Erosion and filtration of cohesive soils

It is commonly accepted that cracks can be sustained in cohesive soils, allowing increased seepage and subsequent erosion of the soil through these cracks. To investigate the erosion rate and the size of particles eroded through a crack, a series of experiments have been performed, following a procedure similar to the pinhole test. In these experiments, a sample of cohesive material is compacted in a 200mm long, 150mm diameter cylinder, and a 3mm pinhole formed through the sample. This modified pinhole experiment is able to determine the erosion rates and the size of particles which can be relied upon for self filtration at the filter interface. Results of the experiments show that there is a correlation between the size of the eroded particles and the original particle size distribution of the soil. For fine grained soils (with d(85)<75 mu m), there is some degree of aggregation, where the eroded particles are larger than the original PSD of the soil. The amount of aggregation appears to be directly related to the fines content of the original soil. No Erosion Filter Tests have been carried out on the same materials. A positive correlation has been found between the size of eroded particles (represented by d(85EP)) and the safe filter boundary (D-15bdy) defined by the NEF test. A relation D-15bdy/d(85EP)=5 is a lower bound for the majority of NEF tests. This introduces a theoretical grounding for the empirical design criteria as determined by NEF tests, described by Sherard and Dunnigan (1989).