Measurements in the Transition Vortex Flow Regime of Mixed Convection Above a Horizontal Heated Plate

1988 ◽  
Vol 110 (2) ◽  
pp. 358-365 ◽  
Author(s):  
S. S. Moharreri ◽  
B. F. Armaly ◽  
T. S. Chen

Experimental results covering the transition vortex flow regime of mixed convection over a heated, horizontal flat plate are presented. A criterion for the onset of vortex instability as a function of critical Reynolds and Grashof numbers was established with the aid of a flow visualization technique. The three-dimensional nature of this flow regime was documented through both velocity and temperature measurements using laser-Doppler and hot/cold-wire anemometers, respectively. A higher buoyancy force, through a higher plate temperature or a larger downstream distance, and/or a lower free-stream velocity, intensifies the strength of the vortices. Velocity and temperature profiles through vortex peaks and valleys are reported to quantify the behavior of these vortices. It has been found from these measurements that the two-dimensional laminar mixed convection flow changes into a transitional three-dimensional vortex flow in a relatively short distance from the leading edge of the plate. The vortex three-dimensional flow continues to intensify as the buoyancy force increases and then develops into a two-dimensional fully turbulent flow at the end of the transition regime. These findings place an upper limit on the applicability of the two-dimensional, laminar boundary layer flow analysis for mixed convection over a heated horizontal flat plate.

2000 ◽  
Author(s):  
A. Li ◽  
B. F. Armaly

Abstract Results from three-dimensional numerical simulation of laminar, buoyancy assisting, mixed convection airflow adjacent to a backward-facing step in a vertical rectangular duct are presented. The Reynolds number, and duct geometry were kept constant at Re = 200, AR = 8, ER = 2, and S = 1 cm. Heat flux at the wall downstream from the step was kept uniform, but its magnitude was varied to cover a Grashof number (Gr) range between 0.0 to 4000. All the other walls in the duct were kept at adiabatic condition. The flow, upstream of the step, is treated as fully developed and isothermal. The relatively small aspect ratio of the channel is selected specifically to focus on the developments of the three-dimensional mixed convection flow in the separated and reattached flow regions downstream from the step. The presented results focus on the effects of increasing the buoyancy force, by increasing the uniform wall heat flux, on the three-dimensional flow and heat transfer characteristics. The flow and thermal fields are symmetric about the duct’s centerline. Vortex generated near the sidewall, is the major contributor to the three dimensional behavior in the flow domain, and that feature increases as the Grashof number increases. Increasing the Grashof number results in an increase in the Nusselt number, the size of the secondary recirculating flow region, the size of the sidewall vortex, and the spanwise flow from the sidewall toward the center of the channel. On the other hand, the size of the primary reattachment region decreases with increasing the Grashof number. That region lifts away and partially detaches from the downstream wall at high Grashof number flow. The maximum Nusselt number occurs near the sidewalls and not at the center of the channel. The effects of the buoyancy force on the distributions of the three-velocity components, temperature, reattachment region, friction coefficient, and Nusselt number are presented, and compared with 2-D results.


Author(s):  
Sufianu A. Aliu ◽  
Richard O. Fagbenle

Simple and familiar perturbation parameters have been employed in applying the corrected Merk series of Chao and Fagbenle to the laminar mixed convection flow over two dimensional or axisymmetric bodies. The governing ordinary differential equations for the first five sets of the resulting universal functions for the velocity and temperature have been given. Numerical solutions were subsequently obtained and the relevant universal functions tabulated with respect to the ‘wedge parameter’ for mixed convection two dimensional flows and with respect to both the ‘wedge parameter’ and ‘shape parameter’ for the axisymmetric case. Using the wall derivatives of these universal functions, friction and heat transfer in mixed convection flows over two dimensional or axisymmetric bodies have been obtained and used in evaluation of skin friction and surface heat transfer.


Author(s):  
Kadeem Dennis ◽  
Kamran Siddiqui

Abstract The boundary layers are known to play key roles in many engineering systems. The hydrodynamic boundary layer found in these systems is often turbulent in nature and heat transfer is involved which further increases flow complexity due to the influence of buoyancy. One of the constituent layers of the turbulent boundary layer, the inner layer, has been established as home to key dynamical turbulent phenomena which can be influenced by the buoyant force. In the mixed convection flow regime, flow inertia and buoyant force are on the same order of magnitude. In this regime, buoyant thermals rising from the wall interact with the inertia-driven turbulent flow field resulting in highly complex three-dimensional flow dynamics. Past research studies conducted in this flow regime have been mostly computational in nature with little experimental work. The current knowledge on the impact of the relative contributions by the buoyant force and flow inertia on turbulent phenomena in the mixed convection flow regime is very limited. This study reports on an investigation into the turbulent flow phenomena present in mixed convection turbulent boundary layer flow over a heated smooth horizontal flat plate. Experiments were performed in a closed loop wind tunnel where the turbulent boundary layer was heated from below. The multi-plane particle image velocimetry (PIV) technique was used to capture two-dimensional velocity fields over two planes with respect to the flow direction. Experiments were conducted over a range of Richardson numbers (Ri) between 0.0 and 2.0 to control the relative contribution of the buoyant force with respect to flow inertia. The measured velocity fields are used to describe the influence of buoyancy on the three-dimensional turbulent boundary layer flow.


1981 ◽  
Vol 103 (2) ◽  
pp. 257-261 ◽  
Author(s):  
A. Moutsoglou ◽  
T. S. Chen ◽  
K. C. Cheng

The vortex instability of laminar, mixed-convection flow over an isothermal, horizontal flat plate is investigated analytically by the linear stability theory. In the analysis, the main flow and thermal fields are treated as non-parallel and the disturbances are assumed to have the form of a stationary longitudinal vortex roll that is periodic in the spanwise direction. Numerical results for the critical Grashof and Reynolds numbers that predict the first occurrence of the vortex rolls are obtained for fluids with Prandtl numbers of 0.7 and 7. It is found that the flow becomes more susceptible to vortex mode of instability as the buoyancy force increases. The present results are compared with available experimental data and also with analytical results from the wave mode of instability.


1995 ◽  
Vol 302 ◽  
pp. 91-115 ◽  
Author(s):  
Sergey A. Suslov ◽  
Samuel Paolucci

We have examined the linear stability of the fully developed mixed-convection flow in a differentially heated tall vertical channel under non-Boussinesq conditions. The Three-dimensional analysis of the stability problem was reduced to an equivalent two-dimensional one by the use of Squire's transformation. The resulting eigenvalue problem was solved using an integral Chebyshev pseudo-spectral method. Although Squire's theorem cannot be proved analytically, two-dimensional disturbances are found to be the most unstable in all cases. The influence of the non-Boussinesq effects on the stability was studied. We have investigated the dependence of the critical Grashof and Reynolds numbers on the temperature difference. The results show that four different modes of instability are possible, two of which are new and due entirely to non-Boussinesq effects.


Sign in / Sign up

Export Citation Format

Share Document