PSI - Issue 66

Aditya Khanna et al. / Procedia Structural Integrity 66 (2024) 370–380 Author name / Structural Integrity Procedia 00 (2025) 000–000 Figure 6 shows the underlying Plasticity Induced Crack Closure (PICC) relationship for the additively manufactured material. It is clearly demonstrated that the PICC relationship between and does not depend upon the direction of crack growth relative to the weld deposition direction. Hence, any observed differences in crack closure, and subsequently crack growth rates, observed between the two crack growth directions is largely due to the differences in residual stress distribution perpendicular to the crack propagation paths. 6. Conclusion The outcomes of several experimental studies have demonstrated that fatigue life of AM components, which can be expressed with S-N curves, can be very different under the same applied stress if loaded in different directions. The significant differences in fatigue life of SDSS fabricated by WAAM were confirmed by direct crack growth rate measurements in CT specimens for crack growth longitudinal (parallel) to the weld deposition direction vs. crack growth transverse (perpendicular) to the weld deposition direction (Sales et al., 2024). In the same study, the ASTM compliance-offset method was applied to experimentally determine the effective SIF range, Δ eff . It was found that the ⁄ vs. Δ eff curves are almost identical for the two crack growth directions, i.e., the observed differences in fatigue crack growth rates at nominally applied SIF ranges were due to different levels of crack closure in the two directions. However, the ASTM E647 compliance-offset method does not allow for the evaluation of the relative contribution of two main mechanisms to crack closure; for long cracks these are Plasticity-Induced Crack Closure (PICC) and residual stress effect. The PICC is largely affected by the loading history, which can be evaluated or simply recorded during operation of the structural component. As mentioned in the Introduction, the residual stress field can be very different in the test samples and in the actual structural component. If ignored, this difference can contribute to errors in fatigue life evaluation, potentially, leading to non-conservative fatigue life assessments. Therefore, it is important to remove the influence of residual stress field during processing fatigue test results. The present work extends our previous analysis (Sales et al., 2024) by estimating the contributions of the residual stress field to the crack tip closure values and the effective stress intensity factor range. Two methods based on the change in back-face strain with crack growth are applied to determine the residual SIF as a function of the crack length. Application of these methods does not require any additional instrumentation or signal processing compared to a standard fatigue test. After processing fatigue test results, the residual SIF was superimposed on the applied SIF range to determine the actual stress ratio at the crack tip, act . It was demonstrated that SDSS fabricated with WAAM this actual stress intensity factor ratio, can be significantly different to the applied stress ratio, ap , especially if the applied stress ratio is close to zero ( ap ~0) . It was also shown that plotting the opening load ratio, , against act significantly reduces the scatter in the experimental results and better matches theoretical expectations. Moreover, the use of the actual stress intensity factor ratio, which removes the contribution of the residual stress on the crack closure levels and the effective stress intensity factor range, provides a more accurate estimation of PICC values for different scenarios of fatigue loading. The latter is specifically important for the development and validation of the advanced fatigue life assessment procedures. 379 10

1.00

0.90

0.80

S1L S2T Eq. (10)

0.70

0.60

-0.2

0

0.2

0.4

0.6

Fig. 6. Opening load ratio vs act for all values of applied stress ratio ( ap ) and normalised crack length in the interval 0.25< ⁄ <0.54 .