Investigation Of Two-Phase Viscous Liquid Flow

Compressor suction and discharge lines in refrigeration systems must be designed to move compressor lubricants along the pipeline. This paper presents experimental and analytical results that describe the characteristics of two-phase, viscous liquid flows. Smooth-walled, flat plate test sections have been designed to allow flow visualization studies of the oil flow in different flow orientations. Experimental pressure drop and liquid film thickness results obtained with air and 300-SUS alkybenzene oil will be presented. A variety of interfacial shear stress models obtained from the general two-phase flow literature have been applied to predict liquid film thickness. Comparisons between measured and predicted film thickness indicate that the measured film thickness is thinner, as expected, from a smooth surface approximation. Predictions of amount of oil holdup in large system pipes will also be presented. NOMENCLATURE D Tube diameter (m) f Friction factor fi Interfacial friction factor fs Smooth wall friction factor h Mean film thickness (m) L h Liquid film thickness (m) + L h Dimensionless liquid film thickness r Tube radius (m) Re Reynolds number h D Re Reynolds number on hydraulic diameter ReLF Liquid film Reynolds number v Re Vapor Reynolds number u Velocity (m/s) ui Interfacial Velocity (m/s) uv Vapor Velocity (m/s) * L u Liquid friction velocity (m/s) * s u Smooth wall friction velocity (m/s) * v u Vapor friction velocity (m/s) y Radial distance (m) y Dimensionless radial distance Greeks Γm Mass flow rate of liquid film per unit width of wall surface (kg/(m⋅s)) ε Wall-roughness height (m) μ Dynamic viscosity (Kg/(m⋅s)) ν Kinematic viscosity (m/s) ρ Density (kg/m) δ Surface tension (N/m) τ Shear stress Subscripts i Interfacial L Liquid v Vapor INTRODUCTION During the operation of a refrigeration system, the lubricating oil in the compressor invariably leaks through the seals and mixes with the refrigerant. Generally, the lubricant flows as a thin film along the walls of the pipe, driven by the turbulent flow of refrigerant vapor. Oils accumulated in the compressor discharge and suction lines must be circulated through the system by the vapor drag on the oil. If the vapor drag is insufficient, oil holdup increases. This can decrease the oil level in the compressor crankcase and lead to poor compressor lubrication and compressor failure. Literature for Oil Return Information in the literature for oil return in refrigeration systems is limited to a few rule-of-thumb correlations and experimental observations. Sharpe [1967] studied a glass P shaped trap between a horizontal and vertical length of suction line in a small refrigeration unit. For vapor velocities in the order of 3 m/s, oil transport by a slow moving oil film was observed. Jacobs et al [1976] studied oil transport with refrigerant vapor by watching the liquid accumulating in the lower unit of the test section. Tests were conducted for combinations of R-12, R-22 and 150 (~32 mm/s) and 300 SUS (~66 mm/s) napthenic oil. A criterion for minimum transport of oil in vertical tubes was correlated with buoyancy forces (i.e., the density difference between the liquid and the vapor, and the momentum flux of the vapor) in terms of minimum tonnage of the refrigeration unit. Hwang et al. [2000] investigated the flow characteristics of R134a and three kinds of oils (mineral ISO 10, alkylbenzene ISO 8, and alkylbenzene ISO 10) in the vertical upward suction line of the refrigeration system. Churn flow and annular flow were observed in their experiments. Mean oil film thickness was estimated from the amount of oil stored in the tube. They found that the oil with poor miscibility and high viscosity caused a larger oil volume stored in the tube. But with the increase of refrigerant mass flow rate, the influence of oil type and viscosity becomes less dominant. Fukuta and Yanagisawa [2000] studied the flow characteristics of air/oil mixtures in vertical upward pipes (inner diameter 8 mm and 10 mm). Two different mineral oils, VG56 (260 mm/s) and VG20 (73 mm/s), were used. They found that the air velocity was dominant over the transition of the flow pattern. Average oil film thickness was obtained by measuring the capacitance between two thin electrodes. The oil film thickness was found to increase with the increase of the oil flow rate, however, the fractional thickness increase was smaller than that of flow rate. Literature for Two-Phase Flow In addition to a few investigations related to the oil return problem, there are a large number of publications focused on the adiabatic, two-phase flow, which are mostly concerned with air-water or water-steam mixtures. Currently, very little information is available in the literature describing the characteristics of a viscous liquid film driven by a turbulent vapor. Friction Factor Friction factor is a very important parameter to characterize the system pressure drop. For single-phase flow, Churchill [1977] developed a clever correlation that combines the friction factor for the laminar, transition and turbulent flow regimes in circular pipes as below: