抄録
Drained and undrained shear strength and creep characteristics of saturated Hokota sand were evaluated by triaxial compression (TC) tests. Among the specimens, the degree of compaction Dc = ρd/(ρd)max x100 %, where ρd is the dry density and (ρd)max is the maximum dry density for the compaction energy level (CEL) equal to the standard Proctor (1.0Ec), and the degree of saturation, Sr, were changed. Compressive strength qmax in undrained TC increases significantly with Dc, becoming significantly higher and lower than drained qmax as Dc increases and decreases from about 90 %. This result indicates that, despite that the drained strength is often used irrespective of drain condition in seismic design, the undrained strength should be used where relevant. For the same Dc, the effects of Sr at compaction on qmax are insignificant. In practice, CEL used in a laboratory test is often equal to 1.0Ec (Ec is compaction energy) and the field compaction is done at a water content higher than its optimum value for 1.0Ec, (wopt)1Ec. However, the actual CEL in the field is variable and usually unknown. CEL in the modern earthwork can easily exceed 1.0Ec. In that case, w higher than (wopt)1Ec may become too high in that, despite an increase in CEL, ρd may not increase noticeably while the strength and stiffness may not increase efficiently. Besides, Sr may become too high and the risk of over-compaction becomes high. On the other hand, "the optimum degree of saturation (Sr)opt" defined as Sr where (ρd)max is obtained at a given CEL is independent of CEL. So, it is recommended to control the field Sr to be equal to (Sr)opt so that (ρd)max is obtained for the current unknown CEL, while ensuring the (ρd)max value to be high enough to achieve the design drained or undrained qmax value.
