A laboratory permeameter study of geotextile-soil retention in cyclic flow

In the absence of an extensive body of laboratory and field data, empirical criteria for soil retention in dynamic or cyclic flow are not yet well-defined with reference to a margin of safety. A performance-based approach is taken in this study: the method of investigation involves laboratory tests on a total of seven geotextiles (needle-punched nonwoven and woven materials) and a total of four uniformly-graded soils (non-plastic fine sand and coarse silt). Filtration compatibility in unidirectional and cyclic flow reversal is evaluated using two rigid-wall permeameters: a small bench-mounted device, and a large floor-mounted device. Analysis of the results addresses the effects of specimen size (small and large), sidewall friction and stress distribution, and examines the influence of filter ratio (AOS/Dn), hydraulic gradient (i) and confining stress (σʹ) over a range of cyclic flow reversal times or wave period (T). A novel analytical framework is proposed from the permeameter test results, to unify AOS/Dn and a hydromechanical index that accounts for the combined effect of hydraulic gradient and confining stress. The framework provides a distinction between the benign actions of mass loss through the geotextile by washout, in contrast to the more problematic action of piping. A filter ratio AOS/D₈₅ appears better-suited to interpretation of the data than AOS/D₅₀. The framework is used to examine the margin of safety inherent in current design guidance. Independent verification of the framework through comparison with other laboratory studies, and a consideration of field observations reported by others, leads to a recommendation that AOS/D₈₅ ≤ 1 to address undue conservatism in design guidance.Applied Science, Faculty ofCivil Engineering, Department ofGraduat