SHAKING TABLE TEST AND SEISMIC BEHAVIOR OF JUNCTION WALLS (L-SHAped masonry wall)
University of Moratuwa, Moratuwa, [email protected]
Dr C.S. Lewangamage
Department of Civil Engineering, University of Moratuwa, Moratuwa.
Abstract: The present paper describes the performance of masonry walls against different types of seismic performance with some modifications in their arrangement and material used. The proper experimental study is needed to analyse the detail performance of masonry structure in both planes (in-plan & out-of-plane) and it was done under several categories under dynamic shaking table. There were several types of research conducted some basic problems in masonry type structures and elements without any reinforcement and then they summarized the results with the performances of different types of seismic waves. After that, they did some modifications in masonry arrangement, brick types and analysed the performance. Also introduced different types of reinforcement like Polypropylene (PP-band), Glass Fiber Reinforced Polymer (GFRP), Carbon Fiber Reinforced Polymer (CFRP) or combinations of these and observe the improvements in the crack initiation, shear strength behaviour, displacement capacity and energy absorption. And also, some experiments were done for large-scale building and figure out the results of seismic studies. The subject of this study is mainly about, the past results up to now under shaking table test and use of different types of reinforcements and how they performed with and without reinforcement in both planes. The limitations of the experiments and how to overcome those problems with proto-type models (scaled parameters), factor selections for earthquake waves are also described in this paper.
Keywords: Masonry wall; Shaking table test; cracks; in-plane; out-of-plane
Masonry structure is those structures which are built by individual units together with the help of mortar. Easy of construction, Low cost and Highly durable, Fire resistance, Architectural appearance, Heat & Sound insulation and freely available materials are some of the advantages of masonry-based buildings. Masonry structures are widely used in most of the developing countries. In early periods, masonry buildings were built with a less priority against seismic risks. But from the last few decades losses due to severe earthquake all around the world shows that the importance of resistance against the Seismic load in masonry buildings.
Sri Lanka is located in an aseismic zone away from major 12 or 13 plate boundaries or any active faults. However, there was an incident which cost more than 2000 deaths & collapses of more than 200 houses due to the earthquake on 14th April 1615 in Colombo. since this earthquake, there were several minor earthquakes were observed within the region around Sri Lanka. Also, the formation of a new plate boundary dividing Indo-Australian plate increase the possibility of the earthquake events. Considering above facts, Sri Lanka is no more in the safety region against seismic attacks. So, the study on masonry walls against seismic loads and retrofitting methods in masonry walls become more important. This research mainly focusses on the In-Plane & out-plane seismic behaviour of the Junction Wall (L-shaped) with(GFRP) and without reinforcement.
2. Seismic Behaviour of URM walls
Most of the buildings in all around the world are built by using unreinforced masonry (URM) walls. These types of building more vulnerable to seismic waves. For earthquake disaster mitigation in developing countries, it has been essential to analyse the behaviour of non-engineered masonry structures and seismic performance using one directional horizontal shaking table. In order to study, 3m*3m*3m model house with openings which was constructed by Pakistan made bricks were tested. The main purpose of below test is an understanding dynamic collapse characteristic of masonry.
The test house walls were built up with English bond and three out of four have the opening for doors and windows. The weight of the total test house around 10.23ton. The base plate of the model was assembled by H steel at the centre of shaking table. Acceleration and displacement were measured after application of several earthquake waves to the shaking table. Servo and strain gage-type acceleration sensor was used to measure acceleration & image processing technique were used to measure displacement. Several input waves are given to the shaking table and total collapse occurred against 1995 JMA KOBE (NS) 100cm/s (110%) earthquake wave (Hanazato et al., 2008). The tested wave list and damage of the specimens were listed below table 1.(Hanazato et al., 2008)
Table 1: Input wave list of shaking table test
Damage of the specimen
2003 Iran Bam Eq. L (EW ) TS=0.79 75cm/s
2003 Iran Bam Eq. L (EW ) TS=0.79 100cm/s
1995 JMA KOBE (NS) 100cm/s (110%)
Sin 15Hz 1G 50 second
Sin 1Hz 10cm 0.4G 20 second
Pulse Shock 1 40cm/s
Pulse Shock 2 -40cm/s
Pulse Shock 3 30cm/s
No Growth crack
2003 Iran Bam Eq. L (EW) TS=0.79 100cm/s
1995 JMA KOBE (NS) 100cm/s (110%)
The crack pattern occurred in wall with openings shown in below figure 1. (Hanazato et al., 2008)
Figure 1 (Initial Cracks and Responses of South Wall in Pulse Shock No.7)
The above shaking table test, one of the outlines of masonry house collapse process was shown. Careful construction work would make a strong house. Also crack pattern by numerical analysis and shaking table test was quite similar.
3. Seismic Behaviour of Reinforced walls
In the past few decades, there were many researchers has been focused on the dynamic response of masonry type structures. Earthquake Engineering Research Center of the University of California Berkley has conducted shake table tests on four different single-story house models, unreinforced and partially reinforced, to determine their seismic capacity using local masonry construction characteristics (Manos, 1983). Zarnic et al. carried dynamic table tests on scaled reinforced concrete masonry infilled models and determined the damage pattern of in filled masonry structures. This test was conducted to two models and test results were compared (R. Zarnic, 2000). Also, some of the masonry walls tested against seismic waves by using masonry strengthened material such as Basalt Fiber Reinforced Polymer (BFRP). Confined masonry structure, (in explained concrete beams and columns used to confine the masonry walls) also tested by some researchers and evaluated the structural capacity, stiffness, deformation, failure patterns & energy dissipations properties.
Polypropylene (PP band) and Glass fibre reinforced polymer (FRP) are some important reinforcement techniques used to mitigate the earthquake effects in normal URM walls. M. Umair Saleem et al. carried a detailed dynamic analysis of URM, PP-band retrofitted, FRP retrofitted and FRP + PP-band retrofitted house models is performed on timber roof burnt brick masonry structures (Saleem, Numada, Amin, & Meguro, 2016). Very few amounts of (0.006 FRP reinforcement ratio) FRP material is used to control the cost factor. The amount of FRP was calculated based upon series of diagonal compression tests with different reinforcement ratios and FRP layouts (Saleem et al., 2016). In masonry walls, FRP reinforcements act like a strong shear links in between two masonry units. The application of FRP on masonry walls was restricted only on the outer face of the wall & the shape of vertical strips.
3.1 Scale factors for Models
Normally shaking table test have some restrictions to do the practical’s. The main reason is the availability of the shaking table and capacity of the shaking table. And also, it is very hard to construct and test the seismic behaviour of the large-scale building. Because of that, it is very important to construct the model with the suitable scale factors. Normally proto-type house models or masonry walls were constructed by following the similitude laws. Cauchy similitude law and Fraud similitude law are most commonly used similitude laws for proto-type modelling. Cauchy laws give the relationship between inertial and elastic restoring forces whereas Froude similitudes laws become more important when significant gravitational forces are under consideration as Froude similitudes laws give the relationship between inertial and gravitational forces (Carvalho, 1998). Normally we can use both similitude criterions for these kinds of experimental programs. Due to the limited size of shaking tables and requirement of additional weight for Froude similitude law, Cauchy similitude laws have picked rather than Froude law for modelling proto-type samples.
In the design of models, the basic rule is it should be reproducing the dynamic behaviour of the actual model, failure mechanism & physical characteristics. If the proto-type model was adapted a geometrical scale factor of 1:4, the relations for the different parameters in terms of the geometrical scale factor according to Cauchy law showed in below Table 2.(Saleem et al., 2016)
Table 2: Scale factors for Cauchy similitude law
It was difficult to similitude all the strength characteristics of the actual one and its prototype model at the same time, therefore to have the same failure mode, the shear strength of the building model take as the main criterion and it is reduced as per scale in above Table 2.
3.2 Retrofitting scheme
Normally Retrofitting of additional reinforcement to the basic masonry is carried after the construction and curing process. As mentioned above, PP band, FRP, PP band + FRP retrofitting finished after 28 days of curing to check the effectiveness of this kind of reinforcement against seismic waves.
3.3 Final Results of behaviour of different retrofitting techniques
URM types masonry structures are highly vulnerable to small ground motions. These types of masonry may not remain serviceable condition after affect of small input motions. Retrofitting by using PP-band has increased the strength capacity of non-reinforced masonry, but experiments show that some of the bricks did not remain their actual positions after the severe input motion. The Crack was initiated above four house models from the doors & windows openings towards the corner position. FRP of small quantity not enough to prevent these crack initiations. The combination of PP-band & FRP retrofitted model has performed very well rather than those of PP-band alone FRP alone retrofitting units. Also, these combinations of retrofitting remain service limit even for a higher value of 1.2g motion input. The detail results of these four housing units against seismic waves showed in below Table 3.(Saleem et al., 2016)
Table 3: Summary of Shaking table test results above four types of retrofitting models
4. Out-of-plane dynamic behaviour of URM walls and FRP reinforced walls
Masonry failure can be assured both In-plane failure & Out-of-plane failure. The major disadvantage of Masonry brick wall is it is low resistance against tensile stresses and lateral loads. The walls which are perpendicular to the seismic wave subjected to out-of-plane bending results in out-of-plane failure featuring vertical cracks at the wall corners and middle of the walls which may due to the inadequate flexural strength of unreinforced masonry or due to lack of integrity of an adjoining structural. Only a few researchers revealed the out-of-plane behaviour of the masonry walls/ buildings due to the inability of taking account of wall interaction with the other part of the building and fragile nature of the brick masonry. So, it is necessary of analysing the seismic behaviour of the brick masonry wall junction (“L-shaped”) in terms of crack initiation, lateral drifts and failure patterns and comparison between the un-reinforcement wall and reinforcement (GFRP) wall.
Tu et al. investigated the behaviour of masonry against out-of-plane for infills in RC frames by shaking table test on one story full scaled buildings. Four specimen samples were loaded under excitations with increasing amplitude until collapse. Results showed that boundary elements as concrete columns and thickness of the masonry walls can increase the out-of-plane resistance of masonry panels significantly (Y.H. Tu, 2010). Out-of-plane seismic response of a U-shaped tuff masonry wall was studied on the shaking table in Rome, results show that the U-shaped masonry tuff behaves as a rigid member undergoing rocking motion at the base and some vertical crack ocurred in between main façade and transverse walls due to poor transversal bond (O. Al Shawa, 2012). Also, the shaking table experiment for full-scale stone masonry facade was conducted by Cost et al. and results obtained that, one-sided rocking happened before the collapse and finally overturning happened (A.A. Costa, 2013).
Due to lack of literature on the effect of retrofitting on out-of-plane dynamic behaviour, Razieh et al. conducted out-of-plane shaking table tests on three I-shaped masonry walls. Specially one of above masonry wall strengthened by unidirectional glass fibre reinforced strips and results were observed. From this experiment, the behaviour of URM specimen is compared to the reinforced masonry specimens and the improvement & effectiveness of strengthening with GFRP composites is analysed. The main expectation of this experiment was improvement in out-of-plane performance of URM walls by adding glass fibre reinforcement. Figure 2 shows the arrangement of specimens.(Sistani Nezhad, Kabir, & Banazadeh, 2016)
Figure 2. The arrangement of URM masonry and FRP reinforced masonry
In above experiment, the behaviour of the walls is studied in the case of displacement capacity, damage state and base shear. A rocking mechanism is observed under the end test record with the difference, that the URM wall damage was occurring through the height of the wall and the major crack line in the reinforced specimen was observed at the base while the other parts of masonry wall remained undamaged. This indicates that a special attention must be paid to base connection of masonry specimen at the reinforcing of masonry walls. An improvement in the shear strength and hysteretic dissipated energy of the reinforced masonry specimen is observed in accordance with the URM wall(Sistani Nezhad et al., 2016).
Masonry structures have more vulnerable to minor to moderate earthquake motions. There were several experiments were done to improve the masonry building in terms of fundamental frequency, displacement capacity, shear strength, crack patterns and energy absorption. Also, most of the experiment revealed the in-plane behaviour against seismic motion under shaking table test and very few number of researchers carried out-of-plane behaviour due to some limitations and fragile nature of masonry walls. In masonry structures, crack initiation is very high in openings of the walls, corners and wall junctions. So, introducing proper reinforcement technique to mitigate cracks should be cost-effective and easy to implement. A large number of experimental results and proper study on wall junction with and without reinforcement under dynamic behaviour are the major requirement to overcome these problems against seismic loads.
I would like to acknowledge Miss.Indunil, Research assistant, University of Moratuwa, who provide some useful materials to complete this Literature review.
A.A. Costa, A. A. (2013). Out-of-plane behaviour of a full scale stone masonry façade. J.Earthquake Eng. Struct.
Carvalho, E. (1998). Seismic testing of structures. Paris, France: Proceedings of the 11th European Conference on Earthquake Engineering.
Manos, G. C. (1983). Shaking table study of single-story masonry houses.
O. Al Shawa, G. F. (2012). In Out-of-plane seismic behaviour of rocking masonry walls (pp. 949-968). J. Earthquake Eng. Struct.
R. Zarnic, S. G. (2000). Shaking table tests of 1:4 reduced scale models of masonry infilled reinforced concrete frame buildings.
Y.H. Tu, T. C. (2010). In Out-of-plane shaking table tests on unreinforced masonry panels in RC frames (pp. 3925-3935). J. Eng. Struct.
Hanazato, T., Minowa, C., Narafu, T., Imai, H., Kobayashi, Q. A. K., Ishiyama, Y., & Nakagawa, T. (2008). Shaking Table Test Of Model House Of Brick Masonry For Seismic Construction. 14th World Conference on Earthquake Engineering (14WCEE).
Saleem, M. U., Numada, M., Amin, M. N., & Meguro, K. (2016). Seismic response of PP-band and FRP retrofitted house models under shake table testing. Construction and Building Materials, 111, 298–316. https://doi.org/10.1016/j.conbuildmat.2016.02.073
Sistani Nezhad, R., Kabir, M. Z., & Banazadeh, M. (2016). Shaking table test of fibre reinforced masonry walls under out-of-plane loading. Construction and Building Materials, 120, 89–103. https://doi.org/10.1016/j.conbuildmat.2016.05.097