Chemical composition and weld cooling time effects on heat-affected zone hardness of line pipe steels
International Journal of Pressure Vessels and Piping
In the past few years, the pipeline industry in North America has faced a reported increase of pipeline girth weld failures predominantly in grade X70 pipe welded with cellulosic consumables. Weld strength matching and HAZ softening were two factors believed to have contributed to the concentration of strain during, or soon after, construction which ultimately lead to an overload failure of the girth welds. The work reported in this paper aimed to establish and clarify the correlation between steel chemical composition and the HAZ hardness in typical production girth welds and a range of weld thermal simulations. The first stage was to calibrate the influence of weld cooling rate of both thermal simulations and real welds in steels over a wide range of chemical composition, typical of commercial high strength API pipe. Second, the influence of Pcm, carbon content, Mo content and cooling time in terms of minimum and maximum hardness were investigated. The comparison between simulated HAZ's and real welds showed that thermal simulation produces samples that adequately duplicate the real condition in terms of maximum and minimum hardness and can accurately assess both hardening & softening which occur in different regions of the HAZ. The simulated samples provided increased areas of different HAZ regions in order to more accurately identify the range of hardness values across the HAZ which was then correlated to real weld hardness profiles. It is concluded that ultimate hardness in the HAZ is directly related to steel chemical composition in terms of Pcm and weld cooling rate. It is however apparent that the relative percentage change in hardness from the original parent pipe hardness (also called % HAZ softening), which is important in determining the strain capacity of the weld joint, is principally dependent upon steel processing, especially the degree and severity of thermomechanical processing. This is evident by the fact that same API pipe grades have a wide range in chemical composition and most importantly Pcm, and the response to field welding conditions can be quite different depending on weld heat input, pipe wall thickness and preheat, all of which define the weld cooling rate.
Open Access Status
This publication is not available as open access
University of Wollongong