Content uploaded by Marco Corradi
Author content
All content in this area was uploaded by Marco Corradi on Nov 15, 2019
Content may be subject to copyright.
DETERMINATION OF SHEAR STRENGTH OF MASONRY PANELS
THROUGH DIFFERENT TESTS
Antonio Borri, Giulio Castori, Marco Corradi*
University of Perugia, Dept. of Engineering Via Duranti, 93 - 06125 Perugia, Italy
* corresponding author: mcorradi@strutture.unipg.it
KEYWORDS: Masonry panels, shear tests, shear strength.
ABSTRACT
This paper addresses the problem of evaluation of strength of masonry walls. In-plane
behaviour of masonry panels has been studied under monotonic diagonal-compression and
shear-compression loading in quasi-static test facility. The results of 35 laboratory and in-situ
tests are analysed to show that in the case of the diagonal compression test results are lower
than the strength of masonry walls evaluated trough the shear-compression test, highlighting
the problem of choosing the test which best simulates to the real behavior of the masonry
when stressed by lateral loads. A presentation is also given of the results of a F.E.
investigation for shear strength evaluation of masonry walls. F.E. modeling non-linear
procedure was used for the representation of masonry panels. The numerical simulations are
compared with experimental results and the reliability of the different finite element models is
discussed, thus confirming the different shear strength values measured in the experimental
campaign.
INTRODUCTION
Uncertainties are inherent in all engineering projects and particularly for rehabilitation and
reinforcement of existing masonry structures due to design assumptions and modeling
methods. As far as material properties are concerned, these can be specified during the design
phase and uncertainties arise due to discrepancies between these assumed values and those
corresponding to the properties of the materials that are actually found or used. Historic
masonry buildings in urban centres have not been conceived to resist seismic loads and an
important mechanic parameter is the masonry shear strength. Site investigations can typically
offer estimates of shear characteristics and researchers recently developed different testing
methods in order to calculate the shear parameters of masonry panels.
Investigations of construction failures during the recent earthquakes, such as the 2009 Aquila
and 2012 Emilia show that low masonry shear strength are among the major causes of failure
[1-3]. Many difficulties could be eliminated if better technical information regarding the
mechanical characteristics of masonry walls are available. In their calculations structural
engineers and technicians have often referred to not well-identified parameters for different
kinds of masonry walls found in scarce bibliography studies. Solid brick and multiple-leaf
stone masonry walls are commonly encountered in historic buildings dating back to ancient
times and up to the first half of 20th century. The classification and the analysis of historical
masonry typologies were conducted in the past with different purposes in mind [4-5].
However these contributions have rarely included an experimental part regarding the
mechanical characteristics of masonry due to the uncertainty in the determination of
mechanical characteristic from in-situ tests. Masonry walls have been classified with regard to
the constituting materials, section dimensions, bonding patterns and mortar types, but very
rarely with regard to their mechanical properties.
When studying literature with respect to shear tests on masonry, different types of test are
found. In this paper two types of test will be distinguished, that are characterized by the way
in which the load is applied.
The first type is the shear compression test. A masonry wall panel with bed joints in
horizontal direction is supported at the lower and at the upper sides. It is loaded in–plane by a
horizontal force placed at mid-span. It was first performed in-site by Turnšek and Sheppard
[6] in Slovenia. Several shear-compression tests were carried out on panels from buildings in
the city of Ljubljana. The compression stress was equal to that effectively hanging over the
panels, but not completely well-defined. In the recent past in Italy many historical
constructions have been tested: Chiostrini and Vignoli [7-8] applied the shear compression
test on some historical buildings in Tuscany fixing the compression stress using oil jacks
positioned over the panels. Other experimental campaigns were carried out by Corradi et al.
[9-11], Valluzzi et al. [12], Luciano and Sacco [13], Costa et al. [14].
The second type is the diagonal compression test. The loading is applied by means of a
compression force only and the bed joints are at an angle with the loading direction. The
diagonal compression test is clearly defined by ASTM [15] and RILEM [16] specifications, to
which this experimental work refers. Several experimental campaigns were carried out in the
past, but most results refer to new masonry walls made of hollow bricks or concrete masonry
units [17-20]. With regard to historic masonry walls, during the last decades diagonal tests
were carried out on masonry panels in site or in laboratory by Valluzzi et al. [21], Gabor et al.
[22] and recently by Brignola et al. [23], Calderini et al. [24], Alecci et al. [25].
Beside variations in geometrical and material properties, mortar quality, this study aim to
validate the two test methods and to discuss and compare the results in terms of shear strength
for similar wall panels. The authors have performed a series of both analytical and
experimental studies on the mechanical characterization of historic masonry walls, and most
of the results have either been published or are currently under consideration for publication.
TEST SETUP AND PROCEDURES
(1) Diagonal compression test
The diagonal compression test was designed in order to evaluate the shear strength and the
shear elastic modulus of the masonry. The laboratory procedure is normalized by ASTM and
RILEM standards. The two codes show some differences both in the interpretation of test
results and in the evaluation of the mechanical properties from the experimental test. Both
testing procedures involved rotation of the URM wall panel by 45° and vertical loading along
one of the wall’s diagonals.
Two groups of masonry specimens were tested. They had two different dimensions: 1200 x
1200 mm and 500 x 500 mm with different thicknesses. For masonry specimens tested in-site
(all 1200x1200 mm), the panel remained anchored to the rest of masonry wall through a part
of the 700 mm of the lower horizontal edge. The remaining three edges and a part of the
fourth were cut and isolated from the rest of the masonry wall.
Fig. 1 shows the schematic and scene picture of the test setup in this experimental program.
The load is monotonically applied by a 1000kN-hydraulic actuator (50 kN- for 500x500 mm
panels) positioned on the corners of the panels between an H shaped metal element and a
loading plate, which when loaded developed tension forces in two steel rods positioned
between the steel loading plates, consequently compressing the wall panel diagonally.
4 LVDTs were used to measure the diagonal shortening and expansion on both sides of
masonry panels. All experimental data about load, time, displacements during entire test were
recorded through an electric datum acquisition system. The tests were performed with many
cycles of loading and unloading, increasing the jack action gradually until the failure of the
panel to identify the diagonal shear strength and the degradation of the shear stiffness.
According to ASTM standard, this test was introduced to simulate a pure shear stress state. In
these conditions the Mohr circle of the stress state is centered in the origins of axes.
Shear stress, Ss of masonry, equal to the principal tensile stress
I, at applied load P was
determined by using the following equation:
n
Is AP
S707.0
(1)
in which An is the net cross-horizontal section of the panel, determined as the average of the
width and height of the specimen multiplied by its thickness. From the diagonal compression
test it is possible to determine the shear modulus G. In the experimental analysis the angular
strain was evaluated:
gHV
(2)
where:
V = diagonal shortening,
H= diagonal extension, g= gage length
Fig. 2 represents the stress states defined by ASTM and Rilem Standards with the Mohr’s
circle representation. According to ASTM Standard, the centre of Mohr’s circle is in the
origin of the axis, so the maximum principal (tensile) stress I is equal to both principal
compression III and shear stress xy.
The RILEM interpretation of the test results is obtained modeling the masonry panel as if it is
an isotropic and homogeneous material and running a linear elastic analysis: the stress state at
the centre of the specimen is not a pure shear state:
n
xy
n
yx A
P
A
P05.156.0
(3)
According to this interpretation, it is possible to evaluate the tensile strength ft of masonry by:
n
tA
P
f5.0
(4)
and using the Turnšek and Cacovic [26] formulation, the shear strength
D0
from a diagonal-
compression test is given by:
n
t
DA
P
f
35.1
0
(5)
The compressive and tensile strain values have been calculated from the relative displacement
between two control points in each diagonal (gage length).
(2) Shear-compression tests
The shear strength is evaluated here with the shear-compression test as the average shear
stress in a panel subjected to a vertical compression and to a horizontal load in its plane. The
investigation was carried out on two types of masonry panels: large panels measured 1800
mm in height and 900 mm in width while small panels tested only in laboratory measured
1000 mm in height and 500 mm in width.
Fig. 3 shows the in-site test set-up. Large specimens were placed in the test set-up and firstly
subjected to the desired level of compression which was kept constant during the test. The
level of compression applied, corresponding to approx. 20-40% of the estimated masonry
compressive strength, was 0.20÷0.30 MPa (typical for a two or three-storey building). The
axial load was applied by means of two 500kN hydraulic jacks placed between a support
frame and an upper spreader beam. Applied loads and corresponding displacements were
recorded with a frequency of 2 Hz.
All the tests were performed under force control (monotonic up to the point of failure), using
a 1000 kN hydraulic jack, at a load increment rate approximately equal to 0.25 kN/s.
The presence of the apparatus overhanging the panel was not enough to constitute a perfect
constraint (Fig. 4). The upper half of the panel was able to translate and rotate while the lower
half, connected to the rest of the masonry, could be considered as a perfect constraint. This
caused a lack of symmetry in shear distribution between the upper and lower halves of the
panel, which was taken into account during the elaboration of the data. As consequence of
this lack of symmetry, the lower half of the panel resulted always more stressed and the
failure always occurred here. Sixteen displacement transducers were adopted: eight W50 (50
mm of maximum deformation) were applied on the main façades of the wall to record the
diagonal displacements (4 shortenings and 4 extensions), respectively, whereas 6 other
transducers W50 were placed along each side of one vertical edge (at the base, the center
point, the top of the panel) and two more transducers were placed on one side to measure
vertical movements on the edge of one side of the panel and eventual rotations at the top of
the panel.
In order to evaluate the shear strength 0T of the masonry, the well-known Turnšek and
Cacovic formulation is assumed:
t
tfb
f
Dt
T0
max 1
(6)
where:
Tt
f0
5.1
(7)
and ft represents the tensile strength of masonry and b is a parameter which was assumed to be
dependent on the panel aspect ratio H/D (H=height of the panel, D=width of the panel) and
accounts for the distribution of shear stress.
0
is the vertical compression stress and T is the
maximum shear load in the lower half of the panel.
max
is the shear strength affected by the
vertical compression stress
0
according to the Turnšek formulation. The parameter b takes
into account the variability and distribution of the shear stresses at the center of the wall and
was assumed equal to 1.5 in the Turnšek and Cacovic formulation. Recently in the new Italian
Seismic Code [27], according to the formulation proposed by Benedetti and Tomasevic [28],
this parameter with aspect ratio H/D=1 has been assumed equal to 1.
EXPERIMENTAL WORK
(1) Test matrix
The evaluation of the shear strength of URM panels was the main thrust of this study. The
total number of specimens is thirty-five, nineteen of which were manufactured in laboratory,
and sixteen large panels were cut from six existing buildings. The in-site tests were carried
out on historic constructions located in central Italy while the laboratory tests on both large
and small panels were conducted at the Lastru laboratory of the University of Perugia located
in Terni.
The test matrix of shear-compression and diagonal compression tests was based on the
following panels:
1. In-site testing: a) 5 full-scale tests carried out at the Farnetta building (2 diagonal
compression tests and 3 shear compression tests); b) 3 full-scale tests carried out at the
Belfiore building (2 diagonal compression tests and 1 shear compression test) and one
full-scale test carried out at the Vescia building (shear compression test); c) 2 full-
scale tests carried out at the Ponte Postignano building (1 diagonal compression test
and 1 shear compression test); d) 2 full-scale tests at the San Felice building (1
diagonal compression test and 1 shear compression test); e) 4 full-scale tests at the
Colle Umberto building (2 diagonal compression test and 2 shear compression test).
2. Laboratory testing: a) 10 tests on large panels (3 diagonal compression tests and 7
shear compression tests); b) 9 tests on small panels (4 diagonal compression tests and
5 shear compression tests).
All panels are identified by a four index code, in which the first indicates the type of test
(CD=diagonal compression, TC=shear-compression); the second identification number of the
panel; the third location of the structure from which the panels were obtained (B=Belfiore,
V=Vescia, F=Farnetta, P=Ponte , U=Colle Umberto, S=San Felice, L=Laboratory), while the
fourth index indicates the type of intervention carried out (in this case the fourth index is
always OR because this paper reports only the results on un-strengthened panels, with the
exception of tests identified by codes V-T-07-IN in which the strengthening technique using
preventive injections resulted as not effective).
(2) Description of the panels tested in-situ
Considering the importance of the type of masonry assemblage a brief description of the
buildings where the panels were cut is reported in this section. The panels were cut from 6
buildings located in Farnetta, Vescia, Belfiore, Ponte, Colle Umberto and San Felice. A photo
of the masonry pattern of each building is reported in Fig. 5.
The mortars of all buildings are rather weak and all lime-based in consideration to the absence
of portlandite and of silicates of calcium and aluminum (with the exception of the poor
cement based mortar of Ponte building) . The chemical analysis shows that the main
differences are in relation to the period of construction of the buildings: the Ponte mortar has
a high weight ratio cement/aggregates. The other buildings, constructed before Ponte, have a
smaller value of this ratio and the mortars have small quartz traces removed by the erosive
action of water. The walls of five buildings (Vescia, Ponte, Farnetta, Colle Umberto and
Belfiore) are made of barely cut calcareous white- and pink-color stones (stone rubble
masonry). The dimensions of the stones vary for the different buildings from which the
panels were cut. Larger stones were present in Belfiore and Vescia and (average dimension of
the longest edge equal to 300 mm) while smaller calcareous stones constituted the panels at
Farnetta, Colle Umberto and Ponte (average dimension of the longest edge equal to 200-250
mm).
The characteristics of the two types of stone were obtained on cylindrical specimens 70 mm
in diameter and 150 mm in height, cored from irregular cut stones. The compressive strength
was 57.5 MPa for pink color calcareous stone and 36.0 for white-color calcareous stones. The
weight density of these stones is sufficiently constant and average values equal to 2330 kg/m3
for the pink color one and 2485 kg/m3 for the white color one were measured.
The results show a significant scattering in the data of the compression tests carried out on the
sponge travertine of the building of Ponte. This depends on the high inhomogeneity of the
stone due to the presence of large and frequent voids. The average values of the weight
density and of the compression strength are respectively equal to 1335 kg/m3 e 2.66 MPa.
The first building located in the countryside of the village of Farnetta in Umbria was
constructed at the beginning of the 20th century as a rural house (farm). The walls of the two
storeys farm are constructed from URM, with a timber framed trussed roof; the walls are
made with a two leaf masonry of irregular white calcareous stones with weak connections and
a thickness of 480 mm.
The buildings of Belfiore and Vescia are two stories high. The masonry texture of these
buildings is very similar and made of stone double-leaf walls with double solid brick courses
interposed at intervals of 800-1200 mm. The two external leaves, approximately 180-240 mm
thick each, consisted of rough-shaped calcareous limestone white- and pink-colored blocks
bonded in sub-horizontal courses. The double-leaf walls had a thickness of 480 mm.
Connecting stones are not present.
The Ponte building is three floors high and was constructed just before 1955 to serve as a
residence. The masonry assemblage is made with irregularly cut stones with a maximum
thickness of 480 mm and the wall is a double leaf masonry made of white- and pink-
limestones and sponge travertine. Connecting stones between the leave are not present.
The stone masonry building of Colle Umberto is an abandoned rural farm located in the north
of Umbria. Colle Umberto farm is a double-gabled building. The building was probably built
in the 18-19th century and enlarged during the 20th century. White-calcareous stone masonry
walls are characterized by two thinness’s: oldest masonry walls are about 600 mm thick and
20th century walls are 480 mm thick.
The building of San Felice is located in the Italian region of Emilia. The building is outside
the city walls and it is from the 18th century. Its layout is quite simple, as a two-storey
building. Solid clay bricks and broad mortar joints (average height 10-15 mm) were used for
the construction of this building. The URM walls of the buildings consist of multi-leaf clay
brick masonry constructed in common bond. The average thickness of the wall section of the
building is approximately 300 mm. The exterior solid brick walls are two single leaf URM
walls connected by transverse bricks (2-4 bricks/m2). The interior and exterior of the
perimeter walls and the internal walls are plastered with a weak 10 mm to 20 mm thick lime-
sand plaster.
(3) Description of the panels built in laboratory
The laboratory investigation was carried out on panels of small and large dimensions. Large
specimens were made of solid clay bricks (dimensions: 240x120x55 mm, bulk density: 970
kg/m3) or barely cut calcareous pink stones (Fig. 6). All panels were constructed by
experienced masons following traditional construction procedures. Clay solid bricks were
supplied by a local Manufacturer (Fornaci briziarelli Marsciano S.p.A). In all specimens the
joints (bed and head) consisted of a general purpose masonry lime-based mortar and were
made approximately 10 mm thick.
The mean compressive and flexural strengths of the clay-brick masonry units and mortar are
shown in Tab. 1. Mortar used for the masonry panels had the following mix composition:
sand/ inorganic binder weight ratio =3.0 and water/binder weight ratio = 0.85. The binder is
produced by Colacem S.p.A. and its commercial name is “Calce Idrata Colacem”.
(4) Test results
Large panels
Twenty-six (10 in laboratory, 16 in site) large panels failed due a shear-friction crack along a
diagonal. The results of in-site tests show significantly different values depending on the type
of test carried out on the masonry panels. Considering only the double-leaf walls in the
building of Belfiore shear strengths were 0.0705 MPa and 0.0337 MPa respectively for the
shear-compression test and the diagonal compression test (Tab. 2).
Similar results were obtained for the panels tested in Ponte: a shear strength 0T of 0.0837
MPa was measured for panel 15, submitted to a shear-compression test, and a shear strength
0D of 0.0270 MPa for panel 13 (diagonal compression test). For these buildings the results of
shear-compression tests are about 170% higher than the results of diagonal compression tests
(respectively +127% and +210% higher).
The shear stress, calculated from Eq. (5 ) and (6), vs angular strain response, is compared in
Figs. 7-9. Fig. 7 shows the shear stress – angular strain response for two similar panels of
Colle Umberto while Figs. 8 and 9 shows the same comparison respectively at San Felice and
Belfiore/Vescia buildings. It must be noted that the shear stress max from the shear-
compression tests is affected by the vertical compression stress 0 applied on the top of the
panel. Using Eq. (7) it is possible to calculate the shear strength 0T and compare it with the
shear strength 0D . Tab. 3 shows the ratio between the results of shear strength obtained from
the two test methods. Working with eight couples of results (16 shear tests) related to the five
above-mentioned buildings of Belfiore and Vescia, Colle Umberto, San Felice and Farnetta,
the ratio r=
0T /
0D varied from 1.13 to 3.10 (average value 2.26). Surely, due to the very few
number of tests carried out, the above quoted correlation must be investigated by a bigger
number of tests. However, the emerging line seems quite correct and hence they allow the
following considerations.
With regard of shear-compression tests carried out in laboratory on masonry stone panels, the
maximum shear strength 0T of the walls varied between 0.036 and 0.104 MPa, but evident
cracks pattern already started at a stress level varying from 0.025 to 0.071 MPa. The average
value of shear strength 0T of these panels was 0.073 MPa (Tab. 4).
Only one panel with this masonry texture (double-leaf roughly cut stones) was tested in
laboratory in diagonal compression (CD-08-L-OR) and a shear strength 0D of 0.020 MPa
was measured. The cracks had a diagonal pattern, located on both panel surfaces and in the
transverse sections. The ratio r of these tests is 3.67. Higher values of shear strength 0D were
detected for the two solid brick panels (shear strength 0.043 and 0.053 MPa), whereas shear
strengths 0T of brick panels tested in shear-compression were much higher, varying from
0.094 to 0.144 MPa.
Tab. 5 shows the comparison between the results of diagonal- and shear-compression tests:
the ratio r is always higher than 1. For solid bricks panels it was 2.44.
Small panels
Small specimens were tested in laboratory according to both test methods. For a reliable
correlation study with the prototype, one of the most important considerations is the
appropriate modelling as per similitude relations. However the aim of this part of the
experimental work was not to simulate the behavior of large panels, but to compare the results
of shear strength of small specimens. For practical considerations, the brick geometric,
modulus and density ratio are assumed unity.
Shear tests on small-scale panels, having nominal dimensions of 500x500x120 mm (diagonal
compression tests) (Figs. 10a and 11a) revealed an average shear strength 0D of 0.0354 MPa
(Tab. 6).
The panels tested in shear-compression consist of single-leaf clay brick masonry constructed
in common bond (horizontal courses) (Figs. 10b and 11b) nor in a masonry bond pattern with
vertical courses (Fig. 10c). These tests were performed in order to remove the influence of
masonry bond pattern from results. From these tests it appears evident that the results are
connected to both the type of test and the bond pattern (Fig. 12). Identical panels (for
dimensions and mix design) have been tested and different results have been found. The shear
compression tests lead to different values of strength 0T when different bond patterns are
used (0T = 0.0897 MPa (bricks arranged in horizontal courses), 0T=0.0375 MPa (vertical
courses)). Tab. 7 shows the comparison between the results of diagonal- and shear-
compression tests: the ratio r=
0T /
0D is always higher than 1 and varied between 1.06 and
2.53.
Discussion
Once it is assumed that a ratio exists between the results of the two shear tests, we face the
problem of choosing the one more representative of the real behavior of masonry walls
stressed by horizontal loads typical of seismic actions. Diagonal compression tests allow the
panel a free deformation, since its four sides are free from any kind of constraints, with the
exception of the small portion of masonry that permits the connection between the panel and
the rest of the masonry wall. Numerical calculations demonstrated its un-influence and the
panel can be considered completely un-constrained. This situation may be assumed to be
representative of masonry spandrels in which the vertical compression stress 0 may be
considered equal to zero and the effect of confinement is very limited.
On the contrary, during the shear-compression test, the two square halves resulting from the
division of the panels in two parts have therefore a common edge. This causes an effect of
confinement from one half to the other. These are also constrained by the presence, on the
upper part of the panel, of the apparatus overhanging the panel (steel plates, jacks, rods) and,
by the bottom one, of the remaining masonry constituting the wall.
The most common seismic verifications for buildings, constituted of 2-3 floors with walls
characterized by a low slenderness ratio, assume the vertical masonry elements between
adjacent openings as infinitely stiff. In analogy with the shear-compression tests, the failure of
these elements occurs when the shear strength is not able to absorb the seismic loads. The
strains along the vertical edges of masonry elements are free, while an effect of confinement
is produced by the remaining overhanging part of the masonry wall (Turnšek and Sheppard
[6], Chiostrini et al. [7-8], Lagomarsino and Giovinazzi [29]). This situation is typical for
masonry piers or heavily confined load-bearing walls in which the absence of openings
prevents any kind of masonry deformation.
A behavior such as this is easy to verify from the analysis of damages to constructions struck
by the earthquake, in which the failure condition occurs when the tensile strength in the center
of the masonry panel is achieved.
One other important aspect of the problem concerns the bond pattern of the panels, which
affects the shear strength. Fig. 13 shows two different bond patterns: it is clear that the shear
strength of the second wall is bigger. Experiments on brick panels also demonstrate that the
shear strength from shear-compression test method is higher compared to the one from
diagonal compression-test. The ratio r assumed high values varying between 2.44 and 2.53 for
brick panels arranged in a common (horizontal courses) bond pattern. The collapse
mechanism of two brick panels is shown in Fig. 14: following the rupture in the centre of the
panel, the contribution of friction during the test is different for the two test methods. The
slide mechanism in the shear compression test may contribute to increase the shear load
while during a diagonal compression test the diagonal crack causes the separation of the two
parts of the panel without significant friction contribution.
NUMERICAL ANALISYS
In this paragraph non-linear static analysis has been carried out on the 3D model of the
masonry panels tested in accordance with ASTM E-519 [15] (diagonal compression) and
shear compression. A three-dimensional finite element model has been developed to account
for geometric and material nonlinear behavior of masonry panels and was used to model the
shear tests on the masonry panels. The numerical simulation was carried out using ANSYS
code ver. 14.5; the panels were modeled using isoparametric elements (SOLID65) with the
hypothesis of plane stress. Panel for diagonal compression and shear-compression tests are
characterized by the dimensions respectively of 1200x1200x400 mm and 1800x900x400 mm.
A compression stress 0 of 0.30 MPa was applied over the panels tested in shear-compression.
A typical finite element discretization of the masonry panels and steel apparatus used in the
present study is shown in Fig. 15. Structural steel sections (spreader beam, C-shaped
elements, loading plate) were modeled as an elastic material. The panels tested in diagonal
compression and shear-compression was respectively modeled with 24x24x8 and 18x36x8
elements. A smeared model with homogenized properties has been used. The masonry is
modelled as a isotropic continuum.
Values concerning the physical properties of masonry material have been established on
statistical analysis of test data found in the literature (Corradi et al. [9]). The average elastic
Young ES modulus, own weight W, Poisson ratio and tensile strength fwt of masonry values
used in the FE analyses were: ES = 1000, 2000 and 3000 MPa, W=17.5 kN/m3, = 0.25,
fwt=0.02, 0.03, 0.04, 0.05 and 0.06 MPa.
The finite element model developed herein has been used to analyze the nonlinear response
of masonry is highly discontinuous due to cracking. Fig. 16 reports the tensional results and
displacements for both types of panels. Fig. 17 reports the crushing and the cracking path
obtained in the non-linear analysis due to increasing shear load. The analyses carry out on the
non-linear 3D F.E. model has also permitted to evaluate the shear strength capacity of the two
test types. The predicted shear strength 0 vs. masonry tensile strength fwt response for three
values of masonry Young modulus Es of 1000, 2000 and 3000 MPa is shown in Fig. 18. In
this figure two theoretical predictions are shown for each test method. It is evident from Fig.
19 that the ratio r=
0T /
0D is always higher than 1. In the case of ES=1000 MPa, the ratio
value varied between 1.23 and 1.89 and all the values were higher than 1.
CONCLUSIONS
An experimental research on the shear behavior of 35 masonry panels tested in diagonal- and
shear-compression has been presented. Based on the results obtained from the experimental
program, the following conclusions can be stated:
1. The diagonal compression and the shear-compression tests were carried out on the same
type of masonry. This allowed to identify a significant differentiation both in site and in
laboratory between the results obtained from the two test methods. It was noted that the ratio
r=
0T /
0D between the results of shear strength for the two tests is always higher than 1,
highlighting the problem of choosing the test which best simulates to the real behavior of the
masonry when stressed by lateral loads.
2. The experimental results of laboratory tests carried out on small and large panels made of
solid bricks confirmed that the type of test has an influence on the shear strength. The
difference between shear results from the two test methods seems to be even bigger for brick
panels arranged in horizontal courses compared to rubble stone panels.
3. The laboratory experimental campaign also analyzed the influence of bond pattern for
small solid-brick panels highlighting the importance of this parameter. Bricks panels arranged
in vertical courses exhibited a lower shear strength compared to the same panels arranged in
standard horizontal course. The ratio r generally assumed high values varying between 2.44
and 2.53 for brick panels arranged in a standard bond pattern.
4. FE analysis of the shear tests has shown that the ratio r=
0T /
0D is always bigger than 1
ranging between 1.23 and 1.89. For low masonry tensile strength the ratio assumed the
highest values. Despite tests results, the value of this ratio never exceeded 1.89 for all FE
analyses carried out.
5. Ongoing and new developments are focused on the influence of masonry assemblage and
of the panel geometrical and mechanical characteristics on the formulation for the prediction
of the shear strength measured with two test methods.
ACKNOWLEDGEMENTS
The experimental program was carried out with the help of Giulio Befani and Fabrizio
Cristofari, Undergraduate Research Assistants, and Alessandro de Maria, Alessio Molinari
and Romina Sisti, Graduate Research Assistants. The authors are grateful for their
contributions.
REFERENCES
[1] Carocci, C.F. 2012. Small centres damaged by 2009 L’Aquila earthquake: on site analyses
of historical masonry aggregates, Bull Earthquake Eng,10:45–71.
[2] Ural, A., Dogangun, A., Sezen, H. and Angın, Z. 2012. Seismic performance of masonry
buildings during the 2007 Bala, Turkey earthquakes, Nat Hazards 60:1013–1026.
[3] Furukawa, A. and Ohta, Y. 2009. Failure process of masonry buildings during earthquake
and associated casualty risk evaluation, Nat Hazards, 49:25–51.
[4] Borri. A. and De Maria, A. 2009. Indice di qualità muraria: applicazioni nell’ambito delle
NTC 2008 [in Italian], L'Edilizia, Ed. De Lettera, 46-54.
[5] Binda, L., Saisi, A. and Tiraboschi, C. 2000. Investigation procedures for the diagnosis of
historic masonries, Construction and Building Materials 14, 199-233.
[6] Turnšek, V. and Sheppard, P.F. 1980. The shear and flexural resistance of masonry walls,
Proc. Research Conference on Earthquake Eng., Skopje, Macedonia, Sept. 8–13, 517–573
[7] Chiostrini, S. and Vignoli, A. 1994. In-situ determination of the strength properties of
masonry walls by destructive shear and compression tests. Masonry Int.;7(3):87–96.
[8] Chiostrini, S., Galano. L. and Vignoli, A. 2000. On the determination of strength of
ancient masonry walls via experimental tests. Proc. 12th World Conference on Earthquake
Engineering, cd-rom, Aukland, New Zealand, January 30–February 4, paper No. 2564.
[9] Corradi, M., Borri, A. and Vignoli, A. 2003. Experimental study on the determination of
strength of masonry walls. Journal of Construction and Building Material, 17: 325–37.
[10] Corradi, M., Tedeschi, C., Binda, L. and Borri, A. 2008. Experimental evaluation of
shear and compression strength of masonry wall before and after reinforcement: Deep
repointing. Journal of Construction and Building Material, 22: 463–472.
[11] Borri, A., Castori, G., Corradi, M. and Speranzini, E. 2011. Shear behavior of
unreinforced and reinforced masonry panels subjected to in situ diagonal compression tests.
Constr Build Mater; 25:4403–14.
[12] Valluzzi, M. R., Tinazzi, D. and Modena, C. 2002. Shear behavior of masonry panels
strengthened by FRP laminates, Journal of Construction and Building Materials, 16:409–416.
[13] Luciano, R. and Sacco, E. 1998. Damage of masonry panels reinforced by FRP sheets.
Int. J. Solids and Structures, Elsevier Science Ldt, 35-15, 1723-1741.
[14] Costa, A., Arêde, A., Costa, A., Oliveira, C. 2011. In situ cyclic tests on existing stone
masonry walls and strengthening solutions, Earthquake Engineering & Structural Dynamics,
40, 4, 449–471.
[15] ASTM E 519-81 1981, Standard Test Method for Diagonal Tension (Shear) in Masonry
Assemblages, American Society for Testing Materials.
[16] RILEM TC. 1994, 76-LUM. Diagonal tensile strength tests of small wall specimens,
1991. In RILEM, Recommendations for the Testing and Use of Constructions Materials.
London: E& FN SPON, 488–489.
[17] Correa, M.R.S., Page, A.W., Han, Y. and Masia, M. 2006. Experimental and theoretical
study of the shear behaviour of URM walls with openings. Masonry Int;19(1): 1–14.
[18] Tomazevic, M., Lutman, M. and Petkovic, L.J. 1996. Seismic behaviour of masonry
walls: experimental simulation. J Struct Eng – ASCE;122(9):1040–7.
[19] Sing-Sang, P., Totoev, Y. and Page, A.W. 2008. Earthquake shear load in load-bearing
masonry buildings. Australian J Struct Eng;8(2):171–89.
[20] Roca, P. and Araiza, G. 2010. Shear response of brick masonry small assemblages
strengthened with bonded FRP laminates for in-plane reinforcement. Construction and
Building Materials 24, 1372–1384.
[21] Valluzzi, M. R., Da Porto, F. and Modena, C. 2004. Behavior and modeling of
strengthened three-leaf stone masonry walls. Materials and Structures / Matériaux et
Constructions, 37, 184-192.
[22] Gabor, A., Bennani, A., Jacquelin, E. and Lebon, F. 2006. Modelling approaches of the
in-plane shear behavior of unreinforced and FRP strengthened masonry panels, Composite
Structures, 74: 277–288.
[23] Brignola, A., Frumento, S., Lagomarsino, S. and Podestà, S. 2009. Identification of Shear
Parameters of Masonry Panels Through the In-Situ Diagonal compression test, International
Journal of Architectural Heritage, 3: 1, 52 -73.
[24] Calderini, C., Cattari, S. and Lagomarsino, S. 2010. The use of the diagonal compression
test to identify the shear mechanical parameters of masonry. Constructions and Building
Materlials, 24, 677-685.
[25] Alecci, V., Fagone, M. and Rotunno, T. 2013. De Stefano M. Shear strength of brick
masonry walls assembled with different types of mortar, Construction and Building Materials
40, 1038–1045.
[26] Turnšek, V. and Cacovic, F. 1971. Some experimental results on the strength of brick
masonry walls. Proc. of the 2nd International Brick Masonry Conference, Stoke-on-Trent,
149-156.
[27] Italian Seismic Code, D.M. 14.01.08, G.U. No.9 – 04.02.08 [in Italian] 2008.
[28] Benedetti, D. and Tomasevic, M. 1984. Sulla verifica sismica di costruzioni in muratura.
Ingegneria sismica, 1(0), 9-16 [in Italian].
[29] Lagomarsino, S. and Giovinazzi, S. 2006. Macroseismic and mechanical models for the
vulnerability and damage assessment of current buildings. Bull Earthq Eng 2006;4:415-43
Tab. 1: Results from mechanical characterization of bricks and lime-based mortar
for masonry panels tested in lab.
Sample size
Compression
stress
(MPa)
Flexural
Strength
(MPa)
Solid clay bricks
30
20.99
7.39
Mortar
21
0.549
-
42
-
0.281
Tab. 2: Results from the diagonal compression and the shear-compression tests
(masonry panels tested in site)
Test No.
Panel
dimensions
(mm)
Masonry
texture
Failure
Load
(kN)
Compression
stress 0
(MPa)
Shear
Strength 0
(MPa)
Shear
Strength SS
(MPa)
CD-03-F-OR
1200x1190x480
1
37.0
-
0.0215
0.046
CD-04-F-OR
1200x1200x480
1
37.9
-
0.0219
0.047
CD-01-B-OR
1200x1220x480
1A
58.8
-
0.0337
0.072
CD-13-P-OR
1230x1220x480
1
47.7
-
0.0270
0.059
CD-02-U-OR
1170x1180x480
1
31.2
-
0.0188
0.039
CD-06-U-OR
1190x1200x600
1
44.1
-
0.0209
0.043
CD-09-S-OR
1180x1200x280
2
19.6
-
0.0195
0.041
TC-01-F-OR
860x480x1820
1
34.3*
0.147
0.0250
-
TC-02-F-OR
863x480x1800
1
37.0*
0.184
0.0241
-
TC-05-F-OR
900x480x1800
1
62.5*
0.183
0.0531
-
TC-04-B-OR
880x480x1830
1A
88.3*
0.308
0.0705
-
TC-07-V-IN
930x480x1830
1A
100.5*
0.287
0.0823
-
TC-15-P-OR
880x480x1820
1
74.4*
0.122
0.0837
-
TC-16-U-OR
890x480x1820
1
36.1
0.100
0.0320
-
TC-17-U-OR
900x600x1810
1
36.9
0.100
0.0430
-
TC-18-S-OR
880x280x1890
2
40.6
0.200
0.0604
-
Masonry textures: 1: double-leaf stone panel, 1A: double-leaf stone masonry with two solid brick courses at
intervals of 800-1200 mm, 2: solid bricks; CD=diagonal compression, TC=shear-compression, B=Belfiore,
V=Vescia, F=Farnetta, P=Ponte , U=Colle Umberto, S=San Felice, L=Laboratory.
Tab. 3: Comparison of the results from the diagonal compression and the shear-compression
tests
(masonry panels tested in site)
Test No.
Test type
Texture
r=
0T /
0D
CD-01-B-OR
Diagonal compression
Double-leaf rubble stone masonry with two solid
brick courses at intervals of 800-1200 mm
2.27
TC-04-B-OR
Shear-compression
TC-07-V-IN
Shear-compression
CD-13-P-OR
Diagonal compression
Double-leaf rubble stone masonry
3.10
TC-15-P-OR
Shear-compression
CD-03-F-OR
Diagonal compression
CD-04-F-OR
Diagonal compression
TC-01-F-OR
Shear-compression
Double-leaf rubble stone masonry
1.13
TC-02-F-OR
Shear-compression
TC-05-F-OR
Shear-compression
CD-02-U-OR
Diagonal compression
Double-leaf rubble stone masonry
1.70
TC-16-U-OR
Shear-compression
CD-02-U-OR
Diagonal compression
Double-leaf rubble stone masonry
2.29
TC-17-U-OR
Shear-compression
CD-09-S-OR
Diagonal compression
Double-leaf solid brick masonry
3.10
TC-18-S-OR
Shear-compression
Tab. 4: Results of laboratory tests carried out on masonry panels.
Test No.
Panel
dimensions
(mm)
Masonry
texture
Failure
Load
(kN)
Compression
stress 0
(MPa)
Shear
Strength 0
(MPa)
Shear
Strength SS
(MPa)
CD-08-L-OR
1200x1200x480
1
34.8
-
0.020
0.044
TC-35-L-OR
900x510x1805
1
109.3
0.208
0.104
-
TC-36-L-OR
900x490x1810
1
52.0
0.208
0.036
-
TC-37-L-OR
900x510x1805
1
80.7
0.188
0.071
-
TC-39-L-OR
900x486x1900
1
87.6
0.209
0.083
-
CD-20-L-OR
1190x1200x245
2
38.1
-
0.043
0.090
CD-21-L-OR
1190x1200x245
2
46.5
-
0.053
0.117
TC-22-L-OR
890x1810x245
2
84.1
0.482
0.144
-
TC-42-L-OR
900x1790x250
2
61.3
0.397
0.094
-
TC-44-L-OR
925x1800x250
2
70.8
0.386
0.113
-
Masonry textures: 1: double-leaf stone panel, 2: solid bricks
Tab.5: Comparison of the results from the diagonal compression and the shear-compression
tests
(masonry panels tested in laboratory)
Test No.
Test type
Texture
r2=
0T /
0D
CD-08-L-OR
Diagonal compression
TC-35-L-OR
Shear-compression
TC-36-L-OR
Shear-compression
Double-leaf rubble stone masonry
3.67
TC-37-L-OR
Shear-compression
TC-39-L-OR
Shear-compression
Solid bricks
CD-20-L-OR
Diagonal compression
CD-21-L-OR
Diagonal compression
TC-22-L-OR
Shear-compression
2.44
TC-42-L-OR
Shear-compression
TC-44-L-OR
Shear-compression
Tab. 6: Results of laboratory tests carried out on small brick panels.
Test No.
Panel
dimensions
(mm)
Failure
Load
(kN)
Compression
stress 0
(MPa)
Shear Strength
0
(MPa)
Shear Strength SS
(MPa)
Texture
CD-70-L-OR
500x510x120
3.78
-
0.0210
0.0445
1
CD-71-L-OR
500x500x120
5.60
-
0.0311
0.0660
1
CD-72-L-OR
500x505x120
8.42
-
0.0468
0.0992
1
CD-73-L-OR
500x500x120
7.68
-
0.0427
0.0905
1
TC-80-L-OR
1000x510x120
12.81
0.33
0.0700
-
1
TC-81-L-OR
1010x500x120
16.77
0.32
0.1080
-
1
TC-82-L-OR
1000x500x120
15.42
0.35
0.0910
-
1
TC-83-L-OR
1000x510x120
8.55
0.32
0.0360
-
2
TC-84-L-OR
1010x510x120
7.76
0.23
0.0390
-
2
Bond pattern: 1=horizontal courses, 2=vertical courses.
Tab. 7: Comparison of the results from the diagonal compression and the shear-compression
tests
(small brick panels)
Test No.
Test type
Bond pattern
r=
0T /
0D
CD-70-F-OR
Diagonal compression
CD-71-F-OR
Diagonal compression
CD-72-F-OR
Diagonal compression
CD-73-F-OR
Diagonal compression
Solid bricks (bond pattern 1 for Shear-
2.53
TC-80-F-OR
Shear-compression
Compression Test)
TC-81-F-OR
Shear-compression
TC-82-F-OR
Shear-compression
CD-70-F-OR
Diagonal compression
CD-71-F-OR
Diagonal compression
CD-72-F-OR
Diagonal compression
CD-73-F-OR
Diagonal compression
Solid bricks (bond pattern 2 for Shear-
1.06
TC-83-F-OR
Shear-compression
Compression Test)
TC-84-F-OR
Shear-compression
Bond pattern: 1=horizontal courses, 2=vertical courses.
Figure 1: Typical in-plane diagonal compression test setup.
Figure 2: Interpretation of the diagonal compression test according to ASTM and RILEM, by
the Mohr’s circle representation.
900
1800
Figure 3: Typical in-plane shear-compression test setup.
Figure 4: Layout of a shear-compression test in laboratory.
(a)
(b)
(c)
(d)
(e)
Figure 5: Front view of masonry wall panels (a) Belfiore, (b) Farnetta, (c) Ponte , (d)
Colle Umberto, (e) San Felice.
Figure 6: View of some specimens during construction.
Figure 7: Shear stress – angular strain response (Colle Umberto building).
Figure 8: Shear stress – angular strain response (San Felice building).
Figure 9: Shear stress – angular strain response (Belfiore and Vescia buildings).
P
P
H/2
H/2
H
0
a) b)
H/2
H/2
H
0
c)
Figure 10: Tests on small scale brick panels: a) diagonal compression; b) and c) shear-
compression.
(a)
(b)
Figure 11: Diagonal compression (a) and shear-compression (b) tests carried out on small-
scale brick panels.
Figure 12: Shear stress – angular strain response (small panels, standard bond pattern).
1)
2)
Figure 13: Influence of different bond patterns on shear strength.
P
P
H/2
H/2
H
0
a) b)
Figure 14: Crack pattern of brick panels tested under diagonal- and shear-compression.
Figure 15: Modeling of diagonal- and shear-compression tests using the code Ansys.
Metal element
Shear load
Compression load
(a)
(b)
Figure 16: FE analysis predictions: tension.
Figure 17: FE analysis predictions: cracks and crushing.
Figure 18: FE Analysis predictions: shear strength vs. tensile strength of masonry fwt for three
values of masonry Young modulus Es of 1000, 2000 and 3000 MPa.
Figure 19: Analysis predictions: ratio r=0T / 0D vs. tensile strength of masonry fwt for three
values of masonry Young modulus Es of 1000, 2000 and 3000 MPa.
