Thermal analysis of the arc welding process: Part II. Effect of variation of thermophysical properties with temperature

This article is Part II of a two-part series on the thermal analysis of the are welding process. In Part I, general solutions for the temperature rise distribution in are welding of short workpieces were developed based on Ja eger's classical moving heat source theory for a plane disc heat source wit h a pseudo-Gaussian distribution of heat intensity and constant values of t hermophysical properties at one temperature (400 degreesC). This was extend ed in this investigation (Part II) to consider different thermophysical pro perties at different temperatures (from room temperature (25 degreesC) to 1 300 degreesC) for a mild steel work material. The objective is to develop a rationale for the selection of an appropriate temperature for the choice o f the thermophysical properties for the thermal analysis of are welding. Si nce the quality of the weld for a given work material depends both on the t hermodynamic and kinetic considerations, namely, the maximum temperatures a nd the temperature gradients (cooling rates) in appropriate sections of the welded part including the weld bead and the heat-affected zone (HAZ), they were determined in this investigation. The main output parameters from a t hermal point of view are the widths and the depths of the melt pool (MP) an d the HAZ at the weld joint. Although the length of the weld pool is also a consideration, if the entire length participates in the welding process, w hich is generally the case, then this is not such an important consideratio n. It is found that for welds produced in a conductive mode only (i.e., not considering the case of deep penetrating welds produced with keyhole mode) , the values of the widths and the depths of the MP and the HAZs are nearly the same (within 10 to 20 pct), irrespective of the values of thermal prop erties for temperatures in the range of 400 degreesC to 1300 degreesC. Henc e, the emphasis on the need to consider variable thermal properties with te mperature in welding appears to be somewhat exaggerated. Also, based on the thermal analysis of the welding process, it appears that the room-temperat ure thermophysical properties may not be appropriate, as rightly pointed ou t by other researchers. The thermal history and the cooling rates were also determined analytically for are welding of long workpieces, where quasiste ady-state conditions are established and the boundary effects can be ignore d, as well as short workpieces, where transient conditions prevail and boun dary effects need to be considered. This information can then be used in th e appropriate time-temperature-transformation (TTT) diagram for a given ste el work material to investigate the nature of the metallurgical transformat ion and the resulting microstructure in the welding process both in the wel d bead and in the adjacent HAZs on either side.