Stochastic and systematic deviations of creep experiments in martensitic steels

Beschreibung

Extrapolating the lifetime of creep exposed materials is difficult, and a lot of strategies are provided for this task. Since tests can last up to 10+ years, the main challenge is to estimate the long-term behavior of the materials in question out of short-term experiments with the help of materials modeling in order to save time and experimental costs. However, one of the most important input parameters for the success of this strategy are thoroughly obtained experimental creep data. Literature sources indicate a huge scatter between results stemming from apparently identical creep experiments. Potential reasons for deviating results include slightly different material starting condition including composition and heat treatment, general inhomogeneity within a block of produced material, deviations from the nominal testing temperature, oxidation at the sample surface, interruptions of the creep tests etc. The presented research is examining residual impact factors within satisfying the standard for creep testing. A set of identical creep experiments (same temperature, stress, source of material etc.) with the material P91 is set up. The varying parameters in the study are the sample diameter, measuring type (interrupted and uninterrupted) and sample preparation order (heat treatment/machining). It is shown to what extent the variation of these parameters can lead to systematically varying results in the creep experiment. The results of the experiments are discussed, as well as potential strategies for reasonable accuracy of creep modelling.

Zeitraum	23 Mai 2023
Ereignistitel	6th International ECCC Creep & Fracture Conference
Veranstaltungstyp	Konferenz
Ort	Großbritannien/Vereinigtes KönigreichAuf Karte anzeigen