Simultaneous natural convective heat transfer from the top, bottom and side surfaces of two-sided inclined square plates having various thicknesses has been numerically investigated. The aim of this work is to determine whether the plate thickness has a significant influence on the heat transfer rates from the plate surfaces when the plate is inclined to the horizontal and to determine how the heat transfer rate varies with this angle of inclination. The upper, lower and side surfaces of the plate have been assumed to be isothermal and at the same temperature which is higher than that of the surrounding fluid. The range of conditions considered is such that laminar, transitional, and turbulent flow occur over the plate. The numerical solution has been obtained using the commercial CFD solver ANSYS FLUENT © . In this study, results have only been obtained for the case where the plate is exposed to air. Inclination angles of between 0 and 40 degrees from the horizontal and plate dimensionless thicknesses (thickness-to-side length ratios) of between 0 and 0.3 have been considered. Variations of the mean Nusselt number with Rayleigh number for the top surface, bottom surface, side surface and that averaged over all heated surfaces of the plate for various inclination angles and for various plate dimensionless thicknesses have been obtained.

A numerical study of natural convective heat transfer from an upward facing, heated horizontal isothermal surface imbedded in a large flat adiabatic surface has been undertaken. On the heated surface is a series of triangular shaped waves. Laminar, transitional, and turbulent flow conditions have been considered. The flow has been assumed to be two-dimensional and steady. The fluid properties have been assumed constant except for the density change with temperature giving rise to the buoyancy forces. This was with treated using the Boussinesq approach. The numerical solution has been obtained using the commercial CFD solver ANSYS FLUENT © . The k -epsilon turbulence model with full account being taken of buoyancy force effects has been employed. The heat transfer rate from the heated surface expressed in terms of a Nusselt number is dependent on the Rayleigh number, the number of waves, the height of the waves relative to the width of the heated surface, and the Prandtl number. This study obtained results for a Prandtl number of 0.74 which is effectively the value for air. An investigation of the effect of the Rayleigh number, the dimensionless height of the surface waves, and the number of surface waves on the Nusselt number has been undertaken.

The purpose of the present work was to investigate numerically the effect of the top and/or bottom blind openings on the convective heat transfer from a window fitted with a double-layered top down-bottom up honeycomb blind system. Top down-bottom up systems that utilize so-called honeycomb (or cellular) blinds can be opened at the top and/or the bottom. When a honeycomb blind is fully closed there are two or more vertical blind portions and a series of horizontal or nearly horizontal blind portions which join the vertical portions and form a column of cells. This gives the blind system its honeycomb or cellular structure. When opening a honeycomb blind the vertical portions of the blind bend or fold allowing the overall height of the blind to decrease. A double-layered honeycomb blind is constructed with three vertical blind portions and two columns of cells. A recessed window has been considered in the present study and only the convective heat transfer from the window to the surrounding room has been investigated. The surfaces of the blind are assumed to offer no resistance to heat transfer. The commercial CFD solver ANSYS FLUENT © has been used to obtain the solution. Over the range of parameters considered in this study, both laminar and turbulent flow can occur. The k-ε turbulence model has been used in obtaining the solutions. The convective heat transfer rate from the inner surface of the window, expressed in terms of a mean Nusselt number based on the window height and the difference between the window and the air temperatures, will depend on the Rayleigh number, also based on the window height, and the difference between the window and the air temperatures, the dimensionless top and bottom blind openings, and the dimensionless window recess depth. Variations of the mean Nusselt number with Rayleigh number for various values of these other parameters have been obtained and the results used to study how these other parameters affect the window heat transfer rate.

Natural convective heat transfer from the top and bottom surfaces of a thin circular isothermal horizontal plate which, in general, has a centrally placed adiabatic section has been numerically investigated. The temperature of the plate surfaces is higher than the temperature of the surrounding fluid. The range of conditions considered is such that laminar, transitional, and turbulent flow occurs over the plate. The heat transfer from the upper and lower surfaces of the plate as well as the mean heat transfer rate from the entire surface of the plate have been considered. The flow has been assumed to be axisymmetric and steady. The k -epsilon turbulence model with account being taken of buoyancy force effects has been used and the solution has been obtained using the commercial CFD solver ANSYS FLUENT © . The heat transfer rate from the heated plate has been expressed in terms of a Nusselt number based on the outside plate diameter and the difference between the plate temperature and the fluid temperature far from the plate. The mean Nusselt number is dependent on the Rayleigh number, the ratio of the diameter of the inner adiabatic section to the outer plate diameter, and the Prandtl number. Results have only been obtained for a Prandtl number of 0.74, i.e., effectively the value for air. The variations of the mean Nusselt number averaged over both the upper and lower surfaces and of the mean Nusselt numbers for the upper surface and for the lower surface with Rayleigh number for various adiabatic section diameter ratios have been studied. The use of a reference length scale to allow the correlation of these mean Nusselt number-Rayleigh number variations has been investigated.

Heat transfer from a window recessed into a wall when there is a plane blind mounted over the window, the blind being in the same plane as the wall, has been considered. In such situations the blind is sometimes narrower than the window, i.e., there are gaps between the vertical edges of the blind and the vertical edges of the wall window opening. In the present study the effect of the size of the blind edge gaps on the window heat transfer has been numerically investigated. Conditions under which the flow over the window-blind system is laminar and under which it is turbulent have been considered. The present study, as in many previous window heat transfer studies, has only considered the convective heat transfer from the window to the surrounding room. The flow has been assumed to be steady and symmetrical about the window-blind vertical center-plane. The Boussinesq approach has been used and it has been assumed that the “window”, which is modeled as a plane surface, is at a uniform temperature. Attention has been restricted to the case where the window is at a higher temperature than the temperature of the room air, this room air being assumed to be at a uniform temperature far from the window. The solution has been obtained by numerically solving the governing equations using the commercial CFD solver FLUENT © . The k -epsilon turbulence model has been used in obtaining the solution. Results have only been obtained for a Prandtl number of 0.74, i.e., effectively the value for air. The effect of the dimensionless size of the blind side gaps on the window Nusselt number has been studied for various Rayleigh numbers; the results being used to determine under what conditions the blind side openings have a significant effect on the window heat transfer rate.

Many smaller churches and similar buildings in Canada have been converted into small theaters. Such theatres are often not fitted with an air-conditioning system. For performances in the fall these theaters sometimes rely on buoyancy driven natural ventilation to moderate the indoor air temperature. Such ventilation systems usually involve near floor inlet vents and a roof level air discharge system. A preliminary numerical study of the effect of inlet vent position and size on the performance of such a system has been undertaken. A simple model of a typical theater building of the type considered has been used. The heat generated by the audience has been represented by a uniform heat flux distributed over the audience area. Inlet vents have been assumed to be located low on the side walls of the theater and the air-flow leaving the theatre has been assumed to be through vents at the top of a chimney system. The flow has been assumed to be steady and symmetrical about the vertical center-line through the building. The Boussinesq approach has been adopted. The standard k-epsilon turbulence model has been used. The solution has been obtained using the commercial CFD solver ANSYS FLUENT © .

Natural convective heat transfer from a horizontal flat rectangular isothermal heated element imbedded in a flat rectangular adiabatic surface has been numerically studied. The surface of the heated rectangular element is in the same plane as the surface of the surrounding adiabatic material. A rectangular flat horizontal adiabatic surface is mounted parallel to and at a relatively short distance from the heated element. The heated element is facing upwards with the covering surface above the element. For the conditions considered laminar, transitional, and turbulent flows can occur. The flow has been assumed to be steady. Constant fluid properties have been assumed except for the density change with temperature which gives rise to the buoyancy forces. This was dealt with using the Boussinesq approach. To obtain the solution, the commercial CFD solver ANSYS FLUENT © was used to numerically solve the governing equations. The k -epsilon turbulence model was employed with account being taken of buoyancy force effects. The effects of the dimensionless distance of the rectangular covering surface from the heated rectangular element and of the ratio of the side lengths of the rectangular element on the variation of the Nusselt number with Rayleigh number have been examined.

Natural convective heat transfer from a vertical isothermal cylinder mounted on a flat adiabatic base has been numerically studied. The cylinder has an exposed top surface. The cylinder is relatively very short, i.e., has a height that is equal to or less than the cylinder diameter. Both the cases where the cylinder is pointing upward and where it is pointing downward have been considered. The governing equations have been numerically solved using the commercial CFD solver ANSYS FLUENT © . Results have only been obtained for Prandtl number = 0.74. The mean heat transfer rates have been expressed in terms of a Nusselt number, consideration being given both to the heat transfer rate from the entire cylinder surface and to the heat transfer rates from the side and top surfaces of the cylinder. The effect of the dimensionless cylinder height–to–diameter ratio on the Nusselt number variation has been studied in detail.

In summer when the air-conditioning system is in use cool air from a floor-mounted vent located beneath a window often flows over the warm window. The presence of a blind system over the window will, in general, influence the effect of the vent flow on the convective heat transfer rate from the window. The effect of a Top-Down, Bottom-Up plane blind system and a cool air vent flow on the heat transfer rate from a recessed window has therefore been numerically studied here. The actual situation considered in this study is an approximate model of real situations. The window is represented by a plane isothermal section recessed into the wall, this window section being hotter than the room air far from the window. The floor-mounted vent is assumed to be located against the wall and to have a uniform discharge velocity which is normal to the vent surface. The flow has been assumed to be two-dimensional, i.e., the effect of the window and vent width has not been considered. The flow has been assumed to be steady and situations involving both laminar and turbulent flow have been considered. The fluid properties have been assumed constant except for the density change with temperature that gives rise to the buoyancy forces, this being dealt with using the Boussinesq approach. The governing equations have been solved using the commercial CFD code ANSYS FLUENT © , the k -epsilon turbulence model having been used. The solution has the following parameters: the Rayleigh number, the Reynolds number based on the vent discharge velocity, the dimensionless depth that the window is recessed, the Prandtl number, the dimensionless top and bottom blind opening, the dimensionless size of the air vent, and the dimensionless vent discharge temperature to undisturbed air temperature difference. Results have only been obtained for a Prandtl number of 0.74 and for fixed values of the dimensionless depth that the window is recessed, the dimensionless size of the air vent, and the dimensionless vent discharge temperature difference. The effects of the other dimensionless variables on the window Nusselt number have been numerically studied.

Natural convective heat transfer from isothermal rectangular cylinders which have an exposed upper surface has been numerically studied. The cylinders considered have high aspect ratios, i.e., have high width-to-depth ratios, and are relatively short, i.e., have a “height” that is of the same order of magnitude as their width. The cylinders considered are mounted on a plane adiabatic base, the cylinders being normal to the plane base with the cylinders pointing either vertically upwards or vertically downwards. One of the main aims of the present work was to numerically determine how the depth-to-width ratio of the rectangular cylinder influences the mean heat transfer rate from the cylinder when this depth-to-width ratio is large. The flow has also been assumed to be steady and laminar and it has been assumed that the fluid properties are constant except for the density change with temperature which gives rise to the buoyancy forces, this having been treated by using the Boussinesq approach. The solution has been obtained by numerically solving the governing equations using the commercial CFD solver, ANSYS FLUENT © . The solution is dependent on the Rayleigh number, the ratio of the width to the height of the heated cylinder, the ratio of the width to the depth of the heated cylinder, the Prandtl number, Pr , and on whether the cylinder is pointing vertically upwards or vertically downwards. Because of the applications that motivated this study, results have only been obtained for a Prandtl number of 0.74, i.e., effectively the value for air. A range of the other governing parameters has been considered and the effects of these governing parameters on the Nusselt number variation have been examined.

Mixed convective heat transfer from an isothermal cylinder with a rectangular cross-section and a relatively large height-to-width ratio has been numerically studied. The axis of the cylinder is horizontal with the longer sides of the rectangular cylinder being vertical. There is a vertical forced flow over the cylinder. The flow conditions considered are such that in general mixed forced and natural convective flow exists. Both the case where the buoyancy forces act in the same direction as the forced flow (assisting flow) and the case where they act in the opposite direction to the forced flow (opposing flow) have been considered. The flow has been assumed to be two-dimensional and the Boussinesq approximation has been adopted. Attention has been restricted to the flow of air and results have therefore been obtained for a Prandtl number of 0.74. The flow conditions considered are such that laminar or turbulent flow can exist. The main attention is this work has been directed at determining the effect of the flow parameters on the mean heat transfer rate from the cylinder and on determining the conditions under which the flow can be assumed to be forced convective and under which it can be assumed to be natural convective.

Top Down – Bottom Up blinds have become quite popular in recent times. However the effects of such blind systems on the convective heat transfer from the window to the surrounding room have not been extensively studied and the effect of solar irradiation of the blind on the window heat transfer has not received significant attention. The purpose of the present work was therefore to numerically investigate the effect of solar irradiation of Top Down – Bottom Up slatted blinds on this convective heat transfer. An approximate model of the window-blind system has been adopted. The solar radiation falling on the blinds is assumed to produce a uniform rate of heat generation in the blind. The Boussinesq approximation has been used. Radiant heat transfer effects have been neglected. Conditions under which laminar, transitional and turbulent flows occur have been considered. The main emphasis is on the effect of the magnitude of the irradiation and of the size of the blind openings at the top and bottom of the window on the convective heat transfer rate from the window to the room.

Top Down–Bottom Up plane blinds have become popular in recent times. These are blinds that can be both raised at the bottom and lowered at the top. Such blinds have the potential to reduce energy consumption in buildings by allowing the controlled use of sunlight to illuminate the building (daylighting) and/or the use passive solar room heating while still providing shade from the direct sunlight and privacy for the occupants. The effect of such blind systems on the convective heat transfer rate from the window to the room to which it is exposed has not been extensively studied. The purpose of the present work therefore was to numerically investigate the effect of the blind openings with Top Down-Bottom Up plane blinds on the convective heat transfer from the window. The cases where the flow over the window-blind system is laminar or turbulent have been considered. The present study, as is the case in many previous window heat transfer studies, considers only the convective heat transfer from the window. In the present study the mean flow has been assumed to be steady and two-dimensional. The Boussinesq approach has been used and it has been assumed that the “window” is at a uniform temperature. The solution has been obtained by numerically solving the governing equations using the commercial CFD solver, FLUENT © . The standard k -epsilon turbulence model with full account being taken of the effect of the buoyancy forces has been used. Results have been obtained for a Prandtl number of 0.74, which is essentially the value for air. The effects of the top and bottom dimensionless blind openings, of the dimensionless depth to which the window is recessed and of the Rayleigh number on the Nusselt number have been studied.

The effect of thermal mass is becoming increasingly important under the scope of low energy use building construction and energy conservation. In this paper the dynamic behaviour of the effect of the thermal mass of external walls on the development of solar-assisted buoyancy-driven ventilation flow rates, air flow patterns and temperature distributions in a simple three-storied atrium building was investigated using unsteady Reynolds Averaged Navier-Stokes equations (RANS) modelling approach. The SST-k-omega turbulence model and the Discrete Transfer Radiation Model (DTRM) were used in the CFD simulations. The unsteady-state governing equations were solved using a commercial CFD solver FLUENT © . The numerical results were obtained for indoor transient thermal conditions by assuming the external concrete walls of the building under three situations: (i) walls with internal insulation covering (ii) walls with external insulation covering, and (iii) walls without insulation covering. The buoyancy-driven ventilation volume flow rates and internal temperature variations from 8:00 to 15:00hrs on July 15, 2011 were calculated using CFD simulations. Numerical results obtained indicate that the transient CFD simulations provide important information about the effect of the thermal mass on the thermal behaviour of the building. It was found that the thermal mass of external walls with external insulation covering was a better option which helped to control the diurnal temperature swings making the conditions within the building more comfortable. Heat stored in the walls can be useful for the night-time buoyancy-driven ventilation when there is no solar radiation.

Natural convective flow through a vertical plane channel has been considered. The walls of the lower portion of the channel are heated to a uniform temperature, both of the heated walls being at the same temperature. The walls of the upper part of the channel are unheated, i.e., are adiabatic. The reason for undertaking the study arose from the fact that using an adiabatic upper wall section can, through the so-called chimney effect, increase the flow rate through the channel and therefore increase the heat transfer rate from the lower heated wall section. However if the upper adiabatic wall section is too long the increased pressure drop due to viscous effects can lead to a reduced flow rate through the channel and to a reduced heat transfer rate. Therefore a need existed to examine in more detail the effect that the height of the upper adiabatic wall section has on the heat transfer rate. The flow has been assumed to be steady and the Boussinesq approximation has been adopted. The solutions have been obtained using the commercial CFD code FLUENT © . The solution has the Rayleigh number, the Prandtl number, the ratio of the channel width to the height of the heated channel wall section, and the ratio of the height of the adiabatic channel wall section to the height of the heated wall section as parameters. Results have only been obtained for a Prandtl number of 0.74 (the value for air at temperatures near ambient temperature). Results have been obtained for a wide range of values of the remaining parameters and the effects of these parameters on the mean Nusselt number have been studied.

Natural convective heat transfer from a wide isothermal plate which has a “wavy” surface, i.e., has a surface which periodically rises and falls, has been numerically studied. The surface waves run in the horizontal direction, i.e., are normal to the direction of flow over the surface, and have relatively small amplitude. Attention has been restricted to the case where the waves have a rectangular cross-sectional shape. The plate is, in general, inclined to the vertical, consideration only being given to inclination angles at which the heated plate is facing upwards. The range of Rayleigh numbers considered extends from values that for a non-wavy vertical plate would be associated with laminar flow to values that would be associated with fully turbulent flow. The flow has been assumed to be steady and fluid properties have been assumed constant except for the density change with temperature that gives rise to the buoyancy forces, this being treated by means of the Boussinesq approximation. The Reynolds averaged governing equations in conjunction with a standard k -epsilon turbulence model with buoyancy force effects fully accounted for have been used in obtaining the solution. The governing equations have been solved using the commercial cfd code FLUENT. The solution has the following parameters: (i) the Rayleigh number based on the height of the heated plate, (ii) the Prandtl number, (iii) the ratios of the amplitude of the surface waviness and of the pitch of the surface waves to the height of the plate, and (iv) the angle of inclination of the plate to the vertical. Results have only been obtained for a Prandtl number of 0.74. The effects of the other dimensionless variables on the mean surface Nusselt number have been numerically studied.

Natural convective heat transfer from a wide isothermal plate which has a wavy surface, i.e., has a surface which periodically rises and falls, has been numerically studied. The main purpose of the study was to examine the effect of the surface waviness on the conditions under which transition from laminar to turbulent flow occurred and to study the effect of the surface waviness on the heat transfer rate. The surface waves, which have a saw-tooth cross-sectional shape, are normal to the direction of flow over the surface and have a relatively small amplitude. The range of Rayleigh numbers considered in the present study extends from values that for a non-wavy plate would be associated with laminar flow to values that would be associated with fully turbulent flow. The flow has been assumed to be steady and fluid properties have been assumed constant except for the density change with temperature that gives rise to the buoyancy forces, this being treated by means of the Boussinesq type approximation. A standard k -epsilon turbulence model with full account being taken of the effects of the buoyancy forces has been used in obtaining the solution. The solution has been obtained using the commercial CFD solver FLUENT. The solution has the following parameters: the Rayleigh number based on the plate height, the Prandtl number, the dimensionless amplitude of the surface waviness, and the dimensionless pitch of the surface waviness. Results have been obtained for a Prandtl number of 0.7 and for a single dimensionless pitch value for Rayleigh numbers between approximately 10 6 and 10 12 . The effects of Rayleigh number and dimensionless amplitude on the mean heat transfer rate have been studied. It is convenient in presenting the results to introduce two mean heat transfer rates, one based on the total surface area and the other based on the projected frontal area of the surface.

Most numerical studies of convective heat transfer between a window-blind system and a room are based on the assumption that the flow remains laminar. However, in the case of larger windows it is to be expected that transition to turbulent flow will occur in the system. The aim of the present study was to numerically determine when transition to turbulent flow occurs in a recessed window system and the effect of a simple partially open plane blind on when transition occurs. An approximate model of a recessed window that is covered by a partially open plane blind has been considered. The inner surface of the window is modeled as a plane vertical isothermal surface and the blind as a thin plane surface that offers no resistance to heat transfer. The fluid properties have been assumed constant except for the density change with temperature that gives rise to the buoyancy forces, this being dealt with using the Boussinesq approach. Radiant heat transfer effects have been neglected. The k -epsilon turbulence model with the full effects of the buoyancy forces being accounted for has been used in obtaining the solution. The governing equations have been solved using the commercial finite-volume based cfd code FLUENT. The solution has as parameters: (1) the Rayleigh number, (2) the Prandtl number, (3) the dimensionless ‘window’ recess depth, (4) the dimensionless blind opening, and (5) whether the ‘window’ surface is at a higher or lower temperature than the room air. Because of the application being considered results have only been obtained for Pr = 0.7 and for the case where the ‘window’ surface is at a higher temperature than the room air. The effect of transition on the mean Nusselt number variation with Rayleigh number with various blind openings for various dimensionless window recess depths has in particular been studied.

The horizontal frame members that often protrude from the inner surface of a window can significantly effect the convective heat transfer rate from this inner surface to the room. The purpose of the present numerical study was to determine how the size of a pair of horizontal frame members effect this heat transfer rate. The flow has been assumed to be steady and conditions under which laminar, transitional, and turbulent flows occur are considered. Fluid properties have been assumed constant except for the density change with temperature that gives rise to the buoyancy forces, this being dealt with using the Boussinesq approach. The governing equations have been solved using the FLUENT commercial CFD code. The k-epsilon turbulence model with standard wall functions and with buoyancy force effects fully accounted for has been used. The solution has the following parameters: the Rayleigh number, the Prandtl number, the dimensionless window recess depth, and the dimensionless width and depth of the frame members. Results have been obtained for a Prandtl number of 0.74.