Dissimilar metal welds (DMWs) between austenitic and ferritic steel tubing and piping are commonly employed in high-temperature applications in energy conversion systems. Differences in coefficient of thermal expansion between the two types of steel induce thermal stresses at the welds and local metallurgical changes near the low alloy steel/weld metal interface due to prolonged service at an elevated temperature. These phenomena, together with the differences in creep behavior of the materials joined, render the DMWs more prone to failure than welds between similar steels. This has been reflected in relatively high failure rates in DMWs in certain service applications (e.g., in utility power plant boiler tubing). Typically these welds fail by low ductility cracking in the low alloy steel at, or very close to, the fusion line. A project, sponsored by the Electric Power Research Institute (EPRI) and managed by the Metal Properties Council (MPC), has made significant headway over the last three years in understanding the failure modes and causes involved and in developing methods to assess residual life of DMWs. Welds from service in superheaters and reheater tubes and from laboratory simulation tests were examined to establish metallurgical characteristics and failure modes. Three failure modes were identified: (i) Prior austenite grain boundary cracking in the ferritic steel, one or two grains away from the fusion line; this mode was mainly observed in DMWs made with stainless steel filler metal. (ii) Cracking along the weld interface, which occurred in DMWs made with nickel-base filler metal. (iii) Propagation of cracks initiating from oxide notches formed at the weld outside surface; this mode occurred mainly in thin-walled tubes. Creep damage induced by steady and cyclic loading was found to be the predominant mechanism for damage and failure; therefore a dependence of damage on loading levels and service temperature was established. It was also determined that failure susceptibility in DMWs made with nickel-base filler was strongly influenced by the type of microstructure that formed at the low alloy steel/weld metal interface. The technique developed for estimating the condition and remaining life of DMWs in service involves detailed assessment of loading histories to which the welds are subjected, along with the use of empirical quantitative relationships established from both laboratory and service data. The methodology assumes that damage results from the combined effects of self damage (caused by thermal cycling of materials of different expansion coefficients) and service loadings, including both primary loads (e.g., pressure and deadweight) and secondary, or cyclic, loads due to the constrained thermal expansion of the system as a whole. The technique, Prediction Of Damage In Service (designated PODIS), has been found to adequately predict levels of damage in stainless-based DMWs in service. It is currently being developed further to embrace nickel-based DMWs.
Modeling, Prediction and Validation of Thermal Cycles, Residual Stresses and Distortion in Type 316 LN Stainless Steel Weld Joint made by TIG Welding Process
