• Transactions of the Korean Society of Mechanical Engineers
• /
• v.16 no.6
• /
• pp.1195-1204
• /
• 1992

The natural convection from upward and downward facing horizontal isothermal plate immersed in water is studied numerically. The temperature of the plate is from 0.0 .deg. C to 8.0 .deg. C and the ambient water temperature is from 1.0 .deg. C to 10.0 .deg. C. Numerical results are presented for the velocity profiles, temperature profiles, local heat transfer coefficients, and average Nusselt numbers over the entire flow fields. Flow patterns are shown in the upward and downward facing surfaces at different ambient water temperatures. For the upward facing surface, there are upflow and unsteady flow. And the regions of the ambient water temperatures which give rise to the upflow are more extensive as the temperatures of the isothermal surface become more distant from the density extremum temperature. For the downward facing surface, only the downflow region is shown. For the upward facing horizontal isothermal surface, the average Nusselt number(= N $u_{1}$$^{*}$) is 28.86(Ra)$^{0.01}$. And for the downward facing surface, the average Nusselt number(= N $u_{2}$$^{*}$) is $C_{2}$(Ra)$^{0.2}$ and the values of $C_{2}$ are enlarged in the range of 0.785 .leq. $C_{2}$ .leq. 1.250 as increasing of the temperatures of the isothermal surface.ace.ace.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.12 no.3
• /
• pp.541-550
• /
• 1988

Natural convection in a square enclosure was investigated numerically for low prandtl number fluids using Finite Element Method. In case of Ra=10$^{4}$, 10$^{5}$ and 10$^{6}$ the temperature gradient decreases gradually at the lower end of the hot wall(or at the upper end of the cold wall) as prandtl number decreases in the range of 0.01 .leq. Pr .leq. 10. Maximum heat transfer occurs at a somewhat higher point from the lower end of hot wall(or at somewhat lower point from the upper end of the cold wall) and it draws near to the lower end of the hot wall(or draws near to the upper end of the cold wall) with increasing prandtl number. The flow in the enclosure appears as an Unicell Pattern for Ra .leq. 10$^{4}$ and secondarily flows(or tertiary flows) appears in the core region for Ra .geq. 10$^{5}$ . The line joining the center of secondary cells skewes in a clockwise direction as the Prandtl number decreases.

• Solar Energy
• /
• v.9 no.3
• /
• pp.44-54
• /
• 1989

Heat transfer phenomena during inward melting process of the phase change material were studied experimentally. N-docosane paraffin [$C_{22}H_{46}$] is used for phase change material and its melting temperature is $42.5^{\circ}C$. Experiments were performed for melting of an initially no-sub cooled or subcooled solid in a horizontal cylinder, in order to compare and investigate the radial temperature distribution, ratio of melting and melted mass, various energy components stored from the cylinder wall, figure of the melting front in the horizontal cylinder. The solid-liquid interface motion during phase change was recorded photographically. The experimental results reaffirmed the dominant role played by the conduction at early stage, by the natural convection at longer time during inward melting in the horizontal cylinder. Ratio of melting and melted mass are more influenced by wall temperature, rather than by the initial temperature of solid. The latent energy is the largest contributor to the total stored energy.

• International Journal of Air-Conditioning and Refrigeration
• /
• v.10 no.1
• /
• pp.31-39
• /
• 2002

Microencapsulated pcm (MPCM) particles are mixed with distilled water and utilized to evaluate its characteristics and performance as a thermal storage medium transporting heat. For the present study, tetradecane ($C_14$$H_30$, $T_m$=5.5$^{\circ}C$) is capsulated in the core, coated with the melamine for their surface. The size of particles is well-controlled under 10$\mu$m in the process of in-situ polymerization with melamine-formaldehyde resin. For the experiment, the concentractions of slurries are prepared for 20 wt%, 30 wt%, and 40 wt%. The results are compared with those of water and 100% tetradecane oil. The pure water and tetradecane start solidifying within 20 minutes after introducing cooling water into the thermal storage tank whose flow rates are varied by 125 cc/min, 250 cc/min, and 500 cc/min. However, MPCM slurries are required relatively longer period of time for their phase change than pure phase change materials. That is, the entrained MPCM particles restrict their heat transfer in terms of natural convection and conduction to them.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.13 no.2
• /
• pp.80-87
• /
• 2001

Microencapsulated PCM particles are mixed with distilled water and utilized to evaluate its characteristics and performance as a thermal storage medium transporting heat. For the present study, tetradecane(C$_14H_30, T_m=5.5^{\circ}C$) is capsulated in the core with the melamine of its surface. The size of particles is well-controlled under 10${\mu}{\textrm}{m}$ in the way of in-situ polymerization with melamine-formaldehyde resin. For the experiment, the concentrations of slurries are prepared for 20wt%, 30wt%, and 40wt%. The results are compared with those of water and 100% tetradecane oil. The pure water and tetradecane start solidifying within 20 minutes after introducing cooling water into the thermal storage tank whose tank whose flow rates are varied by 125cc/min, 250cc/min, and 500cc/min. However, MicroPCM slurries are required relatively longer period of time for their phase change than pure phase change materials. That is, the entrained MicroPCM particles control its heat transfer in terms of natural convection and conducting to them.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.17 no.3
• /
• pp.223-231
• /
• 2005

When test tubes for PCM with melting point lower than a room temperature are installed vertically as the T-history method proposes, there exists a temperature distribution in the longitudinal direction by natural convection, which lowers the precision of measured heat-of-fusion. The purpose of the present work is to improve the precision by arranging the test tubes horizontally, while maintaining the simplicity and convenience. Assuming that the amount of heat transfer is very small through the vapor space formed in the upper part of the tubes by volumetric change, the obtained value by the T-history method using the latent heat period bounded by two inflection points is in good agreement with that of the literature. Also, the scattering of measured values by the proposed method decreases markedly compared to that of vertical positioning.

• Journal of Energy Engineering
• /
• v.24 no.1
• /
• pp.78-87
• /
• 2015

Focusing effect of a metallic layer in a severe accident depending on the aspect ratios and cooling conditions of top plate and side wall was investigated. Experiments were carried out for Rayleigh numbers and aspect ratio in the range of $8.49{\times}10^7{\sim}5.43{\times}10^9$, 0.135~0.541 respectively. In order to achieve high Rayleigh numbers, the heat transfer experiments were replaced by mass transfer experiments based on the heat and mass transfer analogy. A sulfuric acid-copper sulfate ($H_2SO4-CuSO_4$) electroplating system was adopted as the mass transfer system. The experimental results agreed well with the Rayleigh-Benard natural convection correlations of Dropkin and Somerscales and Globe and Dropkin. When compared with the standard Rayleigh-Benard problem, the cooling by the side wall is even higher than the top. For a shorter height, the interaction between the heated and cooled plumes increases due to decrease of the height. Thus, the heat transfer increases.

• Journal of Mechanical Science and Technology
• /
• v.18 no.2
• /
• pp.294-301
• /
• 2004

The natural convective heat transfer in a rectangular enclosure with a heating source has been studied by experiment and numerical analysis. The governing equations were solved by a finite volume method, a SIMPLE algorithm was adopted to solve a pressure term. The parameters for the numerical study are positions and surface temperatures of a heating source i.e., Y /H =0.25, 0.5, 0.75 and 11$^{\circ}C$ $\leq$ΔT$\leq$59$^{\circ}C$. The results of isotherms and velocity vectors have been represented, and the numerical results showed a good agreement with experimental values. Based on the numerical results, the mean Nusselt number of the rectangular enclosure wall could be expressed as a function of Grashof number.

• Nuclear Engineering and Technology
• /
• v.26 no.4
• /
• pp.516-525
• /
• 1994

Since the TMI and Chernobyl accidents, many passive safety features are suggested in advanced reactors in order to enhance the safety in future nuclear power plants. In order to verify the effectiveness and provide the data for detailed design of passive cooling system, in the present work, the effects of air inlet position and external condition on the natural circulated air flow rate and the natural and forced convective heat transfer coefficient have been investigated for the one-side heated closed path such as the passive containment cooling system of the Westinghouse's AP-600. A series of experiments have been peformed with the 1/26th scaled segment type test facility of the AP-600 passive containment. Under natural and forced convection, the air velocities and temperatures are measured at several points of the air flow path. The experimental result are compared with a simple one-dimensional model and it shows a good agreement.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.35 no.1
• /
• pp.33-42
• /
• 2011

In this study, we investigated the natural convection on the outer surface of a vertical pipe by performing mass transfer experiments using fluids with high Pr number using the concept of analogy between heat and mass transfer. A cupric acid-copper sulfate electroplating system was adopted as the mass transfer system. Tests were performed for $Ra_H$ numbers from $1.4{\times}10^9$ to $4{\times}10^{13}$, Pr numbers from 2,094 to 4,173, and diameters from 0.005 m to 0.035 m. The test results for laminar flow conditions were in good agreement with the correlations reported by King, Jakob and Linke, McAdam, and Bottemanne, and those for turbulent conditions with the correlations presented by Fouad for a vertical plate and also proved the dependence on Pr numbers. The obtained correlations were $Nu_H=0.55Ra^{0.25}_H$ for laminar and $Nu_H=0.12Ra^{0.28}_HPr^{0.1}$ for turbulent. The transition between laminar and turbulent occurs at $Ra_H$ of about $10^{12}$.