A RESIN IMPREGNATION TECHNIQUE FOR THE DETERMINATION OF THE DENSITY VARIATIONS IN COMPLETED SPECIMENS OF DRY COHESIONLESS SOIL

A resin impregnation technique has been developed to measure the density variations within laboratory test specimens of dry sand, used for small-scale model testing. A low viscosity, non-toxic resin was selected after a number of trials. The resin is injected in the completed soil specimen using hypodermic needles. The impregnated samples are excavated after the resin has cured, and their density is measured. An advantage of the technique is that the density tests are performed after the test specimen is completed. The resin shows negligible volume change during setting and curing. It also shows a rapid setting time after a period of inactivity, which allows the efficient impregnation of relatively large volumes or sand. Calibration tests have shown good correlation between the densities of impregnated samples and control specimens. Examples of the application of the technique are given. Its use has contributed to the improvement of the uniformity of specimens of dry sand used for small-scale modelling of spill-through abutments and free embedded cantilever walls.