Cemil İnan
Published: 2000
Total Pages:
Get eBook
In this study , understanding of how frost accumulates on evaporator surfaces and affects the heat transfer , airside pressure drop, and refrigerator performance are investigated. In order to make a realistic model of evaporator and simulate the refrigerator performance changes while more frost was deposited on the surfaces of evaporator, heat and mass transfer mechaisms of a refrigerator cabinet were also studied. The study was made in five individual chapters. After a short introduction in Chapter-l, in the Chapters 2 and 3 open door heat and mass transfer analysis and experimental study were made. In Chapter-4, the mathematical model of air distribution system was introduced. Finally in Chapter-5, the mathematical model of evaporator was given. Solvers for all models that were developed in the chapters from 2 to 5 were written in the EES environment. Results of the studies are surnrnarized in chapter basis below . Chapter-2; a physİcal model for estimatİng the cabinet flow and heat transfer rates occuring when door is opened were introduced. Model was based on the momentum and energy equations for two-dimensional flow in Cartesian coordinates. The model includes effect of shelves, which changes from none to three, in all possible combinations. Verification was made by using the experimental measurements, and flow visualization techniques. Experimental study was made on a commercially available domestic refrigerator cabinet. Transient velocity and temperature profiles were measured along the height of the cavity opening. Results of open door studies were summarized by set of equations that shows the parameters contributing to the flow and heat transfer. To solve these coupled equations, a solver was developed in EES environment. It was found that in an empty cabinet (without shelf) the inflow area was larger then the outflow area. This shows that the flow is accelerating along with the cabinet height. The maximum air velocity was measured at a point very close to bottom wall. Flow regimes in an open cavity were classified as three characteristic periods starting right after the door opening. Those were initial transient period (0 to 7-10 seconds), transition period (7 -1 0 to 20 seconds ), and quasi-steady period, ( after 20 seconds ). During the first period, one volumetric air exchange is established, and both inlet and outlet velocities are high. After one volumetric air exchange, the flow is mainly driven by wall temperatures. As wall temperature increases, velocity decreases. The conveciİon heat transfer coefficient decreases when the number of shelves are increased. Experimental studİes showed that, it was acceptable to take wall temperatures constant for the first 20 seconds of door opening. The model results and experimental results agreed within the error levels of +/-10% for wall temperatures above 3 C, and+/- 30% for below 3 C. Chapter-3; by help of heat and mass transfer analogy, the latent and sensible (convective + radiation ) heat transfer, and mass transfer percentages in the total cabinet load were analyzed. According to this study , the daily heat gain of the cabinet for 20 times of door opening was increased by 15% to 32% for 30% RH to 1OO%RH respectİvely .In these analysis; ambient temperature was 25 C, wall temperature was 5 C, duration of each door opening was 20 seconds, number of the shelves were three, and shelves were equally spaced. The percentages of heat gain coming from door openings will, of course, be increased in the refrigerators having better insulaion. The latent and sensible loads caused by door opening were found to be equal for ambient conditions of 60% RH. The breakdown of modes of heat transfer that occur during door opening period is given below .The conduction heat transfer from cabinet walls during the 20 s of door opening is neglected. Radiation heat transfer: 8.2% Convection (sensible) heat transfer: 42% Latent heat transfer: 49.8% Chapter-4; a mathematİcal model for analyzing the air distribution system of a nofrost refrigerator was developed. Model was based on conservation of momentum and the first law of the thermodynamics. A solver was developed in EES environment. A commercial refrigerator cabinet was used to compare the model result with the measurements and the CFD analysis. The estimated airflow rate, and static pressure were found compatible to measurements with less then 3% error. The difference between model and CFD analysis were maximum 15%. The results were found promising, and the solver could be used effectively in the areas of initial design of the air distribution system, selection of a fan, and balancing the airflow rates distributed between compartments of refrigerator .Chapter-5; a refrigerator evaporator, tube on sheet type, was analyzed. Calculatİon procedure for obtaining the heat and mass transfer characteristics of the refrigerator evaporators that operate under dry and frosting conditions has been introduced. A mathematical model that was used to obtain heat transfer coefficients, and pressure drop characteristics was developed. The solver is capable of calculating the frost thİckness, density , frost mass, overall heat transfer coefficient, exit temperature and humidity, and pressure drop values for each individual tube row for quasi-steady time steps. . As it is indicated in the literature, during the frost formation, both thickness and density increase with time. This phenomenon is due to the fact that some percentage of mass transfer goes to increase the thickness, and the remaining goes to increase density (this process can be called as densification of the frost). These percentages are found very important parameters on calculation of evaporator characteristics. Unlike the literature, where densification is calculated by using the analytical methods, these percentages were chosen as constant parameters. According to the study, when the densification ratio was increased from 0.2 to 0.5, the pressure drop decreased approximately three times, the accumulated frost mass increased 4%, and the heat transfer coefficient increased 5% (at the end of 12 hours operation time). By taking the densification ratio equal to 0.3, the effect of following parameters were investigated, the operation time is 12 hours. Effect of the evaporation temperature: Increasing the evaporator temperature from -35 C to -25 C results; 10% decrease on heat transfer coefficient, half of pressure drop, and 5 times lower frost mass. Effect of the airflow rate: When the flow rate is increased from 5 l/s to 25 l/s, the UA value increases 3 times, pressure drop ncreases more than 10 times, and frost mass increases 3.5 times. Simulation of the real refrigerator case was studied. Airflow rate of each time step was calculated by using the updated pressure drop values of whole air distribution system and fan characteristic curves, as well. In order to do this, the solver developed in chapter-4 was integrated to solver of Chapter-5. Evaporator UA value and pressure drop values were obtained for three different inlet air relative humidity values. Accordİng to this study , UA showed first increasing trend until a peak value was reached, then relatively sharp decreasing trend that ended when the evaporator was completely blocked. On the other hand, pressure drop calculations showed exponentially increasing trend all the time. In addition, the pressure drop and frost thickness values were increased more rapidly for higher relative humidity values. Complete evaporator blockage is occurred after 45 hours and 110 hours of continuous operation time for inlet humidity values of 80%, and 50%, respectİvely . In oroer to show how all individual models can be used in calculation of quasi-steady temperature and humidity values of cabinet air, a sample run in a freshfood cabinet was made. Results of the evaporator model was compared to the experimental studies of Seker (1999). The estimated UA value was found 10% higher than the experimental value. However, the calculated pressure-drop value was 25% lower than the measured value. It should be noted that the experimental study was obtained for only 2 hours of operatİon time, therefore the model results will be reexamined after having more experimental results.
