CC BY
9
МЕХАНИКА. ТРАНСПОРТ. МАШИНОСТРОЕНИЕ
0
Gao Jianping, Yin Hongyan
УДК 693.15
SHAKING TABLE TEST OF A 1/5 SCALED INTEGRATED MASONRY-IN-FRAME SYSTEM MODEL
1. INTRODUCTION.
In recent years, it becomes increasingly popular nationwide to add more stories to the existing buildings due to its particular techno-economy advantages such as investing little, occupying no more land, beautifying the cities' appearance, shortening the construction duration, sharing the infrastructure and extending the service life of existing buildings, etc. Especially, story-adding suits the China's situations because of excessively more people, relatively less land and the underdeveloped economy, which plays an important role in mitigating the pressure of needing for more building sites, improving the residential conditionJ-expediting the upgrade and modernization of the cities' shabby zones. The integrated masonry-in-frame system is often used in practice, but its seismic performance and seismic design methods are not clear for lack of experimental and theoretical research. This sort of structural systems, therefore, are not encouraged to be employed currently in the Chinese Code: Technical Specifications for the Story-adding of Existing Masonry Structures (CECS78Je96). Consequently, underlying problems maybe endanger the safety of those buildings of their type constructed or to be constructed in the seismic regions, since no
relevant building codes to follow in the design and construction, which in turn affects its wide application to the practice. Based on the above background, model test was carried out for the first time, to investigate into its seismic behavior and calculating model.
2. SEISMIC DESIGN CONCEPT PROPOSED.
The innovative design concept to be presented originated from the upgrading of the Teaching Building of a college (Gao Jianping et al. 2003). That is, the new and old structure are separated from each other in elevation, whereas in the horizontal direction, they are connected by junctions to resist lateral force together, simultaneous with the seismic retrofitting of the masonry building. A new-style junction designed plays a key role in realizing the design concept. As shown in figure1, the external columns on either side of the masonry building, erected closely against walls, are connected to the two rows of internal columns that break through slabs at the center corridor. Two thin beams that sandwich the internal wall panel are linked by shear keys, which will help transfer horizontal forces between the two structural systems. The holes are designed to be 50mm deeper than the height of thin beam and shear key to permit the free settlement of the frame structure. In the mean time, both sides of some transverse walls
Fig. 1. Junction between new frame column/beam and existing masonry at floor level.
ИРКУТСКИЙ ГОСУДАРСТВЕННЫЙ УНИВЕРСИТЕТ ПУТЕЙ СООБЩЕНИЯ
are thickened by 100mm-thick concrete with steel meshes bolted to the walls.
3. OUTLINE OF THE TEST
3.1 Design of the Model.
Considering the capacity of the
shaking table to be used, a 1:5 scale 2-bay 2-story RC frame that spanned the 2-story brick masonry built earlier was constructed according to the similitude requirements. The plan and elevation of the model were shown in Figure 2. The story height of the masonry and frame was 580mm, 620mm, respectively, and the total height of the model was 2.72m. The dimension of the model brick was 48x38x24mm, its compressive strength was 7.5 MPa. The compressive strength of the cement mortar was 7.5 MPa. The floor slab was 40mm thick, with concrete strength of 20 MPa. Similar to the prototype, the model was transverse wall load-bearing, and continuity beams were cast at the second story. The reinforcement of the columns and beams were constructed according to the Chinese practice of seismic detailing. A reinforced concrete slab of 80mm thick was cast, on which the 2-story masonry was anchored. The elevation of column root was the same as that of the base slab of the masonry. To prevent the masonry from overturning or sliding, a bottom beam of 300 x 150mm was cast around the original masonry building model, and 8 bars of 20mm in diameter were anchored into the bottom beams (2 bars each side). The physical parameters derived from similitude requirements were showed in Table 1.
3.2 Instrumentation and Experimental Setup.
The test was carried out at the Structural Testing Laboratory of the Harbin Institute of Technology, Harbin, China. Displacement transducers and accelerometers were used to measure the lateral displacement and acceleration at each story. Figure 3 shows the experimental setup and instrumentations for the
Fig. 2 Plan and elevation (Unit: mm).
Fig. 3. The test model on the shaking table.
shaking table test. The instrumentation was confined to only Axis-2 due to the limitation in the number of available channels. The experimental results were interpreted, assuming that the behavior of Axis-2 represented that of the whole model structure in earthquake simulation test. This model was then subjected to the shaking table motions simulating Taft N69W, El-Centro N21E and Tianjin, respectively. The earthquake motions were input along the
Table 1
Physical parameters according to the similitude requirements
Dimension Modulus Stress Frequency Acceleration Time and Period
0.2 1 1 2.236 5 0.447
МЕХАНИКА. ТРАНСПОРТ. МАШИНОСТРОЕНИЕ
©
transverse direction. The three time-history records were adjusted by adjusting the PGA and compressing the time scale by the factor of V5 according to similitude law. Therefore, compensation for the difference in the mass was artificially made by adding iron blocks. The effective weight of the model with iron blocks was estimated to be 130.6 kN while the weight ideally required by the similitude law was 140.1 kN. The error was approximately 7%.
4. RESULTS OF EARTHQUAKE SIMULATION TESTS AND INTERPRETATION.
4.1 Global Responses.
From these tests, the natural frequencies and damping ratios of the model were obtained by using the Fourier transform method. Table 2 showed that natural frequency tended to decrease and damping ratio tended to increase as the model experienced higher levels of ground motions. The fact that damping ratios were about 5J¥ accorded with the actual condition of RC structures and masonry buildings.
As shown in figure 4, the first modal shape of the upper two stories was of typical shear mode at elastic stage. The modal shape of the lower two story was a little steep. Obviously, it was the brick panels that contributed to the greater stiffness. At the transition floor, the lateral curve of the first modal form still assumed continuity, indicating that the modal shape could be considered as a shear mode as a whole.
4.2. Crack Development and Failure Mode.
As the excitation increased from 0.15g to 0.4g (El-Centro), four diagonal cracks at the walls were observed to extend to the column edge from the first to the second story at 1-axis and 3-axis (Figure 5). During this stage, only a slight change of dynamic characteristic (natural frequency) was found, which indicated that the model performed from elastic state to a slightly cracked state. No cracks were visible at the internal walls of 2-axis and other external longitudinal walls. As the amplitude of the excitation increased to 0.4g
(Taft), the shaking table system was motivated by itself, which caused the stretching and widening of the original cracks. Except for few minor cracks noticed, no other big cracks occurred. Diagonal cracks developed at the walls under the lateral force and vertical load, which could be attributed to the deficient shear resistant capacity of brick masonry. While the typical shear failure mode indicated that the vibration model was inter-story drift model. During the whole run of the excitations, there were no horizontal cracks observed at the walls, the frame columns did not show any apparent cracks, too. The model did not show serious damage even after the excitation increased and reached the target PGA at the run of EL-Centro (0.6g), Taft (0.76g) , Tianjin (0.88g).
4.3. Displacement and Acceleration Response.
Typical measured time history of absolute displacement and acceleration of each story was shown in figure 6. The acceleration response increased as the excitation increased, each story shifted along the same direction at the same time. The maximum acceleration amplification factors increased with the floor as shown in figure 7, however, which decreased after the model cracked, owing to the degradation of stiffness and increase of damping ratios resulting from the occurring of cracks. The acceleration amplification factors of the top story ranged from 1.3 to 2.5. Figure 8 showed that the distribution of shear forces was of a ladder shape, indicating the behavior of the model was governed mostly by the first frequency. The time history of the floor displacement of each
Fig. 4. Measured vs calculated 1st modal form.
Natural frequency and damping ratio
Table 2
Identification of test Natural frequency Dampinq ratio(%)
After El-centro 0.3q 7.5415 0.46
After El-centro 0.6q 7.2482 0.50
After Tianjin 0.45 q 7.8168 0.47
After Tianjin 0.6 q 7.7800 0.55
Fig. 5. Crack pattern of the brick panels at Axis-2 and Axis-3.
Fig. 6. Time histories of absolute displacement and acceleration for TAFT (PGA=0.4g).
Fig. 7. Maximum acceleration amplification factors.
Fig. 8. Measured story shear envelopes.
story was of typical inter-story mode, and acceleration response was maintained in the shape of an inverted triangle approximately. From the profiles of measured inter-story drift indices in Figure 9, it could be noted that the inter-story drift changed abruptly at the transition floor (the third story) as well as the adjacent floor (the second story). The relationship between the top displacement and
the base shear was shown in Figure 10, in general, the model behaved linear-elastically under the design ground motion.
A time-history analysis program was developed using the inter-story drift model, the MATLAB language, and the Higher-order Single Step B-Method (Wang Huanding et al 1999), a numerical integration method with higher accuracy. The calculated time history of
МЕХАНИКА. ТРАНСПОРТ. МАШИНОСТРОЕНИЕ
ши АО ОО ©
1
G J cent ro fcnis
Fig. 9. Story-drift displacement.
Fig. 10. Top displacement vs base shear.
Fig. 11. Measured vs Calculated Time History of Acceleration For the Top Floor Level.
acceleration for the top floor agreed well with the test results, as shown in Fig.11, which proved it appropriate to adopt the inter-story drift model as calculating model.
5. SUMMARY AND CONCLUSIONS.
1.The seismic response recorded, modal shape identified and crack pattern observed all come to the conclusion that the vibration model and failure pattern is inter-story drift model.
2. The transition floor as well as the floors adjacent to it may be the weakest story. Therefore, attention should be paid to control the rigidness of inter-story within a reasonable range when designing this type of building.
3.Test and analysis results show that the brick masonry building of brittle failure can be remodeled into ductile frame-structural-wall system, using the upgrading and story-adding
method proposed herein, while the seismic potential of existing building can be better utilized and the seismic performance of the complex structural system can be improved accordingly.
BIBLIOGRAPHY
1. Gao jianping et al.(2003). The Structural Design for adding two more stories to an existing masonry structure. Structure in low temperature region, 95: P.26-27.
2. Wang Huanding et al. (2003). A high order single step-B method for nonlinear structural dynamic analysis. Journal of Harbin Institute of Technology. 10: P.113-119.
