Welding is an essential and fundamental technology for manufacturing. When metals are joined by welding, the strength and other qualities of the products vary greatly depending on the welding conditions. In this paper, authors focus on ring mash welding used for powertrain components that transmit engine drive. This welding is required to be strong enough to withstand large loads. However, it is difficult to find appropriate welding conditions because there are many welding parameters, such as the welding current, applied force and shapes to be welded. Moreover, since the welding point is inside the base metal, it is difficult to understand the phenomena and the deformation process during welding. As its deformation is finished in several tens of milliseconds, visualization of the phenomena is difficult. Therefore, it takes much time to determine welding conditions through repeated experiments and to develop and improve the welding process.The numerical simulation is useful to understand these phenomena. If welding phenomena can be analyzed by a numerical model, it is possible to predict the results by trying different welding conditions and shapes of base metal on the numerical analysis, and it is easy to select more appropriate ones. Since the numerical analysis of welding phenomena is supported by theoretical analysis and plays an important role in quality assurance of welded products, the authors tried to make the numerical model using the fundamental experimental results. In this study, the visualization of the phenomena and the relationship between the welding current and the applied force made clear using the numerical simulation. The effect of welding conditions on the welding results when they are varied.