A non-local approach to model the combined effects of forging defects and shot-peening on the fatigue strength of a pearlitic steel

Abstract

This study focuses on the influence on fatigue behaviour of the surface integrity of a steel connecting rod. The component is hot-forged and shot-blasted, producing a complex surface state with large surface defects and high residual stresses. In a prior study, the surface was thoroughly characterized and fatigue tests were performed. Several different surface states were analysed in order to quantify the influence of the various surface aspects. These tests showed that the forging defects and the residual stresses are the most influential aspects of the surface. The goal of this paper is to develop an approach capable of taking into account the influence of both these aspects on fatigue behaviour. Two methods were developed. First, using surface scans of the fatigue specimens, the forging defects were fitted with ellipsoids so as to determine their size and shape. This allows to easily compare the numerous defects and test various criteria in order to identify the critical defect of each specimen. The second method used was the finite element simulation of the defects based on real topography scans. Using a non-local approach based on the theory of critical distances, the simulations were used to accurately predict the influence of the defects' geometry. The residual stress profiles were integrated in the simulations using Dang Van's criterion. The predictions are accurate and show the importance of taking into account the real defect geometry when estimating the fatigue strength.

A probabilistic approach to study the effect of machined surface states on HCF behavior of a AA7050 alloy

“…The second category is composed of non-local approaches, like the critical distance [16] originally proposed by Taylor [17,18] and that considers a mean stress value evaluated over a line, a surface or a given volume. The efficiency of non local approaches has been demonstrated in a number of applications, for instance for notched components [19], specimens containing internal defects [20] or surface defects [21].…”

HCF of AA7050 alloy containing surface defects: Study of the statistical size effect

“…cycle fatigue limit, the macroscopic fatigue limit σ a can be expressed via the following expression: (11) where K tmax is the maximal equivalent stress concentration factor which correspond to the maximal equivalent stress divided by the nominal applied stress. = K σ σ t max applied max (12) In the case of polished specimen and a simple uniaxial loading, the surface stress heterogeneity factor H ms could be easily obtained by an analytical calculation [13].…”

“…Eqs. (11) and (13) are used to predict the fatigue limit in the presence of 1 and 22 defects. A comparison of experimental results and predictions is shown in Fig.…”

“…Thus, even the smallest size defect has an impact on the predicted fatigue limit which is clearly not the case. In literature, some studies [7,8,11] consider the stress not in the highly stressed point but at a given distance or in a non local way as a mean value over a given volume. This is a simple way to reflect the asymptotic response of small defects.…”

“…The models dealing with the gradient effect in a deterministic way are often built by using a non local approach like the one initially proposed by Taylor [7,8]. The efficiency of non local approaches has been demonstrated in a number of applications, for instance for notched components [9], specimens containing internal defects [10] or surface defects [11].…”