EXPERIMENTAL STUDIES ON CONDENSATION AUGMENTATION OF REFRIGERANT BY TURBULENCE PROMOTER

Abstract

In the present work heat transfer augmentation has been studied caper cattail during condensation of 022 vapour inside horizontal tube with full-length tape inserts as well as segmented tape inserts in 1/4, 1/2 and 3/4 length of the full-length tube. The plain tube data were also obtained for the sake of comparison. The experimental set-up is mainly a well instrumented 5 ton refrigeration capacity) (17.5 kW of refrigeration) vapour compression system driven by an open compressor. The system consisted of a pre-condenser, test-condenser, after-condenser, expansion valve, needle valve, refrigeration bY-pase line, evaPeraber and 0 receive' which permitted a wide range of operating conditions. The test-condenser consisted of four identical length of 12)2 ram TD, 15.8 mm OD and 0.95 nl long hard drawn copper tubes. These test tubes were located concentrically inside an outer copper tube of 50 mm ID. Thus, the two concentric tubes formed Che counter flow test-section. These sections were placed in series ; with the refrigerant flowing inside the inner tube while the coolant water was flowing in the outer annular space. Outer wall temperatures of the inner tube were measured at four said locations. At each of these four locations, three, copper-constantan thermocouples were welded at the top, side and bottom positions. Experiments were conducted in the following ranges of parameter. working fluid t 022 Inlet degree of superheat: 5 to 10 k Refrigerant mass velocity: 209.37, 236.77, 281.72 326.85 and 372.03 kg/a-m2 Twist ratio : 5.9, 9.05 and 14.95 cooling water flow rate 200-1200 kg/hr Coolant water temperature. 20 to 30.0 Average condensing temp. : 37.5-53.8"C Vapour quality range 0.99-0.065 Average cooling heat flux: 21312.01-4088.89 W/m2 range The average heat transfer and average vapour quality were calculated for each test-sections and for each run and for all flow conditions. The variation of heat transfer coefficients with' the main parameters, such as mass flux, vapour quality, vapour mass velocity, vapour Reynolds number has been studied. The heat transfer coefficient increases with the increase of vapour mass velocity, mass flux, and vapour quality. A comparison was made between the plain tube experimental data and predicted heat trasfer coefficients calculated with the help of correlations suggested by (i) Akers, at al. [1], (ii) Akers-Rosson [2] (iii) ibis [3],(iv) Azer, et al. [10] and (v) Tendon [64]. It was found that the best agreement occurs with the correlation of Akers-Rosson [2]. Also a generalized best fitting Akers-Rosson type correlation was developed. Experimental data during swirl flow with full-length tape inserts were also obtained for three twist ratios, i.e. 14.95, 9.05 and 5.9. The insertion of a full-length tape in the condenser tube has produced an increasing effect on the heat transfer coefficients. The increase in these values was comparatively small in the experimental range of parameters, and varies from nearly 9 to 26%. Although the gain is generally maximum for lowest twist ratio of y=5.9, it is not very consistent and the trend marginally shifts towards y 9.05 at high mass velocity. The increase of heat transfer during swirl flow is a complex function of mass velocity and twist ratio. The effect is more pronounced for the low mass flux than high mass flux. The increase of heat transfer is about 26% at 209.30 kg/s-m2 where as these values drop down to nearly 9% at 372.03 kg/s-m2. The heat transfer data were compared with the known correlations for swirl flow with twisted tape insert i.e. of Royal and Bergles [52], hula and Bergles [39] and Ramakrishna and Azer [55]. The experimental data did not agree satisfactorily with any of these correlations, though the best agreement was found with Royal and heroics [52] correlations where the deviation were in between -20%to 050% for all twist ratios. Generalized correlation has been developed for the swirl flow condensing heat transfer coefficients based on the modification of the Akers-Ronson [21 plain flow correlation by inclusion of a factor accounting for tangential velocity effects. The correlations developed are as follows: Nu, 39.79 (pri,1/3 ()lp_w11/6 (Os )0.1027 2000 < Revs < 25000 Nu = 0.645 (Prg)1/35_)l/6 (Revs)0.507 eT. Cpl 25000 < Revs < 70,000 Attempts were then made to develop correlations with different combination of the factors accounting for the enhancement in heat transfer rate. One more set of correlation was developed as follows )1/6 (Reyvs)0.0905(y)-0.068]Nu = 49.881. (Pr,)1/3 x Cpl ...(3) 2000 0 Reyvs < 25000 Nu = 0.440 (001)1/3 (lIg___)1/6 (Rsy,„)0.545(,)-0.01668 aT.CP, 25000 < Reyvs < 70,000 Heat transfer data were also obtained for R22 vapour condensing inside the same horizontal tube with segmented tape inserts in 1/4, 1/2 and 3/4 of the full-length of the test- sections i.e. in first one, two or three tubes, respectively. The insertion of tape in a section of the condenser has also been found to increase the average heat transfer coefficient in comparison to plain flow values although the increase is relatively small. The increase is again found to be a complex faction of mass-velocity and also of segmental tape length in this case too. The effect produced by the insertion of segmented tapes are more pronounced and effective at low mass flux. The variation of heat transfer coefficients are found to be more random at high mass flux than at low mass flux. In general, the heat transfer coefficients are found to be more when the length of the tape insert is more. It is the highest for full length tape insert. However, in a few cases the increase has been found to be negative. With the highly unstable nature of heat transfer in refrigerant condenser, the reverse trend can be simply due to large fluctuations caused by this unstable nature. No correlations have been possible with the segmented tape inserts.