A direct numerical simulation (DNS) of flow around a two-dimensional circular cylinder by the higher-order finite difference scheme (FDS) is conducted. Inflow and outflow boundary conditions constructed and successfully tested originally on a spatially developing mixing layer are applied to the above mentioned flow with R_e=226. Comparisons of the general flowfield and the statistics with those obtained in an extended computational domain arc made. In addition, the appropriateness of the outflow boundary condition imposing the zero gradient of the velocity and pressure is investigated. From these studies, the following conclusions are obtained. (1) For the newly proposed outflow boundary condition, the velocity and pressure fields are highly correlated to the results obtained in the extended computational domain. The correlation coefficient of the pressure with that obtained in the extended computational domain is higher than 0.999. (2) For the conventional zero gradient condition, a significant discrepancy from results obtained in the extended domain is observed for both velocity and pressure fields, and the correlation coefficient of the pressure is as low as 0.966. (3) The conventional zero gradient condition overestimates the drag coefficient.

View full abstract

The droplet entrainment in an annular flow plays an important role in heat and mass transfer. owever the fundamental experimental data such as film thickness, entrainment flow rate and disturbance wave velocity in the case of an inclined upward annular flow are very few compared with similar data in the case of two-phase flow in vertical and horizontal pipelines. In this paper, experimental data on the distribution of entrainment flow rate and the onset of liquid entrainment in an inclined upward annular flow are presented and the effects of the inclination angle are examined. In this experiment, the isokinetic sampling probe technique was used for the measurement of the entrained droplet flow rate and the ranges of inclination angle examined were 0, 30, 45, 60°and 75°from horizontal. As a result, it was clarified that the difference in droplet mass flux between the upper and lower tubes becomes small as pipe inclination increases, the entrainment fraction increases as both the flow rate and the inclination increase and the critical gas velocity for the onset of entrainment becomes small as the liquid film Reynolds number and the inclination increase. Furthermore, Ishii & Mishima's correlation for the entrainment fraction was modified for application to the inclined upward annular flow.

View full abstract

It is well known that the fluid dynamics of arterial blood flow plays an important role in arterial disease. Periodic blood flow through a stenosed tube was analyzed numerically. The bi-viscosity model is used as a constitutive equation for blood, and the flow is assumed to be periodic, incompressible and axisymmetric. Effects of pulsation and the rheological property of blood are considered. The flow pattern, separation region and the distributions of pressure and shear stress at the wall are obtained. The results show that the non-Newtonian property reduces the strength of vortex downstream of the stenosis and has considerable influence on the flow even at high Stokes and Reynolds numbers, provided the pulsatile ow has a stagnant period.

View full abstract

The fluid flowing in heated curved pipes is subjected to centrifugal and buoyant forces. Fully developed laminar flow in horizontal, loosely coiled pipes is investigated through similarity arguments and computational studies. The thermal boundary condition at the wall is uniform wall heat flux axially and uniform wall temperature peripherally. Flow and heat transfer are characterized by three parameters: Dean number, Prandtl number and buoyancy parameter. Detailed structures of velocity and temperature fields are shown for a wide range of these parameters.

View full abstract

Experinlental investigations arc performed on flooding which occurs in vertical circular and annular channels. Water film flows down along the channel surface through porous cylinders, and air is raised from the channel bottom by a suction pump. The ooding velocity of the circular tube is proportional to the diameter ill agreement with earlier investigations For the annular channel, a ratio of liquid flow rates of both channel surfaces has a strong effect on the flooding velocity. The conventional method cannot predict the flooding velocity. In the present study, a characteristic diameter of the annular channel is determined by the ratio of liquid flow rates for Reynolds number of liquid and gas. The flooding velocity is approximately indicated as a uniform curvature with the Reynolds number of liquid.

View full abstract

In this paper the limiting fluid velocity allowing steady state transport of solids, called the 'minimum transport velocity', is determined with regard to low power consumption. An empirical equation to predict the minimum velocity is derived for fully developed flow in a horizontal pipe, and an experiment on the minimum velocity of coarse particles in the flow is conducted. In the case of a short pipeline, the above mentioned equation gives results which agree relatively well with the experimental data.

View full abstract

An algebraic model of the turbulent transport of Reynolds stress is proposed, based on an examination of the budget of the transport equation for the third-moment of fluctuating velocity. Combined with a basic version of the second-moment closure, the new model is applied to the calculation of a two-dimensional free mixing layer. The results show a slight modification of the conventional gradient transport model for the turbulent diffusion, and yield a better agreement with experimental results in terms of the amplitude of the third-moment. The inclusion of a simple model for the pressure diffusion term indicates the possibility of further improvement of the turbulent transfer process in free flows.

View full abstract

A wall point source plume, that is emitted to a wall turbulent shear flow, will be mixed and convected by the turbulent flow which scale varies with distance from the wall. This feature makes the wall point source plume show the bimodal feature in its concentration fluctuation held. In this paper, the statistical properties of the concentration fluctuation of the wall point source plume in a turbulent pipe flow will be mainly reported. Vertical profiles of the integral scale in several cross sections downstream from a wall point source show self-similarity and bimodal distributions like as in a point source plume in a turbulent boundary layer on a flat plate. In addition, vertical profiles of the dissipation of the concentration fluctuation show also self-similarity. Kullback-Leibler divergence of PDFs of the concentration will also be discussed.

View full abstract

A method was proposed for measuring the radial void distribution of a two-phase flow in a round tube using neutrons as probes. This method involves (1) assuming an appropriate function which could approximate the radial void distribution, (2) deriving a function to approximate the void distribution projected on a plane perpendicular to the incident neutron beam and along the transverse direction (x-direction) by integrating the function of the radial void distribution, and (3) determining parameters in the function of the radial void distribution by the least squares method. It was shown analytically and experimentally that the effect of scattered neutrons and unparallelness of the incident neutron beam, which might affect the applicability of this method, could be corrected and ignored, respectively, by taking an appropriate distance between the converter and the test section. This method was applied to the measurement of a temporally averaged radial void distribution of an air-water flow in a round tube of 3.90 mm inner diameter. It was revealed that complex void distribution with a saddle-shaped profile could be measured by assuming an appropriate function.

View full abstract

In order to simulate behavior of shock wave propagating in a square kinked duct, a numerical method based on the TVD scheme was developed. Both initial conditions and mesh sizes were tested to simulate the experimental results obtained from exploding wire experiments. Numerical density contours showed qualitatively good agreement with experimental results of schlieren photographs. Measured pressure ratios across the shock front also showed good agreement when they were compared quantitatively with computations along the upper edge, lower edge and center line of the duct. The present method using a narrow high pressure region in a duct as an initial condition was shown to be useful for predicting blast wave decay.

View full abstract

Two types of recently developed improved dynamic sub-grid scale (SGS) models were investigated by simulating turbulent channel flow at R_e=180 normalized by channel half-width and friction velocity. The results of those models were compared with DNS data obtained by Kim et al. in 1987. One model developed by Meneveau et al. in 1994, called, the Lagrangean dynamic SGS model, appeared to show some superiority in eliminating the numerical instability that is often observed in the original dynamic SGS model proposed by Germano et al. In 1991, because the results of the model showed good agreement with those of the original dynamic SGS model without statistically homogeneous direction of turbulent flow that is required in the original one. The results of the other model developed by Vreman et al. in 1994, which is called the dynamic mixed SGS model showed better agreement with DNS data than that of the original dynamic SGS model but some numerical instability was still observed.

View full abstract

A massively parallel computer, which has a large number of processors, is expected to become the main instrument for scientific numerical analysis including CFD field.It has been pointed out that efficiency of parallel computing becomes worse when the number of processors increases and granularity comes to fine. So an acceleration technique should therefore be introduced to achieve nearly maximum performance of massively parallel computers. Some researchers have used the multigrid method, but this technique turned out to be inappropriate for parallel computing of very fine granularity because efficiency rapidly worsens. The purpose of this study is to propose a numerical technique suitable for calculation of incompressible flows on massively parallel computers. In this regard, we choose the extrapolation method as the accelerative technique, which predicts converged solutions from a sequence of intermediate solutions and is expected to retain its accelerative property for fine granularity. Three existing extrapolation methods ROLE, MPE, ROGE and our newly developed LWE are discussed. Furthermore, ROLE and LWE are applied to numerical analysis of Poisson's equation and the results are discussed.

View full abstract

Jets exhausting into a two-dimensional channel cause self-induced oscillation which is observed under a certain condition due to the instability of the Coanda effect. We call the above-mentioned flows as "flip-flop jets". Their behavior is governed by nonlinear features of the Navier-Stokes equations. Numerical results obtained by unsteady computations were compared, with chaotic data of the Rossler model for the spectrum of velocity variations and the Poincare map. Moreover, wavelet analysis was applied to our data. Since some similarities between the two kinds of data have been shown, we concluded that the flip-flop phenomena exhibit a chaotic behavior.

View full abstract

Here we present detailed experimental results concerning flow-induced vibrations of long-span, shell-type gates in which the upstream gate face consists of vertical and inclined skin plates (also referred to as weir plates). Such shell-type gates possess two degrees-of-freedom, one each in the streamwise (horizontal) and vertical directions, due to bending flexibility in those two directions.The streamwise and vertical vibrations can become closely coupled with each other through the hydrodynamic forces acting on the weir plates, resulting in severe self excited vibrations. A two-dimensional laboratory model of a long-span, shell-type gate was operated with underflow only (i.e., no overflow) at small gate openings with several different inclined weir plate geometries, ranging from a 17.5°to a 65°inclination angle (relative to the horizontal) for the inclined weir plate. By measuring the gate's vertical and horizontal displacements, it was possible to determine the vibration frequency, the excitation ratio (negative damping ratio) and the trajectories of gate motion. These results show that inclination angle of the inclined weir plate angle plays a significant role in determining the gate's susceptibility to this type of dynamic instability. Long-span, shell-type gates with an inclined weir plate angle of about 60°relative to the horizontal were found to be the most unstable.

View full abstract

We discuss the dynamics of 'carangiform' propulsion, which is a swimming mode suitable for high speed and high efficiency. The carangiform motion is modeled as the motion of a 2-joint bending mechanism composed of a streamlined body and a rectangular caudal fin. In order to analyze this model, the one-dimensional slender body theory and the two-dimensional linearized wing theory are used for the body and the caudal fin, respectively. Furthermore, the thrust reduction and energy loss due to the nonlinear fluid force acting on the circular body are considered in order to estimate the propulsive efficiency. In this paper, the analytical model is described and the analytical method in which the slender body theory and the linearized wing theory are combined is discussed. A simple linearized method and the fourth-order Runge-Kutta method are compared to estimate the effect of the nonlinear force on the dynamics of the body.

View full abstract

An object which flies hypersonic speed, such as an aeroassisted orbital transfer vehicle, is affected greatly by not only the bow shock but also the expansion around the shoulder and the recompression of separated flow. Though the experiments to analyze this flow have been performed by some groups recently, there are some difficulties in measuring the ow because of limitations of instrumentation. Both experimental and numerical approaches are necessary to analyze the phenomena. The efficient numerical code which has been developed by the authors for hypersonic thermochemical non-equilibrium flow is applied to simulate the problem. The calculated results of different accuracy's in space, the perfect gas and the experimental data are compared.

View full abstract

This paper describes numerical results of a thermal and chemical nonequilibrium plume of high-temperature air. The gas is expanded from an orifice into low-density stationary air as a free jet. The gas considered here is high-temperature air composed of N₂, O₂, N, O, NO, NO⁺and e^-, and translational-rotational and vibrational-electron temperatures are treated in a two-temperature model. In addition to this model, a six-temperature model using a multi-vibrational temperature model is adopted. The governing equations for computation are axisymmetric Navier-Stokes equations coupled with species vibrational energy, electron energy and species mass conservation equations. The equations have been numerically solved using the second-order upwind TVD scheme of the Harten-Yee type.

View full abstract

In the first report, 3-D linear analysis using a finite-span cascade model was carried out, and it was shown that (1) there are not only 0th-order modes in the spawns direction which correspond to conventional rotating cavitation previously predicted by 2-D linear analysis, but also higher order modes of rotating cavitation and (2) 0th-order modes are weakly influenced by the three-dimensionality of cavitation. The purpose of the present paper is to investigate the effects of the impeller rotation, the hub ratio, and the three-dimensionality of cavitation on rotating cavitation. Three-dimensional linear analysis using an annular cascade model is carried out based on semi-actuator disk theory. It is found that (1) there are many radial modes of rotating cavitation as previously shown using a finite-span cascade model, and that (2) a larger hub ratio has the effect of stabilizing rotating cavitation.

View full abstract

Concerning a flat-helical inducer with two blades, examinations on suction performance, cavitation development and internal flow conditions of the impeller were performed. As the suction pressure is reduced, the balanced cavity development on both blades is destroyed. Alternate blade cavitation, in which cavity development evolves on one blade while weakens on the other, can occur. When the alternate blade cavitation occurs, the theoretical and actual pump heads can decrease quickly. The following investigations were conducted to determine how this phenomenon develops.

View full abstract

The deformation, breakup, and displacement of liquid droplets due to air streams are described. Experiments are conducted using a horizontal air-suction wind tunnel. Uniform liquid droplets produced by a longitudinal vibration of a nozzle are used. Shattering processes of liquid droplets are observed using a stroboscope. The deformation of a liquid droplet varies with airstream velocity. The typical critical breakup pattern of a droplet in this experiment is a bag breakup. Time variation of droplet displacement are investigated in detail. Under a constant airstream velocity, the displacement of a droplet increases as its diameter decreases. When deformation of a droplet is increased to some level, the droplet is moved irregularly, The relationship between dimensionless time and dimensionless displacement is presented for droplets with small deformation.

View full abstract

During surface finishing of cold steel, many surface defects are caused by contact between a roller support and strip. An optimal surface finishing process should support the strip without roller contact in order to obtain high-quality steel products. We have studied stability improvement of the running strip derived from the side plates installed on the pad which has nozzles with a distance much shorter than that of a conventional setup. Our studies for practical use focus on reducing the flow rate for a floating strip by optimizing the shape of the side plates. This study clarifies that it is possible to optimize the shape of the side plates by coinciding with the profile of strip from side view. As a result, stable support with an economical flow rate can be achieved by optimizing the shape of the side plates.

View full abstract

The objective of present study is to show the aerodynamic effectivity of a horizontal axis wind turbine rotor blades with winglets by means of numerical analysis. The winglet used in this study is considered to be an inclined extension of the blade. For the numerical analysis a vortex lattice method with a free wake model was used because the model can be fitted to an arbitrary blade shape and needs no empirical parameter about wake geometry. The calculations were made on the flow held in the rotor wake and the rotor performance, and the results were compared between the rotors with and without winglets. In order to examine the structural effects, the flap bending moment was also compared, The results shows that small installation angle of winglets is found to cause a larger increase in the power coefficient and a smaller increase in the flap bending moment than radially extended rotor blades.

View full abstract

Experiments are carried out to measure the temperature at boiling incipience and observe the bubble formation behaviors on a very small film heater immersed in ethyl alcohol at a pressure in the range from 0.1 to 2.0 MPa and heated with a very high average rate of temperature rise up to 20×10⁶ K/s. The temperature at boiling incipience reaches a constant value at rates higher than about 5×10⁶ K/s and it agrees very well with the homogeneous nucleation temperature for a given pressure At such a high rate, concurrent generation of a large number of tiny bubbles of maximum number density 7.0×10¹⁰l/m² is observed, the diameter becomes smaller and the number density becomes larger with the increase of pressure, and the number density versus the surface temperature agrees with that predicted by the fluctuation nucleation theory. The bubble formation is concluded to be due to fluctuation nucleation.

View full abstract

Linear stability analysis of a thin vapor film in subcooled film boiling on a horizontal cylinder is reported. The effects of liquid inertia, vapor viscosity and compressibility, and heat transfer were taken into account. Theoretical predictions of the heat transfer coefficient at the neutral stability point were compared with experimental data at the minimum-heat-flux point that was obtained during rapid quenching of thin horizontal wires in water and ethanol. At high liquid subcooling, the experimental value was 60% of the theoretical prediction irrespective of the wire diameter and quenching liquid. This difference was considered to be due to the nonuniformity of the vapor film which was neglected in the theoretical analysis

View full abstract

An experiment was performed to simulate a water cooling panel system for decay heat removal from a high-temperature gas-cooled reactor (HTGR) in order to investigate the performance of decay heat removal and the temperature distribution of components of the system. The test apparatus was composed of a pressure vessel 1 m in diameter and 3 m in height, containing heaters with the maximum heating rate of 100 kW which simulated residual heat of the core and cooling panels surrounding the pressure vessel. The analytical code TANPACTST2 was applied to analyze the experimental data to investigate the validity of the analytical method and model proposed. Under the conditions of helium gas at the pressure of 0.73 MPa and temperature of 210°C in the pressure vessel, temperatures of the pressure vessel were well estimated within the errors of -29 to +37°C compared with the experimental data. The analyses indicated that the heat transfer-red to the cooling panel was 11.4% less than the experimental value and the heat transferred by thermal radiation was 74.4% of total heat input. It was also found that the lower head of the pressure vessel was effectively cooled by the natural convection through the flow channels at the upper and lower edges of skirt-type supports of the pressure vessel.

View full abstract

Thermal dispersion in convective flow in porous media has been numerically investigated using a two-dimensional periodic model of porous structure. A macroscopically uniform flow is assumed to pass through a collection of square rods placed regularly in an infinite space, where a temperature gradient is imposed perpendicularly to the flow direction. Due to the periodicity of the model, only one structural unit can be taken for a calculation domain to resolve an entire domain of porous medium, Continuity, Navier-Stokes and energy equations are solved numerically to describe the microscopic velocity and temperature fields at a pore scale. The numerical results thus obtained are integrated over a unit structure to evaluate the thermal dispersion and the molecular diffusion due to tortuosity. The resulting correlation for a high Peclet number range agrees well with available experimental data.

View full abstract

If initial temperature is assumed to be constant, a domain integral is not required for solution of unsteady heat conduction problems without heat generation using the boundary element method (BEM). However, under heat generation or nonuniform initial temperature distribution, the domain integral is necessary. In this report it is shown that the problem of unsteady heat conduction with heat generation and initial temperature can approximately be solved without use of the domain integral by an improved multiple-reciprocity boundary element method. In this method, the domain integral in each step is divided into point, line and area integrals in the case of a two-dimensional body.

View full abstract

In the present study, the heat Flux and temperature distribution of heating elements surrounding a furnace are estimated by using the singular value decomposition (SVD) method by giving the temperature and heat flux along the surface of a heated object in the furnace. The heat flux profile of the heating elements which satisfies the given condition is obtained by determining the rank of the singular-value matrix using L-curve method.

View full abstract

A device was developed to enhance heat transfer from heated horizontal plates. The device consists of six gutters with slits at the center for introduction of air, Heat transfer coefficients were measured by altering the device height above the heated surface. A height of about 10mm was found to give rise to higher heat transfer coefficients. The device augmented heat transfer rates by 1.2∼1.4 times in comparison with the heat transfer rates without the device.

View full abstract

The effects of the Pr number on the countergradient scalar transfer in strong stable stratification are examined by means of a three-dimensional direct numerical simulation (DNS). A DNS based on a finite difference method is applied to unsheared thermally stratified water (Pr=5.0) and air (Pr=0.7) mixing layer flows downstream of turbulence-generation grids. The results show that the countergradient scalar transfer becomes significant in a thermally stratified water ow with a high P/ number and that the scale of Fluid motion which contributes to the countergradient scalar transfer is quite different between thermally stratified water and air flows. In a thermally stratified water ow small-scale motions first contribute to the countergradient scalar transfer, whereas in a thermally stratified air ow the small-scale countergradient transfer does not occur. The results agree well with previous laboratory measurements.

View full abstract

Wave velocities and wave widths were determined using the wave-vein analysis for a wide range of air and water flow rates. Distinct features of huge waves and disturbance waves were easily recognized from the relations of uTave velocities to wave widths under limited flow conditions. Cluster analysis by K-mean algorithm, which deals with three parameters such as wave velocity, wave width and maximum liquid holdup, was applied in order to distinguish between huge waves and disturbance waves under the given experimental conditions. Individual waves distinguished by the cluster analysis correspond to those recognized from the relations between wave velocities and wave widths. Additional information on the interfacial profile detected by semi-supermultiple point electrode probes was utilized to distinguish huge waves from liquid slugs. Flow conditions for the appearance of liquid slugs, huge waves and disturbance waves are clarihed, and a new flow map is Presented.

View full abstract

The so-called generalized Clausius-Clapeyron equation (GCCE), i.e., the static equation of state for "solid-liquid-porous media" systems without liquid flow through the pores, has been experimentally examined by measuring the pore liquid pressure at the solid-liquid interface, in order to verify the accuracy of our method for measuring the unfrozen pore liquid pressure. Several series of experiments were carried out using water-saturated Ohya-tuff as a porous medium specimen which was cooled from the top down. In these experiments the solid-liquid interface was always fixed on the top surface of the specimen. The experimental results imply that the measurement method should be useful for determining the relative value of the unfrozen pore liquid pressure.

View full abstract

We presents an experimental study on effects of thermal conditions on the sink-mark generation of press-formed glass products. In experiments, the temperature of pressed glass was measured with variation of the initial temperature of the plunger and the pressing duration. The depth of the sink-mark was continuously investigated using laser displacement sensors. Obtained results are summarized as follows : (1) The temperature of the glass surface decreases dramatically during pressing. After pressing, it rises again because of reheating. However at the center of the glass, reheating is not observed. (2) The sink-mark grows up only when the temperature of the glass surface is above 700°C. In order to suppress the generation of the sink-mark, it is important to keep the temperature of the glass surface below 700°C or to shorten the period during which it is above 700°C.

View full abstract

In this paper, the temperature of the arc plasma and the arc extinguishing time in the electrical discharge machining (EDM) process are calculated based on the magnetohydrodynamics (MHD)approximation. The temperature of the arc plasma determines the heat Flux at the surface of a workpiece which serves as the main heat source in melting and vaporizing of the workpiece. The arc extinguishing time, which must be shorter than the pulse interval, is closely related to the stabiliy of the machining process and hence the machining rate of the process.In cases for which local thermodynamic equilibrium can be assumed, the plasma is described as continuous matter, and the MHD equations are solved by the finite difference method to detemine the distributions of current density, Fluid velocity and temperature. The calculated results show that almost all of the discharge energy is equally distributed into anode and cathode, and that the energy transfer is mostly determined by the heat conduction but not affected by convection or radiation. Further more it is also found that the plasma is extinguished within a few microseconds after the current supply is ceased for each discharge pulse.

View full abstract

The droplet combustion of n-heptane, benzene and ethyl alcohol was investigated in acoustic fields. The frequencies of the acoustic waves used were around the characteristic frequencies of droplet combustion ; the residence frequency and the diffusion frequency. Under low frequency conditions (70-200 Hz), the experimental results can be interpreted by the mechanism of quasi-steady state combustion, which includes all types of droplet combustion ; the envelope name, wake flame and flame extinction. Under high frequency conditions (300-800 Hz), the combustion is possibly influenced by a diffusional mechanism which is induced by acoustic waves of such high frequency.igh frequency acoustic waves increase the flame luminosity, especially for benzene and n-heptane droplets. This suggests that the high frequency acoustic wave enhances radiant heat transfer from the flame zone to the droplet, and thus the burning rate constant increases.

View full abstract

Using an axisymmetric swirl-type burner with eight inlet slits, the effects of rotation on the structure and characteristics of tubular flames have been investigated for lean methane/air mixtures.Results show that with increasing intensity of rotation, (1) the flame diameter increases and the thickness of the luminous zone decreases, (2) the fuel concentration at extinction decreases, i. e., the stable region expands, (3) the radial temperature distribution becomes an M-shaped profile, and the maximum temperature T_f increases while the local minimum temperature T_c at the center decreases, and (4) the values of T_f and T_c vary in a complicated manner in the axial direction. These results are almost consistent with those of our previous experiment, and hence, it is confirmed that the flame structure and the flame characteristics are affected by rotation. As for these phenomena, several mechanisms such as pressure diffusion, radiative heat loss, viscous energy dissipation, and energy recirculation including diffusion of radical species have been proposed. The present experimental results provide the first, basic information on this practically important phenomenon, i. e., the extension of flammable range by rotation.

View full abstract

By using a glass tube into which a combustible mixture is tangentially injected from one end, the behavior of flames in vortex flow, especially the flame shapes and the flame forming range in fuel concentrations, has been experimentally investigated. To improve the controllability of combustion, an iris diaphragm is inserted in the tube. The results show that a tubular flame with a constant diameter is formed near the lean flammability limit, and that the combustion range in the concentration extends with an increase of the inner diameter in the glass tube. When the iris diaphragm is inserted, the flame changes into spindle- or bugle-shaped, and the length of the tubular flame is shortened with its extinction characteristics unchanged.

View full abstract

Recently, clarifying the turbulent combustion mechanism of hydrogen mixtures and the local burning mechanism in the turbulent flame has become necessary to establish the lean-burn technique in internal combustion engines in order to protect the environment and save energy. In previous work, experimental results indicated that local burning velocity in turbulent combustion of hydrocarbon mixtures changed from the original laminar burning velocity, and this mechanism was qualitatively explained in terms of the effects of the Lewis number and preferential diffusion. The purpose of this paper is to estimate quantitatively the change in the local burning velocity of the turbulent combustion of hydrogen mixtures, caused by preferential diffusion. In addition, it is suggested that the quenching mechanism can possibly be explained regardless of the kind of fuel, when the estimated local burning velocity is used as a reference instead of the laminar burning velocity.

View full abstract

The high-pressure injection system has been installed in small high-speed engines in recent years, in order to improve the atomization process of the spray and to reduce the exhausted particulates from engines. However, cavitation erosion appears to occur in this system. In this study, the formation process of cavitation bubbles was investigated via measurement of the pressure history and that of the power spectrum density, PSD, of the high-frequency component of the pressure in the injection pipe. It is revealed that PSD at the early secondary stage pressure is govemed by the differential pressure, Δp. Then mechanism of the generation of erosion source can thus be explained.It is clear that the characteristic value, f(Δp), controls the erosion level.

View full abstract

The effects of the combustion chamber geometry on flame behavior were investigated in an optically accessible DI diesel engine. The flame behaviors in three different chamber geometries including the production version were observed through both windows installed in the piston crown and the head using a high-speed camera and a pseudo-endoscope system, respectively. Comparing the velocity distribution in the flame inside each chamber measured by applying image processing, it is found that (1) the chamber geometry has a significant effect on the flame velocity, that is, the velocity in the reentrant type chamber is larger than in the dish type chamber during the expansion period. Also, in the observation of the spreading of the flames over the squish area taken by the endoscope, the image shows that (2) the flame distribution inside and outside the chamber is considerably affected by the chamber geometry, and the reentrant type chamber prevented the flame from spreading over its squish area. Correlating the flame behaviors with exhaust smoke, it is found that (3) for further smoke reduction it is important to achieve an active combustion during the period of diffusion combustion and desirable distribution of fuel and that of flame in the chamber simultaneously.

View full abstract

The flow induced noise from an expansion cavity-type muffler increases abruptly for some sizes of mufflers and flow velocities. This phenomenon is strongly related to the resonance of the tail pipe and the cavity of the muffler ; therefore, it is called the predominance of resonance″When the cavity length is increased, the measured resonance frequency shows periodic jumps. The cause of the predominance of resonance is assumed to be the generation of vortex rings with a particular frequency in the cavity. The generation of vortex rings is verified by analyzing fluctuating velocities in the cavity when the predominance of resonance occurs. The phenomenon is induced not only by the tail pipe resonance, but also by the cavity resonance. It is suggested that there exists the feedback phenomenon from the sound to the flow and it is strongly related to the sound pressure level in the cavity.

View full abstract