Premium
Statistical characterization of fracture in the brittle‐ductile transition regime of structural steels
Author(s) -
Sun DongZhi,
Dormagen Detlef,
Dahl Winfried
Publication year - 1987
Publication title -
steel research
Language(s) - English
Resource type - Journals
eISSN - 1869-344X
pISSN - 0177-4832
DOI - 10.1002/srin.198701487
Subject(s) - fracture toughness , weibull distribution , brittleness , materials science , transferability , fracture (geology) , fracture mechanics , stress (linguistics) , toughness , composite material , metallurgy , forensic engineering , mathematics , engineering , statistics , linguistics , philosophy , logit
In the brittle/ductile transition region the transferability of elastic‐plastic fracture mechanics values of steel BS 4360 (Grade 50 D) by means of statistical analysis was tested. The scatterband of J ‐values was reduced with increasing crack length and specimen size. Especially smaller specimens tend to show a more ductile behaviour due to preceding stable crack growth. The scatter of the fracture toughness data can be described by Weibull distribution, but due to differences in the stress state problems arise when the ‘weakest‐link’ model is used to describe the failure probability of larger specimens. This is valid for ductile behaviour, even if stable crack extension is taken into account, and also for pure brittle behaviour. There is need for a parameter describing the stress state of the various geometries which can be incorporated into the ‘weakest‐link'‐model. If the lower limit of the scatterband should be evaluated and successfully used for the description of the failure probability, a thickness dependent number of specimens has to be tested, especially when data of small specimens are used. Initiation values for stable crack extension, derived by the DCPD‐method, describe the lower limit of the scatterband of brittle fracture toughness data obtained at the brittle‐ductile transition regime.