Upload
mehmet-aydin-koemuer
View
246
Download
0
Embed Size (px)
Citation preview
7/28/2019 Dolgu Duvar Deney Resimleri
1/40
7/28/2019 Dolgu Duvar Deney Resimleri
2/40
7/28/2019 Dolgu Duvar Deney Resimleri
3/40
7/28/2019 Dolgu Duvar Deney Resimleri
4/40
7/28/2019 Dolgu Duvar Deney Resimleri
5/40
41
3.3.3 Masonry PrismsTwo types of masonry prisms were constructed and tested in this study. One type was
three-course high with a height to thickness ratio of 3.3 as shown in Figure 3.20.
Figure 3.20 3-High Prisms testing arrangement
A total of 12 prisms of this type including 3 hollow, and 9 fully grouted were constructed.
The prisms were tested for the compressive strength, the modulus of elasticity, and thestress-strain relationship of the masonry in accordance with ASTM C1314-07 Standard
test methods for compressive strength of masonry prisms.
The second type of masonry prisms, referred to as square prisms, was 4-course high, 2-
course wide with a height to thickness ratio of 4.4. A total of 27 prisms of this type
including 9 hollow, 9 partially grouted, and 9 fully grouted were prepared. The partially
grouted prisms were grouted on the two outer cells of the prisms. They were tested for
compressive strength under three loading conditions. Vertical compression refers to a
loading direction which was perpendicular to the bed joint; horizontal compression
indicates that the loading was applied in parallel to the bed joint; and the diagonal
compression was to load the specimen in diagonal direction. The diagonal, vertical and
horizontal loading conditions are shown in Figure 3.21. These prisms were tested to have
7/28/2019 Dolgu Duvar Deney Resimleri
6/40
42
a better understanding of the effect of loading direction on the compressive strength of
masonry assemblage. All prisms were cured together with the walls and tested at
approximately the same time as the walls. LVDTs were mounted at the front and the back
of all the square prisms to obtain deformation readings. The gauge lengths of the LVDTs
were kept at approximately 200 mm. Horizontal and vertical specimens had a similar
testing procedure to the three-high prisms. For testing in the diagonal direction, two
custom-made supports were used for the loaded corners where each support was a V-
shaped joint inside a rectangular box designed to encase the corners and provide a
straight surface for testing as shown in Figure 3.22.
Diagonal Vertical Horizontal
Figure 3.21 Loading Conditions of Compression Test for Square Prisms.
Figure 3.22 Diagonal prism test loading shoe.
7/28/2019 Dolgu Duvar Deney Resimleri
7/40
48
4.2.2.2MortarType S mortar was used in the construction of wall infills. Three batches of mortar were
used in the building of specimens and mortar cubes were made from each batch. A total
of twenty-six 50mm mortar cubes were tested for their 28-day strength according to CSA
A179-04 (2004) Mortar and grout for unit masonry. For batch 1 mortar, nine cubes were
tested for 7-day strength as well for quality control purposes. Figure 4.3 shows an
example of the failure of mortar cubes under compressive testing where most of the
mortar cubes showed a conical shear or pyramidal shape failure. The compressive
strength for mortar cubes are summarized in Table 4.3 where the mean 28-day
compressive strength of all the mortar cubes tested was 13.6MPa. The average 28-day
compressive strengths were 15.1MPa from batch 1 (BM1) and 9.6MPa for Batch 2
(BM2) mortar cubes. Mortar cubes from batch 3 (BM3) attained compressive strength of
19.6MPa. The COVs of all 3 batches mortar strength were well within the specified limit
of 15%. It should be pointed out that BM2 mortar strength was lower than the minimum
28-day strength (12.5MPa) specified in CSA A179 - 04 (2004) for type S mortar under
laboratory conditions.
Figure 4.3 Failure of mortar cubes under compressive testing
7/28/2019 Dolgu Duvar Deney Resimleri
8/40
51
Figure 4.4 Failure of grout cubes under compressive testing
Figure 4.5 Load-displacement diagram of a grout cube in compressive
0
5
10
15
20
25
0 1 2 3 4 5
Load(kN)
Displacement (mm)
7/28/2019 Dolgu Duvar Deney Resimleri
9/40
53
while the grouted column remained practically intact. These are common failure modes
in prisms as reported by Drysdale and Hamid (2005).
Vertical crack (face-shell) Vertical crack (web) Face-shell spalling
Figure 4.6 Failure of 3-high prisms under compressive loading.
Net areas were used for the calculation of compressive strength of prisms. The net area
for the 3-high ungrouted prisms is the outside shell-area of the prism as shown in Figure
4.7. The net area of the 3-high fully grouted prisms was the gross area less the web area
of the unit block as shown in Figure 4.8.
Figure 4.7 Net area for 3-high ungrouted prisms
Figure 4.8 Net area for 3-high grouted prisms
The compressive stress of the BP1 ungrouted and grouted prisms were 12.7MPa and
8.0MPa, a difference of 37% with the difference in net areas of 49%. The higher
7/28/2019 Dolgu Duvar Deney Resimleri
10/40
55
For the horizontally loaded prisms, net area was calculated to be the same as the
vertically loaded prisms with respect to the different grouting situations. For diagonally
loaded prisms, the net area used for the fully grouted diagonally loaded prisms was the
rectangular area where the prisms were in contact with the loading shoes as shown in
Figure 4.11. The net area of the hollow prisms was taken as the faceshell area contained
within the loading shoe. For partially grouted diagonal loaded prisms, the average areas
between the fully grouted and hollow diagonally loaded prisms were used to calculate the
net areas.
Figure 4.11 Diagonally loaded prisms net area
To study the behavior of the prisms in different loading directions, a stress-strain curve of
each prism specimen was obtained and the Modulus of Elasticity, Em, was also
determined. It is a common practice to express Em in terms of compressive strength, f'm,
obtained with loading applied perpendicular to the prism bed joint. In this study, this
practice was followed and thus Em in the following tables is expressed in terms of f'mV,
compressive strength of the vertically loaded prisms. Figure 4.12 shows the stress-strain
curves of partially grouted prisms loaded in three directions. The full-set of stress-strain
curves can be found in Appendix A. It can be seen that partially grouted prisms loaded in
the horizontal direction displayed the lowest modulus of elasticity. However, diagonally
loaded partially grouted prisms had the highest modulus of elasticity. As expected,
Loading
Shoe
Net Area
7/28/2019 Dolgu Duvar Deney Resimleri
11/40
41
4 Preliminary System Identification
This chapter deals with the results of the component and system tests performed before the start
of the actual shake-table experiments involving the application of earthquake records. The
component tests include standard concrete compression tests and concrete split tension tests on
concrete cylinders fabricated during the construction of the test structure. These component tests
also include masonry compression, shear and bending tests performed on masonry prisms and
masonry panels constructed at the same time as the URM infill wall in the test structure. The
system tests are a series of pull-back (snap-back) tests on the test structure at different stages of
completion of the test structure configuration, namely before and after building the URM infill
wall and after the placement of additional mass on the test structure. The results of these
preliminary tests are used to gather the required data for calibrating and validating analytical
models of the test structure as discussed in Chapter 7 and to document the state of the test
structure at the beginning of the shake-table experiments as a point of reference when discussing
the results of these experiments in Chapter 6.
4.1 COMPONENT TESTS
4.1.1 Concrete Cylinder Compression Tests
A total of 30 test cylinders are prepared after each concrete placement for foundation, columns,
and beams and slab in accordance with ASTM C 837-99. The cylinders are kept in the same
environmental conditions as the test structure. Three uniaxial compression tests are performed
for each patch of concrete at different times to monitor the strength gain with time, the last of
which was performed on the day before the start of the shake-table experiments. Mean values
and the coefficient of variation (COV) of each test group are reported in Tables 4.1 and 4.2,
respectively. It can be observed that the compressive strength of the concrete on the day of the
7/28/2019 Dolgu Duvar Deney Resimleri
12/40
42
test is on average 22% higher than its 28-day compressive strength. No clear conclusion can be
made on the COV for the test results due to the small sample size, namely three for each reported
value. However, based on the results in Table 4.2 and for the purpose of practical reliability
analyses as in Chapter 8, one may consider a mean value of the COV of the concrete
compressive strength of 4.5%. The test setup and the resulting relationship of the mean strength-
gain with time for the used concrete are shown in Figure 4.1. The mean strength values obtained
at the time of the shake-table test are used in the computational modeling of the test structure
(Chapter 7) and for system identification purposes (Chapter 6).
Table 4.1 Mean uniaxial concrete compression test results.
Structural
element
First group
[ksi (MPa)]
Second group
[ksi (MPa)]
Third group (start of
shake-table experiments)[ksi (MPa)]
Foundation 3.27 (22.5)@11 day 4.15 (28.6)@28 day 4.98 (34.3)@567 day
Columns 3.04 (20.9)@5 day 4.36 (30.1)@33 day 5.40 (37.2)@552 day
Beams and slab 3.28 (22.6)@10 day 4.53 (31.2)@32 day 5.56 (38.3)@538 day
Table 4.2 COV of uniaxial concrete compression test results.
Structural
element
First group
(%)
Second group
(%)
Third group (start of
shake-table experiments)
(%)Foundation 7.6@11 day 5.6@28 day 1.1@567 day
Columns 4.0@5 day 4.6@33 day 9.1@552 day
Beams and slab 5.4@10 day 1.6@32 day 1.4@538 day
0 100 200 300 400 500 6000
1
2
3
4
5
6
Concrete
strength[ksi]
Time [days]
Beams and slab
Coulmns
Foundation0
10
20
30
40
[
MPa]
(a) Test setup (b) Concrete compressive strength-gain with time
Fig. 4.1 Concrete compressive strength test.
7/28/2019 Dolgu Duvar Deney Resimleri
13/40
43
4.1.2 Concrete Cylinder Split Tension Tests
Concrete cylinder split tension tests are performed on three concrete cylinders constructed using
the columns concrete batch. The tests are carried out at the start of the shake-table experiments.
The split tension tests conform to ASTM C496 and are used to identify the tensile strength of
concrete cylinders as defined in Equation 4.1.
dl
Pfct
2= (4.1)
where P is the maximum load at failure, l and d are the length and diameter of the cylindrical
specimen, respectively. The failure is sudden, with a vertical splitting crack across the section of
the specimen. The individual results as well as their mean value and COV are summarized in
Table 4.3. The mean value of the tensile splitting strength of concrete (about 8% of the
compressive strength) is used for the computational modeling of the test structure as described in
Chapter 7.
Table 4.3 Concrete split tension tests.
SpecimenMaximum load
[kips (kN)]ctf
[psi (MPa)]
1 48.8 (217) 431 (2.97)
2 50.4 (224) 446 (3.08)
3 46.5 (207) 411 (2.83)
Mean 48.6 (216) 429 (2.96)
COV 3.3% Fig. 4.2 Concrete split tension test setup.
4.1.3 Masonry Compression Tests
Three masonry prisms are constructed at the time of the construction of the URM infill wallaccording to the requirements of the ASTM C 1314. The prisms are capped and secured to two
steel plates on top and bottom using Hydrocal gypsum cement, and tested under uniaxial
compression 28 days after the wall construction. Both the axial load and the axial displacement
(measured between the two steel plates) of the masonry prisms are recorded during these axial
compression tests. Figure 4.3 shows the configuration of the masonry prism tests as well as the
typical failure mode consisting of vertical splitting and crushing. The stress-strain curves for the
7/28/2019 Dolgu Duvar Deney Resimleri
14/40
44
three prisms are shown in Figure 4.4 with the individual results, as well as their mean values and
COV, summarized in Table 4.4. In this table mof , mE , mo , and mu indicate the compressive
strength of the masonry, the modulus of elasticity measured as the secant modulus at 75% of the
compressive strength, and strain corresponding to maximum compressive stress and ultimate
strain of masonry corresponding to the residual stress value of momu ff 15.0= , respectively, as
shown in the insert of Table 4.4.
(a) Test setup (b) Typical failure mode
Fig. 4.3 Masonry prism tests.
0 0.005 0.01 0.0150
0.5
1
1.5
2
2.5
3
Strain
Stress[ksi]
0
5
10
15
20
[MPa]
Specimen 3
Specimen 2
Specimen 1
Fig. 4.4 28-day compression stress-strain curves for masonry prisms.
7/28/2019 Dolgu Duvar Deney Resimleri
15/40
45
Table 4.4 28-day uniaxial compression test results for masonry prisms.
Specimen mof
[ksi (MPa)]
mE
[ksi (GPa)]mo mu
1 2.31 (16.0) 948 (6.54) 0.0038 0.0120
2 2.31 (16.0) 812 (5.60) 0.0035 0.00863 2.76 (19.0) 935 (6.45) 0.0040 0.0120
Mean 2.46 (17.0) 898 (6.19) 0.0038 0.0109
COV 10% 8.3% 6.1% 18% Tension Compressionmo
mu
mof
muf
mE
mo0.75 f
Tension Compressionmo
mu
mof
muf
mE
mo0.75 f
Tension Compressionmo
mu
mof
muf
mE
Tension Compressionmo
mu
mof
muf
mE
mo mu
mof
muf
mo
mo mumu
mofmof
mufmuf
mEmE
mo0.75 fmo0.75 f
4.1.4 Masonry Diagonal Tension (Shear) Tests
In order to determine the shear strength of masonry, diagonal tension (shear) tests in accordance
with ASTM E519 are performed on three specimens. The used specimens are 2 -5"2-5" (75
cm75 cm) instead of the usual 44 (122 cm122 cm) as specified in ASTM E519 in order to
facilitate the construction and handling of the specimens. This reduction in size is suggested and
allowed by ASTM E519. The specimens are loaded in compression along the diagonal, and the
applied load and its corresponding vertical and horizontal deformations (along the diagonals) are
recorded. The loading causes almost diagonal cracking (vertical splitting in the testing position)
along an axis parallel to the direction of loading corresponding to a rapid drop in the load-
carrying capacity of the specimen. The force-deformation plots corresponding to the vertical and
horizontal diagonal deformations of the three tested specimens are shown in Figures 4.5 (a) and
(b), respectively. From these plots, note that the horizontal deformation (corresponding to the
crack opening of the vertical splitting cracks) is one order of magnitude higher than the vertical
deformation.
7/28/2019 Dolgu Duvar Deney Resimleri
16/40
46
0 0.01 0.02 0.03 0.04 0.05 0.06 0.070
5
10
15
20
25
30
35
40
45
Vertical deformation [in]
Verticalforce[kip
s]
Specimen 1
Specimen 2
Specimen 3
0 0.5 1 1.5
0
50
100
150
200
[kN]
[mm]
0 0.1 0.2 0.3 0.4 0.5 0.6 0.70
5
10
15
20
25
30
35
40
45
Horizontal deformation [in]
Verticalforce[kip
s]
Specimen 1
Specimen 2
Specimen 3
0 5 10 15
0
50
100
150
200
[mm]
[kN]
(a) Deformation along vertical diagonal
axis of specimen
(b) Deformation along horizontal diagonal
axis of specimen
Fig. 4.5 Diagonal force-deformation plots for masonry shear tests.
The shear strength of the masonry vf is obtained using Equation 4.2.
eff
vA
Pf = (4.2)
where P is the applied peak compressive diagonal force on the specimen and effA is the gross
sectional area of the specimen along its diagonal direction calculated as th2 where h and t
are the side length and thickness of the square specimen, respectively. The applied peak
compressive force and its corresponding shear strength for the three specimens as well as the
mean value (about 11% of the masonry compressive strength) and COV are presented in Table
4.5. The test setup and a typical failure mode are shown in Figure 4.6.
Table 4.5 Masonry shear test results.
SpecimenPeak compressive
load [kips (kN)]
Shear strength
[psi (MPa)]
1 44.2 (197) 283 (1.95)2 41.1 (183) 263 (1.81)
3 38.0 (169) 243 (1.68)
Mean 41.1 (183) 263 (1.81)
COV 7.6%
7/28/2019 Dolgu Duvar Deney Resimleri
17/40
47
(a) Test setup (b) Typical failure mode
Fig. 4.6 Masonry diagonal tension (shear) test.
4.1.5 Masonry Bending Test
To determine the tensile strength of the masonry assembly, a bending test on 2 -5"2-5" (75
cm75 cm) specimen is performed. The test setup is such that the middle third of the span of the
specimen is subjected to pure bending moment (i.e., no shear) as shown in Figure 4.7(a).
Assuming an elastic-brittle behavior for masonry in tension, the tensile strength of the masonry
assembly can be calculated from Equation 4.3:
6/
6/2tb
LP
S
Mft == (4.3)
where is the applied bending moment, S is the section modulus, P is the total applied peak
vertical load, L is the span, and b and t are the width and thickness of the specimen,
respectively. The total applied peak vertical load recorded during the test is 978=P lbs (4.35
kN) which corresponds to 5.69=tf psi (479 kPa) representing only 3% of the masonry
compressive strength and 26% of its shear strength. This relatively low value, compared to those
in more homogeneous materials, such as concrete, is attributed to the mode of failure of themasonry composite (two-phase) material (Loureno 1996), in Figure 4.7(b), which is dominated
by a single vertical crack along the weak plane of the mortar-brick interface.
7/28/2019 Dolgu Duvar Deney Resimleri
18/40
48
(a) Test setup (b) Typical failure mode
Fig. 4.7 Masonry bending test.
4.2 SNAP-BACK TESTS
Pull-back (snap-back) tests are performed on the test structure before and after the URM infill
wall construction to determine the stiffness, natural frequency, and damping ratio of the
structural system before starting the shake-table experiments. These tests are separately
conducted for both the longitudinal (north-south) and transverse (east-west) directions of the test
structure. It is to be noted that the torsional response of this symmetric test structure is not of
interest; therefore, asymmetric snap-back tests of the longitudinal and transverse directions are
not considered in this study. For each test, the structure is pulled in one direction by applying 3
8 kips (1336 kN) lateral force, depending on the stiffness of the test structure, using lever hoist
(come-along), and then released suddenly to allow free vibration. The floor acceleration and
displacements are measured during both the loading (pulling) phase and the free vibration phase
of the test. The force-displacement results of the pull test are used to obtain an estimate of the
stiffness of the test structure. The floor acceleration responses during the free vibration after
releasing the pulling force, both in the time and frequency domains, are analyzed and used to
estimate the natural period of vibration of the test structure and the corresponding damping ratio.A typical configuration and sample test results of the snap-back test is shown in Figure 4.8. The
results in this figure refer to the second snap-back test in the north-south and east-west directions
after building the URM infill wall, post-tensioning of the columns, and installation of additional
mass on the RC slab. The complete results of the snap-back tests are presented in Appendix B.
7/28/2019 Dolgu Duvar Deney Resimleri
19/40
70 4. Experimental tests on FRP strengthened masonry wall panels
also includes a comparison of the reinforcement schemes, and comparisons of the
results with other tests from the literature.
4.2 Experimental programThe Diagonal Tension/Shear Test involves subjecting a square section of ma-
sonry, with height and length both equal to 1.2 m, to a compressive load applied
along the diagonal. A schematic of the test is shown in Figure 4.1a. A photograph
of the test is shown in Figure 4.1b.
(a) Test schematic (b) Setup URM-1
Figure 4.1: The Diagonal Tension/Shear Test
The panels were constructed from solid clay masonry units with nominal di-
mensions 230 mm long, 110 mm wide and 76 mm high. Five batches of mortar
were used in the construction of the panels, all having a mix ratio of 1:1:6 (ce-
ment:lime:sand by volume). The mortar joints were 10 mm thick. These are the
same material specifications as used for the pull tests presented in Chapter 3.
The mortar batches used to construct each panel are presented in Table 4.1.
The flexural tensile bond strength of each mortar batch was determined using the
bond wrench test, AS3700-2001, Standards Australia (2001c). The bond wrench
test is described in further detail in Section 5.3.1. The average flexural tensile bondstrength (coefficient of variation in brackets) of each mortar batch is also presen-
7/28/2019 Dolgu Duvar Deney Resimleri
20/40
122 5. Finite element modelling
Table 5.1: Summary of all bond wrench results
Test Mortar batch Bond Strength (MPa) COV (%)Compression test 1:1:6 mortar 1.22 31
(Section 5.3.2) Han (2008) specimens
(1:1:6 + air entrainer) 0.176 -
Torsion test Series 1 (1:1:6 + air entrainer) 0.14 26
(Section 5.3.3) Series 2 (1:1:6) 1.74 11
Pull tests 1 1.84 23
(Section 3.2.1) 2 1.73 22
3 1.22 31
Wall Panel tests 1 1.25 51
(Section 4.2) 1+W 0.65 34
2 0.49 37
2+W 0.29 46
3 0.47 47
3+W 0.31 57
4 0.57 48
5 1.26 32
5+W 0.41 59
5.3.2 Compression tests on masonry prismsCompression tests on 7-brick high masonry prisms were used to determine the
elastic properties of the brick unit and mortar joint, as well as the compressive
strength of the masonry. Five prisms were constructed using the same clay brick
and mortar specification used to construct the pull test specimens and the wall
panels tested in diagonal tension/shear. The flexural bond strength of these speci-
mens was 1.22 MPa (COV 31%), determined using AS 3700 bond wrench test (Stan-
dards Australia, 2001c). This value was approximately equal to the bond strengths
of the strongest panels tested (URM-1, URM-2 and H4A - see Table 4.1 in Chap-
ter 4).
The compression specimens were constructed and tested in accordance with
AS3700 Appendix C (Standards Australia, 2001b). Specimens were constructed 7
bricks high to achieve a height-to-thickness ratio greater than 5 to minimise theinfluence of platen restraint. A photograph of the test is shown in Fig. 5.7.
Potentiometers were placed on both sides of the specimen to measure the dis-
placement across a mortar joint and across 3 bricks to calculate the strain in the
mortar joint and masonry respectively (as recommended by Drysdale et al. (1994)).
Potentiometers were not used to measure the displacement within a single brick
unit because they were not sensitive enough to measure the small brick displa-
cement. Potentiometers were mounted onto brackets that were screwed onto the
specimen at fine target points to allow the gauge lengths for displacement measu-
rement to be determined accurately.
To improve the determination of the elastic modulus from the compressiontest each specimen was loaded and then unloaded three times before being loa-
7/28/2019 Dolgu Duvar Deney Resimleri
21/40
5.3 Compression tests on masonry prisms 123
Figure 5.7: Compression test setup
ded to failure. Specimens were loaded to approximately 40 % of their predicted
peak load, before unloading, to capture the elastic loading range and minimise
non-recoverable damage. Specimen 1 was loaded to 200 kN before unloading (ba-
sed on an estimate ofPc = 500 kN); specimens 2-4 were loaded to 260 kN before
unloading (40% ofPc specimen 1); and specimen 5 was loaded to 300 kN before
unloading (approximately 40% of average ofPc for first four specimens). The dis-
placements recorded from the second, third and final load cycles were averaged
and used in the calculations to determine the elastic modulus values of the ma-
sonry and the mortar (displacements recorded from the first load cycle were igno-
red). All of the compression tests were stopped once the ultimate load was reachedto avoid damaging the potentiometers.
All of the specimens failed by crushing in the mortar joint and vertical cracking
through the front and back faces of the brick units. The ultimate load (Pc) and cor-
responding maximum compressive stress (fc), masonry strain at fc, and the elastic
modulus of the mortar (Emor) and masonry (Emas) are shown in Table 5.2. The
elastic modulii of the mortar (Emor) and masonry (Emas) were determined as the
gradients of the compressive stress-strain curves (for mortar and masonry respec-
tively) between 5 and 33% of the maximum compressive strength (Drysdale et al.,
1994).
The average elastic modulus of a single brick unit Eunit was determined indi-rectly using the average values ofEmor and Emas and by considering compatibility
7/28/2019 Dolgu Duvar Deney Resimleri
22/40
124 5. Finite element modelling
of displacements between masonry, brick unit and mortar joint. This calculation
was required because brick unit displacement (used to calculate strain) was not re-corded. The total masonry displacement is equal to the sum of the displacement
of the units and mortar. The masonry displacement across 3 bricks and 3 mortar
joint is equal to:
mas= 3unit+3mor (5.6)
The displacements are calculated using:
mas=PLmas
EmasA(5.7)
unit=PLunitEunitA
(5.8)
mas=PLmor
EmorA(5.9)
Where P=compression load, A=bedded area of prism, Lmas=258 mm, Lunit=76
mm, Lmor=10 mm. By substituting Equations 5.7, 5.8, and 5.9 into Equation 5.6,
Eunit was determined as 27592 MPa.
Table 5.2: Compression test results (bond strength = 1.22 MPa)
Specimen Pc (kN) fc (MPa) Masonry strain at fc Emor (MPa) Emas (MPa)1 664.88 26.28 0.0013 4801 17698
2 651.73 25.76 0.0030 8047 18895
3 970.51 38.36 0.0027 5067 17909
4 894.86 35.37 0.0025 2650 18157
5 873.10 34.51 0.0023 4854 18415
Average 811.12 32.06 0.0025 5084 18215
The average shear modulus values of the brick unit (Gunit) and mortar (Gmor)
were calculated as 11497 MPa and 2118 MPa respectively, using Equation 5.10 and
Equation 5.11. A Poissons ratio (
) equal to 0.2 was adopted for both the brick unitand the mortar (Loureno, 1996a).
Gunit=Eunit
2(1+)(5.10)
Gmor=Emor
2(1+)(5.11)
The experimentally determined elastic properties of the brick unit and mortar
joint were valid for the actual dimensions of the unit and the joint. As expanded
units and zero-thickness mortar joints were used in the FE model, adjustments
to the elastic properties were required to achieve an equivalent overall elastic res-ponse. A method that alters the elastic properties of the interface elements and
7/28/2019 Dolgu Duvar Deney Resimleri
23/40
5.3 Compression tests on masonry prisms 125
leaves the enlarged unit properties untouched is described in Rots (1997) and Lou-
reno (1996a). The normal elastic stiffness (kn) and the shear elastic stiffness (ks)of the mortar joint interface element were altered using Equation 5.12 and Equa-
tion 5.13, respectively, where hmor = thickness of mortar joint = 10 mm. The nor-
mal elastic stiffness (kn) was calculated as 623 N/mm3 and the shear elastic stiff-
ness (ks) was calculated as 260 N/mm3.
kn=EunitEmor
hmor(EunitEmor)(5.12)
ks=GunitGmor
hmor(GunitGmor)(5.13)
In addition to the maximum compressive stress (fc), the equivalent plastic re-
lative displacement (p) and the compressive fracture energy (Gc) were also re-
quired to model compression failure. The equivalent plastic relative displacement
(p) was calculated using Equation 5.14 as 0.024 mm in order to obtain a total ma-
sonry strain of 0.25% at fc (Table 5.2) (Loureno, 1996a). In Equation 5.14 hunit is
the height of the brick unit = 76 mm.
p=
0.0025fc
1
Eunit+
1
kn(hunit+hmor)
fc (5.14)
As each compression test was stopped just after the ultimate load was reached the
compressive fracture energy was not recorded. The compressive fracture energy
was estimated as 25 N/mm using Equation 5.15 (Loureno, 1996a).
Gc= 15+0.43fc0.0036f2c (5.15)
To estimate the elastic and compression properties for masonry panels with a
weaker bond strength (the average bond strength for some of the panels tested was
as low as 0.29 MPa) the results of Han (2008) were used. Han tested five masonry
prisms constructed using a similar clay brick (as the current investigation), but a
weaker mortar was used. This mortar consisted of cement:lime:sand in propor-
tions of 1:1:6 by volume with eight times the recommended dose of air entrainingagent added to deliberately create low bond strength. The bond strength of these
specimens was 0.176 MPa. The average values of the ultimate load (Pc), maximum
compressive stress (fc), and the elastic modulus of the mortar (Emor), masonry
(Emas) and brick unit(Eunit) are shown in Table 5.3. The masonry strain at fc was
not reported.
Table 5.3: Compression test average results from Han (2008) (bond strength = 0.176
MPa)
Pc (kN) fc (MPa) Emor (MPa) Emas (MPa) Eunit (MPa)
516 20.0 2772 18135 35360
7/28/2019 Dolgu Duvar Deney Resimleri
24/40
126 5. Finite element modelling
From these tests the values of fc = 20 MPa and Emor = 2772 MPa were adop-
ted to represent masonry with a bond strength of 0.176 MPa. For consistency, theelastic modulus of the brick unit (Eunit) from the previously described test series
(equal to 27592 MPa) was kept. The input properties required for the mortar joint
interface elements were calculated the same way as described previously, and are
shown in Table 5.4. For the calculation ofp, the masonry strain at fc was assu-
med as 0.2% (Loureno, 1996a). The input properties required for the mortar joint
interface elements determined for masonry with a bond strength of 1.22 MPa, are
also shown in Table 5.4.
Table 5.4: FE model input properties determined from compression tests
Property Bond strength = 0.176 MPa Bond strength = 1.22 MPakn (N/mm
3) 308 623
ks (N/mm3) 128 260
fc (MPa) 20 32
Gc (N/mm) 22 25
p (mm) 0.010 0.024
5.3.3 Torsion testTo characterise the shear behaviour of the mortar joint the torsion test (shown
in Figure 5.8), developed by Masia et al. (2006, 2007), was used. In this test anannular masonry specimen, which contains a single bed joint (Figure 5.8a), is sub-
jected to combined compressive stress (normal to the bed joint) and torsion. The
torsion test produces close to uniform distributions of normal and shear stress,
thus allowing the shear behaviour at a point to be characterised.
As part of the current investigation a set of torsion tests were performed on
specimens constructed using the same brick and mortar as the wall panel tests.
The specimens were prepared by coring a complete annular specimen through the
height of a pre-cast masonry couplet. These specimens were prepared differently
from previous torsion tests, reported in Masia et al. (2007). In Masia et al.s tests the
specimens were prepared by coring annular sections from two separate solid units
first, and then bonding them together with mortar. After testing, and then analy-
sing the results of the current investigation (joints cast before coring) it was found
that the joint shear strengths were lower than expected when compared to joints
that were cast after coring (as in Masia et al. (2007)). The reduced joint strength
was thought to be caused by damage to the joint during the coring procedure. The
results from the current investigation were unreliable and therefore were not used
for the characterisation of the shear behaviour. The results of tests conducted by
Masia et al. (2007) were used instead.
Torsion tests by Masia et al. (2007)
This section outlines the specimens tested by Masia et al., the testing proce-dure they used, and their results. Torsion tests were performed on specimens
7/28/2019 Dolgu Duvar Deney Resimleri
25/40
37
Figure 3.4 Triplet Test setup
Figure 3.5 Shear stress vs. Lateral compressive stress graph (Average shear
stresses)
7/28/2019 Dolgu Duvar Deney Resimleri
26/40159
CHAPTER 6
PANEL TESTS
6.1 GENERALBefore the frame tests, two series of panel tests were conducted to obtain
information about the behavior of the strengthened masonry walls. The information
gathered in panel tests were used to model the frame tests in analytical evaluation.
In these tests, square masonry walls having dimensions of 700 700 mm and width
of 69 mm were loaded in diagonal direction.
Test set-up was prepared between two heavy concrete support blocks. Test
specimen was placed on thin metal plates parallel to floor. Steel plate was oiled and
sat on ball roller supports to ensure friction free movement of panel specimens.
Steel heads were placed to corners of the wall specimen in the diagonal direction
and were attached with gypsum. Dial gages were placed in six directions to measure
displacements on the wall. Test set-up is illustrated in Figures 6.1 and 6.2.
Figure 6.1 Test Set-up of Panel Tests
7/28/2019 Dolgu Duvar Deney Resimleri
27/40160
Figure 6.2 General View of Panel Tests
6.2 PANEL TESTS6.2.1 First Series Panel Tests
In the scope of the first series, 12 tests were conducted. First, 6 reference wall
specimens, composed of 3 non-plastered and 3 plastered, were tested. Then, 6
plastered wall specimens strengthened in different ways were tested. Plastered wall
specimens were produced of 10 mm plaster thickness on both sides. 10 mm
thickness of mortar with 2% volumetric ratio of steel fibers was applied on one side
of 3 specimens. To the remaining 3 specimens, 20 mm thickness of mortar with 2%
volumetric ratio of steel fibers was applied again on one side. Specimen properties
are given in Table 6.1.
Mix proportions of the mortar used for the first series brick laying are presented in
Table 6.2 and mix proportions of the mortar used for plastering are given in Table
6.3.Mix proportions for 1 m3 of the mortar with steel fibers applied on the plaster of
the first series panel specimens are shown in Table 6.4.
7/28/2019 Dolgu Duvar Deney Resimleri
28/40176
Figure 6.31 Views of SF2P-1
Figure 6.32 Views of SF2P-2
Figure 6.33 Views of SF2P-3
Figure 6.34 View of 2SNPP-1 Figure 6.35 View of 2SNPP-2
7/28/2019 Dolgu Duvar Deney Resimleri
29/40178
Figure 6.44 View of 2SSF1P-3 Figure 6.45 View of 2SSF2P-1
Figure 6.46 Views of 2SSF2P-2
Figure 6.47 View of 2SSF1PD-1 Figure 6.48 View of 2SSF1PD-2
Figure 6.49 View of 2SSF1PD-3 Figure 6.50 View of 2SSF2PD-1
7/28/2019 Dolgu Duvar Deney Resimleri
30/40
Figure5:Bondwrenchtestapparatusandjointfailure
2.5.Masonrytensilebondstrength(Splittingtest)Becausethefailureofthewallettesalsoinvolvedsomeverticalsplitting,forlatermodellingpurposes,
itwas
also
useful
to
determine
the
transverse
strength
of
the
masonry
composite.
This
was
achieved
using thesplitting testreportedbyAli (7),seeFigure6,performedonspecimenswhichwerebuilt
accordingtoreference(7)andwerecuredinairinthelaboratory.Usingthisprocedure,thetransverse
strengthisgivenby:
T
CF
Dt
where/ 4
hlD
andhandlarethespecimenheightandwidth,respectively.Inaddition,tdenotes
the
specimen
thickness;F
is
the
applied
load
andC
a
constant
of
0.648.
This
constant
depends
on
brick/jointstiffnessandthechosenvaluewasbasedonmoduliofelasticityratioofbrickandmortar,
Eb/Em,ofapproximately2,seealso(7).
Using this approach, the mean transverse tensile strength of the five specimens was found tobe
0.62MPawithacoefficientofvariationof22.4%.DetailedresultsaregiveninTableA4.
Figure6:Tensilebondtestapparatusandsplittingfailure
7/28/2019 Dolgu Duvar Deney Resimleri
31/40
TableA7 inAppendixA shows the specimendimensions, failure loadsaswellas the compressive
strengthandmodulusofelasticity(asasecantmodulusattheloadof30%oftheultimate)forallfour
specimens.ThemeanvalueforthemodulusofelasticityEmobtainedfromtestsonspecimens1and4
was6.81GPa.Themeanvalueforthemasonrycompressivestrengthobtainedfromthetestswasonly
ahalfof thatobtained from theprism tests inaccordancewith theAustraliancode (seeTableA6).
Apartfromanysizeeffects,itisalsopossiblethatthespecimensmayhavebeendamagedduringthe
cuttingoutprocess.The failuremodesof these specimens (Figure9)alsodiffered from thatof the
masonrytripletsindicatingthatthetypeofspecimenmayhavealsoplayedarole.
Figure9:Compressiontestoncutoutspecimenanditsfailure
7/28/2019 Dolgu Duvar Deney Resimleri
32/40
The objectives of these small-scale tests were (1) to understand the impact of a thin layer
of ECC on unreinforced masonry, (2) to examine the performance of different ECC
retrofit schemes, and (3) to help develop a retrofit scheme for unreinforced masonry
infills in non-ductile reinforced concrete frames.
The small-scale tests were compression tests, flexural tests and triplet (shear) tests
(Figure 3.2). Compression tests of masonry prisms were conducted representing the
compression strut of a masonry infill under in-plane lateral loading. Flexural tests of
brick beams using a quarter point bending configuration with the constant moment region
intended to approximately represent direct tension (in particular in the ECC), were
performed to investigate the approximate response of tension struts in the masonry infill.
Triplet specimens with ECC in the joints between the bricks were tested in shear to
evaluate the ECC-brick bond in shear. With the small-scale tests different reinforcement
ratios, as well as ECC-masonry bonding techniques were examined.
(a) (b) (c)
Figure 3.2. Schematic of the small-scale test set-ups (a) compression test, (b) flexural
test, and (c) triplet test.
3.2. Compression Experiments
Masonry prisms with and without retrofit were fabricated and tested in compression. The
procedure followed for the fabrication of the specimens, the design of the different
______________________________________________________________________________________Chapter 3 Small-Scale Tests
40
7/28/2019 Dolgu Duvar Deney Resimleri
33/40
research had triple-wythe masonry infills and the intention was to use a 39 mm (1.5 in.)
ECC layer for its retrofit. A thickness of 39 mm (1.5 in.) is the maximum thickness that
the sprayable ECC was reported to reach when sprayed on a vertical surface (Kim et al.,
2003) at the time these experiments were designed.
(a) (b) (c) (d) (e) (f) (g)Figure 3.3. Schematic of the different variables tested, a) Tall plain specimen, b) Short
plain specimen, c) ECC retrofit, d) ECC retrofit with stitch dowels, e) lightly reinforced
ECC, f) lightly reinforced ECC with stitch dowels, and g) heavily reinforced ECC with
stitch dowels. All dimensions are in mm.
3.2.2. Fabrication of Masonry Prisms
All specimens tested in compression had four mortar joints no thicker than 13 mm (0.5
in.) each. As indicated in Figure 3.3, the specimens of group (a) were taller than those of
groups (b) through (g). The tall plain prisms had a height of approximately 343 mm (13.5
in.). The specimens of groups (b) through (g) had a height of approximately 267 mm
(10.5 in.) with the top and bottom bricks being cut down to 20 mm (0.8 in.) in thickness.
The height of the short specimens was controlled by the maximum specimen height that
the Forney compression tester at Stanford could accommodate when modified to give the
full compressive stress-strain response of the specimen. However, due to limitations in
the free rotation of the loading plates of the compression tester when modified, all
specimens were tested in the Powell laboratory, at The University of California, San
______________________________________________________________________________________Chapter 3 Small-Scale Tests
42
7/28/2019 Dolgu Duvar Deney Resimleri
34/40
Diego.
A professional mason from a local masonry company (Walton & Sons Masonry Inc.),
was hired to build the brick specimens in order to ensure real practice conditions (Figure
3.4). All fabrication was performed in one day in the laboratory of the John A. Blume
Earthquake Engineering Center at Stanford University.
200 mm200 mm
Figure 3.4. Fabrication of masonry specimens for compression tests.
The materials used were:
Yellow clay bricks 94 mm x 58 mm x 196 mm (3.7 in. x 2.3 in. x 7.7 in.), grade
MW, type FBS, and manufactured to meet ASTM C216-10.
Mortar consisting of 1 part cement (Type I/II), 1 part lime (Type S) and 5 parts
sand (Oly 1) by volume. The above types of cement, lime and sand were
recommended by the masonry company. This type of mortar is similar to Type
N which uses a 1:1:6 mix and results in a mortar with low compressive strength
(ASTM C270-10).
The type of bricks and mortar used were recommended by the Professional Advisory
Panel (PAP) of the project to represent the mechanical properties of the materials used for
the construction of masonry infills of a building that served as the project's prototype
______________________________________________________________________________________Chapter 3 Small-Scale Tests
43
7/28/2019 Dolgu Duvar Deney Resimleri
35/40
Table 3.13. ECC-brick interface shear strength of TEB specimens.
Joints are named based on Figure 3.79.
TEB specimen Joint 1 Joint 2 f's ECC-brick
1 1' 2 2' MPa (psi )1 R R* S S* 2.08 (302)
2 R R* S S* 2.38 (346)
3 S* S R* R 1.72 (250)
4 R R* S S* 2.00 (289)
5 R R* S S* 2.08 (302)
Average f 's ECC-brick = 2.05 (298)
S.D. f's ECC-brick = 0.21 (31)
* toweled surface
94 mm
58 mm
94 mm
58 mm
Figure 3.80. TE triplet specimen: Brick-ECC interface failure
94 mm
58 mm
94 mm
58 mm
3.81. TEB triplet specimen: Brick shear failure
______________________________________________________________________________________Chapter 3 Small-Scale Tests
128
7/28/2019 Dolgu Duvar Deney Resimleri
36/40
42
Figure 2.32: equipment set up for vertical compression tests: the steel
profiles fitted onto the Metrocom 3000 kN press (left), a wall ready to be
crashed inside the press (right)
Figure 2.33: diagonal compression tests set up, steel supports for
diagonal compression tests (left), wall ready to be crashed inside the
press (right)
Figure 2.34: suspension system of the upper steel profile
7/28/2019 Dolgu Duvar Deney Resimleri
37/40
47
Figure 2.45: equipment for compression test parallel to holes on hollowbrick walls (front and back view of the panel)
Figure 2.46: equipment for compression test orthogonal to holes on
hollow brick walls (front and back view of the panel)
Figure 2.47: equipment for diagonal on hollow brick walls (front and
back view of the panel)
7/28/2019 Dolgu Duvar Deney Resimleri
38/40
48
Figure 2.48: equipment for compression test parallel to holes on half-full
brick walls (front and back view of the panel)
Figure 2.49: equipment for compression test orthogonal to holes on half-
full brick walls (front and back view of the panel)
Figure 2.50: equipment for diagonal compression test half-full brick
walls (front and back view of the panel)
7/28/2019 Dolgu Duvar Deney Resimleri
39/40
Plugging this raw data into the formula, along with the measurements and parameters as defined
by the equation, the tensile strengths are found, shown in Table 2.
Table 2: Bond Wrench Flexural Strength Results
Gross Area flexural Tensile Strength (psi)
PRISM
6(H + H)
I$
( + )
I
Tensile Strength
1 142.4 7.2 135.2 psi
12 109.2 5.8 103.4 psi
15 72.2 4.1 84.8 psi
17 162.6 8.1 154.5 psi
26 93.06 5.1 88.0 psi
28 113.5 6.0 107.5 psi
average 112.2 psi
Taking the average of these values, neglecting the values of Prisms 8 and 15, yield an average of
112.2 psi for the tensile strength of the mortar joints. However, there is a wide range of values,
anywhere from 84 psi to 154 psi, which indicates that the accuracy of these values is not certain.
3. SHEAR TEST
a. MethodsThe second major material testing conducted is the shear test in which a triplet of bricks aremade with the center brick protruding on the top, as shown in Fig. 5.
Prior to being tested, the triplets are capped using hydrostone and following ASTM 1552-07standards (Standard Practice for Capping Concrete Masonry Units). This is done so that the two
bottom prisms are flat against the floor
and the top of the prism is perfectly flat.
To ensure the top is flat, a level is used inall directions to find the best possible fit of
the capping with the triplets.
However, it was found that the friction
forces between the hydrostone capping
and the bottom plate impacted the resultsof the testing. As a result, a steel plate was
placed under the hydrostone, along with a
small roller. The roller was used to eliminate the friction forces at the bottom of the triplets, andthe steel plate, which is hot glued directly onto the hydrostone, prevented the roller from digging
in and crushing the hydrostone capping. This set up is shown in a close up view on Fig. 7.
Figure 5: Shear Test Setup
7/28/2019 Dolgu Duvar Deney Resimleri
40/40
The triplets are then tested in the MTS 500k machine where the
central brick is subjected to a downward force while the entire
prism has a horizontal load at various stresses. An in-housedesign for the test arrangement was used, as there are no
ASTM Standards regarding this shear test. Following some
tweaking to the apparatus and the construction, the shear test asused in the experimented is shown in Fig. 6.
Stresses exerted on the mortar joints are 50 psi, 100 psi, 150psi, and 200 psi. Using the formula:
= (2)
the forces to be exerted in compression on the prisms are
determined as shown in Table 3. This axial compressive stressis generated by tightening screws on plates surrounding the
triplets, using an external load cell (shown in blue in Fig. 6)to measure the amount of force placed axially. Levels areused to ensure that the plates and the prisms are both as close
to perpendicular with the base plate as possible. Wooden
blocks, shown in Fig. 6, are placed on the sides as safetyprecautions, preventing the metal plates or the prisms from
hitting the MTS machine upon the sudden failure of the
mortar joint. In almost all the triplet tests, only one side of themortar joint failed while the other remained intact.
Table 3: Shear Test Axial Forces
Stress (psi) Force (kips)50 1.172
100 2.344
150 3.416
200 4.688
The results from this test are recorded through a data acquisition system which records both axialand shear forces. The shear displacements are recorded using a pair of pots located on the center
brick, as shown in Fig. 7 in addition to the recording of the distance being pushed downward
onto the prism. This is then post-processed and plotted to form coherent and readable graphs.
b. ResultsThe results of the shear data is in forces and displacements; this is then converted to shear vs.
displacement, as shown in Fig. 8.
Figure 6: Shear Testing
Figure 7: Shear Test, a close up
view of the plates and the rollers