1、AN-04-1 -1 Experimental Study of Supercritical CO2 Gas Cooling in a Microchannel Gas Cooler Yuan Zhao, Ph.D. ABSTRACT An experimental study was conducted to investigate the heat transfer characteristics of supercritical CO, gas cooling down in a microchannel gas cooler over a range of operating cond
2、itions encountered in typical residential heatpumps. The microchannels used in the present study had a hydraulic diam- eter of approximately 1 mm. The experiments were conducted to evaluate the heat transferperformance of the microchannel gas cooler at different test conditions by varying airjlow ra
3、tes, air temperatures, refrigerant inlet temperatures, and mass jlow rates. All experimental results are tabulated in thepresent paper: It was,found that the refrigerant masspow rate is the dominant factor for the capacity of a CO, gas coolel; and a signijkantportion of the heat transfer in a CO, ga
4、s cooler was carried out in the heat exchanger module on the refrigerant inlet side. The temperature and pressure of CO2 signgcantly affect the heat transfer andjluidjlow characteristics due to the fact that some important thermal physical properties of CO, (such as spec$c heat, density, viscosity)
5、are strongly depen- dent on its temperature and pressure. All experiments were successfully conducted with an energy balance of +3%. INTRODUCTION Unlike commonly used refngerants that operate solely in the subcritical region, CO, offers more variation in thermody- namic and transport properties due
6、to the fact that CO, systems typically operate in a region that extends to the supercritical area. Since the critical temperature of CO, is 3 1 OC, the heat rejection above 31C is not by condensation, as in conven- tional systems, but by gas cooling. This difference will require some changes in heat
7、 exchanger design. Michael M. Ohadi, Ph.D. Fellow ASHRAE Zhao et al. (2000a, 2000b, 200 1,2003) presented a series of studies on the heat transfer and fluid flow characteristics of CO,. They indicated that CO, exhibited unique characteristics on heat transfer and pressure drop due to its outstanding
8、 ther- mophysical properties, such as much lower viscosity, lower surface tension, large vapor density, and smaller density ratio between liquid and vapor phases. However, as also indicated there, one of the main difficulties in applying CO, as a refrig- erant was the significantly higher system ope
9、rating pressure, especially in gas coolers (up to 12 MPa). Conventional diam- eter tubes cannot tolerate pressure this high. The advent of microchannel tubes has overcome this problem and facilitated the fabrication of CO, heat exchangers. Very lightweight, microchannel tubes can tolerate very high
10、pressures. For instance, a microchannel with a hydraulic diameter of 0.8 mm and a wall thickness of 0.3 mm can easily withstand an oper- ating pressure of 14 MPa. Baldantoni (200 1) showed the feasi- bility of manufacturing heat exchangers with extruded aluminum tubes (microchannels) for high-pressu
11、re HVAC systems. He found that in order to allow for high operating pressures, heat exchangers for CO, could be designed using extruded sections with circular channels. This solution successfully minimizes both wall thickness and weight. Another advantage of microchannels is their very large contact
12、 surface area with the fluid per unit volume, which means they can provide outstanding heat transfer perfor- mance. Owing to their unique heat transfer performance and pressure tolerance, they are now used routinely in most auto- motive condensers and have recently become the subject of study for us
13、e as automotive evaporators. The advantages of CO, microchannel heat exchangers are due, not only to the high performance of microchannel heat Yuan Zhao is an assistant research scientist at Advanced Thermal Environmental Concepts Inc., College Park, Md. Michael Ohadi is a profes- sor in the Mechani
14、cal Engineering Department, University of Maryland, College Park, Md. 02004 ASHRAE. 291 Table 1. Thermophysical Properties of CO, and R-134a at 10C Refrigerant CO2 R-134a psat (MPa) 4.502 0.414 Latent Heat (!clkg) 196.8 190.9 Surface Tension mN/m) 2.67 10.3 Liq. Density (kg/m3) 861.5 1260.2 Vap. Den
15、sity (kg/m3) 135.3 20.2 Lia. Viscositv (upas) 86.7 254.3 Vap. Viscosity (pPas) 16.1 11.4 Lia C, (kJkn.K) 3.01 1.37 Gas Coolu Incicor Chamber Compressor Vap. Cp (kJkg.K) 2.62 0.93 transfer and the environmentally friendly nature of CO, but also to the fact that microchannels and CO2 can offset the Fi
16、gure I Schematic of refrigerant loop. weaknesses of each other, as indicated by Zhao et al. (2003). One of the main weaknesses of microchannels is the tremen- dous flow resistance. Fortunately, CO, has very low viscosity and high vapor phase density, as shown in Table 1. Lower viscosity and higher v
17、apor density corresponds to a lower pressure drop as refrigerant flows through the exchangers. It suggests that the mass flow rate of CO, in microchannel heat exchangers can be designed to be much larger. On the other hand, microchannels are also suitable for high operating pres- sures, which is one
18、 of the main disadvantages of CO,. value), which are typically around 40C for their test condi- tions. Notwithstanding the available information in the litera- ture, a clear understanding of the performance and potential of CO2 microchannel gas coolers is still lacking. Therefore, the objective of t
19、he present work was to characterize the perfor- mance of the latest generation of CO, gas coolers over a selected range of operating parameters of interest to refriger- ationheat pump applications. TEST FACILITY AND APPARATUS Research on microchannel heat exchangers for CO, is relatively new, and th
20、e available information is limited. Pettersen et al. (1 998, 2000) developed a microchannel heat exchanger for CO, and experimentally evaluated the overall heat transfer coefficient. They indicated that refrigerant-side heat transfer coefficients are higher than those of fluorocar- bons and, therefo
21、re, the internal surface areas of heat exchang- ers could be reduced. Smaller tube and manifold dimensions reduce the heat exchanger size compared to those using R- 134a. The temperature difference between the inlet air and the outlet refrigerant is much lower in CO, gas coolers than in baseline HFC
22、 and/or HCFC system condensers of equal size and capacity. The reduced refrigerant exit temperature had a noticeable influence on the coefficient of performance, and it appeared that the microchannel heat exchanger had the best overall heat transfer coefficient. Pitla et al. (2000) numerically analy
23、zed heat exchangers for transcritical CO, systems. They suggested that experimen- tal results were hard to predict when the operating conditions were close to the critical point. Kuang et al. (2003) presented a systematic experimental study of supercritical CO, gas cooling in microchannels. The expe
24、rimental results showed unique heat transfer characteris- tics of supercritical CO, in microchannels. Heat transfer coef- ficients reached maximum values at pseudo-critical temperatures (when the specific heat Cp reaches the maximum The test loop (as shown in Figure 1) used in this study, which incl
25、uded an evaporator and a condenserlgas cooler, was designed to measure the capacity of microchannel heat exchangers. The gas cooler was the main test section compo- nent of the present experiments, in which a supercritical CO, gas released heat to an external flow of air. CO, was driven by a compres
26、sor to the gas cooler and returns to the evaporator after it is expanded in an expansion valve, completing the cycle. During the operation, the gas cooler and evaporator are separated from each other through the use of different air ducts in separate rooms, thus allowing for independent fine control
27、 of the inlet airstream conditions (including temperature and relative humidity) for each heat exchanger. Outdoor Duct The outdoor duct, shown in Figure 2, was built inside an environmental chamber. The outdoor duct houses the micro- channel gas cooler. It is constructed from polypropylene, The duct
28、 contains screens, upstream and downstream thermocou- ple grids, a gas cooler, an air mixer, an obstruction meter, and a large fan. Screens are used to make the airstream uniform, while thermocouple grids measure the bulk temperatures of the airstream. The fan, controlled by a variable-speed motor,
29、was placed at the outlet of the duct where it draws air through 292 ASH RAE Transactions: Symposia 1.0 mm 7 I I . I - :- l i I- 9 ,x S T OC AM OM ST F AM -Air Mixer F-Fan S - Screen GC -Gas Cooler OM - Obstruction Meter T Thennocoupie Grid (3 x 3) I I 16.0 mm - (a) Cross-section of a microchannel tu
30、be. Figure 2 Schematic of outdoor duct. 6 gHeat = 2.2 kW OHeat = 4.2 kW (i) nHeat = 4.2 kW (li) 2 O 60 b O CI U - $00 ARTItest range I -6 -1 I Air Flow Rate (m3ih) Figure 3 Energy balance for the obstruction jlow meter: the duct. The air duct is insulated with 25 mm thickness of thermal insulation m
31、aterial (k = 0.04 W/mK). An obstruction flow meter was designed and fabricated to measure the airflow rates inside the duct. The flow meter was made of 117 circular holes with a diameter of 25 mm. Since these small holes were uniformly deployed across almost the entire cross-sectional area of the te
32、st duct, the flow and ther- mal fields were relatively uniform for the present situation. Fin strip heaters were used to calibrate the obstruction flow meter. Figure 3 shows the energy balance for the obstruction flow meter. For the airflow rate range of interest to this project, the obstruction flo
33、w meter can measure airflow rate within +2% (for more details, see Zhao et al. 2001, Appendix I). Microchannel Gas Cooler The tested gas cooler was constructed with microchannel tubes with a hydraulic diameter of approximately 1 .O mm, as (b) Picture of fins and microchannels. CO, (c) Schematic of a
34、 microchannel unit slab. Figure 4 Microchannel heat exchanger: shown in Figure 4a. The gas cooler is made from several microchannel unit slabs. A schematic diagram of one of these unit slabs is shown in Figure 4c. It is important to note that the two halves of the heat exchange slab are noncommunica
35、tive, so that flexibility in choosing refrigerant paths may be ensured. The specifications of each unit slab are as follows: One unit slab is 348 mm long and 430 mm high. One unit has two passes of 17 parallel microchannels, and its overall airside surface area is 3 m2. Header is single tube, 348 mm
36、 long, 21.3 mm OC, 17.7 mrn ID. Stubs are made of Al 3003-0 tubing with OD of 9.5 mm and ID of 5.4 mm. ASHRAE Transactions: Symposia 293 Louvered fin density: i6 finsI25.4 mm; fin height: 8.0 Data Reduction and Energy Balance mm. Total flow cross-sectional area: 17 x 10 x (3.14 x 1 x 1) 14 = 133.5 m
37、m2. Refrigerant-side heat transfer area: 34 x 10 x (3.14 x 1) x 430 = 0.46 m2. The gas cooler, shown in Figure 5, is composed of ten microchannel unit slabs. Two slabs were set into a frame side by side and then stacked parallel to the airflow five units deep. The refhgerant from the discharge of th
38、e compressor was brought to the rear of this unit, divided by two, and then routed through the two parallel stacks of slabs until it exited at the front of the stacks where the two streams were recombined. This exit point was the point where the incoming airstream entered. This design is a counter-c
39、rossflow pattern, which improves the heat transfer performance. Eleven T-type ther- mocouples with uncertainties within fO.OSC were mounted to measure the refrigerant temperatures along the flow path, as shown in Figure 5 by the letter ? These thermocouples help to determine the heat transfer rate o
40、f each individual slab. TI 1 The inlet and outlet air-side and refrigerant-side pressures and temperatures were measured during the gas cooler tests. The measured data were used to determine the capacity and pressure drop. The air-side capacity was calculated from air enthalpy change and mass flow r
41、ate. The air enthalpy change was determined by the temperature difference multiplied by air specific heat. Air specific heat was based on dry air prop- erties. Q. air =m. air Cp (air-out- air-in) (1) The refrigerant-side capacity was determined by the mass flow rate and enthalpy difference between t
42、he inlet and outlet of the gas cooler, as shown in Equation 2. Qref = ref. (ref- in - ref-out) (2) Uncertainties in the experimental data were calculated based on the propagation of errors, described by Kline and McClintock (1953). The instrumentation accuracies and the associated uncertainties for
43、various parameters were utilized to calculate a total uncertainty within f3% in determining the capacities. Details of the uncertainty calculations can be found in Zhao et al. (2001). Energy balances were checked for all reported experi- mental data using Equation 3. EB = Qref- Qair Qref (3) Figure
44、6 depicts the energy balance results for all reported experimental results, clearly indicating that energy balances for all reported data were within I3%. EXPERIMENTAL TEST RESULTS Experimental tests for the microchannel gas cooler were conducted in an outdoor chamber. The system was allowed HX5 -6
45、-! 1 3 5 7 9 11 13 15 17 19 21 Serial No. Figure 5 Schematic and photo of the gas cooler layout. Figure 6 Energy balance for gas cooler test results. 294 ASHRAE Transactions: Symposia 10 1 6- E O 4- o 120 2 O Tair-in = 21 C 8/ Ref. Pressure: 6.9 MPa - 12.5 MPa Ref. Inlet temperature: 79 C - 120 C I
46、00 O o O Figure 7 CapaciQ vs. refrigerant massflow rate. about two hours to reach a steady-state condition for any specific test conditions. Data were taken once the steady-state conditions were achieved and remained constant for 30 minutes. Data were recorded for 40 minutes for each test condition
47、and averaged over time. In the present study, all properties related to air were based on dry air properties. Figure 7 shows the gas cooler capacity vs. mass flow rate of the refrigerant for an air inlet temperature of 21OC. The refrigerant-side inlet pressure (Pi,) was varied from 6.9 MPa to 12.5 M
48、Pa. As seen in the figure, the mass flow rate of the refrigerant has a significant effect on the capacity of the gas cooler. This is expected because the rehgerant capacity rate (.Iref. C,) is typically smaller than that of the air-side ( .Iojr. C, ) in order to lower the CO, outlet temperature and
49、thus provide larger cooling capacities and acceptable COP for CO, air-conditioning and heat pump systems. In this case, increasing refrigerant mass flow rate results in the significant augmentation of the capacity of the heat exchanger. These results also show that, on the refrigerant side, the mass flow rate plays a more important role than some other parameters, such as the inlet pressure and temperature of the refrigerant. Thus, the refrigerant mass flow rate is one of the primary factors that affects the capacity of a gas cooler. Figure 8 shows typica
