Usually, Trichloroethane has been used for the de-oiling and cleaning of machine parts. But its production and import have been prohibited since 1995 because of its possibility to destroy the ozone layer. Generally for biological and environmental safety, the de-oiling should be done with the physical method instead of the chemical method using detergent or solvent. As one of the physical method, a cleaning by a flashing water flow through a packed bed of machine parts has been proposed. The hot water was injected and flashed into steam and water in the packed bed kept at the low pressure less than an atmospheric pressure. In the present study, a prototype flashing water cleaner was designed and its de-oiling ability was experimentally investigated with 900 pieces of M6 bolts. The de-oiling ability was strongly affected with the newly-defined non-dimensional parameter consisting of the injection duration, the average kinetic pressure of flashing two-phase flow defined at vacant container flow area and the viscosity of oil. The de-oiling rate more than 95% was successfully achieved at the parameter larger than 3×10⁵.

View full abstract

This paper shows the results of n-octane two droplets combustion in DC electric field following our previous study. Fiber-supported two droplets are burned under microgravity in air at atmosphere pressure and room temperature. Droplets are arrayed in the directions of electric field with 10 mm distance between droplets and in this condition both flames are individual and have very little interaction between them without an electric field. The equivalent applied elctric field is around 0 to 120 kV/m. The results show that flames are principally deformed to cathode with electric field, however, luminous flame at anode-side end is observed for anode-side flame while blue flame is observed at anode-side end of cathode-side flame. In addition, the increase in burning rate constants for both flames are observed and the increase in burning rate constant for cathode-side flame is larger than that for anode-side flame. The increase in burning rate constant is consistent to information from flame intensities. For cathode-side flame, an analogy to droplet combustion in force convection is discussed in terms of flame deformation and burning rate constants and a relation between them shows good agreement to that under forced convection quantitatively. The increase for anode-side flame is also mentioned from flame deformations.

View full abstract

In the present study, we focused on the length of the tip cavitation as a cavitation parameter for experimental investigations of the thermodynamic effect on a cavitating inducer. Comparison of the length of tip cavity in liquid nitrogen (76 K and 80 K) with that in cold water (296 K) allowed us to estimate the strength of the thermodynamic effect. In addition, unstable cavitation phenomena occur when the cavity length develops over the throat. Sub-synchronous rotating cavitation appears both in liquid nitrogen and in cold water. In contrast, cavitation surge appears only in cold water at a lower cavitation number. From these experimental results, thermodynamic effects on cavitation performances and cavitation instabilities were discussed.

View full abstract

A method of identification of swirling motion, with swirl function defined in this paper, and its application are presented. Flow characteristic can be classified with eigenvalue of velocity gradient tensor, and complex eigenvalue indicates that flow is swirling motion. Here the imaginary part of the complex eigenvalue is defined as swirl function, and the local maximum point of swirling function is assumed to be the axis of swirling motion. The swirl function can be considered as a physical property which corresponds to angular velocity of swirling, which is invariant in transformation of coordinate. This method enables to identify the swirling motion hidden in complicated flow, which is difficult to identify with velocity field or streamline, and also estimate the intensity of swirling.

View full abstract

Three-dimensional large-eddy simulation (LES) is applied to a particle-laden swirling jet, and the effect of swirl on particle diffusion is investigated. The trajectories of all particles are individually pursued with a Lagrangian method. The particles with different diameters are uniformly injected into the non-swirling and swirling flows with different Swirling numbers. The result shows that the trajectories of the particles with different diameters are quite different. The smaller particles have the peak of the particle number densities at central axis, and diffuse across swirling jet. On the other hand, the larger particles at the exit of the nozzle migrate outward by the centrifugal force, but the particles gradually migrate inward with the axial distance increases. The inward movement of the particle in the downstream region is dominated by turbulent motions, which transport the particles inward near central axis. Furthermore, above behavior is shifted upstream with the particle diameter decrease because the smaller particles are easily transferred by smaller inward fluid motions existing in the wide range of the swirling jet.

View full abstract

In preceding report, we proposed Relaxation equations for the smoothed particle hydrodynamics (SPH) method suited to suppress the numerical disorder around the boundary of 2 phase flow as gas and liquid. The effect of the equations was discussed for uniformly distributed particles having same smoothing length. But, when the smoothing length of particles is not constant, it causes numerical disorder again. Additionally, there was a difficulty that the momentum conservation is not satisfied in the Relaxation equation. This paper discusses the problem of numerical disorder caused by the difference of the smoothed length between particle, and we propose a new calculation method suited to suppress the numerical disorder. Additionally, we discuss about the treatment of smoothing length, which was not clearly discussed previously, and we propose a revised Relaxation equation.

View full abstract

Two-dimensional direct numerical simulations have been performed in order to investigate the relationship between surface wave structures and internal flow of water falling films by using the MARS (Multi-interfaces Advection and Reconstruction Solver) method. The numerical results indicate that the velocity profile of falling film is mostly parabolic and similar to the Nusselt theory. However the magnitude of the velocity profile is rather different from that of the Nusselt theory. In this paper, the velocity profiles in a falling film are categorized into four regions : recovery region, development region, break-down region and capillary wave region. Moreover, small vortices can be found in the capillary wave region. Additionally velocity profiles are consistent with a velocity profile obtained from boundary layer theory in a development region. Here, surface velocity is defined as a main flow velocity of a boundary layer, and film thickness is defined as a boundary layer thickness. So, it is considered that the development of velocity in a falling film is similar to that of boundary layer.

View full abstract

The present study restructured a surface capturing method using the optimized WENO scheme for numerical simulations of free surface flows in order to decrease influences of numerical diffusion upon interfaces. Main modifications are as follows. The weight function in the optimized WENO scheme was modified in consideration of interface gradient. A compression parameter in the artificial compression method was modified to avoid over-compression in smooth regions and reduction of numerical accuracy. The method correcting the numerical flux on cell faces was restructured in order to suppress occurrence of non-physical volume fractions. The present surface capturing method was applied to two convective problems. Numerical diffusions are very low around discontinuous regions, and the present method can capture interfaces more sharply than the previous one proposed by the authors. The present method has good conservation property at the time of interface transport as well as the previous surface tracking methods and surface capturing methods. There-fore, the calculation accuracy was greatly improved.

View full abstract

Large eddy simulation (LES) using a mixed-time-scale (MTS) SGS model is applied to the intake flows in simplified internal combustion engine geometry. A modified colocated grid system is employed so as to perform calculations stably with a central difference scheme for convection terms. The results are compared with corresponding experimental data and the computational results obtained by using the low-Reynolds-number linear k-εand cubic nonlinear k-ε-A₂ turbulence models. The LES results show the best agreement with experimental data in three computational cases, not only in the mean velocity profiles but also in the profiles of turbulent energy. These results suggest that LES using the MTS SGS model is an effective method for accurately predicting the performance of a combustion engine involving turbulent diffusion of spray and combustion flame. In addition, the results are compared with those obtained by a quasi-direct simulation employing the QUICK scheme for the convection term and no turbulence model. It is clarified that the numerical viscosity from the QUICK scheme has a great influence on the computational results and decreases the prediction accuracy.

View full abstract

A polygon motor consists of several mirror faces and it is one of the core components for a laser beam printer and a copier. Under high-speed rotation, aeroacoustic noise with a peak frequency of rotational speed times the number of mirror faces and its harmonics is prominently generated from a polygon motor covered with a casing. This paper describes a method to predict such noise that is generated from the flow in between the rotating mirror and stationary casing, propagates through the casing and is radiated into the ambient air. The unsteady flow is firstly computed by large-eddy simulation. The computed pressure fluctuations on the inner wall of the casing are then fed to structural analysis based on a dynamic explicit finite element method that computes the propagation of the elastic waves in the casing. Resulting external surface velocities are finally used for the computation of the acoustical field. A good correlation between the computed and measured results has been observed both in terms of velocity spectra on the casing and spatial distribution of sound pressure level. The proposed method thus seems a promising engineering tool for predicting and identifying the generation mechanism of such noise that is generated from unsteady flow, propagates through solid walls and is radiated to ambient air, in general.

View full abstract

Flow-induced in-line oscillation of two tandem rectangular cylinders has been experimentally studied by free-oscillation tests in a wind tunnel. Only one cylinder of two cylinders was elastically supported easily to move in the in-line direction for reduced mass-damping parameter Cn≈1.5 ; the other was fixed to the tunnel sidewalls. The gap between two tandem cylinders was changed from s=0.3 to 3. We measured the response amplitudes of the oscillatory cylinder in the in-line direction and the vortex-shedding frequency in the wake. The flow around the tandem cylinders was visualized by the smoke-wire method. As a result of the upstream cylinder, an excitation appeared in a wide region of reduced velocity, Vr=2.5 to 5 at gap ratios of s=0.3 to 2. This excitation was mainly induced by symmetric vortex shedding, the other excitation of Vr=4.5 to 6.5 at wide gap ratios of s=0.75 to 1.5, was induced by alternate Karman vortex shedding, accompanied with unstable limit-cycle. The downstream cylinder had an excitation region in a wide range of Vr=3 to 5.5 for the narrow gap distance of s=0.3 to 0.75, which also seems to be induced by alternate Karman vortex shedding. Furthermore, the downstream cylinder oscillated as a buffeting phenomenon influenced by wake-fluctuation of the upstream cylinder, when gap distance was greater than s=1.

View full abstract

This paper describes the fluid force acting on a square cylinder of finite length in the locking-in region. The experiment was carried out in an N. P. L. blow-down type wind-tunnel with a working section of 500 mm×500 mm×2000 mm at Reynolds number of 1.25×10⁴. The cylinder was forced to oscillate sinusoidally in the lift direction. The power spectrum of the fluctuating velocity in the wake behind a square cylinder was measured to show the locking-in region in the present experiment. The time-mean pressure distribution and fluctuation pressure distribution on the square cylinder were measured for the displacement in the vibration. Consequently it was found that the mean drag and fluctuating lift increase and become maximum in the locking-in region, while the base pressure in the rear surface of the cylinder becomes low and attains minimum.

View full abstract

This paper describes the flow past an inclined flat plate of finite width placed on a ground plane. The experiment was carried out in the 800 mm diameter Eiffel type wind tunnel having the working section of 1.5 m length at Reynolds number of 5.1×10⁴ mainly. The surface pressure distributions on the flat plate and on the ground plane around the flat plate were measured, and the drag coefficient and lift coefficient were determined from them. The power spectrum of the streamwise component of fluctuating velocity was measured to confirm the presence of arch type vortices shedding behind the flat plate. The vortices generated in the flow field around a flat plate were also observed by flow-visualization. Consequently, it was found that the arch type vortices are shed in the range of inclining angle of 60°≤β≤90°for W/H=2 and 4, while the drag coefficient attains maximum near β=90°and the lift coefficient decreases for W/H=2, 4 and 7, as β increases from 30°to 135°.

View full abstract

It is known that the low-frequency fluctuation about 1/20 of vortex shedding frequency exists in the irregularity of the vortex shedding from a two-dimensional circular cylinder. Local vortex shedding frequency in spanwise direction is considered to differ from one point to another due to different local Strouhal number caused by flow three-dimensionality. Therefore, nonlinear interaction between vortex shedding cells of different frequencies is considered to be the cause of the low-frequency fluctuation. In this study, two distinctive different vortex shedding frequencies have been forced to occur at a local spanwise position with use of a stepped circular cylinder, and modified particular phase difference TSC (Trans-Spectral Coherence) analysis was conducted to investigate the presence of quadratic (or second order) nonlinear interaction in the wake. It is expected that the quadratic nonlinear interaction is generated as the result of complicated interaction between spanwise vortex cells of different frequencies.

View full abstract

The pressure-sensitive paint is very useful and easy way to obtain the pressure distribution on surfaces. Usually the pressure distribution by the paint shows good agreement with that obtained by other methods, like a pressure tap. Recently, the paint has been adopted to the high Knudsen number regime for pressure measurement, and showed possibility of pressure measurement for wide range of flows. The mechanism of the pressure-sensitive paint is mainly discussed within the luminescence intensity and the oxygen quenching of luminescent molecules. Still, it is unclear that the luminescence intensity is really proportional to the pressure, or another related value, especially in the high Knudsen number regime. Here, in this study, we tried to relate the flux and the pressure of oxygen to the luminescence intensity of the pressure-sensitive paint, and clear out the limit of application from the gas dynamical point of view.

View full abstract

A novel definition of liquid-vapor interface at the molecular level is proposed, which can capture the local and instantaneous fluctuation of the interface. This definition is not based on the time- and space-averaged density profile, but on the instantaneous particle density distribution of molecules. Previously, the Gibbs dividing surface, which is the mathematical difinition decided from static density profile, is used to analyze the structure or interfacial physics. However, in this physical picture of the interface, the fluctuation itself and structure embedded in the fluctuation cannot be extracted. In this paper, we discuss the relation between the Gibbsian interface defined thermodynamically and the statistical-averaged interface of the local and instantaneous interface presented here. Furthermore, the interfacial fluctuation is evaluated by using the local and instantaneous interface, and an asymptotic character of the fluctuation is compared with the fluctuation predicted by the macroscopic capillary wave theory.

View full abstract

The present aim is to clarify the combination effect of curvature and streamwise thickness of a ring with rectangular cross sections, which is immersed perpendicularly to uniform flow. In a wind tunnel, the authors carry out measurements using a pressure manometer and hot-wire anemometers.and flow visualisations by a smoke-wire method, for Re=5.0×10³-1.5×10⁴, d/w=3.0-16.0 and t/w=0.28-1.80, where Re, d, w and t are the Reynolds number, mean diameter, cross-section width and cross-section thickness of the ring, respectively. Consequently, the Strouhal number St, which is a reduced frequency of regular vortex shedding from the ring, is independent of d/w, t/w and Re. On the other hand, base suction depends not on Re, but on d/w and t/w. Considering d/w and t/w as governing parameters, and regarding to base suction, flow visualisation, velocity fluctuation, mean-velocity and turbulence-intensity profiles and four-points velocity correlation, we can classify flow into four flow modes; viz., an over-critical axisymmetrical mode, a sub-citical axisymmetrical mode, a sub-critical non-axisymmetrical mode and a sub-critical less-correlation mode.

View full abstract

This research deals with the switching mechanism of a flip-flop jet nozzle with a connecting tube, which is based on the measurements of pressure in two chambers, velocity in the connecting tube and velocity distribution between two inside walls of the nozzle, i. e., reattachment walls. The authors particularly focus on the details of switching flow field inside the nozzle, using a Ultrasound-Velocity-Profile monitor (UVP monitor). As a result, two re-circulating flows, viz., two vortical structures, are shown on both side walls inside the nozzle. By means of the simultaneous observation of chamber pressures and connecting-tube velocity with UVP results, we show a coherent scenario of this jet-oscillation phenomenon.

View full abstract

This study investigated on the characteristics of flow-induced vibrations of two circular cylinders in staggered arrangement. The response characteristics on elastically supported two circular cylinders were examined by a free vibration test. Measurements were performed for stagger angle α=5°, 10°, 15°, 25°, 45°and 60°, and the spacing between two cylinders L/D=0.1-3.2, where α is the angle between the free-stream flow and the line connecting the centers of two cylinders, L is the gap width between two cylinders, and D is the diameter of two cylinders. The new findings for characteristics of flow-induced vibrations of two cylinders in staggered arrangement are as follows. (i) There were seven patterns of flow-induced vibration with changing stagger angle a and spacing ratio L/D, (ii) the generation mechanism of flow-induced vibration was clarified by doing flow visualization test and generation regions of flow-induced vibration at each stagger angle were arranged, (iii) when the cylinder was perturbed at stagger angle 10 and 15 degrees, flow-induced vibration genetated two cylinders was a divergent vibration, (iv) there were two patterns of flow-induced vibration at critical spacing L/D in which exist two flow patterns.

View full abstract

The reduction of drag of a circular disk with the axial-length/diameter ratio c/d<0.5 was investigated. The drag coefficient is 1.17 to 1.0 in the range 0< c/d <0.5 regardless of Reynolds number. To control the flow field around the circular disk, the front edge was cut to a stepwise shape with smaller-diameter d₁. The step height and step length, a= (d-d₁) /2 and b, are a/d =0.1 and b/d=0.2, respectively. Experiments were performed c/d=0.30, 0.40 and Re= (0.8-8) ×10⁴. The shear layer separated from the corner of the small diameter-section reattaches on the front corner of the large-diameter section. Then, an annular vortex is formed at the step region and the pressure coefficient on the step region becomes negative. This causes thrust on the step region of the disk. In addition, the drag force decreases due to the smaller area of the front surface. As the results, the wake width decreases and the base pressure coefficient rises. The drag coefficient of the circular disk with a step decreases by about one-forth of a standard circular disk.

View full abstract

Laminar-Turbulent transition of a boundary layer induced by a single roughness element in the inlet region of a circular pipe was experimentally investigated. Mean and fluctuating velocity components in the axial direction were measured by a hot-wire anemometer. The Reynolds number based on the pipe diameter and the bulk velocity was 20 000. A stationary turbulent region was formed downstream of the roughness element. The contour lines of the turbulent region were wedge-shaped once they unfolded onto a flat plane. The virtual origin of the turbulent region equaled the roughness position. The contour lines of intermittency, which were initially staight, then curved to the outside. This condition is similar to that of the turbulence wedge on a flat plate. The fluctuating velocity was as its maximum not at the center of the region but at an interface region between the turbulent region and surrounding laminar region. The fluctuation within the interface region was large.

View full abstract

Several experiments have reported that Departure from Nucleate Boiling (DNB) in a Pressurized Water Reactor (PWR) fuel assembly often takes place in upstream part of a grid spacer, and it was supposed that bubble accumulation induced by flow around an obstacle is one of reasons of the DNB inception. Hence, experiments were conducted to investigate the motion of small bubbles in high-speed flows in a vertical duct containing an obstacle. The bubble trajectories and averaged bubble distributions were measured by an image processing method. The experimental results indicated that small bubbles migrate toward the opposite wall of the obstacle near the upstream edge of the obstacle. The lateral translation of bubbles was larger than that of the liquid phase, which could be induced by the difference in inertia between bubbles and the liquid phase.

View full abstract

For the famous formulas of pipes and ducts, for example, Weisbach's formulas of the pressure loss coefficients in entrance flush with wall at arbitrary angle, abrupt contraction, miter bend, and so on, it is still not clear that some of these formulas could be derived from a simple flow model or not. In this paper, an analysis was conducted by using linear momentum theorem for these equations from the macroscopic point of view. It was found that these formulas could be derived as modified ones of the equations obtained by using the momentum theorem. In the first report, the pressure loss coefficients of gradual and abrupt contraction, entrance flush with wall at arbitrary angle, and the effect of rounded entrance (r/D) on the loss coefficient were examined.

View full abstract

For the famous formulas of pipes and ducts, for example, Weisbach's formulas of the pressure loss coefficients in entrance flush with wall at arbitrary angle, abrupt contraction, miter bend, and so on, it is still not clear whether some of these formulas could be derived from a simple flow model or not. In this paper, analysis was conducted by using linear momentum theorem for these equations from to the macroscopic point of view. It was found that these formulas could be derived as modified equations of the equations obtained by using the momentum theorem. In the second report, the pressure loss coefficient of miter bends (elbows) and the effect of inner rounded corner (r/D) were examined.

View full abstract

We have investigated the particle orientational distribution of a non-dilute colloidal dispersion, which is composed of ferromagnetic rodlike particles with a magnetic moment normal to the particle axis, subject to a simple shear flow. The mean field approximation is applied to take into account the magnetic interaction between rodlike particles. In the present analysis, these spheroidal particles are assumed to suffer from the rotational Brownian motion in a simple shear flow and an external magnetic field. The basic equation of the orientational distribution function has been derived from the balance of the torques and solved numerically. The results obtained here are summarized as follows. For a very strong magnetic field, the magnetic moment of the rodlike particle is significantly restricted in the field direction, so that the particle approximately points to a direction normal to the flow direction (and also to the magnetic field direction). For the case of very strong magnetic interactions between particles, the particle can rotate around the axis of the clusters of the particles, although the magnetic moment nearly points to the field direction. Due to this characteristic motion of the particle, the viscosity for the case of strong particle-particle interactions indicates smaller values than those for the case of weak interactions.

View full abstract

We propose a new type of propulsive equipment for a ship in this paper. It is directly driven by the high-pressure wave generated by air-propane mixture combustion. The propulsive equipment is a simple structure consisting of a combustion tube and a semi-open-type nozzle. In order to clarify the mechanism of thrust generation, we measured the pressure fluctuation on the nozzle wall and the thrust for various angles of ejecting the pressure wave, and also we observed the flow in the two-dimensional nozzle by using a high speed motion camera. Water flow in the nozzle is accelerated momentarily by the pressure wave, and the bubble is exhausted to downstream while their expansion. It was clarified that the thrust strongly depends on the nozzle-wall angle, and the thrust for the nozzle-wall angle 45 degrees is higher than others in this experiment. The thrust of the nozzle was 42 N when the 400 kPa pressure wave is driven with added oxygen.

View full abstract

This paper describes the generation of cavitation by pulsed water jetting and its capability for utilization. A submerged cavitating water jet (SCAWJ) was realized by a system that we developed. In this study, we visualized the cavitating water-jet in water pool by laser-sheet imaging technique. And to evaluate the capability for surface treatment of metal by impinging SCAWJ, the impulse force of SCAWJ and the eroded depth and area of metal by SCAWJ were measured. From the experimental result, it was found that the present pulsed SCAWJ system have a capability for shotless-peening of metal surface at longer stand-off distance by means of strong impulse force derived from collapse of cavity generated by cavitating jet.

View full abstract

In centrifugal pumps for artificial hearts, a magnetic drive and lightly loaded journal bearing system are often used to avoid leakage and to minimize the damage on blood cells. In such system, the rigidity of the bearing is small and the impeller usually rotates over the critical speed. For such cases, the rotordynamic fluid forces play an important role for the stability of operation. In the present study, the characteristics of the rotordynamic fluid forces on the impeller were examined. The destabilizing fluid force which encourages the whirling motion of the impeller occurred in the range of 0<ω/Ω?0.75, where ω and Ω are the angular velocities of the whirling motion and the rotation of the impeller, respectively. The steady fluid forces in the case without the whirling motion were also investigated. The magnitude of the steady fluid force was consistent with the unsteady fluid force in ω/Ω≅0. The forces estimated from the unsteady pressure distribution on the casing were in agreement with the force measured directly by 4-axis sensor. The effects of leakage flow were also examined. It was found that the leakage flow has a stabilizing effect on the whirling motion of the impeller.

View full abstract

In the present paper, a new method for the suppression of unsteady cavitation in a turbopump inducer is proposed. An accumulator with a large volume is placed at the upstream of the inducer to equalize the pressure around the periphery. Through the experiments with the accumulator, it was found that the occurrence region of rotating cavitation at design flow rate decreases and higher order cavitation instabilities are suppressed. However, at low flow rate, higher order cavitation instabilities were found when the clearance opened to the accumulator was too large.

View full abstract

In the range of very low specific speed n_s, improvement of efficiency is the most important subject in order to apply turbo machine to such a low specific speed range. The purpose of the present study is focused on the performance improvement of pump in the range of n_s?80 [m, m³/min, min^-1], and CFD simulation is performed to determine the reason why efficiency is very low. The result shows that a rapid increase of leakage rate and disk friction rate with a decrease of n_s is the main reason of low efficiency. Also, the result shows that if the leakage loss is reduced by use of Labyrinth seal at n_s=25, the disk friction rapidly increases in spite of the leakage reduction and pump overall efficiency becomes only 5% higher.

View full abstract

A cross-flow wind turbine was tested for the performance under the various conditions of number of blades and blade setting angle, respectively to investigate the influence of these parameters on the cross-flow wind turbine performance. The cases of number of blades were 6, 8, 10, 12 and 16 and about blade setting angle, we tried 30, 45, 60 and 90 degrees. The results showed that there were roughly three types of the performance curves, which were single smooth parabolic line, single peaky curve and curve with two peaks. The rotor with 12 blades showed the highest porformance and the optimum blade setting angle was 45 degrees.

View full abstract

One of the well known cures for the annuloaortic ectasia is a vascular grafting surgery using a composite graft as an artificial ascending aorta and an artificial valve. Although, almost all conventional grafts do not have Valsalva sinus at the aortic root, the influence of omitting the Valsalva sinus have not yet been clarified. In order to understand the effect of the Valsalva sinuses at aortic root on the valve characteristics and flow field around the valve and furthermore the coronary blood flow, the authors have fabricated some realistic models of aortic valve using three-dimensional modeling machine, and have investigated the effect of the sinus existence by particle image velocimetry (PIV). As the result, the fluctuation of flow rate at coronary artery in the case of the straight wall model was larger than the Valsalva sinus model. The valve leaflets were pulled to the left ventricle, which could cause a diastolic regurgitant flow from the aorta into the left ventricle. The leaflet suction in the Valsalva sinus model was weaker than the straight wall model. Flow visualization results showed that vortex flow occurred in the Valsalva sinus due to jet flow through the valve leaflets during systole. At early diastole, flow into coronary artery around the Valsalva sinus was smoothly than the straight wall model. Consequently it was concluded from a standpoint of flow and valve characteristics that the Valsalva sinus had an important role for the aortic valve and the inlet of the coronary circulation.

View full abstract

Experimental investigations have been carried out for combined convective flows of air induced around uniformly heated, horizontal cylinders. Three cases of aiding, opposing and cross flows were treated. The experiments covered the ranges of the Reynolds and modified Reyleigh numbers as ; Re_d=50-900 and Ra^*_d=5×10⁴-3×10⁶. The flow fields around the cylinders were visualized with smoke. The results showed that separation points of forced flow gradually shift from the sides to the upper edge of the cylinder with increasing wall heat fluxes. The local heat transfer coefficients of the cylinders were also measured. Although the local coefficients show complex variations with the forced flow velocities and the wall heat fluxes, the overall coefficients become higher than those estimated from the pure forced and natural convections throughout the cases of aiding, opposing and cross flows. Moreover, it was found that the overall Nusselt numbers and the separation points are well predicted by the non-dimensional parameter (Gr^*_d/NU_dRe²_d).

View full abstract

An enhancement technique is developed for natural convection heat transfer from a vertical heated plate with inclined fins, which are attached on the vertical heated plate to isolate a hot air flow from a cold air flow. Experiments are performed in air for inclination angles of the inclined fins in the range of 30 to 90 degrees measured from a horizontal plane, for height of 25 to 50 mm, and for fin pitch of 20 to 60 mm. The convective heat transfer rate for the vertical heated plate with inclined fins at the inclination angle of 60 degree is found to be 19% higher than that for the vertical heated plate with vertical fins. A dimensionless equation on the natural convection heat transfer of the vertical heated plate with inclined fins is presented.

View full abstract

The most part of energy losses in power system such as fuel cells is due to the heat released by the exhaust gas to atmosphere. The exhaust gas consists of non-condensable gas and steam with sensible and latent heat. As a lot of latent heat is included in the exhaust gas, its recovery is very important to improve the power system efficiency. Based on the previous basic studies, a thermal hydraulic prediction method for latent heat recovery exchangers was proposed. For the condensation of steam on heat transfer tubes, the modified Sherwood number taking account of the mass absorption effect on the wall was used. Two kinds of compact heat exchanger with staggered banks of bare tubes of 10.5 or 4 mm in outer diameter was designed with the prediction method. The more compactness was obtained with the smaller tubes at a designed heat recovery. The thermal hydraulic behavior in the compact heat exchangers was experimentally studied with air-steam mixture gas. In the parametric experiments varying the steam mass concentration, the temperature distributions of cooling water and mixture gas were measured. The experimental results agreed well with the prediction proposed in this study and the more compactness with the smaller tubes was proved.

View full abstract

The present study performs numerical simulations of turbulent flow transition of separated flow and heat transfer around a two-dimensional rib mounted in a laminar boundary layer. The separated shear layer becomes unstable by Kelvin-Helmholtz instability and generates two-dimensional vortex. This vortex becomes three-dimensional and collapses in the latter half of the separation bubble. As a result, transition from laminar to turbulent flow occurs in the separated shear layer. Streamwise vortices exist downstream of the reattachment flow region. Low flapping motion and transition of the separated shear layer are influenced by three-dimensional dynamics upstream of the separation bubble. Large-scale vortices around the reattachment flow region have great effects upon heat transfer. Downstream of the reattachment point, the surface friction coefficient and Nusselt number are different from profiles in the laminar boundary layer, and approach the distributions of the turbulent boundary layer.

View full abstract

This paper presents three-dimensional simulation of separated and reattached flow and heat transfer around a surface mounted cube in a laminar boundary layer. The shear layer separated at the leading edge of the cube becomes unstable, and a hairpin vortex is generated periodically. The hairpin vortex moves downstream with time and grows to the streamwise vortices stretched to the streamwise direction. The head and legs of the hairpin vortex and horseshoe vortices have great effects upon the heat transfer near the wall. The shed hairpin vortex forms alternating positive and negative fluctuations near the wall, and decreases gradually toward the downstream extending in the spanwise direction. The head of the hairpin vortex locally generates a high turbulent area at the center of the spanwise direction. Another high turbulent area across the center of the spanwise direction is generated by interference of horseshoe vortices.

View full abstract

Experimental and numerical studies were performed to clarify the axial mass and heat transport mechanism of reciprocating flow inside ribbed circular tubes. Water was used for the working fluid. The test section consisted of a circular tube of 16 mm in inside diameter and 800 mm in length. Ribs were installed on the tube wall with a given interval in axial direction and one end of the tube was connected to a heat source and the other end to a heat sink. A crank-piston mechanism was utilized to create reciprocating flow with no net through flow. Numerical simulations were also carried out under the same conditions as the experiment. Since it was confirmed that the results of the simulation agreed very well with those of experiment, the simulation was used to explore the details of the flow and heat transport mechanisms in the present study. The present study disclosed that the periodic formation and extinction of flow separation behind the ribs constituted the axial mass and heat transport through the “trap and release” mechanism in the reciprocating cycles. It was also shown that heat transport performance of the ribbed tube was far superior to that of plain tubes or dream pipes.

View full abstract

The authors have conducted measurements of liquid-vapor behavior in the vicinity of a heating surface for saturated and subcooled pool boiling on an upward-facing copper surface by using a conductance probe method. The previous paper (Transactions of JSME, Series B, pp. 2951-2958, vol. 70, no. 699 (2004)) reported that thicknesses of a liquid rich layer (a so-called macrolayer) forming in subcooled boiling are comparable to or thicker than those formed near the CHF in suturated boiling. This paper examines the dryout behavior of the heating surface by utilizing the feature that a thin conductance probe very close to the heating surface can detect the formation and dryout of the macrolayer. It was found that the dryout of the macrolayer formed beneath a vapor mass occurs in the latter half of hovering period of the vapor mass. Two-dimensional measurements conducted at 121 grid points in a 1 mm×1 mm area at the center of the heating surface showed that the dryout commences at specific areas and spreads over the heating surface as the heat flux approaches the CHF. Furthermore, transient measurements of wall void fractions from nucleate boiling to transition boiling were conducted under the transient heating mode, showing that the wall void fraction takes small values (< 10%) in nucleate boiling region, and then steeply increases in transition boiling region. All these findings strongly suggest that the macrolayer dryout model is the most appropriate model of the CHF for saturated and subcooled pool boiling of water on the upward copper surface.

View full abstract

By installing HFC 134 a on the air conditioning machine designed for HCFC 22, it was proved that a more efficient heat pump system could be operated without changing mineral lubricant oil. The circulation mechanism of mineral oil with HFC 134 a was investigated in comparison with that of HCFC 22. It was shown by the flow visualization that the mineral oil was not solved in condenced HFC 134 a, but droplets of mineral oil were transported with condensed HFC 134 a at the outlet of condenser. The circulation of mineral oil with HFC 134 a was also confirmed by extracting the working fluid from the tubing of the air conditioning machine under operation.

View full abstract

Application of advanced steam condition for the old oil fired units is very useful to improve the plant thermal efficiency, as a result, it reduces CO₂ emission and achieves the economical generating cost. Development of the heat resistant ferritic rotor combined with rotor surface cooling system is very important to realize the steam turbine with the advanced steam condition. The full scale ferritic turbine rotor has been manufactured to verify the producibility and the quality of materials. This paper describes how to develop heat resistant ferritic turbine rotors for advanced steam condition for future.

View full abstract

PEFC was simulated with an equivalent circuit having four resistances and three capacitances whose values were estimated by AC, impedance method, and each overvoltage (OV) of which total OV consists was then calculated by integral calculus from current density i for each resistance R. This method enables us to investigate influence of characteristics of gas diffusion layers (GDL) upon each OV that varies with relative humidity of supplied gas. As a result, cathode activation OV was found to decrease using GDL with micro porous layer (MPL) at any humidity. In addition, using different types of GDL, cathode concentration OV was found to decrease due to efficient gas diffusion in GDL at high humidity, and due to air permeability at low humidity. Moreover, ohmic resistance of PEFC obtained by AC impedance method decreased with increase of i at humidity under saturation. This result suggests that humidity of membrane electrode assembly (MEA) depends on i at humidity under saturation. The above results will offer design of efficient GDL, which improves PEFC performance.

View full abstract

We have experimentally investigated influences of cathode channel heights on vapor permeation flux through gas diffusion layer (GDL) of polymer electrolyte fuel cell (PEFC). Vapor permeation flux through GDL was remarkably influenced by channel velocity. In the analysis of these experiments with a vapor transport model, it was shown that the resistance associated with the vapor-concentration boundary layer was dominant in the low channel velocity less than 0.5m/s, and the resistance for the diffusivity of GDL became dominant as channel velocity increases. Influences of the channel height on the vapor permeation flux were verified by the power generation tests using a single cell. In the cell tests, anode channel height was fixed at 0.5 mm and cathode channel heights were changed at 0.2 mm, 0.5 mm, 1 mm and 3 mm. Under the high humidity condition, the cell performance was improved by using a separator with the shallow channel height. Such a cell performance was well correlated with the vapor permeation flux through GDL. In addition, influences of oxidant utilization and water management layer (WML) on the cell performance were evaluated.

View full abstract

In this paper, the basic studies on the high temperature steam gasification system which is named as HyPR-MEET system are carried out using the Japanese cedar and pure polythylene (PE) particles as feedstocks. First, the characteristic of steam gasification is investigated. With each feedstock, the hydrogen concentration increases as the gasification temperature rises, but the increasing rate of hydrogen concentration decreases over the temperature of 1 073 K for Japanese cedar and 1 173 K for PE. The increase of the steam/carbon molar ratio also increases the hydrogen concentration. Second, the effectiveness of the Ru/Al₂O₃ catalyst for steam reforming of tar components is investigated. With use of this catalyst, tar components can be reformed at the temperature of 973 K, while the temperature of 1 173 K is required without use of any catalysts. Furthermore, light hydrocarbon gases are also reformed with this catalyst, and the hydrogen concentration in the reformed gas can reach the value of 55 vol% for Japanese cedar, and of 68 vol% for PE.

View full abstract

Experimental work was performed to investigate formation of soot precursor and PM in diesel combustion process using a flow reactor and a laminar co-flow diffusion burner. Thermally decomposed aliphatic diesel fuel produced in the flow reactor was supplied directly to the burner as fuel. Measurements were conducted in the flow reactor, in the flame, and the exhaust for light hydrocarbons, low molecular weight aromatics, PAHs, soot precursor, and PM. Results show that many PAHs ranging from two to five-member ring are already formed as well as light hydrocarbon such as C₂H₄, C₂H₂ and CH₄ during pyrolysis in the flow reactor. Exhaust PM is subject to be formed when abundant C₂H₂, Benzene, PAHs are produced during high temperature thermal decomposition of aliphatic diesel fuel, and the structure of PM has long chain-like aggregation. Measured fluorescence spectrometry indicates 464 nm peak. This peak is caused by soot precursor, which is in the SOF extracted from PM.

View full abstract

In the present study, behavior of thermo-chemical fluid in a combustion chamber of engine was numerically analyzed using the finite difference method with Navier-Stokes equations including chemical reactions. As the position of a piston changes with the lapse of time, grid points composing a computational mesh are not always on the piston head. Therefore, values of conserved quantities on the piston head were found by series expansion from near grid points and were utilized in the calculation. Aiming at capturing the feature of premixed compression ignition, a stoichiometric gas mixture of hydrogen and air was used with a detailed reaction mechanism. Numerical method executed in the present case is the semi-implicit MacCormack-TVD scheme with the second order accuracy. It was found from the calculated results that a new computation approach to the premixed compression ignition could obtain valid solutions for us without grid regeneration.

View full abstract

LES (Large-eddy simulation) of swirling cold flows in a pulverized coal combustion furnace is performed and validated by comparing with experiment and RANS (Reynolds-Averaged Navier-Stokes) simulation with the standard k-ε model. Unstructured grids and energy-conserving scheme are employed to obtain robust and accurate solutions at high swirl numbers. The results show that LES is in good agreement with the experiment in terms of the recirculation flow, though RANS overestimates. This is considered due to the capability of prediction of turbulent mixing. The effect of secondary and tertiary swirler angles on the behavior of the recirculation flow are also well predicted by LES. The recirculation flow affects the flame stabilization and combustion efficiency related to the particle residence time in the furnace. It can be said, therefore, that LES is strongly recommended for the accurate prediction of the strong swirling flows in such furnaces.

View full abstract

This paper described an availability of acetone-OH simultaneous PLIF scheme for combustion diagnostics on turbulent premixed flame. Distinctive signal gap is found between acetone (seeded in the premixed gas) and OH (combustion product) with single excitation wavelength and single CCD camera to achieve relatively low cost measurement of flame zone. Flame zone thickness includes much flame information e.g. the location of heat release region, local burning velocity. Considered turbulent condition throughout the present study is relatively weak and categorized in laminar flamelet regime. As the equivalence ratio of premixed mixture (acetone seeded methane-air premixed gas) is varied from 0.64 to 0.90 (burning velocity is increased), measured local flame zone thickness (mode value) is decreased, however 3-D movement of the observed flame front makes the measured flame zone thickness wider than the laminar flame condition. In addition, results show that flame front curvature (convex toward unburned is positive) is positively correlated with flame zone thickness. It is ensured an applicability of the present scheme to survey the relation between flame zone thickness and flame front curvature in turbulent premixed flame.

View full abstract

Steady flamelet method, flamelet/progress-variable method and chemtables used for the methods are evaluated on DNS of two-dimensional diffusion jet flame by comparing with the direct approach. The results show that the steady flamelet methods with the chemtables considering the distribution of scalar dissipation rate in Z space are superior to those with chemtables not considering it. This is not conditioned by the solution, i.e. whether the governing equations are solved in the physical space or Z space. The accuracy is improved by flamelet/progress-variable method. Although product mass fraction has been used as the progress-variable for the flamelet/progress-variable method, it is found that enthalpy is preferable in terms of the accuracy and the application to complicated fields like a spray conbustion.

View full abstract

The combined probability density function (PDF) /moment method which had been developed in the previous report has been evaluated in the frame of a detailed chemical kinetics. The method was implemented in the configuration of a turbulent jet diffusion flame and is evaluated through the comparison with the conventional PDF method with the detailed chemical kinetics, the flamelet model method with the flame structure on the one-step irreversible reaction and laser experiments. The present method overestimates the effect of finite reaction rate to lead to a slight deviation from the experimental values, especially in the near-field. This is caused by the ignition which is required in the PDF methods. Thus, the improvement in terms of the ignition is required. The present result is much better than the result of the conventional PDF method, reflecting the usefulness of the combined PDF/moment method.

View full abstract