SEDL / STP / STP798-EB / STP33214S

Characterization of the Variability in Fatigue Crack Propagation Data

Ostergaard, DF
Engineer, Alcoa Technical Center, Alcoa Center, Pa.

Hillberry, BM
Professor, School of Mechanical Engineering, Purdue University, West Lafayette, Ind.

Pages: 19    Published: Jan 1983

Download this paper for $25 PDF (344K)          View License Agreement
Abstract

The variability in fatigue crack propagation is studied through the analysis of 68 sets of replica constant amplitude fatigue crack growth data. Each of the a versus N data sets are fit to the integral of a crack growth rate equation through a finite integral optimization routine. The routine determines the best fit crack growth parameters of the crack growth rate equation. Through characterization of the variability in the crack growth parameters and utilizing statistical theory, the variability in the a versus N data is predicted. The finite integral method of obtaining the crack growth parameters was shown to very accurately reproduce the a versus N data. The variability in the crack growth parameters showed a strong linear trend, and a regression analysis on the parameters provided information to accurately predict the variability in the fatigue crack propagation data.