TIG and A-TIG welding experimental investigations and comparison with simulation Part 2 – arc constriction and arc temperature A. Berthier 1 , P. Paillard* 1 , M. Carin 2 , S. Pellerin 3 and F. Valensi 3 In this part II, the comparison of physical mechanisms between tungsten inert gas (TIG) and active TIG (A-TIG) welding is shown. The plasma was monitored by a high speed camera to present the arc constriction phenomena while passing from TIG to A-TIG. The elemental analysis and the arc temperature measured by optical emission spectroscopy were performed according to the type of welding and the different fluxes in A-TIG welding. The two-dimensional axial symmetric model presented in part I was used to simulate the flow behaviour in the melting pool realised on a stainless steel disc (304L) melted by a stationary heat source and to study the influence of energy density. Keywords: TIG welding, Activating flux, Arc constriction, Arc temperature, Simulation Introduction The active tungsten inert gas (A-TIG) welding process was developed in the early 1960s by the Paton Welding Institute (Kiev, Ukraine) to improve the tungsten inert gas (TIG) welding process. In A-TIG welding, a fine layer of activating flux, constituted of inorganic powder, is deposited on the steel plate before welding. The penetration depth was multi- plied by a factor of 1?5–4 while passing from the process TIG to A-TIG, depending on alloys. In this study, we considered only three mechanisms in the A-TIG welding process: the Marangoni effect combined with Lorentz forces and the arc constriction effect. The chemical mechanisms were already studied in part I. The electrical arc constriction is generated by the dissociation and ionisation of the components of the fluxes, as shown in Fig. 1. The use of fluxes such as fluorides, chlorides and oxides supports the arc constric- tion mechanism. 1 The arc constriction can have two consequences: an energy densification on the metal and/or an increase in the arc temperature. The arc temperature depends on the first ionisation potential of the metallic element. 2 The presence of oxides leads to a constriction of the plasma column and an increase in the arc temperature, which may explain the observed effects on the weld bead. 3 The fluorides (once dissociated) have great affinity for electrons. Arc constriction depends on the ability of the flux to combine with electrons. In A-TIG welding with fluorides or chlorides, the Marangoni convection remains centrifugal as in TIG welding. Some fluorides do not affect the chemistry of the melting zone. Consequently, some fluorides have only one effect on the arc, i.e., increasing the arc energy density. This is due to a global or local increase in the arc temperature or in the arc temperature gradient, depending on the fluoride. 4 Experimental Flux preparation and material conditions Before flux application and welding, the workpiece was cleaned with acetone and dried. In the present work, different fluxes were used. The first was a commercial flux composed of TiO 2 , Cr 2 O 3 ,V 2 O 5 , MgF 2 and MgCl 2 . The second flux contains the mixture of three oxides, such as TiO 2 , Cr 2 O 3 and V 2 O 5 , and TiO 2 , Cr 2 O 3 , K 2 Cr 2 O 7 , MgF 2 and BaF 2 were also studied separately. A planetary crusher was used to grind 15 g of powder (oxides and halides) with 15 mL of ethanol in a stainless steel bowl for 20 min at 360 rev min 21 in order to prepare the activating flux. The flux was deposited on the steel plate before welding by pulverisation using an aerograph. The geometry of the fusion line was pro- duced on austenitic stainless steel 304L rectangular plates (20065064 mm). The chemical composition of steel is 0? 013C–0?53Si–1?61Mn–0?017P–0? 011S–19?67Cr– 0?08Mo–9?95Ni–0?005Al–0? 072Co–0? 11Cu–0?019Ti–0? 046V (wt-%). Welding was conducted using a Fronius Magic Wave 2200 current supply and a Servisoud automatic welding bench for dynamic experiments. 1 LUNAM - Institut des Mate ´ riaux Jean Rouxel (IMN), UMR Universite ´ de Nantes CNRS, 2 rue de la Houssinie ` re, BP 32229, Nantes Cedex 3 44322, France 2 Laboratoire d’Inge ´ nierie des MATe ´ riaux de Bretagne (LIMATB), Universite ´ de Bretagne-Sud, Centre de Recherche, Rue Saint Maude ´, Lorient Cedex 56321, France 3 GREMI, Universite ´ d’Orle ´ ans/CNRS, Rue Gaston Berger, BP 4043, Bourges cedex 18028, France *Corresponding author, email pascal.paillard@univ-nantes.fr ß 2012 Institute of Materials, Minerals and Mining Published by Maney on behalf of the Institute Received 9 March 2012; accepted 30 March 2012 DOI 10.1179/1362171812Y.0000000025 Science and Technology of Welding and Joining 2012 VOL 17 NO 8 616