• Journal of Mechanical Science and Technology
• /
• v.17 no.9
• /
• pp.1388-1396
• /
• 2003

The hydrodynamic instability of the three-dimensional boundary layer on a rotating disk introduces a periodic modulation of the mean flow in the form of stationary cross flow vortices. The instability labeled Type II by Faller occurs first at lower Reynolds number than that of well known Type I instability. Detailed numerical values of the amplification rates, neutral curves and other characteristics of the two instabilities have been calculated over a wide range of parameters. Presented are the neutral stability results concerning the two instability modes by solving the appropriate linear stability equations reformulated not only by considering whole convective terms but also by correcting some errors in the previous stability equations. The present stability results agree with the previously known ones within reasonable limit. Consequently, the flow is found to be always stable for a disturbance whose dimensionless wave number is greater than 0.75. Some spatial amplification contours have been computed for the stationary disturbance wave, whose azimuth angle $\varepsilon$= 11.29$^{\circ}$ to 15$^{\circ}$ and for the moving disturbance wave, whose azimuth angle $\varepsilon$ = 12.5$^{\circ}$ to 15$^{\circ}$. Also, some temporal amplification contours have been computed for the stationary disturbance wave, whose azimuth angle $\varepsilon$= 11.29$^{\circ}$ to 15$^{\circ}$ and for the moving disturbance wave, whose azimuth angle $\varepsilon$= 12$^{\circ}$ to 15$^{\circ}$. The flow instability was observed by using a white titanium tetrachloride gas over rotating disk system. When the numerical results are compared to the present experimental data, the numerical results agree quantitatively, indicating the existence of the selective frequency mechanism.

• Applied Chemistry for Engineering
• /
• v.22 no.6
• /
• pp.703-710
• /
• 2011

As the spacer in the membrane module provide the channel space to flow the feed solution smoothly and induce the flow turbulence, it could help to reduce both the concentration polarization and to take the long-term operation of membrane modules with high permeate flux by mixing the accumulated contaminants on the membrane surface into the bulk solution. In this study, the concentration distribution in membrane module with respect to the spacers which have the cross-sectional shapes of circle, cross, diamond and hexagon, the angles of spacer configuration, solute rejection and permeate flux were interpreted and optimized numerically using the "COMSOL Multiphysics" software. The concentration on the membrane surface was kept the lowest level for the cross-shape among the above four types of spacers. Also the 30 degree spacer configuration was showed as the most efficient case. The concentrations on the membrane surface at the module outlet for without spacer and the cross shape with the 30 degree spacer configuration were 2.09 and 1.29 times higher than those at inlet, respectively. The reduction effect of concentration polarization increased rapidly as the permeate flux increased.

• The KSFM Journal of Fluid Machinery
• /
• v.19 no.1
• /
• pp.18-23
• /
• 2016

A HVAC(Heating Ventilation and Air Conditioning) is adapted to increase the comfort of the cabin environment for train. The train HVAC duct system has very long duct and many outlets due to the shape of a train set. the duct cross section shape is limited by a roof structure and equipments. Therefore, the pressure distribution and flow uniformity is an important performance indicator for the duct system. In this study, the existing blow down type HVAC duct system for a train was supplemented to improve the flow uniformity by applying a design method combining design of experiment (DOE) with numerical analysis. The design variables and the test sets were selected and the performance for each test set was evaluated using CFD(Computational Fluid Dynamics). The influence of each design variable on the system performance was analysed based on the results of the performance evaluation on the test sets. Furthermore, the optimized model, whose the flow uniformity was improved was produced using the direct optimization(gradient-based method). Finally, the performance of the optimized model was evaluated using numerical analysis, and it was confirmed that its flow uniformity has indeed improved.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.8 no.2
• /
• pp.267-278
• /
• 1996

Steady, laminar, forced convection flow and heat transfer in the entrance region of an internally finned circular duct with a finite thermal conductivity has been analyzed numerically. The problem under investigation is a three-dimensional boundary layer problem, and is solved by employing a marching-type procedure which involves solution of a series of 2-dimensional elliptic problems in the cross-stream plane. Two types of inlet hydrodynamic conditions are considered : (a) uniform velocity flow and (b) fully developed flow. From the above inlet conditions, the effects of the fin height(h), fin number(N) and conductivity ratio($k_r$) on the flow and thermal characteristics are investigated. The numerical results show that the height and number of fins, and ratio of the solid to fluid thermal conductivity have pronounced effect on the solution. Considering pressure drop, optimized dimensionless fin height is 0.4.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.17 no.2
• /
• pp.101-109
• /
• 2005

The purpose of this study is to investigate the performance of outdoor heat exchanger for heat pump using carbon dioxide. Two types of fin and tube heat exchangers (2 rows for type A and 3 rows for B) are tested. Both heat exchangers have counter-cross flow and 1-circuit arrangement. Test results such as heat transfer rate, pressure drop characteristics and temperature distribution in the heat exchanger are shown with respect to mass flow rate of refrigerant and frontal air velocity For cooling mode, the minimum temperature difference between air and refrigerant of type B is smaller than that of type A by $1^{circ}C$, but the pressure loss of air side is much higher for type B by $29\%$. It is found that a large temperature gradient of carbon dioxide during gas cooling Process Promotes thermal conduction through tube wall and fins which results in degradation of heat transfer performance. For heating mode operation, type B heat exchanger shows higher heat transfer performance compared to type A. However, because pressure loss of refrigerant side of type B is much greater than that of type A, the refrigerant outlet pressure of type B becomes lower than that of type A.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.14 no.1
• /
• pp.2557-2569
• /
• 1972

Hydraulic computations to determine the elevation of canal bottom and mater surface for box type concrete culverts are discussed. Velocity and cross sectional area of flow are computed from Manning's formula. Aad then head loss and velocity head are considered to determine the elevation of bottom and water surface. For stress analysis, 13.5 ton live load and earth pressure are considered. Also longitudinal stress of box culverts is checked.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.11 no.4
• /
• pp.448-456
• /
• 1999

Variations of thickness and effective thermal conductivity of frost forming on the horizontal] cylinder with respect to time were measured under cross flow. The local heat flux around the cylinder was determined by measuring the radial temperature distribution in the cylinder having small holes drilled axially in which T-type thermocouples were inserted, then by using one dimensional cylindrical heat conduction equation. The thickness and the surface temperature of the frost layer around the cylinder were measured periodically while developing the frost. Each experiment was peformed by varying the Reynolds number, the temperature, and the humidity condition. Specially the dew point temperature of the most cases was below the freezing point. Experimental data showed that the frost layers on the front and the rear surface were thicker than those on the top and the bottom one which was near the separation point. The thickness and effective thermal conductivity of the frost layer were affected by inlet air velocity, temperature, and humidity. Moreover, the effective thermal conductivity and the effective thermal resistance increase with respect to time.

• Membrane and Water Treatment
• /
• v.12 no.1
• /
• pp.23-35
• /
• 2021

A POM-membrane was fabricated by immobilizing a keggin type polyoxometalate (POM) H5PV2Mo10O40 onto the surface of microporous flat-sheet polymeric polyvinylidene fluoride (PVFD) membrane using a chemical deposition method. The POM-membrane was characterized by FT-IR, SEM and EDX to confirm existing of the POM onto the membrane surface. The POM-membrane was used to remove an anionic textile dye (Reactive Black 5 named as an RB5) from aqueous phases with a cross-flow membrane filtration and a batch adsorption system. The dye removal efficiency of the POM-membrane using the cross-flow membrane filtration system and the batch adsorption system was about 88% and 98%, respectively. The influence factors such as contact time, adsorbent dosage, pH, and initial dye concentration were investigated to understand the adsorption mechanism of the RB5 dye onto the POM-membrane. To find the best fitting isotherm model, Langmuir, Freundlich, BET and Harkins-Jura isotherm models were used to analyze the experimental data. The isotherm analysis showed that the Langmuir isotherm model was found to the best fit for the adsorption data (R2 = 0.9982, qmax = 24.87 mg/g). Also, adsorption kinetic models showed the pseudo second order kinetic model was found the best model to fit the experimental data (R2 = 0.9989, q = 8.29 mg/g, C0 = 15 ppm). Moreover, after four times regeneration with HNO3 acid, the POM-membrane showed high regenerability without losing dye adsorption capacity.

• Journal of Biosystems Engineering
• /
• v.34 no.5
• /
• pp.358-362
• /
• 2009

This study was conducted to evaluate the performance of a 12 tons capacity circulating concurrent flow rice dryer. An actual scale dryer with the capacity of 12 tons was developed to obtain a faster drying rate of 0.8~1.2%(w.b.)/h, while maintaining a lower drying energy consumption of 5,000 kJ/kg-water and achieving a drying quality that was comparable to a conventional cross-flow rice dryer. The Test-1 was conducted at $110^{\circ}C\;-\;20\;cmm/m^2$ and the Test-2 was conducted at $120-110-100-90^{\circ}C\;-\;20\;cmm/m^2$ under the same conditions as Test-1. In Test-1, the drying rate, drying energy consumption and crack ratio were 0.98 %(w.b.)/h, 4,573 kJ/kg-water and 3.2%, respectively. In Test-2, the drying rate, drying energy consumption and crack ratio were 0.74 %(w.b.)/h, 4,790 kJ/kg-water and 4.0%, respectively. The results of these tests demonstrated that this concurrent-flow dryer reached the desired drying rate, drying energy consumption and crack ratio.

• Journal of computational fluids engineering
• /
• v.19 no.1
• /
• pp.47-56
• /
• 2014

Since a typical plate heat exchanger is made up of a huge number of unitary cells, it may be impossible to predict the aero-thermal performance of the full scale heat exchanger through three-dimensional numerical simulation due to the enormous amount of computing resources and time required. In the present study, a simple flow-network model using the friction factor correlation and a thermal-network model based on the effectiveness-number of transfer units (${\varepsilon}$-NTU) method has been developed. The complicated flow pattern inside the cross-corrugated heat exchanger has been modeled into flow and thermal networks. Using this model, the heat transfer between neighboring streams can be considered, and the pressure drop and the heat transfer rate of full-scale heat exchanger matrix are calculated. In the calculation, the aero-thermal performance of each unitary cell of the heat exchanger matrix was evaluated using correlations of the Fanning friction factor f and the Nusselt number Nu, which were calculated by unitary-cell CFD model.