ASHRAE IJHVAC 7-2-2001 International Journal of Heating Ventilating Air-Conditioning and Refrigerating Research《供暖 通风 空调和制冷研究的国际期刊 第7卷第2号 2001年4月》.pdf

《ASHRAE IJHVAC 7-2-2001 International Journal of Heating Ventilating Air-Conditioning and Refrigerating Research《供暖 通风 空调和制冷研究的国际期刊 第7卷第2号 2001年4月》.pdf》由会员分享，可在线阅读，更多相关《ASHRAE IJHVAC 7-2-2001 International Journal of Heating Ventilating Air-Conditioning and Refrigerating Research《供暖 通风 空调和制冷研究的国际期刊 第7卷第2号 2001年4月》.pdf（124页珍藏版）》请在麦多课文档分享上搜索。

1、 AUTHOR INSTRUCTIONS Manuscripts submitted to the International Journal of Heating. Ventilating. Air-conditioning and Refrigerating Research should repre- sent original work of lasting significance. They should be submitted exclusively toHVAC however, only a few investigators have reported this para

2、meter. The fin pitch pt (= nDiln), shown in Figure 2, is often reported instead of the number of fins, n. When tubes are installed in heat exchangers, mandrels are used to expand the tubes into fin plate holes. Almost all investiga- tions have used unexpanded tubes. The effect of mandrel deformation

3、 on the microfin surface is unknown. Tatsumi and Ooizumi (1985) describe a number of the early microfin tube manufacturing techniques. As discussed by Chiang (1993), microfin popularity has grown because of the sig- nificant enhancement of heat transfer, typically equal to or greater than the enhanc

4、ed to smooth tube surface area ratio (It0 and Kimura 1979 and Kimura and Ito 1981). Enhanced to smooth tube surface area ratios generally range from 1.5 to 1.7. The enhancement effect is achieved with relatively small increases in material (16 to 20%) and relatively minor modifications to manu- fact

5、uring processes. An additional benefit of the microfin tube is that the pressure drops are gen- erally higher by 20% or less compared to smooth tube pressure drops. Fin height requires further definition due to the mechanisms that may be important for enhancement effects. For example, when fins are

6、able to protrude through the liquid film layer, VOLUME 7, NUMBER 2, APRIL 2001 127 Figure 2. Schematic of the microfin geometry 2.5 i I I I I E E 2.0 1.5 1 .o O o O 8 O i 0.0 0.2 0.4 0.6 0.8 1 .o QUALITY Figure 3. Film thickness distribution prediction of R-22/R-114 (76.5%/23.5% by mole fraction) at

7、 500 kg/m2.s and saturated liquid pressure 435 kPa in a 9.1 mm diameter tube surface tension effects between the fins and the liquid phase may have important wetting charac- teristics (Shah et al. 1999). When fins are below the liquid phase surface, they are buried within the liquid phase where surf

8、ace tension effects lose significance. In general, refrigeration tubing is oriented horizontally in order to ensure return of lubricating oil to the compressor. In horizon- tal tubes, both conditions may exist with a relatively deep liquid layer at the tube bottom and a relatively thin liquid film l

9、ayer at the top of the tube. Figure 3 shows a prediction of the film thickness distribution around the circumference of a horizontal tube with a mixture of R-22 and R-114 over a range of qualities at a mass flux of 500 kg/m2.s (Hurlburt and Newel1 1997). The liquid film thickness decreases significa

10、ntly as one moves from the bottom toward the top of the tube. Various factors affect the film profile. Higher mass fluxes, lower vapor densities, and smaller tube diameters tend to make the film thickness more uniform around the tube circumference for both condensation and vaporization. The vapor 12

11、8 HVAC&R RESEARCH density depends upon the refrigerant used in specific applications. A high-pressure refrigerant replacing a lower pressure refrigerant will tend to have a less uniform film at similar mass flux and quality conditions because its vapor density will tend to be higher. In terms of mic

12、rofin per- formance, the higher pressure refrigerant may have a much thinner liquid film in the top region of a tube with surface tension playing an important role in increasing heat transfer, while the lower pressure refrigerant may have a much more uniform film that is thick enough to circum- fere

13、ntially cover the fins, thus reducing heat transfer. Even with the same refrigerant, condenser and evaporator conditions are significantly different with condensers generally operating close to the annular-stratified flow boundary and evaporators operating mostly within the annular flow region. The

14、kinetic energy of the relatively low-density vapor in an evaporator is large enough to keep the liquid phase spread around the tube perimeter. The high-density vapor in a condenser has a lower velocity that is unable to keep liquid distributed around the tube perimeter. These differences are suffici

15、ent to cause important differences in the basic flow field configuration, which may be significant in terms of the effectiveness of microfin performance. A means of characterizing microfins in two-phase flow would be defining the fin in terms of an average liquid film thickness. The height of fins w

16、ith zero thickness would be the same as the average liquid film thickness in a two-phase flow when: e1 0.5 Di 2 - - -(1-a ) where a is the void fraction. Fins with finite thickness displace volume, requiring less liquid (higher void fraction) for flooding the groove regions. Equation (1) also assume

17、s no liquid phase entrainnient in the vapor. Liquid entrainment is generally small for vapor velocities less than 30 m/s. Void fraction prediction is a subject area that has received little attention in the refrigeration field. The results by Graham et al. (1998), Kopke et al. (1998), Wilson et al.

18、(1998), and Yashar et al. (1998) have helped characterize refrigerant void fraction effects for both smooth and microfin tubes. More details will be discussed in a later section on void fraction. Generally, the void fraction is less than about 0.9 in refrigeration tubing at typical mass fluxes. In r

19、eality, most of the condensation and evaporation process occurs at void fractions where there is more than enough liquid to cover the fins if the liquid is evenly distributed around the tube. Mass fluxes range from 25 kg/m2.s to 700 kg/m2.s with domestic refrigerators at the low end and residen- tia

20、l/commercial air-conditioning systems at the higher end. Fin heights greater than the average film thickness will have fin surfaces that interact directly with both liquid and vapor phases, while fins with heights less than the average liquid film will tend to have effects that are domi- nated by in

21、teraction with the liquid phase only. Smooth Tube Trends The general behavior of low, medium, and high pressure refrigerants on void fraction, pres- sure drop, and heat transfer trends is presented for internally smooth (unfinned) horizontal tubes (Hurlburt and Newel1 1997). Table 1 shows a variety

22、of vapor and liquid properties for represen- tative refrigerants in each of these categories. In smooth tubes, the primary factor causing pressure drop is the interaction between the liq- uid and vapor phases at the liquid-vapor interface. In contrast, the wall shear effect between the fluid and the

23、 wall in a single-phase flow is the primary factor causing pressure drop. Vapor momentum dissipated at this interface is transmitted through the liquid layer to the tube wall. In contrast, heat transfer is controlled by the relatively thin liquid viscous sublayer and buffer regions adjacent to the t

24、ube wall. A mechanistic picture of the liquid film assumes the VOLUME 7, NUMBER 2. APRIL 2001 129 Table 1. Comparison of Saturation Properties for R-11, R-12, R-22, R-123, R-l34a, and R410A to Air-Water T, Pv PI pV PI pVx1O7 p1x107 kv kl CP,. CPJ “C kPa kPa kg/m3 kg/m3 Paas Pa-s W/m*K W/m.K kJ/kg*K

25、kJ/kg.K Prl R-11 O 40 40 2.5 1532 99 5338 0.007 0.095 0.55 0.85 4.8 40 175 40 959 40 1538 40 154 R-134a O 292 40 1018 R-12 O 308 R-22 O 498 R-123 O 33 R-410A O 796 175 308 959 498 538 33 154 292 018 798 9.7 1440 114 3504 17.9 1397 114 2570 55.9 1254 136 1737 21.1 1279 117 2205 65.7 1127 141 1424 2.2

26、 1522 96 5920 9.6 1423 111 3623 14.2 1294 109 2857 49.1 1144 131 1775 29.4 1195 119 1730 0.009 0.009 0.01 1 0.010 0.013 0.008 0.010 0.0 12 0.016 0.012 0.082 0.078 0.061 0.103 0.079 0.086 0.074 0.094 0.074 0.112 0.60 0.89 3.8 0.61 0.93 3.1 0.73 1.02 2.9 0.68 1.15 2.5 0.89 1.31 2.4 0.64 0.97 6.7 0.71

27、1.03 5.0 0.83 1.32 4.0 1.01 1.49 3.6 0.88 1.43 2.2 40 2403 2411 94.9 1003 151 1005 0.014 0.078 1.29 1.74 2.2 Air 20 101 1.2 180 0.026 1 .o Water 20 101 1001 1 O000 0.60 4.2 7.0 Sources: Refrigerant data are taken from Gallagher et al. (1993), and ir and water data are taken from Incropera and DeWitt

28、 (1996). Table 2. Percentage of Thermal Resistance, R, in the Three Liquid Film Layers for Condensation of R-134a in a 7.75 mm Diameter Tube Quality R, tv%b tb%b tl%b 0.2 51.4 39.3 9.3 3.9 19.7 76.4 0.3 51.8 39.6 8.6 4.5 22.3 73.2 0.4 52.3 40.0 7.8 5.2 25.8 69.0 0.5 52.8 40.4 6.8 6.0 30.0 64.0 0.6 5

29、3.5 40.9 5.7 7.1 35.7 57.2 0.7 54.3 41.5 4.2 9.1 45.5 45.4 0.8 55.5 42.4 2.2 12.2 61 .O 26.8 0.9 58.3 41.7 0.0 20.0 80.0 0.0 flux of 300 kg/m2.s bpercentage dimensionless thicknesses, t, of the liquid films viscous, buffer region and turbulent layers flow to be very similar to “law of the wall” turb

30、ulent flows in single-phase pipe flow (Traviss et al. 1973). Viscous sublayer, buffer region, and turbulent region areas are assumed to exist in the liquid layer. As the quality is increased, at a given mass flux, the vapor velocity increased and that increased the vapor-liquid interaction. Pressure

31、 drop increases while liquid film thickness decreases. Decreasing film thickness results in an increase in both void fraction and heat transfer. From Figure 3, the average liquid layer is quite thin in refrigerant flows. Table 2 shows the relative percentages of the sublayer, buffer region, and turb

32、ulent region thickness. It also shows the relative percentages of each layers contribution to the thermal resistance in condensation heat transfer for R-134a at a mass flux of 300 kg/m2.s in a 7.75 mm diameter tube at a saturation temperature of 5C (Hurlburt and Newel1 1997). Figure 4 is a schematic

33、 of pressure gradient trends for low, medium, and high-pressure refrig- erants in a 10 mm diameter, smooth horizontal tube at a mass flux of 250 kg/m2.s at condenser sat- uration temperature conditions. The trends are taken from data collected by Sacks, (1975) and predictions from the refrigerant pr

34、essure drop correlation by Souza (1995). At a given mass flux 130 HVACLRRESEARCH and quality, low pressure refrigerants have higher vapor velocities, causing higher relative pres- sure drops than those for higher pressure refrigerants. Medium and high pressure refrigerants are more similar in pressu

35、re drop. At similar mass fluxes and qualities, evaporators will have rela- tively higher pressure drops than condensers due to lower saturation pressures as a result of lower operating temperatures. Properties shown in Table 1 provide an indication of the vapor density range for two temperature leve

36、ls. Evaporators generally have lower mass fluxes than condensers in order to obtain reasonable pressure drops. Also, evaporators and condensers have opposing pressure change effects due to deceleration and acceleration of the refrigerant changing phase. Figure 5 shows a schematic of condensation hea

37、t transfer coefficient trends for low, medium, and high pressure refrigerants. The ranges are drawn based on a number of heat transfer correlations (Hurlburt and Newell 1997, Traviss et al. 1973, Cavallini and Zecchin 1974, Shah 1979, and Chen et al. 1987). As discussed by Hurlburt and Newell (1997)

38、, predic- tions by Traviss et al. (1973) and Cavallini and Zecchin (1974) tend to be on the high side of the bands shown. Predictions by Chen (1987) tend to be at the low end of the bands, and the correlation by Shah (1979) is generally in the middle of the shaded bands. The difference between mediu

39、m and high pressure refrigerants is not significant enough to show a general difference between the two categories. Other property differences such as thermal conductiv- ity, specific heat, and viscosity that affect the thermal resistance structure of the viscous sub- layer and buffer regions are si

40、gnificant enough to blur the differences in this range of refrigerant effects. Heat transfer effects in an evaporator are more difficult to categorize than condenser effects due to two opposing effects. Nucleate boiling will tend to enhance heat transfer above the heat transfer coefficient levels sh

41、own in Figure 5. Generally, this enhancement occurs during the low quality and low mass flux conditions when vapor velocities are lower. Tube wall dryout in an evaporator tube results in a degradation of heat transfer. The top portion of the tube wall may dry out under low quality and low mass flux

42、conditions as well because vapor velocities may be insufficient for wetting the top of a tube. As shown in Figure 3, a fairly broad circumferential region has a very thin film where dryout regions may form. The combination of nucleate boiling 10 E . $5 a 2 O MID-PRESSURE (R-12. R-134a) (R-22, R-410A

43、) O 0.2 0.4 0.6 0.8 1 O QUALITY Figure 4. Schematic of pressure drop variation for low, medium, and high pressure refrigerants in a 10 mm diameter tube at a mass flux of 250 kg/m2.s and a temperature of 30C 8 6 s E . B r 2 o LOW PRESSURE (R-11, R-123) E O 0.2 0.4 0.6 0.8 1.0 QUALITY Figure 5. Condensation heat transfer coefficients for low, medium, and high pressure refrigerants in a 10 mm diameter tube at a mass flux of 250 kg/m2.s and a temperature of 30C