Published: Jan 1994
| ||Format||Pages||Price|| |
|PDF (412K)||25||$25||  ADD TO CART|
|Complete Source PDF (15M)||25||$131||  ADD TO CART|
A random loading fatigue test program has been conducted on materials used in pressurized water reactors. The four materials considered were low alloy steels and austenitic stainless steels, including both base metals and associated welds.
The present study concerns crack initiation in the high cycle fatigue domain. Standardized Gaussian stationary sequences were chosen. Both analytical and numerical calculations have been performed in order to determine the parameters necessary for the accurate prediction of fatigue life under random loading.
The general objective of the study is to provide a better knowledge of safety margins taken in the design of nuclear components.
Analysis of results has shown: 1. for the irregularity factor, I = 99% (narrow band spectrum—overall mean stress equal to zero), the calculation of damage is possible using both analytical and numerical methods; and 2. for the other irregularity factors, I ≠ 99% (wide band spectrum—variable mean stress), the damage calculation seems possible using an analytical method or a semianalytical method, but it is difficult to perform.
Numerical calculations have been performed using a rainflow counting method of cycles, the Haigh diagram, constant amplitude S-N curve without fatigue limit, and the linear Miner rule of damage cumulation.
Numerical calculations led to predicted fatigue lives slightly higher than those obtained experimentally, for I = 99% and 70%. This excess of nonconservatism is accentuated in the case of I = 30%, except in the case of the austenitic stainless steel.
Good agreement was observed between damage values obtained using analytical and numerical calculations for the case of I = 99%.
high-cycle fatigue, random loading, stationary Gaussian sequences, damage calculations, rainflow counting method, austenitic stainless steel, low alloy steels, welds, fracture (materials), fatigue (materials), testing methods, test automation, data analysis
Research engineer, CEA-Saclay, Gif-sur-Yvette,
Senior research engineer, IRSID-Unieux, Firminy,
Le Duff, J-A
Engineer, FRAMATOME, Tour Fiat, Paris La Défense,
Senior research engineer, Westinghouse, Pittsburgh, PA
Paper ID: STP13955S