PSI - Issue 66

Aditya Khanna et al. / Procedia Structural Integrity 66 (2024) 370–380 Author name / Structural Integrity Procedia 00 (2025) 000–000

374

5

Eq. (2) can be applied to obtain the residual SIF if the residual strain is experimentally measured as a function of crack (or cut) length. For example, Kerr et al. (2011) demonstrated the application of the influence function method by wire cutting additively manufactured compact tension samples. For noisy experimental data, it is recommended that the derivative of the residual strain with respect to the crack length, res ( ) ⁄ , be obtained by differentiating a polynomial fitted to the experimental data rather than using finite difference methods. For instance, a second order polynomial (parabola) can be fitted to 2 +1 consecutive data points centered about the evaluation point. This approach (incremental polynomial fitting) is also recommended by ASTM E647 for a related problem of experimentally evaluating crack growth rates ( ⁄ ). 2.2. Derivative ratio method Assuming that the global response of the specimen is linear-elastic, Eq. (2) can be applied to any arbitrary loading condition. Hence. at maximum applied load during cyclic loading, the stress intensity factor can be written as max ( ) = ( ) max ( ) . (4) As mentioned previously, max ( ) represents the maximum strain magnitude in the absence of residual stresses and can be evaluated using the open crack compliance as max ( ) = ̅ 0 ( ) max ⁄ . Taking the ratio of Eqs. (4) and (2), the residual SIF can be written as res ( ) ≈ max , ap ( ) res ( ) max ( ) . (5) Eq. (5) was referred to as the derivative ratio method for determining the residual SIF by Smudde et al. (2023). In Eq. (5), it is important to note that max , ap denotes the maximum SIF applied to the crack. Since the residual stress field is superimposed on the cyclic loading, the actual maximum and minimum values of the SIF at the crack tip are: max = max , ap + res , min = min , ap + res . (6) The derivative ratio method is more straightforward to implement for fatigue crack growth testing compared to the influence function method since the maximum back-face strain and maximum applied SIF are already computed as part of routine data processing. The infinitesimal quantities res and max can be calculated over several consecutive cycles using either finite difference or polynomial fitting methods. The latter is preferred as it is more robust to experimental noise. 3. Specimens and Loading Conditions The present work performs additional data analysis on the experimental data collected by the authors in a recent experimental campaign (Sales et al., 2024) on CT Specimens made from Super Duplex Stainless Steel (SDSS) using WAAM. The WAAM deposition parameters, specimen dimensions, loading history, and detailed results for the fatigue crack growth rates and crack tip opening loads are provided in Sales et al. (2024). Here, we reiterate the most essential points relevant to the present analysis. Standard CT specimens ( = 50.8 mm, =12.7 mm) as per ASTM E647 specifications were machined from the WAAM printed test wall such that one of the specimens had a starter notch oriented longitudinal (parallel) to the deposition direction (S1L), while the other specimen had a starter notch transverse (perpendicular) to the deposition direction (S2T). Only one specimen per orientation was used for fatigue testing as this was considered sufficient for the qualitative nature of the present study. However, it is acknowledged that replicate testing on multiple specimens per orientation would be preferred in other situations, e.g. when collecting fatigue test data for design or life assessment purposes.