_STP_511-1972.pdf

PROBABILISTIC ASPEQS OF FATIGUE A symposium presented ai the Seventy-fourth Annual Meeting AMERICAN SOCIETY FOR TESTING AND MATERIALS Atlantic City,N.J.,27 June-2 July 1971 ASTM SPECIAL TECHNICAL PUBLICATION 511 Robert A.Heller,symposium chairman list price$19.75 04-511000-30 AMERICAN SOCIETY FOR TESTING AND MATERIALS 1916 Race Street,Philadelphia,Pa.19103 Copyright by ASTM Intl(all rights reserved);Mon Dec 21 11:08:31 EST 2015Downloaded/printed byUniversity of Washington(University of Washington)pursuant to License Agreement.No further reproductions authorized.BY AMERICAN SOCIETY FOR TESTING AND MATERIALS 1972 Library of Congress Catalog Card Number:72-75900 NOTE The Society is not responsible,as a body,for the statements and opinions advanced in this publication.Printed in Baltimore,Md.July 1972 Copyright by ASTM Intl(all rights reserved);Mon Dec 21 11:08:31 EST 2015Downloaded/printed byUniversity of Washington(University of Washington)pursuant to License Agreement.No further reproductions authorized.Foreword The Symposium on Probabilistic Aspects of Fatigue was presented at the Seventy-fourth Annual Meeting of ASTM held in Atlantic City,N.J.,27 June-2 July 1971.The sponsor of this symposium was ASTM Committee E-9 on Fatigue.Robert A.Heller,Virginia Polytechnic Institute,presided as symposium chairman.Copyright by ASTM Intl(all rights reserved);Mon Dec 21 11:08:31 EST 2015Downloaded/printed byUniversity of Washington(University of Washington)pursuant to License Agreement.No further reproductions authorized.Related ASTM Publications Fatigue at High Temperatures,STP 459,(1969),$11.25 Effects of Environment and Complex Load History on Fatigue Life,STP 462,(1970),$22.00 Manual on Low Cycle Fatigue Testing,STP 465,(1970),$12.50 Copyright by ASTM Intl(all rights reserved);Mon Dec 21 11:08:31 EST 2015Downloaded/printed byUniversity of Washington(University of Washington)pursuant to License Agreement.No further reproductions authorized.Contents Introduction 1 New Method for the Statistical Evaluation of Constant Stress Amplitude Fatigue-Test ResultsF.A.Bastenaire 3 Estimating the Median Fatigue Limit for Very Small Up-and-Down Quantal Response Tests and for S-N Data with Runouts R.E.Little 29 Regression Models for the Effect of Stress Ratio on Fatigue-Crack Growth RateB.Mukherjee and D.J.Burns 43 Comparison of Scatter under Program and Random Loading and Influencing FactorsG.H.Jacoby and H.Nowack 61 Investigation of Fatigue Life and Residual Strength of Wing Panel for Reliability furposesSigge Eggwertz 75 A Reliability Approach to the Fatigue of Structures.O.Payne 106 Discussion 150 Random Fatigue of 2024-T3 Aluminum Alloy under Two Spectra with Identical Peak-Probability Density Functions/?.C.Linsley and B.M.Hillberry 156 Application of the Monte Carlo Technique to Fatigue-Failure Analysis under Random Loading.Itagaki and M.Shinozuka 168 On the Probabilistic Determination of Scatter Factors using Miners Rule in Fatigue-Life Studies5.C Sat/nrfe-i 185 Copyright by ASTM Intl(all rights reserved);Mon Dec 21 11:08:31 EST 2015Downloaded/printed byUniversity of Washington(University of Washington)pursuant to License Agreement.No further reproductions authorized.STP511-EB/JUI.1972 Introduction The analysis and design of engineering structures and systems are,in most cases,carried out with the assumption that loads,environmental factors,and material properties are deterministic quantities,though it is generally recog-nized that most design parameters have statistical variations.Early experiments have already revealed a much wider scatter in fatigue test results than in most static tests and hence researchers have been concerned with the statistical interpretation of data.Because extrapolation and interpolation based on limited experimental evidence have been the aim of most tests,attention was initially focused on various suitable methods of data plotting,curve fitting,and determination of distribution functions for fatigue life.Today regression analysis and mathe-matical techniques of curve fitting,aided by electronic computers,have replaced curve fitting by eye.The use of the normal and logarithmic normal distributions has been supplemented with the extremal distributions of Weibull and Gumbel and various other transformations which either fit the data better or are based on an underlying physical process.To date proponents of various theories are not in complete agreement on the most suitable distribution functions for the description of the statistical variations of fatigue life.Response curves and confidence bands for fatigue strength and their relationship to Ufe distributions have also received considerable attention in the past.Methods are still being sought to determine a statistically justifiable fatigue limit based on a minimum number of test results.The recognition of the effects of variable loads on fatigue life has led to the extensive study of random processes and their application to the description of fa