International Journal of Air-Conditioning and Refrigeration

This paper presents a review of differences and similarities of in-tube heat transfer and pressure drop between ammonia (NH₃) and carbon dioxide (CO₂) from the perspective of the design of heat exchangers for NH₃ two-stage and CO₂/NH₃ cascade refrigeration systems. The focus is on differences in thermophysical properties and thus different characteristics of heat transfer and pressure drop. A brief summary of published literatures about CO₂/NH₃ cascade refrigeration systems is provided and literature review of available correlations and developed correlations are presented for flow boiling and condensation heat transfer and pressure drop. Because of large deviation of calculated values with exiting correlations from measured results, a new correlation to predict flow condensation heat transfer coefficients was developed based on experimental results for CO₂ at -15°C. From comparison of measured and predicted values, it is shown that some correlations, previously published in open literature, can be used to calculate flow boiling heat transfer coefficients for NH₃ at -20°C, if a flow pattern can be appropriately determined for a flow condition. Also, it is presented that existing correlations can predict well the heat transfer coefficients for CO₂ flow boiling at -15 and -30°C. It is shown that some correlations can predict pressure drop relatively well for NH₃ and CO₂ two-phase flow. The NH₃ and CO₂ flow evaporation heat transfer and pressure drop characteristics at -40°C are compared with predicted values.

In this study, a numerical study has been carried out to analyze the factors of the efficiency decrease at backward flow situation in an axial fan with adjustable blades. The analysis is done with the pitch angle of 36° on the forward flow and of -26° on the backward flow. The numerical results show that the air flow rates of the pitch angle of 36° and -6° are calculated to be 285 cubic meter per min (CMM) and 212 CMM, respectively. The results are similar to the experimental results by Chang et al.¹ The results had the maximum error of 10.6% compared with the experimental results. The fan efficiency decrease is caused by the fact that the axial fan used for this study was designed for the forward flow. As the results, the pitch angle of -26° has induced the recirculation around the impeller blade, impeller cover and downstream. This recirculation caused the large decrease in total pressure coefficient. This turned out to be the main cause of the efficiency decrease.

This study presents experimental frictional pressure drop for R410A/oil mixture flow boiling in small horizontal smooth tubes with inside diameters of 4.18 mm and 2.0 mm. Experimental conditions cover nominal oil concentrations from 0 to 5%. The test results show that the presence of oil enhances two-phase frictional pressure drop about 0–120% and 0–90% at present test conditions for 4.18 mm I.D. smooth tube and 2.0 mm I.D. smooth tube, respectively, and the enhanced effect is more evident at higher vapor qualities where the local oil concentrations are higher. A new correlation to predict the local frictional pressure drop of R410A/oil mixture flow boiling inside conventional size and small smooth tubes is developed based on local properties of refrigerant–oil mixture, and the experimental data of 4.18 mm I.D. and 2.0 mm I.D. smooth tubes and that of 6.34 mm I.D. smooth tube (Hu et al., 2008) are well-correlated with the new correlation.

At the Institute of Thermal Engineering, a small-capacity ammonia/water absorption heat pumping unit with about 5 kW cooling capacity has been constructed and tested in the laboratory. For all heat exchangers, standard plate heat exchangers have been used. The heat pump has been designed to operate in a wide operating range in order to enable different temperature levels for ice production, residential cooling, and heating applications.

To identify potential for improvements, a thermodynamic analysis using both the first and the second law of thermodynamics has been carried out in order to locate irreversibilities associated with inefficient processes and to calculate the exergy loss of each component. Therefore the exergy losses of the main components have been separated into one part which is caused by the heat transfer and another part which originates from internal irreversibilities, e.g., from mixing processes. Furthermore, sensitivity calculations have been performed by varying the temperature levels of the heat sources and the heat sink in order to investigate the different component losses at different operating conditions.

The results show that the component exergy losses depend particular on the temperature level of both the heat sources and the heat sink. At low temperature lifts, the exergy losses of the absorber and generator are dominant and with increasing temperature lift, the exergy losses of the dephlegmator and the rectification column become considerable.

Capillary tubes are widely used as a refrigerant flow control device in small refrigeration systems. Since the flow behavior inside the capillary tube is complex, many physical models are necessary to predict the characteristics of the refrigerant flow in a capillary tube. In the present paper, refrigerant flow characteristics inside the capillary tube have been studied to find out recommended empirical correlations of influential parameters. A numerical capillary model is developed to predict the refrigerant characteristics. Various empirical correlations regarding single-phase friction factor, two-phase viscosity, two-phase frictional multiplier and metastable flow are examined using this numerical capillary model. Calculated results are compared with experimental data to examine the accuracy in terms of required capillary tube length and mass flow rate. Based on the comparison, recommended correlations are selected to be used for capillary flow analysis.

In this study, the performance characteristics of a bypass two-circuit refrigeration cycle designed for refrigerators were investigated in a bench scale refrigeration system. The bypass two-circuit refrigeration cycle with R600a operates in two modes of refrigerator/freezer-mode (RF-mode) and freezer only-mode (F only-mode) by switching refrigerant flow path. The performance of the bypass two-circuit refrigeration cycle was measured by varying the condenser capacity, the length of the suction line heat exchanger (SLHX), and the refrigerant charge amount with adoption of the refrigerant storage vessel. The COPs of the bypass two-circuit refrigeration cycle at RF-mode and F only-mode increased by 10% and 17%, respectively, with the increase of the condenser capacity and SLHX length. In addition, the COP of the system at F only-mode increased by 10% with the adoption of the refrigerant storage vessel.

In the present work, a new method is proposed for the measurement of the ice packing factor (IPF) of an ice slurry using the index of refraction. The purpose of the new method is to improve the resolution of the measurement and to increase its resistance to electric noise as compared to the standard IPF measurement technique that relies on the measurement of the freezing temperature. These two methods are similar in that they both obtain a concentration of aqueous solution from the measured physical quantities and calculate the IPF using the relationship between concentration and IPF. Both methods were also compared with the results from the calorimeter method obtaining the IPF directly. The two methods are in good agreement (within 5%), which demonstrate the validity of the newly proposed method.

An experimental study was performed to measure the ice concentration of an ice slurry pipe flow in real time. In the present work, we suggest a refractive index measurement method and compare it to freezing point and density measurement methods. To measure the refractive index of the solution, ice particles in the ice slurry should be completely removed and a hydro-cyclone was used in this study instead of a mesh. The results from the measurement method by the refractive index coincided with those by the density within a ±5% error range when the real-time solution density was used. Even though the density measurement method showed good resolution, the results using the initial density of the solution were more than 10% in error compared to those using the real-time density. And the density measurement method has an error range 1.5 times greater than the refractive index measurement method.