PSI - Issue 57

David Mellé et al. / Procedia Structural Integrity 57 (2024) 61–72

63

David Melle´ / Structural Integrity Procedia 00 (2023) 000–000

3

stresses, to homogeneize as much as possible the microstructure and to obtain a suitable α + β microstructure instead of the as-built α ′ microstructure, heat treatments are applied to the coupons.

2.2. Heat treatments

Multiple heat treatment steps have been applied to the coupons after building. The first is a stress relieving of the building plate at 730 ◦ C for 2 h. This is done before the coupons were separated from the building plate. After this stress relieving step, the coupons are separated using EDM and heat treated using a two-step heat treatment of 940 ◦ C for 1 h followed by 650 ◦ C for 2 h. A low cooling rate (lower than 5 ◦ Cmin − 1 ) is used after each of these two steps. After this heat treatment, the coupon heads are machined and some of the net-shape coupons are chemically etched. This surface treatment is described in the following paragraph. All heat treatments are done without isostatic pressure application (no HIP) and in a controlled atmosphere (argon for the building plate stress relieving and a secondary vaccum for the heat treatment and post-machining stress relief). One of the aims of this work is to characterize the e ff ect of chemical etching on the fatigue life, the initiation mech anisms and the population of the surface micro-geometric features. These micro-geometric features can be surface connected gas pores, surface connected lack-of-fusion or surface valleys inherent to the layer-by-layer manufacturing process. Part of the coupons are then chemically etched for 30 min in the solution developped by the IRT M2P in the ”After ALM” project. The solution contains hydrofluoric acid. During the treatment, the temperature is controlled and the solution is slowly stirred. This etching time corresponds, according to previous studies, to the almost optimal treatment time for this material. As described in the previous paragraph, two surface states (net-shape and 30 min etched) are presented in this work. Fatigue bending tests were conducted on those surface states using a Rumul Cracktronic resonant machine. The specimens are flat coupons whose geometry is presented in Figure 1. Once correctly mounted in the testing machine, the loading mode is in pure bending. All fatigue tests were realized under bending moment controlled conditions (which correspond to a stress controlled configuration) at a stress ratio R = 0 . 1 at ambient temperature. The test frequency which depends on the coupon geometry is approximately 80 Hz and a frequency drop of 1 Hz is used as a criterion for the tests to be stopped. The specimens were tested using the locati methodology with constant stress steps of δσ = 25 MPa. The starting stress value depends on the surface state and on the surface scan (on some low roughness specimens, the first steps were omitted in order to diminish the experimental campaign time). This means a fatigue strength value is determined for each specimen which corresponds to the fatigue life used during the locati procedure, here 10 7 cycles. This fatigue strength value, σ FL , is determined using the post-treatment formula proposed by Lanning et al. (Lanning et al. (2005)) based on the final stress level applied during locati procedure, σ f in , and on the life reached on this final step, N r (cf. Equation 1). As the level of porosity and pore size distribution may di ff er from one coupon to another, this procedure is more appropriate than multiple coupons procedures like the staircase methodology (Dixon and Mood (1948)). σ FL = σ f in − δσ 1 − N r 10 7 (1) 2.5. Surface scan procedure Having flat specimens makes it possible to fully scan the gauge length of the coupons before and after fatigue testing. These surface measurements were made using a Brucker Contour GT interferometer equiped with a × 5 ob jective and × 1 field adapter. The in-plane resolution in this configuration is around 2 µ m. The vertical resolution is approximately 10 nm. Surfaces of around 22 mm × 6 . 7 mm were scanned. These surface scans were used to isolate the surface connected porosities and surface valleys. Several measurements were then made on each defect to establish its 2.3. Surface treatments 2.4. Test procedure