Effects of specimen geometry on tensile ductility, The

Abstract

The effects of specimen geometry on tensile ductility, with specific focus on methods to correlate total elongation values between different sheet type specimen geometries, were evaluated for five sheet steel alloys using six different geometries. Alloys included DQSK (drawing quality special killed), TRIP590, DP980, QP1180, and M1500. All materials were 1.4 mm thick and uncoated. Specimen geometries had gage lengths ranging from 10 to 80 mm and gage widths from 2.4 to 25 mm. All specimens were tested in room temperature air using a constant crosshead velocity determined for each specimen geometry to produce an engineering strain rate of 0.001 s^-1. Strain data were gathered using digital image correlation (DIC). Ultimate tensile strengths (UTS) and total elongations (TE) for the alloys evaluated with ASTM E8 standard tensile specimens ranged from 332 to 1617 MPa and 5.6 to 44.7 pct, respectively. Engineering stress-strain curves showed that smaller sample geometries resulted in greater total elongations. Tensile deformation behavior for a given alloy was independent of specimen geometry up to the point of tensile instability. Beyond instability, curves for a specific alloy diverged, indicating that variations in total elongation with specimen geometry were primarily influenced by post-uniform strains. Total plastic elongations (TPE), defined as TE less elastic strains, were used to evaluate the original Oliver equation upon which the correlation method outlined by ISO 2566/1 is based. Fit quality in terms of coefficient of determination (R^2) ranged from 0.84 to 0.99 for the alloys studied, indicating that reasonable correlation results were possible despite the fact that all alloys except DQSK were outside the recommended applicability of ISO 2566/1 based on strength and/or condition. The average correlation exponent a for the geometries and alloys studied was 0.4, equivalent to the general value a = 0.4 recommended by ISO 2566/1. However, best fits of TPE data for alloys DQSK and M1500 were obtained with a-values of 0.16 and 0.75, respectively, indicating that correlations can be optimized by evaluating a with respect to specific alloys. Fractured tensile samples were systematically catalogued to determine the effect of necking mode on total elongation. Results showed that the appearance of a localized neck, characterized in sheet type specimens by plane-strain localization along a defined angle relative to the tensile axis followed by fracture along the neck, resulted in lower total elongations compared to specimens on which the initial diffuse neck was not followed by formation of a localized neck. Localized necking generally occurred on specimens with larger gage widths, but also occurred on specimens with smaller gage widths as material strength increased. Analyses of incremental strain rate data confirmed that total elongation generally increased as strains within the neck became less localized. Necking strains were less localized as specimen gage width decreased. Analyses of DIC strain profiles showed that neck lengths along the tensile axis were independent of specimen geometry. The absence of neck length dependency, combined with the dependence of necking mode on gage width and the subsequent influence of necking mode on total elongation, indicate that total elongation for sheet type tensile specimens is more sensitive to gage width than gage length for the range of geometries considered.