酿造的
管道(软件)
材料科学
机械工程
工程类
地理
考古
作者
Kenneth George,Ming Gao,Ravi Krishnamurthy
标识
DOI:10.1520/stp164320220033
摘要
This paper qualitatively explores challenges encountered in obtaining JQ values during a failure investigation. The source material was a section of vintage seam-welded X52 transmission pipeline with wrinkle bends. Circumferential cracks developed at the apex of a few bends after a period of service. Metallographic samples taken from the apex of the bend revealed a Widmanstätten microstructure with minimal deformed grains, suggesting that a normalization heat treatment may have been applied after bending. Charpy impact results demonstrated that the bend region had a ductile-brittle transition temperature (DBTT) at or above the local environmental conditions due to localized microstructural changes. The details of the failure analysis are not the focus of this paper. Instead, this paper presents ASTM E1820 J-R testing on a small matrix of samples taken from the virgin pipe wall and the wrinkle bend areas. During J-R testing, the pipe wall material exhibited non-straight crack fronts due to microstructural inhomogeneity through-wall, and the wrinkle-bent material exhibited asperities and remaining ligaments in the crack wake. These anomalies presented challenges in obtaining accurate crack length measurements with the compliance technique alone. The direct-current electrical potential difference (DCPD) method was added as a simultaneous alternative crack measurement to improve the chances of getting usable crack length predictions during J-R testing. The accuracy of the compliance measurement was less affected by uneven crack fronts than DCPD. Conversely, the accuracy of the DPCP measurements was less affected by asperities and remaining ligaments in crack wake. These are unique and exaggerated test conditions. However, the results may be useful in parsing sources of error between compliance and DCPD crack length measurements. A quantitative analysis of the results is in progress.
科研通智能强力驱动
Strongly Powered by AbleSci AI