Theoretical study of hydrodynamic flow in thermal plasma devices

In this paper we focus our discussion on thermal plasmas created by electric arc. Due to the electrons mobility, the energy transferred by Joule effect to the gas, leads to temperature around 10 000 K. Thermal plasmas are very used in numerous processes like spraying, cutting, welding, waste treatment and are also present in numerous situations such as in interruption systems: high and low voltage circuit breakers or in natural phenomena: lightning strike. In order to be improved, thermal plasma processes necessitate a better understanding of the medium. Numerical modeling is now largely used to analyze and to predict the behavior of arcs and hydrodynamic flows to improve thermal plasma processes. Due to the increase of computer performances and to the availability of commercial computational fluid dynamics codes, the 2D modeling mainly developed in the 80’s and 90’s have been progressively replaced for a few years by 3D models. But is there always a real interest to develop such kind of codes if we consider the increase of accuracy in comparison with the increase of CPU time and the level of the assumptions? In order to give some elements for the discussion, we compare in the first part two physical conditions: one based on the flow injection and the other on the arc attachment, using 2D and 3D models. The legitimacy of the 2D model in then studied. In the second part, we illustrate through example the necessity of the use of 3D models due to complexities in configurations. Finally the difficulty of the 3D models validation is underlined.