SUEMASA Naoaki Tokyo Institute of Technology, Faculty of Engineering, Research Associate, 工学部, 助手 (80206383)
TAKEMURA Jiro Tokyo Institute of Technology, Faculty of Engineering, Associate Professor, 工学部, 助教授 (40179669)
HIROOKA Akihiko Tokyo Institute of Technology, Faculty of Engineering, Research Associate, 工学部, 助手 (70238400)
|Budget Amount *help
¥12,100,000 (Direct Cost : ¥12,100,000)
Fiscal Year 1991 : ¥3,400,000 (Direct Cost : ¥3,400,000)
Fiscal Year 1990 : ¥8,700,000 (Direct Cost : ¥8,700,000)
In order to investigate the dynamic behavior of soils with extremely low plasticity (Ip=2,5 ; silty soils), a series of cyclic triaxial tests was conducted on these soils and two others with Ip equal to 10 and 30. Extremely low plasticity soil speci13EA\ : mens were prepared by freezing in a specially designed mold since it is not possible to trim specimens from preconsolidated chunk of these soils. The cyclic triaxial tests were carried out on both normally and over consolidated soils with varying i13EA\ : nitial void ratio, initial shear stress (consolidated isotropically or anisotropically) and cyclic stress ratio. Sinusoidal waves with the frequency ranging between 0.01Hz and 1Hz were applied to the specimens. Conclusions derived from these tests a13EA\ : re as follows.
1.For a higher stress ratio, a lower frequency and a smaller plasticity index, during cyclic loading, a higher pore pressure is generated and the reduction in stiffness is more marked.
2.For normally consolidated spec
imen, cyclic loading causes the soil to build up a higher pore pressure and as a result a larger reduction of the stiffness than for over-consolidated specimen.
3.There is a marked difference in the dynamic response of low plasticity soils between specimens with and without initial shear stress. When the initial shear stress is small, liquefaction was observed and the stiffness is almost equal to 0 during c13EA\ : yclic loading. Otherwise, large irreversible plastic axial strain is induced by cyclic loading without total loss of stiffness.
4.If the same initial shear stress is applied to soil specimen, there is a unique relationship between the irreversible axial strain and the excess pore pressure caused by cyclic loading.
Dynamic centrifuge tests were also carried out using a rigid box in which a model caisson sits on a soil layr so as to study the dynamic response of these soils as a soil mass. Two types of input waves were applied to the soil mass, namely a sinuso13EA\ : idal and random wave simulating the El Centro record. In particular, for extremely low plasticity soil, dynamic centrifuge tests using overconsolidated soil were conducted to investigate the effect of the initial void ratio on the dynamic response. Conclusions derivel from these tests are as follows.
1.When a structure sitting on a soil layr is subjected to an actual earthquake, a larger settlement will result for soils with a smaller plasticity index.
2.In case of soils with extremely low plasticity, it was observed that the rigidity of the ground was reduced during the early stage of shaking and as a result, the acceleration amplitudes of the model casson was much smaller.
3.In this study, the settlement of the model caisson induced by the shaking is slightly larger for over-consolidated soil layr than for normally consolidated one. For this reason, the observed amplification of the model caisson's acceleration is gr13EA\ : eater for the overconsolidated case. Thus the overconsolidated soil layr was subjected to a higher cyclic stress because of its higher stiffness. This implies that the stability of soil layrs consisting of soil with higher stiffness is not always better, when the soil-structure interaction is taken into account.
4.The ratio of the observed acceleration spectrum of caisson to the input acceleration spectrum is larger at the dominant frequency but smaller in the high fequency region for the higher plasticity soil layr.
Using a laminar box, dynamic centrifuge model tests were also carried out to compare with the test results obtained using a rigid box under the same conditions. The ovserved settlement of the caisson is smaller for the laminar box than for the stron13EA\ : g box. Since in the rigid box, shear deformations are restricted by the two side plates, the input waves tend to be transmitted to the model caisson with less energy dissipated, and therefore imposing a more severe vibration on the caisson. However, 13EA\ : the general trend that larger settlement can be induced by shaking for soils with lower plasticity is confirmed in the test results using a laminar box.