1、4682 Heat Transfer Augmentation by Segmented Tape Inserts During Condensation of R-22 Inside a Horizontal Tube Kailash Nath Agrawal, Ph.D. Ravi Kumar, Ph.D. Sachida Nand Lal, Ph.D. ABSTRACT The augmentation in heat transfer by twisted tape inserts has been studied for forced convection condensation
2、of R-22 vapor inside a tube. The inserts areput in the entire length and in 1/4,1/2, and 3/4 lengths ofthe test condenser: Thepitch ratio, y, for the full-length tape and the segmented tapes is 5.9. The refrigerantjlow rate, G, has been taken as 21 O, 237,282, 327, and 372 kg/s-m2. The enhancement i
3、n heat transfer during condensation 0fR-22 has been attained up to 25% for the full- length twisted tape insert. However, in a certain range of refrigerant jlow rates, the 1/2 segmented tape insert has outperformed the full-length tape. It is a general observation that all the twisted tape inserts a
4、re more effective at low refrig- erant flow rates; however, the full-length twisted tape insert remains efective at the higher refrigerant jlow rate as well. INTRODUCTION AND BACKGROUND The theory of heat transfer during condensation of vapor is quite old and is supposed to have been first given by
5、Nusselt (1 9 16). Since then, considerable research has been carried out in this area and a few notable reports have come out in the last few decades (Shah 1979; Dobson and Chato 1998; Agrawal et al. 2002; Cavallini et al. 2003). However, these investigations were carried out for plain flow condensa
6、tion. In earlier years, many investigators (Kirov 1949; Bergles 1973) mentioned a few convective heat transfer augmentation techniques with the help of insertion devices, viz., twisted tapes, coiled wires, baffles, etc. The purpose of these devices was to create turbu- lence in the flow field. It wa
7、s found that these techniques were quite effective in enhancing the heat transfer rate in single- phase heat transfer. These devices were, therefore, termed “turbulence promoters.” Considering the usefulness of turbu- .iari Krishna Varma, Ph.D. lence promoters in single-phase flow, their performance
8、 has been tested in two-phase systems for the last several years (Luu and Bergals 1980; Kaushik and Azar 1988; Schlager et al. 1990) such as in refngerant condensers. It has been found that the turbulence promoters are good heat transfer augmen- tation devices for refrigerant condensation. However,
9、the improvement in heat transfer depends upon the range of exper- imental parameters. Authors have been conducting experimental research to evaluate the performance of full-length twisted tapes for the last a few years (La1 1993; Agrawal et al. 1998; Behabadi et al. 2000). One difficulty that was ge
10、nerally encountered during the experimentation was the fitting of these turbulence promoters in the entire length of the test condenser. Therefore, an experimental investigation was planned to study the enhancement in heat transfer with the twisted tape inserts only in a part ofthe test condenser, k
11、nown as segmented tape insert. EXPERIMENTAL PROGRAM The schematic layout of the experimental facility used in the present investigation is shown in Figure 1. The experimen- tal setup is a well-instrumented, 5 ton vapor compression refrigeration unit driven by an open type R-22 compressor. The unit i
12、s designed to yield a wide range of experimental parameters (see Table 1). The main constituents of the test faciliy are the test condenser, pre-condenser, after-condenser, evaporator, a water-cooled condenser, instrumentation, and the accessories. Experimental Setup Figure 2 shows the sectional arr
13、angement of one test section. The test section is made of hard drawn copper tube Kailash Nath Agrawal is a professor and Ravi Kumar is an assistant professor in the Mechanical and Industrial Engineering Department, Indian Institute of Technology Roorkee, Roorkee, India. Sachida Nand La1 is a profess
14、or at Bhagalpur Engineering College, Bhagalpur, India. Hari Krishna Varma is director of Ideal Institute of Technology, Ghaziabad, India. 02004 ASHRAE. 1 43 e I“ “ Y “,II“ ,* “III%. “ “ YIIIir “ I ,I I II I - - - 4- I coolant water temperature condensing temperature of R-22, T, I 20 to 30C 37.5- 53.
15、8“C Figure 1 Schematic diagram of experimental setup. vapor quality range, X average cooling heat flux, q with 12.7 mrn inside diameter, di, 15.8 mm outside diameter, do, and length of 950 mm. This tube is located concentrically inside a copper tube of 50 mm inside diameter; thus, the two concentric
16、 tubes form a counterfiow annulus. Four such test sections were installed in series to form the complete test condenser. In the test condenser, the refrigerant vapor flowed inside the inner tube, whereas the cooling water flowed in the annular space. The outer wall temperatures of the inner tube wer
17、e measured at four axial locations in each test section. At each of these four locations, the copper-constantan thermo- couples were fixed on the top, side, and bottom. Thus, alto- gether, on each test section tube, there were 12 thermocouples, and a total of 48 thermocouples were fixed on the whole
18、 of the test condenser. The entire test condenser was completely insu- lated with glass wool to prevent heat loss. An analysis was carried out to quantify the heat loss to the surroundings from the test section. It was found that the heat loss was up to 1 .O% of the heat transfer during condensation
19、 inside the test section. Each test section of the test condenser was also instru- mented to measure the inlet and outlet temperatures of the cooling water and also the inlet and outlet temperatures of refrigerant R-22. Pressure taps were provided to measure the pressure at inlet and outlet of each
20、test section and also for the whole test condenser. The operating parameters of the present investigation are given in Table 1. 1.0- 0.10 21 to 4 kW/m2 Twisted Tapes The twisted tapes were used for creating swirl in the fluid flow. These tapes were made from 0.5 mm thick stainless steel flats. The w
21、idth of the strips was cut approximately 5 percent greater than the inside tube diameter of the test section to allow for contraction in the twisting. The edges of the strips were gently filed to remove any abrasions and make it smooth. A full-length tape fitted inside the tube is shown in Figure 2.
22、 The 114 length segmental tape was fitted only in section A; the 112 length segmental tape was fitted in A and 50 IZ 7 mi Cnst200+ 2oo200-cJy17s seaion at x-x X win Figure 2 Details of test section. Table 1. Range of Operating Parameters I working fluid 1 R-22 refrigerant mass velocity, G 210,237,28
23、2,327, and 372, cooling water flow rates, m, 200- 1200 kdh Icondensing pressure of R-22. P I 14.4 to 21.9 bar segmented tape length V4, 112, and 314 length of full length tube I twist ratio for taDes, y 15.9 B; and the 314 length tape was fitted in A, B, and C (see Figure 1). It has been observed th
24、at, in general, the twisted tape with the twist ratioy of 5.9 gives the greatest heat transfer improve- ment (Agrawal et al. 1998; Behabadi et al. 2000). Hence, this twist ratio was adopted for the segmental inserts of different lengths. The heat transfer data were acquired for the conden- sation of
25、 R-22 vapor inside the same horizontal tube under the swirl flow condition with segmented tape inserts in 114, 112, and 314 lengths of the test condenser, i.e., up to the first test section (A), up to the second test section (B), and up to the third test section (C), respectively. This employed inse
26、rtions of segments in three lengths, Le., 1 O00 mm, 2 1 O0 mm, and 3200 mm inside the test condenser. The refrigerant flow rates, G, were the same as were used in the swirl flow tests with full- length tapes and the plain flow. Calculation Procedure 1. The outside average tube wall temperature of th
27、e test section at one station was calculated by Equation 1 : 144 ASH RAE Transactions: Research 2. The average temperature of the outside wall of one test section was calculated from the average of four axial stations, Equation 2: 4 CTws 1 Two = 4 3. The radial heat flux, q, for a test section was e
28、valuated by dividing the heat flow out of that test section, Q, by its outer wall surface area, A,. The heat flow out of a test section, Q, was the heat carried away by the cooling water. Hence, it was calculated by taking the product of mass flow rate of the cooling water flowing in annular space,
29、specific heat of water, and the temperature rise of cooling water while pass- ing through the test section. Since and A, = xd,L heat flux is, therefore, 4. The radial temperature drop across the test section tube wall, ATw was evaluated by the Equation 4. (4) 5. The average inside tube wall temperat
30、ure, Tm, was evalu- ated by adding the temperature drop across the wall to the measured average outside wall temperature already evalu- ated by Equation 2. Twi = Tw,+ATw (5) 6. The average static pressure in the test section was deter- mined by taking the mean of inlet and exit pressure for each tes
31、t section. The average heat transfer coefficient, h, was computed with the help of the radial heat flux of the tube wall, q, the aver- age tube inside wall temperature, Tm, and the average temperature of condensing vapor, Ts, using Equation 6. 7. 8. As shown in Figure 1, a pre-condenser was installe
32、d ahead of the test condenser. By regulating a predetermined mass flow rate of water in this condenser, a desired vapor qual- ity could be achieved at the test condenser inlet. The data were collected for a vapor quality range of approximately 8000 5000 5 3000 5 2000 $1500 t ? 500 E .- o 1000 8 800
33、w m 2 300 200 150 1 O0 nass flow me, G, kS/s-m* 1.2101 O 2500 7 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 vapor quality, X Figure 3 Comparison ofheat transfer coeficient for aplain pow at diferent refrigerant mass pow rates. 5 to 10 K superheat at the inlet of the test condenser to about O. 10
34、 at the outlet of the test condenser. An uncertainty analysis of the experimental results was also carried out. The accuracy in the measurement of the test section tube diameter was O. 1 mm and that in the measurement of the test section length tube was 1 mm. The test section tube wall temperature a
35、nd cooling water temperature were measured with an accuracy of O. 1C and O.O5“C, respectively. The cooling water mass flow rate was measured within an error band of I2%. The vapor pressure of R-22 was measured with an accuracy of 0.5 psi, which yielded an uncertainty in the vapor saturation temperat
36、ure of 0.2“C. The uncertainty in the determination of heat transfer coefficient, h, was calculated and found to be within 7% (La1 1993). RESULTS AND DISCUSSION In order to study the relative performance of segmented tape inserts, experimental data were also collected for plain flow and for twisted t
37、ape inserts for the entire length of the test condenser. The variation of heat transfer coefficient in these cases is discussed in the following sections. Condensation During Plain Flow Figure 3 is drawn between the condensing side heat trans- fer coefficient, h, and the vapor quality, X, of R-22, d
38、uring condensation of vapor inside the test condenser tube for plain flow (no inserts). During condensation of vapor inside the horizontal tubes, for the entire refrigerant flow rates, the vapor quality is nearly unity at the entrance of the test condenser and it approaches O. 15 at the exit of the
39、test condenser. It is apparent from Figure 3 that the heat transfer coeffi- cient, h, and the vapor quality, X, decrease down the test ASHRAE Transactions: Research 145 8000 5000 y 3000 s =- 2000 .D f 1000 F- 1500 8 800 t 5 500 2 300 200 150 1 O0 O 114 1R W4 1 segment oi tw$ted tape inam 1“I“l“ y, ,
40、 ., ),I, , I, , 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 vapor quality, X Figure 4 Cornpurison of heut transfer coe8cient for segmented tape inserts, G=21 O kgs-m2. section tubes. In fact, during condensation of vapor in a hori- zontal tube, the dryness fraction of condensing vapor reduces in
41、 the downstream direction. This causes a thick layer of condensate on the inside tube surface of the test section, which offers greater resistance in heat transfer from refrigerant to the cooling water flowing around the test section tube. This results in a lower value of heat transfer coefficient n
42、ear the exit of the test section. The increase in refrigerant mass flow rate, G, enhances the Reynolds number of refrigerant flowing inside the test condenser, causing higher value of heat transfer coef- ficient, h. As shown in Figure 3, the average value of the heat transfer coefficient, h, is lowe
43、st, 1233 W/m2-K, for the lowest refrigerant flow rate of 210 kgls-m2, and it has the highest value, 21 15 Wlm2-K, for the refrigerant flow rate of 372 kg/ s-m2. The increase in the average heat transfer coefficient, h, is marginal for the step increase of flow rate from 210 kg/s-m2 to 237 kg/s-m2. A
44、fter that, the increase in average heat transfer coefficient was nearly uniform. The results of the present investigation for the condensa- tion of R-22, inside the tube, for a plain flow, are compared with the predictions of other investigators. The experimental data are in agreement with the Akers
45、 Rosson correlation (1960) in an error band of +40% to -30%. The correlation developed by Tandon et al. (1985) predicts the experimental data in a range of 35% to 40%. For two-phase flow conden- sation studies, the deviation of the above order is quite accepted to establish the integrity of an exper
46、imental setup (McAdams 1949). The following correlation, Equation 7, has been developed between the heat transfer coefficient, h, mass flow rate, G, and dryness fraction, X, of the condensing vapor. This correlation predicts 75% of experimental data in an error band of f20%. (7) Condensation with Fu
47、ll-length Tape Inserts Heat transfer data were also obtained for the condensation of R-22 vapor with full-length tape inserts over the same range of operating parameters (La1 1993; Behabadi et al. 2000). The condensation heat transfer coefficient, h, was determined for the tubes fitted with full-len
48、gth twisted tapes and inserts of different twist ratios, y. The twisted tape with the twist ratio, y, of 5.9 enhances the condensing heat transfer coefficient, h, up to 25% in comparison to that forplain flow tube (La1 1993). Condensation with Segmented Tape Inserts Figure 4 is drawn to show the var
49、iation of condensing side heat transfer coefficient, h, with the vapor quality for different segmented tape insert lengths. The refrigerant mass flow rate has remained constant at 2 1 O kgls-m2 for all the test runs in this figure. The results are scattered because the condensation of refrigerants inside the condenser tube is a highly unstablized phenomenon. The two phases flow together inside the condenser tube at the same time the phase transformation is also taking place. It is observed that the heat transfer coefficient decreases as the condensation proceeds, Le
