EXPERIMENTAL AND NUMERICAL MODELING OF TOOL DESIGN EFFECTS ON FSW OF Al-Zn AND Al-Cu ALLOYS

Abstract

Present research deals with modeling the effects of tool designs and process parameters on friction stir welding (FSW) of AA7039 and dissimilar AA2219-AA7039 butt joints. FSW tools with varying probe diameter, shoulder diameter and shoulder surface geometries were utilized together with varying process parameters such as welding speed, tool rotational speed and axial force for producing AA7039 and AA2219-AA7039 butt welds. The tool shoulder diameter, probe diameter and shoulder surface geometries were varied by following design of experiment (DOE) technique. Experiments were conducted as per the DOE and response surface methodology (RSM) was used to predict the effect of tool geometries on the welds. Ultimate tensile strength (UTS), percentage of elongation (PEL) and weld cross section (WCA) were the responses from the experiment. Regression modeling results were observed to be adequate for predicting the weld responses. Computational fluid dynamics (CFD) model was also developed to predict the thermal conditions and material flow of aluminum welds with respect to different tool geometries and process parameters. The CFD model exhibited close agreement with the experimental temperature distributions with respect to the different FSW tool geometries and process parameters for joining of AA7039 and AA2219 butt joints. The present investigation discusses the effects of threaded friction stir welding (FSW) tool geometries on AA7039 aluminum alloy welds. However, the effect of tool shoulder geometry and concavity of the shoulder surface on the welds were also studied. Experiments were conducted to study the effect of tool probe and shoulder geometries on AA7039 aluminum welds with respect to weld tensile strength, cross section area and % elongation. A full factorial design matrix was utilized to manufacture twenty seven FSW tools having different levels of threaded probe diameter, shoulder diameter and shoulder surface concavity. A mathematical model was developed to predict the effects of the tool geometries on the welds using response surface regression analysis. The interaction effects of the control factors (tool geometrical parameters) on the responses such as weld strength, weld cross section and % elongation werestudied. The modeling methodology developed in the present investigation found to be adequate for predicting the effects of critical FSW tool geometrical factors on the weld. To study the effect of tool geometry and process variables on the ultimate tensile strength and % elongation of friction stir welding (FSW) of dissimilar AA2219-AA7039 butt joints, a plan of full factorial design (DOE) was implemented with different combinations of control factors. It iii was observed that, the extent of tool shoulder flat surface and tool rotational speed influenced the weld quality significantly. A mathematical model was also developed using response surface regression analysis to predict the effects of tool geometry and process variables on dissimilar AA2219-AA7039 welds. The weld microstructures and hardness evolution were investigated by employing electron backscatter diffraction (EBSD) technique and Vickers micro hardness testing machine respectively. The microstructural observations indicated that the grain size at advancing side (AA 2219 side) is much finer as compared to retreating side (AA7039 side). Hardness along the central line of the specimen showed uneven hardness distribution. The hardness values of the weld zone were lower than that of in base materials. The material flow and thermal conditions of AA 2219 and AA7039 FSW joints were studied by computational fluid dynamics (CFD) modeling. Four types of tools with threaded probes and varying shoulder geometries were investigated to determine the geometrical effects on material flow and thermal conditions in the AA 2219 and AA7039 welds. Experimental conditions such as the process variables were utilized as the input to the model together with the tool geometries. The CFD model exhibited most of the characteristics of friction stir welding such as differences in the advancing and retreating side temperature profiles. The resulting material flow closely indicated the shape of the welds produced with the respective tools. The predicted and observed thermal profiles also closely matched indicating the adequacy of the model