It is well known that without a proper regularization scheme the finite element analysis of softening materials leads to mesh dependent response. The results of the analysis should be as less as possible dependent on the choice of the finite…
It is well known that without a proper regularization scheme the finite element analysis of softening materials leads to mesh dependent response. The results of the analysis should be as less as possible dependent on the choice of the finite element type and on the finite element discretization. In the present article the mesh sensitivity study of the compact tension specimen loaded under quasi-static and dynamic loading conditions is carried out. It is shown that relatively simple regularization scheme based on the element size independent energy dissipation (crack band method) leads to mesh insensitive results. In the analysis the constitutive law based on the rate sensitive microplane model for concrete is used. In the model the tensile response is based on the bi-linear crack-opening softening law. It is shown that, in spite of the simple regularization scheme, the rate dependent peak load, failure mode, crack branching and crack velocity are realistically predicted for different types and sizes of finite elements.