HEAT TRANSFER AND PRESSURE DROP IN FORCED CONVECTION BOILING OF REFRIGERANT MIXTURE INSIDE HORIZONTAL TUBES

Abstract

A review of literature reveals that binary refrigerant mixtures involving non-isothermal phase change results in thermodynamic advantage of performance. The feasibility of practical refrigeration systems using mixed refrigerants requires generation of performance data. The objective of this investigation is to study the heat transfer and pressure drop behaviour of mixture of refrigerants 13 and 12. The present work provides experimental data on several compositions of the mix-tures during flow boiling in a horizontal tube. The test evaporator consisted of two horizontal identical stainless steel tubes, each 2.35 m long, having inside and outside diameters as 9.52 and 12.50 mm, respectively. The test evaporator tubes were indepen-dently heated by passing controlled stabilized low voltage alternating current directly through them. These tubes were connected by a smooth copper U--bend with glass tubes to provide electrical isolation at the inlet and the exit of both the test sections. Copper constantan thermocouples were mounted at 24 locations, at a regular interval of 200 mm, on the top, side and bottom positions of the test evaporator. Refrigerant temperatures and pressures were measured at the inlet and the exit of the expansion valve, the first test section and the second test section. An oil separator was installed in the ( iv) discharge line to make circulating refrigerant effective-ly free of lubricating oil. A preheater was used to obtain the desired quality of refrigerant mixture at the inlet of the first test section. A back pressure regulating valve was used to maintain a constant pressure at the test section outlet. Measurements of power input to the test sections and preheater, pressures and temperatures were recorded. Since flow boiling shows random fluctuations, three to four readings were taken for each run to get represen-tative mean steady state values. The experimental data was collected for the following ranges of the operating variables. Concentration of : 0 to 20 percent by mass in R-13 in R-12 the interval of 5 percent Flow rate : 60 - 120 kg/hr. 2 (234 to. 454 kg/m .$) Heat flux : 5000 to 17000 W/m2 Thermodynamic and transport properties of the pure R-i2 and mixtures of refrigerants 13 and 12 have been deter-mined using equations from literature [1,12,13,46]. The observed pressure drop across each test tube was found to be small. It vas, therefore, considered reasonable to assume a linear variation of fluid pressure along the test length. Starting with the assumption that refrigerant temperature varied linearly along the test length, the correct temperature was obtained that satis-fied-mass and energy balance at each station. In addition, (r) the local composition of the liquid and the vapour phases and the quality of the flowing mixture were also computed. The two-phase flow pressure drops were measured and examined for the applicability of two existing correlations, viz. those of Martinelli-Nelson and Chisholm-Sutherland; the agreement was not satisfactory. However, the latter method has been found to be quite successful for pure R-12. Also it could be established that pressure drop is a function of mixture composition. Correlations of the following general forms, related tn both the above methods, were developed to predict the pressure drop on the basis of the pressure drop measurements of this investigation. However, correla-tions based on Chisholm-Sutherland pressure drbp method were relatively superior to those based on Martinelli-Nelson method. ApTp/ LipCl (1+C) n1 n, °Thu/ C2 (1-0 4 where op is calculated with the help of either MN or..: CS method. The variation of the circumferential wall tempera-tures and the refrigerant temperature were studied during two-phase flow. Random and complex variations of wall temperatures were observed at the top, side and bottom (vi) positions of thermocouples along the test length. The refrigerant temperature decreased from the entrance to the exit of the test evaporator for R-12; it was due to pressure drop. Due to non-isothermal boiling of binary mixtures, in general, there was an increase in refrigerant temperature from the entrance to the exit of the evaporator. It was found that the effect of quality, heat flux, mass velocity on the boiling of mixtures is similar to that of boiling of pure refrigerants. The effect of mixture composition was found to be complex. The heat transfer coefficients increased at higher concentrations of R-13 at high vapour qualities. The trend is reversed at low quality. This reversal of trend tends to occur at about 0.3 quality. Mixtures with 15 percent R-13, however, showed different effect of increase of vapour quality.