PSI - Issue 7

E. Vacchieri et al. / Procedia Structural Integrity 7 (2017) 182–189

189

8

E. Vacchieri et al. / Structural Integrity Procedia 00 (2017) 000–000

(cycles) of a surface crack between 0 and l corresponds to the extra-time needed for a crack initiation in the substrate. The linear relationship between the crack size k and L k blade (number of cycles for which a single blade shows at least a crack of size k with the probability level x ) and between the crack length and its depth have been used to calculate the Weibull probability function for k ≡ d c , for which a single blade shows a crack of d c depth in the most critical location. The maximum crack depth in the two considered critical locations have been plotted against the NSU, as shown in Fig.6-b. A linear relationship with NSU is observed for both the locations and it means a linear crack propagation rate with increasing NSU. This plot is very important as can be used to predict the limit of NSU considering the critical penetration d c for part repairability, as shown in Fig.6-b, or for the part structural integrity. The developed safe life procedure estimates the life of GT components taking into account the new operating regime environment. It has been successfully applied to the first stage blade of an Ansaldo F-class engine as a very good agreement between the calculated life map and the critical locations highlighted in service is achieved not only qualitatively but also in terms of number of cycle to crack initiation. The use of Weibull statistics to field feedback data allows to achieve a life estimation based on field information that is in agreement with the evaluation of the developed lifing procedure. Weibull statistics can employed in di ff erent ways depending on the perspective of the user, e.g. repairability or mechanical integrity. The relationships between crack length-penetration and between maximum crack depth-NSU are very important because they can be exploited to have a rough and quick evaluation of crack propagation without the use of complicate and time-consuming fracture mechanic calculations. S.R. Holdsworth, Creep-Fatigue Testing and Assessment Guideline Material Property Data Requirements for Component Assessment, 1019778, EPRI Technical Report, December 2010. S.R. Holdsworth, Component Assessment Data Requirements from Creep-Fatigue Tests, JAI103583, Journal ASTM International 8 (3), 2011. M.E. Melis, E.V. Zaratesky and R. August, Probabilistic Analysis of aircraft gas turbine Disk Life and Reliability, Journal of Propulsion and Power, AIAA Transactions, Vol.15, No. 5, 15 Sept-Oct. 1999, pp.658-666. R.H. Salzman, “Applying Weibull methods in gas turbine component data analysis”, Proceedings of PWR2005, ASME Power, April 5-7, 2005, Chicago, Illinois USA. J.D. Summers-Smith, Fault Diagnosis as an Aid to process Machine Reliability, Quality and Reliability Engineering International 5 (3), 1989, pp.203-205. E. Vacchieri, S.R. Holdsworth, A. Costa, E. Poggio, A. Riva, P. Villari, Creep-fatigue interaction in two gas turbine Ni based superalloys subjected to service-like conditions, Materials at High Temperatures 31 (4), 2014, pp. 348–356. E. Vacchieri, A. Costa, E. Poggio, P. Villari, S.R. Holdsworth, Creep-fatigue life assessment strategy for gas turbine blades and vanes: safe life procedure development, assessment and experimental validation, ASME Turbo Expo 2016, 13 th − 17 th June 2016, Seoul, South Korea. E. Vacchieri, ”Creep-Fatigue Interaction and Small Size Testing Techniques”, Diss. ETH No. 23893, 2016, Zu¨rich, Switzerland. E. Vacchieri, A. Costa, P. Guarnone, E. Poggio, A. Sanguineti, P. Villari, S.R. Holdsworth, Service-like TMF tests for life assessment of a SX GT blade, Materials Science and Technology 33 (9), 2017. E.V. Zaratesky, J.S. Litt and R.C. Hendricks, Determination of Turbine Blade Life from Engine Field Data, Journal of Propulsion and Power 28 (6), Nov-Dec. 2012, pp. 1156-1167. 4. Conclusions References