A computational model of gas tungsten arc welding of stainless steel: the importance of considering the different metal vapours simultaneously

A 2D computational model of the mixing of multiple metal vapours into a helium arc in gas tungsten arc welding of stainless steel is presented. The combined diffusion coefficient method, extended to three-gas mixtures, is used to treat helium-chromium-iron and helium- manganese-iron plasmas. It is found that all metal vapours penetrate to the arc centre and reach the cathode, with iron vapour confined near the cathode tip, while chromium and manganese vapours accumulate about 1.5 mm above the tip. The predicted distributions of chromium, manganese and iron show reasonable agreement with published photographic images and radial distributions of atomic line emission intensities. The results are also consistent with published measurements of the deposition of the metals on the cathode surface. A detailed examination of the influence of the different diffusion coefficients, net emission coefficients and vapour pressures of the metals on the metal vapour transport in the arc plasma is presented. It is shown that cataphoresis (diffusion due to applied electric fields) leads to the penetration of the metal vapours into the arc. The different distribution of iron vapour from those of chromium and manganese vapours near the cathode is strongly influenced by the lower ordinary diffusion coefficients of iron at low temperatures. Radiative emission is found to be important since it leads to cooling of the arc, which decreases the influence of cataphoresis. The vapour pressure only influences the concentration of the metal vapour close to the workpiece. Results for the two-gas helium-chromium and helium-iron systems are compared to those for the three-gas helium-chromium-iron system. It is shown that it is important to consider the different metal vapours simultaneously to obtain an accurate calculation of the metal vapour and arc temperature distributions.