Hydromechanical coupling at dam-foundation joint

When fracture occurs in a concrete dam, the crack mouth is typically exposed to water. Very often this phenomenon occurs at the dam-foundation joint and is driven also by the fluid pressure inside the crack. Since the joint is the weakest point in the structure, this evolutionary process determines the load bearing capacity of the dam. In this paper the cracked joint is analyzed through the cohesive model proposed by [Cocchetti et al., 2002], which takes into account the coupled degradation of normal and tangential strength. The water pressure inside the crack, which reduces fracture energy and increases the driving forces, is analyzed through the model proposed by [Reich et al., 1994], [Bruhwiler and Saouma, 1995] . Some numerical results are presented which refer to the benchmark problem proposed in 1999 by the International Commission On Large Dams (see [ICOLD, 1999]). The first crack is induced by tensile stress at the upstream edge. This propagation is stable in load control even when the water pressure is applied inside the crack. Initially the Mode I displacement discontinuity dominates. As the crack grows, the Mode II contributions become important. If the tensile strength of the rock is high, vertical branching of the crack is prevented and the next crisis occurs at the downstream edge. It is a sliding crack, under high compressive stresses in dry conditions. There is a phase in which both cracks grow in load control. Since in exceptional flood conditions the load cannot be reduced, the maximum load carrying capacity is achieved when the load control condition is lost.