Experimental investigations on the initiation of external suffusion

This research study presents the experimental investigations conducted on a laboratory constructed model in the laboratory of hydrodynamic, surveying the phenomenon of external suffusion on the interface of three different soil layers: clay/moraine, sand and gravel.;Several different types of models (in geometry or component materials) were constructed and subjected to a water level increase in the upstream preventing any over topping, which resulted in hydraulic gradient changes.;The first series of tests, including seven type of test, consisted of models with different geometries and gravel grain size distributions. The obtained results showed no evidence of considerable suffusion on the interface of clay/moraine and sand, while the visual and quantitative data proved the presence of suffusion on the interface of sand and gravel.;It was shown that increasing the length of the filter layer or using step like slopes may not have a great influence on the initiation of suffusion, whereas the fine-grained gravel has a great impact on the erosion rate.;This work surveyed the process of external suffusion on the interface of soil layers in an embankment, caused by water augmentation in the upstream. Considering the constructed models as representing a part of an embankment, the clay/moraine formed the core, the sand was used as filter and the gravel performed the role of the pervious layer.;The second series of tests, consisting of four types, was divided into two parts. Using the most critical test from series one, four tests were executed: two type with different sand grain size distributions, finer and coarser compared to the critical type from series one, and two test types with gravel with different grain size distributions, more and less porosity compared to the critical test type from series one.;The results revealed that types of sand with finer grains wash out more easily, and coarser sands have a lower rate of suffusion.;Also, gravel with lower porosity can prevent suffusion easier than gravel with higher porosity. With all of this data as the result, focusing on the critical hydraulic gradient on the initiation of movement, one could conclude that despite the differences in the type of tests, the critical hydraulic gradient has approximately the same value (≈ 0.0332 +/- 0.0007).