The Taylor vortex flow exhibits several flow patterns (modes) under the same conditions at small aspect ratios. The primary mode that is formed by increasing Re gradually from zero has a unique pattern with two normal cells having inward flow direction on the plate, and the secondary modes that are formed by increasing Re rapidly have several flow patterns with the normal cell or one or two anomalous cells that flow outward on the end plate. We used the flow visualization technique for the Taylor vortex flow and counted the number of modes appearing according to the various ways of increasing Re. The entropy was calculated, and the complication of the bifurcation of the modes was clarified. Finally, the processes of the formation of modes were illustrated and discussed.

View full abstract

In order to clarify structures of high-speed submerged water jets and to predict the jet working capacity, we systematically measured the cavitation noise around the jets using a hydrophone, for several nozzles. The power spectra and the rms values of the cavitation noise are carefully analyzed. Clearly, cavitation noise created by the jets is related to the behavior of cavitation bubble clouds, and the cavitation noise characteristics are markedly changed with the injection pressure and nozzle geometry.

View full abstract

High-Speed submerged water jets have two characteristic peaks with respect to the stand-off distance, in the same manner as high-speed water jets in air. In order to clarify the behavior of the impulsive pressures at these two peaks, we systematically carried out erosion tests using an aluminum-alloy specimen in a wide range of the injection pressure P₁ from 10 to 70 MPa, and microscopically observed the eroded surfaces by SEM. We find that the mass loss rate Δm^^· at the 1st peak is decidedly due to the water-column impact, so that it is very sensitive to the injection pressure P₁, while Δm^^· at the 2nd peak is due to the shock pressures of cavitation bubble collapse, so that it is not necessarily sensitive to P₁. We also indicate that the form and the speed of erosion change markedly with the differences in the jet structure with the stand-off distance x/D.

View full abstract

A correction formula is proposed for wall effects on unsteady separated flow around bluff bodies, which is derived from numerical results by the discrete vortex method on two-dimensional separated flows around an inclined flat plate, a square cylinder and an elliptic cylinder. The present formula is found to estimate reasonably well the wall effects on the mean and fluctuating flow characteristics in a wide range of the blockage ratio and the angle attack.

View full abstract

Steady and unsteady flow in a channel with an oscillating constriction is studied numerically and experimentally. A new one-dimensional flow theory is presented to describe the flow through the constriction. Both the flow separation and reattachment are included in the theory. The separated flow region is marked from a jet flow by a dividing streamline which is assumed on the basis of an available experimental observation. An experiment is performed using a two-dimensional rigid flow channel with the constriction oscillating sinusoidally with low frequencies to measure the relationship between the flow rate and the pressure distribution. Good agreement between experimental and numerical results has been obtained. Additionally, numerical simulation is carried out to evaluate the dynamic flow characteristics at a high frequency of the sinusoidal oscillation of the constriction which cannot be realized by the experimental model. The present theory is applicable to separated flow in rapidly diverging tubes such as glottis and collapsible tubes.

View full abstract

This paper describes numerical analysis for the flow behavior in the groove related to the turbulence of the flow over rows of two-dimensional ribs on the ground plane for various values of S/D using a finite-element method. In this analysis, the two-dimensional Navier-Stokes equation was formulated on stream function ψ and vorticity ω by the Galerkin method. The value of stream function at each time was calculated using a sub-implicit scheme. As a result, it was found that the results of the numerical analysis were close to those of the flow visualization using a water channel, and that this analysis was adequate to investigate the flow over rows of two-dimensional ribs on the ground plane. It was also confirmed that stable vortices were existent in the grooves for S/D≤5 and the vortices were unstable behind the ribs for S/D≥7.

View full abstract

A detailed experimental investigation was carried out to examine the three-dimensional boundary layer characteristics in a radial inflow turbine scroll. Some basic flow phenomena and growth of secondary flow were also investigated. In the inlet region of the scroll, the incoming boundary layer begins to assume a skewed nature, namely the radially inward secondary flow caused by the radial pressure gradient. From the inlet region to one-third of the scroll circumference, the secondary flow increases largely that most of the low-momentum fluid in the incoming boundary layer is transported to the nozzle region. The succeeding elimination of the low-momentum fluid in the boundary layer suppresses growth of the boundary layer further downstream, where the boundary layer shows a self-similar velocity profile. The distributions of the boundary layer properties in the scroll correspond well to those of the flow properties at the nozzle. The behavior of the boundary layer in the scroll is found to affect the circumferential nonuniformity of the nozzle flow field.

View full abstract

In this paper, molecular dynamical analyses of an atomic beam passing through a slit are presented. In the case where microscopic phenomena, such as in rarefied gas flow involving wall interaction as well as intermolecular collision, must be taken into account, the molecular dynamical method (MD method) is effective to analyze the individual behavior of constituent particles such as atoms, molecules or ions. Since the atomic beam here is highly rarefied, beam atoms suffer, at most, a few interatomic collisions in the process. Therefore, collision with the slit surface is cosidered to be more dominant than interatomic collision. Wall interaction has been modeled numerically, for example, as mirror reflection or Maxwell reflection, although such conventional models, in many cases, do not describe the real interaction with the wall surface. Here, a mixed reflection model is employed in order to simulate the beam propagation phenomena through a slit.

View full abstract

Flows through parallel Plates containing a cylinder are numerically calculated for nematic liquid crystalline polymers using the simplified model of the Leslie-Ericksen theory. A tensor expression for the orientation of the director is used in the basic equations in order to avoid the difficulty in calculation due to the tumbling of the director. The orientation of the director is affected by the contraction and expansion flows, and tumbling occurs. An area where the molecule aligns in the direction of the flow exists beyond the cylinder. It extends downstream as the parameter λ approaches the value of one. The flow field does not change significantly from that of Newtonian fluids.

View full abstract

The effect of uniform suction or injection on the stability of free-convection boundary layer over a vertical uniform flux plate is theoretically investigated. The characteristics of the free-convection boundary layer flow with uniform suction or injection are calculated. The perturbation equation describing the temporal evolution of the disturbances is solved numerically for the velocity and temperature profiles obtained, and the stability curves are presented for various values of suction/injection parameters. The critical Grashof number G_C increases when the suction parameter increases, while G_C decreases when injection parameter increases. This implies that the free-convection boundary layer with uniform suction becomes stable.

View full abstract

Previous investigations on the flow in diffusers suggest that study on the flow characteristics in the elbow draft tubes of Francis turbines, especially in the case of strong swirling flow involved, is still required. As one of the steps to investigate the surge phenomena in the elbow draft tubes, we carried out a series of time-averaged measurements in a conventional draft tube, which was installed in an air test rig, by using a five-hole probe. Flow pattern and pressure distributions in the draft tube were studied in the range of swirl rate m_ex from 0 to 1.8. The variations of area-mean pressure and mass-averaged pressure at each measuring section are discussed, and the effect of swirl rate m_ex on the pressure recovery coefficients of a typical draft tube is also shown.

View full abstract

An experimental study is carried out for a swirling flow pneumatic conveying system in a vertical pipeline in terms of the power consumption and additional pressure drop. The test section consists of a pipeline 12m in height with an inside diameter of 80 mm. The three initial swirl numbers from 0.58 to 1.12, the average gas velocities from 9 to 23m/s and the solid-gas ratios from 1.5 to 23 were investigated. Polystyrene, polyethylene and polyvinyl pellets with mean diameters of 1.7, 3.1 and 4.3 mm, respectively, were transported as test particles. It is found that the critical and minimum gas velocities of swirling flow pneumatic conveying were lower than those of axial flow pneumatic conveying, the maximum reduction rates being 13% and 20%, respectively. In the lower gas velocity range, the power consumption and additional pressure drop were decreased by the swirling flow pneumatic conveying.

View full abstract

The purpose of this research is to examine the details of hysteresis which appears in the pressure-flow relation of collapsible tubes such as veins exposed to negative transmural pressure. To exclude human error in setting the flow rate, an electric control valve was introduced and was automatically controlled with a personal computer. Tube deformation was measured simultaneously with a laser displacement meter to obtain the displacement distributions. The hysteresis in which the pressure drop for increasing flow is larger than that for decreasing flow can be seen in the range of low flow rate. The tube shows self-excited oscillations for increasing flow and does not oscillate for decreasing flow in the range of high flow rate. It was found that the hysteresis for the low flow range is caused not only by elastic hysteresis but also by fluid dynamic hysteresis. It was also found that the lack of state uniqueness in the high flow range is caused by multiplicity which is due to different initial conditions for a nonlinear system.

View full abstract

New inflow and outflow boundary conditions are presented. These conditions are tested for two incompressibility problems. One is a problem in which a vortex convected by a uniform mean flow passes an outflow boundary. Another is a two-dimensional spatially developing mixing layer. New boundary conditions which give highly accurate results for both test problems consist of convection-viscous conditions and an approximate transport equation for pressure. The applicability of the convection-viscous conditions is verified by numerical tests which show that conventional boundary conditions do not approximate the velocity or pressure conditions on the boundary. The pressure transport equation proposed in this work is used to obtain the pressure at the inflow and outflow boundaries from a previous time step pressure field. As a result, we conclude that the new boundary conditions are appropriate for direct numerical simulations of spatially developing flows.

View full abstract

The presence of water droplets in the flow of a steam turbine leads to erosion of moving blades because of the large blade tip speed and higher local wetness. Since erosion resistance is strongly influenced by the impact droplet velocity and size as well as the hardness of the blade material, estimating the distributions of droplet size and frequency is important for establishing erosion criteria. In this paper we describe an estimation method for droplet size of dispersed water flow in the wake of a fixed blade, which can be extended to a moving blade. The LDV system was applied to measure velocity and size of droplets in the wake near the trailing edge of a fixed blade, under conditions simulating those found in the last stage of a low-pressure steam turbine. Test results show that droplet sizes in the wake of a fixed blade can be approximately evaluated by a normal distribution function depending on the mean droplet diameter. It is concluded that the size and frequency distributions of the droplets which impact on a moving blade can be calculated using these characteristics.

View full abstract

A k-ε model has been reconstructed with the aid of direct numerical simulation (DNS) data. In the present model, particular attention is paid to obtaining the best near-wall solutions for both the turbulent kinetic energy and its dissipation rate. The proposed model was applied to the prediction of channel and boundary layer flows, and the relevant solutions were compared with the DNS data as well as the results calculated with existing recent models. These comparisons have proven that the proposed model has sufficient capability to mimic the DNS data, while the results from other models, except for one recent model, are not very satisfactory.

View full abstract

The arrangement of tubes in a tube-bundle straightener which is used as an element in laminar flowmeters and a component in flow devices can be classified into two different patterns ; a staggered matrix and a side-by-side matrix. Fluid resistance in the straightener depends on the arrangement, the length of the tubes and Reynolds number. In this experiment, we found that a noticeable change of the pressure-loss coefficient for each matrix is caused by the transition from laminar to turbulent flow in the circular tubes and the star conduits which are elements of the tube-bundle straightener. Using conventional hot-wire measurement at the exit of the tubes, it was also observed that turbulent slag appeared at different Reynolds numbers for each circular tube and star conduit.

View full abstract

An experimental study of a fully developed turbulent flow in an elliptical duct with an aspect ratio 2 : 1 was carrid out using a Laser-Doppler anemometer. Special attention was paid for the measurement of Reynolds stresses and the secondary flow of the second kind. In addition to these results, axial mean velocity and turbulent intensity of the central region were measured in the developing flow. Although it is very difficult to measure the secondary flow of the second kind because of its small intensity, this measurement system was able to determine a large eddy caused by the secondary flow in a quarter cross section. At the same time, a clear statement of the distribution of five of the six Reynolds stresses was clarified, showing their characteristic aspects. The axial mean velocity of the central region reaches maximum intensity in the middle of developing. This maximum intensity is a result of shear layer interaction effects as the boundary layers developing along the wall of the elliptical duct begin to merge.

View full abstract

Measurements of turbulent quantities are presented for a low-Reynolds-number turbulent two-dimensional rotating channel flow. Turbulence intensities, Reynolds shear stress, correlation coefficient, skewness and flatness factors, four-quadrant classification and power spectra are considered. Contributions of production and Coriolis terms in Reynolds stress transport equations are estimated. Coriolis terms in the transport equations of normal Reynolds stress are very small in comparison with production term, even in the case where mean velocity profiles are influenced by the effect of the Coriolis force. However, the contribution of the Coriolis term is great in the transport of Reynolds shear stress.

View full abstract

Mechanisms of particle dispersion have been experimentally clarified in wake regions of obstacles. The present experiments have been performed in the wake regions of a pair of circular cylinders and trianglar cylinders. Detailed measurements of particle and gas velocities and particle number density are presented using a three-beam laser Doppler velocimeter. Particles of which mean diameter was 42μm were released into the flow at the center of the gaps between two cylinders and the initial point of the wake region of the trianglar cylinders. Particles are affected by the large-scale eddy in the wake region, and particle diffusivity is well correlated with the Stokes number which is defined as the ratio of the particle relaxation time to the characteristic time scale of the turbulence. This result indicates that the mechanism of the interaction between particles and large-scale eddies is independent of flow configuration.

View full abstract

A three-dimensional particle tracking velocimeter for turbulent air flow measurement was developed. As a tracer particle, spherical plastic capsules and He-filled bubbles were tested. The plastic capsule's traceability was anticipated to be insufficient for high-frequency components of turbulence because its density was thirty times larger than that of air, its diameter was small enough compared with the turbulent microscale, and the image was clear enough to track. On the other hand, the bubble, which was neutrally buoyant, had a relatively large diameter and its image had two peaks in brightness. The previous 3-D PTV system was modified so that two successive instantaneous particle images could be recorded on two successive TV fields with an arbitrary time interval. In order to evaluate the present technique, a fully developed turbulent air flow in a square-cross-section duct was measured. Cross-stream distributions of the mean velocities and of the full Reynolds stress tensor components were obtained and some of them were compared with a separate hot-wire measurement. The agreement was generally good for both tracers. However, part of the plastic particles adhered to the bottom wall of the duct and caused serious errors in the measurement.

View full abstract

Higher-order far-field boundary conditions utilizing the Riemann-like variables have been applied for solving the two-dimensional compressible Euler equations. The governing equations are discretized by a central finite-difference approximation and integrated by the rational Runge-Kutta (RRK) scheme. Higher-order far-field computational boundary conditions in external flows presented by Verhoff et al. are incorporated into the basic scheme in order to investigate the effects on the numerical solution. Numerical calculations are carried out for external flows past a NACA 0012 airfoil at the uniform flow Mach number M_∞=0.55, angle of attack α=2°, M_∞=0.8 and α=1.25°.Compared with the conventional low-order far-field boundary conditions, the present method gives a comparable lift coefficient (C_L) and pressure coefficient (C_P) with a much smaller number of grid points, while it shows a slower convergence rate. It is found that this yields reduction of the number of grid points by approximately 50∼70% and total CPU time by 50∼70% in obtaining comparable results.

View full abstract

Vane compressors used for automotive air conditioners maintain 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 contact, and chattering phenomena which cause noise and partial wear occur. We theoretically analyzed the vane behavior under start-up operation with pressure rise in the back chamber due to discharge flow from the compressor. The validity of the theoretical analysis was confirmed by comparison with experimental results. The initial position of the vanes and initial condition of fluid in the back chamber greatly affect the vane behavior after start-up. Flow resistance of a channel, through which fluid in the back chamber communicates with the discharge fluid, also affect the vane behavior under start-up operation.

View full abstract

In order to increase the noise-reducing effect of a perforated tube, a plug is combined with a perforated tube. The plug is designed to change an underexpanded jet into a parallel flow. We derived the expression of a sonic line in an annular nozzle. The flow field downstream of the sonic line was, then, calculated straightforwardly by the method of characteristics. A couple of plugs were designed, manufactured and tested on convergent nozzles. Acoustical measurements and optical observation with a Toepler-Schlieren system were carried out. The plug was able to reduce shock strength but did not cancel it. The combination of a plug and a perforated tube considerably improved the noise-reducing effect of a plug nozzle.

View full abstract

A theoretical study is carried out to clarify the possibility of the application of the temperature -sensitive electrically conducting magnetic fluid to the working fluid in a liquid-metal MHD power -generation system. The equations governing a one-dimensional boiling two-phase duct flow of liquid-metal-based magnetic fluid which contains Fe-Ni alloy particles in a transverse magnetic field are numerically solved. The analytical results of the boiling two-phase flow characteristics and power-generation characteristics of the magnetic fluid are compared with those of an electrically conducting nonmagnetic fluid. It is found that better driving force or power-generation characteristics are obtained using the temperature-sensitive electrically conducting magnetic fluid as a working fluid. These results on the characteristics of the electrically conducting magnetic fluid will contribute to the development of a new MHD power-generation system using boiling two-phase magnetic fluid flows.

View full abstract

For detecting the rotating stall observed at a low flow rate in a centrifugal blower with a backward-leaning blade impeller, blade-to-blade variation of velocity fluctuation was measured in the impeller, immediately downstream of the impeller and in the vaneless diffuser. Measurement was conducted by means of the spot-focus-type semiconductor laser 2-focus velocimeter developed by the authors. This L2F velocimeter has a narrow-band spatial-filter effect within ±1.5 degrees of flow angle fluctuation. The fluctuations of velocity and flow angle were analyzed on the basis of the probability density distribution. The change of the blade-to-blade relative velocity distribution due to flow rate was clearly shown, and it was determined that the origin of rotating stall was in the vaneless diffuser because the blade-to-blade velocity fluctuation distribution in the impeller was hardly affected by the large velocity fluctuation due to the rotating stall which occurred in the vaneless diffuser.

View full abstract

Fundamental knowledge for detecting and predicting the inception of rotating stall was investigated. The measured blade-to-blade variation of velocity fluctuation was analyzed on the conception of the flow-fluctuation ellipse which indicates the fluctuation ranges of velocity vector and correlation coefficient based on the covariance between the tangential and radial components of velocity fluctuation. It was cleanly indicated that, (1) the inception of rotating stall was closely related to the three-dimensional separation of boundary layer on the diffuser wall, and (2) the rotating stall inception could be predicted experimentally by detecting the variation of correlation coefficient at the impeller exit. That is, it is presumed that the rotating stall will occur if the value of correlation coefficient becomes close to zero and then larger than zero, on if the local separation zone originated in the vaneless diffuser reaches the impeller exit.

View full abstract

A fan using the Weis-Fogh mechanism was built and its characteristics were studied. Two wings were set to move out-of-phase by an arm which rotates changing its directions in a square channel. Characteristics for two types of arms and wings were investigated. The best performance was achieved with a combination of an elastic arm and elastic wings which elastically deform. The Weis-Fogh mechanism is shown to be applicable to fans as well as pumps or propulsion equipment for ships. However, further investigations are required for diminishing the noise and vibration.

View full abstract

A series of experimental studies was made on the suction characteristics of a water ring vacuum pump sucking air with 100% humidity at 20, 30, and 40°C, which were identical to the water ring temperature. It was found that the characteristic curves at different temperatures agreed quite well with each other regarding the partial pressure of air. However, the characteristics are considerably inferior to the characteristics observed in conventional tests, where air is dry in the suction pipe. It means that the partial pressure of air in the impeller at the suction port is not in equiliblium with the partial pressure of air in the suction pipe.

View full abstract

Effects of outlet blade angle and rotational speed on the two-phase-flow pump performance are discussed based on the results calculated by a three-dimensional bubbly flow model with fixed cavity. Critical volumetric flow ratio of air to whole fluid, at which the fixed cavity on the shroud near the impeller inlet section expands rapidly to fill the section, decreases with an increase in the outlet blade angle and increases with an increase in the rotational speed. To predict the theoretical head. for the higher volumetric flow ratio, the one-dimensional two-fluid model by Furuya is also applied and modified for low-specific-speed pumps.

View full abstract

In this study, we show that hybrid oils including dispersed graphite whiskers, graphite fibers and natural graphite powders show marked effects in the reduction of the sound power level of high-frequency noise which is caused by teeth sliding against each other under a high-pressure condition in an accelerated gear pump. Furthermore, the source of this noise is analyzed and evaluated on the kind of materials used as solid lubricans in this study.

View full abstract

Critical heat flux during forced convective subcooled boiling on a rectangular heated surface of 40 or 60 mm in length and 7 mm in width, which is supplied with subcooled liquids through a small round jet which impinges at the center of the surface has been studied experimentally employing three different liquids (water, R113, and R22) at pressures from 1 to 25 bar under high subcooling condition up to 115 K. A flow model based on observation of the change in the flow aspect with increasing heat flux shows that the flow state can be divided into two regions : first, a subcooled region including weak boiling where the liquid is kept subcooled ; and second, a saturated region where the liquid is kept at the saturation temperature and most of the liquid splashes out as droplets. A criterion is proposed that the CHF in the saturated region can be determined from the existing equation for the CHF for the saturated liquid, while for the same heat flux in the subcooled region, the equation can be derived from a simple heat balance under the assumption that the subcooled liquid is heated to the saturation temperature at the leading edge of the saturation region. By eliminating the length of the subcooled region from the two equations and after revising them slightly, a generalized correlation can be derived, which yields about 82% of the CHF data with an accuracy of ±20%.

View full abstract

An experimental study has been made of the critical heat flux (CHF) of natural convective boiling on uniformly heated inner tubes in vertical annular tubes. The experiment was performed at a pressure of P=0.1 to 3.1 MPa for the clearance of 0.4 to 4.0 mm, the heated tube diameter of 5 to 10.6 mm, the annular tube length of L=58 to 840 mm and three kinds of liquids. The effects of the pressure in which the density ratio of the test fluid ρ_ν/ρ_l varies from 6.24×10^-4 to 0.16 and of the ratio of L/D_he=2 to 500, where an equivalent heated length, D_he, serves as a parameter connecting the clearance with the heated tube diameter, on the CHF, are mainly discussed. The experiment shows that the CHF obtained depends on the ratio of L/D_he and then a generalized correlation can be derived which predicts the CHF data well.

View full abstract

The aspect of condensation shifts from dropwise to film condensation as the cooling intensity increases on a lyophobic surface. The condensation curve, which shows the change of the heat flux against the surface subcooling, exhibits the characteristic N-shape. In order to understand the characteristics of the condensation curve, the measurement of the local heat transfer quantities for ethylene glycol vapor under the various vapor temperatures and velocities and the observation of the aspect of condensate were performed using the relatively long surface concerning the effect of the condensate flow. On the basis of both the results of those experiments and the previous report for propylene glycol vapor, the dependencies of the maximum heat flux on the position of the heat transfer surface and the quantitative relations among the basic parameters determining the characteristics of local condensation curve were clarified.

View full abstract

Operating limit of a closed two-phase thermosyphon was experimentally investigated with a mixture of ethanol and water as the working fluid. Even if the heat input was less than the maximum value, two kinds of oscillations of the system pressure p and temperatures of both the heating and cooling walls and fluid were observed ; i. e., (1) all the temperatures oscillated in phase with p (oscillation I ), and (2) the heating wall temperature at the wall of the liquid pool region oscillated out of phase with p (oscillation II ) . These oscillations, which did not appear in the case of the pure working fluid, were considered to occur due to the intermittent violent boiling of the pool of the binary liquid in the heating section. At the onset of the operating limit, five types of increasing patterns of the increase in temperatures and p were found. In three cases, dryout of the heating wall in the liquid pool was initiated at first and triggered by the violent oscillation II, which was the unique feature in the case of a binary working fluid.

View full abstract

Three-dimensional natural convection heat transfer in a saturated porous cube heated from below is analyzed by a finite-element method. The Rayleigh numbers between 50 and 140 are considered. When the Darcy-modified Rayleigh number is high, the convection pattern can take various modes in the cube for a given value of the Rayleigh number. It is found that the convection modes of (1, 0, 1) and (1, 1, 1) are the only stable ones below Ra = 100. Moreover, the (1, 0, 2) mode also becomes stable when Ra is greater than 100. The other modes, however, are unstable at 50≤ Ra ≤100 in spite of the fact that the corresponding Rayleigh numbers exceed critical values according to the linear stability theory. The Nusselt numbers for respective stable modes are also calculated.

View full abstract

The governing equations for turbulent flow through porous media consisting of packed spheres are examined and a O-equation model is proposed. In the process of deriving the governing equations by using the local volume-averaging technique, we consider the eddy viscosity (thermal conductivity) as the algebraic sum of the eddy viscosities (thermal conductivities) estimated from two types of vortices : (1) interstitial vortices of the order of the thickness of the gap width, √(K), which develop in micro-boundary layers between particles, and (2) pseudo vortices of the order of the particle diameter, d_p, which reflect the forced distortion of flow due to interruption of the solid particles. Furthermore, we propose the O-equation model, and show that the present turbulence model can predict the flow and heat-transfer characteristics at high Reynolds number, i. e., Forchheimer's flow resistance and the dispersion effect in thermal conductivity, and these phenomena can be explained in terms of vortex transport in porous media.

View full abstract

The temperature and the streamwise and normal velocities in the free-convection boundary layer along the vertical plate with uniform heat generation in air were measured along with the surface temperature on the plate. The downstream growth of the temperature and velocity disturbances from the end of the laminar to the beginning of the fully turbulent region were investigated, considering the initial disturbance level. When the initial disturbance level was relatively low, the intensity profiles of the velosities in the transition region had two peaks, which were previously found by Gebhart et al. in the streamwise velocity intensity profiles using water. However, when the downstream growth of the disturbances was compared using X^*(=Gγ^*1/4_x) as a streamwise dimensionless distance, the present growth rates of the temperature and velocity disturbances were found to be 5∼6 times larger than those for the previous case of water.

View full abstract

The vibrational energy of molecules in phase-changing processes was studied by measuring the infrared absorption spectra with FTIR (Fourier transform infrared) spectrometer. IR spectra of (NO)₂ (ν₁, ν₄), CO₂(ν₂, ν₃), N₂O(ν₁, ν₂, ν₃), and C₂H₂ (ν₃, ν₅) were Compared in the gas, Cluster, and solid phases. Absorptions of solid-phase molecules are observed at a blue or red shifted band from those in the gas phase, but in the clustering phase they are always blue shifted compared with those in solid phase. This leads to two kinds of vibrational energy changes in the phase-changing process from gas to solid ; blue and red shift, and red and red shift. These results observed can be explained reasonably by the vibrational mechanism of the ε, σ and m effects which characterize the effective potential and mass of molecules.

View full abstract

A numerical estimation was performed for the laminar flow and heat transfer by a porous medium in a flow passage with a miter bend. Heat transfer between solid material and fluid was considered in the numerical analysis. The calculations were performed for the Darcy number Da = 1∼2.5×10^-6, the ratio of thermal conductivity of solid material and fluid RRM = 1∼1000, the parameter of heat transfer between solid material and fluid H_g = 10∼1000 and the Reynolds number Re = 300. Local heat transfer and dimensionless stream function were compared between the flow with and without porous medium. The behavior of the flow, the solid and gas temperatures and the Nusselt number with these parameters was represented. It was found that the porous medium with low Da, high RRM and H_g should be selected to enhance the local heat transfer.

View full abstract

The purpose of this report is to predict the cavitation intensity using sound pressure on a heating surface, through experimental investigation and numerical analysis. Sound pressure measured using a small hydrophone was not in accordance with the cavitation intensity because of the difference in acoustic impedance of the hydrophone and water. Nevertheless, the sound pressure calculated by the finitely bounded Green's function can predict the measured cavitation intensity profile. It is elucidated through this calculation that the larger the sound pressure on the heating surface became, the greater was the cavitation intensity, and that the cavitation intensity on the side wall was far less than that on the bottom.

View full abstract

The mechanism of heat-transfer enhancement achieved by the artificial introduction of large-scale longitudinal vortical structures is clarified experimentally. The longitudinal vortices are generated by placing a single half-delta wing vortex generator in an otherwise laminar boundary layer on a flat plate. A rotating probe technique with a miniature slant hot-wire sensor is employed for the measurement of highly skewed three-dimensional velocity fields downstream of the vortex generator. It is shown that the local maxima of heat-transfer enhancement correspond well to the downwash side of the vortical structures, where boundary-layer thinning takes place as a result of entrainment of high-speed outer-layer fluid. On the upwash side of the vortices where heat-transfer enhancement is insignificant, energetic velocity fluctuations are found to appear as a result of onset of local turbulence transition, which then leads to the substantial improvement in heat transfer further downstream. From a detailed analysis of power spectra and probability density distributions of the streamwise velocity fluctuations measured with an I-probe, it is revealed that turbulence characteristics in the local turbulence transition regions are analogous to those reported for a flat-plate turbulent boundary layer, particularly when the wall normal positions are normalized appropriately either by the local viscous length scale or the local boundary layer thickness.

View full abstract

An experimental investigation on wire-coil-inserted tubes as a turbulence promoter has been undertaken for enhancement of heat transfer and reduction of flow friction factor in water flow. The performance of wire-coil-inserted tubes has been evaluated in terms of two different criteria based on the fixed flow rate and pumping power. In order to enhance the heat transfer rate and reduce the How friction factor, some experiments have been carried out wherein coils with various lengths were partially inserted into the tube. As a result, it is noted that there is an optimum arrangement of the wire coil in the tube for the enhancement of heat transfer and the reduction of friction factor.

View full abstract

Heat transfer enhancement of heaters in Stirling engines plays an important role in improvement of the thermal efficiency of the engines. Effective use of wall radiation in heaters working at high temperatures enhances the heat transfer between combustion gas and heater tubes. In this paper, we report the results of numerical calculation of the improvement in heater performance by the heat transfer enhancement, as well as also the resultant improvement in engine efficiency. Heat transfer calculation is coupled with the engine analysis incorporating detailed key phenomena. As a result, the engine efficiency is shown to be improved more than 10 % by the effective use of radiation in the heater, and the effects of air excess ratio, temperature efficiency of the preheater, maximum tube temperature etc. are also discussed.

View full abstract

The purpose of this study is to examine the relation between the surface temperature change and the change of inner surface of the container which results from mercury corrosion. Even if the mercury heat pipe is manufactured very carefully, when one is heated for the first time, the evaporator surface temperature becomes very high, but 30 hours later, it will decrease to normal level. In order to examine the cause, 2 mercury heat pipes were used. One of them was heated for 10 hours and the other for 52 hours under vapor temperature of 400°C. As a result of these tests, it is concluded that the time required for improving the wettability of the inner surface is about 30 hours.

View full abstract

The characteristics of a heat exchanger for the melting of piled snow were investigated theoretically and experimentally. By assuming steady state distributions in temperature and water content in a snow body, a quasi-state model of the melting of piled snow was presented and was expanded to the performance of heat exchanger operating with snow melting. The predicted results were in agreement with the experimental results obtained using compacted snow on a rubber heat exchanger. In case of the melting of subcooled snow, the melting efficiency becomes higher with a higher inlet brine temperature because a shorter melting time leads to a lower heat loss to surroundings.

View full abstract

In this paper we describe an alternative snow-melting system for the Joetsu Shinkansen Line which runs through a heavy snowfall district. This system consists of a cogeneration system, heat pumps and thermal storage tanks. In order to verify its performance and to establish a control procedure, a dynamic simulation of the system based on the actual equipment data, was carried out under actual load conditions. As a result, the thermal storage tanks were found to play the important role of reducing the start/stop frequency of the heat generation equipment for unpredictable intermit-tent snow load. It contributes to reduce the total energy consumption. This alternative system concludes about 30% less primary fuel than the conventional plant.

View full abstract

In pure rotational CARS (Coherent Anti-Stokes Raman Scattering) through multiple four-color interaction, the rotational energies are excited with two photons of different frequencies from a broadband dye laser. We apply this technique to measurement of temperature in nitrogen supersonic free jets. This method can measure low temperature easier and more accurately than Q-branch vibrational CARS, because rotational lines of pure rotational CARS are separated enough so that they are not affected by pressure. The number densities just behind the nozzle exit are higher than the least number density (5.6×10²³ m^-3) required for temperature determination at room temperature with our apparatus. Therefore, the measured temperatures are in good agreement with the theoretical values under the assumption of isentropic expansion at that region.

View full abstract

Extinction characteristics of stretched flames established in a stagnation-point flow between premixed mixtures and inert gases have been experimentally investigated. The extinction limit was mapped as a function of fuel concentration and mixture velocity, and the flame location and the maximum flame temperature at the extinction limit have been measured. Methane and propane are used as fuels, and nitrogen, argon, carbon dioxide, and helium are used as inert gases. Results show that the flammable range becomes narrower than that of the adiabatic twin flame, especially for rich-propane flames. The flames are extinguished close to the stagnation surface for the lean methane/air mixture and rich-propane/air mixture, but the flames are extinguished at some distance from the stagnation surface for the rich methane/air mixture and lean propane/air mixture. Such flame behavior at the extinction limit is similar to that of the adiabatic twin flame. The flame temperature at the extinction limit is almost the same for the flames which are extinguished close to the stagnation surface, but different for the flames which are extinguished at a finite distance from the stagnation surface. The flame which is extinguished at a finite distance from the stagnation surface is insensitive to the nature of the stagnation surface.

View full abstract

The effect of water on gasification reactivity is studied experimentally for the case of gasification of coal-water mixture (CWM). In this study, CWM, dry coal and dry coal with water vapor, the total amount of which is as same as that contained in the CWM are used. In all experiments, the chemical species produced, the residual volatile matter (VM), fixed carbon (FC), C, H and N molar numbers in both VM and FC, the distribution of particle diameter, the specific surface area and the structure of coal surface are analyzed along the axis of furnace. As a result, the gasification reactivity of CWM is the highest among the three fuels, since water vapor directly reacts with the dense chemical species around a particle. This phenomenon plays a part for the increase of specific surface area and the improvement of gasification efficiency.

View full abstract