Search for: All records

This paper presents the results of soil characterization and element tests of Ottawa F65 sand. The data presented is intended to be used as calibration material for the prediction exercise conducted as part of the Liquefaction Experiments and Analysis Project (LEAP 2017). The databank generated includes soil specific gravity tests, particle size analysis, hydraulic conductivity tests, maximum and minimum void ratio tests, and cyclic triaxial stress-controlled tests. An effort was made to ensure the consistency and repeatability of the test results by reducing the sources of variability in the sample preparations and increasing the number of tests. The uniformity of the soil was evaluated by conducting tests on samples from five different batches. The results showed that the sand is uniform among the five batches. Due to significant variability in previously reported maximum and minimum void ratio results, the effects of the test operator were studied by comparing test results obtained from three different operators. For the triaxial tests, a constant height dry pluviation method was used for sample preparation. To eliminate the effect of the human error in maintaining a constant drop height and to ensure consistency of the sand fabric between different samples, a device was developed to facilitate the sample preparation. The cyclic triaxial experiments were performed using three different soil densities, and a liquefaction strength curve was obtained for each density based on a 2.5% single amplitude axial strain criteria. The developed databank in this study was made publicly available for the community through DesignSafe. 

Constitutive relations used to describe the stress-strain-strength behavior of soils in cyclic loading are known to play a critical role on our ability to predict the response of geo-structures to seismic loading. The extent and intricacies of this role, however, are highly problem-dependent and often difficult to discern from the effects of other ingredients of a numerical simulation. Moreover, realistic assessments of constitutive models and numerical analysis techniques require detailed comparisons of their performances with reliable experimental observations. The experimental data that have been produced in the course of recent Liquefaction Experiments and Analysis Projects (LEAP-2015 and LEAP-2017) provide an opportunity for a more thorough assessment of the capabilities and limitations of constitutive models for sands over a wide range of strains. The LEAP experimental data along with a large number of cyclic element tests are used here to explore the performance of several constitutive models in numerical simulation of soil liquefaction and its effects on lateral spreading of mildly sloping grounds.