Stress-path laboratory tests to characterise the cyclic behaviour of piles driven in sands

AbstractRecent publications have emphasised the importance of addressing cyclic behaviour when designing piled foundations. Laboratory tests may be conducted to provide site-specific cyclic soil characteristics, but questions arise concerning: (i) how to take into account the pile installation process and (ii) how to apply the results to assess pile capacity and deformation responses under cyclic loads. This paper describes an investigation into the cyclic behaviour of Dunkerque and NE34 Fontainebleau sands, performed to support and to help analyse field-scale and model pile cyclic loading tests on the same soils. Forty computer-controlled, locally instrumented, cyclic and static triaxial tests were performed, following testing schemes that were developed to reflect the conditions adjacent to the pile shafts. Assessments were made of how the cyclic variations in stress imposed during installation and the period allowing for the two types of sand to creep following such ‘installation’ effects, affect the response to subsequent cycling. Constant-volume cyclic tests, involving up to 4500 cycles, were imposed from alternative sets of initial conditions that revealed the relationships among the cyclic deviatoric amplitude, the changes in mean effective stress and the number of cycles as well as the permanent strain accumulation and the cyclic stiffness characteristics. Monotonic compression and extension tests are also presented for both sands to help frame their strength, stiffness and critical state properties. A synthesis with previously obtained cyclic pile test trends confirms the practical applicability of the results obtained