-150 -100 -50 0 50 0 50 100 150 200 distance from the surface temperature 1/4-T 3/4-T °C mm TT41J T0-dyn T0-stat Fig. 2 Effect of specimen location on the ductile-brittle transition temperature determined by dynamic (To dyn) and static (To stat) master curve procedure and Charpy-V impact tests (TT41J) STATISTICAL ANALYSIS The influence of the depth position is not relevant for specimen locations between 1/4- and 3/4-thickness. Thus, the results are assumed to come from a material with identical but strongly scattering toughness from which a data set arises consisting of about 150 observations. This is a base for a statistical analysis to find appropriate fits and to perform hypothesis tests for various parameters and assumptions. For this purpose, the data were newly arranged in 10 subsets of isothermal series with 8 to 22 observations. The validation of the Weibull-Wallin distribution, the temperature dependence of To, and the accuracy of the To determination as function of the number of tests were tested. The statistical treatment used the Maximum-Likelihoodmethod and the Kolmogorov- Smirnov-test, mainly [6]. The same investigation was conducted for the static data set. This help to separate effects due to the material from the effects caused by the loading condition.
RkJQdWJsaXNoZXIy MjM0NDE=