International Concrete Abstracts Portal

When designing concrete structures, fatigue related problems are not among the first that come to mind. However, structures subjected to strong cyclic loads such as those associated with destructive earthquakes experience strength and stiffness degradation that are most aptly described as a low-cycle fatigue phenomenon and are related to the damage accumulated under such loading. This paper briefly discusses the various elements of a rational, i.e. mechanics-based design methodology. Results of an experimental test program are summarized, in which 4-inch cubes with or without fiber reinforcement are subjected to uni- and biaxial cyclic compression until failure. The review concludes with a brief review of the various aspects of material behavior that need to be modeled, if the response of reinforced concrete members is to be simulated numerically.

Confinement is the key to the performance of reinforced concrete structures when ductility demands are of primary interest. Hence dilatancy and restraining effects are critical for the behavior of reinforced concrete under seismic environments. In fact, restrained dilatancy is the determinant factor ensuring strength and ductility of reinforced concrete members in compression. In this paper, the issue of the dilatancy of concrete at different levels of active confinement is revisited. Experimental observations on 150x300 mm concrete cylinders, which were recently tested in a large capacity triaxial chamber, are presented. For the analysis of the dilatancy data, the elastoplastic concrete model known as the Extended Leon Model is applied. The study is focused on the volumetric behavior of concrete, which in plasticity terminlogy refers to inelastic dilatancy and the concomitant issue of normality. In particular, the test data is examined within the framework of the non-associated flow theory of plasticity. In this context, the origin of discontinuous failure mechanisms in the high confinement regime is questioned, where inelastic dilatancy together with the loss of axisymmetry are the primary reasons for localized failure in the form of discontinuous faulting.

The load-deformation response of R/C membrane elements (panels) subjected to reversed cyclic shear showed that the orientation of the steel bars with respect to the principal coordinate of the applied stresses has a strong effect on the pinching effect in the post-yield hysteretic loops. When the steel bars were oriented in the directions of the applied principal stresses, there was no pinching effect. When the steel bars were oriented at an angle of 45’ to the applied principal stresses, there was severe pinching effect. It was obvious that the pinching effect is caused by the orientation of the steel bars, rather than the bond slips between the steel bars and the concrete as surmised by many researchers. A non-linear analytical model capable of describing this pinching behavior is presented in this paper. The model is actually an extension of the fixed-angle softened truss model (FA-STM) proposed by Hsu and his colleagues for monotonic loading. The extension of FA-STM for application to reversed cyclic loading requires new constitutive models for concrete and steel in the unloading and reloading ranges. This rational theory satisfies Navier’s three principles of the mechanics of materials: equilibrium, compatibility and constitutive relationships of materials. The validity of this theory is illustrated by comparing the behavior of three panels with three different steel bar angles. The predicted cyclic behavior compared well with the experimental behavior, except in the descending branch.

A stress hybrid element that incorporates an internal displacement dis-continuity is presented for the modeling of concrete fracture. This stress hybrid formulation is superior to similar stiffness-based embedded crack formulations in that it explicitly accounts for boundary tractions so that the equilibrium of the traction fields at the element boundary and the internal crack interface can be enforced in a consistent manner. As a consequence, it also allows for the modeling of crack initiation in an accurate and consistent manner. Numerical examples are provided to compare the performance of the new element to that of a smeared crack model and to demonstrate its superiority in capturing the sliding shear behavior of fractured concrete. The element achieves the realism of the discrete crack approach without the need for remeshing or knowing the location and orientation of a crack a priori.

As a structural material, reinforced concrete requires bond between plain concrete and reinforcing steel. Accurate numerical modeling of structures that exhibit severe bond-stress demand requires explicit representation of bond-zone response. A bond element is presented for use in high-resolution finite element modeling of reinforced concrete structures subjected to general loading. The model is defined by a bond stress versus slip relationship and a relationship between maximum bond strength and the concrete and steel stress-strain state. A finite element implementation of the model is proposed that enables a one-or two-dimensional representation of bond-zone action. Non-local modeling is used to incorporate the dependence of bond strength on the concrete and steel material state. Comparisons of simulated and observed response for systems with uniform and variable bond-zone conditions are presented.