CONDENSATION OF REFRIGERANTS OVER SINGLE HORIZONTAL INTEGRAL-FIN TUBES

Abstract

The surface condensers are widely used in the refrigeration and air-conditioning industries, petrochemical industries, process industries, thermal and atomic power plants and allied industries. Numerous augmentation techniques for energy conservation and for improving the vapour to the coolant heat transfer rate have been reported in literature. These techniques can reduce the size of condensers in a significant manner. A thorough review of literature shows that, the heat transfer coefficient can be enhanced many folds by a simple technique of providing integral-fins over the tube surface. The present investigation is an endeavour to study the augmentation of heat transfer by enhancing the condensing side heat transfer coefficient during condensation of the saturated vapours of refrigerants R-245fa and R-236fa over 18 single horizontal tubes including 15 different integral-fin tubes. R-245fa and R-236fa, are new environment friendly refrigerants and potential substitutes for R-11, R-113 and R-123 as they have zero ozone depletion potential. The investigation was initiated by acquiring data for the condensation of refrigerant, R-134a, R-245fa and R-236fa over a plain tube in the experimental set-up. The results for the condensation over a plain tube established the integrity of the experimental set-up. The plain tube results are also used as reference data for the comparison of the performance of other finned tubes. In past, several investigations have been carried out to find the optimum fin density (fin-spacing) for the condensation of R-11 over circular integral-fin tubes (CIFTs) and it has been established by these investigators that a CIFT with 1417 fpm fin density produces maximum increase in heat transfer rates and yields 2 to 3 times augmentation in heat transfer coefficient. R-245fa and R-236fa are new refrigerants in the market and the information regarding their thermal performance during condensation over horizontal tubes is not available in the literature. Therefore, experiments have been conducted for the condensation of R-245fa and R-236fa. The fin density was systematically varied by four CIFTs (787 fpm, 1102 fpm, 1417 fpm, and 1732 fpm) for R-245fa and five CIFTs (787 fpm, 1102 fpm, 1417 fpm, 1732 fpm and 2048 fpm) for R-236fa to establish the best value. The 1102 fpm and 1417 fpm fin density tubes turned out to be the best performing tubes with an enhancement factor, EF, equal to 7.92 and 7.87 for the condensation of R-245fa and R-236fa respectively. All the data for the condensation of R-245fa and R-236fa have been acquired by keeping the temperature of saturated vapours constant at 318±0.5 K. (approximate temperature of refrigerant in the condenser of a refrigeration plant) and cooling water flow rate was varied from 400 kg/h to 1300 kg/h in the increment of 50 kg/h. In order to further increase the heat transfer coefficient, the spine integral-fin tubes (SIFTs) are introduced. The spines are generated by cutting axial slots on the surface of best performing CIFT. The depth of slots is purposely kept less than the height of the circular fins to ensure the proper drainage of the condensate from the tube surface. The experiments have also been performed for the condensation of R-245fa and R-236fa vapours over these tubes. It is discovered that in comparison to CIFT, the SIFTs increase the heat transfer coefficient further by approximately 23 percent for the condensation of R-245fa and by 10 percent for the condensation of R-236fa. The following equations have been developed to correlate the ratio of heat transfer coefficients of SIFT-1 and CIFT-2 for R-245fa and of SIFT-2 and CIFT-3 for R-236fa at given heat flux for the condensation. iv (h0)SIFT -1 = 0.86 qoi (ho )GIFT-2 and ( h =1.094 (ho)CIFT-3 q Similarly, the following equations have been developed to correlate the ratio of heat transfer coefficients for SIFT-1 and CIFT-2 & SIFT-2 and CIFT-3 with each other for the condensation of R-245fa and R-236fa. (k )sIFT-1 .\■ ,.(k )CIFT-2 AT), =1.21 and ( k )siFT-2 =1.1 (ho )cm-T-3 AT, Hence, it can be concluded that the SIFT outperforms the CIFT for the condensation of R-245fa and R-236fa in the range of ATf investigated. Later, to investigate the position on the tube surface where spines are most effective, partially-spined circular integral-fin tubes (PCIFTs) are manufactured. These tubes have the spines either on the upper half or on the lower half of the surface of the best performing CIFT. The dividing plane lies on the axis of the tube. The spines are more effective in the lower half of the tubes for the condensation of R-245fa and R-236fa and improve the performance of best performing CIFT by approximately 19 percent for the condensation of R-245fa and 7 percent for the condensation of R-236fa. An uncertainty analysis of the experimental results has also been carried out. The uncertainties in condensing side heat transfer coefficient for the condensation of R-245fa and R-236fa have remained in the range of 4-6 percent. The experimental heat transfer coefficients for the condensation of the pure vapour of R-245fa and R-236fa do not show a satisfactory agreement with different models. However, the measured heat transfer coefficients of R-245fa are in the best agreement with those predicted by Honda and Nozu model (1987). The data are overpredicted in a range of 12 to 28 percent for the condensation of R-245fa. Whereas, for the condensation of R-236fa the data are in the best agreement with those predicted by Adamek and Webb model (1990). This model over predicts the best data in the range of 9 to 25 percent. The condensing side heat transfer coefficient has also been determined by the modified Wilson plot technique. For the condensation of R-245fa the modified Wilson plot technique underpredicted the heat transfer coefficient in a range of 9 to 30 percent, whereas, for the condensation of R-236fa the underprediction has been in a range of 14 to 29 percent. From the present experimental results a correlation has also been developed between different dimensionless numbers to find the condensing side heat transfer coefficient. The dimensionless heat transfer coefficient is expressed in the form of Condensation number (CN) and has been correlated with the condensate Reynolds number (Re), condensate Weber number (We) and non-dimensional tube geometry(Y). CN = 0.098 Re-Y3weo 48y1.1 Where, 112 CN =h„[ ; -° k p2 g 4 Re = th AlPf rh = p.AF .V f Weber number 26 1 1) (We) = Op pfx of rt rb Pg vi