Abstract

Inclusions found in 4330 high strength low alloy steel, play an important role in fatigue failure of components in corrosive environments. This paper uses FEA-derived SCF?s and theoretical elastic stress singularities at surface-breaking non-metallic inclusions in 4330 steel, to explain a possible contributory mechanism for the development of pitting and fatigue crack growth. Previous work on the nature of theoretical elastic stress singularities[1] demonstrated that not all singularities are the same and can be usefully characterized by their ?strength?. Different stress components will generally have different singularity strengths (akin to SCF?s being a function of stress component, local geometry, and how the stress ?flow? is disturbed). The previous work[1] also demonstrated that singularities are generally not realised in practice (i.e. they are theoretical in nature, for various reasons). In the case of dissimilar joints e.g., there is likely to be a ?non-abrupt? transition from one material to another due to melting, alloying, diffusion etc. Nevertheless, in the case of inclusions, the transition region will be narrow. While infinite stresses won?t be realized, a large increase in local stress will arise, which could cause failure at the interface. Propagation of interface cracks could then result in inclusion detachment, leaving a pit. It is argued that stronger elastic stress singularities will give rise to earlier damage. It is proposed herein how the theoretical strength of singularities and associated SCF?s can provide insight to the formation of early corrosion pits and their characteristic shapes. These stresses will no doubt interact with variables such as residual stresses, fatigue, corrosion, erosion, dissolution etc. The size, number, density, physical and chemical properties of inclusions, will clearly influence the pitting process (including corrosion & dissolution). While it has been observed that inclusions in 4330 are often globular, for the purposes of this study, they were considered as 2D plane strain circular. The techniques developed previously[1], for determining singularity strength, involves using a very fine finite element mesh in the vicinity of the singularity and fitting a Power Law to the stress distribution (with an R-squared fit parameter close to unity). The Power Law variable is indicative of singularity strength, e.g -0.5 being the singularity strength for a mode 1 closed crack in fracture mechanics. For a free-surface dissimilar material singularity, the material parameter that influences singularity strength is the moduli ratio. To determine the influence of this, various FEA sensitivity studies were carried out. It was assumed that the elastic modulus of homogeneous inclusions varied over a wide range for different depths of surface-breaking inclusions in a block under tension. The inclusions had different levels of truncation due to forming & machining of the free surface. The singularities and SCF?s for this case were determined, both with and without inclusions. Cases with empty pits, provide an indication of where on the pit wall, fatigue crack development is most likely. A study of SCF versus moduli ratio was also carried out for a block with a centrally located subsurface inclusion (no singularities in this case). The FEA results are qualitatively compared with results from corrosion fatigue tests carried out using 4330 steel specimens. A novel corrosion attachment was produced for RR Moore machines, using layered manufacture. The design allows specimens to experience an interaction of cyclic stressing and corrosion. [1] Wood J. et al, Theoretical elastic stress singularities ? much maligned and misunderstood, Proceedings of NWC, San Diego, 2015.