A unified numerical modeling of stationary tungsten-inert-gas welding process

  • PDF / 3,812,438 Bytes
  • 10 Pages / 612 x 792 pts (letter) Page_size
  • 3 Downloads / 178 Views

DOWNLOAD

REPORT


I. INTRODUCTION

HEAT transfer from the arc plasma to the weld pool plays an important role in the determination of the weld penetration in the arc-welding process.[1] Details of the fluid flow in the weld pool are important in determining weld shape. Taking account of all these phenomena is necessary for the development of a numerical model of the tungsteninert-gas (TIG) welding process because there is close interaction between the arc plasma and the weld pool. For example, there are four driving forces of fluid flow in the weld pool.[2] These are the drag force of the cathode jet on the liquid surface, the buoyancy force, the electromagnetic force due to the self-magnetic field of the welding current, and the surface-tension gradient force of the weld pool, as shown in Figure 1.[2] These driving forces are dependent not only on the physical properties of the weld metal but also the properties of the plasma state.[1] Therefore, a unified numerical model accounting for both plasma and weld-metal processes is important for predicting the TIG arc-welding properties. Modeling the arc-welding process has been tried by a number of researchers.[2–15] However, almost every numerical model has treated either only the arc plasma[3–8] or only the weld pool.[2,9–13] Then, calculated predictions, for example, for the weld pool, require distributions of heat flux and MANABU TANAKA, Research Associate, HIDENORI TERASAKI, Ph.D Student, and MASAO USHIO, Professor, are with the Joining and Welding Research Institute, Osaka University, Osaka 567-0047, Japan. Contact e-mail: [email protected] JOHN J. LOWKE, Chief Research Scientist, is with the Department of Telecommunications and Industrial Physics, CSIRO, Lindfield, NSW 2070, Australia. Manuscript submitted November 1, 2001. METALLURGICAL AND MATERIALS TRANSACTIONS A

current density to be specified at the anode surface. Recently, modeling the combined arc plasma and the weld-pool phenomena has been tried for stationary welding,[14,15] but the calculated results of the arc plasma and the weld pool were made separately, without interaction between the plasma and the weld pool. In the present article, we use a unified numerical model of stationary TIG arc welding. The basic model and procedure is that of Sansonnens et al.[16] but it is extended to include melting of the anode, with inclusion of convective effects in the weld pool. We give predictions of the twodimensional distributions of temperature and velocity in the whole region of the TIG welding process and also the predicted profile of weld penetration. Furthermore, quantitative values of the energy balance for the various plasma and electrode regions are given. We did not take into consideration the metal-vapor phenomenon from the weld pool and depression of the weld-pool surface. II. MODEL OF TIG WELDING PROCESS A. Governing Equations The tungsten cathode, arc plasma, and anode are described relative to a cylindrical coordinate, assuming rotational symmetry around the arc axis. The calculation domain is shown in Fig