Out-Of-Plane Behavior of Confined

Masonry Walls Subjected to Concentrated

Loads (One-Way Bending)
Dante Navarrete-Macias,a) M.EERI, Jorge Varela-Rivera,b) M.EERI, and
Luis Fernandez-Baqueiro,c) M.EERI

This paper presents the results of a study on the out-of-plane seismic behavior
of confined masonry walls. Five confined walls were tested under reverse cyclic
loads. The variables studied were the axial stress and the wall aspect ratio.
Analytical out-of-plane strength of walls was calculated considering the strengths
of the wall panel and the concrete confining elements. The former was deter-
mined using the unidirectional strut method and the latter using a plastic analysis.
It was observed that for walls with the same aspect ratio, as the axial stress
increases, the out-of-plane strength increases. For walls with the same axial stress,
as the aspect ratio increases, the strength decreases. Based on comparisons
between analytical and experimental results, it was concluded that the models
developed in this work predict accurately the out-of-plane strength of the
walls. [DOI: 10.1193/061715EQS097M]

INTRODUCTION
Confined masonry walls are commonly used in many countries of Latin America, Europe
and Asia. This type of masonry consists of an unreinforced wall panel surrounded by flexible
reinforced concrete confining elements. Typically, the width and height of the cross section
of these elements are equal to the wall thickness. In confined masonry, the wall panel is
constructed first and later the confining elements. For housing, in general, the structural sys-
tem consists of a combination of confined walls and concrete slabs. Confined walls are
required to withstand both in-plane and out-of-plane loads.
The out-of-plane strength of masonry walls can be related to two types of loads: distrib-
uted loads acting on the wall surface or concentrated loads applied at the top of the wall. In
the first case, distributed loads can be related to seismic forces due to wall self-weight or
uniform pressures due to wind forces. In the second case, concentrated loads are related
to seismic forces due to floor self-weight. For the case of masonry walls subjected to dis-
tributed loads, there are some research studies for unreinforced walls (Lawrence 1983,
Anderson 1984, Drysdale and Essawy 1988, Middleton and Drysdale 1995, Lam et al. 2003,
Griffith et al. 2004, Griffith and Vakulic 2007), reinforced walls (Button and Mayes 1992,
Abbout et al. 1996, Zhang et al. 2001), infill walls (Dawe and Seah 1989, Abrams et al. 1996,
Klingner et al. 1996, Flanagan and Bennett 1999, Henderson et al. 2003, Moghadam and

a)
Ph. D. Student, College of Engineering, Autonomous University of Yucatan. navarrete.md@hotmail.com
b)
Professor, College of Engineering, Autonomous University of Yucatan. vrivera@correo.uady.mx
c)
Professor, College of Engineering, Autonomous University of Yucatan. luis.fernandez@correo.uady.mx

2317
Earthquake Spectra, Volume 32, No. 4, pages 2317–2335, November 2016; © 2016, Earthquake Engineering Research Institute
2318 NAVARRETE-MACIAS ET AL.

Goudarzi 2010), and confined walls (Kasemi et al. 2010, Varela-Rivera et al. 2011, Varela-
Rivera et al. 2012a and 2012b, Moreno-Herrera et al. 2014, Singhal and Rai 2014). Similarly,
for the case of masonry walls subjected to concentrated loads, there are few research studies
for reinforced walls (Da Porto et al. 2010 and 2011), and infill walls (Tu et al. 2007,
Tu et al. 2010, Chen et al. 2012).
Based on these studies, it is concluded that the main variables that affect the out-of-plane
behavior of masonry walls are wall support conditions, wall aspect ratio (height over length),
wall slenderness ratio (height over thickness), axial stress, wall openings, compressive
strength of masonry, stiffness of the surrounding elements, and previous damage as a result
of in-plane loads. Another conclusion is that wall cracking pattern depends on the type of
load applied to the wall. For example, for rectangular walls with distributed loads, wall crack-
ing pattern is defined by horizontal, vertical and diagonal cracks (Moreno-Herrera et al.
2014), but for concentrated loads, only by horizontal cracks (Figure 1; Tu et al. 2007,
Tu et al. 2010, Chen et al. 2012). This out-of-plane cracking pattern is similar to that observed
on confined walls after earthquakes (Jagadish et al. 2003, Kaushik et al. 2006, Sepulveda
et al. 2008, Brzec et al. 2010). The cracking patterns divide the walls into a number of wall
segments. For uniform distributed loads, there are at least two wall segments along the length
and height of the wall, but, for concentrated loads, there is only one wall segment along the
entire height of the wall (Figure 1). This shows that for walls subjected to distributed loads,
arching action is a two-way mechanism (Moreno-Herrera et al. 2014), but, for concentrated
loads, it is a one-way mechanism (Tu et al. 2007, Tu et al. 2010, Chen et al. 2012). Another
conclusion is that the behavior of confined walls and other types of masonry walls is dif-
ferent. This is mainly associated with wall reinforcement details and construction procedures.
For example, differences between confined and infill walls are as follows: confined walls are
surrounded by flexible reinforced concrete confining elements; however, infill walls are sur-
rounded by a stiff concrete or structural steel frame. In addition, for infill walls, the frame is
constructed first and later the masonry wall panel. This type of construction causes gaps
between the frame and the wall panel. Because of these gaps, axial force is transferred
only to the frame for infill walls. Axial force is transferred to the wall panel for the case
of confined walls.
The out-of-plane strength of masonry walls has been determined using different analy-
tical methods. For example, the yielding line method has been used for unreinforced walls

Distributed loads Concentrated loads

Horizontal
crack A B

Wall Horizontal Horizontal

segments cracks cracks
Wall
Vertical segment
crack
Diagonal Section A-A’ Horizontal B’
A’ Section B-B’
crack cracks

Figure 1. Cracking pattern for walls with distributed and concentrated loads.
OUT-OF-PLANE BEHAVIOR OF CONFINED MASONRY WALLS 2319

(Drysdale and Essawy 1988, Middleton and Drysdale 1995, Lawrence and Page 1999),
reinforced walls (Zhang et al. 2001), and confined walls (Varela-Rivera et al. 2011, Varela-
Rivera et al. 2012a). The failure line method has been used for unreinforced walls (Drysdale
and Essawy 1998) and confined walls (Varela-Rivera et al. 2011, Varela-Rivera et al. 2012a).
The compressive strut method has been used for infill walls (Abrams et al. 1996, Klingner
et al. 1996, Tu et al. 2007, Tu et al. 2010) and confined walls (Varela-Rivera et al. 2011,
Varela-Rivera et al. 2012a). Finally, the spring-strut and the bidirectional strut methods have
been used for confined walls (Varela-Rivera et al. 2011, Varela-Rivera et al. 2012a and
2012b, Moreno-Herrera et al. 2014).
Based on comparisons between experimental and analytical out-of-plane strengths car-
ried by different authors (Varela-Rivera et al. 2011, Varela-Rivera et al. 2012a and 2012b,
Moreno-Herrera et al. 2014), it was concluded that the bidirectional strut method predicts
more accurately the out-of-plane strength of confined walls. This method is an iterative pro-
cedure based on two-way arching action. It is assumed that after the formation of a given
cracking pattern the wall is divided into segments. As the wall segments rotate, compressive
strut forces are developed in the wall segments. The strength is determined by the out-of-
plane equilibrium between the external loads and the compressive strut forces. This method
considers two types of failure of the unreinforced panel: crushing of masonry and snap
through failure of wall segments. In the first case, the compressive forces increase and even-
tually reach the axial compressive strength of masonry. In the second case, the out-of-plane
displacements increase and the wall segments eventually collapse. The above method was
developed for confined walls subjected to out-of-plane uniform distributed loads. For the
case of concentrated loads, this method needs to be modified to take into account the for-
mation of a single wall segment along the wall height and that the corresponding arching
action is only in one direction.
The objective of this paper is to study the out-of-plane seismic behavior of confined walls
subjected to concentrated loads (one-way bending). As far as the authors know, there is no
previous research on confined walls subjected to this type of load in the literature. Results of
five confined masonry walls subject to out-of-plane concentrated loads are presented. The
variables studied were the axial stress and the wall aspect ratio. The out-of-plane strength of
the walls was calculated as the contribution of the strengths of the wall panel and the concrete
confining elements. The former was determined using the unidirectional strut method,
which is a modified version of the bidirectional strut method. The latter was calculated
using a plastic analysis. Analytical out-of-plane strengths are compared with corresponding
experimental values.

EXPERIMENTAL PROGRAM
Five full-scale confined walls were considered in this study (walls M1 to M5). The geo-
metry of walls M1 to M3 was 3.49
2.55 m (length
height), 2.45
2.55 m for wall M4,
and 1.41
2.55 m for wall M5. Walls were constructed using hollow clay bricks with nom-
inal dimensions of 0.12
0.12
0.25 m (thickness
height
length). Bricks were
obtained from a single batch. Walls were constructed in half running bond by a qualified
worker. Mortar used in construction was in proportion by volume 1∶2∶7 (Portland cement:
lime: sand). Mortar was placed on both the face shells and the head joints. Average thickness
2320 NAVARRETE-MACIAS ET AL.

Top CE Vertical CE Bottom CE

LR 4#3
TR #2@200 mm
LR 8#3
TR #2@120 mm

LR 4#3
Cross section dimensions in mm
TR #2@150 mm

Figure 2. Cross-sectional dimensions and steel reinforcement details for confining elements.

of the mortar joint was equal to 10 mm. Cross-section dimensions and steel reinforcement
details of confining elements (CE) for the walls are presented in Figure 2. Longitudinal rein-
forcement (LR) consisted of deformed steel bars with nominal yield strength of 412 MPa.
Transverse reinforcement (TR) consisted of plain steel bars with nominal yield strength of
228 MPa. All requirements used for the confining elements were based on those minimum
specified in the Mexico City Masonry Technical Norm (MTN°2004).
Walls M1, M2, and M3, with an aspect ratio of 0.73, were tested using axial stresses of
0.11, 0.23, and 0.35 MPa, respectively. Walls M4 and M5, with aspect ratios of 1.04 and
1.80, respectively, were tested using an axial stress of 0.23 MPa. Axial stresses of 0.11,
0.23, and 0.35 MPa correspond to interior walls of one, two, and three stories high, respec-
tively. Each wall was subjected to constant axial stress and reverse cyclic out-of-plane con-
centrated loads until failure. Axial stress was applied first and later the concentrated loads.
The wall test setup was divided into three parts: an axial load system, a support system, and
an out-of-plane load system. The axial load system was formed by two swivel beams, two
simple supported spreader beams, four steel threaded bars and two hydraulic actuators with
capacity of 117 kN (Figure 3a). Axial stress was maintained constant by using a mechanical
“load maintainer” (Edison 1994). The support system consisted of a reinforced concrete
slab, two structural steel frames, and four steel wheels (Figure 3b). The concrete slab
was connected along the length of the wall by using steel threaded bars. Wheels were placed
between the concrete slab and the structural frames. This support system allows the out-of-
plane horizontal displacement and restricts the rotation at the top of the wall. Walls were
attached to the laboratory strong floor by using concrete blocks and steel threaded bars.
Details of support system are presented in Figure 3b. The out-of-plane load system consisted
of a reaction steel frame, a two-way hydraulic actuator and a connection steel beam
(Figure 4).
Axial load was measured using four donut load cells with capacities of 44 and 111 kN.
Out-of-plane load was measured using a pin load cell with a capacity of 400 kN. Axial and
out-of-plane loads were verified by using pressure transducers with capacities of 69 and
34.5 MPa, respectively. Out-of plane horizontal displacements at top of the wall were
OUT-OF-PLANE BEHAVIOR OF CONFINED MASONRY WALLS 2321

Steel
Plan View Plan View frame

Threaded Concrete
Swivel Steel
bar slab
beam wheels

Spreader Concrete Wall

beam blocks
Threaded Steel
Concrete wheels
bar Steel
slab
frame
Lateral View Lateral View
Swivel Concrete slab
beam
Concrete Steel
Spreader
slab wheels
beam Steel
Threaded Wall
bar frame
Wall Threaded
Hydraulic bars Threaded
actuator bars
Strong floor Strong floor Strong floor Strong floor
(a) (b)

Figure 3. Test setup: (a) axial load system and (b) support system.

Connection
steell bbeam C
Concrete
slab

Reaction
steel frame Hydraulic
actuator
Wall

Strong floor Strong floor

Figure 4. Out-of-plane load system.

measured using 127 and 381 mm linear potentiometers. Relative displacements between the
concrete slab and the wall, the concrete slab and the steel frames, and the wall and the strong
floor were measured using 25 mm and 50 mm linear potentiometers. Strain gauges were
attached at the top and bottom parts of the longitudinal reinforcement of vertical confining
elements. A schematic view of wall instrumentation is presented in Figure 5. Overall views of
wall M3 before testing is presented in Figure 6. The out-of-plane loading history used for the
walls consisted of six initial reverse cycles controlled by load and subsequent cycles
2322 NAVARRETE-MACIAS ET AL.

Linear
i
potentiometer
Hydraulic Load
actuator cell

Pressure
transducers
Donut
load cell Hydraulic
H d li
Donut pump
load cell Hydraulic
actuator
Hydraulic
pump

Pressure L d
Load
Transducer maintainer

Figure 5. Schematic view of wall instrumentation.

Figure 6. Overall views of wall M3 before testing.

controlled by displacement (Figure 7). Walls were first loaded to the east direction and later
to the west direction. In Figure 7, V y is the lateral load associated with the yielding of the
flexural reinforcement of the vertical confining elements. Loading history was based on that
proposed in MTN (2004).
The axial compressive strength of concrete, units and masonry was determined according
to NMX-C-083-ONNCCE-2002 (National Organization for Norms and Certification for
Construction and Buildings (ONNCCE)), NMX-C-036-ONNCCE-2004 and NMX-C-
464-ONNCCE-2010, respectively. Procedures presented in these norms are similar to
those specified in corresponding ASTM standards. Samples of six specimens were consid-
ered in each case.
OUT-OF-PLANE BEHAVIOR OF CONFINED MASONRY WALLS 2323

Out-of-plane
load Control
East Drift
Load Displacements
Vy ratio
8%

Increment
6%

of 2 %
0.5 V y 4%
0.25 V y 2%

Cycles

West

Figure 7. Out-of-plane loading history.

EXPERIMENTAL RESULTS
The average axial compressive strength of concrete (f c ) was equal to 17.92, 14.92, 18.70,
15.55, and 16.51 MPa for walls M1 to M5, respectively. Corresponding coefficients of var-
iations were equal to 0.03, 0.02, 0.02, 0.06, and 0.07, respectively. The average axial com-
pressive strength of units (f p ) was equal to 18.9 MPa with a coefficient of variation of 0.14.
The average axial compressive strength (f m ) and modulus of elasticity (E m ) of masonry were
equal to 6.48 and 5,705 MPa, respectively. Corresponding coefficients of variations were
equal to 0.05 and 0.12, respectively. All values were calculated using gross properties of
corresponding cross-sections.
The behavior of the confined walls was in general similar. Horizontal cracks were
observed first in the two bottom brick courses of the west wall face and between the
two top brick courses of the east wall face, while loading to the east direction (Figure 8a).
When the load was reversed, horizontal cracks formed between the two bottom brick courses
of the east wall face and between the two top brick courses of the west wall face (Figure 8b).
These horizontal cracks, initially, did not form along the entire length of the wall. Yielding of
the flexural reinforcement was observed at the top and bottom ends of the vertical confining
elements. Yielding was observed in both loading directions. Horizontal cracks were observed
along the top and bottom parts of the vertical confining elements over a length of about
0.45 m (Figure 8c). A vertical crack, located at the wall center, was observed for walls
M1, M3, and M4 (Figure 8d). Crushing of masonry was observed at the two top corners
of the wall panel in both wall faces. Failure of walls was related to crushing of concrete
at the top and bottom ends of the vertical confining elements (Figure 8d). Testing of
walls was stopped at a maximum out-of-plane top displacement of 175 mm because the
hydraulic actuator ran out of stroke. The final cracking patterns of the west face of the
walls are presented in Figure 9. The out-of-plane load – drift ratio curves for the walls
are presented in Figure 10. Drift ratios reported were calculated using the out-of-pane dis-
placements measured at top of the walls. Out-of-plane strengths (V me ) were equal to 10.95,
2324 NAVARRETE-MACIAS ET AL.

V V
West East West East
Wall Horizontal Horizontal
crack crack

Horizontal Wall
Horizontal
crack crack

(a) (b)
Horizontal Crushing of
Crushing crack masonry
West East of masonry

Vertical Horizontal
Crushing of
confining cracks
Horizontal Vertical concrete
element crack crack
Horizontal Crushing
cracks of concrete

(c) (d)

Figure 8. Cracking propagation of walls, (a) first horizontal cracks in wall panel while loading to
the east direction, (b) horizontal cracks in wall panel while loading to the west direction, (c) hor-
izontal cracks along the vertical confining elements, and (d) vertical crack at the center of wall,
crushing of masonry and concrete.

Figure 9. Final cracking patterns of the west face of walls.

OUT-OF-PLANE BEHAVIOR OF CONFINED MASONRY WALLS 2325

15

Out-of-plane load (kN)

15 15
M2

Out-of-plane load (kN)

Out-of-plane load (kN)

M1 10 M3
10 10
5
5 5

0 0 0

-5 -5 -5

-10 -10 -10

-15 -15 -15

-10 -5 0 5 10 -10 -5 0 5 10 -10 -5 0 5 10
Drift ratio (%) Drift ratio (%) Drift ratio (%)
15

Out-of-plane load (kN)

15 M5
Out-of-plane load (kN)

M4 10
10

5 5

0 0

-5 -5

-10 -10

-15 -15
-10 -5 0 5 10 -10 -5 0 5 10
Drift ratio (%) Drift ratio (%)

Figure 10. Out-of-plane load - drift ratio curves for walls.

11.22, 14.07, 9.96, and 8.93 kN for walls M1 to M5, respectively. Corresponding drift ratios
were equal to 2.4%, 2.8%, 2.6%, 3.2%, and 3.4%, respectively.

DISCUSSION OF EXPERIMENTAL RESULTS

The final cracking pattern of walls was in general similar. Horizontal cracks were
observed first between the two top and the bottom brick courses on both wall faces. Horizontal
cracks were not formed between the top brick course and the top confining element because
concrete penetrated in the brick cells during pouring. The length of these horizontal cracks
was initially smaller than the wall length. Horizontal cracks were observed later at the top and
bottom parts of the concrete confining elements. A vertical crack was observed for walls M1,
M3, and M4 at the end of the tests. This vertical crack was related with the relative displace-
ment observed between the wall panel and the bottom horizontal confining element. This
displacement caused an out-of-plane bending of the wall panel about the vertical direction.
The out-of-plane behavior of each wall was linear elastic up to the formation of the first
horizontal cracks. After this, the behavior was nonlinear (Figure 10). This nonlinear behavior
was related to the presence of new horizontal cracks in the wall panel, horizontal cracks in
both vertical confining elements and the yielding of the flexural reinforcement of these ele-
ments. After the out-of-plane strength was reached, the wall strength decreased. This reduc-
tion was associated with the progressive crushing of masonry observed in the two top brick
courses. In addition, as the out-of-plane displacement increased, the out-of-plane component
of the compressive struts also decreased causing a reduction in the wall panel strength.
Figures 11a and 11b show the envelope curves of walls with different axial stress (M1 to
M3) and aspect ratios (M2, M4, and M5), respectively. In this figure, circular markers repre-
sent the out-of-plane strength of the walls. Figure 11a shows that, for walls with the same
2326 NAVARRETE-MACIAS ET AL.

15 15
Out-of-plane load (kN)

10 10

Out-of-plane load (kN)

5 5

0 0.11 MPa
0 0.73
1.04
0.23 MPa
-5 -5 1.80
0.35 MPa

-10 -10

-15 -15
-10 -5 0 5 10 -10 -5 0 5 10
Drift ratio (%) Drift ratio (%)

(a) (b)

Figure 11. Out-of-plane load - drift ratio curve envelopes: (a) effect of axial stress and (b) effect
of wall aspect ratio.

aspect ratio, as the axial stress increased the out-of-plane strength increased. This was related
to the additional vertical displacement restriction provided by the axial stress. This restriction
increased the compressive strut forces for a given out-of-plane displacement. The slight dif-
ference in the out-of-plane strength between walls M2 and M1 is associated with the lower
axial compressive strength of concrete of wall M2 compared with M1. The ratio between the
out-of-plane strength of wall M3 and M1 was 1.28. Figure 11b shows that, for walls with the
same axial stress, as the aspect ratio increased, the out-of-plane strength of the walls
decreased. As the wall length increased, the contribution of the wall panel strength increased
but the contribution of the confining elements remained the same. The ratio between the out-
of-plane strength of wall M2 and M5 was 1.25. Figure 11 shows that the out-of-plane strength
of the walls was associated with drift ratios between 2.4% and 3.4%. In-plane allowable drift
ratios for confined masonry walls (NTM 2004) are smaller than out-of-plane drift ratios
observed in this work; therefore, out-of-plane and in-plane strengths cannot be added.
The out-of-plane strength contribution must be calculated using the drift ratio associated
with the in-plane strength. Even though maximum wall drift ratios were greater than
5.0% (Figure 11), walls did not collapsed. Vertical confining elements provided enough
out-of-plane support for the walls to avoid snap through of wall segments.

ANALYTICAL MODELS
The analytical out-of-plane strength of the walls was calculated as the contribution of the
strengths of the wall panel and the vertical confining elements. The wall panel strength was
determined using the unidirectional strut method, which is a modified version of the bidir-
ectional strut method (Moreno-Herrera et al. 2014). This last method was developed for uni-
formly distributed loads instead of concentrated loads as those studied in this work. The
bidirectional strut method was modified to consider: (1) the formation of a one-way arching
action mechanism instead of a two-way mechanism and (2) that only one wall segment was
formed along the wall height instead of two.
OUT-OF-PLANE BEHAVIOR OF CONFINED MASONRY WALLS 2327

Frame
A element B V
V

Sectiion A-A’

Sectioon B-B’
Vertical Guided
Horizontal strip support
crack
Vertical
strips Strut Y
forces Node Fixed
X support
Z

A’ B’
(a) (b)

Figure 12. Unidirectional strut method: (a) wall divided into vertical strips and (b) frame divided
into n frame elements and boundary conditions.

The unidirectional strut method is as follows. For a given cracking pattern, the wall is
divided into vertical strips (Figure 12a). As the top of the wall displaces, compressive strut
forces are developed in each strip. These forces are transferred to the horizontal confining
elements. The frame, defined by these elements, is divided into n frame elements
(Figure 12b). Frame elements with six degrees of freedom per node are considered: three
rotational displacements (θ) and three translational displacements (u; McGuire et al.
2000). The boundary conditions for this model are fixed supports at the bottom of the
frame and restricted rotational displacements θx at the top of the frame (Figure 12b).
The iterative procedure to determine the wall panel strength is as follows:
1. An out-of-plane translational displacement at the top of the wall panel (δt ;
Figure 13a) is assumed. The in-plane translational displacements are calculated
(uC ; Figure 13a). Rigid body rotation of strips is assumed.
2. A uniformly distributed force (F A ) acting on the frame is assumed (Figure 13b). An
arbitrary value of F A is initially used. Axial load (P) is included as a uniformly

(a) (b)

Figure 13. (a) Out-of-plane and in-plane translational displacements and (b) in-plane uniformly
distributed forces and fixed-end forces.
2328 NAVARRETE-MACIAS ET AL.

distributed load on the top confining element (Figure 13b). Fixed-end forces (M A ,
M B , RA , and RB ) for each element of the frame and corresponding force vector {F A }
are calculated.
3. The displacements of the frame {U A } caused by {F A } are calculated (Equation 1),
where [K] is the global stiffness matrix of the frame. This stiffness is constant and it
is calculated using the gross properties of the frame cross sections:

fU A g ¼ ½K1 · fF A g
EQ-TARGET;temp:intralink-;e1;41;564 (1)
4. The normalized deformed shape vector {N} is calculated (Equation 2), where uAc is
the in-plane vertical displacement at the center of the top confining element obtained
from {U A }:
1
(2) fNg ¼ fU A g ·
uAc
EQ-TARGET;temp:intralink-;e2;41;500

5. The displacement vector {U C } is calculated (Equation 3). This vector is associated

with the in-plane displacement of the central vertical strip (uC ) calculated assuming
a rigid body rotation (Figure 13a). The normalized deformed shape vector {N} is
used to simplify the calculation of the displacement vector {U C }:

fU C g ¼ fNg · uC
EQ-TARGET;temp:intralink-;e3;41;413 (3)
C C
6. The force vector {F } associated with the displacement vector {U } is calculated
(Equation 4):

fF C g ¼ ½KfU C g
EQ-TARGET;temp:intralink-;e4;41;361 (4)
7. Axial deformation of each strip is calculated (Equation 5) assuming a triangular δai
bearing stress distribution and a linear stress-strain relationship up to crushing of the
masonry. In Equation 5, f m is the axial compressive strength of the masonry, hi is the
length of the strip, and δcri (Equation 6) is the axial deformation associated with
crushing of masonry. In Equation 6, εm is the crushing strain of the masonry
equal to 0.003 (Drysdale et.al. 1999). δcr i is calculated assuming a linear strain dis-
tribution over the length of the strip (Figure 14a):
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
  ffi
u
u2 · F · cos δsi · δs · δcr
t i hi i i
δai ¼ (5)
f m · hi
EQ-TARGET;temp:intralink-;e5;41;249

εm · hi
(6) δcr
i ¼
2
EQ-TARGET;temp:intralink-;e6;41;193

8. The total in-plane displacement for each strip uTi is calculated considering the axial
deformation of the masonry (Equation 7). The total displacement vector {U T } is
assembled:

uTi ¼ uCi þ δai

EQ-TARGET;temp:intralink-;e7;41;125 (7)
9. The force vector {F T }
are calculated (Equation 8), where is set equal to {F A } {F T }.
Steps 3 to 9 are repeated iteratively until the displacement convergence criterion is
OUT-OF-PLANE BEHAVIOR OF CONFINED MASONRY WALLS 2329

(a) (b) (c)

Figure 14. (a) Linear strain distribution in the strip, (b) geometric relationship used for calcula-
tion of bi , and (c) angles αi , βi , and γ i .

obtained (Equation 9). In this equation, Tol is a tolerance value and kU T k is the
norm of the displacement {U T } (Equation 10):

fF T g ¼ ½KfU T g
EQ-TARGET;temp:intralink-;e8;62;400 (8)

kU T j k  kU T j1 k
≤ Tol (9)
kU T j1 k
EQ-TARGET;temp:intralink-;e9;62;370

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
kU T k ¼
EQ-TARGET;temp:intralink-;e10;62;329 fU T gT · fU T g (10)
10. The contact widths between the vertical strip and the confining elements (bi ) are
calculated using a geometric relationship (Equation 11, Figure 14b):
δai
bi ¼ h (11)
δsi i
EQ-TARGET;temp:intralink-;e11;62;276

11. Axial stress of each strip (f ei ) is calculated assuming a linear stress-strain relationship
(Equation 12):
δai
f ei ¼ fm (12)
δcr
EQ-TARGET;temp:intralink-;e12;62;213

i
12. Based on the in-plane equilibrium of forces between F i and C i , the compressive strut
force (C i ) is calculated for each strip (Equation 13, Figure 14c):
Fi
Ci ¼ (13)
cosðγ i Þ
EQ-TARGET;temp:intralink-;e13;62;150

13. Based on the out-of-plane equilibrium between acting external forces and the com-
pressive struts, an out-of-plane strength V is calculated (Equation 14), where αi , βi ,
and γ i are calculated using Equations 15 to 17, respectively (Figure 14c).
2330 NAVARRETE-MACIAS ET AL.

In Equation 19, k 2 is the ratio of the distance from the extreme compressive fiber to
the resultant compressive strut:
X
n
V¼
EQ-TARGET;temp:intralink-;e14;41;615 C i · sinðγ i Þ (14)
i¼1

γ l ¼ β i  αi
EQ-TARGET;temp:intralink-;e15;41;567 (15)
 s
δi
αi ¼
EQ-TARGET;temp:intralink-;e16;41;548 sin1 (16)
hi
 
t  2 · k2 · bi
βi ¼ tan1 (17)
hi  4 · k 2 · δai
EQ-TARGET;temp:intralink-;e17;41;509

14. A new out-of-plane displacement δt is assumed. Steps 1 to 13 are repeated until the
stresses f ei are equal to f m or the out-of-plane translational displacement at the top of
the wall (δt ) is equal to the wall thickness. In the first case, wall failure is related to
crushing of masonry. In the second case, the out-of-plane components of the com-
pressive struts are equal to zero; therefore, wall failure is associated with snap
through of wall segments. These two failure types are related to the wall panel.
A wall panel strength–displacement curve was calculated for each wall studied.
On the other hand, the out-of-plane strength of the vertical confining elements was deter-
mined using a plastic analysis. A simplified bilinear moment-curvature diagram was calcu-
lated for each confining element. Yielding and flexural moments ðM y ,M u Þ together with
corresponding curvatures (Φy and Φu ) were calculated. The stress-strain curve used for con-
crete was based on Scott et al. (1982). The stress-strain curve used for the steel reinforcement
was based on Rodriguez y Botero (1996). The maximum strain of concrete was equal to
0.003 (Wight and Macgregor 2009). Failure of vertical confining elements was related to
crushing of concrete. Wall boundary conditions used in the plastic analysis are presented
in Figure 10b. A bilinear out-of-plane strength-displacement curve was calculated for the
vertical confining elements. The out-of-plane yielding and flexural strengths ðV y ,V u Þ
were determined using Equations 18 and 19, respectively. Out-of-plane yielding and ultimate
displacements ðΔy ,Δu Þ were calculated using Equations 20 and 21. In Equation 21, Lp is the
plastic hinge length calculated using Equation 22 (Paul and Priestley°1992). In this equation,
d b is the diameter of the flexural steel reinforcement. The analytical out-of-plane strength
(V ma ) of the walls was calculated as the maximum value of the sum of the strengths of
the wall panel and the confining elements for a given out-of-plane displacement (Figure 15):
4 · MY
Vy ¼ (18)
h0
EQ-TARGET;temp:intralink-;e18;41;187

4 · Mu
Vu ¼ (19)
h0
EQ-TARGET;temp:intralink-;e19;41;146

 2
0
Φy · h2
Δy ¼ 2 · (20)
3
EQ-TARGET;temp:intralink-;e20;41;115
OUT-OF-PLANE BEHAVIOR OF CONFINED MASONRY WALLS 2331

16 16 16

Out-of-plane load (kN)

Out-of-plane load (kN)

Out-of-plane load (kN)

Out-of-plane M1 Out-of-plane M2 Out-of-plane M3
14 14 strength 14
strength strength
12 12 12
10 10 10
8 Confining 8 Confining
Confining 8
elements elements
6 Wall 6 elements Wall 6 strength Wall
strength
4 panel 4 strength panel panel
4
strength strength strength
2 2 2
0 0 0
0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
Drift ratio (%) Drift ratio (%) Drift ratio (%)
16 16
Out-of-plane load (kN)

Out-of-plane load (kN)

Out-of-plane M4 M5
14 strength 14 Out-of-plane
12 12 strength
10 10
8 8
Confining Confining Wall
6 Wall 6
elements elements panel
4 panel 4
strength strength strength
strength
2 2
0 0
0 1 2 3 4 5 0 1 2 3 4 5
Drift ratio (%) Drift ratio (%)

Figure 15. Analytical strengths of confined walls, wall panels and confining elements.

  0 
ðΦu  Φy Þ h 2
Δu ¼ Δy þ 2 · Lp ·  · Lp (21)
2 2 3
EQ-TARGET;temp:intralink-;e21;62;387

Lp ¼ 0.08 · h 0 þ 0.88 · d b
EQ-TARGET;temp:intralink-;e22;62;348 (22)

Analytical out-of-plane strengths (V ma ) and corresponding experimental strengths (V me )

are compared in Table 1. This table shows that experimental strengths are predicted accu-
rately by the analytical models developed in this work.
The out-of-plane strength of confined walls is calculated in codes considering only the
contribution of the vertical confining elements (Norm-E.070-Peru 2006, NSR-98-Colombia
1998, MTN-Mexico 2004). With this assumption, the out-of-plane strength of the walls is

Table 1. Analytical and experimental strengths

Wall P (MPa) H∕L V ma (kN) V me (kN) V ma ∕V me

M1 0.11 0.73 10.71 10.95 0.98
M2 0.23 0.73 11.76 11.22 1.05
M3 0.35 0.73 12.91 14.07 0.92
M4 0.23 1.04 10.61 9.96 1.06
M5 0.23 1.80 9.69 8.93 1.08
Average 1.02
COV 0.065
2332 NAVARRETE-MACIAS ET AL.

underestimated as expected (Figure 15). On the other hand, the out-of-plane behavior of the
vertical confining elements is different than that observed for the walls. For example, no
strength degradation should be observed in the confining elements (Figure 15) but strength
degradation was observed in the walls (Figure 10).

CONCLUSIONS
Five confined walls were tested in the laboratory under reverse cyclic loads. Analytical
out-of-plane strength of walls was calculated using the unidirectional strut method and a
plastic analysis. Based on the experimental and analytical results obtained in this work,
the following conclusions are presented.
• The final cracking pattern of walls was in general similar. Horizontal cracks were
observed first between the two top and bottom brick courses on both wall faces.
Horizontal cracks were observed later at the top and bottom parts of the concrete
confining elements. For walls M1, M3, and M4, a vertical crack at the center of the
wall was observed at the end of the tests.
• The out-of-plane behavior of walls was linear elastic until the first masonry cracks.
After that, the behavior was nonlinear. This nonlinear behavior is related to the
presence of new horizontal cracks in the wall panel, horizontal cracks in vertical
confining elements and the yielding of the flexural reinforcement of these elements.
• For walls with the same aspect ratio, as the axial stress increased the out-of-plane
strength increased. This was related to the additional vertical displacement restric-
tion provided by the axial load. For walls with the same axial stress, as the aspect
ratio increased, the out-of-plane strength of the walls decreased. This was related to
the increment of the wall panel strength as the wall length increased.
• The out-of-plane strength of the walls was associated with drift ratios greater than
those related to the in-plane strength of confined masonry walls; therefore, the out-
of-plane and in-plane strength cannot be added. The out-of-plane strength contribu-
tion must be calculated using the drift ratio related to the in-plane strength.
• The out-of-plane strength of the walls was accurately predicted with the analytical
models developed in this work. Walls with different axial loads and aspect ratio
were considered.

REFERENCES
About, B. E., Ahmad, A. H., and Harris, H. G., 1996. Flexural behavior of reinforced concrete
masonry walls under out-of-plane monotonic loads, Structural Journal, ACI, 93, 327–335.
Abrams, D. P., Angel, R., and Uzarski, J., 1996. Out-of-plane strength of unreinforced masonry
infill panels, Earthquake Spectra 12, 825–844.
Anderson, C., 1984. Arching action in transverse laterally loaded masonry wall panels, Structural
Engineer 62B, 12–23.
Brzev, S., Astroza, M., and Monroi Yadlinn, M., 2010. Performance of Confined Masonry
Buildings in the February 27. Chile Earthquake.
Button, M. R., and Mayes, R. L., 1992. Out-of-plane seismic response of reinforced masonry
walls, Journal of Structural Engineering, ASCE, 118.
OUT-OF-PLANE BEHAVIOR OF CONFINED MASONRY WALLS 2333

Chen, W.-W., Yeh, Y.-K., Hwang, S.-J., Lu, C.-H., and Chen, C.-C., 2012. Out-of-plane seismic
behavior and CFRP retrofitting of RC frames infilled with brick walls, Engineering Structures
34, 213–224.
Colombian Association for Earthquake Engineering, 1998. Norm NSR-98: Colombian Norms for
Earthquake Resistant Design and Construction – Structural Masonry, Colombian Association
for Earthquake Engineering, Bogota, Colombia.
Da Porto, F., Mosele, F., and Modena, C., 2010. Experimental testing of tall reinforced masonry
walls under out of plane actions, Construction and Building Materials 24, 2559–2571.
Da Porto, F., Mosele, F., and Modena, C., 2011. Cyclic out of plane behavior of tall reinforced
masonry walls under P-Δ effects, Engineering Structures 33, 287–297.
Dawe, J. L., and Seah, C. K., 1989. Out-of-plane resistance of concrete masonry infilled panels,
Canadian Journal of Civil Engineering 16, 854–856.
Drysdale, R. G., and Essawy, A. S., 1988. Out-of-plane bending of concrete block walls, Journal
of Structural Engineering, ASCE, 114, 121–133.
Drysdale, R. G., Hamid, A. A., and Baker, L. R., 1999. Masonry Structures: Behavior and
Design, 2nd Edition., The Masonry Society, 888 pp.
Edison, 1994. Edison Hydraulic Load Maintainers, Operation and Maintenance Manual, Edison
Hydraulic Load Maintainers, Paradise, CA.
Flanagan, R. D., and Bennett, R. M., 1999. Bidirectional behavior of structural clay tile infilled
frames, Journal of Structural Engineering, ASCE, 125, 236–244.
Government of Federal District (CTN), 2004. Mexico City concrete technical norm, Federal
District Official Gazette, Federal District, Mexico.
Government of Federal District (MTN), 2004. Mexico City masonry technical norm, Federal
District Official Gazette, Federal District, Mexico.
Griffith, M., Lam, N., Wilson, J., and Doherty, K., 2004. Experimental investigation of
unreinforced brick masonry walls in flexure, Journal of Structural Engineering 130,
423–432.
Griffith, M. C., and Vaculik, J., 2007. Out-of-plane flexural strength of unreinforced clay brick
masonry walls, TMS Journal 25, 53–68.
Henderson, R. C., Fricke, K. E., Jones, W. D., Beavers, J. E., and Bennett, R. M., 2003. Summary
of a large and small-scale unreinforced masonry infill test program, Journal of Structural Engi-
neering, ASCE, 129, 1667–1675.
Jagadish, K., Raghunath, S., and Nanjunda, S., 2003. Behaviour of masonry structures during the
Bhuj earthquake of January 2001, Earth Planet Sci 112, 431–440.
Kaushik, H., Dasgupta, K, Sahoo, D., and Kharel, G., 2006. Performance of structures during the
Sikkim earthquake of 14 February 2006, Current Science 91, 449–455.
Kazemi, M. T., Asl, M. H., Bakhshi, A., and Rofooei, R., 2010. Shaking table study of a full scale
single storey confined brick masonry building, Scientia Iranica 17, 184–193
Klingner, R., Rubiano, N., Bashandy, T., and Sweeney, S., 1996. Evaluation and analytical ver-
ification of shaking table data from infilled frames, Part two, Proceedings of the Seventh North
American Masonry Conference, 2–5 June 1996.
Lam, N. T. K., Griffith, M., Wilson, J., and Doherty, K., 2003. Time-history analysis of URM
walls in out-of-plane flexure, Engineering Structures 25, 743–754.
Lawrence, S., 1983. Behaviour of Brick Masonry Walls under Lateral Loading, Ph.D. Thesis, The
University of New South Wales, Australia.
2334 NAVARRETE-MACIAS ET AL.

Lawrence, S., and Page, A. W., 1999. Design of Clay Masonry for Wind and Earthquake,
Clay Brick and Paver Institute, Australia.
McGuire, W., Gallagher, R. H., and Ziemian, R. D., 2000. Matrix Structural Analysis,
2nd Edition, Wiley, 480 pp.
Middleton, A. C., and Drysdale, R. G., 1995. Flexural capacities of concrete block walls with
openings, Seventh Canadian Masonry Symposium, Hamilton, Ontario.
Moghaddam, H., and Goudarzi, N., 2010. Transverse resistance of masonry infills, ACI Struc-
tural Journal 107, 461–467.
Moreno-Herrera, J. A., Varela-Rivera, J. L., and Fernandez-Baqueiro, L. E., 2014. Bidirectional
strut method: out-of-plane strength of confined masonry walls, Canadian Journal of Civil
Engineering 41, 1029–1035.
National Organization for Norms and Certification for Construction and Buildings, 2001. Norm
NMX-C-061-ONNCCE-2001: Construction Industry – Cement – Determination of the Com-
pressive Strength of Hydraulic Cements, National Organization for Norms and Certification
for Construction and Buildings (ONNCCE), Mexico.
National Organization for Norms and Certification for Construction and Buildings, 2002. Norm
NMX-C-083-ONNCCE-2002: Construction Industry - Concrete – Determination of the Com-
pressive Strength of Concrete Cylinders, National Organization for Norms and Certification
for Construction and Buildings (ONNCCE), Mexico.
National Organization for Norms and Certification for Construction and Buildings, 2010. Norm
NMX-C-464-ONNCCE-2010: Construction Industry – Masonry - Determination of the Com-
pressive Strength and the Modulus of Elasticity of Masonry Prisms, National Organization for
Norms and Certification for Construction and Buildings (ONNCCE), Mexico.
National Training for the Construction Industry, 2006. Technical Norm-E.070–Masonry.
National Training for the Construction Industry (SENCICO), Lima, Peru.
Rodríguez, M., and Botero, J. C., 1996. Aspectos del comportamiento sísmico de estructuras de
concreto reforzado considerando las propiedades mecánicas de aceros de refuerzo produci-
dos en México, UNAM, Series of II, 575. [in Spanish].
Sepulveda, A, Astroza, M., Kausel, E., Campos, J., Casas, E., Rebolledo, S., and Verdugo, R.,
2008. New findings on the 1958 Las Melosas earthquake sequence, Central Chile: Implica-
tions for seismic hazard related to shallow crustal earthquakes in subduction zones, Journal of
Earthquake Engineering 12, 432–455.
Singhal, V., and Rai, D. C., 2014. Seismic behavior of confined masonry walls when subjected to
in-plane and out-of-plane loading, Proceedings of the 10th National Conference on Earth-
quake Engineering, Earthquake Engineering Research Institute, Oakland, CA.
Scott, B. D., Park, R., and Priestley, M. J. N., 1982. Stress-strain behavior of concrete confined by
overlapping hoops at low and high strain rates, ACI Journal 79, 13–27.
Tu, Y. H., Liu, P. M., and Lin, H. P., 2007. Out of plane seismic behavior of unreinforced
masonry in filled walls, Structures Congress: New Horizons and Better Practices, ASCE.
Tu, Y. H., Chuang, T. H., Liu, P. M., and Yang, Y. S., 2010. Out of plane shaking table tests on
unreinforced masonry panels in RC frames, Engineering Structures 32, 3925–3935.
Varela-Rivera, J. L., Navarrete-Macias, D., Fernandez-Baqueiro, L. E., and Moreno, E. I., 2011.
Out-of-plane behaviour of confined masonry walls, Engineering Structures 33, 1734–1741.
Varela-Rivera, J., Moreno-Herrera, J., Lopez-Gutierrez, I., and Fernandez-Baqueiro, L., 2012a.
Out-of-plane strength of confined masonry walls, Journal of Structural Engineering,
ASCE, 138.
OUT-OF-PLANE BEHAVIOR OF CONFINED MASONRY WALLS 2335

Varela-Rivera, J., Polanco-May, M., Fernandez-Baqueiro, L., and Moreno, E. I., 2012b. Confined
masonry walls subject to combined axial loads and out-of-plane uniform pressures, Canadian
Journal of Civil Engineering 39, 439–447.
Wight, J. K., and MacGregor, J. G., 2009. Reinforced Concrete, Mechanics and Design,
5th Edition, Pearson Prentice Hall. Upper Saddle River, NJ, 1112 pp.
Zhang, X., Singh, S., Bull, D., and Cooke, N., 2001. Out of plane performance of reinforced
masonry walls with openings, Journal of Structural Engineering 127.
(Received 17 June 2015; accepted 8 April 2016)