Module 4: Interferometry
  Lecture 21: Applications - buoyant convection
 

The present work is concerned with visualizing isotherms in square cavity at various instants of time. The top and bottom walls of the cavity are respectively cold and hot and their temperatures are maintained constant for all time. This is the classical Rayleigh-Benard problem with confining side walls. The original problem of convection in an infinite fluid layer of small thickness admits three-dimensional cellular flow as solution. This is, however, considerably modified in the presences of confining side walls [111, 112] and at higher Rayleigh numbers. The formation of longitudinal rolls with their axes aligned normal to the shortest side has been observed in experiments on a horizontal fluid layer [41]. It is to be expected that when symmetry planes between adjacent cells are replaced by adiabatic walls. There will be no fundamental change in the flow pattern. This assumption has also been made implicitly in Tolpadi and Kuehn [8]. Fluid contained in the cavity whose vertical cross-section is a square and is long in the third dimension in the horizontal plane, is thus expected to exhibit cellular motion with the cell axis parallel to the longest side (Figure: 4.38(a)). If the length exceeds a critical value, two cells with the same direction of vorticity appear in the cavity. Since the cells are unidirectional, the temperature distribution within the two cells is similar. Hence the average temperature distribution obtained by optical projection is representative of the flow field at any section along the cavity length.

Interferograms have been obtained in the present work by orienting the light beam parallel to the longest dimension of the cavity. Rayleigh numbers considered in the study are . These are large in comparison to the critical Rayleigh number for the infinite fluid layer. These are, however, close to the critical value for a cavity with confining side walls [113]. Fringe pattern have been thinned using image processing operations. The thinned fringes near the cold wall have been used to compute the local Nusselt number. Images have been acquired during the transient evolution of the thermal field as well as at steady state. The images show distinct flow patterns during the early transient phase in comparison to steady state.

The apparatus used to study buoyancy–driven motion of air in a cavity is shown in Figure 4.38(b). the cavity is of a square cross-section with its width and height being adjustable in the range of 4 to 6 cm.The top wall is cooled with the help of chilled water from a constant temperature bath to a temperature of around while the bottom wall is maintained at a temperature of around (close to room temperature). The temperatures of both the walls are maintained constant for the entire duration of the experiment. In the present study, special precautions have been taken so  the top and bottom walls are maintained constant for the entire duration of the experiment. In the present study, special precautions have been taken so that the top and bottom walls out of thin brass sheets and exposing them to large volume flow rates of water. The side walls are made of 12.5 mm thick perspex sheets and padded using thermocole insulation. To avoid temperature tanks containing hot and cold water. The upper wall is cooled to a temperature below the ambient value and convection is initiated in this part of the cavity. The heat transfer rates at the hot and cold walls are unequal during the transient process and approach each other at steady state. Since the active boundary in the present work is the cold upper wall, the heat transfer rates have been computed in this region.