Staircase法による疲れ試験 : 試験片の数が少ない場合(続報)

概要

論文の詳細を見る
In the conventional method of the fatigue test, the accuracy of the endurance limit cannot be estimated. On the other hand the statistical methods represented by the probit method, the staircase method and so on require large number of test pieces. In the previous report the fatigue test by staircase method with small samples was presented. However it was assumed that the standard deviation σ of the endurance limit is known, and the accuracy of the endurance limit was not thoroughly discussed. This report treats the case in which σ is unknown. Since the size of the step in the staircase method must be approximately equal to σ, it is necessary to find σ. The fatigue test is performed in the neighborhood where the S-log N curve is represented approximately by the straight line having a negative slope. The experimental results are plotted in the S-log N diagram, and the regression analysis is applied to find σ. This method is presented and thebasic assumptions are considered. The methods of calculating the confidence limit of the mean value and the tolerane limit of the endurance limit are discussed. "The tolerance limit" corresponds to "the lower limit of the endurance limit", and it is defined from the statistical point of view. At the end of this report these methods are applied to some experimental examples.
一般社団法人日本機械学会の論文
1961-07-25