1  Flexural Behavior of Confined Masonry Walls

2  Jorge Varela-Rivera, a) M.EERI, Luis Fernandez-Baqueiro, b)

M.EERI, Jose
3  Gamboa-Villegas c) and Adda Prieto-Coyoc d)

4  This paper presents the results of a study on the flexural behavior of

5  confined masonry walls. Six walls were tested in the laboratory under reverse
6  cyclic loads. The variables studied were the wall aspect ratio and the wall axial
7  compressive stress. The behavior of the walls was characterized by yielding of
8  the longitudinal steel reinforcement followed by vertical and diagonal cracks.
9  The failure of the walls was associated with crushing of the concrete of the
10  vertical confining elements. The flexural strength increases as the wall aspect
11  ratio or the wall axial stress increase. The flexural strength of the walls was
12  validated using flexural theory. A displacement ductility capacity of 6 and a
13  drift ratio capacity of 1% were proposed for the walls. A hysteretic model
14  based on that proposed by Takeda was developed. This model represented well
15  the flexural behavior of the walls studied.

16 

                                                            
a)
Professor, Autonomous University of Yucatan, Mexico. vrivera@correo.uady.mx
b)
Professor, Autonomous University of Yucatan, Mexico. luis.fernandez@correo.uady.mx
c)
Former graduate student, Autonomous University of Yucatan, Mexico. atticus_beto@hotmail.com
d)
Former graduate student, Autonomous University of Yucatan, Mexico. addapc@gmail.com

Varela-Rivera - 1
 
17  INTRODUCTION

18  In many countries of Latin America, central and south Asia, and eastern and southern
19  Europe, confined masonry walls are widely used as a structural system (Riahi et al. 2009).
20  Confined masonry consists of an unreinforced wall panel with flexible reinforced concrete
21  confining elements around its perimeter. In this type of construction, the wall panel is
22  constructed first and later the confining elements are concrete cast. The use of confined
23  walls in Mexico is very common because of their low construction cost and ease of
24  construction. Its use in the United States of America is scarce but could increase in the
25  future.

26  The shear behavior of confined walls under in-plane lateral loads has been widely
27  studied. There are several experimental studies carried out by different authors. The main
28  variables studied are the unit type (Meli 1979, San Bartolome and Quiun 2010),
29  combination of clay and concrete units (Tena-Colunga et al. 2009), types and quantities of
30  steel reinforcement in confining elements (Treviño et al. 2004, Quiroz et al. 2014), wall
31  axial load (Urzua et al. 2001), wall aspect ratio (height over length) (San Bartolome et al.
32  1992, Perez Gavilan et al. 2015), wall openings and type of reinforcement around openings
33  (Flores et al. 2004). In general, these studies considered the shear behavior of confined
34  walls with aspect ratios smaller than or equal to one. Only one study considered wall
35  aspect ratios greater than one (Perez Gavilan et al. 2015). The walls considered by those
36  authors were constructed using clay or concrete units. The shear behavior of the walls was
37  characterized by diagonal cracks that eventually formed the traditional “X” final cracking
38  pattern. The failure of the walls was mainly associated with propagation of diagonal cracks
39  into the top and bottom ends of the vertical confining elements. The shear strength of the
40  walls was associated with the formation of the first diagonal crack. It was observed that the
41  confining elements increased the loading and deformation capacity of the walls after
42  reaching their shear strength.

43  In contrast, the flexural behavior of confined walls under in-plane lateral loads has not
44  been studied. Confined walls are mainly used for housing. In low masonry structures, the
45  wall aspect ratios tend to be smaller than or equal to one. The prescribed minimum amount
46  of longitudinal (vertical) steel reinforcement for those walls (FDG, 2004a) induce shear

Varela-Rivera - 2
 
47  behavior. The walls tested under those geometrical and reinforcement conditions had shear
48  failure as described before. The prescribed minimum amounts of longitudinal and
49  transverse steel reinforcement for confined walls are based on shear behavior rather than
50  flexural behavior (FDG, 2004a). For example, closely-spaced steel hoops are required at
51  the ends of vertical confining elements to retard the propagation of diagonal cracks into
52  these elements. Reduction in the amount of the longitudinal steel reinforcement below the
53  minimum might be used to induce flexural behavior. The flexural strength of walls will be
54  reduced but the shear strength will remain the same. For taller masonry structures, wall
55  aspect ratios tend to be much greater than one. In this case, the prescribed minimum
56  amount of longitudinal steel reinforcement might induce flexural behavior rather than shear
57  behavior. Flexural behavior is more desirable in seismic zones.

58  There are some studies related to the in-plane flexural behavior of other types of
59  masonry walls. In the case of reinforced walls, the main variables that have been studied
60  are the quantity and distribution of reinforcement (Yoshimura et al. 2000, Shedid et al.
61  2008), wall axial load (Tanner et al. 2005, Varela et al. 2006, Shedid et al. 2008) and wall
62  aspect ratio (Tanner et al. 2005, Varela et al. 2006). The flexural behavior of those walls
63  was, in general, characterized by flexural cracks over the wall height followed by yielding
64  of the longitudinal steel reinforcement. The failure of the walls was mainly associated with
65  crushing of the masonry or the concrete at the wall ends. Flexural strength of walls was
66  maintained over a certain maximum displacement ductility.

67  The objective of this paper is to study the flexural behavior of confined walls subjected
68  to reverse cyclic loads. As far as the authors know, there is no previous research on
69  confined walls subjected to this type of load in the literature. Results of six confined walls
70  subjected to lateral loads are presented. The variables studied were the wall aspect ratio
71  and the wall axial compressive stress. The final cracking patterns of the walls are
72  presented. The lateral load – drift ratio curves for the walls are analyzed. A discussion
73  related with flexural strength, displacement ductility and drift ratios is presented. A
74  hysteretic model is developed based on experimental data.

Varela-Rivera - 3
 
75  EXPERIMENTAL PROGRAM

76  Six full-scale confined walls were tested in the laboratory (walls M1 to M6). The study
77  variables were the wall aspect ratio (H/L) and the axial compressive stress () (Table 1).
78  Nominal dimensions of hollow clay bricks were 115 200 320 mm (thickness height
79  length). Bricks were obtained from a single batch. Walls were designed to induce
80  flexural behavior. Table 1 shows details of each confined wall. In this table, H, L and t are
81  the wall height, length and thickness, respectively. The wall height was measured up to the
82  point of load application. The cross-section dimensions of the vertical confining elements
83  were 115 115 mm (width height). The longitudinal steel reinforcement in those
84  elements consisted of a 1#3 (95 mm) bar. The corresponding steel reinforcement ratio ()
85  is included in Table 1. No transverse reinforcement was placed on the vertical confining
86  elements. Longitudinal reinforcement consisted of deformed steel bars with nominal yield
87  strength of 412 MPa. The amount of longitudinal steel reinforcement of the walls was
88  smaller than the minimum amount prescribed in the Mexico City Masonry Technical Norm
89  (FDG 2004a).

90  Table 1. Details of confined walls.

H L t  
Wall H/L
(m) (m) (mm) MPa 
M1 2.91 2.54 115 1.1 0.24 0.024
M2 2.91 1.88 115 1.5 0.24 0.033
M3 2.91 1.88 115 1.5 0.47 0.033
M4 2.91 1.22 115 2.4 0.24 0.051
M5 2.91 1.22 115 2.4 0.47 0.051
M6 2.91 1.22 115 2.4 0.71 0.051
91 

92  The axial compressive strength of the concrete of the vertical confining elements was
93  determined according to NMX-C-083-ONNCCE-2002 (National Organization for Norms
94  and Certification for Construction and Buildings (ONNCCE)). The axial compressive
95  strength of the bricks was determined according to NMX-C-036-ONNCCE-2004. The
96  axial compressive strength and modulus of elasticity of the masonry were determined
97  according to NMX-C-464-ONNCCE-2010. The tensile strength of the steel longitudinal
98  reinforcement of the vertical confining elements was determined according to NMX-C-407-

Varela-Rivera - 4
 
 99  ONNCEE-2002. Procedures presented in these norms are similar to those specified in
100  corresponding ASTM standards.

101  Walls were constructed in half running bond by a qualified worker. Brick courses were
102  laid using mortar in proportion by volume 1:3 (Portland cement: sand). Mortar was placed
103  on both the face shells and the head joints. The average thickness of the mortar joints was
104  equal to 10 mm. Construction of the walls was as follows: first, the seven bottom brick
105  courses were laid and later the corresponding part of the vertical confining elements were
106  concrete cast. Second, the last five brick courses were laid and the corresponding
107  remaining part of the vertical confining elements were cast. Finally, the top confining
108  element was cast.

109  Each confined wall was tested with constant axial load and reverse monotonic cyclic
110  lateral loads until failure. The axial load for each wall was calculated using the
111  corresponding axial compressive stress, wall length and wall thickness (Table 1). Axial
112  load was applied using a swivel beam, a spreader beam, two threaded rods and a hydraulic
113  actuator (Figure 1a). Pressure in the actuator was maintained constant during the test using
114  a mechanical load maintainer (Edison 1994). Axial load was measured using two donut
115  type load cells. This load was verified using a pressure transducer. Lateral loads were
116  applied using a steel frame, a loading steel beam, and a two-way hydraulic actuator (Figure
117  1b). Lateral load was measured using a tension-compression pin load cell. This load was
118  verified by using two pressure transducers. Wall specimens were attached to the lab
119  reaction floor.

120  Horizontal and vertical wall displacements, and shortening or lengthening of the wall
121  diagonals were measured using linear string potentiometers (SP). Relative displacements
122  between the loading beam and the wall, the wall and the wall foundation, and the wall
123  foundation and the reaction floor were measured using linear potentiometers (LP). Strain
124  gages were attached to the longitudinal steel reinforcement of both vertical confining
125  elements. Two strain gages were located at the bottom of each bar. A typical view of wall
126  instrumentation is presented in Figure 1c. In this figure, PLC and DLC refers to the pin
127  type load cell and the donut type load cell, respectively.

Varela-Rivera - 5
 
128  The loading history used to test the walls was based on the protocol established in the
129  Mexico City Masonry Technical Norm (FDG 2004a). This loading history has six initial
130  reverse cycles controlled by load and subsequent cycles controlled by drift ratios. The
131  maximum target load was associated with yielding of the longitudinal steel reinforcement
132  of the vertical confining elements. The target load for the first two cycles was equal to one
133  quarter of the maximum target load, the third and fourth cycles to one half of the maximum
134  target load, and the fifth and sixth cycles to the maximum target load. After that,
135  increments of drift ratios of 0.002 were applied.

136 

137 

138  Figure 1. (a) Axial load test setup, (b) lateral load test setup, and (c) typical view of wall
139  instrumentation.

140  EXPERIMENTAL RESULTS

141  The average axial compressive strength of the concrete (fc) of walls M1 to M6 was
142  equal to 17.49, 18.10, 20.00, 18.36, 22.16 and 21.23 MPa, respectively. Corresponding

Varela-Rivera - 6
 
143  coefficient of variation (CV) was equal to 0.08, 0.04, 0.03, 0.08, 0.02 and 0.01,
144  respectively. The average axial compressive strength of the units (fp) was equal to 16.33
145  MPa with a CV of 0.06. The average axial compressive strength (fm) and modulus of
146  elasticity (Em) of masonry were equal to 9.08 and 5077 MPa, respectively. Corresponding
147  CV was equal to 0.04 and 0.12, respectively. The average yielding strength of the
148  longitudinal steel reinforcement was equal to 445.7 MPa with a CV of 0.02. All values
149  were calculated using gross properties of corresponding cross-sections.

150  The flexural behavior of walls was, in general, similar. First, a horizontal flexural crack
151  was observed at the joint between the first brick course and the concrete foundation
152  together with horizontal flexural cracks on the bottom part of the vertical confining
153  elements. After this, yielding of the longitudinal steel reinforcement at the bottom end of
154  both vertical confining elements was reached. As the drift ratio was increased, horizontal
155  flexural cracks propagated into the wall panel and new flexural cracks were observed along
156  the height of the vertical confining elements. Vertical cracks were observed for the walls
157  M2 to M6. A single vertical crack for the walls M2, M4 and M5 and two vertical cracks
158  for the walls M3 and M6. Diagonal shear cracks were observed on the wall panels. These
159  cracks propagated from the top part of the wall panels to the existing vertical cracks. Out-
160  of-plane buckling of one of the vertical confining elements was observed for the wall M5.
161  The failure of the walls was associated with crushing of the concrete at the bottom part of
162  the vertical confining elements. The final cracking patterns of the walls are presented in
163  Figure 2. The lateral load – drift ratio curves of the walls are presented in Figure 3.

164 

165 

Varela-Rivera - 7
 
166 
167  Figure 2. Final cracking patterns of walls.

168 

169 

170 

Varela-Rivera - 8
 
171 

172  Figure 3. Lateral load – drift ratio curves of walls.

173  DISCUSSION OF EXPERIMENTAL RESULTS

174  The experimental average flexural strengths of the walls (Me) are presented in Table 2.
175  These strengths were calculated using the corresponding positive and negative maximum
176  observed lateral loads. Table 2 shows that, as expected, for the walls with the same axial
177  compressive stress (walls M1, M2 and M4), the flexural strength of walls increases as the
178  wall aspect ratio increases. For walls with the same aspect ratio (M4 to M6), the flexural
179  strength increases as the axial compressive stress increases. The analytical flexural
180  strengths of the walls (Ma) are also presented in Table 2. These strengths were calculated
181  using flexural theory. A rectangular block was used for the compressive stresses of
182  concrete (FDG 2004b). This type of block was used because the wall neutral axis was
183  located within the height of the vertical confining element. A stress-strain relationship of
184  the steel including strain hardening was used (Rodriguez and Botero 1994). Table 2 shows
185  that there is a good agreement between analytical and experimental flexural strengths. The
186  ratio between Ma/Me varied from 0.98 to 1.06.

187  Maximum horizontal displacements (m) were selected for the walls (Table 2). These
188  displacements were limited by one of the following events: a 10% strength degradation or
189  the presence of the first diagonal crack on the wall, whichever happened first. The
190  corresponding displacement ductilities () and drift ratios () of the walls are included in
191  Table 2. The yielding displacements of the walls (y) (Table 2) were calculated using the
192  average readings of the strain gages. Table 2 shows that, for the walls with the same axial

Varela-Rivera - 9
 
193  compressive stress (walls M1, M2 and M4), drift ratio increases as the wall aspect ratio
194  increases. The corresponding displacement ductilities did not follow that trend because
195  walls M1, M2 and M4 had the same amount of longitudinal steel reinforcement but
196  different steel reinforcement ratios () (Table 2). For the walls with the same aspect ratio
197  (walls M4 to M6), as expected, drift ratio and displacement ductility increase as the axial
198  compressive stress decreases. The displacement ductilities of the walls varied from 6.06 to
199  14.31. The drift ratios varied from 0.96% to 2.41%. These values showed that the
200  confined walls studied had a good deformation capacity under lateral loads. The minimum
201  displacement ductility and drift ratio of the walls were about 6 and 1%, respectively. Based
202  on those minimum values, a displacement ductility capacity of 6 and a drift ratio capacity
203  of 1% are proposed for the confined walls studied. The ductility capacity of 6 is greater
204  than the value of 3.5 reported for reinforced AAC walls with flexural behavior (Varela et
205  al. 2006). The drift ratio capacity of 1% is equal to that reported for those reinforced AAC
206  walls but is greater than the values proposed for confined walls with shear behavior, 0.4%
207  and 0.6% for hollow and solid units, respectively (FDG 2004a).

208  Table 2. Experimental flexural strengths, analytical flexural strengths, drift ratios and displacement
209  ductilities of walls.
  Me Ma Ma/Me m y  
Wall H/L
(MPa)  (kN-m) (kN-m) (mm) (mm)
M1 1.1 0.24 0.024 237.71 237.71 1.00 28.0 3.0 9.33 0.96
M2 1.5 0.24 0.033 151.82 158.37 1.04 51.5 3.6 14.31 1.77
M3 1.5 0.47 0.033 201.47 199.76 0.99 30.4 2.4 12.66 1.04
M4 2.4 0.24 0.051 79.33 83.90 1.06 70.0 7.9 8.86 2.41
M5 2.4 0.47 0.051 102.73 104.16 1.01 40.0 5.2 7.69 1.37
M6 2.4 0.71 0.051 130.13 126.99 0.98 30.3 5.0 6.06 1.04
210  Vertical cracks were observed for the walls M2 to M6, one for the walls M2, M4 and
211  M5 and two for the walls M3 and M6. These cracks formed on the wall side in
212  compression. The vertical cracks were, in general, located on the bottom brick courses at
213  about 150 mm from the joint between the vertical confining elements and the wall panel
214  (Figure 2). The vertical cracks were associated with the brick bond pattern used in
215  construction and the non-uniform vertical deformation along the wall length. The mortar
216  head joints of alternating end bricks were vertically aligned (Figure 2). It was observed
217  during testing that vertical cracks formed first at those head joints and then propagated into

Varela-Rivera - 10
 
218  the bricks. This was related to the smaller compressive strength of the mortar of joints
219  compared with that of the bricks. The non-uniform vertical deformation along the wall
220  length was caused by the difference between the modulus of elasticity of concrete and
221  masonry. The modular ratio between concrete and masonry was about 3. For example,
222  under only axial load, the axial stress on the wall is uniform but the bricks close to the
223  vertical confining elements tend to deform less than those located at the wall midlength.
224  This deformation gradient was greater for the walls with the smaller length. Wall M1 with
225  the largest length did not have any vertical crack. Under lateral loads, the compressive
226  stress increases at the corresponding wall end. Vertical cracks on walls with aspect ratios
227  greater than one have also been observed by other authors (Sosa 2013, Perez Gavilan
228  2015).

229  The vertical cracks divided the walls into wall segments. As the “effective” wall length
230  decreased, the shear strength of the walls also decreased. This strength reduction triggered
231  the formation of the diagonal cracks observed on the walls. Analytical shear strengths (Va)
232  were calculated using Equation 1 (Fernandez et al. 2014). In this equation, vm is the
233  average diagonal compressive strength of the masonry (shear strength) of 0.83 MPa as
234  reported by the manufacturer,  is the inverse of the wall aspect ratio (L/H), At is the wall
235  cross-section area of the wall and P is the wall axial load. The values of  and At were
236  calculated using reduced wall lengths (Lr) (Table 3). These reduced lengths were measured
237  before diagonal cracking. The analytical shear strengths of the walls (Va) are compared in
238  Table 3 with the corresponding observed lateral loads at first diagonal cracking (Ve). In this
239  table, P is the total axial load on the wall including wall and loading beams self-weight.
240  Table 3 indicates a good agreement between analytical and experimental shear strengths.
241  The ratio between Va/Ve varied between 0.97 to 1.07.

242  Table 3. Analytical and experimental shear strengths of walls.

 L Lr P Va Ve Va/Ve
Wall H/L
(MPa) (m) (m) (kN) (kN) (kN)
M2 1.5 0.24 1.88 1.44 73.46 53.33 52.72 1.01
M3 1.5 0.47 1.88 1.32 121.39 62.71 64.70 0.97
M4 2.4 0.24 1.22 0.95 45.77 27.35 25.94 1.05
M5 2.4 0.47 1.22 0.95 78.79 37.25 34.72 1.07
M6 2.4 0.71 1.22 0.69 120.57 43.35 44.77 0.97
243 

Varela-Rivera - 11
 
244  Va 0.46 vm α At 0.3 P (1)

245  The secant stiffness at yielding of the steel longitudinal reinforcement (Ky) of the walls
246  M1 to M6 was equal to 19959, 8463, 21772, 2737, 5574, and 6775 kN/m, respectively.
247  Each yielding stiffness was calculated using only the first positive cycles. Figure 4 shows
248  the normalized stiffness degradation curves for the walls as a function of the displacement
249  ductility. The secant stiffness (Ki) of each positive cycle was divided by the secant yielding
250  stiffness (Ky). Load cycles up to the values of maximum displacements were included.
251  Figure 4 shows that the stiffness degradation of the walls was similar. For displacement
252  ductilities of 2, 4, and 6, the stiffness degradation was about 45, 70, and 80%, respectively.

253 

254  Figure 4. Normalized stiffness degradation curves of walls.

255 

256  ANALYTICAL MODELS

257  A hysteretic model was developed to represent the flexural behavior of the walls
258  (Figure 5). This model was based on that proposed by Takeda (CANNY, 1999). The
259  bilinear loading branch was defined by the yielding stiffness (Ky) and the post yielding
260  stiffness (Kpy) of the walls. The unloading branch of the model was defined by the
261  unloading stiffness (Ku) and the slope of the target straight line U-U´. Degradation of the
262  unloading stiffness (Ku) was considered as a function of the wall displacements.

Varela-Rivera - 12
 
263 

264  Figure 5. Hysteretic model for walls.

265 

266  The yielding stiffness of the walls (Ky) was defined by Equation 2. In this equation,  is
267  a hysteretic parameter, Vy is the lateral load associated with yielding of the longitudinal
268  steel reinforcement, Gm is the shear modulus of elasticity of masonry, Ig is the gross
269  moment of inertia of the cross-section area, and Ag is the gross cross-section area. Values
270  of  were back calculated for walls M1 to M6 using each corresponding experimental
271  secant yielding stiffness. The values of  were equal to 0.36, 0.28, 0.72, 0.25, 0.51 and
272  0.62, respectively. For the walls with the same axial compressive stress (walls M1, M2 and
273  M4), the values of  varied between 0.25 and 0.36 with a mean value of 0.30. For the walls
274  with the same aspect ratio (walls M4 to M6), the values of  increase as the axial
275  compressive stress increases. The values of  were equal to 0.25, 0.51 and 0.62,
276  respectively. The post yielding stiffness (Kpy) was defined as a function of the initial
277  yielding stiffness (Ky) (Equation 3). Values of  were calculated for walls M1 to M6 using
278  the loading parts of the lateral load-drift ratio curves after yielding of the longitudinal steel
279  reinforcement. A straight line was fitted in all cases. The values of  were equal to 5, 7, 4,
280  4, 3 and 9%, respectively. The mean value of  was equal to 5%.
‐1
Vy H3 1.2 Vy H
281  K y β 3 Em Ig Gm Ag
(2)

282  K py γ K y (3)

Varela-Rivera - 13
 
283  The unloading branches of the lateral load-drift ratio curves of the walls were divided in
284  two parts, an elastic unloading branch and a damage unloading branch. The elastic part of
285  the unloading stiffness (Ku) was defined by Equation 4. In this equation,  is a hysteretic
286  parameter, Vm is the lateral force at a given load cycle and dm its corresponding
287  displacement. Values of  were calculated for the walls M1 to M6 using the elastic
288  unloading parts (positive and negative) of the lateral load-drift ratio curves. The
289  degradation of the unloading stiffness (Ku) was considered by using different maximum
290  displacements (dm). A straight line was fitted in all cases. The values of  were equal to
291  1.7, 1.6, 1.6, 1.4, 1.4 and 1.4, respectively. It was observed that the values of  were
292  similar. The average value of  was equal to 1.5. The damage part of the unloading
293  stiffness was defined by the slope of the target straight line U-U´ (Figure 5). This slope
294  (stiffness, Ku-u) was defined as a function of the yielding stiffness (Ky) (Equation 5). Values
295  of  were calculated for walls M1 to M6 using the end points of the elastic unloading parts
296  (positive and negative) of the lateral load-drift ratio curves. The values of  were equal to
297  5, 5, 7, 5, 10 and 15%, respectively. For the walls with the same axial compressive stress
298  (walls M1, M2 and M4), the values of  were the same (5%). For the walls with the same
299  aspect ratio (walls M4 to M6), the values of  increase as the axial compressive stress
300  increases. The values of  were equal to 5, 10 and 15%, respectively. Pinching of the
301  cycles was not included in the hysteretic model because it was not observed up to the
302  values of maximum displacements of the walls.
θ Vy Vm
303  K u θ Vy (4)
K ye dm

304  (5)

305  A hysteretic model was proposed for the walls with the same axial compressive stress
306  (walls M1, M2 and M4). The selected hysteretic parameters were  = 0.3, = 5%, = 1.6
307  and = 5%. The experimental and analytical lateral load – drift ratio curves of walls M4
308  and M6 are presented in Figure 6. Wall M1 and M4 had the minimum and maximum wall
309  aspect ratio, respectively. Experimental and analytical curves are represented by solid and
310  dashed lines, respectively. Figure 6 shows, in general, a good agreement between both
311  experimental and analytical curves up to the value of drift ratio capacity of 1%.

Varela-Rivera - 14
 
312 

313  Figure 6. Experimental and analytical lateral load – drift ratio curves for walls M4 and M6.

314  CONCLUSIONS

315  Six confined masonry walls were tested in the laboratory under reverse cyclic lateral
316  loads. Walls were designed to induce flexural behavior. Based on the results obtained in
317  this work, the following conclusions are presented.

318   The flexural behavior of the walls was characterized by yielding of the longitudinal
319  steel reinforcement followed by vertical and diagonal cracks. The failure of the
320  walls was associated with crushing of the concrete at the bottom ends of the vertical
321  confining elements.

322   As expected, flexural strength of walls increases as the aspect ratio or the axial load
323  increase. Flexural strength of walls can be determined using flexural theory.

324   The confined walls studied had a good deformation capacity under lateral loads.
325  The displacement ductilities varied from 6.06 to 14.31 and the drift ratios from
326  0.96% to 2.41%. Based on the observed minimum values, a displacement ductility
327  capacity of 6 and a drift ratio capacity of 1% are proposed for the walls.

328   The vertical cracks observed on the walls caused a reduction in their shear strength.
329  Because of this reduction, diagonal shear cracks were observed in the walls.
330  Vertical cracks were associated with the brick bond pattern used in construction and
331  the non-uniform vertical deformation on the wall.

Varela-Rivera - 15
 
332   A hysteretic model was developed to represent the flexural behavior of the walls.
333  The model was based on that proposed by Takeda. In general, hysteretic parameters
334  depended on the wall aspect ratio and the wall axial compressive stress. The
335  proposed hysteretic model represented well the observed flexural behavior of the
336  walls studied.

337  REFERENCES

338  CANNY, 1999. A 3-Dimensional Nonlinear Static / Dynamic Structural Analysis Program,
339  Technical Manual, CANNY Structural, Vancouver, Canada.
340  Edison, 1994. Edison hydraulic Load Maintainers, Operation and Maintenance Manual. Edison
341  hydraulic load maintainers, Paradise, California, USA.

342  Flores, L. E., Mendoza J. A., and Reyes C., 2004. Ensaye de Muros de Mampostería con y sin
343  Refuerzo Alrededor de la Abertura (Testing of Masonry Structures with and without
344  Reinforcement on Openings), In Proceedings of the XIV National Congress of Structural
345  Engineering, October, 2004, Acapulco, Mexico. [in Spanish]
346  Federal District Government (FDG), 2004a. Technical Norms for the Construction and Design of
347  Masonry Structures. Federal District Official Gazette. Mexico City, Mexico. [in Spanish]

348  Federal District Government (FDG), 2004b. Technical Norms for the Construction and Design of
349  Concrete Structures. Federal District Official Gazette. Mexico City, Mexico. [in Spanish]

350  Fernandez-Baqueiro, L. E., Sosa-Moreno, M. I., and Varela-Rivera, J. L., 2014. Resistencia en el
351  Plano de Muros de Mampostería: Efecto de la Relación de Aspecto (In-plane Strength of
352  Confined Walls: Effect of Aspect Ratio), In Proceedings of the XIV National Congress of
353  Structural Engineering, November, 2014, Puerto Vallarta, Mexico. [in Spanish]

354  Meli, R., 1979. Comportamiento Sísmico de Muros de Mampostería (Seismic Behavior of Masonry
355  Walls), Tech. Rep. Serie Azul: Instituto de Ingeniería, UNAM, No. 352, Mexico City, Mexico.
356  [in Spanish]

357  National Organization for Norms and Certification for Construction and Buildings, 2004. Norm
358  NMX-C-036-ONNCCE-2004. Construction Industry – Blocks and Bricks – Compressive
359  Strength – Test Methods. National Organization for Norms and Certification for Construction
360  and Buildings (ONNCCE), Mexico.

361  National Organization for Norms and Certification for Construction and Buildings, 2002. Norm
362  NMX-C-083-ONNCCE-2002. Construction Industry - Concrete – Determination of the

Varela-Rivera - 16
 
363  Compressive Strength of Concrete Cylinders. National Organization for Norms and
364  Certification for Construction and Buildings (ONNCCE), Mexico.

365  National Organization for Norms and Certification for Construction and Buildings, 2010. Norm
366  NMX-C-464-ONNCCE-2010. Construction Industry – Masonry – Determination of the
367  Compressive Strength and the Modulus of Elasticity of Masonry Prisms. National Organization
368  for Norms and Certification for Construction and Buildings (ONNCCE), Mexico.

369  National Organization for Norms and Certification for Construction and Buildings, 2002. Norm
370  NMX-C-407-ONNCCE-2002. Construction Industry – Deformed Steel Bars for Concrete –
371  Specifications and testing Methods. National Organization for Norms and Certification for
372  Construction and Buildings (ONNCCE), Mexico.
373  Perez Gavilan, J. J., Flores, L. E., and Alcocer, S. M., 2015. An Experimental Study of Confined
374  Masonry Walls with Varying Aspect Ratio, Earthquake Spectra, 31:2, 945-968.

375  Quiroz, L., Maruyama, Y., and Zavala, C., 2014. Cyclic Behavior of Peruvian Confined Masonry
376  Walls and Calibration of Numerical Model Using Genetic Algorithms, Journal of Engineering
377  Structures, 75, 561-576.
378  Riahi, Z., Elwood, K., and Alcocer, S. M., 2009. Backbone Model for Confined Masonry Walls for
379  Performance Based Seismic Design, Journal of Structural Engineering, 125:6, 644-654.

380  Rodríguez, M., and Botero, J. C., 1994. Aspectos del Comportamiento Sísmico de Estructuras de
381  Concreto Reforzado Considerando las Propiedades Mecánicas de Aceros de Refuerzo
382  Producidos en México (Fundamental Aspects of the Seismic Behavior of Concrete Structures
383  Considering Mechanical Properties of Mexican Steel Reinforcements), Tech. Rep. Instituto de
384  Ingeniería, UNAM, No. 575. Mexico City, Mexico [in Spanish].
385  San Bartolomé, A., Quiun, D., and Torrealva, D., 1992. Seismic Behaviour of a Three-Story Half
386  Scale Confined Masonry Structure, In Proceedings of Earthquake Engineering, Tenth World
387  Conference, 1992, Balkema, Rotterdam.
388  San Bartolome, A., and Quiun, D., 2010. Diseño Sísmico de Edificaciones de Albañilería Confinada
389  (Seismic Design of Confined Masonry Structures), Revista Ciencia, Mexico City, Mexico, 13:2,
390  161-185. [in Spanish]

391  Shedid, M. T., Drysdale, R. G., El-Dakhakhni, W. W., 2008. Behavior of Fully Grouted Reinforced
392  Concrete Masonry Shear Walls Failing in Flexure: Experimental Results. Journal of Structural
393  Engineering, ASCE, 134:11, 1754-1767.
394  Sosa, M., 2013. Muros de Mampostería Confinada de Bloques Huecos de Concreto, Sujetos a
395  Cargas Laterales en su Plano (Confined Masonry Walls Constructed with Hollow Concrete

Varela-Rivera - 17
 
396  Blocks Subjected to In-Plane Lateral Loads). M.S. Thesis, Autonomous University of Yucatan,
397  Mexico.

398  Tanner, J. E., Varela, J. L., Klinger, R. E., and Brightman, M. J., 2005. Seismic Testing of
399  Autoclaved Aerated Concrete Shear-Walls: A Comprehensive Review, ACI Structural
400  Journal, 102:3, 374-382.
401  Tena-Colunga, A., Juárez-Ángeles A., and Salinas-Vallejos, V. H., 2009. Cyclic Behavior of
402  Combined and Confined Masonry Walls, Journal of Engineering Structures, 31:1, 240-259.

403  Treviño, E. L., Alcocer, S. M., Flores, L. E., Larrua, R., Zarate, J. M., and Gallegos, L., 2004.
404  Investigación Experimental del Comportamiento de Muros de Mampostería Confinada de
405  Bloques de Concreto Sometidos a Cargas Laterales Cíclicas Reversibles Reforzados con Acero
406  de Grados 60 y 42 (Experimental Study on the Behavior of Confined Walls, Reinforced with 60
407  and 42 grade Steel and Constructed with Concrete Blocks, Subjected to Reverse Cyclic Loads),
408  In Proceedings of the XIV National Congress of Structural Engineering, October, 2004,
409  Acapulco, Mexico. [in Spanish]

410  Urzua, D. A., Padilla, M. R., and Loza, J. R., 2001. Influencia de la Carga Vertical en la Resistencia
411  Sísmica de Muros de Mampostería Confinada con Materiales Pumíticos Típicos de Guadalajara
412  (Influence of the Axial Load on the Strength of Confined Masonry Walls Constructed with
413  Materials from Guadalajara), In Proceedings of the XIII National Congress on Earthquake
414  Engineering, October 2001, Guadalajara, Mexico. [in Spanish]

415  Yoshimura, K., Kikuchi, K., Kuroki, M., Liu, L., and Ma, L., 2000. Effect of wall reinforcements,
416  applied lateral forces and vertical axial loads on seismic behavior of confined concrete masonry
417  walls, In Proceedings of the 12 World Conference on Earthquake Engineering, 30-January to 4-
418  February, 2000, Auckland, New Zealand.

419  Varela, J. L., Tanner, J. E., and Klinger, R. E., 2006. Development of Seismic Force
420  Reduction and Displacement Amplification Factors for Autoclaved Aerated Concrete
421  Structures, Earthquake Spectra, 22:1, 267-286.

Varela-Rivera - 18