Plastic flow of mild steel (En8) at different strain-rates under abruptly-changing deformation paths

Abstract

Strain-gauged thin-walled tubular specimens of annealed medium carbon steel (En8) were tested at room temperature in combined twisting and extension using a closed-loop, servo-controlled, electro-hydraulic biaxial testing machine. Bilinear deformation paths of twisting at a constant rate followed by extension at three different rates were investigated. Precise measurements of the resulting torque and load, together with the controlled deformation parameters, were recorded as functions of time. This study extends earlier work ( Meguid, Malvern and Campbell, 1979, J. Engng Mater. Technol. 101, 248) in which a notable feature of this particular type of bilinear testing was reported: namely, that it was possible to obtain almost the entire positive quadrant of the initial yield locus from a single run without unloading or reloading (neutral loading). Here, particular attention has been given to the effect of the axial strain-rate on the shape of these “initial” yield loci. Attention has also been given to the effect of the sudden change of direction in the deformation path upon the deviatoric stress and the plastic strain-rate vectors. The results indicate that there exist appreciable differences between the Mises equivalent stress and equivalent plastic strain curves (up to strains of order 2%) for the three axial strain-rates investigated. These differences are attributed to the rate-sensitivity of the material. The results also show a much slower alignment of the deviatoric stress vector direction to the direction of the plastic strain-rate vector than had been expected. Comparisons with two theoretical analyses of a bilinear deformation path of quasistatic twisting followed by extension at a constant strain-rate are made, one using Perzyna's (1966) viscoplastic constitutive law for rate-sensitive (but non-strainhardening) material and the other using a rate-independent theory. Refinements in the test procedures now reveal that significant differences exist between calculated and measured axial and shear stresses. In contrast, both theories gave reasonably good agreement with the time history of the equivalent stress and the stress trajectory, in agreement with the earlier work.