This paper proposes a new outer boundary condition called the "periodic-shell boundary condition" which is used for molecular dynamics simulations on a flow around a body. The present boundary condition is devised to diminish the influences of the limitedness of simulation regions by making molecules inside regions interact with imaginary molecules outside which are replicas of actual molecules existing near boundary surfaces. In using this condition we need no information such as temperature, densities, and averaged velocities at boundaries, which is in contrast with the case of ordinal uniform-flow conditions where such information is indispensable. To clarify the availability of the present boundary condition we have carried out simulations of a rarefied gas flow past a circular cylinder. The results of simulations show that the periodic-shell boundary condition significantly diminishes the distortions of the flow fields compared with conventional conditions such as uniform-flow and free-outflow conditions. Such improvements enable us to use smaller simulation regions and lead to a dramatic decrease in computation times.

抄録全体を表示

Effects of the stress overshoot on the transmitting force generated in some cases of squeezing flows are analytically investigated using a constitutive equation. A nondimensional factor (O_s) that corresponds to the magnitude of the stress overshoot, and an apparent relaxation time (λ_O) are defined by extending the definition in the simple shear flow. The magnitude and the peak time of overshoot in the force induced in normal or reverse squeezing flow are respectively correlated with the O_s and λ_O. In the case of successive squeezing flows consisting of normal squeezing in the first half and reverse squeezing in the second half and vice versa, the force generated in the second squeezing is only dependent on the normalized time t/λ_O. When the two surfaces are subjected to sinusoidal oscillations in the region of high Deborah number, the amplitude of the force and phase lag become large with an increase in O_s, and separating forces time-averaged during the sinusoidal oscillation decrease with an increase in O_s, especially in the extreme case, the two surfaces being forced to approach with sinusoidal oscillation.

抄録全体を表示

For a translational symmetric case, the change rate of an elastic stress is approximated by the spin term alone. The contribution of the derived elastic stress to the momentum equation is decoupled from a velocity component in each coordinate. Then, the body force caused by the elastic stress is approximated by a linear source term within an iteration loop. The following cases are demonstrated : (1) a square duct, (2) a lid driven swirl flow in a cylinder, (3) a free surface movement of coaxial rotating cylinders.

抄録全体を表示

The interesting phenomenon of nonperiodical flow rate switching between a high level and a low level was found under certain flow conditions in a loop with a cross pipe. It is caused by generation and collapse of swirling flow through opposite laterals of the cross pipe. In this paper, control of this phenomenon is experimentally investigated by using air-flow test equipment at Re≒2×10⁵. It is clear that the phenomenon depends on cross pipe structure, and a slight difference in structure brings the flow rate to a stable high level or a stable low level. As the turbulence intensity in the flow is high, the switching does not occur because swirling flow becomes difficult to form. The effects of the control flow injection for stabilizing such a phenomenon are also confirmed.

抄録全体を表示

The flow in a pipe downstream from a 90°branch was investigated using air-flow test equipment at Re≒1.5×10⁵. The tested pipe system was composed of a tee and a long elbow. Mean and fluctuating velocities were measured by hot wire probes. The test results clarified the flow pattern and the effective factors. The flow pattern downstream from the elbow depends on the elbow offset angles as the distance between the tee and elbow becomes short. The asymmetry of the profile of mean velocity U^^^-/V_m at the branch pipe inlet section increases as the distributing discharge ratio decreases. The mean velocity profile downstream from the elbow is almost independent of the distributing discharge ratio at the elbow offset angles of 0°and 90°. At the elbow offset angles of 90°and 180°, the turbulence intensity √(u^^^-2)//V_m downstream from the elbow increases as the distributing discharge ratio decreases.

抄録全体を表示

The fluctuating velocity and pressure downstream of a pipe orifice were experimentally investigated at Re_d≒1.7×10⁵. For the standard sharp-edged orifices, 0.4∼0.8 in diameter ratio, the inner velocity fluctuations and the wall pressure fluctuations were measured by means of hot wire probes and pressure transducers. It is clear that the turbulence intensity on the orifice jet periphery is about twice as high as that in the flow near the wall surface at a short distance from the orifice. More high-frequency components are included in the spectrum of the fluctuating wall pressures in the downstream reverse-flow region than in the irreverse-flow region further downstream. The reverse-flow region was detected by the flow direction detector and was visualized by a spark tracing method.

抄録全体を表示

Effects of swirling flow on the boundary layer and performance in a divergent channel, whose cross section changes from annular to rectangular, were investigated experimentally. At the cross section with flat side walls and concave corners, the secondary flow on the flat wall somewhat promotes the reverse flow, namely, separation. On the contrary, the flow on the concave wall suppresses the separation. In addition, the swirling component markedly affects the hydraulic loss in the annular passage, but scarecely affects the loss in the downstream passage. In the downstream passage, however, the pressure recovery cannot be disregarded as the swirling component becomes large.

抄録全体を表示

Numerical calculations by a finite-difference method were made for a steady three-dimensional flow inside a cylindrical housing when a disk having radial slits and its housing were rotated. Three types of flows for the following conditions were examined at the same relative angular velocity between them: (1) The slitted disk rotated and the housing was stationary. (2) The slitted disk and the housing rotated in opposite directions. (3) The slitted disk was stationary and the housing rotated. With the increase in the relative angular velocity, the difference in flow inside the housing due to the rotating conditions becomes pronounced. For the counter-rotating condition, the radial flow induced inside the housing was suppressed and the tangential velocity distribution tended to be linear, showing the smallest torque transmitted between the disk and the housing.

抄録全体を表示

To obtain a flow development process which is important for turbulent generation, we directly simulated the three-dimensional circular Couette flows with high Reynolds number using the third-order upwind finite difference scheme. In the experiments we observed flow states by a visualization technique. The results of the simulations are in fairly good agreement with the experiments. At the very early stage, the cells and vortices appear near the inner cylinder. The distribution of the centrifugal force caused by the velocity distribution makes the vortices grow to the outer cylinder. When the vortices have reached to the outer cylinder, they begin to interact with each other. At that moment, the flows become turbulent or chaotic, and soon after that the turbulent or chaotic flows disappear and new cells, with a height two or three times higher than the former cells, appear in the whole entire region of the flow. But in the case of higher Reynolds number (Re=1500), turbulent flows still exist in a cell. The flows become steady states with velocity fluctuations.

抄録全体を表示

A new adaptive control method for fuzzy reasoning is proposed, in which the fuzzy controller works with a fluctuation generator for adaptation. The fluctuation generator which produces Gaussian noise is combined with antecedent parts of a fuzzy controller to fluctuate membership functions in each processing time. The learning parts of the system update the membership functions if the output from the controller is more suitable for faster convergence of the Poisson equation. Two-dimensional numerical simulation of the flow around a moving plate is conducted with the fuzzy controller using this new technique. The control variables, i. e., input to the controller, are the interval of the time step and the over-relaxation parameter of the Poisson equation. As a results it is found that this method is adaptive and time effective. The new adaptive method reduced the computational time to 80% compared that for a fuzzy controller without learning.

抄録全体を表示

This paper analyzes experimentally the influence of the geometrical configuration on the aerodynamic drag using a bluff body such as a box-type (fastback) vehicle. For a box-type vehicle, the aerodynamic drag depends on the angle of the front window and the rear window. Therefore, these angles were selected as the configuration parameters. By changing the window angle and the ratio of the model length to the bending window position, the aerodynamic drag characteristics were investigated. As a result, it is found that, the aerodynamic drag decreases at the front window angle of 20∼30°and increases drastically at the rear window angle of 30°when the ratio of the model length to the bending rear window position is more than 0.2. On the other hand, it is also found that the wake of the box-type vehicle consists of the arch vortex on the rear window and the trailing vortices behind the rear window, by means of the flow visualization and the velocity vector distribution. Based on understanding the wake structure, it is clarified that the aerodynamic drag which increases drastically at the rear configuration is caused by the arch vortex on the rear window.

抄録全体を表示

This paper analyzes experimentally the influence of the geometrical configuration on the aerodynamic drag using a bluff body such as a notchback vehicle. The effective parameters which reduce the aerodynamic drag were clarified by parametric study of the geometrical configuration. These effective parameters were defined as the apparent angle. For the notchback, it is found that the critical geometries which increase the aerodynamic drag, such as the critical geometry for the box type, exist when using these apparent angles. On the other hand, it is also found that the wake of these bluff bodies consists of the arch vortex and the trailing vortices, by means of flow visualization and velocity vector distributions. As a result, based on understanding of the wake structure, it is clarified that the aerodynamic drag increases at a certain rear apparent angle which is caused by the arch vortex. The apparent angle is proposed for the effective configuration parameter.

抄録全体を表示

Effects of a favorable pressure gradient on the average time between bursts were studied by two different methods : the modified U-level method, and a proposed detection scheme that applies the U-level method to ejections and sweeps. Experiments were carried out in a boundary layer along a flat wall of two-dimensional asymmetric contraction. The results of experiments show that the average time between bursts scaled with inner variables increases with a decrement of the pressure gradient parameter Δ_p, and the ratio of the Eulerian micro-time scale to the average time between bursts gives a constant value of 0.1 at Y⁺=15. The proposed method also indicates that bursts still remain throughout the reverse transition process from a turbulent boundary layer to a laminar one.

抄録全体を表示

Cycle-to-cycle variations in the time period of a Doppler signal cycle on which higher frequency noise is superimposed give rise to velocity turbulence errors in a commercially available counter processor. The errors simulated and measured by the dual-beam LDV are found to increase in lower turbulent flows at lower count-cycle numbers and higher shift frequencies. The measured turbulence was corrected to some extent by the simulation described in this study.

抄録全体を表示

It is intended in this report to discuss the structure of near-wall turbulence by drawing pictures of flow fields of specified concerns using three-dimensional computer graphics. The flow taken here as the material is a two-dimensional channel flow obtained by a large-scale computer simulation. The discussion is focused on the streaks which are observed in all kinds of near-wall turbulence as a marked feature. It turned out that impinging, which is the key phenomenon in high-speed streaks, plays a key role in the near-wall turbulence as well as the lift-up and break-up of low-speed streaks. Most of the important events take place in these sparsely distributed spots. Turbulent shear stress is found only sparsely where it is produced, and the whole process of shear stress mostly restricts itself to within a small confined area. It was confirmed that although the pressure diffusion and the redistribution of shear stress are important when considered separately, they almost cancel one another and as a result the splitting of the velocity-pressure gradient term into these two terms is not beneficial in turbulence modelling.

抄録全体を表示

This paper describes a numerical analysis of flow in an induction-type electromagnetic pump. Laminar and turblent flows in a conducting fluid between two coaxial cylindrical walls to which were applied an azimuthally traveling sinusoidal magnetic field are investigated numerically. In order to consider the demagnetizing effect quantitatively, induced magnetic fields due to induced currents are calculated based on the Biot-Savart's law. Several characteristic curves are obtained from numerical results. They are characterized by the Hartmann number, magnetic Reynolds number, azimuthal pressure gradient, pole pitch and curvature of the channel. Influence of these parameters on the flow are discussed in detail.

抄録全体を表示

In our previous study, the SOLA method, which is a numerical solution method for incompressible viscous flows, was modified to take into account the boundary conditions during the iteration process. This modification increased the computational efficiency. In the present study, the modified SOLA method was extended to be applicable to an adiabatic two-fluid model for two-phase flows. In this method, the pressure correction equation, which is defined by the volume balance between two phases, is iteratively solved in terms of not only pressure but also velocities and volume fractions of both phases. This makes the algorithm simple and therefore makes it easy to realiz in a program. This advantage implies possibilities for future improvement and upgrade.

抄録全体を表示

In the first report of the present study, a numerical solution method for an adiabatic two-fluid model was developed based on the modified SOLA (Solution Algorithm) method. In this second report, a numerical treatment is discussed for the case when a phase occupying a computational cell is changed with time from single phase to two-phase or vice versa, which is defined as a phase transition. In addition, the numerical method with a treatment for the phase transition is applied to a numerical benchmark problem. The problem is to calculate two-dimensional sedimentation due to gravitational force. The problem is suitable to discuss the numerical method for an adiabatic two-fluid model including the treatment for the phase transition. The calculated results are qualitatively well understood, and therefore, it is confirmed that the present method is applicable to the transient two-phase flow analyses including the phase transition.

抄録全体を表示

In our previous paper, the modified Galerkin method (MGM) was proposed for the one-dimensional convection-diffusion equation as one of the most efficient methods. In this paper, the MGM concept has been analytically extended to the two-dimensional convection-diffusion equation. The triangular linear and the rectangular linear elements were investigated by the error analysis technique combined with an analytical solution. Then, six correction coefficients were obtained for MGM and the Galerkin method (GL). At the same time, two finite difference methods were investigated using the error analysis technique. Then, all of these numerical methods were evaluated from the viewpoints of phase error and dissipation error. The MGM was proven to be the most efficient numerical method among these five methods. Through numerical experiments by MGM and GL, we confirmed the effectiveness of MGM and the general correspondence between the error analysis and the numerical simulation.

抄録全体を表示

The Reynolds-stress budgets in channel flow are calculated using a second-moment closure that was previously validated in various turbulent flows. The results are compared with direct simulation (DNS) data. The redistribution model is considered to correspond to the velocity pressure-gradient term Π_ij, in view of the fact that there exists a spit of Π_ij which yields a redistributive term close to Π_ij itself. The comparison shows that, in general, the redistribution model behaves properly, indicating a reason for the high capability of the present model. However, the model produces a suppressed turbulent diffusion and an exaggerated dissipation as compared to those of the DNS data. When the redistribution is split into the rapid term and the slow term, the individual model is unsatisfactory; in this respect, the present model should be improved for wider applicability.

抄録全体を表示

Flows around a flat plate during a pitching motion in a uniform crossflow are studied numerically. The flat plate makes a cycle of pitching motion about its mid-chord position at a constant angular velocity. The body-fitted grid generation with moving boundaries is used to obtain the numerical solution of the incompressible Navier-Stokes equations. The time-dependent curvilinear coordinate system which coincides with a contour of a moving boundary is transformed into a fixed rectangular coordinate system in a calculational plane. The changes in the streamlines, C_D and C_L, during the pitching motion are obtained. The results of present analysis show good agreement with those of previous experiments.

抄録全体を表示

A method is presented for the aerodynamic characteristics of the cascade of porous sails with the same shape. In this analysis the authors apply a series of Jacobi's polynomials to express the pressure distribution and chordwise shape. As the result of this numerical analysis, the effects of porosity and relaxation on the coefficient of lift and the position of the center of pressure are shown in diagrams as functions of solidity and stagger angle. With an increase in the porosity, the critical relaxation varies greatly with the stagger angle. The maximum eigenvalues of tension coefficient depend on the stagger angle and solidity, but when the porosity is large they do not vary with the stagger angle.

抄録全体を表示

System concepts have been dealt with by applying the power augmented ram wing for large ground effect craft. Trend analysis and parametric studies on the effects of its dynamic stabilities have been conducted to evaluate some kinds of tandem wing crafts. The necessary ranges and tendencies of the aerodynamic characteristics were determined for each of the systems of enhanced lift wings. By controlling the appropriate tail unit we achieved a simulation of the flare maneuvers.

抄録全体を表示

This third report deals with friction losses in adjustable blade runners for axial-flow hydroturbines. A method is proposed to extract the friction coefficient for the runner loss from the measured performance characteristics of bulb turbines. As a result of the loss analysis for two model bulb turbines based on the above method under a wide range of runner vane setting angles and wicket gate openings, an algorithm is presented for computing the friction loss in a runner using the well-established friction coefficient for a flat plate with the same Reynolds number and surface roughness as the blade at the root mean square radius.

抄録全体を表示

This study is concerned with a kind of oil-free mechanical vacuum pump with spiral grooved rotors. To improve the performance of the pumps at pressures above 1000Pa, sleeves which have triangular or rectangular grooves were tested instead of the smooth ones ordinarily used. The pressure difference between suction and discharge sides was measured for various pairs of rotors and sleeves. The measured values were normalized and plotted against the Reynolds number defined as Re=4ρUm/μ(U : peripheral speed, m : hydraulic mean groove depth, ρ : density, μ : viscosity). Grooves on sleeves serve as paths for leakage in the laminar regime (Re<1000), but they activate the generation and conveyance of vortices when the flow becomes turbulent, and thus improve the performance. The improvement becomes marked as the groove pitch on the sleeve decreases and as the helix angle of the groove on the sleeve increases.

抄録全体を表示

Low cycle pressure oscillations often occur in cavitating pumps operated in the range of low flow rates. Several types of oscillations have been found to exist, corresponding to flow rate and NPSH. The instability phenomena occurring in a centrifugal pump system at extremely low flow rate conditions were investigated by measuring pressure fluctuations at the suction pipe and observing the cavitation behavior there with a high-speed video system. The cavitation compliance and mass flow gain factor were calculated with the aid of experimental data. The transfer matrix method was used to analyze the system stability, and the results were compared with the experimental ones.

抄録全体を表示

A wind tunnel simulation of the convective planetary boundary layer was performed. The temperature of a flat wind tunnel floor was increased almost uniformly up to 80°C to form an unstable temperature boundary layer. Velocity, temperature and kinematic heat flux on the wind tunnel floor were measured using LDV, resistance thermometer and thin-film-type heat flow meters. The measured turbulent intensity showed good agreement with those obtained from field experiments and a water tank experiment when normalized by the convective velocity. The universal functions for velocity and temperature profiles were applied and empirical constants were optimized. The length scales of turbulent motion were estimated by means of the universal functions.

抄録全体を表示

A semiconductor laser with a wide active stripe was utilized as the power source of a laser 2-focus velocimeter. The focus was expanded to about 0.2mm wide to form the two parallel focal planes. The backscattered light from seeding particles was detected by photodiodes, and the time-of-flight between two foci was counted by means of a digital clock. The flow velocity was measured in a circular jet stream, and the results were compared with those measured simultaneously by Pitot tube. After checking the accuracy of measurements, the blade-to-blade distribution of velocity and flow angle were measured at the exit of a centrifugal impeller to confirm the applicability of the semiconductor-L2F with band focus.

抄録全体を表示

Functionally gradient materials (FGM) with distributed propertied have attracted special interest as advanced heat-shielding/structural materials in future space applications. The objective of the present study is to evaluate the validity and/or usefulness of the solution of the temperature rise in FGM, the front surface of which is subjected to pulse^- or stepwise heating. Applicability of the approximate solution has been investigated by comparing it with an exact analytical solution for the FGM in which distributed thermophysical properties have certain profiles. When the FGM is composed of conventional solid materials, the appropriateness of the approximate solution for the FGM has been shown, inasmuch as the temperature response near the rear surface is concerned. The approximate solution is also compared with the solution for the multilayered material. Because of its simplicity and fair degree of agreement, the approximate solution is anticipated to be used for not only qualitative but also quantitative prediction of the temperature rise near the rear surface of the FGM, from the engineering point of view.

抄録全体を表示

This paper is concerned with the absorption enhancement in the process of ammonia vapor absorption into pure water and aqueous solutions of LiBr with 50% by mass. The Marangonian convection was generated violently with addition of n-octanol in the concentration range of 0.05-1% by mass, and high absorption enhancement was obtained in a binary system as well as in a ternary one. Furthermore, the effect of various surface active agents was studied. Straight-chain hydrocarbons with six to ten atoms were selected as surfactants. The mass input by ammonia absorption increased to almost 200% for a binary system with addition of n-heptyl alcohol (C₇H₁₅OH), and to almost 300% for a ternary one with n-hexyl alcohol (C₆H₁₃OH), compared with the case without any additives.

抄録全体を表示

We have developed a new apparatus to measure the thermal diffusivity of liquids by the forced Rayleigh scattering method using a CO₂ laser. In the case of employing the infrared wavelength laser as a heating source, it is less necessary to add dye substances in samples than with the previously used Ar⁺ laser. The purpose of the present study is to develop a practical forced Rayleigh scattering apparatus applicable to a wide range of substances and conditions. In the present study, we have invented a free-surface-heating technique which does not require any cell windows. In order to demonstrate the applicability of the present apparatus without adding dye substances, we have preliminarily measured the thermal diffusivity of molten KNO₃ and a molten carbonate mixture (Li₂CO₃-Na₂CO₃-K₂CO₃=43.5-31.5-25.0mol%) in the temperature range up to about 1000K.

抄録全体を表示

This paper is the first report of an experimental study on a new highly efficient cooling method for hot strip mill work-roll cooling. The boiling heat transfer characteristics of a heated horizontal cylinder revolving in a pool or in annular narrow spaces were investigated experimentally. The obtained data were compared with the data on a stationary cylinder with the same narrow space conditions and the data on a revolving cylinder without narrow space. Experimental results show that the cylinder revolving in the small narrow space receives a much greater heat transfer promotion than that in the case without narrow space conditions, and the boiling heat transfer characteristics of the cylinder revolving in the small narrow space are not greatly affected by revolving speed and heat flux, but are mainly governed by the gap dimensions of the narrow space.

抄録全体を表示

Experimental investigations have been carried out on a natural convection over an horizontal, heated plate facing upward. The plate was heated with a uniform heat flux and was placed in a large water tank. The flow over the plate and the temperature distribution of the test plate were visualized with dye and liquid crystal thermometry, respectively. The local heat transfer coefficients were also measured in the direction from the edge to the center of the plate in the range of the local, modified Rayleigh numbers from 10₂ to 10¹¹. The results revealed that the transitional region appear in between the laminar boundary layer region near the plate edge and the irregular, disturbed flow region near the central part of the plate. The result of the local heat transfer coefficients also confirmed the existence of the above transitional region.

抄録全体を表示

Radiative heating and cooling of solid particles are studied numerically by considering the vibrational motions of solid particles and their relationships to the radiative absorption and emission with the molecular dynamics method. The radiative absorption and emission are assumed as a square-shaped function of the vibrational frequency and the Lennard-Jones potential is used for estimating the forces between particles. The change in the solid temperature is proportional to the difference between the products of the vibrational probability and the energy intensity of the absorption state and that of the emission state.

抄録全体を表示

Heat transfer and pressure drop performances are experimentally studied for the flow through wire-fin cores which consist of closely packed thin wire arrays. A modified single-blow transient test method is used to evaluate the performance. The range of the Reynolds number tested is 20-200. Results of the heat transfer and pressure drop are compared with those of the conventional compact heat exchanger surfaces. It is shown that the wire-fin surfaces yield markedly higher heat transfer performance. The flow visualization methods of dye injection and hydrogen-bubble generation are employed to investigate the entire flow field in the cores. Steady laminar flow, vortex shedding, oscillating flow, and turbulent flow are observed in sequence along the flow in the cores.

抄録全体を表示

Until today, size reduction of the bearing housing in induction motors has been made based mainly on past experiences. However, size reduction based on these experiences sometimes has induced excessive temperature in the bearing. Therefore, in order to better design the housing, it is necessary to analyze heat flow and temperature distribution around the bearing. One of the factors which largely affects the accuracy of the results of these analyses is the heat transfer coefficient from bearing to lubricant. However, practical papers with respect to its heat transfer have yet to be published. In this paper, the following experimental equation of its heat transfer in the case of oil bath lubrication was obtained through experimental study intended for practical use. When the Prandtl number and Reynolds number are expressed by Pr and Re, the Nusselt number Nu can be expressed as follows : Nu=0.27 Pr^1/3Re^0.53.

抄録全体を表示

This paper describes extensive numerical studies on a gas diffusion flame gasturbine combustor based on an interactive steering supercomputing simulation system. The system formed by CRAY Y-MP, IRIS4D and Ultra-Net could visualize the analysis process and alter the numerical condition during computation. The result of computation showed a complicated unsteady 3D flow caused by wall jet and temperature increase as a result of combustion reaction. The present study leads to the conclusion that the interactive steering simulation system could be applied to the complicated thermo-fluid analysis / such as of a gasturbine combustor.

抄録全体を表示

With the use of aquifer thermal energy storage for heating and cooling, one of the main aims of this study was to establish a method for keeping energy consumption levels low while heating or cooling utilizing a large quantity of stored water. Therefore we have proposed continuous cooling and heating requirements whereby the difference between the ambient air temperature and the water temperature is minimized. This paper reports the subjective ratings of the room thermal sensations by a large number of Japanese subjects while the thermal condition was monitored. Radiation, thermal convection and evaporation were analyzed using a cylindrical model to approximate the human body. The results indicated that thermal comfort was obtained under the following conditions : (1) the ratio of the radiation heat transfer coefficient to that of convection was 2.34 for cooling and 2.19 for heating, and (2) the ratio of heat loss by evaporation to the sum of radiation and convection was 0.052 for cooling and 0.050 for heating. Finally, a new thermal design process based on thermal comfort was proposed.

抄録全体を表示

A new purification method for a water supply system utilizing an aquifer was proposed, i.e., artificial groundwater charging through a biofilter such as a ferric-colloid and microorganism layer. The main objective of this paper is the observation and investigation of filter effects for iron bacteria and light synthetic microorganisms. Experimental investigation led to the following : (1) the iron bacteria which acts as a catalyst in the formation of the ferric-colloidal substance in the aquifer is gallionella, whose stalk works as a kernel of a chemical reaction, (2) The filter effect through the ferric colloid was nearly 100% when the permeability of the filter decreased to less than 10% of the initial one, making use of carbon ink of particle size less than 0.1 micron in diameter, (3) The purifying effect of organic matter due to biofilter reactions such as ulothrix aequalis, navicula falaisiensis and formidium fenue was confirmed on the basis of decreasing COD value under lighting by halogen lamp, (4) A new numerical simulation modeling for the purifying effect of the biofilter was proposed, and was confirmed by experiment.

抄録全体を表示

In order to simulate the behavior of heat pumps, theoretical models of heat exchangers and compressors are developed. The distribution of refrigerant flow rate between two paths in the heat exchangers can be satisfactorily simulated. Unbalanced distribution of the refrigerant flow rate among multiple paths frequently occurs in the condenser in contrast with in the evaporator. This different feature is ascribed to the different orders of magnitude of frictional and gravitational components of pressure drop in the condenser and the evaporator. The new compressor model requires neither the polytropic exponent nor the volumetric efficiency. The present models can estimate the mass of refrigerant existing in the heat exchangers and the compressor.

抄録全体を表示

As a new type of louvre blind cryopump, a parallel-louvre-blind cryopump with a rectangular chamber with opening distance B, louvre width C and depth D, is proposed. The influence of louvre size ratio R_c(=B/C) and pump depth size ratio R_d(=D/B) on the transmission probability P_bc is calculated by the Monte Carlo simulation and measured in model experiments. The following results are obtained. (1) The transmission probability of a cryopump designed with a dimensional ratio of R_d=1.28, is calculated as 0.7 at ratios of R_C=0.7∼0.9, which means that this type of cryopump can theoretically atttain a high pumping speed of more than 300 m³·s^-1/m² per opening area. (2) The transmission probability of the cryopump strongly depends on R_d and decreases rapidly for R_d less than 1.0 in the case of R_c=0.92.

抄録全体を表示

The spontaneous ignition of gaseous fuel injected perpendicular to a steady hot air stream was studied at atmospheric pressure. Ethylene and hydrogen were used as fuel. Ignition delay was measured by detecting ultraviolet radiation of OH radicals. Ignition delay curves in the Arrhenius coordinate indicated straight lines of large gradient in the low-temperature region, but their gradient became gradually smaller with the increase of the temperature. In order to examine the cause of this bending of the ignition delay line, changes were made in the physical processes of ignition phenomena such as mixing or diffusion, so that the effect on ignition delay could be experimentally investigated. As a result, it was revealed that the ignition delay behavior indicated by a straight line in the low-temperature region is controlled by chemical processes, and the behavior indicated by a small-gradient line in the high-temperature region is controlled by physical processes.

抄録全体を表示

For sprays of practical interest, the region where vigorous vaporization occurs is confined to a thin layer on their outer edges. The cubical array of droplets occupying a semi-infinite space provides a microscopic prototype to investigate this vaporization layer structure, upon which interactive droplet vaporization/combustion characteristics are examined to derive rational boundary conditions to be imposed at the spray edge when the spray is regarded as a continuum medium on large scales. It is found that not only the local droplet number density but also the gas-to-liquid density ratio plays an important role in excitation of prevailing vaporization/combustion modes, leading to their criteria in terms of the droplet separation distance as a function of the gas-to-liquid density ratio.

抄録全体を表示

Current transport in the boundary layer of combustion gas plasma flow on a cold electrode changes from diffuse mode to micro-arc (about 1A/Arc) and then to high-current big-arc (about 100A/Arc) mode with the increase in mean current density. This study intends to clarify the big-arc formation mechanism and to extract the pure effects of basic factors, particularly electrode temperature and bulk plasma temperature, based on the statistical analysis at the arc transition state. While the critical current density for the transition from diffuse to micro-arc mode largely increases with the increase in electrode temperature, its effect on the transition from micro-arc to big-arc mode is not so significant. It is the bulk plasma temperature that changes the critical micro-arc current for the transition to big-arc mode. Micro-arc currents fluctuate and some micro-arcs with arc current beyond the critical value will grow up to big-arcs breaking through the thermal boundary layer. Most of the micro-arcs with arc current below the critical value remain as micro-arcs adjacent to the electrode wall.

抄録全体を表示

The performance characteristics of pulsed jet ignition in a turbulent flow with swirl are confirmed by tests carried out in a disk-shaped combustion vessel with lean methane-air mixtures, in comparison to those of plasma jet ignition and standard spark plug ignition. The combustion process was monitored by pressure measurements and high-speed schlieren photography. It was found that the effect of cavity volume on combustion enhancement in a swirl flow was smaller than that in a quiescent mixture, and it can be considered that there was no effect of orifice diameter. The lean flammability limit was most extended by pulsed jet ignition in both quiescent and swirl flow.

抄録全体を表示

Carburetor engines with a mechanical governor sometimes exhibit a self-excited oscillation in rotating speed, called hunting. In this paper, two different mechanisms of hunting, mechanical hunting and lean hunting, are discussed. Through a simple system of equations, the mechanism of mechanical hunting is explained, and it is clarified that the dynamic characteristic of the mechanical governor plays a significant role in the deterioration of system stability. On the other hand, a new mathematical formulation is developed for the mechanism of lean hunting where a differential equation for time-varying air-fuel ratio is included, and the effect of the dynamic characteristic of the governor is excluded. A numerical simulation is performed to analyze the process where lean hunting is induced as the air-fuel ratio becomes larger, or as the delay in fuel evaporation becomes larger, or as the load becomes smaller. The results of the simulation agree with the experimental fact previously reported.

抄録全体を表示

Behavior measurements have been made with two displacement sensors and an underseal pressure sensor, using an overhanging eccentric shaft-type single-rotor research engine eguipped with a packaged multichannel slip ring. The two-piece seal was tilted to the leading and trailing sides of a seal slot during its travel along the rotor housing surface and vibrated on the top end of the leading side of the slot as a fulcrum after the shift from the trailing to the leading side of the slot after the minor axis on the spark plug side. As for the three-piece seal, its top part was also tilted in all operating conditions, although its bottom part made effective area contact with the side of the slot under light load conditions up to medium engine speeds. The working chamber pressure was induced in the underseal in the same manner as with the two-piece type.

抄録全体を表示