Numerical simulations of the steady shear flows of a nematic phase are performed using nonequilibrium molecular dynamics. The SLLOD algorithm is developed for application to the steady shear flows of ellipsoids of revolution that interact via the Gay-Berne potential. The system composed of particles interacting via the Gay-Berne potential forms various phases including a nematic one. In the initial stage of simple shear How of the nematic phase, the order parameter significantly decreases as the director rotates rapidly. The director, however, is inclined at a nearly constant angle regardless of shear rate in the steady state. Rheological properties, such as shear viscosity and normal stress differences, are examined.

View full abstract

We present the effects of a ground plate (stationary or moved) on the Magnus effect of a rotating circular cylinder. Measurements were carried out for Reynolds number Re=6.6×104 and velocity ratio α=1. The surface pressure distributions on the cylinder were measured with a highly sensitive pressure transducer for the various clearances between the cylinder and the ground plate. The drag and lift forces on the cylinder were calculated from those surface pressure distributions. As a result, it was found that the lift suddenly changes at a certain clearance, and this behavior is not caused by the transition of the boundary layer from laminar to turbulent. This phenomenon is related to a change in the flow pattern, and each pattern was measured with a hot-wire anemometer. These flow patterns show that one of the flows through the clearance is along the cylinder surface, and another is along the ground plate. It is interesting that this phenomenon is unstable and hysteretic

View full abstract

Vibration and acoustic resonance of tube banks of a boiler heat exchanger are some of the most important problem. The main causes of these problems is induced cross flow through the tube banks that gives rise to fluctuation at some frequency. However, there is no report that explained velocity fluctuations in closely arranged circular cylinders. This present paper is concerned with the excitation mechanism of velocity fluctuation in closely arranged staggered tube banks. We measured the frequency of velocity fluctuation for nine arrays in the range of Re_s=1.0×10⁴ to 9.4×10⁴. In these results, Strouhal numbers tend to increase as longitudinal spacing ratio decrease and we measured several Strouhal numbers at the tube arrangement with X_T=2.O, X_L=1.4 and X_T=2.O, X_L=1.8. Analysis of the frequency was done by the use of a linear instability model in which instability occurs due to the unbalance between friction and pressure difference. The result of this calculation agrees quite well with available experimental data.

View full abstract

Rarefied gas flows in narrow spaces are affected by the surface roughness of the walls, but the interaction between gas molecules and solid surfaces is not yet sufficiently understood. The effects of surface roughness on the conductance of passages composed of two flat plates are studied experimentally and theoretically. Experiments are performed on both passages with generally treated surfaces and ones with two-dimensional triangular roughness of constant height cut by machining. The latter roughness is chosen as the model of roughness for calculation by the Monte Carlo method to simplify the problem. The experimental results show that the conductance decreases with increase in the angle of the local roughness slope. The calculated results agree well with the experimental results in the case of two-dimensional roughness.

View full abstract

When a compression wave propagating in a straight circular tube reaches the tube's open end, reflection and radiation of the compression wave occurr. The reflected compression wave propagates in the circular tube toward its portal as an expansion wave, and the emitted compression wave extends into the surrounding area from the open end as an impulsive wave. During the process of reflection and radiation, the flow at the open end is very complex due to three-dimensional phenomena. The purpose of this paper is to investigate the mechanism of generating the impulsive wave from the compression wave. A numerical analysis is carried out for the emission of a propagating compression wave from an open end of a circular tube with an infinite baffle plate at the open end, using the TVD method. The process of generating the impulsive wave and the influence of compression wave length on its generation process are discussed, and the directivity of the impulsive wave at near field is shown.

View full abstract

This paper provides numerical solutions of vorticity- velocity equations for incompressible viscous flows around a single or double circular cylinder revolving in a closed circular region. We perform the computation for three different motions of the circular cylinder; nutating motion, rotating motion and rolling motion. The computed results show that the center of rotating flow is closer to the geometrical center in the closed region in the order of rolling, nutating and rotating motions for the case of a single circular cylinder, and the streamlines are more deformed around the geometrical center in the aforementioned order for the case of a double circular cylinder. Furthermore, it is shown that the vorticity contours for a single circular cylinder are similar to those for a double circular cylinder in the gap between a circular cylinder and the wall of the closed region.

View full abstract

This study aims to analyze the basic factors affecting interfacial characteristics of liquid film flow driven by gas stream. An experimental method, in which liquid film thickness could be adjusted freely by varying the inclination angle of flow channel and interfacial shear force could be estimated from the angle, was developed. The effects of gas-liquid relative velocity, liquid film thickness, liquid flow rate, channel height, liquid viscosity and surface tension upon the interfacial shear force were examined. The effects of these basic factors upon wave amplitude, wave length and wave velocity of the interfacial wave were also examined. The results for the disturbed wave regime were expressed in terms of coefficient of the exponent. It was found that the motion of liquid film was dominated by drag forces exerted by gas stream on the disturbed waves.

View full abstract

The characteristics of pressure wave propagation in a vacuum tube have been investigated experimentally from the viewpoint of vacuum protection in the beamlines of a synchrotron radiation facility. Baffle plates having a single orifice of 5, 10 or 15 mm in diameter were installed in shock tubes 5 m in length, and 36.6 or 68.8 mm in diameter, in order to slow the pressure wave or shock wave propagation as a model for the beamline. To evaluate the decay of pressure waves, pressure changes with time at several locations along the side wall as well as at the end wall of the tube were measured. The results show that the effect of the orifices on pressure wave propagation and its decay is significant. The present investigation may contribute to the design and construction of high energy synchrotron radiation facilities with long beamlines.

View full abstract

Experiments were made on the oscillation of a single spike of a magnetic fluid droplet under a vertical alternating magnetic field. A kind of resonance was found to occur in certain ranges of the frequency and the amplitude of the magnetic field. Three types of oscillation patterns were observed, i.e., mode 1 with one node, mode 2 with two nodes and a peculiar oscillation between these modes. The frequencies of models 1 and 2 were twice that of the magnetic field, while the frequency of the peculiar oscillation was the same as that of the magnetic field.

View full abstract

An approximate analysis was made of the oscillation of a single spike of a magnetic fluid droplet under a vertical alternating magnetic field. The shape of the spike was assumed to be semiellipsoidal. Two resonances were found to occur, which corresponded to mode 1 and the peculiar oscillation observed in a previous experiment. The resonance frequencies of both oscillations decreased with the increase in both the amplitude of the magnetic field and the volume of the droplet.

View full abstract

A numerical simulation was made of the oscillation of a spike of a magnetic fluid droplet under a vertical alternating magnetic field, and the time-dependent changes of the droplet shape and the inner flow field were estimated. The resonance was found to occur when the timings of the changes of the magnetic field and the radial flow motion coincide. A peculiar oscillation occurs when an annular peak is formed exactly at the maximum of the magnetic held. The oscillation of mode 2 occurs in when the magnetic field becomes its maximum exactly before the formation of the annular peak. The oscillation of mode 1 occurs when the single peak is formed exactly at the maximum of the magnetic field.

View full abstract

A new type of self-induced free surface oscillation (swell flapping) was discovered in a system with a jet, a free surface and a structure. An upward round jet was injected into a cylindrical tank from the bottom center. The jet impinged on the bottom of a cylindrical rod (UIS) which was set just beneath the free surface, resulting in separation of the jet. In a certain condition, the separated jet oscillated, forming a flapping swell of the free surface around UIS, i.e., swell flapping. The inlet jet did not oscillate. The onset condition of the swell flapping depended mainly on the jet inlet velocity and UIS depth. The frequency of the swell flapping was different from that of sloshing, and depended mainly on jet inlet velocity and UIS depth. The oscillation was considered to be excited by the interaction between the separated jet and the free surface.

View full abstract

Vortex structure in the stagnation region of an excited plane impinging jet has been experimentally investigated by digital particle image velocimetry (DPIV). Jet excitation was utilized to lock the phase of the vortex motion and all the component of phase averaged vorticity were obtained. Selective amplification of the wall jet stream component of RMS vorticity on the stagnation streamline was recognized. Phase-averaged vorticity isosurfaces revealed that the rib vortices in the braid region and vortex potato just downstream of the spanwise rollers merge into the counter-rotating vortex pairs in the vicinity of the wall. The vorticity of the vortex pairs was augmented by the continuous stretching in the stagnation region. Production terms in the transport equation of the phase-averaged vorticity and principal axis of the strain rate tensor were evaluated to understand the transport mechanism of seeds of the counter-rotating vortices.

View full abstract

The structures of two opposed supersonic free jets with the same and different source pressures are studied by numerical simulation using the piecewise linear method (PLM) based on Godunov's method. For different source pressures, the ratio between the source pressures is set at 1.0 or 2.0. The ratio of the higher source pressure to the back pressure is varied between 50 and l00. The simulation results are analyzed and show agreement with the theoretical and experimetal results with respect to the flow field structures, internal structure of the interacting region and the positions of the shock waves. A dead air region is formed in the vicinity of the slip line. The flow field structure depends significantly on the pressure in the interacting region, where the pressures of the two jets balance.

View full abstract

In the present paper, drag reduction in turbulent pipe flow of water by means of implanting microfibers, instead of injecting or premixing them, has been investigated. Experiments using FRP pipes with microfibers implanted on the inside wall have been conducted at various Reynolds numbers (2000-40000), as well as with various types of microfibers. A maximum drag reduction of about 50% has been obtained using one type of microfiber, and the most effective length of microfibers in terms of the law governig the wall variables has been determined to be 6-8 wall units. The direction of implanted microfibers also was found to have an effect on the drag reduction of pipe flow.

View full abstract

The fundamental problem of the decelerating laminar boundary layer defined in the title has been analyzed theoretically. The original, phenomenological viscous diffusion theory and the numerical analysis using a rational nondimensionalizing scaling technique based on the said phenomenological theory, which were both developed in our previous research for the similar unsteady problem of a linearly retarded semi-infinite flat plate, are applied successively to the present problem. As the result the interesting flow behaviors of the decelerating stagnation boundary layer over the entire time region from the distant past to the deceleration terminal at which the wall has just stopped, and further up to the distant future where the wall is continuously at rest, have been clarified not only numerically but also phenomenologically.

View full abstract

The fundamental problem of the decelerating laminar boundary layer defined in the title has been analyzed theoretically. The original, phenomenological viscous diffusion theory and the numerical analysis using a rational nondimensionalizing scaling technique based on the said phenomenological theory, which were both developed in our previous research for the simpler, linearly retarded flows on a semi-infinite flat plate and near a front stagnation point, are expanded and applied successively to the present problem. As the result the interesting flow behaviors of the decelerating stagnation boundary layer over the whole time region from the far past to the deceleration terminal at which the wall has just stopped, and further up to the distant future where the wall is permanently at rest, have been clarified not only numerically but also phenomenologically

View full abstract

In this study, GS-SGS energy transfer is investigated based on a DNS data base of three-dimensional homogeneous isotropic turbulence and turbulent mixing layer. In the case with a Gaussian filter, the Bardina model for the Cross term predicts the overall features of the energy transfer, but shows large energy transfer in the high-wave-number region. For the Reynolds term, the Smagorinsky model shows large energy transfer compared with the exact one. The Bardina model for the Reynolds term shows the same tendency as is observed for the Cross term. In the case with a Gaussian and sharp cut-off filter which corresponds to the real LES, almost all models for the SGS stress show incorrect GS-SGS energy transfer. The Bardina model for the Cross and Reynolds terms also predicts the overall features of the energy transfer even for the turbulent mixing layer, but shows energy transfer from SGS to GS in the high-wave-number region.

View full abstract

A three-dimensional particle tracking velocimetry (3D PTV) technique has been developed, which enables us to obtain markedly larger number of velocity vectors than can be obtained by previous techniques. Instead of the usual stereoscopic image recordings, the present 3D PTV technique visualizes an entire three-dimensional flow with scanning laser-light sheets generated from a pair of optical scanners and the images are taken by a high-speed video recording system synchronized with the scannings. Digital image analyses for the derivation of velocity components are based on a numerical procedure, in which several improvements have been made in terms of the extraction of particle images, the determination of their positions, and the derivation of velocity components. The present 3D PTV technique was applied to rotating fluids accompanied by Ekman boundary layers and their complicated secondary flow patterns as well as the primary circulations are quantitatively captured.

View full abstract

We made an experimental and a theoretical investigation on the turbulent noise generated by a bladed multiple-disk fan. In the theory, we assumed that the turbulent noise consists of two sources: the turbulence of boundary layer on the disk surface and the vortex shedding from the trailing edge of the blades, and that these two sources are independent of each other. To estimate the acoustic power due to the latter source, we proposed a new method to calculate the wake width which is an important parameter to control the noise level. The effects of the four parameters: the span length of the blade, the number of blades, the inner radius of the impeller and the setting angle of the blade, on the turbulent noise were investigated. The agreement between the theoretical and the experimental results was satisfactory.

View full abstract

When a compression wave generated by a train entering a tunnel propagates through the tunnel and arrives at the tunnel exit, an impulsive pressure wave (micro-pressure wave or tunnel sonic boom) is radiated from the tunnel exit. Since the impulsive wave is closely related to the form of the compression wave arriving at the tunnel exit, improvement of the train nose shape is one means of reducing the impulsive wave. In this work, numerical optimization of the train nose shape for reducing the impulsive wave is performed by nonlinear programming combined with numerical simulation. The cross-sectional area distribution of the train nose is determined by the nonlinear programming, while the effect of the train nose shape on the form of the compression wave is estimated by the numerical simulation which solves the unsteady axisymmetric compressible Euler equations by finite volume method, TVD scheme and sliding grid method. The effect of the nose shape obtained by the optimization is confirmed by experiments using models.

View full abstract

A numerical analysis is developed for sound reduction by cavities in a duct, focusing on the effect of cavity length on the transmitted sound. A linearized unsteady Euler equation for acoustic disturbances is solved by a finite difference scheme to obtain the multidimensional structures of the entire sound field. The validity of the present method is verified using a solution based on the plane wave theory and an available analytical solution. The present results show that the level of transmitted sound is reduced at specific values of the cavity length, with tendencies opposite those obtained using the plane wave theory. The results also indicate that a series of cavities yields a more effective reduction of the transmitted sound level.

View full abstract

A flow mechanism that leads to the onset of positively sloped head-flow characteristics (referred to as stall onset) was investigated for a series of unshrouded mixed-flow pump impellers having the nominal specific speed of 560 (rpm, m³/min, m). A criterion that indicates the stall onset was proposed using the critical value of a theoretical pump head. The stall onset was caused by the discontinuous reduction of the theoretical pump head due to the blockage effects of impeller inlet recirculation, which was induced by the extensive flow separation on the casing wall at the latter half of the impeller. The stall onset was prevented when the impeller inlet recirculation occurred before the theoretical pump head reached its critical value as the flow rate was reduced. The blockage effects of the impeller inlet recirculation, in this case, developed continuously and decreased the diffusion of the flow within the impeller flow passages, thus suppressing the onset of extensive flow separation on the casing wall.

View full abstract

In order to improve spanwise deflecting flow generated near the last stage of an axial-flow turbomachinery in a partial flow rate operating range, a cascade with shape-controlled airfoils actuated by shape memory alloys (SMAs) has been developed. A two-way change in the blade shape is realized by using spring and SMA rods in which three different shapes are memorized. The flow field around a developed cascade was measured in a wind tunnel in order to determine the effect of the shape change. As a result of this study, it is proven that a cascade developed with shape controlled airfoils is capable of suppressing spanwise deflecting flow by changing the blade shape and restricting the flow passage. It is effective especially to a steam turbine in which SMA plays the roles of a temperature sensor as well as an actuator because the temperature of the exhaust steam is dependent on the operational conditions.

View full abstract

A test of liquid flow through semipermeable membranes was done in an attempt to utilize the salinity difference between sea water and pure water for electric power generation. Pure water and salt water were filled separately into two chambers divided by semipermeable membranes of a reverse osmosis (RO) module. Salt water was forced to flow along the semipermeable membranes by a pump. Pure water was passed through the semipermeable membranes and into the salt water chamber by osmotic pressure. The pressure difference and the flow rate across the flow control valve installed in the pipeline of the pure water chamber were measured. The valve served as a load for a water turbine. Compared with theoretical considerations, the measured values of pressure difference and flow rate suggest the high possibility of generating electric power by utilizing the system proposed here.

View full abstract

Critical Heat Flux (CHF) of subcooled flow boiling with water in a tube with internal twisted tape under a nonuniform heating condition was experimentally investigated by direct current heating of a stainless steel tube. The boiling curve of the subcooled flow under a high heat flux was measured to confirm the characteristics of the nucleate boiling curve. The Net Vapor Generation (NVG) point almost agreed with Levy correlation. By assuming that the bubble boundary layer disrupts when the heat flux is lower than the NVG heat flux, the increasing rate of the CHF with internal twisted tape under a nonuniform heating condition was explained assuming an alternate development and disruption of bubble boundary layer.

View full abstract

The solution process of a single gas bubble in liquid was simulated by directly solving gas diffusion equation in liquid, the vorticity transport equation and the Poisson equation for the stream function in order to estimate the effect of natural convection due to the difference in concentration on the solution process. Solution times of a single gas bubble under a plate were measured experimentally in order to estimate the diffusion coefficient of gas in liquid. Both the experimental and theoretical results reveal that solution time is affected by the natural convection due to the concentration difference in the case of large initial bubble radius. In the region where the effect of the density induced natural convection is not large, the process of solution is governed by two nondimensional parameters and consequently, the conditions where the process of solution is not affected by natural convection are clarified.

View full abstract

In order to clarify the mechanism of heat and mass transports in drying and wetting processes of a mortar concrete slab, a numerical analysis is presented based on microscale structures consisting of gel pores, capillaries, and macrocavities. The capillary pressure is estimated using the pore density function measured by the mercury injection method, and the unsaturated permeability is calculated by introducing the connection rate of capillaries. The numerical results of the unsteady state saturation are in good agreement with those of the experiments in both drying and wetting processes. The liquid and vapor fluxes and the gas and liquid pressures are also discussed in detail.

View full abstract

This paper is concerned with the theoretical basis for the determination of the condensation coefficient of vapor by means of a shock tube. Film condensation on the shock-tube endwall behind a reflected shock wave is analysed on the basis of the first author's gas-dynamics theory. It is clarified that there exists a transition phenomenon during the growth of the liquid film, that is, the liquid film grows in proportion to time immediately after the reflection of the shock wave, and after a transition time, it grows in proportion to the square root of time. The transition phenomenon between these growths is caused by the change in heat conduction characteristics at the endwall. The reason why the condensation coefficient must be determined before the transition is clarified.

View full abstract

Convection heat transfer and pressure drop measurements were performed with a rectangular duct, having a cooled upper and a heated lower surface, which was packed with spherical particles. Air was used as the test fluid and four kinds of spherical particles having different diameters and thermal conductivities were used as the packing materials. The ratio of the diameter of the spherical particle to the distance between the cooled and the heated surfaces, d/H, was varied from 0.173 to 1. The thermal conductivity of the particle layer was also measured under the still air condition. The thermal conductivity of the particle layer was not affected by the value of d/H. In the case of the one stage arrangement of spherical particles (d/H = 1), the flow resistance took on a remarkably small value compared with the flow resistance of a homogeneous spherical particle layer. Moreover the flow resistance of the particle layer formed with some layers of particles was able to be predicted by combining the flow characteristic of the one stage particle layer and that of the homogeneous spherical particle layer. The heat transfer coefficient of the particle layer was larger than that of turbulent air flow on a flat plate. At a constant superficial air velocity, there existed the value of d/H which gave a maximum value of the average heat transfer coefficient. Nondimensional heat transfer correlation equations were derived in terms of parameters expressing the average characteristics of the spherical particle layers.

View full abstract

Three-dimensional simulations of laminar separated and reattached flow and heat transfer over a blunt flat plate in a channel are presented. Numerical calculations of Navier-Stokes equations and energy equation are carried out using the finite difference method. The results of three-dimensional calculations are compared with the two-dimensional ones, and show the effects of the endwall. It is clarified from the numerical results that the reattachment length increases with Reynolds number and the flow in the recirculation region becomes three-dimensional. The reattachment line is curved by the endwall effects. Two-dimensionality of the flow is reduced as Reynolds number increases. The horseshoe vortex near the endwall greatly influences the heat transfer in the redeveloping flow region.

View full abstract

To avoid the complicated system and additional pump power required for a gas-solid suspension flow, the authors proposed a new concept of an impinging air jet laden with solid particles recirculating in an enclosure, and the heat transfer characteristics were experimentally studied. As the first step of this study, this paper deals with the flow characteristics of gas-solid suspension in the enclosure. Since a preliminary experiment showed that the configuration of the cell and the position of the gas exhaust hole strongly affect the flow pattern and the impinging particle volume fraction, gas-solid two-phase flow within two types of cells, namely parabolic plane cell and inclined plane cell, was examined. In addition to the observation of flow pattern and measurement of pressure drop, the volume fraction distribution of particles was obtained through an optical experiment. By comparing the results of the two cells, it was clearly shown that the flow pattern within the parabolic plane cell is a smooth two phase flow pattern and thus the volume fraction of impinging particles is higher than that of the inclined plane cell. The results enable us to predict that the parabolic plane cell has more favorable characteristics for heat transfer enhancement.

View full abstract

This paper deals with the heat transfer characteristics of the impinging air jet laden with solid particles recirculating in an enclosure, which was proposed in our previous paper. The particles enclosed in the test cell were glass, aluminum and ceramic with the diameter ranging from 50μm to 600μm, and a 50μm thick stainless steel foil glued on the bottom of the test cell was used as a heat transfer surface. Results clearly show that the heat transfer coefficient in the cell containing solid particles is about 30% higher than that of the same cell without particles; the heat transfer coefficient of the gas-solid suspension within the cell is about 120% higher than the single phase flow of a conventional impinging air jet without the cell. Under our experimental conditions, it can be considered that the heat tranfer enhancement observed in the present study is mainly caused by the increase in turbulence of gas flow due to particle motion, since the heat transfer coefficient is hardly affected by the volume fraction of solid particles in the stagnation region, which is influenced by the number of particles enclosed in the test cell.

View full abstract

Two-dmensional solidification of mixtures with supercooling was studied experimentally using aqueous solution tilled in filling a circular freezing cell. It was clarified that the solidification process consisted of the following characteristic stages: free crystal growth in a supercooling state, recovery of the recalescence temperature field with development of a mushy zone, and solidification under the quasi-steady state condition. Microbehaviors such as the crystal growth velocity and their number density were also clarified in connection with the supercooled fields and the geometrical configuration of the freezing cell.

View full abstract

Enhancement of heat transfer was investigated experimentally on natural convection adjacent to a vertical heated plate. In order to promote heat transfer from the heated plate, a V-shaped diverter plate of which the edge faced upstream was attached onto the surface of the vertical plate. The diverter plate redirects high-temperature fluids toward both sides of the plate and introduces low temperature ambient fluids behind the plate instead. These two mechanisms enhance heat transfer, in particular, in the region behind the diverter plate. Local heat transfer coefficients around the diverter plate were measured using water as a test fluid. The coefficients behind the diverter plate reached 2.5 times of those without the diverter plate. The optimum heights and angles of the the diverter plate that make the heat transfer maximum were also studied experimentally.

View full abstract

A numerical analysis of convective heat transfer has been performed for a three-dimensional turbulent flow developing in a 180°-bent tube with straight inlet and outlet sections using several two equation heat transfer models. The calculated results for three types of turbulent model for temperature field are compared with the available experimental data. Althouth the calculated results using the three models predict well the experimental data of time-averaged temperature, one of these models shows a tendency to overpredict the experimental result. Near the bend inlet the decrease of local Nusselt number, which is caused by laminarization, is observed experimentally at the inner wall. By comparing this decrease of local Nusselt number with the calculated results, we find some discrepancies among the heat transfer models. As regards the difference among the models, the eddy diffusivity for heat is examined quantitatively. As a result of this examination, it is clarified that a reasonable prediction is dependent on the model of the eddy diffusivity for heat and that the mixed time scale, which is a parameter of the eddy diffusivity for heat, plays an important role in predicting the temperature distributions. Therefore, it is necessary to propose on elaborated mixed time scale in order to predict precisely the convective heat transfer in both the near-wall region and a region remote from the wall,

View full abstract

The effects of free stream turbulence on turbulent boundary layer were calculated using a k-εtwo-equation model. The calculations were performed with respect to velocity profiles on a flat plate, wall shear stress, turbulence energy, integral length scales of turbulence and decay of free stream turbulence, and the results were compared with experimental results. The energy of free stream turbulence and the dissipation values at the leading edge of flat plate were used, as the initial conditions for calculation. These initial values of dissipation were determined from the integral length scales of free stream turbulence at the leading edge. The calculated wall shear stress increased with the free stream turbulence and integral length scales of turbulence. The velocity profiles and turbulence energy agreed well with the experimental results, and the effects of free stream turbulence on the wall shear stress agreed fairly well with those observed in experiments.

View full abstract

The effects of free stream turbulence on turbulent heat transfer were calculated with two equation model for heat. The calculations were performed with respect to turbulent boundary layers along a flat plate with a uniform wall heat flux. From the measured values of near wall thermal turbulence, an improvement on turbulence model function has been made in turbulent heat flux equation. The results for the heat transfer rate, logarithmic distribution of temperature profile, the eddy diffusivity of heat and turbulent Prandtl number agreed comparatively well with the experimental values.

View full abstract

This paper describes an experimental and a semi-empirical study of impingement cooling characteristics of heat sinks with longitudinal fins, which are suitable for LSI packages. The experiments are performed with a variety of longitudinal fins. For enhancing impingement cooling, one slot-shaped inlet orifice (slit) over the center of the heat sink is found to be the best structure. The optimum orifice width is about 1/6 of base width of the heat sink. The thermal resistance at the same volumetric flow rate and orifice width varies little with size of a gap between the fin top and inlet orifice ; however the gap of about 2 mm slightly reduces it. Correlations are proposed between the thermal resistance of the heat sink and the longitudinal fins. The estimation error for the predicted thermal resistance is found to be within ±25% of the experimental data.

View full abstract

This paper reports on the impingement cooling characteristics of a heat sink with thin longitudinal fins of 0.2 mm thickness, which are spaced with a fin-pitch of 0.5∼2.0 mm. Air cooling the heat sinks comes from slot-shaped orifices positioned above the heat-sink centers. The gap between the fin tops and the inlet orifice is 0∼10 mm. The thermal resistance of thin longitudinal fins is about 50∼57% of that of the thick longitudinal fins now in commercial use. The cooling performance of thin-plate fins is almost the same as that of optimally arranged pin fins with the same total surface area. A maximum value 6 times the heat transfer rate of a flat plate with the same base area was observed for the thin-plate fins. The performance of impingement cooling systems is almost unaffected by the gap between the fin tops and the inlet orifice (or, for channel cooling, the upper wall). On the other hand, performance of channel-cooling systems decreases markedly with widening of the gap.

View full abstract

We present an experimental study on the effects of thermal conditions on sink-mark generation of press-formed glass products. In experiments, the depth of sink-marks of glass was investigated with variation of the weight of the glass, the initial temperature of the glass, the initial temperature of a plunger, the pressing time, and the pressing pressure. The obtained results can be summarized as follows (1) The depth of the sink- mark increases linearly with the weight of the glass. This is explained by the fact that the cause of the sink-mark is basically shrinking of the glass. (2) It decreases exponentially with lower initial temperature of the glass. This is explained by the fact that the viscosity of the glass increases exponentially with lower temperature. (3) It is reduced with lower initial temperature of the plunger, longer pressing time and higher pressing pressure. This is explained by the fact that the glass surface, where the sink-mark is generated, is cooled more effectively under these conditions. (4) It is most Influenced by the initial temperature of the plunger, and least influenced by the pressing pressure.

View full abstract

The thermal behavior in a stratified heat storage tank was analyzed during the charging process. The present model is characterized by the application of a strict boundary condition in the tank outlet and well represents a real system. The mixing effect, which is known to disrupt the stratification, was also included in the analysis. The solution was obtained in two forms, exact (long-time) and approximate (short-time), since the former is difficult to apply to a general problem because of slow convergence and complicated calculation. The results were compared to that of Cabelli; our solution can be applied more generally, and includs completely that of Cabelli.

View full abstract

Transient characteristics of a ractangular latent heat storage vessel packed with shape-stabilized phase change (solid-liquid) material(PCM) are investigated numerically with the finite difference technique. It is found that the heat storage characteristics are greatly affected by the flow direction of the heat transfer medium since the natural and forced convection coexist in the heat storage vessel, That is, it is clarified that the effective thermal efficiency of the latent heat storage system is obtained with downward flow along a vertical PCM for the heat storage process. The effects of the inlet velocity and the inlet temperature of the heat transfer medium on transient heat characteristics of the latent heat storage system are also revealed in the present study.

View full abstract

Recently, the absorption refrigerator has attracted much attention because it does not use freon as the coolant and it can be driven by waste heat. The absorption refrigerator contains much solution and coolant in it. Therefore, it requires a long start-up time because of the large heat capacities of the solution and the coolant. Moreover, it is known that much more heat is necessary at start-up than in steady-state operation. If it is driven by waste hot water of e. g. a combustion gas engine, the hot water return temperature becomes low at the start-up which adversely affects the engine. The purpose of this study is to construct an analytical model of the start-up characteristics of the single-effect absorption refrigerator driven by waste hot water, which can be used as a tool for investigating the problem of the start-up. Moreover, the validity of this model is confirmed by comparing the simulation results with the experimental results.

View full abstract

The objective of this series of investigations is to develop compact heat pumps which exhibit sufficient heat output performance and high COP at temperatures below -10°C. This report investigates the extent of improvement in efficiency, system size and the maximum heat output by the combination of a heat storage system with a gas-engine heat pump (HS-GHP system). From a cycle simulation analysis, it was found that the HS-GHP has the potential to exhibit better efficiency and significantly high heat output with a smaller system size compared to the conventional GHP system. This improved efficiency was realized by the storage of latent heat of liquid working fluid at the exit of a condenser without increasing any compression work.

View full abstract

Three dimensional measurements of velocity and gas flux measurements of O₂, CO₂ flux near burner of oil furnace were demonstrated to show a flame holding mechanism and fluctuation characteristics. The use of convective flux in flame was not accurate but found to be applicable in whole combustion region in a furnace. Two pairs of recirculating flow region rotated in swirl direction due to the slit flow and its timescale was about 5-10 msec. The timescale of turbulence and CO, CO₂ gas variation were compared in order to show the flame holding mechanism in a transparent evaporating region. The comparison results show that the timescale of turbulence and chemical species fluctuation were the same order.

View full abstract

The OH radicals and soot in an unsteady spray flame, which was achieved in a rapid compression machine, were visualized simultaneously by the laser-induced fluorescence and the laser-induced scattering techniques, The fuel mixture consisting of 90% paraffin hydrocarbon (reference fuel) and l0% polypropylene glycol was used to reduce the optical attenuation caused by dense soot cloud. The simultaneous images of the fluorescence from OH radicals and the scattering from soot show that the soot and OH radicals exist separately from each other in the leading portion of the spray flame, and neither soot nor OH radicals exist in the center part of the quasisteady region of spray during injection.

View full abstract

To study cycle-to-cycle variation of diesel spray direction, spray shape, needle valve motion and sac inner flow of multi-hole-type nozzles were imaged using a high-speed drum camera and two CCD cameras. It was observed that needle valve offset and one-sided sac inner flow during fuel injection induced the fluctuation of spray direction and deformation of spray shape. Experiments on the effect of cycle-to-cycle variation of the spray on emissions under no load and lower speed of the engine revealed that the suppression of the variation yielded reduced hydrocarbon emission and white smoke.

View full abstract

Spray flames inherently show complicated behaviors due to the group-forming character of droplets and the preferential flame propagation through the most favorable routes. In order to observe the detailed structure of spray flames, the light emission signals in OH- and CH-bands, the Mie scattering signals from droplets, and the size and velocity of droplets with a phase Doppler anemometer (PDA) were monitored simultaneously in a premixed spray flame stabilized by an annular hydrogen pilot flame. It was experimentally confirmed that the homogeneous premixed spray was divided into a number of droplet clusters due to the preferential flame propagation in the process of combustion, while there existed no droplet clusters in an approaching nonburning premixed spray flow. The group combustion number for each droplet cluster Gc was estimated from the present experimental data. It was found that the mean value of Gc fell within the range of the external and internal group combustion mode according to the definition of Chiu et al., and that the observed burning behavior of clusters corresponded well to their classification of group combustion mode.

View full abstract