A Probabilistic Method for the Fatigue Life Assessment of Powder Metallurgy Parts of Aircraft Engines

1996 ◽  
Vol 118 (2) ◽  
pp. 411-415
Author(s):  
R. Krafft ◽  
S. Mosset

This paper will present a probabilistic approach developed in order to assess the fatigue life of aircraft engine parts (turbine disks) obtained by powder metallurgy technique. First of all, the main issues will be pointed out and the theoretical principles of the method will be described. Then the design implications and the experimental correlation will be emphasized. The scale effect is a major concern for the fatigue life assessment of a powder metallurgy part. It no longer allows the designer to evaluate the life of a massive part directly from experimental results based on small specimen fatigue tests as is done in the classical methodology. In order to describe this scale effect correctly, incubation sites (inhomogeneities like ceramic inclusions) must be characterized. The size of these inhomogeneities and their positions in the part appeared to be the most relevant parameters. Hence the methodology developed at SNECMA integrates the scale effect scatter through a binomial probability distribution as well as a temperature and stress-dependent life evaluation for each inhomogeneity size and position. The life calculation of a part implies an analysis of its whole volume and surface. An iterative process determines the number of cycles corresponding to a global reliability level requirement for the part. The complete methodology is then validated by comparing the calculated initiation distribution with experimental results on small specimens and test disks.

Author(s):  
R. Krafft ◽  
S. Mosset

This paper will present a probabilistic approach developed in order to assess the fatigue life of aircraft engine parts (turbine disks) obtained by powder metallurgy technique. First of all, the main issues will be pointed out and the theoretical principles of the method will be described. Then the design implications and the experimental correlation will be emphasised. The scale effect is a major concern for the fatigue life assessment of a powder metallurgy part. It no longer allows the designer to directly evaluate the life of a massive part from experimental results based on small specimen fatigue tests as it is done in the classical methodology. In order to correctly describe this scale effect, incubation sites (inhomogeneities like ceramic inclusions) must be characterised. The size of these inhomogeneities and their positions in the part appeared to be the most relevant parameters. Hence the methodology developed at Snecma integrates the scale effect scatter through a binomial probability distribution as well as a temperature and stress dependent life evaluation for each inhomogeneity size and position. The life calculation of a part implies an analysis of its whole volume and surface. An iterative process determines the number of cycles corresponding to a global reliability level requirement for the part. The complete methodology is then validated by comparing the calculated initiation distribution with experimental results on small specimens and test disks.


Author(s):  
Esakki Muthu Shanmugam ◽  
Raghu V. Prakash ◽  
Shakthivel Ammaiappan

The fatigue life of the titanium alloy axial compressor rotor blade was estimated based on stress based life method. The fatigue life of the compressor blade was evaluated through incremental amplitude test method. The incremental amplitude test method involves cumulative fatigue damage at different stress levels by using Miner’s Hypothesis. The probabilistic analysis of fatigue life was carried out by Weibull distribution method. The analytical and test methods results were compared and found satisfactory.


Author(s):  
Jürgen Rudolph ◽  
Paul Wilhelm ◽  
Armin Roth ◽  
Matthias Herbst ◽  
Matthias C. Kammerer ◽  
...  

Fatigue life assessment for pressurized components in nuclear power plants (NPPs) is an essential part of the aging management (AM) ensuring safe and long term operation (LTO). For fatigue life assessment different codes and standards provide a variety of methodologies with variable complexity. Major fatigue life influencing factors such as temperature, surface finish, multiaxiality, loading history and others are often considered in a more or less global way by combined overall reduction factors covering multiple mechanisms together. Other effects such as the environment or hold times are often considered not at all or otherwise with high levels of conservatism resulting in large discrepancies between calculated fatigue life and practical experience from power plant operation. In order to reduce this inadequacy a more accurate fatigue lifetime assessment concept including individual fatigue life influencing factors in a mechanistic manner is required. Nevertheless, these amendments are to fit into the existing basic engineering approach of design against fatigue failure as it is implemented in nuclear standards and design codes. In the framework of an ongoing three years German cooperation R&D project with participation of the Materials Testing Institute MPA University of Stuttgart and AREVA GmbH (Erlangen) it is the aim to both improve the state of the art based on an experimental program for some of the main fatigue life influencing factors and on the derivation of a practicable engineering fatigue assessment procedure. Within this fatigue assessment procedure the dominant fatigue life influencing factors are considered individually. The experimental program covers fatigue test results for austenitic and ferritic piping materials including a dissimilar metal weld. Within the testing program strain controlled fatigue tests were performed with and without hold-times in air and high temperature water environments. Smooth and notched specimens provide a database to study the influence of notches and multiaxiality. These results are used to state on the applicability of commonly used failure hypothesis like von Mises and Tresca in comparison to advanced fatigue damage parameters. In addition to constant amplitude strain controlled fatigue testing load spectra were investigated. Thereby fatigue cumulative damage models like Miner’s rule can be evaluated. This publication constitutes a follow-up to a previous paper [1] and targets at the presentation of experimental results in conjunction with potentials with an improved fatigue assessment concept. In addition the methodology of the concept is applied to experimental results on fatigue life assessments for piping materials published by other organizations. Requirements for further experimental investigations towards the verification of a closed concept are formulated.


Sign in / Sign up

Export Citation Format

Share Document