Seismic response of stone masonry building with wooden band

Full Text

1. Introduction The historical and traditional structures of Karnali Zone of Sinja Valley (Nepal) are constructed with excessive use of stone masonry and timber elements. Even today, use of masonry walls cannot be avoided in developing countries, worldwide like our country. The timber elements are used in these houses in the form of beams, columns, joists, doors, windows, band and band connectors and other decorative elements [1]. Timber door, windows and other decorative elements not only provide pleasant aesthetic view but also impart structural stability and in controlling localized stresses. However, the structural strength of these houses against the possible earthquakes is limited. The situation calls for the need of seismic analysis of the buildings so that appropriate strengthening techniques can be applied. Stone masonry house are highly vulnerable in shear, bending, and torsions. Due to these stress on masonry, out of failure and in plane failure is common vulnerable phenomenon. The appropriate modeling of the building like those of stone masonry is important to assess in analysis for performance and response of the structure [2]. As per field observation, most of houses of Karnali Zone (Nepal) made of stone masonry with compacted mud thatched roof. Buildings are not designed properly in terms of seismic performance and vulnerability. Besides that Karnali Zone (Nepal) by the virtue of active faults in the vicinity, several places worldwide are located in highly seismic prone zone, constructed highly steep terrain and the soil strata is found composed of very weak soil. The structures constructed over such terrain and with soft strata are not very much favorable for resisting seismic forces, which may subject to high amplification of ground shaking effect [3]. The most area of Sinja Valley (Nepal) still contains traditional building constructed with masonry and timber. Neither significant researches nor have detailed studies of the loads or bearing capacities of traditional houses of Sinja Valley been carried out. Usually, most of the construction or repair works are done in a very simple way without considering seismic effects [1-6]. A lot of research works have been conducted on new construction materials and technologies but the research works regarding retrofitting, rehabilitation, repair, strengthening of traditional house are limited. Besides that this tradition also affects the cultural heritage housing construction practices and lose our traditional architectural value day by day. In view of structural performance masonry structures have limited resisting capacity against earthquake (Figures 1 and 2) [7]. So, it is vital to have a study to address the present status of the structural capacity of the traditional house whether they are capable of withstanding the possible future seismic impact. The responsibility of a structural engineer is not only limited to construction of modern structures, but also to preserve the traditional structures which reflect the state of civilization, tradition and culture [8; 9]. In this regard, the present study becomes an essential step in the strengthening of traditional house for our future generations and the study of the Sinja Valley (Nepal) housing trend [10; 11]. Likewise the analysis is done enhancing seismic impact worldwide. Figure 1. Failure mode of masonry houses: a - out of plane collapses of load bearing masonry wall in Bhaktapur; b - heavily damaged masonry structure in Chautara due to out of plane collapse of majority of walls; c - delaminating of the masonry observed in stone masonry wall in Solukhumbu (Everest base camp area); d - common practice of mortar placement for masonry construction [12] Figure 2. Failure mode of masonry houses: a - pounding and progressive failure on the building situated on edge; b - complete collapse of row houses in Baluwa (near epicenter; c - progressive failure on row houses and good performance of timber frames; d - stone masonry failure 2. Methods The research plan is shown in Figure 3. The traditional building as shown in Figures 4 and 5 is usually rectangular in plan and stretched over two storey's height. The length of the plan is 7.7 m with facades of various widths but most 6.92 m, the house is raised vertically over two storeys with a partition wall running up the height, creating front and back rooms. Timber frames are provided at certain interval 1.7 m, parallel to the facade. Sometimes timber frames are replaced by stone walls in order to create rooms. The typical inter storey height is between 1.75 m for ground floor and 2.75 m for first floor. The ground floor is used for animals and first floor is used for human beings. Generally small size of opening are provided where size of doors are 0.90 m width and 1.5 m height whereas window size are 1.2 m width and 1 m height. During the construction, the modern construction materials like concrete, bricks, steel were not available in the proposed site frequently [1]. Figure 3. Flow chart of methodology Using structural analysis program it’s not easy to create model as in RCC or steel building. However using various links emends the model of masonry building shall be created whose applications are as followings [13]: 1) the links as hooks, springs, plastic wanes are created so as to meet criteria of nodal points among joist and beams, beams and posts stone masonry walls and stiffening beams and connectors; 2) those links need for optimizations of modal so as give approximate final output results by nearest partial fixity among all nodes of structures; 3) those links are placed in separate model and the results are verified with manual results; 4) the main applications of such links are to create partial fixity and pinned joints among nodes and stone masonry with connectors and stiffened beams. Figure 4. Ground floor plan of model house Figure 5. First floor plan of model house 3. Results and discussion The link element is used to connect two joints together. Each link element may exhibit up to three different types of behaviour: linear, non-linear, and frequency-dependent, according to the types of properties as signed to that element and the type of analysis being perform [14]. Figure 6. Comparison of period Fundamental time period of hook, masonry, bare frame, composite, gap, spring and rigid models obtained as 0.0613, 0.183, 0.264, 0.063, 0.0603, 0.061 and 0.051second respectively (Figure 6). Maximum fundamental time period obtained is in care of bare frame and minimum in care of rigid. There is no significance difference in time period among hook, composite, gap and spring. It can be seen that influence of timber band is significant to increase global lateral stiffness of the building that caused decrease in fundamental time period in hook, composite, gap and spring. Band is not modeled inn care of bare frame and masonry care. Similarly rigid diaphragm also play significant role for enhance the stiffness of the building [15]. Base shear data presented at Tables 1-4 and Figures 7-9. Table 1 Base shear calculation, manually Description of items L, m B, m t, m Unit weight, kN/m3 Weight, kN Storey height, m, ground floor 1.5 Thickness of mud, m 0.2 Plank thickness 0.05 Thickness of wall 0.35 Size of joist 0.18 0.2 Size of beam 0.2 0.225 Size of post 0.225 0.225 Total longitudinal length 7.7 Total transverse length 6.92 Unit weight of mud 15 Unit weight of wood 8.5 Unit weight of stone masonry 22 Load calculation Mud load calculation 159.85 Self weight of plank 22.646 Weight of joist 21.175 Weight of beam 11.781 Weight of post 23.237 Weight of wall longitudinal 592.9 Weight of transverse wall 399.63 Seismic load due to live load 26.642 Total lumped mass at roof 1231.2 Total lumped mass 1257.9 Total weight, w 2789.1 Time period 0.15 Importance factor 1 Response reduction factor 1.5 Sa/g 2.5 Ah 0.0970 Vb 156.73 Table 2 Comparison of base shear Model designation Base shear along X-direction, Vx, KN Base shear along Y-direction, Vy, KN Gap 158.81 171.12 Hook 153.38 159.90 Bare frame 31.28 25.10 Masonry 144.85 110.67 Composite 153.38 159.90 Spring 138.58 146.39 Semi rigid 155.22 162.39 Rigid 156.10 170.10 Figure 7. Base shear of various test models Figure 8. Displacement about X-axis of various modeled cases Table 3 Displacement about X-axis, mm Joint Height, m Bare frame Masonry Composite Hook Gap Spring Semi rigid Rigid 720 4.5 2.416 0.224 0.185 0.098 0.1172 0.1183 0.1089 0.068 754 3.45 2.251 0.219 0.099 0.081 0.096 0.1106 0.094 0.063 11 2.7 2.1166 0.2165 0.085 0.0705 0.083 0.092 0.08 0.062 13 2.15 1.4436 0.1748 0.070 0.06 0.071 0.074 0.067 0.059 146 1.6 1.092 0.1467 0.060 0.051 0.059 0.063 0.054 0.055 15 1.2 0.84 0.106 0.054 0.042 0.051 0.054 0.049 0.0395 145 0.5 0.22 0.0298 0.035 0.028 0.0336 0.034 0.032 0.028 38 0 0 0 0 0 0 0 0 0 Figure 9. Displacement about Y-axis of various modeled cases Table 4 Displacement about Y-axis, mm Joint Height, m Bare frame Masonry Composite Hook Gap Spring Semi rigid Rigid 720 4.5 1.49 0.21 0.116 0.1 0.117 0.096 0.1 0.059 754 3.45 1.422 0.153 0.098 0.08 0.096 0.079 0.088 0.063 11 2.7 1.352 0.105 0.086 0.07 0.083 0.068 0.078 0.062 13 2.15 0.805 0.089 0.075 0.06 0.071 0.058 0.063 0.059 146 1.6 0.664 0.074 0.061 0.05 0.059 0.048 0.058 0.055 15 1.2 0.567 0.064 0.055 0.04 0.051 0.04 0.04 0.039 145 0.5 0.171 0.029 0.039 0.03 0.033 0.027 0.031 0.028 38 0 0 0 0 0 0 0 0 0 Output of shell element internal stresses. The basic shell element stresses are identified as S11, S22, S12, S13, and S23. You might expect that there would also be an S21, but S21 is always equal to S12, so it is not actually necessary to report S21. Sij stresses (where i can be equal to 1 or 2 and j can be equal to 1, 2 or 3) are stresses that occur on face i of an element in direction j. Direction j refers to the local axis direction of the shell element. Thus S11 stresses occur on face 1 of the element (perpendicular to the local 1 axis) and are acting in the direction parallel to the local 1 axis (that is, the stresses act normal to face 1). As another example, S12 stresses occur on face 1 of the element (perpendicular to the local 1 axis) and are acting in the direction parallel to the local 2 axis (that is, the stresses act parallel to face 1, like shearing stresses). The Figure 10 shows examples of each of these basic types of shell stresses. Structural analysis program reports internal stresses for shell elements at the four corner points of the appropriate face of the element [15]. Figure 10. Stresses on thick shell elements Table 5 Stress at various members and links of building S.N. Stress Bare Masonry Composite Gap Hook Spring Rigid, N/mm2 Frame 1 S11 T 24.19 20.347 6.531 6.594 6.581 6.58 6.134 C -22.81 -19.391 -4.866 -4.84 -4.36 -4.867 -5.82 2 S22 T 34.83 20.027 8.923 6.422 6.536 6.54 6.061 C -33.02 -19.531 -4.953 -3.81 -3.89 -3.896 -5.806 3 S12 T 0.78 5.275 0.855 0.754 0.762 0.762 0.584 C -0.623 -4.849 -0.922 -0.93 -0.93 -0.932 -0.579 4 S13 T 0.561 0.273 0.713 0.459 0.46 0.46 0.904 C -0.515 -0.512 -3.27 -2.01 -2.04 -2.047 -4.056 5 S23 T 0.683 0.285 1.206 0.86 0.876 0.876 1.406 C -0.507 -0.286 -0.806 -0.42 -0.42 -0.428 -1.042 Table 6 Response of test model (natural time period, sec., and displacement, mm) Description Time period from model Time period from IS 1893:2002 ∆x -∆x ∆y -∆y Mode-1 Mode-2 Mode-3 Bare frame 0.26 0.251 0.231 0.15 3.082 -3.082 1.815 -1.815 Masonry 0.18 0.15 0.11 0.15 0.392 -0.392 0.205 -0.205 Composite 0.063 0.058 0.043 0.15 0.142 -0.142 0.147 -0.147 Hook 0.061 0.058 0.045 0.15 0.151 -0.151 0.164 -0.164 Gap 0.0603 0.057 0.0449 0.15 0.155 0.155 0.182 -0.182 Spring 0.065 0.058 0.045 0.15 0.178 -0.178 0.193 -0.193 Semi rigid 0.055 0.052 0.043 0.15 0.10 -0.10 0.095 -0.095 Rigid 0.051 0.048 0.041 0.15 0.068 -0.068 0.0738 -0.0738 Table 7 Stress on first floor roof, N/mm2 Descriptions Point 1 corner Point 2 corner Point 3 corner Point 4 corner Point 5 middle Bare frame (S11) 0.85 1.12 0.80 1.02 -0.24 Bare frame (S22) 0.091 0.10 -0.13 -0.46 -0.58 Masonry (S11) 0.094 0.121 0.76 0.58 0.309 Masonry (S22) -0.348 -0.1966 -0.275 -0.368 -0.51 Composite (S11) -0.33 -0.061 0.105 0.172 0.3772 Composite (S22) 0.884 0.410 0.438 0.660 -0.678 Rigid (S11) 0.192 0.185 0.113 0.097 -0.39 Gap (S11) 0.364 0.757 0.132 0.227 0.098 Gap (S22) 0.382 0.309 -0.104 0.45 -0.350 Hook (S11) 0.362 0.626 0.136 0.220 0.331 Hook (S22) 0.055 -0.0094 -0.152 -0.182 -0.458 Spring (S11) 0.254 0.621 0.173 0.273 -0.028 Table 8 Stress on wall, N/mm2 Descriptions Point 1 corner Point 2 corner Point 3 corner Point 4 corner Point 5 middle Masonry (S11) 0.0281 0.034 0.031 0.030 -0.013 Masonry (S22) -0.064 0.0203 0.155 -0.1269 0.10 Rigid (S11) 0.010 0.0115 -0.051 -0.0237 0.0055 Composite (S11) 0.0020 -0.0079 -0.0072 0.0017 0.0024 Composite (S22) 0.0051 0.0011 0.0050 0.0093 0.00288 Higher stress found in the connection of timbers, connector inters face and plank area this scenario show that timber members are must responsible for withstand all types of stress of house and increase the seismic performance of the buildings. Among all the model analysis the axial stress along X-axis S11 and S22 is found in masonry and bare frame model and in case of gap, hook, composite, rigid and spring model have less (Tables 5-8). It be clearly seen that model with wooden band and band connector having less amount of stress, i.e. wooden member responsible for to counteract out of plane failure and in plane failure. 4. Conclusion The modified structure having joint connecting elements such as gap, and spring perform better than the existing one but still lacks fulfilling the required purpose hence another modification is made reducing the size of opening and its placement is at center of wall and finally linked elements and connectors are introduced between timber elements in the model also enhanced the better response of the seismic performance and under seismic performance of the structure. Specific conclusions: - introduction of timber joist, beam and column in stone masonry house increase the base shear and reduces the time period and increase the stiffness of the structure. Finally, the response of structure against seismic force is improved by using connectors; - although timber frames and bands enhance the structural performance under seismic excitation in plane and out of plane stresses, where as the major contributing element to withstand external load is stone masonry as load path shown; - doors, windows, bands and band connector contribute in controlling the localized stress and create the box effect of the house globally and perform the good behaviors under the seismic forces. From the result it can be conclude that there is different contribution in lateral stiffness of the building model of different connecting element.

About the authors

Govinda Khatri

Mid-Western University

Author for correspondence.
Email: govindkhec@gmail.com

lecturer of the Faculty of Engineering

Post Box 21700, Birendranagar-9, Surkhet, Federal Democratic Republic of Nepal

Govind Prasad Lamichhane

Pokhara University

Email: govindkhec@gmail.com

Dean in Mid-Western University, Associate Professor of the Faculty of Science and Technology, PhD in Technical Sciences

Pokhara Metropolitan City-30, Lekhnath, Kaski, Federal Democratic Republic of Nepal

References