RESPONSE OF CONFINED CONCRETE SUBJECTED TO EARTHQUAKE TYPE LOADINGS

The concept of envelope curve is experimentally confirmed and extended to confined normal weight concrete as well as plain and confined light weight concrete. For slowly applied loading the envelope curve as defined by the monotonic curve may be exceeded. This observation limits the validity of the envelope concept, but it is not too severe for earthquake loadings.;A rheological stochastic model for uniaxial loading is proposed which depends on three parameters: K, which is associated with the initial stiffness and a and b which are associated with the probabilistic distribution of the rheological properties of the elements. The strain softening, the concept of envelope, the path dependency and the stiffness degradation are simulated well by the model.;The uniaxial model is extended to a biaxial with two parameters. Predictions for proportional and non-proportional monotonic as well as cyclic loading compare well with experimental results. The model gives the (sigma)-(epsilon) curves in the principal axes. Rotation of loading and shear stress cannot be handled.;The proposed analytical expressions for the envelope curve are used to predict the behavior of large size columns with longitudinal and lateral reinforcement. The predictions compare favourably with experimental results. From a parametric study conducted using these analytical expressions it was revealed that (1) columns subjected to axial load and monotonically increasing lateral load may exhibit a drop in moment following a first peak. The ACI method accurately predicts the first peak; (2) the drop in moment depends on concrete strength, shape of column, amount of confinement and axial load; (3) everything else being equal, the square column studied here, designed according to ACI method, has higher moment capacity at large strains or curvature than the round ones; (4) the contribution of unconfined concrete is rapidly decreasing at large strains; (5) the contribution of steel is increasing up to high strain due to strain hardening; (6) the contribution of confined concrete is significant (up to 30 percent); (7) the ACI method of accounting for the influence of compressive strength and amount of confinement seems adequate for high strength concrete columns; (8) load-history, steel debonding, cracking and shear are some parameters that influence the sectional behavior, but they were not studied in this work.