The Round Robin Test on Normalization Method under Static Loading Conditions

It is well known that ferritic steels can display a lower fracture resistance values under rapid loading rates due to Dynamic Strain Aging (DSA) phenomenon. In order to consider DSA in application of Leak-Before-Break (LBB) in nuclear power plant. Korean Standard Review Plan (KSRP) 3.6.3-1 on LBB recommends to conduct dynamic loading J-R test for ferritic steel piping material. While the unloading compliance method is well established for J-R determination, it can not be applied to dynamic loading situation because of the impracticability of unloading at such high rates. Alternatively, the dc-electric potential drop (DCPD) method has been employed for crack length measurements in continuous dynamic loading. However, a severe shortcoming of the DCPD method was encountered in testing ferromagnetic materials with unclear definition of crack initiation point due to spurious potential peaks arising in the early stage of rapid loading. Recently, DCPD method was excluded from ASTM Method E1820-99 because of difficulties for application as well as the particular accuracy issue. The normalization method, developed by Landes et al., has no fundamental shortcoming in its application to dynamic fracture tests since only load and load line displacement need to be measured in order to determine J-R curves. In 2001, the normalization method was accepted as ASTM Method E1820-01 for use as a high rate J-R tests. Prior to its approval by ASTM, the High Rate Round Robin was conducted under the coordination of ASTM Subcommittee E08.08.02 for the verification of accuracy and reproducibility in 2000. In the High Rate Round Robin, however, the results of normalization method could not be compared with any of other well-established methods because no suitable method is available as a reference due to aforementioned difficulties. With the new Annex A15 developed for its implementation added in ASTM Method E1820-01, the need for systematic validation on the consistency of the normalization method with a well-established standard method became apparent. The present work has been conducted to meet this need