Issue 31

H.F.S.G. Pereira et alii, Frattura ed Integrità Strutturale, 31 (2015) 54-66; DOI: 10.3221/IGF-ESIS.31.05 61 Fig. 11 presents the results obtained for the three meshes when considering the Concrete Damage Plasticity model. As it was expectable the maximum slip was 5 mm. In terms of maximum slip the results are similar to the ones obtained considering an elastic behaviour for concrete. As it is visible in Fig. 11 b), it was verified a sharper peak in meshes 3 and 5, which incremented nearby 3% the peak load. Based in this analysis it was adopted Mesh 4, to carry out the forthcoming analysis. a) b) Figure 11 : Bond stress-slip relationship displacement at failure equal to 5 mm: a) complete curve; b) detail peak curve. Cohesive element thickness The interface was simulated with the use of cohesive elements in-between the rebar and surrounding concrete. Since these are not pure interface elements, a certain thickness must be ascribed to the element. The cohesive element’s thickness used in the parametric study were 0.1, 0.5 and 1.0 mm, respectively. Fig. 12 and 13 show the results for the three thicknesses, adopting an ultimate slip of 5 and 1000 mm, respectively. As foreseeable, with the thickness increase was verified an increase of the maximum pullout load. This occurs because the cohesive elements’ mid surface is farther from the rebar’s surface, consequently, there is an increase of the interface area. Therefore for the same bond stress profile it will result in higher shear stresses at the interface and consequently a higher pullout load. Figure 12 : Bond stress-slip relationship displacement at failure equal to 5 mm, with different thickness of cohesive elements. Figure 13 : Bond stress-slip relationship displacement at failure equal to 1000 mm, with different thickness of cohesive elements. Viscosity coefficient The viscosity coefficient is a regularization parameter for obtaining convergence in damage models that exhibit a softening response. It allows the analysis to converge after the peak load is attained, but the results can be unrealistic if a proper value is not selected. The influence of the viscosity in the pullout response was carried out. For this purpose three quite distinct values for the viscosity coefficient, namely, 0.001, 0.0001 and 0.00001 were selected. These analyses were carried out for two linear bond stress – slip relationships, adopting an ultimate slip of 5 and 1000 mm, respectively. As it can be observed in Fig. 14, for the local bond law defined with an ultimate slip of 5 mm, the viscosity coefficient only does not