The effect of a nonuniform magnetic field on bubble behavior in a pipe flow of magnetic fluid is numerically analyzed in order to realize effective boiling and high performance in a new energy conversion system using boiling two-phase flow of magnetic fluid. The governing equations of a single bubble in Poiseuille flow of magnetic fluid are presented and the translational motion and growth process of bubble are numerically calculated considering the effect of various additional forces which act on the bubble. It is found that the bubble velocity in a flow of magnetic fluid is strongly influenced by the magnetic field strength due to the effect of magnetic body force. Also, the effect of various additional forces which influence bubble behavior is accurately estimated. These fundamental studies show that precise control of bubble behavior is possible by effective use of the magnetic force acting on the fluid.

View full abstract

A fully three-dimensional numerical procedure based on the two-fluid model is proposed for the prediction of developing turbulent bubbly fllow in a rotating square-sectioned duct. The single-phase nonlinear k-ε modelis extended to the two-phase flows. This model yields normal Reynolds stress anisotropies which allow for a more accurate prediction of the secondary fllow and phase distribution in rotating square-sectioned ducts. The primary and secondary flows are predicted well for single phase flow. The numerical results obtained for bubbly flow show trends similar to the measured results for the rotating ducts.

View full abstract

In order to investigate the complex behavior of collapsing bubbles in cloud cavitation, numerical simulations have been performed by solving the 2-D governing equations for gas-liquid two-phase media. Here, the equation of state for such media is derived from the locally homogeneous assumption, so that the apparent compressibility can be taken into account. The interface of a bubble or a cloud of bubbles is treated as a contact discontinuity surface, where the values of density and void fraction are allowed to jump. When the void fraction in a bubble cloud is almost 100%, the collapsing behavior agrees well with that of a single bubble. Numerical results of the interaction between some bubbles and an incident shock wave show qualitative agreement with experimental results. Therefore, complex interaction phenomena between multibubbles can be simulated when some small bubbles are arranged in a bubble cloud by distributing the corresponding void fractions nonuniformly.

View full abstract

Laser Doppler velocimeter measurements were made to investigate bubbling jet characteristics in an aeration tank at a pressure of 200kPa. The data were compared with previous measurements at atmospheric and reduced pressures. Bubble frequencies at the nozzle outlet were correlated with the mass flow rate of gas rather than the volumetric flow rate. In the far field where the buoyancy force of bubbles prevails, the axial and radial distributions of the mean velocity components, the r. m. s. values of turbulence components, the Reynolds shear stress and the skewness and flatness factors of the turbulence components obtained at an elevated pressure agreed well with those obtained at the atmospheric pressure for the same volumetric gas flow rate. Consequently, the liquid flow characteristics including the turbulence structure in the far field are not influenced by an increase in surface pressure as long as the volumetric gas flow rate is the same.

View full abstract

Particle trajectories in a typical rotating flow of liquid were calculated for a single-sphere with density lower than that of the liquid, the initial flow velocity being zero. A general form of momentum equation is employed, which consists of a buoyant force, a drag force, a lift force, an added-mass force, a pressure gradient force and a Basset history force. The effects of the lift force acting on the spherical particle were shown by the computation reqults of the particle trajectories. The calculated results of the dynamic equilibrium radius of the particle fit the experimental results.

View full abstract

In this paper, a numerical simulation for the axial and swirling sas-solid flows in a vertical pipe is carried out with an Eulerian approach for the gas and a stochastic Lagrangian approach for particles, where particle-particle and particle-wall collisions are taken into consideration. The k-ε turbulence model is used to characterize the time and length scales of the gas-phase turbulence. Models predicting the particle source and additional pressure loss are used. Numerical results are presented for polyethylene pellets of 3.2mm diameter conveyed through a pipeline of 12m height with an inner diameter of 80mm, solid mass flow rates of 0.03kg/s and 0.084kg/s, and gas velocity varying from 11m/s to 18m/s. The axial and radial distribution of particles, the particle concentration, the particle velocity, gas velocity, turbulent kinetic energy and turbulent energy dissipation rate are obtained. The numerical results agree with the experimental data.

View full abstract

Experiments were performed on the flow past a circular cylinder in silica dioxide suspensions with C_w=0.5∼1.0wt% and in carbon black suspensions with C_w=0.3∼1.0wt%. Two cylinders with diameters between 10mm and 20mm located in a horizontal duct were tested in terms of the pressure distribution. It was shown that the separation point shifts downstream although the Strouhal number is almost the same as that of tap water. Consquently, from the result of an integration along the pressure distribution curve it is clarified that the drag reduction phenomenon in water/fine solid particle suspension is caused by the shift of the separation point in a certain Reynolds number range.

View full abstract

In this study, a two-dimensional solid-liquid two-phase flow past a circular cylinder was simulated using the boundary element method combined with the vortex method. As a result of calculation, the flow pattern and the motion of particles were shown. It was found that the motion of particles did not follow the fluid movement in the Karman vortex street. In the cases of mixing ratios 0.25%, 1.0% and 1.5%, there is no difference in the Strouhal number of vortex shedding between the two-phase flow and the single-phase flow. In the case of mixing ratio 2.3%, two patterns of vortex shedding, that is Karman vortex type and twin vortex type, were shown, and the value of the Strouhal number was approximately 40% lower than the Strouhal number in the single-phase flow. The spin speed and size of the particle had an effect on vortex shedding.

View full abstract

We have measured the velocity of an accelerating sphere as it falls in water/fine-solid-particle suspensions by means of four ultrasonic transducers. The fluids tested were silicon dioxide suspensions and carbon black suspensions. The concentrations of silicon dioxide suspensions were 0.3, 0.5 and 1.0wt% and those of carbon black suspensions were 0.3 and 0.5wt%. Four spheres between 57.1 and 152.4mm in diameter were tested. From the experimental results, it was shown that although the fall velocity of a sphere in suspensions is higher than that in tap water within a certain range of Reynolds numbers, drag reduction does not occur above the critical Reynolds number.

View full abstract

Experimental studies are carried out to clarify the effect of nozzle shape on the structure and drilling capability of a premixed abrasive water jet in air. Flow visualization and erosion results are obtained with the use of five different nozzles: four conical nozzles with different length focus sections and a quarter circle entrance nozzle with a focus section. Erosion tests are conducted on stainless steel specimens at an injection pressure of 11.9MPa. Aluminum oxides with different grain sizes, #100 and #220, are used as abrasives. The structures of the jets are observed using instantaneous photographs. The optimum nozzle shape is determined through erosion tests and observation of the jets.

View full abstract

A generalized Langevin model described in the cylindrical coordinate system on the basis of the consistency condition up to the second-order moments of velocity has been applied to simulate the plume diffusion field of a high Schmidt number matter from a wall point source in a fully developed horizontal turbulent pipe flow. The plane source of the admixture particles with the number density distribution specified from the experimental data of the mean concentration is set up at some position downstream from the wall point source to avoid the disturbance caused by the injection of matter near the point source, since the present model assumes that the admixture particles have no influence on the fluid motion. From the simulation results, it is found that the simulated vertical and azimuthal profiles of the mean concentration at various downstream cross sections show good agreement with experimental data. The mean streamwise variations of the maximum mean concentration, and the half-widths of the vertical and azimuthal profiles of the mean concentration are also accurately predicted by the present model.

View full abstract

Transonic two-phase flows around an airfoil in moist air are numerically investigated. The fundamental equations are the compressible Navier-Stokes equations and model equations of the homogeneous nucleation and the droplet growth by the classical condensation theory. The fourth-order compact MUSCL TVD scheme is used to obtain not only the shock but also a weak condensation shock clearly. The speed of sound and the total energy for moist air are estimated by the homogeneous assumption and the thermodynamic relations. Finally, the affects of the condensation shock and non-equilibrium phase change on airfoil performance are discussed.

View full abstract

Analysis of gas transport in oscillatory flow though an airway model with complex geometry such as a single bifurcation, relies on the theoretical understanding of gas exchange during High-frequency oscillation; HFO. Our purpose was to develop a method for measuring the mean effective diffusivity in an oscillatory flow between two points, by applying Laplace integration, and the Laplace parameter s are related to the accuracy of th measured effective diffusivity, D_eff. In this study, numerical analysis was carried out to investigate the Laplace Transform Method. We defined the parameter K required to obtain optimum s, and found that K≃0.1. Then we applied this method to evaluate D_eff in a straight tube and a single symmetrical bifurcation. The theoretical results for a straight tube agreed a factor of Dn²/α⁴, that proved the effect of the secondary flow.

View full abstract

Flow structures of wakes near two circular cylinders forming a cross shape in contact each other in a uniform flow are studied from the mean velocity and turbulent intensity profiles, and flow visualizations by tuft-stick and smoke-wire methods. In the wake behind the cross part of the two cylinders, there is a characteristic region in which the mean velocity is high and the turbulence intensity is low. According to the flow visualization results, the high velocity and low turbulence region is due to jet bleeding from the neighborhood of a contact point of two cylinders. In addition, the bleeding jet produces secondary flows which go from the the uniform flow region toward the bleeding jet region, and go away from the jet region along each cylinder axis. Then, those secondary flows form eight large scale longitudinal vortices around the wake center. Therefore, the mean velocity becomes high and the turbulent intensity becomes low in the wake center region, down-stream of about x/d=10, as compared with those in the quasi two-dimensional region out of the center region.

View full abstract

We have presented the effects of a ground plate (stationary or moving) on the Magnus effect of a rotating cylinder in the first report and showed that the lift suddenly changed at a certain H/D, and the phenomenon was related to a change in the flow pattern. Details of the phenomenon that is unstable and hysteretic are studied experimentally. To determine the critical points of the hysteretic phenomenon appearing with the changing in flow patterns with α, the velocity vector is measured by a hot-wire anemometer for the clearance H/D and velocity ratio α. As a result, it is found that two different flow patterns exist in a certain area enclosed by H/D and α. We call this area a "bistable area". The influence of Reynolds number on the bistable area is also discussed.

View full abstract

An application of a neural network to jets with self-induced oscillation exhausting into a two-dimensional channel is presented. No established method nor suitable parameter to control flip-flop jets is known. One new approach for neural networks is a recurrent network. We find that if the flip-flop is periodic, the recurrent network can be useful to reproduce the original periodic orbit obtained by the time delay method. Another application of the recurrent network employed the flow pattern from velocity fields of the neural network. In this case, the results depend on the number of reference points and on the number of teaching patterns in computed fields given. A promising result is obtained for the control of flip-flop jets using a recurrent network.

View full abstract

The apparent elongational viscosity of concentrated polymer solutions was measured using a spinline rheometer; 0.1wt% and 1.0wt% solutions of polyacrylamide (PAA) showed a remarkable stretch-thickening property in elongational viscosity. Furthermore, numerical simulation of the elongational flow in the spinline rheometer was carried out using the modified Giesekus model. The apparent elongational viscosity was examined for various values of material constants in the modified Giesekus model including the values for the solutions of PAA and compared with the experimental results. It was found that the gravity force significantly affected the flow in the spinline and the apparent elongational viscosity in the simulation of the solutions of PAA.

View full abstract

In the previous paper, three existing extrapolation methods and the newly developed one, LWE, were discussed. Furthermore ROLE, one of the existing extrapolation methods, and LWE were applied to numerical analysis of Poisson's equation and their suitability for massively parallel computing was shown. In this paper, we clarify mathematically that the extrapolation methods described in the previous paper can be applied to the iterative solution of a nonlinear equation. We select the coupled method as a numerical algorithm for incompressible flow because it is simple and its procedures can be efficiently parallelized. We combine the coupled method and the extrapolation method, and create a parallel code which is appropriate for massively parallel computers. Our code is ported on AP1000 and its performance is discussed.

View full abstract

In this paper we propose a new thermal conductivity measurement of a two-dimensional anisotropic heat conductor using an unsteady point source of heat. This method improves upon the principal thermal conductivity measurement of a steady point heat source and sink which was proposed by us in our previous report. This new method has two merits. First, it does not use a heat sink of the controlled cooling system. Second, it simplifies the calculation of two principal thermal conductivities and the principal rotating angle from the measured coordinates. To show the validity of this method, we have measured the two principal thermal conductivities and the principal rotating angle of the high-T_c superconductor Er₁Ba₂Cu₃O_7-y. The zero-resistance and onset temperature of this sample were 95.9K and 102K, respectively. In the superconductive transition range, both principal thermal conductivities had the peak values with decreasing temperature. Further, the principal rotating angle changed greatly from the normal conducting state to the superconductive state. The temperature dependences of the two thermal conductivities of this sample are similar to the Y₁Ba₂Cu₃O_7-y compound system which was reported in the results of our previous experiment.

View full abstract

This work was undertaken experimentally and analytically to acquire a better understanding of condensation heat transfer of binary vapors of immiscible liquids on horizontal tube banks. The characteristics of condensation heat transfer on lower tubes, which received condensates inundated at the upper tube banks, were clarified through detailed experiments using an azeotropic mixture of carbon tetrachloride and water as binary vapors. Due to the combination of the organic immutable behavior and the extreme behavior of water droplets, standing droplets and continuously detaching droplets on the surface, a model of condensation was developed taking into account the sweeping effect and the inundation effect. The predictions obtained using by this model agree well with the experimental results.

View full abstract

A direct numerical simulation has been carried out to investigate viscous flow in an anisotropic porous medium. A collection of square rods placed in an infinite two-dimensional space has been proposed as a numerical model of microscopic porous structure. The degree of anisotropy was varied by dislocating alternate rows of square rods. Extensive calculations were carried out for various sets of the macroscopic flow angle, Reynolds number, porosity and degree of anisotropy. The numerical results thus obtained were integrated over space to determine the directional permeability from a purely theoretical. The predicted directional permeability closely follows the formula proposed by Scheidegger.

View full abstract

A simple model for the heat transfer and pressure drop characteristics in a layer of spherical particles is proposed in the present study. The layer of spherical particles bounded by two parallel boundary walls is considered to consist in two regions, one of which is the near-wall region of particles within a half-diameter of the spherical particle from the wall, and the other is the core region of particles over a half-diameter of the spherical particle away from the wall. The characteristics of the near-wall region, such as the permeability, the Forchheimer coefficient, and the thermal dispersion coefficient, are determined based on the experimental data for a one-stage spherical particle layer. For the core region, the permeability and the Forchheimer coefficient are evaluated using previous correlations for a homogeneous spherical particle layer, while the thermal dispersion coefficient is modified from that given by the previous correlation for a homogeneous spherical particle layer to include the effect of the thermal conductivity ratio of the spherical particle and the fluid. The effective thermal conductivity of the layer of spherical particles is assumed to be constant in the present model. The validity of the model is proved by comparing the results of this analysis with experimental results under various conditions.

View full abstract

The linearity principle of scalar transport, proposed by Pope (Phys. Fluids 1983), has long been ignored in existing models. An examination of existing scalar transport models indicates that some of the terms employed do not satisfy the principle. We propose a new model which satisfies the linearity principle and takes into account the effect of the Prandtl number. The model is applied to scalar transport in a turbulent channel flow for Pr=0.71 and 0.025. The results obtained are in good agreement with DN S data.

View full abstract

The effect of ultrasonic vibration on heat transfer during rapid quenching of 0.3 and 0.5mm dia horizontal platinum wires was studied experimentally using water and ethanol as test liquids. The frequencies of ultrasonic vibration ν were 24 and 44kHz. The power input to the transducer P ranged from 10 to 280W. The falling velocity of the wire was set at 0.6 and 1.0m/s. The initial wire temperature ranged from 500 to 1200K and the liquid subcooling from 40 to 70K. The wall superheat at the first minimum heat flux point (M1 point) initially increased with increasing P and approached a constant value at high P. The heat flux after the M1 point increased considerably due to the application of ultrasonic vibration. The effect of ultrasonic vibration was more significant for ν=24kHz at high subcooling. In order to understand the mechanism of heat transfer enhancement, the distributions of sound pressure and cavitation intensity in the liquid pool were also measured. The heat transfer enhancement showed a close correlation with the effective value of sound pressure.

View full abstract

In order to determine the heat transfer characteristics of mist cooling in the range of small degree of superheating, detailed experiments have been conducted on mist cooling heat transfer from a circular superheated surface. The experimental results indicate that the heat flux curve obtained is a function of parameters such as sprayed mass flow rate and degree of superheating, in relation to the boiling and evaporation behavior of the liquid film formed on the heated surface. The effects of size, velocity and temperature of spray droplets on heat flux are also clarified. The mechanism of mist cooling heat transfer is discussed and theoretical and experimental results are compared.

View full abstract

Transpiration cooling is known as one of the most effective cooling methods. Most papers on transpiration cooling dealt with the coolant of a plate heated by convection. In a very hot environment which requires transpiration cooling, however, the heat flux by radiation plays an important role. Very few papers focussing on this problem have been published. It has been pointed out that a liquid coolant is more efficient than a gaseous coolant because of its latent heat. In the present study, distilled water was used to cool a flat plate heated by radiation. The experimental results on transpiration cooling were compared with the theoretical ones and good agreement was found.

View full abstract

To improve the performance of ice energy storage systems, the problem of freezing on the cooling surface must be addressed. One of the authors has proposed a new ice-making method in which a magnetic fluid prevents water from freezing on the cooling surface. Thus, we investigated experimentally the effect of shape control of magnetic fluid using a magnet on the supercooling phenomenon. The changes of the degree of supercooling and the heat flux with time are shown for several cases. The shape changes and motion of magnetic fluid, which are controlled by controlling the magnetic field, have a slight effect on the supercooling phenomenon. However, exposure of the cooling surface covered by a film of magnetic fluid reduces the critical degree of supercooling.

View full abstract

It is important that Al films fill contact holes on substrates in order to produce high-density devices. In this paper, free surface profiles of Al films deposited in a high-temperature reflow process on substrates with and without trenches are calculated using molecular dynamics simulation. We use an atomic-scale model to analyze microscale nucleation on the substrates. We calculate the effects on nucleation of the initial configuration and temperature of the Al film and the bond energy between Al and the substrates. We find that nucleation occurs significantly when the Al film is initially thicker at the edge of the trench than at the bottom of the trench.

View full abstract

In order to analyze the energy flow in the Carnot cycle, 3 concepts of the first law of thermodynamics are strictly applied. These are the law of conservation of energy, the equal treatment of all forms of energy regardless of form and the consecutive conversion of energy. For example, the energy supplied to the cycle is summed up not only heat energy but also other forms, and, the energy efficiency [η_EN] is adopted rather than the thermal efficiency [η_th] . As a result, it is shown that the energy efficiency of the Carnot cycle is a function of not only the higher and the lower source temperature but also the compression ratio in the isothermal process and it is zero when the compression ratio is 1.

View full abstract

Production and destruction of nitrogen oxides in the dilution process of fuel-rich burnt gases are investigated based on detailed chemical kinetics. From the analysis of combustion processes with mixing of methane-air fuel-rich burnt gas and air, it is shown that fast mixing due to strong turbulence enables a great deal reduction in NO_x concentration, and that a further reduction cannot be achieved for a rich mixture of equivalence ratio higher than 1.2. The NO production rates of elementary reactions indicate that NO is formed by the destruction of HNO at low NO concentration and is deoxidized into HNO at high NO concentration. It is also shown that NO_x concentration can be halved if the temperature decreases with the volume expansion in a high-speed piston-engine motion prior to the heat release due to dilution. Furthermore, the effect of temperature fluctuation on the dilution path is discussed from the practical point of view for reducing NO_x in two-stage combustion.

View full abstract

Characteristics of the combustion of a lean mixture and a nitrogen-diluted mixture in a spark-ignition engine were investigated under the conditions that the fluctuations of mixture strength and charge quantity were very small. The rate of heat release and the combustion duration were determined by a one-zone model from pressure data, and the effects of turbulence characteristics and laminar burning velocity on the combustion duration were discussed. Consequently, it was found that 1% combustion duration varied linearly with the parameter (L/u')^3/3S_L^-2/3ν^1/3, where L is the integral length scale of turbulence, u' turbulence intensity, SQL laminar burning velocity, and ν kinetic viscosity. The 1% combustion duration was a dominant factor for determining 100% combustion duration.

View full abstract

High-pressure injection systems have been installed for the purpose of reducing toxic exhaust gas from diesel engines in recent years. However, the cavitation phenomenon and erosion occur in the high-pressure pipe system due to the large pressure fluctuations at the end of fuel injection. Therefore, there is strong demand for clarification of the cavitation phenomenon under a fluctuating pressure field. For this reason, behavior of cavitation bubbles in liquid fuel under a fluctuating pressure field in a pipeline was investigated in the experiments presented here.

View full abstract

In order to reduce NO_x emission and increase the brake thermal efficiency of a stationary diesel engine, a new prechamber, which is placed at the center of a cylinder head with a smaller volume, and a stratified fuel-water injection system were installed in a single cylinder diesel engine with 170mm bore and 180mm stroke, and it was tested to investigate its potential. As a result, the stratified fuel-water injection system reduced NO_x and smoke emissions with a smaller reduction in brake thermal efficiency. A NO_x concentration of 88ppm (excess O₂ 13%) at the brake thermal efficiency of 39% was obtained using this new prechamber combustion system in combination with the stratified fuel-water injection system.

View full abstract

In order to reduce fuel consumption, smoke and NO_x emissions of a diesel engine, the combustion and cycle characteristics of direct injection (DI) and indirect injection (IDI) diesel engines were investigated from the viewpoint of mixture formation energy, degree of constant volume of heat release and heat loss. As a result, the following conclusions were obtained. (1) The larger the mixture formation energy, which consists of jet, fuel injection and swirl energy, the better the combustion. (2) The combustion of the IDI diesel engine with larger mixture formation energy is better than that of the DI engine. (3) The fuel consumption of the DI diesel engine with higher degree of constant volume of heat release and lower heat loss is lower than that of the IDI engine. This lower heat loss is due to the reduction of specific surface area and coefficient of gas heat transfer.

View full abstract