Publication

Abstract : Extraction of mechanical properties of in-service materials through small specimen test techniques has become attractive in the recent years. Of the available small specimen test techniques, automated ball indentation (ABI) and small punch test (SPT) methods have proved to be more promising. These test methods are basically non-destructive in nature and are proficient enough to extract the flow properties of the materials using small volume specimen. In this work, tensile properties of a pressure vessel steel (P12) have been estimated through ABI and SPT. The objective is to compare the capability of these test methods in determining the tensile properties. The influence of lubrication (between the indenter and the specimen) on the ABI response is also investigated. The ABI response is found to be similar and the effect on the tensile properties was under 2 %. The tensile properties estimated from ABI and SPT are found to be in good agreement with conventional tensile test results. Nevertheless, in case of SPT, the error in the estimation of yield strength and ultimate tensile strength using empirical correlations is significantly high. However, the use of analytical formulations to convert the SPT load–displacement response to stress–strain curve are found to be reliable, since the error in the estimated properties is considerably less. The ABI process is numerically simulated to study the stress–strain field beneath the indenter. The maximum strain occurs at the edge of the contact indicating the material displacement along the radial direction. The plastic zone beneath the indenter resemble hemispherical shape which is in agreement with the expanding cavity model. The nature of stress changes from compressive (right below the indentation axis) to tensile at the edge of contact. This indicates that radial cracks may initiate on the specimen and propagate outwards. The pile-up is significantly higher in the case of frictionless contact between the indenter and the specimen but found to converge for a value of around 0.2.