Transactions of the Architectural Institute of Japan
Online ISSN : 2433-0027
Print ISSN : 0387-1185
ISSN-L : 0387-1185
DEFORMATION CAPACITY OF DEFECTIVE WELDED JOINTS : Part 3 Non-linear fracture mechanical consideration on the size of defect and the plate width under the monotonic tensile tests
MORIHISA FUJIMOTOTADAO NAKAGOMIMITSURU IZUMI
Author information
JOURNAL FREE ACCESS

1983 Volume 328 Pages 71-82

Details
Abstract
The monotonic tensile tests of cruciform butt weld joints with incomplete root penetration were carried out to investigate the strength and the deformation capacity dependency on the size of defect and the plate width. And the test results were considered by means of J-integral analysis that is one of the non-linear fracture mechanics. The conclusions are summarized as follows. (1) The crack initiation was recognized before the maximum load in every specimens and the all fractures were occured due to stable crack growth. (2) Any difference of the strength and the deformation capacity cannot be recognized between the defect length and the defect height those defective areas are equal. (3) The crack length and the plate width dependency on the J-integral value was analyzed by the elastic-plastic finite element method. From the results, the relations between the J-integral value and the average elongation that gauge length is 300mm almost depend on only the crack length. (4) The equivalent through-thickness defect length is more reasonable than the ratio of defective area in the estimation of the maximum gross stress and the average elongation of defective welded joints. The gross maximum stress and the average elongation those are estimated by the equivalent through-thickness defect length are not depend on the plate width. This result corresponds to the result of J-integral analysis. (5) J-integral value and crack tip opening displacement those are parameters of fracture toughness and the average elongation have some relations each other. However, the relations are so different from the past studies that more experiments are needed to equationalize the relations.
Content from these authors
© 1983 Architectural Institute of Japan
Previous article Next article
feedback
Top