 64 25.6
Time (t) St = SL √(Sε Sγ/ Sf) 2 3.16
Frequency 1/ St 0.5 0.316
Velocity Sv = √(SεSf/ Sγ) 2 1.26
Acceleration Sf /(SL Sγ) 1 0.4
Table 5.14 and 5.15 give base shear and story drift corresponding to limit states of prototype
single and double story building respectively. The table also gives ground motion defined by
peak ground acceleration. The peak ground acceleration is determined by multiplying
149
maximum shake table acceleration with acceleration scale factor (0.4). The same acceleration
scale factor is used to convert the base shear force for model to prototype building.
Table 5.14 Parameters of seismic resistance for single story building
Description of limit state Base shear
Coefficient
Story Rotation
angle (%)
Maximum Ground
Acceleration
(g)
Elastic limit (R125) 0.482 0.25 0.42
Maximum Resistance (R175) 0.778 1.18 0.81
Ultimate state (R150) 0.18 2.1 1.15
Table 5.15 Parameters of seismic resistance for double story building
Description of limit state Base shear
Coefficient
Story Rotation
angle (%)
Maximum Ground
Acceleration
(g)
Elastic limit (R30) 0.180 0.37 0.254
Maximum Resistance (R60) 0.272 0.55 0.461
Ultimate state 0.134 0.80 1.02
It could be seen from the tables that both single and double story models designed according
to Eurocode could be used in high seismic zones of Pakistan Building code 2007. Both the
models properly constructed and designed could resist strong earthquake. However, the
double story model because of its small wall area in the short direction of the ground floor
would significantly be damaged. Separation of walls from tie column and/or collapse of walls
would result as a consequence of strong earthquake.
5.5 COMPARING SINGLE AND DOUBLE STORY BUILDING
Single story building is comparatively more rigid than the double story building. The wall
density ratio of single story in the short direction (parallel to excitation) is 5.64 % as
compared to 3.87 % in the ground floor and 5.99 % in the second floor. This could be the
reason for the double story model that ground floor was more damaged than the first floor.
However, concentration of damage in the ground floor has also been observed by other
researchers. Damage limit state was observed at 0.42 g in the single story model, however
first significantly crack in the double story has been observed at 0.22 g.
150
However, significant degradation has been observed in single story model at the last test run,
the confining elements keep the walls from disintegration. At the last test run, separation of
walls and buckling of model rebars in the confining element were observed. The in-plane
walls in the ground floor of the double story on other hand disintegrated at the last test run.
The confining element could not prevent the walls disintegration at strong ground shaking.
The single story buildings are much stiffer. Designed according to Euro code, the single story
buildings could survive collapse at strong ground motion.
It is concluded that the double story building cracks at lower PGA value than the single story
building. As discussed earlier the walls disintegrated after their separation from tie columns.
It is recommended to provide horizontal reinforcement to connect the tie columns and
masonry walls at ground floor. As the first story is comparatively stiffer than the ground floor
and the damage is also small, the toothing would work to keep the integrity of masonry walls
and tie columns.
151
CHAPTER 6 CONCLUSION AND RECOMMENDATION
The objective of this research work was to determine the behavior of typical confined brick
masonry buildings subjected to seismic loadings. Response modification and ductility ratio of
the confined brick masonry building was evaluated. The research work was aimed at the
growing demand of confined brick masonry buildings after 2005 Kashmir earthquake which
resulted in the destruction of more than 450,000 buildings.
6.1 SUMMARY
A survey of building typology and inventory has been carried out in Peshawar city and
earthquake affected area, including Abbottabad and Mansehra city. About 65 building
drawings have been collected from these cities and analyzed. Typical single story and double
story buildings have been selected on the basis of wall density ratio close to mean value of
the analyzed drawings.
Masonry materials survey has also been made, including survey of mortar, bricks and rebars.
Masonry mortar samples were collected in Peshawar as well as in Abbottabad and Mansehra
city from the construction sites. The survey shows that cement-sand and cement-sand-khaka
(stone dust) mortar were the typical mortar used in the area. The popular mix proportions are
1:6, 1:8 in cement-sand and 1:4:4 in cement-sand-khaka by volume. The mean compressive
strength of masonry mortar was found to be 837 psi (5.77 MPa). Test data of fired clay brick
and rebars was collected from Material Testing Lab, N-W.F.P UET, Peshawar. Mean
compressive strength of the brick was 2350 psi (16.2 MPa). Grade 40 and 60 deformed bar
are mainly used in buildings. Grade 40 bars are mostly used in residential buildings from 3/8
to 6/8 inch diameter. Data about compressive strength of structural concrete was; however,
not collected. Compressive strength of concrete cores cut from existing buildings vary from
1500 (10.34 MPa) to 2500 psi (17.24 MPa). The mean compressive strength of masonry unit
and masonry mortar comply with minimum requirements of Pakistan building code 2007 and
Eurocode 6.
152
The experimental work has been divided into three phases. In the first phase properties of
prototype masonry have been determined. In the second phase, properties of model materials,
including masonry mortar, masonry unit and micro-concrete have been simulated. The
mechanical properties of model masonry have been evaluated. In the last and final phase
model buildings have been fabricated and tested on shake table to evaluate their seismic
response.
In order to fulfill the requirements of complete model similitude law extensive experimental
work has been carried out for modeling masonry mortar, model brick and micro concrete.
Mixes with different proportions have been prepared and tested in compression to simulate
the masonry mortar. Cement-sand, cement-sand-lime, cement-lime-surkhi (brick remains),
cement-lime-marble powder and lime-surkhi were used in different proportions. Almost 40
batches of masonry mortar have been tested. Finally, cement-lime-sand (1:1:5), has been
selected as masonry mortar. Cement-lime-sand (1:1:5), having coarser sand than used in
masonry mortar, has been selected as micro concrete. Cement-lime-surkhi (1:1:2) has been
used for fabricating model masonry unit. Surkhi (burnt brick remains) passed through sieve
no 8 and retained on sieve 30 has been used to simulate specific weight of the prototype brick
as cement-lime-sand resulted in high specific weight. In order to reduce the fabrication
efforts, model bricks were fabricated in three different dimensions. That is actual model brick
of dimensions 2.2x1.1x0.67 (56x27x17 mm) (length x width x height), model brick with
double height 2.2x1.1x1.34 inch (56x27x34 mm) and model brick double in width with
dimensions 2.2x2.2x0.67 (56x56x17 mm) are used in model masonry wallets.
The prototype and model masonry assemblage has been tested in compression, diagonal
compression and cyclic test. Modulus of elasticity and rigidity has been determined. The
compression strength of prototype masonry was 828 psi (5.7 MPa) and modulus of elasticity
was 290 psi (2.0 MPa). Diagonal shear strength of prototype masonry was 51.0 psi (0.35
MPa) and modulus of rigidity was 26.0 psi (0.18 MPa). Model masonry walls of all the three
type of bricks have been tested. It was concluded that model bricks doubled in height and
width resulted in high strengths than actual model brick walls.
153
A small scale model of single and double story building has been tested on single degree of
freedom shake table. Scale factor four and complete model similarity has been adopted for
the models construction on the basis of capacity of the shake table and economy.
Kobe accelerogram compressed in time with square root of scale factor but with the same
PGA value as original record has been used for shaking table test of the reduced model. The
models have been subjected to increasing intensity of vibration in each test run. The models
have been instrumented with accelerometers and displacement transducers connected at the
floor slab at front of the in-plane walls. Four accelerometers and four string pots have been
used in single story model. However, four string pots and two accelerometers have been
connected at each floor in double story models to measure response acceleration and
displacement at the floor level. Additional weights are attached to each floor of the model to
simulate the dead load of flooring. The confining elements (tie columns and bond beams) are
designed according to Eurocode requirements. However, horizontal reinforcement were not
provided to connect tie columns and masonry walls.
Both the models failed in shear. Cracks initiated in masonry walls were confined by the
confining elements. However, with the increased intensity of shaking, the cracks propagated
and damaged the tie columns. During the final intensity of shaking, crushing of masonry
units at the corners of walls, crushing of concrete and buckling of rebars in the tie columns
have been observed. In the case of double story building model, damage was concentrated in
the ground floor. However, the in-plane walls were severely damaged at high intensities of
shaking; the confining elements prevented the collapse of model. The confining elements
also prevented collapse of out-of-plane walls which are generally vulnerable in the URM
buildings. Separation of masonry walls from the tie columns at early stages of shaking could
be attributed to the absence of horizontal reinforcement. It is, therefore, recommended to
provide horizontal reinforcement to connect the tie columns and masonry walls. However,
further research work should be carried out to evaluate the effect of horizontal reinforcement
on the ductility and energy dissipation capacity of confined masonry walls. However, further
research work should be carried out to evaluate the effect of horizontal reinforcement on the
ductility and energy dissipation capacity of confined masonry walls.
154
Seismic resistance and deformation capacity of the model have been evaluated on the basis of
measured acceleration and displacement response. Base shear is determined by multiplying
the response acceleration, measured at the attainment of maximum response displacement
with masses concentrated at floor level. A graph of base shear coefficient, that is base shear
divided by total weight of the model and first story rotation angle, first story drift divided by
height of first story has been plotted. The curve is idealized as bilinear curve. Complete
elastic analysis has also been carried out by modeling the reduced scale building in SAP,
computer software. The response modification factor, R of single story model determined on
the basis of global ductility ratio is 3.04 and the ductility ratio, µ comes out to be 5.13 for
single story building. However, response modification factor determined by dividing elastic
base shear over ultimate base shear comes out to be 2.16 for the single story model. The
response modification factor evaluated for double story model on the basis of ductility ratio
is 2.41 which is conservative value.
In order to extrapolate the shaking table parameters and response characteristics of the model
to the prototype earthquake and building, laws of complete model similarity have been taken
into consideration. However, actual scale factor is to be considered instead of theoretical. The
actual scale factor is determined as the relation between properties of the prototype that are
target values to that of properties measured on model masonry. As the basic failure mode in
the single and double story model was shear, therefore shear strength should be kept as the
basis of the true scale factor. The scale factors for other physical quantities are determined on
the basis of geometric scale factor and the strength factor. The base shear and shake table
accelerations at the three characteristic states have been extrapolated to the prototype
building. The limit states are damage limit state where the first significant crack occurs or
where there is significant drop in the stiffness, the maximum resistance limit state and
ultimate limit state that is before collapse and where the structural walls in the first story are
severely damaged. Base shear and the acceleration at the three limit states are given in table
5.14 and 5.15. The tables are reproduced as given below.
155
Table 6.1 Parameters of Seismic Resistance for Single Story building
Description of limit state Base shear
Coefficient
Story Rotation
angle (%)
Maximum Ground
Acceleration
(g)
Elastic limit (R125) 0.482 0.25 0.42
Maximum Resistance (R175) 0.778 1.18 0.81
Ultimate state (R200) 0.18 2.1 1.15
Table 6.2 Parameters of Seismic Resistance for Double Story building
Description of limit state Base shear
Coefficient
Story Rotation
angle (%)
Maximum Ground
Acceleration
(g)
Elastic limit (R30) 0.180 0.37 0.254
Maximum Resistance (R100) 0.272 0.55 0.461
Ultimate state (R300) 0.134 0.80 1.02
It can be seen that both single and double story building would be able to resist with minor
damage an earthquake with PGA 0.40g and 0.25g and without collapse an earthquake with
PGA 1.1g and 1.0g respectively. It is concluded that properly constructed and designed
single and double story models could be used in high seismic zones (Pakistan Building Code
2007). On the basis of observed behavior and wall density ratio it could be concluded that the
Eurcode requirements are stringent for single story building.
6.2 CONCLUSIONS AND RECOMMENDATIONS
Based on the experimental work, the following conclusions and recommendations are made.
Further, research areas are identified to investigate vulnerability in the masonry buildings and
its subsequent mitigation.
Conclusions:
• Shear is the predominant failure mode in both single and double story models
• The analysis of measured response and observed behavior of the single story model reveals that the typical single story building would withstand with minor damage an earthquake of PGA 0.40g and would not collapse under an earthquake of PGA 1.15 g.
156
• The typical double story building would withstand with minor damage an earthquake of PGA 0.25g and without collapse an earthquake of PGA 1.0g.
• Single and double story confined masonry buildings properly designed and constructed could be used in high seismic zones (zone 3 and 4 of Pakistan Building Code 2007).
• The ground story walls could collapse or severely damaged during strong ground motion (that is PGA 0.8 and higher) because of the absence of horizontal reinforcement.
• The confining elements could prevent collapse of out-of-plane walls of both single and double story buildings at strong earthquake, if proper monolithic behavior of tie columns and masonry walls is achieved.
• The eurocode requirements for the design of confined masonry buildings seems stringent for single story building.
• The provision of toothing in walls could not prevent separation of walls from confining element at strong ground motion.
• The response modification factor of single story building confined according to Eurocode 8 is 3.0. Response modification factor for double story building designed according to Eurocode is determined to be 2.41. The Eurocode requirements for response modification factor are found adequate.
• The mean compressive strength of masonry unit, 2350 psi (16.2 MPa), tested in the Material Laboratory, N-W.F.P UET, Peshawar, comply with the minimum requirements of Pakistan Building Code 2007 and Eurocode 6.
• The mean compressive strength of masonry mortar collected from field in Peshawar, Abbotabad and Mansehra after Kashmir earthquake comply with the minimum requirements of Pakistan Building Code 2007 and Eurocode 6.
Recommendations:
• In order to delay extensive damage to masonry wall, it is recommended to provide horizontal bed joint reinforcement to connect the masonry walls and tie-columns.
• Wall density ratio should not be less than 5 % in seismic zone 3 and 4.
• The building should be regular in plan and elevation.
157
• Minimum thickness of masonry wall should be kept at 9 inch (229 mm).
• Minimum thickness of confining element should be 9 inch (229 mm).
• Minimum 1% longitudinal reinforcement should be provided in the confining element.
• Stirrups of 3/8 inch (10 mm) diameter bars should be provided in the confining elements at a maximum spacing of 6 inch (152 mm) c/c.
• At least 48 diameters splice length should be provided.
• The longitudinal reinforcement of tie-columns should be adequately anchored in foundation and bond beam/floor slab. The reinforcement must be tied with foundation reinforcement at the bottom as well as with the bond beam/floor slab reinforcement at the top.
• Tie-column should be provided at 16 ft (4.2 m) c/c horizontal spacing.
• Not more than two stories building should be constructed in zone 4 and three stories in zone 3.
• Minimum compressive strength of brick, mortar and concrete should be1800 psi (12.4 MPa) , 800 psi (5.5 MPa) and 2000 psi (13.8 MPa), respectively.
6.3 FUTURE RESEARCH WORK
• Numerical analysis of the typical confined brick masonry buildings is recommended.
• As real earthquake is three dimensional phenomenons, experimental testing of typical confined brick masonry building model on 6-degree of freedom shake table is recommended.
• Cyclic testing of full-scale confined brick masonry walls with different dimensions of confining elements and reinforced with different reinforcement ratio should be carried out. The optimum dimensions of confining element and reinforcement ratio could be suggested and made part of the Pakistan Building Code 2007.
• Shake table test of single story building model with smaller dimensions and less reinforcement ratio as required by Eurocode for both horizontal and vertical confining element should be tested and compared with the results of this research.
158
• Effect of different earthquake accelerograms on the typical confined masonry building should be carried out.
• Confined masonry walls with different horizontal reinforcement ratio should be carried out and the effect of horizontal reinforcement on the seismic resistance and ductility should be investigated in future studies.
• Shake table testing of typical confined masonry building with flexible diaphragm should be carried out to study its behavior.
• The study of seismic behavior of typical confined block masonry buildings is recommended for future research work.
159
REFERENCES
Abram D.P., “Seismic Response Pattern for URM Buildings”, Journal of the Masonry
Society, Vol 18, No.1, 2000.
ACI 318-63, “American Concrete Institute Building Code requirements for Reinforced
Concrete”, American Concrete Institute, Michigan, 1963.
ADB-WB. “Preliminary Damage and Needs Assessment-Pakistan 2005 Earthquake”, Asian
Development Bank and World Bank, Islamabad, Pakistan, 2005.
Alcocer, S. M., and Klingner, R. E., “Masonry Research in the America”, Masonry Research
in Americas, ACI (SP-147), pp, 239-262.
Alcocer, S. M., Arias, J. G., Vazquez., A., “Response Assessment of Mexican Confined
Masonry Structures Through Shaking Table Tests”, proc. 13th World Conference on
Earthquake Engineering, Vancouver, B.C., Canada, paper No. 2130.
Ali. Qaisar, “Seismic Risk Assessment of Un-reinforced Brick Masonry Buildings System of
Northern Pakistan”, Ph.D thesis, Department of Civil Engineering, N-W.F.P UET, Peshawar,
2004.
Applied Technology Council (ATC). NEHRP Commentary on the Guidelines for the Seismic
Rehabilitation of Buildings. Publication No.274, Federal Emergency Management Agency,
Washington, D.C. (FEMA-274), 1997b.
Aschheim, M., et al, “Improving The Earthquake Resistance And Sustainability of Confined
Masonry (Mixto) Dwellings in El Salvador” Proc. of the 8th U.S. National Conference on
Earthquake Engineering, San Francisco, USA, Paper No. 1462.
ASTM C67, “Test Methods for Sampling and Testing Brick and Structural Clay Tiles”, 4th
Ed., Philadelphia.
ASTM C 109, “Standard Test Method for Compressive Strength of Hydraulic Cement
Mortars (using 2-inch or 50 mm cube specimens)”, 4th Ed., Philadelphia.
ASTM E519, “Standard Test Method for Diagonal Tension (Shear) in Masonry
Assesmblages”, 4th Ed., Philadelphia.
160
ASTM C1314, “Standard Test Method for Compressive Strength of Masonry Prisms”, 4th Ed,
Philadelphia.
ASTM C615, “Standard Test Method for Compressive Strength of Masonry Prisms”, 4th Ed,
Philadelphia.
Bartolome, A. S., et al, ‘Seismic Behavior of a Three Story Half Scale Confined Masonry