In order to extend the working range of high-speed submerged water jets and to improve its accuracy confidence, the jet working characteristics are studied under the condition of widely varying ambient pressures P₂. The erosion tests were systematically carried out by using aluminum-alloy specimens, and the aspects of the cavitation clouds around the jets were simultaneously observed. Clearly, the pressure P₂ drastically affects the aspects of cavitation and its occurrence region. With increasing P₂, the length of the cavitating jet decreases, so that the standoff distance, where the most severe erosion appears, shifs upstream. In short, the intensity and the distribution of the impulsive pressures induced by the cavitating jets also drastically change with P₂. The impinging erosion characteristics are also sensitive to the standoff distance and the injection time.

抄録全体を表示

Universality of the Strouhal number in the wake flow of a cylinder over a wide range of Reynolds numbers, i.e., 5×10⁴<Re<10⁷, was considered. Eight cylinders with various surface roughnesses were used for the measurement. The universal strouhal numbers considered were Roshko's number, Bearman's number, Griffin's number and that took the measured values of the width of wake streets for the reference length. These universal Strouhal numbers were calculated using the measured values of vortex frequency, surface pressure distributions and drag. Comparisons were made. It was made clear that Bearman's number was the most uniform of all the universal Strouhal numbers considered. It had a uniform value of 0.18 throughout the Reynolds number range. The reason for the nonuniformity of other Strouhal numbers was also considered.

抄録全体を表示

The vortex method with the diffusion velocity model was utilized to simulate mixing shear layers. Our numerical results of the mean-velocity profiles, the momentum thickness and the local width of the mixing layer, at Re=1.8×10⁵, are in good agreement with the experimental results by Oster Wygnanski. Our results also show that the momentum thickness and the local width of the mixing layer increase, and the turbulent intensity decreases as Re number decreases. The similarity of the mean-velocity profiles is observed even when Re number changes.

抄録全体を表示

An incompressible round compound jet issuing from a circular nozzle was simulated by a three-dimensional vortex blob method. The circular nozzle was composed of a source disc and vortex panels. The nascent vortex blobs were released from the edge of the nozzle in such a way that their circulation satisfies Kelvin's conservation law of circulation. The vortex rings released from the edge experienced vortex merging and developed three-dimensional structures with small-scale random structures. Effects of the limitation of maximum vorticity of the vortex blobs on the vortex structures were discussed. The large-scale vortical structure is fairly indendent of the limitation of maximum value of vorticity.

抄録全体を表示

In order to estimate the time- and volume-averaged volumetric fractions in a dispersed multiphase flow, terminal velocities are often used as representative velocities in the drift flux model, slip model and others. Their applicability to the estimation of the two-or three-dimensional structure of a liquid-solid two-phase How was examined in this study. It was clarified that terminal velocities cannot be applied to the prediction of the solid velocity distribution. Hence, the terminal velocities were corrected by considering the phase distribution and pipe wall effects. This correction gave more accurate predictions.

抄録全体を表示

The present paper describes the experimental work on the flow through the tapered branch model with the taper angle of 1°, in which a tapered side branch bifurcates at a right angle from a tapered trunk. The results show that in steady flow the wall shear stress reaches a maximum at the upstream corner of the side branch and varies significantly with position along the proximal wall in the form of a damped sine wave. The variation is closely associated with the secondary flow from the trunk to the side branch. Furthermore, the wall shear stress in the tapered branch varies more moderately than that in the parallel straight branch, and the occurrence of separation flow would be suppressed around the branch point. Therefore, the results of the present study suggest that the tapered branch is more suitable for a human artery than the parallel straight branch.

抄録全体を表示

The loss coefficients and the power-loss coefficients for sharp-edged combining tees with the area ratios of 11.44 and 17.72 are determined experimentally for the branch angles of 60°and 120°. Based on the least squares analyses of the experimental data of both the present authors and previous investigators, empirical equations of loss coefficients for 60°and 120°combining tees are proposed. The effects of both the area ratio and the radius of curvature of the wall at the joining edge on loss coefficients are taken into account in deriving the equations. The comparison of the proposed equations with the experimental data shows that the proposed equations can be used to reliably estimate the energy losses at 60° and 120° combining tees.

抄録全体を表示

Two-phase flow pressure drops at the junction were measured in the Freon loop. The loop has two channels, the main and branch, which combine horizontally at the angle of 90 or 45 degrees. Each pressure drop of the junction is obtained by subtracting the frictional and accelerative pressure drop from the measured pressure difference. The junction pressure drop coefficient, defined as the pressure drop divided by the momentum flux after combination versus the ratio of the momentum flux of the branch to that of the main channel after combining, explained the characteristics of the junction pressure drop well. The junction pressure drop coefficients of 45 degree were less than those of 90 degree, and the values at the branch channel became similar to those of 90 degree when the momentum flux was large.

抄録全体を表示

A numerical simulation of gas-solid two-phase flow is carried out for mixed-size coarse particles and for uniform ones in a horizontal pipe of 40 mm inside diameter for mean air velocities of 12 and 16 m/s and solid loading ratio less than 1.14. Interaction between gas and particles is considered in the mean flow level of gas. The gas flow profile is modified by particles when the loading ratio is large, and the modification is significant in the acceleration region of particles. The main source of particle rotation is collision with the pipe wall and the particle-particle collision decelerates the rotation. The particles lose kinematic energy mostly by collision with the wall.

抄録全体を表示

A generalized Langevin equation developed by Haworth and Pope [Phys. Fluids, 29-2 (1986), 387] is applied to calculate the dispersion of a passive contaminant in turbulent pipe flow. The model coefficients in the equation are determined from an algebraic relation based on the consistency condition for the second-order moments of velocity, which includes the third-order moments. In the present model, the first- and second-order moments are used as the input data, but the third order moments are not specified due to the lack of reliable data. First, we checked on the consistency condition for the simulated velocity field. Second, the long time dispersion was examined. The Eulerian velocity statistics show good agreement with the specified data to the second order. With regard to long time dispersion, the longitudinal distributions of a cross-sectional mean concentration agree well with experiments. It is also found that the proper value of the Kolmogorov constant C₀ is 1.9 for this problem.

抄録全体を表示

Based on the assumption that the fundamental modes of the fluid transfer between subchannels are turbulent mixing, void drift and diversion crossflow in a developing (hydrodynamic nonequilibrium) two-phase flow, we have examined whether the following treatment is applicable or not, i. e., whether the turbulent mixing rate in such a flow can be considered to be equal to the rate in the equilibrium state that the flow will attain. In order to obtain the data needed for the examination, experiments were performed for hydrodynamic nonequilibrium air-water churn flows without pressure differential between the subchannels, using simple vertical multiple channels consisting of two circular subchannels. Data on tracer concentration distribution along the channel axis were obtained for both gas and liquid phases. The analysis of the data showed that the assumption and the treatment mentioned above are valid for all nonequilibrium flows tested. Furthermore, it has been clarified that the effect of turbulent mixing on tracer mass exchanges between the subchannels is more dominant than that of void drift.

抄録全体を表示

The DNS (direct numerical simulation) of the decaying homogeneous anisotropic turbulence was performed by the spectral method on a parallel computer, Fujitsu, AP 1000. A domain decomposition method suitable for FFT (fast fourier transform) was developed. The padding method was applied to eliminate aliasing error. The simulation was parallelized with 16×16×16 to 128×128×128 meshes. A sufficiently high efficiency was obtained up to 512 processors. The anisotropic turbulence decays in the invariant II vs III plane of the anisotropic tensor. The Reynolds stress returns to isotropy after the invariant III becomes positive. The return to isotropy of the dissipation is found to be more rapid than that of the Reynolds stress and takes a different route.

抄録全体を表示

Detailed measurements are performed on boundary layers on a flat plate subjected to incident periodic wakes. The objective of this study is to examine whether the velocity fluctuation associated with the wake passage, which is due to a so-called negative jet, might have any effects on the wake-induced boundary layer transition. A spoked-wheel-type wake generator is used to simulate the unsteady flowfield over the suction surface of a turbine blade or a compressor blade by changing the direction of rotation of the wake generator. Wake-affected heat transfer distributions on the flat plate indicate that the wake passage promotes the boundary layer transition ; however, appreciable difference in the transition onset point appears between the two unsteady flow conditions with the normal rotation and the reverse rotation of the wake generator, emulating a turbine blade case and a compressor blade case, respectively. This difference is also confirmed by the boundary layer measurement using a hot-wire probe, which shows that a turbulent region induced within the boundary layer by the wake passage for the reverse rotation case evolves downstream at a consider-ably slower speed compared to that of the normal rotation case.

抄録全体を表示

Circular coaxial jets have been experimentally studied, and the outer annular jet is excited by the sound wave from speakers. We have measured velocity, turbulent intensity distributions and the power spectra of the fluctuating velocity of the coaxial jets for different configurations of the outer nozzle with a pipe length of L/D₀=0 to 5 and for various values of the velocity ratios of inner jet to outer jet, in order to examine the effects of excitation. For the nozzles with the pipe length of L/D₀≥2 and the forcing frequency of f₀>200Hz, the recirculating region shrinks and moves upstream. The intensity of fluctuating velocity in the outer mixing region for the jets of L/D₀=0 and 0.5 increases at f₀=300∼400 Hz. It is confirmed that the increase of turbulence intensity is related to that of the fundamental frequency component. For the jet of L/D₀=1, turbulence suppression is observed at the forcing frequency of about 600 Hz. It is found that the suppression effect is a consequence of earlier transition of the shear layer vortices. For the nozzle with L/D₀=2, the vortex pairing event has been recorded through phase-locked measurements at the forcing frequencies of about 300∼400 Hz. The velocity fluctuation intensity is confirmed to be enhanced by vortex pairing.

抄録全体を表示

Self-excited oscillation of an axisymmetric jet impinging upon a ring is investigated experimentally. The discreat vortices impinge upon the downstream sharp edge of the ring, in turn producing an upstream influence, thereby closing the loop for sustaining oscillations, as confirmed by the form of phase analysis of velocity and pressure signals. The fundamental frequency component exhibits an overall phase difference of (2nπ) +1/2 (n=0, 1, 2) between the nozzle lip and edge. Furthermore, the fundamental component propagates with phase speed, which is about 0.52∼0.59U_o. Spectral analysis of velocity fluctuation reveals multiple frequency components having substantial amplitudes relative to that of the fundamental frequency component ; using coherence analysis it is shown that the lowest frequency component can interact with the fundamental component either to reinforce itself or to produce an additional frequency component.

抄録全体を表示

The fundamental problem of the decelerating laminar boundary layer defined in the above title has been analyzed theoretically continued from the preceding paper. The difficult problem of numerical instability, which arose inevitably in the preceding numerical work, has been overcome perfectly by a rational nondimensionalizing scaling technique based on the original "semi-quantitative viscous diffusion theory" which has been developed in the preceding paper. And thus the interesting flow behaviors of the decelerating boundary layer over the whole time region from far past to the deceleration terminal at which the plate stopps, and further up to far future where the plate is continuously at rest, have been analyzed numerically and are clarified satisfactorily without any numerical instability or other like obstructions.

抄録全体を表示

A planar shock wave reflecting from a circular concave wall generates a local pressure/temperature amplification at a certain point adjacent to the center of the concave wall. This phenomenon is, so-called, shock wave focusing but, in this case, it is caused by the collision of a pair of triple points formed on the shock reflection process. This mechanism is considered to be governed by the behavior of these triple points and depends on the shape of reflector. In this paper, in order to clarify the contribution of the triple points and its three-shock structure to the focusing mechanism, the difference of an initial angle of concave wall as a shape of reflector was discussed. Particularly, the behavior of a secondary triple point and its slip line behind a normal shock which was formed due to the inverse Mach reflection was investigated by using double exposure holographic interferometer. Numerical simulation was also carried out using Harten-Yee type second order upwind TVD finite difference scheme applied to 2-D Euler equations. The numerical results are compared with the interferograms.

抄録全体を表示

Simplified model equations are proposed for simulations of propagation of weak shock waves in nonuniform liquid. It is explained why the equations can predict shock propagation in medium with arbitary distributions of the density and the acoustic speed which are given as parameters. The TVD scheme is applied to numerically solve the equations. Algorithms are shown for finite difference calculations of the equations. Simulations are carried out for refraction and reflection of weak shock waves by an inclined interface. Computed results are compared with wave patterns obtained from Huygens' principle as well as reflection coefficients of sound waves. Good qualitative agreement is obtained and it is concluded that the present model equations are convenient to predict phenomena induced by shock propagation in nonuniform liquid.

抄録全体を表示

The C-CUP method suitable for both compressible and incompressible fluids is extended to the general curvilinear coordinate system. In this paper, a staggered square grid in the transformed space is used, and the numerical viscosity, which does not depend on the change of metrics due to the rotation of the grid, is proposed. The equation of pressure is solved by the iteration method. As numerical examples, a shock-tube problem, a two-fluid flow and a flow field around a two-dimensional wing were calculated. Numerical results of the shock-tube problem show that the proposed numerical viscosity for the curvilinear coordinate system is effective to stabilize the solution. For the viscous two-fluid flow, the sharp interface between the two fluids is traced by using the density function.

抄録全体を表示

The present study theoretically investigates the aggregate structures of thick chainlike clusters arising in ferromagnetic colloidal dispersions by means of the cluster-moving Monte Carlo method. Pair correlation functions are also evaluated to clarify the internal structures themselves of such clusters. The following aggregation image of thick chainlike clusters has been obtained from the Monte Carlo simulations for a two-dimensional model system. Particles aggregate to form chainlike clusters when magnetic attractions between particles become strong over a certain value. Such thin chainlike clusters prefer linear formation along a magnetic field direction if the strength of magnetic field is much greater than magnetic attractions between particles. In this situation, attractive forces come to act between these nearly linear chainlike clusters, and these forces increase with length of the clusters. Such attractions between the clusters lead to the formation of thick chainlike clusters.

抄録全体を表示

The object of the present study is to evaluate the performances of various microcanonical molecular dynamics algorithms. To do so, the divergence times and energy fluctuations are discussed for a model system, i. e. a three-dimensional Lennard-Jones system under various conditions of number densities and temperatures. The results of the evaluations of superiority or inferiority of algorithms are as follows. The velocity Verlet, leapfrog, and Beeman algorithms are significantly superior to the other algorithms such as the 4-value Gear algorithm since a system does not diverge and the energy conservation law is reasonably satisfied for much larger time intervals. Although these three algorithms show approximately the same performance concerning the properties of divergence times and energy fluctuations, we can conclude that the velocity Verlet algorithm is the most suitable for molecular dynamics simulations of flow problems, since this algorithm is easy to use, requires less computer memory, and evaluates molecular positions and velocities at the same time steps.

抄録全体を表示

The present study investigates the pressure characteristics of a two-dimensional air-cushion pad in order to determine the effect of the fringe of the pad on the cushion pressure. Two-dimensional experimental models are used : in model (1), the width of the fringe is varied, and in model (2) its width is very large. The present experimental results show the following : (1) The effect of the fringe on the cushion pressure is fairly large ; the cushion pressure dicreases remarkably with the increase of the width of the fringe for l/h&le; 5 but the cushion pressure becomes almost constant for l/h&le;5, where l and h are the width of the fringe and the clearance of gap, respectively. (2) The decrease of the cushion pressure is caused by the circulatory flow which is formed in the gap between the surface of the goods and the fringe of the pad, and the pressure at the edge of the fringe is almost atmospheric pressure.

抄録全体を表示

This paper describes extensive computer simulation for aerodynamic noise radiated from a square cylinder using an interactive steering supercomputing simulation system. The unsteady incompressible three-dimensional Navier -Stokes equations are solved by the the finite volume method using a steering system which can visualize the numerical process during calculation and alter the numerical parameter. Using the fluctuating surface pressure of the square cylinder, the far-field sound pressure is calculated based on Lighthill-Curle's equation. The results are compared with those of low noise wind tunnel experiments, and good agreement is observed for the peak spectrum frequency of the sound pressure level.

抄録全体を表示

The present paper deals with a finite element method for the solution of a two-dimensional time-dependent free surface flow problem and its application to the analysis of the low-velocity impact of a rigid body into still water. For the ease of numerical treatment of the free surface expressed by a multiple-valued function, a marker-particle method is employed. Numerous virtual particles are spread over regions occupied by water. Each particle is moved according to the water velocity on a fixed finite element mesh. The velocity field within water regions is calculated by solving the Navier-Stokes equations and the equation of continuity by the finite element method based on quadrilateral elements. The methodological problems arising in the implementation of the marker-particle method on an unstructured mesh of finite elements and the solutions to those problems are discussed. Numerical results are compared with available experimental data and good agreement is obtained.

抄録全体を表示

This paper presents a fast-response simultaneous measurement method of velocity and concentration in the mixing of air and city gas mainly composed of methane. The measurement of city gas concentration and fluctuation utilizes the absorption of a laser light by methane in the mixture. A hot-wire anemometer is used to measure velocity. It is shown that the effect of the change in the thermal conductivity of the mixture on a hot-wire anemometer signal can be compensated by using the measured city gas concentration at the same time and location. This method is applied to the simultaneous measurement of concentration and velocity as well as fluctuation in a coaxial flow of city gas and air.

抄録全体を表示

The flow characteristics of two models of cross flow fan were clarified by flow visualization and LDV investigations. The velocity vector maps, stream lines and vorticity distributions were obtained so that the position of the eccentric vortex in the rotor could be clarified. Furthermore, the aerodynamic characteristics of a blade and the power generated by a blade were clarified. The results showed that the blades of a rotor at θ= 240°to 270° (in discharge area) have good aerodynamic performance and have better work performance than those of the suction area. The effect of changing of casing shape and suction load in suction area on the flow pattern was also clarified.

抄録全体を表示

Steep gradients of fluid density are the influential factor of spurious oscillation in numerical solutions of low Mach number (M«1) compressible flows. The total variation diminishing (TVD) scheme is a promising remedy to overcome this problem and obtain accurate solutions. TVD schemes for high-speed flows are, however, not compatible with commonly used methods in low Mach number flows using pressure-based formulation. In the present study a higher-order TVD scheme is constructed on a modified form of each individual scalar equation of primitive variables. It is thus clarified that the concept of TVD is applicable to low Mach number flows within the framework of the existing numerical method. Results of test problems of the moving interface of two-component gases with the density ratio &ge;4, demonstrate the accurate and robust (wiggle-free) profile of the scheme.

抄録全体を表示

Flow in a linear compressor cascade with low-speed and small-aspect-ratio blades is predicted by a finite-element full 3-D Navier-Stokes solver. The code is applied to the analysis of the secondary flow patterns in several planes normal to the main flow inside and behind the cascade. Moreover, experimental measurements of the cascade flow field are presented. The blade has an aspect ratio of 1.04. Cascade performances are presented in the form of pitch-averaged spanwise variation of outlet flow angles, losses, axial velocity ratios and blade circulations. The analytical results are compared with the experimental data of a cascade. The agreement between the predictions from this numerical analysis and experimental data is shown to be good.

抄録全体を表示

The experimental results obtained from radially curved mixed-flow vaneless diffusers having a constant radius of curvature at the curved part show that the diffuser flow deviation depends on the radius of curved part and the total enthalpy gradient between the diffuser walls. In this study, a design method of the radially curved mixed-flow diffuser whose inlet flow has a total enthalpy gradient is developed. The objective of the optimization is to minimize the flow deviation of the diffuser flow and the design functions are based on the radius of centerline in the meridional section. Some applications which prove the effectiveness of this method are demonstrated.

抄録全体を表示

It is the purpose of the present study to investigate the flow pattern in a fully three-dimensional poppet valve for oil hydraulic power applications and the vibration of the poppet by means of both numerical and experimental flow visualizations. The numerical simulation by a discrete vortex method is used to analyze the annular axisymmetric jet flow issuing from an orifice of the poppet valve. The experimental flow visualization in the poppet valve is conducted by using two industrial fiberscopes which are connected to two high-speed video cameras from two different directions simultaneously. The flow pattern near the orifice with cavitation and the vibration of the poppet are observed in three dimensions. A digital image processing technique for the image obtained by the flow visualization is employed.

抄録全体を表示

All positive displacement pumps generate a flow ripple which interacts with the wave transmission characteristics of the connected circuit to produce pressure ripple. In order to predict the pressure ripple generated in a hydraulic system accurately at the design stage, it is necessary to understand and quantify the wave transmission characteristics of all of the system components as well as the source characteristics of the pump. This paper is concerned with measuring the transfer matrix parameters of hydraulic system components by applying the principle of the existing method which has been developed originally for experimentally determining in-duct acoustic properties. Experiments were conducted on the high-pressure flexible hose, various types of control valves with moving parts, or a hydraulic cylinder, of which no adequate theoretical representatios have been made, as well as the uniform steel pipes for preliminary examination. It has been found from a wide range of tests that the matrix parameters of various types of hydraulic components can be measured with sufficient accuracy for practical use by carefully calibrating pressure transducers and by using dual FFT providing A-D converter with good resolving power. The successful application of the existing measurement principle to the hydraulic components as well as the high quality of the results make it possible to use the test results as valuable materials for both modeling hydraulic components and predicting pressure ripple generated in hydraulic systems accurately at the design stage.

抄録全体を表示

Fluid-borne presure ripple in hydraulic systems may be a major source of the overall system noise levels as well as a major cause of the vibration of system components. The most common way to reduce the pressure ripple is the use of a hydraulic silencer. A number of criteria have been proposed for standard hydraulic silencer performance, but a reliable and a practical method for standard silencer assessment has not been established because these criteria are dependent on the wave transmission characteristics of all other system components as well as the silencer. This paper deals with the application of an experimental technique described in this first paper for the determination of matrix parameters of hydraulic system components. Experiments were conducted for a total of eight silencers of four typical geometrical configurations. It was found from a wide range of tests that the matrix parameters of various types of silencers could be obtained with good repeatability and with lower amount of scatter by carefully calibrating presure transducers and using two-channel FFT providing A-D convertor with a good resolving power. The transmission loss for standard silencer assessment could also be obtained readily by calculation using the experimentally deter-mined matrix parameters and the known relationship. Thus, the successful application of the existing measurement principle to the hydraulic silencer as well as the high quality of the results make it possible to use the test results as valuable materials for both modeling and assessing silencers as well as for producing less noisy hydraulic systems with less fluid-borne pressure ripple.

抄録全体を表示

The air chamber system is a kind of method for transformation from wave energy to mechanical energy. In this system, wave energy operates periodically in the wave power generation plant. In order to improve this system, we propose a new technology with one-way flow and restriction of turbine input. The former is applicable to an impulse turbine, and the latter saves the excess power from the air chamber. In this system, the air duct has a bypath with a turbine. Pressure and flow direction are controlled with only one swing plate valve in the air duct. This paper describes both experimental results at a site of sea-side, and the analytical result of the plant performance.

抄録全体を表示

In this paper we describe an experimental study of a new type of nozzle which generates a high-speed two-phase water jet as a propulsion device for a ship. In order to investigate the performance improvement of this device, two nozzles are designed : The first nozzle has a cascade with 4 blades instead of the bottom plate. The second nozzle has the same structure as that of the nozzle without a bottom, except that the high-pressure chamber is placed at the position where the bottom plate is removed. These nozzles were set in a towing tank and the thrust was measured for the various parameters such as the air injection angle, the pressure of the injection gas, and the towing speed. The thrust increases as compared with that of the nozzle without bottom. The maximum value of the propulsive efficiency is about 20% at the flow velocity of 1.5m/s. It was proven that the increase of the water flow rate in the nozzle is essential for improving the performance of this device.

抄録全体を表示

Characteristics of a thin liquid film flowing concurrently with high-speed gas flow were experimentally investigated in the gas velocity range from 20m/s to 80m/s. Moreover, experimental and analytical investigations were performed on the absorption of CO₂ from CO₂-laden high-speed gas flow into the thin water film flow. We concluded the following. First, there are two layers in the film flow, one is the viscous sublayer and the other is the wavy layer. The velocity is discontinuous at their boundary. Secondly, we verified that the prediction method for the bulk concentration variation of CO₂ in the flow direction, which we proposed in the first report, is also valid in the high-gas-velocity region if the velocity profile in the liquid film and the mixing length are modified. Thirdly, the correlation of average mass transfer for the liquid film flow which we proposed in the second report can also be used in the high-gas-velocity region without any modification.

抄録全体を表示

Flow patterns and mass-transfer characteristics in a channel with arc-shaped walls are investigated experimentally for steady and pulsatile flows. Flow visualizations are performed by the aluminum dust method, and mass transfer rates are determined by the electrochemical method. This channel experiences an earlier transition to turbulence compared with a channel with the sinusoidal walls previously studied. Mass-transfer enhancement depends on the oscillatory fraction of the flow rate and the Reynolds number for net flow ; however, the effect of frequency is weak. The mass-transfer enhancement factor exceeds the value predicted by the quasi-steady model, in the laminar flow regime, but agrees well, in the turbulent flow regime.

抄録全体を表示

Premature dryout due to flow oscillation is a very important factor in designing boiling systems. The flow oscillation depends, in general, on system size and/or configuration, and therefore the relationship between the premature dryout and the flow oscillation has not been fully understood so far. In this investigation, a CHF experiment in a forced flow boiling channel under the oscillatory flow condition system has been conducted. Numerical simulation has also been conducted based on the lumped-parameter model of the boiling channel. The simulation well represents the transient behavior of the dryout under the oscillatory flow condition.

抄録全体を表示

Experiments were conducted to obtain row-by-row heat transfer data during condensation of downward flowing HCFC-123 vapor in staggered bundles of horizontal two-dimensional fin tubes. Two conventional low fin tubes and two finned tubes with fin shapes close to that of a new high-performance fin proposed by Zhu and Honda were tested. The latter tubes had fin dimensions that were optimized by theoretical analysis. The heat transfer characteristics were compared with previous results for in-line bundles of the same test tubes and with theoretical predictions for stagnant vapor. The decrease in heat transfer due to condensate inundation was less significant for the staggered tube bundle. The highest heat transfer coefficient was provided by the staggered bundle of one of the new finned tubes. The theoretical predictions compared well with experimental data at a low vapor velocity and a low to medium condensate inundation rate.

抄録全体を表示

Measurement of condensation characteristic curves, which show the relation between the heat flux/heat transfer coefficient and the surface subcooling over a wide range of a subcooling, was performed for various vapor concentrations and mass velocities of binary mixture of steam and ethanol. The curves divided into several characteristic domains in consequence of the variations of the modes of condensate occurring due to the distribution of the surface tension based upon the concentration distribution on the condensate surface. In the order of surface subcooling, vapor-side dominated region, region of steep increase of heat transfer, transition region and region of ordinary film condensation were observed. It was also shown that the heat flux and the heat transfer coefficient were independent of the position on the heat transfer surface in the steep increase region. In contrast, relatively strong dependency appeared in the transition region.

抄録全体を表示

This paper reports an experimental study conducted by mounting a 31×31×7 mm³ simulated module on the base of a 20-mm-high parallel-plate channel and cooling it by forced convective air flow. The module has a given power input ranging from 1.8 to 3.6 W. The heat transfer process is complex due to the presence of multiple heat flow paths and the coupling of heat conduction in the base and convective heat transfer from the module and base surfaces. In an attempt to understand the effects from different factors upon cooling of the module, some variables are introduced in the design of the test piece, namely, the material of the base and the thermal resistance between the module and the base. Heat transfer data were reduced to macroscopic thermal conductance components such as conjugate conductance U_BB of the board and direct convective thermal conductance U_A on the module.

抄録全体を表示

A circular cylinder of 50 mm diameter (D) was placed in a two-dimensional jet. The flow pattern was changed by mounting two large flat plates symmetrically near the cylinder maintaining a clearance (C) between the cylinder and each plate. The half-angle (ψ) between the plates and the clearance were varied (ψ=20°∼60° ; C/D≤0. 3). The jet velocity and nozzle-to-cylinder distance (L) were also changed. The pressure distribution around the cylinder for the case of C/D≤0. 1 differs markedly from that for the case of C/D≥0. 2. In the former case, pressure fluctuation is suppressed and no peak is observed in the spectrum of velocity fluctuation. Thus, the suppression of self-induced flow oscillation is suggested for the case of C/D≤0. 1. The value of peak Nusselt number changes remarkably with the values of ψ and C/D. The mean Nusselt number is clearly larger in any geometrical condition other than that obtained without an insertion of the plates.

抄録全体を表示

This paper presents the evaluation of heat transfer rate on the surface of turbine moving blades based on their microstructural changes due to aging. γ' strengthen nickel base superalloy are used extensively in turbine moving blades. The aging effect on the change of γ' precipitates is investigated. The average diameter of γ' precipitates increases linearly with time raised to the one-third power for aging temperature. The metal temperature of turbine moving blades used in gas turbine field operations could be estimated by applying the above method. Also, its temperature is obtained numerically from the heat transfer rates which are calculated using some analytical methods. By comparison of both metal temperatures, the heat transfer rates on the surfaces of turbine moving blades used in gas turbine field operations are estimated and are evaluated.

抄録全体を表示

Low-speed upward flow in an uniformly heated vertical flat duct was studied experimentally. The flow pattern was visualized by the dye injection method. The flow was laminar throughout the entire channel when the effect of buoyancy was relatively small. However, the streak line of dye began to oscillate sinusoidally halfway through the channel when the effect of buoyancy became large, and then the flow transited to a turbulent one. The wall temperature oscillated sinusoidally at the transition point, and fluctuated randomly in the turbulent region. The heat transfer coefficient increased rapidly with the flow transition. The effects of Reynolds number and modified Grashof number upon the condition under which flow transition occurred and upon the heat transfer coefficient in the turbulent region were clarified.

抄録全体を表示

In this paper, the modified laser-flash method (LF method) measuring themal diffusivity of low conductivity materials is proposed. What is unique about this method is that the front shielding plate (FSP) is employed. FSP prevents superheating of the irradiated surface and stops the beam from transmitting through when sample is transparent. In order to calculate the thermal diffusivity, the equation describing the rear surface temperature response involving the heat loss effect is derived under assumption that the FSP temperature gradient is negligible because it is so highly conductive. An acrylic resin with copper and aluminum FSP was tested, and the measured thermal diffusivity agreed well with the recommended value in the literature. This modified LF method is useful for measuring high polymer materials such as resin or rubber, especially for the case that their thickness is greater than 2×10^-3m.

抄録全体を表示

A molecular dynamics study has been performed for water over wide ranges of temperature and density. The Carravetta-Clementi intermolecular potential model was applied for a system of 256 molecules. The liquid-vapor coexistence region, near-critical and supercritical regions, and region of high-density compressed liquid were extensively studied. The network of hydrogen bonds is analyzed in the present report. The size, temperature and density dependences, and structure of the network are presented in detail.

抄録全体を表示

A rigorous two-equation heat-transfer model has been constructed with the aid of the most up-to-date direct numerical simulation (DNS) data for wall turbulence with heat transfer. The DNS -data indicate that the near-wall profile of the dissipation rate, ε_t, for temperature variance, t², is completely different from the existing model predictions. In the 1st report, we have performed a critical assessment of existing ε_t-and ε_t-equations for both two-equation and second-order closure models. Based on these assessments, we propose a new dissipation-rate equation for temperature variance, taking into account all the budget terms in the exact ε_t-equation. Comparisons of the present model predictions with the DNS data for a channel flow with heat transfer show an excellent agreement for the profiles of εt itself and the budget in the ε_t-equation.

抄録全体を表示

Noncontact and nondestructive measurements for determining temperature and CO₂-concentration distributions in environmental fields are under investigation by acoustic tomography. A new method to reconstruct the three-dimensional temperature and CO₂-concentration distributions from the projected ultrasonic phase differences is proposed in this paper. A set of programs to evaluate the projected data is made and a numerical reconstruction process is also offered. An experimental system consisting of several pairs of ultrasonic sensors is established to apply our new method to axisymmetric temperature and CO₂-concentration fields. The temperature profiles reconstructed by the present acoustical tomography agree well with the results of the thermocouple measurements. As a result, the procedure proposed in this paper is verified to be reasonable.

抄録全体を表示

Unsteady state natural convection around a horizontal cylinder in the saturated-glass-beads packed bed is experimentally and numerically investigated. The outside wall-temperature changes sinusoidally to clarify periodical temperature changes in the packed bed in Darcy and non-Darcy flows. For Darcy flow regions, clear thermally concentrated regions appear periodically in the porous media for rather large periodic time, but they are unclear for non-Darcy flows. This phenomenon is successfully analyzed based on the Forchheimer-Brinkman-Boussinesq model. Also, good agreement is obtained in consideration of the thermal nonequilibrium between particles and surrounding fluids and the channel effects near the cylinder wall with large particles.

抄録全体を表示

When a closed pipe filled with hot water is cooled, a cavity is generated in the pipe under pressure reduction due to large thermal contraction of water. In this paper, it was made clear that the transient response of pressure reduction in the above cooling cavitation shows the undershoot, and the response was modeled and formulated by the air desorption from the liquid into the cavity. The model parameters identified experimentally were investigated, and the time constant of air desorption sucked through the interface and the time constant of air diffusion within the liquid are obtained as the expressions in terms of the initial total air mass. Comparing the experimental and the calculated results, it was found that the minimum pressure and the final equilibrium pressure are proportional to the initial total air mass, and that the time at the minimum pressure is inversely proportional to the initial total air mass.

抄録全体を表示

Spray cooling experiments were carried out in an aircraft making sequential parabolic flights. CFC-113 and water were used alternately as the coolant which was sprayed onto a plane surface of an electrically heated copper block. Transient cooling of the block was continuously recorded for each experiment to deduce the heat flux (q) vs. surface superheating (ΔT_sat) relationship at a nearly constant gravity level which was either reduced (∼10^-2g_e) or elevated (∼1.5 or 2g_e) from the terrestrial gravity level, 1g_e. The q-ΔT_sat relation for CFC-113 showed a slight shift in the direction of decreasing ΔT_sat and a decrease of the peak heat flux by about 10% with the reduction in gravity. As for water sprayed at a rather low rate, the reduction in gravity resulted in both the shift of the q-ΔT_sat relation in the direction of decreasing ΔT_sat, and an increase in the peak heat flux. Such gravity dependences of heat transfer were lost at a rather high rate of water spraying.

抄録全体を表示