J. Marine Sci. Appl. (2012) 11: 493-503 DOI: 10.1007/s11804-012-1160-z Modeling the Effects of Tool Shoulder and Probe Profile Geometries on Friction Stirred Aluminum Welds Using Response Surface Methodology H. K. Mohanty 1 , M. M. Mahapatra 1 , P. Kumar 1 , P. Biswas 2* and N. R. Mandal 3 1. Mechanical & Industrial Engineering Department, IIT, Roorkee-247667, India. 2. Department of Mechanical Engineering, IIT, Guwahati-781039, India 3. Department of Ocean Engineering & Naval Architecture, IIT, Kharagpur-721302, India. Abstract: The present paper discusses the modeling of tool geometry effects on the friction stir aluminum welds using response surface methodology. The friction stir welding tools were designed with different shoulder and tool probe geometries based on a design matrix. The matrix for the tool designing was made for three types of tools, based on three types of probes, with three levels each for defining the shoulder surface type and probe profile geometries. Then, the effects of tool shoulder and probe geometries on friction stirred aluminum welds were experimentally investigated with respect to weld strength, weld cross section area, grain size of weld and grain size of thermo-mechanically affected zone. These effects were modeled using multiple and response surface regression analysis. The response surface regression modeling were found to be appropriate for defining the friction stir weldment characteristics. Keywords: friction stir welding (FSW); tool geometries; mechanical properties; microstructures; response surface; regression modeling Article ID: 1671-9433(2012)04-0493-11 1 Introduction 1 Friction stir welding (FSW) process has gained popularity for joining the aluminum alloys used in structural fabrication. FSW was invented at the Welding Institute in the early 90’s (Thomas et al., 1993), Cambridge. During FSW, to form the welded joint, a plunged rotating tool with shoulder and protruding pin is utilized. The rotation of the tool provides the frictional heat to the intended weld joint primarily through the shoulder and the plunged tool pin in between the mating surfaces of the joint, facilitates stirring of the joint (Fig. 1) material. The shoulder also prevents the material expulsion and assists material movement around the probe. The heat generated during the FSW process is not severe enough to produce defects those are generally observed during arc welding (Thomas et al., 1999). Fig. 1 Schematic of friction stir welding The main process parameters of FSW (Rajakuma et al., Received date: 2011-08-02. *Corresponding author Email: biren@isical.ac.in © Harbin Engineering University and Springer-Verlag Berlin Heidelberg 2012 2010) are the rotational speed of the tool, tool traverse speed, and vertical pressure on the plates during welding. The FSW tool geometry which involves the geometry of the tool shoulder and tool pin probe is also of immense importance as the weld strength and grain size (Su et al., 2003a) of weld are affected by it. The designing of FSW tool is still an evolving field of research and there can be many possibilities of tool geometries. The effectiveness of FSW joint is strongly influenced by several tool geometry parameters; in particular geometrical parameters such as the height and the shape of the probe and the shoulder surface of the tool both on the metal flow and on the heat generation due to friction forces (Leal et al., 2008). The FSW machine setting parameters like rotational speed of tool, tool traverse speed, vertical pressure on the tool are also important which together with the geometries of tool affect the weld (Fujii et al., 2006). The microstructure evolution during the FSW process (Su et al., 2003b; Biswas and Mandal, 2009; Dawes and Thomas, 1999) is also controlled by the material physical and thermo-mechanical characteristics. The influence of the tool rotation speed (Sato et al., 2002), welding speed (Lee et al., 2003; Boz and Kurt, 2004) and both parameters simultaneously on the microstructure and mechanical properties of different aluminum alloy welds by considering the same tool geometry have been analyzed in detail by some authors. Limited research has been carried out on the effects of tool geometrical structures (Leal et al., 2008; Lee et al., 2003; Boz and Kurt, 2004; Scialpi et al., 2007) on friction stir welds, while much work, focused on the variation of rotation and welding speeds to optimize the