Search

Two stress regime‐dependent creep constitutive equations, that is, two‐regime Norton (2RN) model and Sin‐hyperbolic (Sinh) model, are adopted to numerically investigate the influence of constitutive equation and friction on the measurement of steady‐state creep strain rate based on five types small specimens. Results show that the influence of constitutive equation and friction depends strongly on the type of small specimen. A measurement error of more than 19 times may occur if the 2RN model is used, while the use of the Sinh model is able to generate a consistently lower measurement error. Of all small specimens, the cantilever beam testing (CaBT) is revealed to exhibit the best comprehensive performance in terms of measurement error and response to friction. To validate numerical findings, creep tests on the CaBT specimen of P91 steel are conducted at 600°C. Finally, recommendations on the choice of small specimen for obtaining creep strain rate are made. Highlights: Sinh model is better in describing stress regime‐dependent creep behavior.Influence of friction is strongly dependent on the type of small specimen.Small cantilever beam testing has the best comprehensive performance.Attention should be paid to selecting the constitutive equation and specimen type. [ABSTRACT FROM AUTHOR]

To assess the contribution of residual stress to the improvement of high‐cycle fatigue (HCF) properties, HCF tests are performed on Ti‐6Al‐4V specimens with different residual stress levels but no other evident surface state modification induced by machining. Results indicate that compressive residual stress induced by machining significantly improves the HCF performance of Ti‐6Al‐4V specimens. Of all the specimens, the fatigue strength of specimens with a surface residual stress level of −400 MPa is increased by approximately 16.4%. Furthermore, to accurately predict the fatigue life based on three mean stress correction equations (e.g., Walker, Goodman, and Smith–Watson–Topper), the method of reducing surface residual stress by a reduction factor and then adding to the mean stress is proposed. A reduction factor of 0.55 is found to be appropriate for the alloy under the test condition, with majority of the predicted data falling within a factor of three scatter band. [ABSTRACT FROM AUTHOR]