Investigation of heat transfer inside a cavity having heat sources

In engineering applications, various types of cavity structures are used. Some important applications of these cavities are energy storage systems, solar collectors, double pane windows, electronic equipment cooling and heating of rooms. Many efforts have been made to investigate natural convection heat transfer in closed cavities. Theoretical, numerical or experimental studies are conducted on the natural convection heat transfer in the cavity. Most of the work focuses on two-dimensional and steady state solution of the problem. Natural convection in a closed cavity differs widely in boundary conditions, geometries, fluids and flow regimes. Based on the thermal boundary conditions, enclosures can be classified into two main groups: heating from a horizontal wall and heating from a vertical wall. In the second group, two opposite vertical walls of the cavity are differentially heated while its top, bottom, front and back walls are maintained adiabatic. In this cavity, one vertical wall is heated and the other vertical wall is cooled. The cavity has two active vertical walls. Isothermal vertical walls are called hot and cold walls. Considering the cavity geometry, a parameter which is called the aspect ratio is used. The aspect ratio is the ratio of the cavity height to cavity length. Depending on the aspect ratio, enclosures can be divided into three groups: shallow rectangular, square and tall rectangular. In these groups, the aspect ratio is smaller than one, equal to one and greater than one, respectively. Many studies have been conducted on natural convection in a cavity with adiabatic horizontal walls and differentially heated isothermal vertical walls. In this frame, some researchers investigated square and cubical cavities, or rectangular and rectangular prism cavities. Some other researchers studied natural convection of air in a cavity with partially active vertical walls. The active part is half height of the cavity wall and the inactive part is adiabatic. In these two types of studies, the heat transfer mechanism is generally by natural convection alone. The differentially heated cavities have found applications in many areas. However, in many other applications, the cavity is heated or cooled from only one side. Therefore, it has one active vertical wall. This thesis study focuses on a closed cavity with one active vertical wall. The cavity is cooled from one side with a local heat sink. The temperature gradient on the active wall is larger than that on the opposing wall, which is called the hot wall. These two vertical walls have different values of constant temperatures. The horizontal top and bottom walls are adiabatic. The fluid inside the cavity is air. The angle of inclination of the cavity is zero. Because the vertical walls of the cavity are maintained at the temperatures of constant values, the temperature difference between the walls is resulted in the flow of air inside the cavity. In this study, heat transfer in square and tall rectangular cavities is investigated. Six rectangular cavities with different aspect ratios are studied to find the effect of aspect ratio on the heat transfer in the cavity. In the literatue review, the studies of many researchers are summarized. For each work, the geometry of the cavity, the aspect ratio, the fluid in the cavity, the material and the surface emissivity of the walls, the thermal boundary conditions on the walls of the cavity, the Prandtl number range, the Rayleigh number range and the proposed heat transfer correlations are provided. In this thesis, the heat transfer problem is studied experimentally. For this purpose, an experimental facility is designed and manufactured. Temperature measurements are performed to study the heat transfer in the rectangular cavity. The problem is studied in the steady state. The temperature distribution across the cavity walls is obtained for different cavity heights and depths. The effects of the local sink on the temperature profile, the flow and the heat transfer are obtained. The temperature measurements are plotted in terms of both dimensional and non dimensional graphs. The flow in the cavity is presented in the form of temperature contours. The cavity geometry is non-dimensionalized. The calculations are done on the measurement data to obtain heat transfer characteristics. The heat transfer variables are presented in the form of non-dimensional heat transfer numbers. The heat transfer in the cavity is studied in terms of both average and local values. The change in heat transfer with respect to the cavity geometry, the aspect ratio, the surface emissivity and the temperature difference is presented by means of equations. The variation of the heat transfer data is discussed. The results are compared with the literature. In this study, the heat transfer in one of the rectangular cavities is investigated in three dimensions. For this case, the heat transfer mechanism is combined radiation and natural convection. The experiments are performed to investigate the heat transfer in the directions of the cavity length, height and depth. The temperature distribution in the cavity is measured. The contour maps of temperature measurements are plotted. The variation of the Nusselt number in the cavity is obtained. The average Nusselt number is obtained by means of the two- and three-dimensional approaches. Thus, the difference between the two- and three-dimensional investigations is obtained. After completing the three-dimensional investigation, the heat transfer inside the cavity is studied in two-dimensions for all six cavities. In the frame of this work, the experiments are conducted to study the heat transfer for the aspect ratio range of 1 to 6. The heat transfer mechanism is natural convection and radiation. For each of six cavities, the temperature profile across the cavity length is obtained. The investigation is performed for seven heights at one cavity depth. For each cavity, the variations of the temperature profile and the local Nusselt number along the cavity height are obtained. The average and local heat transfer data are found. Heat transfer correlations are obtained. Having studied the combined radiation and natural convection, the heat transfer mechanism is changed to the natural convection alone. For this purpose, the surface emissivity of the walls of the cavity is reduced. The investigation is performed at the mid-depth of the cavity. In this frame, the aspect ratio is varied between 1 and 6. For all six cavities, the temperature profiles, the isotherms and the heat transfer data are obtained. The comparisons between the average and the local heat transfer data are made. For this work, heat transfer correlations are provided. In the thesis, the effect of thermal boundary conditions is also studied. The aspect ratios of the six cavities are maintained constant. The heat transfer is by natural convection alone. One of the vertical walls of the cavity is cooled with a local heat sink. The opposing vertical wall is heated with a local heat source. Thus, the cavity with two active vertical walls is obtained. For this cavity, the temperature difference between these two walls is increased when compared to the cavity with one active vertical wall. Because of the change in the thermal boundary conditions, the temperature gradients on the walls, the temperature field, the isotherms, and hence, the flow in the cavity are changed. The variation of the local Nusselt number along the cavity height is obtained. The measurements are performed at the mid-depth of the cavity. The relation between the Nusselt and the Rayleigh numbers is presented in terms of a correlation. The effect of radiation is obtained by comparing the combined radiation and natural convection, and the natural convection cases with each other. The effect of the thermal boundary conditions on the opposing vertical wall is found by comparisons between the cavity with one active vertical wall and the cavity with two active vertical walls. A new coefficient has been generated by this comparison. A correlation which shows the variation of this coefficient with respect to the Rayleigh number is presented.