Compression testing techniques to determine the stress/strain behavior of metals subject to finite deformation

Abstract

We have investigated techniques to conduct a valid single-increment compression experiment to large strains, e ≈ 1.0. Our investigation included experiments with many different lubricants, specimen-end geometries, five different materials, and temperatures of 22 °C, 315 °C, and 600 °C. We found that for materials with a work-hardening exponent greater than 0.15 and a slightly positive strain-rate sensitivity, the best technique was to cut concentric grooves into the ends of the compression sample and to use polished platens and a molybdenum-disulfide greasebase lubricant. This technique gave valid stress/strain data and uniform displacement fields (as determined by metallography of the deformed specimen cross sections). At all temperatures, satisfactory results were also obtained using a MOLYKOTE spray lubricant. We investigated both well-lubricated experimental conditions (the concentric specimen grooves and Molygrease lubrication and TEFLON sheet lubrication) as well as poor lubrication conditions. The poorest lubrication was achieved by using a set of platens which had grooves cut in them to prevent any expansion by the ends of the specimen during deformation. Up to a compressive strain of 0.5, the load/displacement data from all of these lubrication conditions were almost identical. This was the case for materials with a work-hardening exponent >0.15. Despite the similarity of the “average” stress/strain data to e = 0.5 for all lubrication conditions, metallographic cross sections showed that the deformation was nonuniform for the poorly lubricated experiments. The results of the experiments show that a single increment compression test can be used to obtain accurate constitutive behavior of many materials to large deformations, e ≈ 1.0. In addition, we have found that the compression test is relatively insensitive to lubrication conditions for many materials through moderate deformations,e ≈ 0.5.