CYCLIC UNDRAINED STRESS-STRAIN BEHAVIOR OF DENSE SANDS BY TORSIONAL SIMPLE SHEAR TEST

Abstract

In the first part of the article a review of laboratory cyclic undrained test methods is presented. The features of the cyclic torsional simple shear device which was newly developed to assess the resistance of saturated dense sand against cyclic undrained loading are described. The test results are represented by the relationship between the relative density D_γ and the cyclic shear stress amplitude normalized by the effective mean principal stress at consolidation τ_cy/σ'_mc for which a certain value of double amplitude shear strain is observed at a certain number of loading cycles. Cyclic undrained tests were performed for 1) isotropically and anisotropically consolidated specimens, 2) specimens initially sheared and not sheared during consolidation and 3) specimens prepared by different methods, using two different sands, Toyoura Sand, a clean, uniform, fine sand having no fine particles and Sengenyama Sand, a medium fine sand including fine particles to some extent. For any test condition, specimens denser than some critical densities showed extremely high resistances which were much larger than expected from the concept that the cyclic undrained strength be proportional to relative density. For identical values of relative dinsity, loose Sengenyama Sand specimens had strengths similar to loose Toyoura Sand specimens, while dense Sengenyama Sand specimens had strengths much less than dense Toyoura Sand specimens. Loose specimens prepared by the air pluviation method had strengths similar to ones prepared by the wet tamping method, while air-pluviated dense specimens had strengths much lower than wet tamped dense specimens. Other factors affecting the cyclic undrained strength of sands were also investigated. Effects of any factor investigated on the cyclic undrained strength were very large for dense specimens, but only slight for loose specimens.