CHAPTER 6 Experimetnal study on steel-concrete-steel sandwich composite shell
6.4 Test results and discussions
6.4.3 Measured deflections of the shell
As aforementioned, LVDTs were installed on both top and bottom steel shells along the centerline of arch direction and width direction, respectively (as illustrated in Fig. 6.5(a)).
Locations of these LVDTs installed on the sandwich composite shells are shown in Figs.
6.5(a), 6.6(a) and 6.6(g). Based on these measurements, the exact deformed shapes of the arch and width centerline at each load stage could be obtained by linking the measurements of these locations. Finally, the deformed shapes of both outer top shell and inner bottom shell of the seven specimens SB1~SB7 are shown in Fig. 6.13. Exact deformed shapes of bottom steel shell of specimens SA1 and SA2 are shown in Figs. 6.14 and 6.15, respectively.
‐ 247 - For specimens SB1~SB7, seven LVDTs were installed on inner bottom whilst eight LVDTs were installed on the outer top steel shells. For the bottom inner steel shell, three LVDTs were installed at the middle arch whilst five transducers were installed along the width direction at middle span. For the outer top steel shell, six LVDTs were installed along the width direction at middle span whilst four LVDTs were installed along middle arch direction (shown in Fig. 6.5(a), Fig. 6.5(g)). From Fig. 6.13(a)~(n), several behaviors are summarized as follows:
1) During the early stage of the loading, the deflections at the loading point are not significantly larger than deflections of other points away from the loading area. This trend can be clearly reflected in the Fig. 6.13 when the applied load is less than 440 kN.
2) During the progress of the applied load increases to the first peak value, the deflections at the loading point increases significantly than the deflections of other points on both inner and outer steel shells. This means the deformation in the shell is a more local behavior rather than a global behavior, which also implies that local punching shear failure occurs and the core material is punched through.
3) Due to all the SCS sandwich composite shells are one-way supported, it is observed that the deflections along the middle arch strip are much smaller than the deflections along the width direction at middle span at the same loading level. This implies the arch strips of the shells act as the main beams transferring the applied load to the support, and of course the middle arch strip takes more loads compared with the edge arch strips. The width strips act as the secondary beams that disperse the load to the edge arch strips.
4) Specimen SB5 with HPC (fck=180 MPa) exhibits largest initial stiffness among all the tested specimens (shown in Figs. 6.10a and 6.0b). The stiffness of load-deflection curve is almost 379 kN/mm compared with a range of 132~180 kN/mm for the other
‐ 248 - specimens.
For specimens SA1 and SA2, unsymmetrical loading was applied. Six LVDTs were installed underneath the inner shell of the specimen. The deformed shapes of SA1 at different loads are shown in Figs. 6.14 and 6.15. From this figure, it can be seen that the half shell near the loading point is pushed down while the opposite side goes up.
Moreover, the deflection at the loading point increases much faster than the deflections of the rest locations. This implies that local punching shear occurs. The deflections at the far end from the loading point increase significantly as the load increases from 800 to 850 kN.
The deflection atPp2 is 85 mm which is much larger than 16.58 mm at Pp1. From the deformed shapes at different loading levels in Fig. 6.13(a) and (b), it can be observed that deflections of the locations in the vicinity of the loading point are significantly larger than that of other locations, which implies that the concrete is locally punched through.
Moreover, the phenomenon that nearer end to the loading point of the shell is pushed down and the far end is pushed up implies a global buckling occurred to SA1 at Pp2.
Fig. 6.15 shows the deformed shapes of the sandwich shell SA2 under different loading stages. From these figures, it can be seen that the whole shell is pushed down, which is different from SA1. It is observed that the deflections at the loading point are only smaller larger than other locations when the load is less than 440 kN. After the applied load is larger than 1000 kN, the deflection at the loading point become much larger than the deflections at other locations, i.e. the punching shear failure starts to develop. After that point, the deflections at the loading point become significantly larger than the deflections of all the other points. The difference becomes extremely larger when the structure achieved the first peak resistancePp1. This implies local punching shear failure takes place.
‐ 249 - From the above observations of deformed shapes of the sandwich shell under different loading levels, the failure modes of the structure can be easily observed through analyzing the deformed shapes at different loading levels. The local large deformation in the vicinity of the loading point takes place that implies the punching shear failure mode occurs to most of the specimens when these shells achieve the first peak resistancePp1. Moreover, SA1 is observed to exhibit different deformed shapes that relates to global buckling failure mode when the sandwich shell achieves the second peak resistancePp2.