An experimental study was carried out on vortices issued from a triangular prism. The experiments were conducted in a hydraulic flume using laser Doppler velocimetry (LDV) and the ensemble-averaging technique to map the unsteady velocity field about the vortex-generating region. From this study, transitional characteristics of vortices issued from a bluff body were clarified. In particular, the dissipating phenomenon of vorticity in the vorticity-concentrating region was clarified. Lastly, behavior of entrained fluid particles carrying vorticity from the separated shear layer was clarified.

View full abstract

The laminar pulsatile flow through a two-dimensional channel constructed of a fixed wall and a moving wall is calculated numerically using the finite difference method. The main purpose of this calculation is to evaluate the effects of the phasa lag between the flow pulsation and the wall movement on the wall shear stress and flow pattern. The calculated results are summarized as follows: i) The phase lag strongly affects the shear stress of the fixed wall. ii) The flow separation of the fixed wall can be prevented by selecting the appropriate phase lag. iii) The phase lag does not strongly affect the separation bubble length of the moving wall.

View full abstract

The objective of this study is to establish the law of resistance for uniform flow of a Bingham fluid in a rectangular open channel. The fluids used in experiments were mixtures of water and bentonite. Their physical properties, plastic viscosity and yield stress, were determined through an experiment on pipe flow, and it was confirmed that the fluids were Bingham fluids. An experiment on uniform flow was carried out using these fluids. An extended Reynolds number was newly defined and the adaptability of the Reynolds number to the Moody diagram was confirmed by the experiment. Consequently, the resistance laws of uniform flow for Bingham fluids were proposed.

View full abstract

The molecular theory of Doi is employed to analyze numerically the unsteady simple shear flow of polymeric liquid crystals. Both the shear stress and the first normal stress difference exhibit an overshoot during the transient behavior. The order parameter also shows an overshoot phenomenon, which is almost the same as that for the first normal stress difference. Contrary to these quantities, the preferred angle of molecules does not express an overshoot but decreases monotonically with time. The overshoot rate decreases remarkably with the mean field potential, which implies that an overshoot phenomenon is not observed for strongly oriented polymeric liquid crystals. While the orientational distribution function under equilibrium conditions takes a symmetric form, it is sharpened and its symmetry is broken in the shear flow. Less molecules exist above the angular position, where the orientational distribution function is maximum, than below the angular position.

View full abstract

In this study, the flow fields of two, three or four interacting parallel supersonic free jets are studied by flow visualization using the planar-laser-induced fluorescence of iodine molecules seeded in the test gas. Centers of orifices are set on a line, vertexes of a triangle or square. The flow fields are visualized on the plane including the jet centerlines and in the cross sections vertical to the centerlines. The flow field structures are clarified using various visualized pictures for each geometrical arrangement of the orifices. In particular, three-dimensional structures of the flow fields do not tend downstream to the structures which are expected from the arrangement of the orifices

View full abstract

The kinetic energy conservation third-order upwind finite difference scheme is used for the convection terms in three-dimensional Navier-Stokes equations to directly simulate the turbulent impinging jet issuing from a round nozzle. The calculated results of mean velocity and turbulent intensity distributions and autocorrelations are compared with experimental data, which are measured using the hot-wire anemometry system. The cell size is roughly six times the Kolmogorovmicroscale; however, calculated turbulent characteristics represent the experimental results fairly well. This means that the simulation well describes the motion of large-scale eddies, which play an important role in the enhancement of heat and mass transfer in the flows. The various features of turbulent impinging jet flow are demonstrated.

View full abstract

The domain decomposition technique, which is widely used in the solution of a flow field on multiprocessors, requires a treatment of new boundaries appearing between subdomains. Poor convergence tends to be introduced due to these new boundaries where otherwise very frequent communication between neighboring processors is required. We determined how frequently processors should communicate with each other so as to obtain the best efficiency. In this study, steady 2-D incompressible laminar and turbulent flows are solved on the transputer system equipped with distributed-type memory. The SIMPLE method is employed as a numerical algorithm for the incompressible flow solution. The performance of parallel programs, such as the efficiency and increase in speed, is discussed for the cavity and sudden expansion flows.

View full abstract

The unsteady flow fields in a narrow-gap flow between two corotating circular disks are investigated numerically by solving the time-dependent Navier-Stokes equations through the SMAC-FEM (simplified marker and cell finite-element method). In each triangular torus element, quadratic velocity elements and a linear pressure element are used. The conclusions are as follows. (1) When disks suddenly start rotating, they induce a secondary flow in the gap flow. As the primary swirling flow pattern develops into a nearly solid rotation, the secondary flow grows and then becomes very weak. (2) Since the viscous force is weaker in the gap flow with larger Reynolds number, the secondary flow is stronger in the early period. (3) The torque coefficients on both sides of a rotating disk decrease rapidly and then become constant. These constants are much smaller than that for a rotating disk in a stationary housing.

View full abstract

This research aims at development of a three dimensional Lagrangian stochastic process model of a particle-pair for predicting the statistics of the scalar fluctuation diffusion field in a non -stationary homogeneous turbulence. In this paper, the three-dimensional stochastic particle-pair model for stationary homogeneous turbulence has been improved so that it is applicable to the non-stationary case. Main points of improvement are as follows; (1) The temporal Lagrangian variation of the velocity fluctuation scale σ_v is incorporated into the Ito's stochastic equation. (2) The effective sphere of the two-particle velocity correlation has been modeled. As a simple example to which this new model can be applied, the scalar fluctuation field of a point source plume in decaying grid-generated turbulence was simulated. The improved model gives a better prediction of the actual scalar fluctuation properties than stationary homogeneous models.

View full abstract

Budgets of turbulent heat flux and temperature variance were calculated by direct numerical simulation of decaying homogeneous isotropic turbulence with heat transport. The computations were executed on 96³ grid points by a spectral method. With a constant mean temperature gradient imposed in one direction, dependence of the thermal turbulence statistics on the Prandtl number, and the initial temperature variance was studied. The Reynolds number of the simulated turbulent flow field is low. The relative magnitude of the dissipation and the pressure temperature gradient correlation in the turbulent heat flux budget markedly changed at three different Prandtl numbers of 0.2, 0.71 and l.5.

View full abstract

A deposition simulator using the DSMC (direct simulation Monte Carlo) method has been developed to investigate deposition profile characteristics on small trenches in semiconductor manufacturing processes from vacuum to atmospheric pressure conditions. This simulator can be applied to several deposition processes such as sputter deposition, plasma chemical vapor deposition, and atomspheric-or low-pressure chemical vapor deposition. In particular, rarefied gas flow fields with low concentrations of reactive gases are calculated using a modified DSMC method (test particle method). In the case of low-pressure processes such as sputter deposition, upstream boundary conditions of the trenches can be calculated by means of flow field analysis in the equipment. The effects of upstream boundary conditions, molecular collisions, sticking coefficients and surface migration on deposition profiles in the trenches were clarified.

View full abstract

In our previous work, we have already computed backward-facing step flow using Large Eddy Simulation (LES) and obtained satisfactory results when compared with experimental data. In this paper, the behavior of time-averaged turbulence models and the influence of the diffusion model in the turbulent energy equation are investigated using the LES data base. It is found that the algebraic stress model describes this flow more accurately as compared with the eddy viscosity model, and that the gradient-diffusion model for the turbulent diffusion term cannot predict well the κ distribution in the recirculating region. A new diffusion model using TSDIA theory is applied and the relationship between the reattachment length and the turbulent energy distribution in the recirculating region is discussed.

View full abstract

The Pressure probe consisting of a fine static tube and a condenser microphone was developed to measure pressure fluctuations in the turbulent flows. The structure and the dimensions of the probe were determined to minimize the error in the measurements of pressure fluctuations. In order to damp the organ-pipe resonance in the static tube, a thin nylon gauze was inserted in front of the diaphragm of the microphone, and the probe exhibited a nearly flat response to 2.0kHz. The dynamic calibration of the probe was carried out in the wake of the cylinder by measuring pressure-and velocity-fluctuations simultaneously, and the measured pressure agreed well with the theoretical pressure predicted from the fluctuating velocity. The probe was also applied to the measurements of pressure fluctuations in jets, and the results suggest that the direct pressure measurements are very effective in detecting the vertical structures in the turbulent shear flows.

View full abstract

The Agency of Industrial Science and Technology, Ministry of International Trade and Industry has been researching and developing three types of 300kW ceramic gas turbines since 1988. The turbine blade's Reynolds numbers based on the basic design are predicted to be in the region of l0⁴∼10⁵. In the field of wind turbines, for example, some studies on the characteristics of airfoils at low Reynolds numbers have been carried out. However, it is not yet obvious how the turbulence affects the characteristics of airfoils. This paper describes the experimental results obtained with use of turbulence screens. It is found that laminar seperated flows, in some cases, reattaches to airfoils by raising turbulence intensity of free stream at Reynolds numbers of 0.5×10⁵ and 1×10⁵, and that lift to drag ratios increase further.

View full abstract

A supercavitating hydrofoil placed in vertical and spanwise shear flow between two parallel planes is approximated to a lifting-line. By introducing two kinds of separation of variables into both a fundamental solution and a main stream velocity distribution, an equation of motion with respect to disturbance pressure is reduced to two ordinary differential equations. The lifting-line equation is obtained from up-wash velocity distribution along a lifting-line derived from solutions of ordinary differential equations. Hydrodynamic force characteristics are expressed and calculated by series summation, whose spectra are solutions of the lifting-line equation, and then cavity distributions are obtained by applying the method proposed by the authors.

View full abstract

Flow behavior of a rotating stall is studied numerically by means of the vortex model which was developed to solve the flows through stalled cascades. The flow features and pressure distributions, as well as the aerodynamic forces and moments exerted on the blades are examined. It is clarified that the strong suction due to the stall and unstall vortices has a considerable influence on the unsteady aerodynamic characteristics of the blades. The development of rotating stall of large amplitudes from various initial disturbances is also shown. Some important aspects of the rotating stall such as inception point, the hysteresis, the number of stall cells and its timewise changes are discussed and compared with the results deduced from a conventional linearized actuator disk theory.

View full abstract

An experimental method to measure the unsteady aerodynamic forces and moments acting on the vibrating circular cascading blades which rotate about the main shaft is proposed. Here, we consider that the cascade is composed of two small-scale circular cascades, one of them is harmonically vibrated and the other is interrupted. Unsteady forces and moments acting on a vibrating blade and a standing blade are measured. This manipulation is repeated for the cascades exchanging the motion between two small-scale cascades. Then, the unsteady forces and moments of the cascading blades producing harmonic vibration are obtained by linearly superposing on the measured datum for each amplitude of the blades, hence, the data of interferences between the cascades are obtained. The results obtained using this method were in the good agreement with one by the conventional method and it was applicable to the determination of the unsteady forces and moments on the cascade.

View full abstract

The double linearization theory is extended to a three-dimensional straight cascade between parallel-plane end walls operating at supersonic axial velocity. Numerical examples are presented to show unsteady aerodynamic responses and flutter boundaries in correlation with cascade parameters. The three-dimensional effects are evaluated by comparing them with the strip theory predictions. To attain a large steady loading on the stability boundary of the first bending vibration, it is advantageous to design blades with angle of attack and camber decreasing from hub to tip. The three-dimensional effects are small, and the strip theory was found to be a useful method for estimation of the unsteady aerodynamic work of the three-dimensional straight cascade.

View full abstract

An experimental study of both noise and aerodynamic characteristics of a multiple-disk fan was carried out with respect to the effects of three parameters: (1) the existence of blades, (2) the gap of disks, (3) the location of blades. It was shown that the bladed multiple-disk fan exhibited higher pressure head and the fan efficiency than the non-bladed multiple-disk fan, though the noise generated from the former was higher by 3 to 6 dB than that of the latter. The fan efficiency and noise increased proportional to the gap of disks, and it was experimentally shown that a 0.7mm gap gave the lowest specific noise level among the impellers tested in this experiment. Concerning the location of blade, when the distance between the trailing edge of the blade and the cutoff was larger than 13mm, the discrete frequency noise generated by the interference between the blade and the cutoff became lower than the level of turbulent noise. We also obtained the experimental result that the multiple-disk fan with a 13mm distance between the blade and the cutoff had the lowest specific noise level.

View full abstract

The effects of four parameters-the setting angle of the blade, the thickness of the disk, the inner radius of the disk and the number of blades-on both noise and fluid dynamic characteristics of bladed multiple-disk fans were experimentally investigated and discussed, with consideration of the measured velocity field at the outlet of the impeller. The results showed that as the setting angle increased, the pressure rise of the fan and the fan noise increased. A thicker disk produced a lower pressure rise of the fan. If the number of blades BB was less than 24, the sound power of the fan was proportional to the number of blades. However, this did not hold for BB of 36, because of the interference among blades. Judging from the specific noise level, it was recommended that the optimum design conditions be number of blades of approximately 24, the span length of 0.7mm and the setting angle of 35°.

View full abstract

Analytical solutions of the wave equation are presented for the reflection and the diffraction of a plane wave by a wedge at attack angle, and are compared with results of simulations for flow fields induced by weak shock waves. An algebraic procedure is proposed to generate grid systems fitting a wedge in which the grid is clustered near the apex of the wedge and the orthogonality holds on the wedge surface. This is an extension of conformal mapping and can simply produce grid systems for simulations. The explicit TVD method is applied to numerically solve the gasdynamic equation for the simulations. Comparisons between results of the simulations and the analytical solutions show that the analytical solutions of the wave equations can generally approximate the flow fields induced by reflection and diffraction of weak shock waves although deviations between the simulations and the analytical solutions are noticeable near the apex of the wedge.

View full abstract

In our previous paper we showed that when the passive boundary layer control method is adequately applied to the pseudo-shock wave in a rectangular duct, the pseudo-shock region can be changed to exhibit a flow pattern similar to that of a shockless state, which is suggested by L. Crocco as a limiting state of the dissipative phenomenon in the pseudo-shock wave. The present paper presents experimental results on the measurement of pressure oscillation of such a nearly shockless state. The upstream Mach number of the pseudo-shock is 1.80. It is shown that the pressure oscillation in the pseudo-shock region is greatly reduced by passive boundary layer control.

View full abstract

In order to investigate the variation in turbulent structure for the d-type rough wall boundary layer disturbed by a cylinder rod, the statistical properties in two cases have been examined experimentally. The rod was positioned at two distances of y_c/δ_o=0.14 (inner layer) and 0.92 (outer layer). For y_c/δ_o=0.14, the production rate is slightly greater than the dissipation rate at the upper shear layer of the cylinder, whereas for y_c/δ_o=0.92 the dissipation rate is significantly larger than the production rate at both upper and lower shear layers of the cylinder. Both skewness and flatness factors are different from those of the undisturbed flow in the region of y/δ≲ 0.2 for y_c/δ_o=0.14 and in the region of y/δ≲ 0.2 for y_c/δ_o=0.92, respectively. The higher-order moments including Reynolds shear stress flux show that the influence of the cylinder disturbance appears as high amplitude of the v fluctuating component for both cases.

View full abstract

It is demonstrated by our numerical experiments that the wavy numerical errors, which are produced when the inviscid Burgers equation is solved by the particle method, are solitons. The error introduced by the Gaussian cored particle method is of the order of the third derivative which works as the dispersion term. As a result, the solitons are created. An empirical equation of the relation between the amplitude of the soliton and the dispersion error is obtained. This equation enables us to solve the K-dV equation. The relation between the diffusion term of the Burgers equation and the dispersion error is also studied.

View full abstract

Aerodynamic acoustic problems near plate edges are treated by a new method based on the linear theory. For sound radiation and generation problems, we can calculate one of the half acoustic fields divided by semi-infinite plates by determining the distributions of the acoustic monopoles. Some numerical calculations confirm the validity of this distributed monopole method. Applying this method, we can impose the Kutta condition explicitly at the trailing edge and explain the feedback mechanism in terms of sound wave effects on the flow field. The generation of vorticity waves at the trailing edge due to the incident sound wave is calculated and the possibility of self-excited tones between the trailing edge and the leading edge is verified. The relationship obtained is the same as that of the edge tone phenomenon, and the necessary amplification of the vortex during its convection is quantified for the sound to become self-excited.

View full abstract

Sound source analysis is numerically carried out for aerodynamic sound generated by uniform flow over a square cylinder at a Reynolds number of 10⁴. The incompressible time-dependent turbulent flow is first calculated by means of Large Eddy Simulation and the sound generation at the far field is then obtained by surface integral on the square cylinder using the Curle equation modified for low Mach number flows. The sound source intensity distributions on the square cylinder, including negative sound source intensity regions, are shown for three observation points. The time-averaged and instantaneous sound source intensity maps can be used to detect dominant noise origin and to consider the relationship between flow and sound generation. It is demonstrated that estimation of the sound source by observing the large-scale structure of flow is difficult. This is because sound is generated by small timewise fluctuation of the flow. Approximate predictions for the sound source and sound directivity based on timewise fluctuations of the drag and lift are also shown.

View full abstract

This paper concerns the development of an impulse turbine with a self-pitch-controlled guide vane, which was proposed by the authors to be used as a wave power generator. In order to clarify the effects of the rotor blade profile and guide vane pitch/chord ratio on the characteristics of this turbine, experimental investigations were performed using small-scale equipment. These include three types of turbine blade rotors with different blade inlet angles γ and two types of guide vanes, that is, splitter type and monovane type. The results show that a high efficiency impulse turbine can be achieved for rotor blade profile with γ=60°. It is also found that, in this case, the optimum pitch/chord ratio of the guide vane (S_g/l_g) is about 0.8 for the splitter type, and S_g/l_g<0.65 for the monovane type. Furthermore, the effects of the nozzle setting angle and diffuser setting angle on the turbine characteristics have been clarified.

View full abstract

The vane compressor used for automotive air conditioners maintains contact between a vane tip and a cylinder wall by centrifugal force and backpressure acting on the vane under steady-state operation. However, under start-up operation, the backpressure is not sufficient to maintain the vane tip contact, and a chattering phenomenon which causes noise and partial wear occurs. We analyzed vane behavior under start-up operation by applying an equation of motion to the vane. We also observed the vane behavior after start-up by visualization in the compressor cylinder. The influences of oil shearing force, inertial force due to rotor acceleration and the backpressure on the vane behavior during start-up operation were clarified. The validity of the analysis was confirmed by comparison of calculated results with experimental ones.

View full abstract

A basic experimental study is carried out to clarify the magnetic field effect on the dynamic characteristics of a magnetic fluid viscous damper. The steady or alternating magnetic field is applied vertically or longitudinally, respectively, to a pipe axis of the damper, which induces a variation of mass concentration of magnetic fluid. The amplitude ratio of output to input oscillations of the damper is measured to examine the magnetic field effect on the dynamic characteristics. It is clarified that the dynamic characteristics are strongly influenced by the applied magnetic field according to the variation of mass concentration of magnetic fluid.

View full abstract

As an approach to the experimental study of various unsteady forces acting on a hydrofoil with cavitation, we have designed a torsional vibration apparatus for a cavitating hydrofoil. The system has a driving part producing sine wave vibration, and two multicomponent load cells for measuring the unsteady fluid forces and canceling the mechanical inertia forces. This paper reports the mechanical and measuring accuracies of the system for future experimental studies and, for example, refers to various steady and unsteady fluid forces in noncavitating or cavitating flow.

View full abstract

The vibration mechanism of a water-type Stirling engine is studied experimentally and theoretically. The theoretical analysis indicates that the effect of the loss coefficients is negligible, and the nonlinear equations derived in the previous paper are simplified. The experimental studies indicate that the vapor pressure in wet conditions plays an important role in the vibration characteristics. Furthermore, a stability analysis was conducted to explain the critical condition of vibration, for comparison of the p-v characteristics curves under self-excited and non-vibration conditions.

View full abstract

Experiments on air/water two-phase flow performance were carried out using a centrifugal pump with five kinds of closed impellers, each of which has a different outlet or inlet blade angle. Results are described as follows. (1) A sudden head drop due to gas accumulation in the impeller occurs at a low inlet void fraction with high inlet or outlet blade angle of impellers. (2) A degraded pump head at a high inlet void fraction becomes higher with increasing outlet blade angle. These results are discussed in terms of numerical calculations of one-dimensional two-phase flow.

View full abstract

Conditional sampling and the point-averaging technique have been introduced to particle-tracking velocimetry using a binary cross-correlation algorithm. This system is applied to the measurement of unsteady flow inside a rotating Savonius rotor, and the quantitative nature of the power mechanisms of the rotor at various rotor angles and tip-speed ratios are newly discussed and the usefulness of this techique is demonstrated. It is found that all velocity fields in and around the rotor vary markedly with the tip-speed ratio, and that the separation of Coanda-like flow on the convex side of the advancing blade is delayed by the velocity fields induced by the rotor rotation and also by the production of vortices downstream of the advancing blade.

View full abstract

Fluid flow in a small model of a Francis-type pump turbine was visualized by means of a tracer method and was imaged using a video camera rotating with the rotor. The relative velocity vectors of the fluid flow were analyzed from the video image using a digital image processing technique based on the correlation method. The flow pattern was discussed in relation to the characteristics of the pump turbine. At a zero incidence condition the reverse flow zone on the suction surface of the blade was much less than that at a shockless inlet condition. An unsteady flow pattern at a low flow rate was also demonstrated. The problems in the image processing technique and countermeasures such as image improvement and size selection of the correlation matrix were also discussed.

View full abstract

Most reciprocating screw injection molding machines are equipped with a back-flow valve which prevents back flow of plasticated material. Behavior of the back-flow valve during the injection process and its effect on the variations in injection moldings mass were investigated through the analysis of data obtained from the measurements of screw position and nozzle pressure. This simpler method proved to be useful in the estimation of screw displacement, nozzle pressure and material leakage which are required for the closure of the back-flow valve. Factors influencing the injection moldings mass variations were clarified from the material leakage for consecutive runs estimated by applying this method and the corresponding measured injection moldings mass.

View full abstract

Experiments on the countercurrent two-phase flow of air and water were conducted using vertical pipes of 10 to 26mm in diameter, to investigate the initiation conditions of liquid ascent. When the liquid film flowed down to the bubbling two-phase mixture in the lower portion in the pipe, liquid ascent began at much lower gas velocities than usual flooding velocities without the bubbling two-phase mixture. In most cases, liquid ascent occurred in a slug flow state. The initiation conditions of liquid ascent were analyzed physically by considering two-phase mixture level swell fluctuating around the mean height.

View full abstract

Three kinds of beads - vinyl chloride, glass and steel beads - with different mean diameters were each packed in a rectangular cell. These packed beads were saturated with NaCl-solution and used as porous media. The solidification process where the porous medium was solidified from one vertical wall of the cell, was investigated experimentally and analytically. In the analysis, the permeability in a mushy region, which was considered to be one of the main factors governing the above solidification process, was experessed as the n-th power function of the volume fraction of the liquid phase in the mushy region. By discussing the results of the distributions of temperature, concentration and the volume fraction of the liquid phase and the results of streamline contours, the influences of the initial concentration of the solution and the mean diameter of beads on the present solidification process were clarified.

View full abstract

Transition processes from deflagration to detonation waves in stoichiometric oxyhydrogen mixtures diluted with nitrogen were observed using high-speed Schlieren photography as well as pressure and ionization current measurements. In this study, effects of obstacles on the transition processes were investigated. The obstacles were multigutter shaped and were installed near an ignition plug on an inner wall of a rectangular tube. Width and depth of the gutter were selected as parameters while composition of the mixture was fixed. It was revealed that there exist optimum values of these parameters for enhancing the transition to detonation.

View full abstract

Supercooling characteristics in quiescent bulk water enclosed in a circular tube were investigated on a basic level through numerous experiments. In the experiments, three kinds of water with different specific resistances were used as the test samples and their freezing temperatures were measured. The critical degree of supercooling depended on the property of the heat transfer surface and the cooling rate, and it became larger for a smoother surface and a higher cooling rate. It was found that the specific resistance of the water had little effect on the critical degree of supercooling . However, the insoluble particles included in the water greatly affected the ice nucleation in the supercooled water and the size effect of the insoluble particles was found to become important in the range of 0.1-1.0 μm in particle radius. For particle radii larger than this range, the critical degree of supercooling was independent of the size, while below this range it sharply increased with a decrease in the particle size.

View full abstract

Numerical analyses have been conducted for combined free and forced laminar convection of low-Prandtl-number fluids in a vertical pipe. The effects of both buoyancy force and axial heat conduction on the hydrodynamic and thermal characteristics are systematically investigated for the fluid heating and cooling with upward flow at Pe=0.7 to 71(Pγ=0.007 to 0.71 with Re=50 and 100) and |Gγ/Re|=10 to 500. At relatively high |Gγ/Re|, flow reversals are produced at the center of a pipe and near the wall in the heating and cooling cases, respectively. The regime of reverse flow is identified for both heating and cooling cases in Pe vs |Gγ/Re| coordinates. The axial conduction effect becomes significant and even dominant at very low Pe numbers. Numerical results are presented for velocity profiles, streamlines showing the occurrence of flow recirculation, temperature profiles, and distributions of the Nusselt number and wall shear stress with |Gγ/Re| as a parameter.

View full abstract

This paper deals with a new defrosting method in which a frost layer is melted by radiative heat energy as an energy source. The far-infrared radiative heat energy having a maximum wavelength of 5.5 μm is selected as the optimum radiative heat energy source for melting of the inclined frost-layer. The inclined frost-layer melting experiments using the radiative heat energy with a discharge of melted water on a cooling copper plate are carried out under various environmental parameters (inclination angle of frost-layer, radiative heat energy flux, air temperature, cooling brine temperature) including porosity of frost-layer as a frost structural factor. The dimensionless correlation equations which predict the time taken for complete frost-layer melting are derived as a function of various nondimensional frost-layer melting parameters.

View full abstract

The melting behavior of a snow layer was investigated experimentally and numerically for the case where the snow layer was melted from the upper surface using radiative heat absorption material (black calcium carbonate powder). The experiments and calculation were carried out under various conditions of sprinkling density of radiative heat absorption material, environmental temperature, radiation heat intensity and snow density. It was clarified that an optimum density of the absorption material existed for the enhancement of snow layer melting. With low sprinkling density of the absorption material, the exposed snow surface, due to the gathering effect of the absorption material, brought about a decrease of the snow melting rate. On the other hand, with high sprinkling density of the absorption material, the snow melting rate also decreased due to increase of the thermal insulation effect of the absorption material. Useful nondimensional correlation equations for snow melting were derived in accordance with the ranges of various parameters.

View full abstract

Heat transfer measurements were performed during forced and natural mixed convections of a rectangular open cavity which was packed with spherical particles arranged in a one step orthorhombic array. Air flowing over the cavity was heated from the bottom surface of the cavity via the particle layer. Three kinds of spherical particles having almost the same diameter of 10 mm and different thermal conductivities were used as the spherical packing material. The cavity depth was varied from 0 mm (flat plate) to 10 mm. The particles suppressed the air motion near the heating surface and decreased the heat transfer coefficient. In the case of particles having large thermal conductivity, those particles behaved as an extended heat transfer surface and turbulence promoter so that the heat transfer coefficient was enhanced. The Nusselt number ratio as a dimensionless heat transfer coefficient was expressed in terms of Reynolds number, ratio of particle diameter to depth of the cavity and modified Prandtl number.

View full abstract

Experiments have been carried out for measuring the average heat transfer coefficients for yawed circular cylinders, 5, 8 mm and 20 mm in diameter, using a wind tunnel. The effect of the yaw of a cylinder on heat transfer characteristics was examined for yaw angles of 16°∼90° at the Reynolds numbers from 2 × 10² to 5 × 10⁴. The flow visualization was also conducted to observe the flow patterns around a cylinder. From these results, experimental formulae were derived to estimate heat transfer from yawed circular cylinders.

View full abstract

Systematic measurements using a wind tunnel were conducted regarding pressure loss characteristics and heat transfer performance of pin-fin heat sinks exposed to air flow in a cross flow direction, varying the pin pitch as a parameter. The pin-fin heat sinks used in the present study were made of thin copper pins with square cross section arranged densely in a rectangular form. The modified single blow transient testing method was applied to the measurement. The flow visualization was also carried out with dye injection technique in order to investigate the flow pattern inside the pin-fin arrays. In the test results, the effects of the pin pitch both in the flow direction and transverse direction on the overall performance were clarified.

View full abstract

A parallel computational method is developed for the Monte Carlo analysis of radiative heat transfer in three-dimensional nongray gas enclosed by gray walls. The highly parallel computer AP1000 is used for the analysis. For the nongray gas, a mixture of water vapor and carbon dioxide is chosen as the absorbing-emitting medium. Three types of parallelization are studied: type (1), event parallelization; type (2), combination of event parallelization and algorithm parallelization, and type (3), memory-saving algorithm of type (2). When 512 cells are used, values of the speed-up ratios of 395, 430, and 370 are obtained for types (1), (2), and (3), respectively. The maximum number of elements which can be treated is increased from 1200 for type (1) to 140000 for type (3).

View full abstract

Heat fluxes from a supersonic high temperature inert gas flow to electrodes in a disk MHD generator are measured under the strong influence of Lorents force. When we compare heat fluxes from the inert gas flow to electrodes with and without magnetic field, it has been observed that strong MHD interaction significantly increase heat fluxes especially in the middle region of the disk generator. This increase of heat fluxes shows a strong correlation with rise of the static pressure induced by the MHD interaction. By raising the inlet temperature of the inert gas, the region with strong MHD interaction expands to upstream because of shortening of the ionization relaxation region. For the same static pressure rise, helium shows more significant increase of heat fluxes than argon, which may ascribed to an unstable fluctuated helium plasma.

View full abstract

The objective of this work is to understand the effects of several parameters on the blowing process involved in glass container production. The process was numerically simulated by the Simplified Marker And Cell (SMAC) method, assuming that glass was an incompressible viscous fluid. It was found that the glass-blowing process was strongly affected by the initial temperature and the viscous stress on the glass surface. It was proven, however, that the blowing process was little affected by the energy dissipation due to viscosity and the heat transfer to the mold.

View full abstract

Evaporation and combustion processes of a single fuel drop in ambient gas pressurized and heated beyond the critical point of the fuel (a supercritical environment) were observed from the lateral side of the drop on a flat hot wall settled in a high-pressure bomb. The ambient gas temperature near the wall was controlled to be the same as the wall temperature. The fuel drop showed different evaporation lifetimes in air and in nitrogen. The difference in the evaporation lifetime was larger under higher ambient pressure. Drop volume decreased gradually in a subcritical environment, but in a supercritical environment the drop maintained close to the initial volume during evaporation and showed a sudden decrease just before extinction. In an ambient temperature range around the critical tempreature of the fuel, longer duration of combustion and smaller flame width were found for the drop in the supercritical ambient pressure than for that in the subcritical ambient pressure.

View full abstract

Numerical calculation of the combustion process of a single fuel droplet immersed in an otherwise quiescent oxidizing gas whose pressure, P, exceeds the critical pressure of the fuel, P_cF, was performed. The burning time, t_b(P), takes a minimum value at P = P_CF. The underlying physics are that (i) at P< P_CF, for which the droplet has a liquid-gas interface throughout its lifetime, the gasification rate is controlled by the phase equilibrium condition, yielding decreasing t_b with increasing P; (ii) at P > P_CF, the droplet loses its surface immediately after ignition and the subsequent burning rate is controlled by the diffusion rates of the fuel and oxidizer, which yields increasing t_b because of the reduced diffusion coefficient with increasing P. To switch (i) to (ii) smoothly and continuously, the diffusion rate should be suppressed at around P= P_CF, which is accomplished under the condition of a vanishing diffusion coefficient at the critical state of the mixture. The behavior of gasification time is similar to that of the burning time, whereas it becomes minimum at a P significantly different from P_CF in the corresponding pure vaporization problem.

View full abstract