Investigation of the crack front shape effect on the DCPD technique – application to the crack propagation under hold-time conditions in DA Inconel 718

Aeroengines manufacturers have to demonstrate that critical components such as turbine disks meet the certification requirements in terms of fatigue crack propagation life, under sinusoidal loading, using the damage tolerance approaches. In order to be more representative of in-service loading conditions, this study will focus on the effect of hold-time at maximum load, involved during trapezoidal cycles, on the crack propagation behaviour of DA Inconel 718, a nickel-based superalloy widely used for turbine disk manufacturing. Crack propagation tests under hold-time have been performed both on SEN- and KB-type specimens, at 550°C and 650°C and for various hold-times (300s up to 3600s), as well as fatigue crack propagation tests (2Hz). Crack length was monitored using the direct current potential drop technique (DCPD). Hold-time cycles lead to increased crack growth rates depending on the hold-time duration, when compared to sinusoidal cycles. Under fatigue loading, transgranular fracture is observed whereas it is intergranular under hold-time conditions. Moreover, a strong crack front curvature is observed under hold-time conditions when compared to fatigue loading (Figure 1-a), undepending on specimen geometry. This is usually observed under hold-time conditions and is often neglected while analysing crack growth rates derived from PD signal. Using finite elements method (FEM), first, calibration functions of the DCPD technique were identified for ideal crack front shapes: straight for SEN and half-circular for KB-type specimens (as observed under fatigue loading). Similar calculations were carried out based on real crack front shapes, reproduced from fracture surfaces obtained under hold-time conditions (Figure 1-b). The macroscopic curvature and the microscopic “saw-teeth” aspect of the crack (Figure 2) front were taken into account in the calibration process. It is demonstrated that small approximations in the crack front shape have a strong effect on the identified calibration functions. It also affects the stress intensity factor which has then been re-computed. This procedure was used to re-analyse crack propagation test results under hold-time conditions, which usually lead to non-ideal crack front shape. The hold-time effect is often associated with the concept of a damaged zone ahead of the crack tip, due to a damaging effect of environment during the hold time. Experimental attempts to reveal this zone were carried out by performing, on the same sample, crack propagation under hold-time conditions followed by fatigue loading. A transient state is observed in the da/dN-ΔK curves obtained during the fatigue step. The crack growth increment before it reaches a stabilized state is associated to the size of the damaged zone. Analysing these tests using non-idealized calibration functions showed that the damaged zone is mostly related to the “unfolding” of the (neglected) crack curvature. It is believed that the size of the damaged zone, defined as the volume of material affected by environment at the crack tip, cannot be measured using the DCPD technique. This very local phenomenon may lead to an uneven, curved and tortuous crack front which will give rise to local high crack growth rates detected by the DCPD technique.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.