No abstract

No abstract

Performing tensile tests on ECAE material reveals a long post-uniform elongation. In order to calculate correct true stress-true strain diagrams, three different approaches are used: measurements of the actual sample geometry, a neck evolution model proposed by Segal [V.M. Segal, S Ferrasse, F. Alford, Mater. Sci. Eng. A442 (2006) 321-326] and an inverse modelling method by finite element (FE) simulations. Keywords: SPD; Strain hardening; Tension testUntil recently it was believed that grain size refinement towards ultrafine-grained material leads to a loss of strain hardening capacity. However, when closely observing the shape change during tensile deformation at room temperature, it can be seen how a very small homogeneous deformation occurs, followed by a long postuniform elongation [2]. Calculating a stress-strain curve from a load-displacement curve assuming homogeneous deformation obviously leads to an erroneous shape of the flow curve (see Fig. 1). A correction method has recently been proposed by Segal et al. [1] by looking at the evolution of the sample boundaries. This leads to a modified set of equations for the calculation of stress, strain and strain rate in the sample neck. In the present paper, experimental verifications by measuring the actual sample cross-section are performed. They provide an improved tensile true stress-true strain curve for AA1050 aluminium after equal channel angular extrusion (ECAE) processing. The inverse finite element FE method provides both the material stress-strain curve and the actual sample cross-section. The computed results depend on the material model assumed. Finally, experimental results are compared with both FE and Segal's approach, allowing their validation.The material used in this study is commercial-purity aluminium (type AA1050). It was delivered hot rolled and fully recrystallized with an initial grain size of 100-300 µm. All ECAE samples were machined with the extrusion direction (ED) parallel to the former rolling direction (RD). The samples had a diameter of 12 mm and a length of 60 mm. ECAE was performed at room temperature following route B C [3] for 8 passes, lubricating with MoS 2 . The ECAE die had an intersection angle Φ = 90°, without rounding (Ψ = 0°). The final grain size was about 1 µm, with more than 70% high angle grain boundaries (when analysed by EBSD as described in Ref.[4]).The tensile samples were machined such that only the homogeneous zone of the ECAE sample would be subjected to deformation. Stainless steel custom-made extension parts with screw fixation were used for mounting. All the tests were performed at room temperature on an Instron 4505 machine with 100 kN load cell. The initial strain rates were 2.5 x 10 -5 and 8.3 x 10 -2 s -1 respectively, for constant crosshead speeds of 2.5 x 10 -7 and 8.3 x 10 -4 m s -1 during the test. For the FE simulations only the high crosshead speed was retained.To determine the initial yield point, the slope of the stress-strain curve during elastic deformation was used to calculate ...

In order to model accurately the anisotropic material behavior during finite element simulations, a precise description of the material yield locus is required. Beside the shape (linked to the material model used), the size (related to the isotropic hardening) and the position (kinematic hardening) of the yield locus, its orientation is of particular interest when large rotations of the material are encountered during the simulations. This paper proposes three distinct methods for the determination of the material yield locus rotation: a method based on the Constant Symmetric Local Velocity Gradient (CSLVG), a corotational method and a method based on the Mandel spin. These methods are compared during simple shear tests of an aluminum sheet.