Convection in a long box driven by heating and cooling on the horizontal boundaries

1996 ◽  
Vol 310 ◽  
pp. 61-87 ◽  
Author(s):  
J. J. Sturman ◽  
G. N. Ivey ◽  
J. R. Taylor

Convection driven by spatially variable heat transfer across the water surface is an important transport mechanism in many geophysical applications. This flow is modelled in a rectangular tank with an aspect ratio, H/L, of 0.1 (where H and L are the tank height and length, respectively). Heat fluxes are applied through horizontal copper plates of length 0.1 L located at the top of one end of the tank and at the bottom of the other end. Experimental flows have been forced with heating at the bottom of the tank and cooling at the top, which gives rise to unstable convection in the end regions. Using water and a glycerol/water mix as the experimental fluids, flow visualization studies and measurements of temperature, velocity and heat flux have been made. Flow visualization studies revealed that complex unsteady turbulent flows occupied the end regions, while cubic velocity profiles characterized the horizontal laminar flow in the interior of the tank. Simple scaling arguments were developed for steady-state velocity and temperature fields, which are in good agreement with the experimental data. In the current experiments the portion of the plates closest to the tank interior (and to the tank endwall in the case of the glycerol/water experiments) were occupied by laminar boundary layers, while the remainder of the plates were occupied by turbulent flow. An effective Rayleigh number Ra* was defined, based upon the portion of the plate occupied by turbulent flow, as was a corresponding modified Nusselt number Nu*. The heat transfer was well predicted by classical Rayleigh-Bénard scaling with the Nusselt number Nu* ∼ Ra*1/3. The range of Ra* was 4.3 × 105 ≤ Ra* ≤ 1.7 × 108. Scaling arguments predicted the triple occupancy of the plates by differing boundary layer regimes within the range of 105 ≤ Ra* ≤ 1014.

2015 ◽  
Vol 764 ◽  
pp. 362-394 ◽  
Author(s):  
T. Dairay ◽  
V. Fortuné ◽  
E. Lamballais ◽  
L.-E. Brizzi

AbstractDirect numerical simulation (DNS) of an impinging jet flow with a nozzle-to-plate distance of two jet diameters and a Reynolds number of 10 000 is carried out at high spatial resolution using high-order numerical methods. The flow configuration is designed to enable the development of a fully turbulent regime with the appearance of a well-marked secondary maximum in the radial distribution of the mean heat transfer. The velocity and temperature statistics are validated with documented experiments. The DNS database is then analysed focusing on the role of unsteady processes to explain the spatial distribution of the heat transfer coefficient at the wall. A phenomenological scenario is proposed on the basis of instantaneous flow visualisations in order to explain the non-monotonic radial evolution of the Nusselt number in the stagnation region. This scenario is then assessed by analysing the wall temperature and the wall shear stress distributions and also through the use of conditional averaging of velocity and temperature fields. On one hand, the heat transfer is primarily driven by the large-scale toroidal primary and secondary vortices emitted periodically. On the other hand, these vortices are subjected to azimuthal distortions associated with the production of radially elongated structures at small scale. These distortions are responsible for the appearance of very high heat transfer zones organised as cold fluid spots on the heated wall. These cold spots are shaped by the radial structures through a filament propagation of the heat transfer. The analysis of probability density functions shows that these strong events are highly intermittent in time and space while contributing essentially to the secondary peak observed in the radial evolution of the Nusselt number.


2006 ◽  
Vol 129 (3) ◽  
pp. 769-777 ◽  
Author(s):  
Paul Lewis ◽  
Mike Wilson ◽  
Gary Lock ◽  
J. Michael Owen

This paper compares heat transfer measurements from a preswirl rotor–stator experiment with three-dimensional (3D) steady-state results from a commercial computational fluid dynamics (CFD) code. The measured distribution of Nusselt number on the rotor surface was obtained from a scaled model of a gas turbine rotor–stator system, where the flow structure is representative of that found in an engine. Computations were carried out using a coupled multigrid Reynolds-averaged Navier-Stokes (RANS) solver with a high Reynolds number k-ε∕k-ω turbulence model. Previous work has identified three parameters governing heat transfer: rotational Reynolds number (Reϕ), preswirl ratio (βp), and the turbulent flow parameter (λT). For this study rotational Reynolds numbers are in the range 0.8×106<Reϕ<1.2×106. The turbulent flow parameter and preswirl ratios varied between 0.12<λT<0.38 and 0.5<βp<1.5, which are comparable to values that occur in industrial gas turbines. Two performance parameters have been calculated: the adiabatic effectiveness for the system, Θb,ad, and the discharge coefficient for the receiver holes, CD. The computations show that, although Θb,ad increases monotonically as βp increases, there is a critical value of βp at which CD is a maximum. At high coolant flow rates, computations have predicted peaks in heat transfer at the radius of the preswirl nozzles. These were discovered during earlier experiments and are associated with the impingement of the preswirl flow on the rotor disk. At lower flow rates, the heat transfer is controlled by boundary-layer effects. The Nusselt number on the rotating disk increases as either Reϕ or λT increases, and is axisymmetric except in the region of the receiver holes, where significant two-dimensional variations are observed. The computed velocity field is used to explain the heat transfer distributions observed in the experiments. The regions of peak heat transfer around the receiver holes are a consequence of the route taken by the flow. Two routes have been identified: “direct,” whereby flow forms a stream tube between the inlet and outlet; and “indirect,” whereby flow mixes with the rotating core of fluid.


2007 ◽  
Vol 11 (4) ◽  
pp. 171-178
Author(s):  
Khalid Alammar

Using the standard k-e turbulence model, an incompressible, axisymmetric turbulent flow with a sudden expansion was simulated. Effect of Prandtl number on heat transfer characteristics downstream of the expansion was investigated. The simulation revealed circulation downstream of the expansion. A secondary circulation (corner eddy) was also predicted. Reattachment was predicted at approximately 10 step heights. Corresponding to Prandtl number of 7.0, a peak Nusselt number 13 times the fully-developed value was predicted. The ratio of peak to fully-developed Nusselt number was shown to decrease with decreasing Prandtl number. Location of maximum Nusselt number was insensitive to Prandtl number.


Author(s):  
Md. Islam ◽  
Z. Chong ◽  
S. Bojanampati

Various technologies have been developed to enhance flow mixing and heat transfer in order to develop an efficient compact heat exchanging devices. Vortex generators/turbulent promoters generate the vortices which reduce the boundary layer thickness and introduce the better mixing of the fluid to enhance the heat transfer. In this research experimental investigations have been carried out to study the effect of delta winglet vortex generator pairs on heat transfer and flow behavior. To generate longitudinal vortex flow, two pairs of the delta winglet vortex generators (DWVG) with the length of 10mm and winglet-pitch to tube-diameter ratio (PR = 4.8) are mounted on the inner wall of a circular tube. The DWVG pairs with two different winglet-height to tube-diameter ratios (Blockage ratio, BR = 0.1 and 0.2), three attack angles (α = 10°, 20°, 30°) and three spacings between leading edges (S = 10, 15 and 20mm) are studied. The experiments were conducted with DWVGs pairs for the air flow range of Reynolds numbers 5000–25000. The influence of the DWVGs on heat transfer and pressure drop was investigated in terms of the Nusselt number and friction factor. The experimental results indicate that DWVG pair in a tube results in a considerable enhancement in Nusselt number (Nu) with some pressure penalty. It is found that DWVG increases Nu up to 85% over the smooth tube. It is also observed that Nusselt number increases with Re, blockage ratio and attack angle. Friction factor decreases with Re but increases with blockage ratio, spacing and attack angle. And 30° DWVG pair with S = 20mm, BR = 0.2 gets the highest friction factor. The Highest thermal performance enhancement (TPE) was noticed for α = 10°, S = 20mm, BR = 0.2 for turbulent flows. To obtain qualitative information on the flow behavior and vortex structures, flow was visualized by laser sheet using smoke as a tracer supplied at the entrance of the test section. The generation and development of longitudinal vortices influenced by DWVG pairs were clearly observed.


2019 ◽  
Vol 29 (11) ◽  
pp. 4130-4141 ◽  
Author(s):  
Abdulmajeed Mohamad ◽  
Mikhail A. Sheremet ◽  
Jan Taler ◽  
Paweł Ocłoń

Purpose Natural convection in differentially heated enclosures has been extensively investigated due to its importance in many industrial applications and has been used as a benchmark solution for testing numerical schemes. However, most of the published works considered uniform heating and cooling of the vertical boundaries. This paper aims to examine non-uniform heating and cooling of the mentioned boundaries. The mentioned case is very common in many electronic cooling devices, thermal storage systems, energy managements in buildings, material processing, etc. Design/methodology/approach Four cases are considered, the left-hand wall’s temperature linearly decreases along the wall, while the right-hand wall’s temperature is kept at a constant, cold temperature. In the second case, the left-hand wall’s temperature linearly increases along the wall, while the right-hand wall’s temperature is kept a constant, cold temperature. The third case, the left-hand wall’s temperature linearly decreases along the wall, while the right-hand wall’s temperature linearly increases along the wall. In the fourth case, the left-hand and the right-hand walls’ temperatures decrease along the wall, symmetry condition. Hence, four scenarios of natural convection in enclosures were covered. Findings It has been found that the average Nusselt number of the mentioned cases is less than the average Nusselt number of the uniformly heated and cooled enclosure, which reflects the physics of the problem. The work quantifies the deficiency in the rate of the heat transfer. Interestingly one of the mentioned cases showed two counter-rotating horizontal circulations. Such a flow structure can be considered for passively, highly controlled mechanism for species mixing processes application. Originality/value Previous works assumed that the vertical boundary is subjected to a constant temperature or to a sinusoidal varying temperature. The subject of the work is to examine the effect of non-uniformly heating and/or cooling vertical boundaries on the rate of heat transfer and flow structure for natural convection in a square enclosure. The temperature either linearly increases or decreases along the vertical coordinate at the boundary. Four scenarios are explored.


2008 ◽  
Vol 130 (5) ◽  
Author(s):  
Abdullatif Ben-Nakhi ◽  
M. M. Eftekhari ◽  
D. I. Loveday

A computational study of steady, laminar, natural convective fluid flow in a partially open square enclosure with a highly conductive thin fin of arbitrary length attached to the hot wall at various levels is considered. The horizontal walls and the partially open vertical wall are adiabatic while the vertical wall facing the partial opening is isothermally hot. The current work investigates the flow modification due to the (a) attachment of a highly conductive thin fin of length equal to 20%, 35%, or 50% of the enclosure width, attached to the hot wall at different heights, and (b) variation of the size and height of the aperture located on the vertical wall facing the hot wall. Furthermore, the study examines the impact of Rayleigh number (104⩽Ra⩽107) and inclination of the enclosure. The problem is put into dimensionless formulation and solved numerically by means of the finite-volume method. The results show that the presence of the fin has counteracting effects on flow and temperature fields. These effects are dependent, in a complex way, on the fin level and length, aperture altitude and size, cavity inclination angle, and Rayleigh number. In general, Nusselt number is directly related to aperture altitude and size. However, after reaching a peak Nusselt number, Nusselt number may decrease slightly if the aperture’s size increases further. The impact of aperture altitude diminishes for large aperture sizes because the geometrical differences decrease. Furthermore, a longer fin causes higher rate of heat transfer to the fluid, although the equivalent finless cavity may have higher heat transfer rate. In general, the volumetric flow rate and the rate of heat loss from the hot surfaces are interrelated and are increasing functions of Rayleigh number. The relationship between Nusselt number and the inclination angle is nonlinear.


Author(s):  
Georgii Glebovich Yankov ◽  
Vladimir Kurganov ◽  
Yury Zeigarnik ◽  
Irina Maslakova

Abstract The review of numerical studies on supercritical pressure (SCP) coolants heat transfer and hydraulic resistance in turbulent flow in vertical round tubes based on Reynolds-averaged Navier-Stokes (RANS) equations and different models for turbulent viscosity is presented. The paper is the first part of the general analysis, the works based on using algebraic turbulence models of different complexity are considered in it. The main attention is paid to Petukhov-Medvetskaya and Popov et al. models. They were developed especially for simulating heat transfer in tubes of the coolants with significantly variable properties (droplet liquids, gases, SCP fluids) under heating and cooling conditions. These predictions were verified on the entire reliable experimental data base. It is shown that in the case of turbulent flow in vertical round tubes these models make it possible predicting heat transfer and hydraulic resistance characteristics of SCP flows that agree well with the existed reliable experimental data on normal and certain modes of deteriorated heat transfer, if significant influence of buoyancy and radical flow restructuring are absent. For the more complicated cases than a flow in round vertical tubes, as well as for the case of rather strong buoyancy effect, more sophisticated prediction techniques must be applied. The state-of-the-art of these methods and the problems of their application are considered in the Part II of the analysis.


1993 ◽  
Vol 115 (4) ◽  
pp. 890-896 ◽  
Author(s):  
R. M. Manglik ◽  
A. E. Bergles

Thermal-hydraulic design correlations are developed to predict isothermal f and Nu for in-tube, turbulent flows with twisted-tape inserts. Experimental data taken for water and ethylene glycol, with y = 3.0, 4.5, and 6.0, are analyzed, and various mechanisms attributed to twisted tapes are identified. Tube blockage and tape-induced vortex mixing are the dominant phenomena that result in increased heat transfer and pressure drop; for loose- to snug-fitting tapes, the fin effects are insignificant. The limiting case of a straight tape insert correlates with the hydraulic-diameter-based smooth tube equation. Tape twist effects are thus isolated by normalizing the data with the asymptotic predictions for y = ∞, and the swirl effects are found to correlate with Re and l/y. The validity of the final correlations is verified by comparing the predictions with previously published data, which include both gases and liquids, under heating and cooling conditions and a wide range of tape geometries, thereby establishing a very generalized applicability. Finally, correlations for laminar (presented in the companion Part I paper) and turbulent flows are combined into single, continuous equations. For isothermal f, the correlation describes most of the available data for laminar-transition-turbulent flows within ±10 percent. For Nu, however, a family of curves is needed due to the nonunique nature of laminar-turbulent transition.


1972 ◽  
Vol 14 (6) ◽  
pp. 393-399
Author(s):  
J. S. Coombs ◽  
S. D. Probert

Experimental determinations were made of the steady-state heat fluxes and velocity profiles in water between two vertical concentric cylinders, the heated inner cylinder being stationary while the outer cylinder was rotated in ambient temperature air. Secondary flows, due to end effects, existed in the annulus at all rotational speeds and profoundly influenced the rate of heat transfer across the annulus. When the circulation of the secondary flows opposed those due to natural convection, the mean Nusselt number decreased almost to unity.


2014 ◽  
Vol 348 ◽  
pp. 162-170
Author(s):  
Pey Shey Wu ◽  
Yi Hung Lin ◽  
Yue Hua Jhuo ◽  
Hsiao Ying Chan

Impingement heat transfer between a circular jet and a semi-spherical concave surface with or without coverage of porous material is investigated experimentally and numerically. For cases with coverage of the porous material on the target plate, a trapping hole for the jet fluid is fabricated. Measured local Nusselt number distributions along a meridian are documented. The flow and temperature fields at the conditions similar to that of experiments were computed with CFD software to support the experimental results and help to explain the physics. Varying parameters include Reynolds number, nozzle-to-plate distance, relative curvature, and a target surface with or without the covered porous material. Results show that the attachment of a porous material increases Nusselt number, with more influence at the stagnation zone than the far field. Increasing Reynolds number usually increases Nusselt number unless it is too high. Although an increase in the nozzle-to-plate distance decreases stagnation Nusselt number, the influence in heat transfer is small in the far field. The trapping-hole diameter should be the same as that of the jet diameter for best heat transfer enhancement.


Sign in / Sign up

Export Citation Format

Share Document