A Simple and Efficient Reformulation of the Classical Manson–Coffin Curve to Predict Lifetime Under Multiaxial Fatigue Loading—Part II: Notches

Abstract

The present study is concerned with the use of the modified Manson–Coffin curve method to estimate the lifetime of notched components subjected to multiaxial cyclic loading. The above criterion postulates that fatigue strength under complex loading paths can efficiently be evaluated in terms of maximum shear strain amplitude, provided that the reference Manson–Coffin curve used to predict the number of cycles to failure is defined by taking into account the actual degree of multiaxiality/nonproportionality of the stress/strain state damaging the assumed crack initiation site. The accuracy and reliability of the above fatigue life estimation technique was checked by considering about 300 experimental results taken from the literature. Such data were generated by testing notched cylindrical samples made of four different metallic materials and subjected to in-phase and out-of-phase biaxial nominal loading. The accuracy of our criterion in taking into account the presence of nonzero mean stresses was also investigated in depth. To calculate the stress/strain quantities needed for the in-field use of the modified Manson–Coffin curve method, notch root stresses and strains were estimated by using not only the well-known analytical tool due to Köttgen et al. (1995, “Pseudo Stress and Pseudo Strain Based Approaches to Multiaxial Notch Analysis,” Fatigue Fract. Eng. Mater. Struct., 18(9), pp. 981–1006) (applied along with the ratchetting plasticity model devised by Jiang and Sehitoglu (1996, “Modelling of Cyclic Ratchetting Plasticity, Part I: Development and Constitutive Relations. Transactions of the ASME,” ASME J. Appl. Mech., 63, pp. 720–725; 1996, “Modelling of Cyclic Ratchetting Plasticity, Part I: Development and Constitutive Relations,” Trans. ASME J. Appl. Mech., 63, pp. 720–725)) but also by taking full advantage of the finite element method to perform some calibration analyses. The systematic use of our approach was seen to result in estimates falling within an error factor of about 3.