Flow Characteristics Of Gas-liquid Two-phase Flow In Small Square Channel

The rapid development of MEMS, micro-electronics, laser technology, and new Materials manufacturing bring about new needs of heat transfer technologies and flow characters in mini-channel. The thermal exchange based on the flow characters in mini-channel is an important challenge of these technologies. In recent years, many researches have paid much attention to the study of mini flow characters and heat transfer. However, the published experimental results and the flow mechanism lead to obvious confusions. The main purpose of the present study is to clarify the mechanisms of mini flow characters. Based on prevailing knowledge of flow patterns, void fraction and pressure drop characters, the experimental study and theoretical analysis on flow characters in mini-channel is performed in this work.Based on the observed flow pattern by high speed camera, experimental study was conducted to investigate flow characteristics for nitrogen and water flow in square mini-channel, with hydraulic diameter of 1mm.The gas and liquid superficial velocity ranges were 0.1-100 and 0.1-5m/s, respectively. Five typical flow pattern images were obtained in both vertical and horizontal flow direction in the square mini-channel. These flow patterns were dispersed bubble flow, capillary bubble flow, slug flow, churn flow, annular flow. Comparisons with the relevant literatures show that both cross-section shape and experimental condition have remarkable effect on the flow pattern transition.The perimeter and area of gas phase have been calculated by a method of image processing techniques. Based on the acquired flow pattern images the void fractions were obtained by computing the parameters with my supposed cylinder model.Comparisons with the available literatures show that Chisholm model provided the best agreement with the experimental data of two-phase flow pressure drop in the vertical square channel .For low superficial velocities condition, Lee&Lee model and Dukler model show a better predictive ability than the other experimental correlations.