Rotation induces centrifugal and Coriolis forces. The former is unimportant in many problems but the latter is the cause of differences between the dynamics of rotating and non-rotating fluids. This article compares the nature of the Ekman layer which plays the key role in rotating fluids and that of the Prandtl boundary layer in non-rotating flow systems. Fluids in rotation are abundant in nature and have important applications in industry. This is an area that needs to be explored. Perspectives of research, both experimental and theoretical, on transport phenomena in rotating systems are presented to encourage researchers to meet the challenge.

抄録全体を表示

Theoretical study of three-dimensional boundary layer transition on a sphere rotating in still fluid has been carried out by a linear stability analysis. It was determined from experimental results that the relative location of spiral vortices which appear in the transition region is fixed against the rotating sphere surface when the rotating effect is large. However as the rotating effect decreases, these vortices begin to move. So far, this phenomenon has not been explained theoretically. In this study, the boundary layer instability and the characteristics of spiral vortices are theoretically investigated by calculating the behavior of the small disturbances and the critical Reynolds number of the boundary layer on a rotating sphere using the linear stability theory.

抄録全体を表示

In this paper, we describe the flow structure of a cavity interacting with a normal shock wave in a supersonic duct. The position of the normal shock wave and the bleed inside the cavity are focused on. Experiments are conducted at Mach number M=1.58. Time-averaged and unsteady pressures are measured and the flows are visualized by the schlieren and the oil flow techniques. The characteristics of the cavity flow with a shock wave interaction, namely, a shock system around the cavity and the separation in the interacting region are recognized. The bleed inside the cavity stabilizes the normal shock wave.

抄録全体を表示

This paper describes a method for visualizing boundary layers in hypersonic flow. The method is applied to the visualization of shock-wave/boundary-layer interaction phenomena in hypersonic flow. In order to develop the visualization method, the radiation of an electric discharge is utilized. First, a visualized boundary layer over a wedge in hypersonic flow is shown. Second, visualized results of the hypersonic shock-wave/boundary-layer interaction phenomena generated by two wedges are demonstrated. Furthermore, the method for visualizing boundary layers is improved, and a flowfield of hypersonic shock-wave/boundary-layer interaction is demonstrated by utilizing the improved method. These experiments are carried out using a hypersonic shock tunnel. The main characteristics of the tunnel are that the Mach number is 10 and the duration is 10 ms.

抄録全体を表示

Two-dimensional Navier-Stokes equations are numerically solved for a shock wave moving in a tube. The internal structure of the shock wave is taken into account as is the viscous layer along the tube wall. Slip conditions are applied along the wall surface. The gas is a monatomic, perfect gas, and the shock Mach numbers are 2.0 and 20.0. The results for the low Mach number case (Ms=2.0) are compared with numerical results using a kinetic model. The N-S solutions, which include slip effects on the wall, show qualitative agreement with the kinetic model solutions for Maxwellian molecules with respect to temperature contours, isopycnics and slip effects along the wall surface.

抄録全体を表示

A numerical analysis has been performed for three dimensional developing turbulent flow in a 90° bent tube with straight inlet and outlet sections by an algebraic Reynolds stress model. To our knowledge, very little has been reported about detailed comparison between calculated results and experimental data containing Reynolds stresses. In calculation, an algebraic Reynolds stress model together with a boundary-fitted coordinate system is applied to a 90° bent tube in order to solve anisotropic turbulent flow precisely. The calculated results display comparatively good agreement with the experimental data of time averaged velocity and secondary vectors. In addition, the present method predicts as a characteristic feature that the intensity of secondary flow near the inner wall is increased immediately downstream from the bend outlet by the pressure gradient. With regard to comparison of Reynolds stresses, the present method is able to reproduce well the distributions of streamwise normal stress and shear stress defined streamwise and radial velocity fluctuation except for the shear stress defined streamwise and circumferential velocity fluctuation. The present calculation has been found to simulate many features of the developing flow in bent tube satisfactorily, but it has a tendency to underpredict the Reynolds stresses.

抄録全体を表示

We report the differences between transitional and steady flows of a free jet. In this calculation, the fifth-order upwind finite difference scheme and Runge-Kutta method are applied to the convective terms and the time developing term, respectively, in three-dimensional Navier-Stokes equations to simulate directly the free jet flow. The calculated flow field ranges from the nozzle exit to the developed turbulent region. The computational cell size is smaller than the Kolmogorov microscale in the developed region. Calculated results are compared with experimental data of flow visualization, velocity fluctuation and FFT analysis, and found to be in fairly good agreement with experimental data. Large-scale vortices which are created in the early transitional flow lead to differences between transitional and steady flows. The direct numerical simulation accurately represents the motions of microscale eddies in a free jet.

抄録全体を表示

Taking into account the fact that the state of turbulence approaches a two-dimensional state close to the wall, a new Reynolds-stress turbulence model is proposed. The rotational properties of the two-dimensional turbulence differ significantly from the three-dimensional one. Under the system rotating about the axis perpendicular to the plane the turbulence correlations between fluctuating velocities remain invariant, while those combined with fluctuating pressure vary. Based on these mathematical properties, a proposal for the pressure-strain correlation is made. Trial computations for turbulent channel flows show that the present Reynolds stress model can reproduce well strong anisotropic near-wall behaviors of Reynolds stresses, which cannot be predicted by previous turbulence models.

抄録全体を表示

The rate of local axial gas transport in oscillatory flow through a model of human central airways was measured to evaluate gas exchange during high-frequency oscillation. A rigid model of human central airways consists of asymmetrical bifurcations up to 3-5 generation, of which geometries were determined by the study of Horsfield et al. (J. Appl. Physiol., 31 (1971), 207.). A bolus of CO₂ washout profiles. The rate of increase of effective diffusivity depends on the local flow conditions and differs with branches. The effective diffusivity in the left main bronchus is 3.2 times greater than that in the straight tube, whereas no significant difference is observed in the right main bronchus. In addition, the presence of a stationary period augments axial gas transport in intermittent oscillatory flow because it provides time for the occurrence of lateral mixing in the radial direction, i.e., axial diffusivity is 1.6 times greater on the average than that in sinusoidally oscillatory flow.

抄録全体を表示

The main purpose of this paper is to develop a consistent method for evaluating boundary and internal stresses in order to satisfy the momentum conservation discretized by a finite element method (FEM) in a LES (Large Eddy Simulation). The paper focuses on the following two aspects : (1) investigation of a stability method to calculate the divergence form for the convection term at a high Reynolds number, particularly in the LES, and (2) development of a post-processing technique to calculate Reynolds stresses consistently with a discretized Navier-Stakes equation. To achive numerical stability, a one-point quadrature is used to integrate the divergence form of the convection term. The post-processing technique has been developed such that the boundary and the internal Reynolds stresses are treated as unknown values and their weak solutions are solved using the Galerkin FEM. The present method has been applied to a turbulent channel flow. As a result, the Reynolds stress by the present method has shown better agreement with the DNS (Direct Numerical Simulation) data than that by the conventional nodal evaluation method.

抄録全体を表示

In casting simulation, flow analysis using an irregular grid is required to predict filling pattern in molds of complicated shapes. For treatment of free surface movement, a new technique based on the VOF (volume fraction of fluid) method has been developed using the FVM method. The inclination of free surface in an irregular cell was estimated using the spatial gradient of F value. A criterion to control the numerical false diffusion of F value on the free surface was adopted in the new algorithm. Two-dimensional analyses of free surface movement in a uniform velocity field were performed to check the accuracy and stability of the calculations using different ways to estimate the spatial gradient of F value. The results showed that the proposed method is suitable for free surface movement treatment using an irregular grid, and proper predictions of free surface movement can be realized using the diffusion control criterion. The accuracy and stability of the analyses are greatly affected by the methods for estimating the spatial gradient of F value, especially under high Courant Condition.

抄録全体を表示

In this study, the applicability of CIP-FEM to numerical simulation of two-dimensional unsteady incompressible viscous fluid flow is discussed. In this procedure, the basic differential equations of the flow are separated into advection and non-advection phases. Analysis of the advection phase is based on an approximate solution given by a cubic polynomial interpolation and the one of the non-advection phase is performed by FEM. The formulation in order to solve the basic equation of viscous fluid by using this procedure is discussed and also described. Numerical simulation of the problem of viscus flow is carried out. The applicability and effectiveness of the proposed method is discussed with the obtained numerical results.

抄録全体を表示

Control of the aerodynamic noise generated from two-dimensional circular and square cylinders are studied experimentally in a low noise wind tunnel. The circular cylinder of 1cm diameter is inclined against the mean flow, whereas the square cylinder of 1cm² cross section is placed normal to the mean flow, with various angles of attack. Mean velocity of the flow is between 20.6m/s and 41.6m/s. Powere spectra of the radiated noise in various angles of inclination or angles of attack are measured in order to examine the changes in peak levels and the Strouhal numbers of the so-called Aeolian tones. For the circular cylinder, it is found that the Strouhal number based on the frequency of the Aeolian tone and the normal component of the mean velocity to the axis of the circular cylinder stays nearly constant for the angle between 0° and 45°. The peak level of the radiated noise decreases rapidly from the inclined angle from 0° to 15°. but it stays nearly constant for the angle over 15°. For the square cylinder, the Strouhal number is smallest and the peak level is highest at the angle of attack of 0°. The peak level decreases and the Strouhal number increases rapidly with increase of angle of attack up to 13°. For the angle above 15°, both the Strouhal number and the peak level stay almost constant. It is concluded that a circular cylinder should be inclined more than 15°. and a square cylinder should have an angle of attack well over 15° in order to reduce aerodynamic noise generation.

抄録全体を表示

Outflow boundary conditions (OBC's) are very important in the numerical simulation of incompressible viscous flows. In this study, we test three types of OBC's in a three dimensional simulation of the Hill's spherical vortex convected in a uniform flow. These OBC's are a typical Sommerfeld Radiation Condition (SRC) type which uses a uniform flow velocity for it's convection velocity, a new SRC type which uses an arithmetic mean velocity of the maximum normal velocity and the minimum normal velocity at the outflow boundary, which we will propose, and a conventional free outflow condition. The results show that of the three, the SRC type which uses an arithmetic mean velocity elicits the most accurate condition.

抄録全体を表示

In this paper, we propose a new GSMAC-FEM (generalized simplified marker and cell-finite element method) which is suited to the numerical analysis of visco-elastic fluids. The equation of continuity and the equation of momentum are solved by the GSMAC-FEM algorithm and the constitutive equation is solved by the finite volume method. This scheme employs the third order MUSCL (Monotone Upstream-centered Schemes for Conservation Law) in order to guarantee the absence of spurious oscillation near the steep gradients of the variable. This method uses a minmod limiter in order to satisfy the TVD (Total Variation Diminishing) condition. The present method employs the simultaneous relaxation of velocity and pressure for the incompresible condition. The flows of Maxwell fluid through two dimensional planer abrupt contraction are calculated by the present method and the effects the Weissenberg number and the Reynolds number are discussed.

抄録全体を表示

A three-dimensional explicit time-dependent full Navier-Stokes computation was carried out for a turbulent subsonic mixer-ejector flow field. The objective was to obtain better insight into the complex flow field and to utilize computational fluid dynamics (CFD) to design a mixer ejector with less jet noise while maintaining high thrust and efficiency. The grid consists of about 660 000 nodes in two blocks. The numerical simulation shows that the flow field was dominated by a system of large streamwise vortices. These vortices are created at the exit of the mixer nozzle, and they exert a significient influence on the mixing process between primary and secondary flows. Computed velocity, total pressure and total temperature distributions in the exit of the mixer ejector were compared with experimental results. Agreement between computational results and experimental ones is generally good.

抄録全体を表示

Coherent structures in a turbulent plane mixing layer are comparatively studied by wavelet transform and proper orthogonal decomposition (POD) analysis, using the database of the mixing layer obtained by Bonnet et al. (1994). The POD analysis yields more coherent vorticity distributions of the large-scale structures in the mixing layer, suggesting that the first mode represents the most dominant structure. The wavelet transform indicates the low-frequency modulation of vorticity of the large-scale structure. The POD analysis fails to show the large-scale structure in the central part of the mixing layer due to the effects of the random components, whereas the wavelet transform can extract the coherent structure if the vortex-passage period is chosen as the scale of the wavelet.

抄録全体を表示

Generally speaking, the finite-element method in computational fluid dynamics is universally accepted, however computation by the CPU is time-consuming and requires large memory capacity for data storage. Therefore development of an analytical formulation to reduce the time and storage required for calculation is desired. In this paper, we propose a novel discrete del operator method in order to overcome these defects. This method is formulated using the discrete del operator as the element coefficient matrices in finite-element analysis, and low-memory and high-speed calculations are carried out. In particular, we examine whether this method is effective for the numerical analysis of natural convection of thermoelectrically conducting fluids in a cubic cavity. Moreover, we estimate the effect of heat transfer enhancement under a weak magnetic field on the Hartmann number.

抄録全体を表示

The afterglow technique was applied to visualize hypersonic flow past a compression corner and that past an expansion corner. The experiments were conducted at the Kobe University hypersonic shock tunnel facility. An electric discharge column was generated in a flow in which there was a region of reverse flow and in a flow which expanded, and afterglow was observed. Good quantitative agreement with a result of Schlieren visualization of a shock wave shape was obtained. The afterglow results were also compared with the results of calculations based on the two-dimensional full Navier Stokes equations. The technique can be used to obtain a qualitative flow velocity profile. Finally, several error sources were discussed.

抄録全体を表示

The dynamics of high-speed phenomena induced by focusing of pulsed laser beam in liquid nitrogen is studied experimentally. Before the growth of a spherical bubble, a series of nonspherical phenomena is observed. The laser-induced breakdown creates luminous plasma around the point of focus. When the pulsed laser beam is switched off, the relaxing plasma and the resulting gas form a nonspherical bubble, which emits a nonspherical shock wave. Then several compression waves following the preceding shock wave are emitted from the bubble. As they overtake the shock wave, they make both the shock wave and the bubble spherical. A possible mechanism for these phenomena is considered.

抄録全体を表示

We propose a new method for the measurement of temperature in high temperature air using the LIPF (laser-induced predissociative fluorescence) of O₂ molecules excited by a broadband ArF excimer laser. In this method, temperature is measured using the temperature dependence of the ratios between the fluorescence intensities integrated in two spectral regions including peaks. The theoretical broadband spectra of O₂ molecules are calculated and compared with the experimental spectra of static hot air (300∼1100K) inside an electric furnace. Temperatures measured using this method agree with those measured by a thermocouple within about ±100 K at temperatures below 1000K.

抄録全体を表示

The sound radiation from a cylindrical duct having a cavity at the open end has been numerically studied. The structures of the entire sound field is numerically obtained for various depths and radii of the cavity. The governing equation for acoustic disturbance is a linearized Euler equation and is solved by a finite difference method. The present study shows that the cavity affects the sound diffraction and reflection at the open end of the duct and changes the radiated sound field. The cavity of a particular depth reduces the sound power significantly and changes the distibutions of intensity and sound pressure level at the open end of the duct.

抄録全体を表示

One of the major limitations of wind power is the inconsistency of wind turbine generator (WTG) output. However, there is a possibility that a multiple wind turbine system could reduce fluctuations in a total output by canceling out each other's variations. Theoretically, the deviation of the multiple wind turbines'fluctuations is reduced at a rate inversely proportional to the square root of the number of turbines. The authors have constructed a hybrid multiple WTG system, which consists of three 1kW wind turbine generators and four 21.6W polycrystal solar panels. The generators are directly connected to the common 24V d.c. electric line system. Measurements of the performance of each generator, such as rotational speed of the turbines, fluctuating electric currents and line voltage, and the fluctuations of each generator's output are analyzed. Our results show that the hybrid multiple WTG system successfully reduces wind turbine output fluctuations.

抄録全体を表示

We have experimentally studied the relationship between the periodic velocity fluctuation in the near wake of a rotating NACA 65 Blade and the generated broad band noise at different setting angles of the blade. The typical periodic velocity fluctuation in the near wake was not clearly observed at a design point but was observed at the attack angle at off design points. The frequency of the produced velocity fluctuation caused by Karman vortex shedding from the trailing edge of the blade increased toward the tip of the blade. Then, the spectrum of noise generated from the blade becomes broad band. In addition, we have proposed a method to predict the spectrum of noisc generated from blades due to Karman vortex shedding. The validity of the method was experimentally verified.

抄録全体を表示

Industrial control valves sometimes generate a peculiar noise under the low-pressure condition, although the noise problem of control valves becomes crucial in the case of high pressure. The purpose of the present study is to clarify the mechanism of a whistle-like tone generated under lowpressure conditions. We set up a model pipe-line which contained a globe valve plug of"equal percentage characteristics". The flow inside the control valve was visualized and observed with a high-speed video camera. We focused particular attention on the effect of the length of a seat ring. It was found in the present experiment that the length of the sea tring affects the peak frequency of noise generated from the control valve. We established an empirical formula between the frequency of whistle-like noise and the seat ring length.

抄録全体を表示

The paper discusses the characteristics of a flapper-nozzle system for water hydraulic servovalves. High pressure water at the supply port is first used as the working fluid for the hydrostatic bearings supporting the spool. Spool valve stiction induced by poor lubrication with water is thus avoided. The fluid is then led to the ends of the spool and is used as the working fllid of the flapper-nozzle system. In the new flapper-nozzle system the circumferential clearance of the spool becomes a laminar restriction that substitutes fixed orifice used in conventional servovalves. The linearity in the pressure-displacement relationship of the new flapper-nozzle system is better than that of conventional fixed orifice systems.

抄録全体を表示

In this paper flow force due to nozzle flow and its influence on the flapper-nozzle system for water hydraulic servovalves are discussed. The flow force acting on the flapper is estimated using the momentum theory, and expressed as a function of the nozzle pressure and flapper-nozzle gap. Analysis shows that the double nozzle flapper gives a quasilinear characteristic to the relationship between the force on the flapper, the flapper gap and the spool velocity. The numerical result based on the theory is used to estimate accuracy of the classical linear approximation. The flow force reduces the magnitude of the change of the nozzle back pressures which drive the spool of the water hydraulic valve, because the input moment acting on the armature is resisted by the moment due to the flow force. Experimental verification of the theory is also given.

抄録全体を表示

An electrohydraulic servovalve that uses clear water as a pressure medium (water hydraulic servovalve) is developed. A conventional electrohydraulic servovalve cannot work with this hostile fluid. This paper explains the structure and functions of components, and the relationship between them, as well as development procedure of the water hydraulic servovalve. The basic idea for the development is to support the spool of the valve with hydrostatic bearings and to lead the water from the bearings to the flapper-nozzle system, which, in this valve contains a laminar restriction upstream from the nozzle instead of the turbulent restriction in the conventional flapper-nozzle system. This paper also explains the design procedure of the hydrostatic bearing in the valve and the commection of electromagnetic elements and the fluidic elements. The manufactured water hydraulic servovalves show good characteristics in their smooth motion. endurance and controllability, including dynamic characteristics.

抄録全体を表示

Flow pattern and performance of a return channel model were investigated experimentally, under various operation points in the turbining mode. The position where the secondary flow conspicuously occurs near the return channel inlet changes gradually from the casing wall to the pressure surface of the return vane, with decreasing the turbine output power. Such a secondary flow induces swirling flow in the downstream passage and makes the through flow under-turning. The features of the hydraulic loss to the change of the turbining peration point are also decided near the return channel inlet, and the loss increases markedly in the low output power range because of the shock loss at the return vane inlet.

抄録全体を表示

Focusing on the effects of liquid velocity and subcooling, forced-convection film boiling heat transfer was investigated experimentally and analytically. In the experiments, a long-vertical cylinder was adopted as the test section and R-113 liquid was used. The measured heat transfer coefficients, which were higher than those predicated with the two-phase boundary layer theory even in a low liquid velocity region, showed a different trend from the values calculated by Dittus-Boelter type correlations based upon the turbulent flow of a liquid or vapor phase and were not dependent upon the distance from the leading edge. The characteristics were quite similar to those of pool film boiling. The vapor-film-unit model developed for natural-convection film boiling was extended and applied to the present conditions. In the model, the Kelvin-Helmholtz instability wavelength was used to determine the length scale of the unit. The model provided good agreement with the experimental results. It was also pointed out that if the viscous sublayer in the turbulent boundary layer, which was formed at the upstream of the heated section in the bulk liquid flow, was much thicker than the vapor film and the boundary layer formed in the liquid along the vapor film, the turbulence of the bulk fiow had little effect on the film boiling heat transfer.

抄録全体を表示

In the present study, the dynamic behavior of liquid-solid contact phenomena in saturated pool boiling is observed from beneath the boiling surface by using a single crystal sapphire plate as the boiling surface. The results indicate that at low wall superheat in the nucleate boiling region, most of the dry areas are the primary dry areas corresponding to the bubble base areas, but secondary dry areas appear through coalescence of the primary dry areas with incresing wall superheat. However, even at wall superheat near the critical heat flux point, the ratio of the total area of primary dry areas to the total dry areas is not so small. This fact indicates that, even at the critical heat flux point, not only secondary dry areas but also primary dry areas contribute to boiling heat transfer. To account for the contribution of all the dry areas equally, the contact-line length density which is the total length of the periphery between liquid-solid contact and dry areas in a unit heat transfer area is measured. The dependence of this new time- and space-averaged quantity on the wall superheat is very similar to the boiling curve measured on the sapphire surface. Using the experimental data of the contact-line length density and the model of evaporation heat flux per unit contact-line length developed by Dhir and liaw, boiling curves are predicted as a function of contact angle. The calculated boiling curve for the contact angle of 30 degrees is in good agreement with the measured boiling curve. The results indicate that the contact-line length density, which represents the ratio of intensively evaporating area to the total heat transfer area, is a more direct measure of boiling heat transfer in high-heat-flux nucleate and transition boiling regions than the liquid-solid contact fraction is.

抄録全体を表示

An experimental study on the boiling heat transfer to water flowing in a tube at pressures of 0.3, 0.5, 1 and 2 MPa was conducted. Based on the data in the subcooled boiling region, a heat transfer model, in which the heat flux was postulated to consist of forced convection and nucleate boiling contributions, was developed. The forced convection heat flux was evaluated by the single-phase convective heat transfer correlation by taking into account the effect of increase in liquid velocity due to net vapor generation. The nucleate boiling heat flux is determined from an empirical correlation that takes into consideration promotive and suppressive effects on the incipience of boiling. The present model was compared with the experimental data and various known correlations and satisfactory results were obtained.

抄録全体を表示

Combined heat transfer characteristics are obtained numerically for three-dimensional natural convection and thermal radiation in a long and wide vertical porous layer with a hexagonal honeycemb core. The porous layer is assumed to be both homogeneous and isotropic. The pure Darcy law for the fluid flow and Rosseland's approximation for the radiation are employed. The numerical methodology is based on an algebraic coordinate transformation technique and the transformed governing equations are solved using the SIMPLE algorithm. The effect of radiation on the heat transfer characteristics is investigated in a wide range of radiation numbers and emprerature ratios, for two Darcy-Rayleigh number values (Ra^*=100, 1000), and for a fixed aspect ratio of H/L=1. The results are presented in the form of combined and convection heat transfer coefficients, and are compared with the corresponding values for pure natural convection.

抄録全体を表示

The combined natural convection and radiation heat transfer characteristics in a vertical porous layer with a hexagonal honeycomb core were investigate experimentally. The temperature distributions on the honeycomb core wall and the combined heat transfer rates through the porous layer were measured. The measurements of the heat transfer were accomplished using the guarded hot plate (GHP) method. The honeycomb core wall was made of paper and large mesh foamed resins were inserted into the honeycomb enclosures. The meeasurements were performed while varying the radiation parameters between 0.5 to 0.65, varying the temperature ratios between 0.01 to 0.1 and varying the Darcy-Rayleigh numbers between 5 to 80, and for a fixed aspect ratio of H/L=1. The experimental results for Nusselt numbers agreed well with our available numerical results.

抄録全体を表示

This paper deals with the visualization of particle recirculation behavior and with the estimation of heat transfer enhancement due to particle convection in a fluidized bed. In order to examine the trajectory of a fluidized particle in a bed, fluorescent particles put into the fluidized bed were illuminated by a UV light, and their motion was recorded on a VCR. The Visualized results enabled us to evaluate the contact period and contact frequency between the fluidized particles and a heat transfer surface immersed in the bed, as well as the particle velocity in each region of the bed. Using the visualized recirculation behavior of a fluidized particle, heat transfer enhancement due to "particle-convection" was estimated, and the contribution of the effect to the total heat transfer was discussed.

抄録全体を表示

Transient natural convection in water around a cooled horizontal cylinder placed at the center of a rectangular enclosure has been studied. The governing equations were solved by a finite difference method, and the flow structure and temperature distributions have been predicted. The purpose of the present study was to examine the effects of the density inversion of water and the initial temperature of water on the cooling process of water. The initial water temperature was varied 4, 6, 8 and 12°C, and the timewise variations of the temperature field and the velocity field as well as the mean and local Nusselt numbers on the cylinder surface were compared. The change in the initial water temperature largely affects fluid flows due to the density inversion of water. Complicated flow patterns are observed for initial water temperatures larger than 4°C. The cooling of water is most effective for an initial water temperature of 12°C.

抄録全体を表示

Flashing of liquid nitrogen in a pressure vessel (cryostat) was observed at depressurization rates from 0.01 to 4.0 MPa/s. The explosive boiling behavior was observed by using a video camera. Pressure and temperature changes in the pressure vessel were measured. In the case of high depressurization rates, mist formation was observed in the vapor phase near the vapor-liquid surface at the early stage of the depressurization rocess. This mist layer becomes more dense as the depressurization rate increases. Observations of mist formation and the calculated temperature drop of the vapor under an adiabatic expansion process show that mist formation depends on the vapor expansion and boiling near the vapor-liquid surface. Mist formation in flashing phenomena play an important role in which the thermal nonequilibrium states between the supercooled vapor and the superheate liquid are relaxed due to depressurization.

抄録全体を表示

A new unsteady-state method that determines thermal contact conductance, or resistance, between two solid materials in contact with each other is presented. The feature of this method is that it enables conductance to be determined by applying an arbitrary-functioned heat input to the specimen. The general measurement principle of the method is derived for systems of one-dimen-sional rectangular coordinates. The principle utilizes the Laplace transform of the heat conduction equations for two materials and the continuity equation of the heat flux at the interface. The equation for determining conductance involves the Laplace integral values of the temperature responses measured at two positions in each sample. A series of numerical simulations is performed to evaluate the measurement errors of this method. Experimental study is performed to verify the validity and usefulness of the method.

抄録全体を表示

A new apparatus to measure the mutual diffusion coeffcient of liquid mixtures based on the forced Rayleigh scattering method using the Soret effect has been developed. The characteristics of the method are (1) a short measurement time of about 1 s, (2) a sample volume of a few mm³, (3) no addition of photochromic dyes, (4) the concentration dependence of the mutual diffusion coeffcient is detectable. The purpose of the present study is to establish the Soret forced Rayleigh scattering method as a reliable technique for the measurement of thermophysical properties. It was found through experimental study on polymer solutions (solute : polystyrene : solvents : ethyl acetate and toluene) that this method has considerable potential for many applications.

抄録全体を表示

The formation process of high-voltage ultra-short pulsed plasma in argon gas has been clarified by computer simulation based on mass conservation equations of electrons and ions coupled with Poisson's equation, including the transport processes of resonance photons which cause photoemission of secondary electrons on the cathode. First, the initial charged particle densities for high-voltage ultra-short pulses were estimated through a simulation of an after-glow period following the end of the former pulse. In this after-glow period, the decay process of plasma is mainly governed by recombinations and a large amount of charged particles, more than 10⁷cm^-3, remain even 20ms after the voltage is turned off. Then, the formation process of the plasma structure in the high-voltage ultra-short pulse was analyzed. The residual charged particles generated by the former pulse assist in the formation of the plasma structure in the next pulse. The electron photoemission has a significant effect on the formation process of th plasma structure, especially a very short time after the voltage rise. The lag time of the current rise obtained through this simulation agrees well with our experimental result. The model developed in this study could become a fundamental basis for applying this type of pulsed plasma to the active control of plasma chemical reactions.

抄録全体を表示

This paper deals with fluidization and heat storage characteristics of shape-stabilized latent heat storage particles packed into a fluidized bed-type heat storage vessel. The shape-stabilized latent heat storage material consists of normal paraffin (pentacosane C₂₅H₅₂, latent heat 164 kJ/kg, melting point of 327.2 K) as a latent heat material, and polyethylene as a shape-stabilizing material. The pressure losses both on fluidized and on fixed particle layers were measured in order to investigate flow behavior in the heat storage vessel. It was found that the pressure loss of the fluidized particle layer was lower than that of the fixed particle layer. Furthermore, the effects of hot air flow rate, inlet air temperature and the amount of heat storage particles packed into the heat storage vessel on the completion time of the heat storage process were investigated. As a result, the nondimensional correlations for the completion time of the heat storage process were expressed in terms of nondimensional pumping power, the Stefan number and the ratio of the packed particle layer height to the diameter of the cylindrical heat storage vessel.

抄録全体を表示

Economic and energy-saving properties of simple-and combined cycle gas turbine cogeneration systems for use as a designated electric supply are compared. In this supply, the properties are influenced by an electricity supply obligation. An optimization approach is adopted to efficiently evaluate the influence on the properties. Using this approach, equipment capacities and a maximum contract demand of town gas are determined to minimize the annual total cost in consideration of operation strategies for energy demand requirements and of the electricity supply obligation. Numerical studies are carried out using the capital unit cost of a cogeneration unit and the unit cost of energy charge of town gas as parameters. The results show that the combined-cycle system is more suitable as a designated electricity supply, but the unit costs for electricity supply by both systems do not become lower than that by power company unless the parameters are very small.

抄録全体を表示

Numerical computations are made of axisymmetric hydrogen laminar jet diffusion flames taking into account detailed chemical kinetics and multicomponent diffusion. The results obtained are as follows. (1) A very high concentration of NO is formed in the flame of 100% H₂ fuel burnt with air. Dilution of H₂ by N₂ reduces the NO concentration significantly. (2) When 100% H₂ is burned with O₂ or O₂ diluted with H₂O, considerably high temperature is generated even though the amount of H₂O diluted is 70%. This is the reason why a small amount (1% by volume) of N₂ in the oxidizer results in the formation of a considerable amount of NO. (3) The amount of NO₂ and N₂O formed are low compared with that of NO.

抄録全体を表示

The local reaction rate and the diffusion rate may be useful values for the study of turbulent diffusion flames. but they cannot be directly measured. In this study, an attempt was made to obtain them by numerical calculation using measured values for a hydrogen jet diffusion flame. The following results were obtained. (1) The correlation between fluctuating velocities u″v″^^^∼that were obtained with the developed method was compared with the measured value. As a result, the method was revealed to be reliable. (2) The molecular diffusion coefficient of hydrogen becomes large and not negligible compared with the turbulence diffusion coefficient, in a high-temperature reaction region of comparatively weak turbulence. (3) It was suggested that the method developed in this study has the capability of giving reasonable values for local reaction rates in a hydrogen jet diffusion flame.

抄録全体を表示

In a premixed flame stabilized by a V-gutter holder, direct numerical calculation was performed aiming at the examination of the interaction of fluctuating vortices and the combustion reaction caused downstream of the V-gutter and the elucidation of the effect of flow fluctuation on the behavior and the stability of the premixed flame. Important dimensionless parameters which affect the phenomenon are the Reynolds number and Damkohler number, and the effect of these parameters on the behavior and the stability of the premixed flame was systematically clarified. The flame configuration is classified according to the Damkohler number. The flame is blown off when the Damkohler number is very small and the Karman vortex street is formed. A flame which adheres to the V-gutter is formed when the Damkohler number is fairly large, and changes into a long one which expands to the calculation area downstream from a short one as the Damkohler number increases. This long flame might be parallel or oblique to the direction of the uniform flow of a premixed mixture. The effect of stabilization due to the density gradient in the velocity shear layer downstream of the V-gutter on the Kelvin-Helmholtz instability is important, and flow is stabilized when the flame is parallel to a premixed mixture flow.

抄録全体を表示

In order to facilitate practical application of hydrogen fueled engines with external mixture formations. it is necessary to determine the cause of backfire and to identify appropriate countermeasures. It has been found in our study that one of the causes of backfire is attributed to the electric discharge which occurs after the ignition of the spark plug when the pressure in the cylinder becomes low enough to discharge the electric charge remaining in the ignition system. The electric charge in the plug cable remains much longer when hydrogen is used as the fuel than when gasoline is used. In other words, the number of ions produced from hydrogen-air combustion after the ignition of the spark plug is less than that from gasoline-air combustion, so it is difficult for the ion current to flow out through the gap of the spark plug. It has been found in our experiments that backfire takes place when the electric discharge occurs during the intake stroke and that engine operation could be performed up to an excess air ratio of λ=1 without any backfire at a low engine speed when the ignition system was modified such that no electric charge remained in the spark plug cable.

抄録全体を表示

In order to reduce fuel consumption in an indirect injection (IDI) diesel engine, a new combustion system that increases the degree of constant volume of heat release and reduces heat loss was investigated using a four-cylinder diesel engine with 98 mm bore and 98 mm stroke. As a result, the new combustion system which consists of a new combustion chamber with a refined jet passage in which the jet passage angle inside the main chamber is smaller than that inside the prechamber, an enlargement of the jet passage area, a miniaturized injection nozzle with a reduced needle diameter and a plunger with an enlarged diameter, was proven to increase the degree of constant volume of heat release, reduce heat loss, and reduce fuel consumption over the wide range of engine operation.

抄録全体を表示

For reducing exhaust emissions, the effect of pilot injection with a short pilot-main interval on combustion was investigated in a turbocharged direct injection diesel engine. Although the amount of pilot injection could be decreased by increasing the Dodge plunger seat diameter, two types of technique for reducing the amount of pilot injection were examined to suppress an increase in smoke due to pilot injection ; one is a reduction in the nozzle-hole diameter and the other is an increase in the spring chamber pressure. Each technique was compared from the view points of combustion, smoke density and the trade-off relationship between NO_x and fuel consumption. Smoke is reduced as the amount of pilot injection decreases while the trade-off between fuel consumption and NO_x is markedly improved by reducing both the nozzle-hole diameter and the amount of pilot injection.

抄録全体を表示

Many studies have been conducted on heat flux or the heat transfer coefficient in an engine in order to increase thermal efficiency and to solve thermal loading problems. The authors showed the relation between Re number and Nu number at near TDC under the motored condition and it was found to be subjected to the equation Nu=a+b·Reⁿ. To examine whether the formula could be adapted to compression and expansion strokes, the formula was compared to the experimental data during compression and expansion strokes. Also, the value of the authors' formula was compared to that of usual formula. The authors' formula is found to agree well with the experimental data, and it was confirmed to predict approximate the heat transfer coefficient in an engine.

抄録全体を表示

The authors have studied the effects of the heat release rate (combustion rate) on emissions in IDI diesel engines. In this study, the heat release rate in main and swirl chambers were separated using measured pressure values in both chambers to estimate the combustion characteristics. As the result, it was found that, in partial load conditions, the reduction in combustion quantity in swirl chamber is effective in reducing the PM emission and that this improvement can be realized by retarding the fuel injection timing and optimizing the injection rate.

抄録全体を表示