A simple model to explain the effect of different boundary conditions in direct tensile tests

Experimental observations suggests that concrete specimens subjected to uniaxial tension may have different responses depending on the imposed kinematic constrains. A simple model is proposed here to explain the experimental findings. Concrete specimens are schematized as a system composed of three trusses. This model can thus consider the non-homogeneity of the material, and the eccentricity between the head of the testing machine and the specimen. Two kinds of tests are considered: imposed uniform displacement (fixed platens) and freely rotating platens. The results of the model show that the tests based on fixed platens provide higher peak loads. Only for a perfect specimen and testing conditions (homogeneous material and absence of geometric eccentricity) the results of a fixed platens test and of a rotating platens test coincide. Furthermore, the same results are obtained by energy balance. (C) 2008 Elsevier Ltd. All rights reserved.