Assessment Of Microstructural And Property Prediction Equations In Structural Welding

Abstract

ABSTRACT With the greater demand for improved structural reliability and increased productivity, fabricators of large structures can no longer afford to engineer by trial and error. The cost of rework of fabricated structures with defects is prohibitive. Sound weld joints must be produced as specified at the first attempt. Consequently, predictive equations that can reliably estimate the microstructures and properties of a weldment from processing parameters are in great demand. A review of the literature was carried out to examine the currently available equations for weld property and microstructure prediction. Empirical equations obtained from simple regressional analysis were shown to be reliable in predicting weld properties and microstructures. However, large amounts of experimental data must be generated for each specific group of materials used for modeling. Theoretical models that take into consideration the thermodynamics and kinetics of phase transformations showed improved prediction ability. It is also concluded that the development of models incorporating welding process parameters and chemical composition in the form of equal property diagrams should be investigated further. INTRODUCTION Compared with other manufacturing processes, welding has established a reputation of being more difficult to control and less likely to achieve consistent quality. This misconception is probably a result of the multiplicity of interrelated control parameters, the complexity of the control relationships and the difficulty of monitoring process performance. When arc welding is chosen as the method of fabrication, suitable welding procedures must be established. A formal qualification of the welding procedure, which involves completion of sample joints according to an agreed welding procedure specification with non-destructive and mechanical testing, must be carried out. During the actual fabrication, testing of the completed structure with nondestructive test techniques is again performed to check the weld quality. If necessary, defects are removed and the joint repaired. Progressive monitoring during production is strongly encouraged to prevent costly rework after final inspection. As the costs involved in these operations are considerable, welded joints must meet the required project specifications at the first attempt. Consequently, the ability to predict the microstructure and mechanical properties of weld joints becomes extremely important. Predictive equations can be empirically developed, based on statistical analysis of experimental data, or theoretically derived according to fundamental principles. This paper reviews a broad range of predictive equations that estimate weldment microstructure and mechanical properties. WELD METAL MICROSTRUCTURE AND PROPERTIES DEVELOPMENT The microstructural development of a weld metal is a function of its chemical composition and cooling rate. In a structural steel weldment, the amounts of hardenability elements and inclusion formers will determine the decomposition behavior of austenite and the phases that result. Consequently, on a continuous cooling transformation diagram, the position of the transformation start curves of grain boundary ferrite, acicular ferrite, bainite, and martensite in the temperature-time space is determined by the chemical composition of the alloy.