Linear Elastic Fracture Mechanics Assessment of a gas turbine vane

The continuous evolution of heavy-duty gas turbines, with the aim of improving the machine performance, implies a significant increase of its working conditions. Regarding turbine expander, front blades & vanes have to withstand extremely harsh operating conditions that lead to damage mechanisms as creep, fatigue and oxidation, which play an important role in the crack initiation and propagation. At Ansaldo Energia Group, it has been studied, by means of Linear Elastic Fracture Mechanics (LEFM) approach, the evaluation of crack propagation behaviour at the most critical locations of a GT second vane for power generation (F-class). Furthermore, it has been developed a Mechanical Integrity (MI) assessment on the FE model where the required creep limits of the material were overcome, according to FE calculations. The application of Fracture Mechanics (FM) approach is mandatory to estimate the component lifing before to achieve the critical crack size. Namely, the evaluation of the number of firing hours or cycles the vane can accomplish with a stable crack propagation can be calculated, avoiding the unstable crack behaviour and catastrophic failure. Via FEM using the Fracture Tool within ANSYS Workbench environment, the whole procedure of crack modelling and evaluation of fracture parameters have been developed, in particular the Stress Intensity Factors (SIFs). Nevertheless, the correct functioning of the Fracture Tool has validated by means of a simple plate with a crack inside. A comparison of FE results with direct calculation by formula has been performed to appreciate the increase of the stress intensity factor KI (Mode I) with the rise of the crack length a. Additionally, by an in-house program, Propagangui, the crack growth due to the effects of creep, fatigue and oxidation phenomena has been studied in the operating conditions. The results obtained from the first analysis show an unexpected behaviour of the KI, which is in contrast with the theoretical approach: a KI reduction has been obtained with the a increase. Moreover, in the SIF assessment, it has been observed that KII and KIII are of the same order of magnitude than KI, but lower. Therefore, it has been chosen KI as the dominant SIF. Results can be explained due to the component complex geometry and to the extreme operating conditions. Furthermore, it seems to be performed a redistribution of the stresses along the critical location with the crack length increase. This fact does not create an unstable increase of the stress around the crack. The stable propagation of the crack has been validated from the second performed analysis (crack growth assessment), where it has not achieved neither plastic collapse nor unstable propagation. Slowdown of the crack growth rate as the crack length increases due to the reduction of KI is shown. So, stresses redistribution helps the achieved propagation of the crack to become slower. To evaluate the component lifing, different end-of-life lines at different "critical" crack sizes have been drawn. These latter will give the number of cycles / firing hours needed to achieve critical crack size. Finally, the propagation length between 2 mm and 5 mm happens within low number of cycles (or firing hours), the propagation to achieve 10 mm crack size needs more time. This behaviour confirms crack slows down as it evolves. So, the behaviour of the crack follows a stable propagation, the evolution of the crack tends to slow down its growth rate and therefore no unstable fracture is expected.