In order to establish the necessary conditions for cavitation tests using ultrapure water, relative air content rates, cavitation noises and bubbles rising from a vibratory cavitation test apparatus are carefully measured with respect to time. The bubble diameters are measured by the ascending velocities of the bubbles using an LDV. It is note worthy that the relative air content rates of ultrapure water are considerably scattered and the bubbles induced by vibration are smaller than those of tap and still water. The bubble sizes are comparable to the size of cavitation nuclei. The cavitation noises are always changing with time, especially the number of high amplitude acoustic pulses, however, there is no effect on water quality in the changes with time.

抄録全体を表示

To evaluate pressure surge caused by a spiral vortex core in an elbow draft tube, the technique for processing the ensemble-averaged pressure data measured at two points at an angle 0f 90 degrees in the cross section of an inlet cone is regarded as the minimum essential. However, this is not applicable to fluctuating pressure data having a random nature, which are sometimes observed in model testing of a Francis turbine. Thus, the present study aims to develop a new method based on spectrum analysis instead of Fourier analysis for the diagnosis of pressure fluctuations for those cases mentioned above. Experiments were conducted using an elbow draft tube having a long inlet cone under three kinds of swirl strength. The dominant frequencies derived from the cross spectrum of fluctuating pressure were classified and correlated with the behavior of the vortex core inside the tube, which was observed by using a high-speed video system.

抄録全体を表示

Velocity and temperature in the stagnation region of a two dimensional impinging jet were simultaneously measured by digital particle image velocimetry (DPIV) and Laser-induced fluorescence (LIF). Instantaneous thermal flow fields and spatial correlation of the vorticity and the turbulent heat flux show that a streamwise vortex pair in the stagnation region sweeps cold fluid toward the wall and ejects high-temperature fluid toward the outer region. The weighted PDF of the turbulent heat flux indicates that the contribution of the ejection to the net heat flux is dominant. The streamwise vortex pair is transported from the free-jet region and the vorticity is amplified by the mean strain in the vicinity of the wall.

抄録全体を表示

The present paper discusses the availability of the periodic shell boundary condition, which the author developed in the previous paper as an outer boundary condition for molecular dynamics simulations, to nonrigid molecular systems. For this purpose, we have conducted molecular dynamics simulations on a flow of Lennard-Jones liquids past a two dimensional circular cylinder. The ordinary boundary condition, i. e., the periodic boundary condition, gives significantly distorted velocity fields : for example, a flow near the boundary surfaces in the cylinder side direction flows along the boundary surfaces, namely, in the uniform flow direction ; the velocities near the boundary surfaces behind the cylinder are much larger compared to the results of the present boundary condition and Navier-Stokes equations ; a pair of vortices behind the circular cylinder are significantly distorted in the cylinder direction. On the other hand, the results of the present boundary condition agree well with those of Navier-Stokes equations qualitatively, and quantatively to a certain degree ; good agreement is obtained concerning the formation of a pair of vortices behind the circular cylinder and also concerning the flow fields in the side region of the cylinder. These results clearly show that the periodic-shell boundary condition is highly useful for non rigid molecular systems, and needless to say, for rigid molecular systems as well.

抄録全体を表示

Experimental studies on drag reduction of a circular cylinder were carried out. To control the flow around a circular cylinder, a rod was set up upstream of the cylinder. The diameter of the cylinder, D, was 40 mm and the Reynolds number was Re = 4.1×10⁴. The diameter of the rod used as the vortex shedder, d, ranged from 1 to 10 mm, and the distance between the axes of the rod and cylinder, L/D, was varied from 1.5 to 3.0. The results obtained were as follows. The separated point on the cylinder moves downstream from 10 to 150 and the Strouhal number increases about 10% in proportion to the rod diameter. In the range of d/D=0.15 to 0.2 and at L/D=2.0, the total drag of the cylinder decreases by 45%.

抄録全体を表示

To control the flow around a square prism, a small rod was set upstream of the prism. The side of the prism, D, was 20 mm and the Reynolds number ranged from 5. 3 × 10³ to 3. 2 × 10⁴. The effects of the diameter of the rod, d, the distance between the axes of the rod and prism, L, and the Reynolds number on the drag reduction of the prism were investigated. The value of the Strouhal number of the prism is classified into three groups. These groups correspond to the flow patterns around the prism, which are related to) the behavior of vortex shedding from the rod. The maximum reduction of drag coefficient is 70% compared with that without the rod at d/D=0.3 and L/D=2.5.

抄録全体を表示

This paper describes the effect of spacing between two adjoining circular cylinders on flow around two-dimensional circular cylinders. The experiment was carried out in an N. P. L. blow-down type wind-tunnel with a working section of 500 mm×500 mm×2000 mm, and between the Reynolds numbers of 6.3×10³ and 1.9×10⁴. The surface-pressure distributions on the circular cylinder were measured and the drag coefficient was determined from these measurements. The flow pattern around circular cylinders was observed. The power spectrum in the turbulent wake behind circular cylinders was also measured. It was found that shedding vortices from circular cylinders exist in the range of S/D≥2.3.

抄録全体を表示

The objective of this paper is to investigate the suppression of the fluid forces acting on a square prism by controlling the separated shear layer, Flow control is established by introducing a fine flat plate in on one side of the separated shear layer. Measurements are carried out with variations of the position, width and attack angle of the flat plate. Subsequently, the steady and unsteady fluid forces, vortex-shedding frequency and flow pattern are systematicallyexamined. Furthermore, such matters as the mechanism of flow control, the nature of the controlled wake, the relationship between the characteristics of the controlled fluid forces and behavior of the flow are discussed in detail on the basis of the obtained results of fluid forces and flow pattern.

抄録全体を表示

The prediction of blocking of solid particles is very important for the design and the operation of a hydraulic transportation system. The estimation method by means of a flow-pattern diagram has been the conventional way in design, but it is not useful an a real-time diagnosis. The flow sounds emitted by particle-particle collisions and those caused by particle-pipe collisions are closely related to the flow patterns. In order to develop a new method to estimate the flow patterns of solid liquid two-phase flows, the relationship between the sounds and the flow patterns are identified experimentally.

抄録全体を表示

Flow and heat transfer in heated pipes rotating about all axis parallel to the pipe axis are reduced to the same problem as those in heated stationary pipes situated horizontally, when the radius of rotation of the pipe is large compared with the pipe radius. The purpose of this study is to clarify the development characteristics, common to those two flows, of mixed convective laminar flow and heat transfer through similarity arguments and computational studies. A new dimensionless axial distance is introduced to correlate the characteristics.

抄録全体を表示

It is well known that free convection is caused by the buoyancy forces of a laminar steady flow in a heated pipe. However, this experimental investigation shows that free convection occurs when the temperature differences between air in a room and water in an unheated pipe are large. Under such experimental conditions, cross-sectional distributions of axial velocities are measured for water as a parameter of ReRa of the Rayleigh and Reynolds numbers. Furthermore, numerical analysis is also presented, and the calculated results are qualitatively compared with the above experimental data of axial velocities and experimental ones in the case of air reported by Mori et al.(1964).

抄録全体を表示

Flow patterns were visualized, and energy losses due to flow branching were measured in laminar dividing flow of glycerine and its solutions in channel junctions with branching angles of 45° (Y-branch) and 90° (T-branch). Two separation zones were observed. The sizes of the separation zones depend on the Reynolds number (Re₁), the branching angle, the ratio of the cross-sectional area of the lateral conduit to that of the main conduit (m) and the dividing flow rate. The energy loss coefficients decrease with increasing Re₁, but their rate of decrease decreases with increasing Re₁ as the sizes of the separation zone increase. The relationship between the energy loss coefficient and the dividing flow rate depends considerably on the size of the separation zone. The loss coefficients for the lateral conduit of the T-branch are larger than those of Y-branch, and those with m=0.5 are larger than those with m=1. The loss coefficients for the main conduit do not depend very strongly on the branching angle or m.

抄録全体を表示

In this study, the process of solidification of fluid with free surface in contact with a solid wall was investigated numerically and experimentally, where temperature of the solid wall was lower than the freezing point of the fluid. In the experiment, we selected a single liquid metal droplet as fluid, and freezing process and transformation of the droplet were observed using a high speed camera. Furthermore, the computer code, which was developed by the authors based on a finite difference method, was employed to analyze the phase change and free surface problems. The numerical results of the transformation of the droplet agreed well with the experimental ones. The analyses showed the intermittent growth of liquid film and the piling steps of solidified layer.

抄録全体を表示

In the previous research, the primitive variable methods have been proposed in the numerical scheme of magnetic field of thermo-electrically conducting fluids. The point to be emphasized in the above mentioned scheme is to satisfy the conservation law of magnetic flux efficiently. In the present paper, we compare the two numerical schemes of magnetic field, namely the cross helicity method and the numerical residual method, concentrating on convergence of Poisson's equation of the velocity and magnetic fields. Moreover, the relationship between the periodicity of low Prandtl number fluids and the convective inhibitation effect by Lorentz force must be clarified quantitatively. The frequency characteristics of kinetic energy, enstrophy, palinstrophy and hydromagnetic cross helicity are calculated by the maximum entropy method.

抄録全体を表示

Experimental investigation was carried out in an attempt to provide data and phenomenological explanation of spherical couette flow in magnetic fluid for the situation where the outer sphere rotates while the inner sphere is stationary. Data for torque characteristics were obtained when a magnetic field was imposed on the fluid, using a bar magnet which was situated in a stationary inner sphere. The torque (torque coefficient) increases, when the density of the magnetic field is increased and when the spherical gap width is increased, resulting in an increase of the effective viscosity. The main flow dominant regime (spherical couette flow with weak secondary flow) appeared to be extended toward higher Reynolds number, suppressing the strong secondary flow, while with further increase of Reynolds number, the torque characteristics approach the non magnetic field case, showing no effect of magnetic field.

抄録全体を表示

The cutting quality of a plasma arc cutter depends greatly on the momentum and energy distributions in the transferred arcjet. In this paper, using the result of integral analysis for the boundary layer equations of transferred arcjet, the correlations between the momentum loss of the arcjet and the operating conditions of a plasma torch such as arc current, gas flow rate, and cutting stand off were investigated. Furthermore, in order to study the actual effect of the momentum loss on the cutting performance of an unmodified plasma cutting torch (Komatsu G940), the relations between the calculated magnitudes of the momentum loss and the maximum cutting speed for the sheet metals of various thicknesses have been investigated by using a numerically controlled plasma cutting table.

抄録全体を表示

A new hybrid streamline upwind finite-element method, which we mentioned in the previous paper, is based on the finite analytic method developed recently in the finite difference method. Using an adjoint discrete del operator and a discrete del operator which are defined on a dual space, it can be anticipated that we can achieve both the reduction of required computer memory and the highly accurate calculation of the finite-element method. In the present paper, we will discuss a steady one-dimensional advection-diffusion problem, for which exist some exact solutions or other numerical solutions. In order to verify the above-mentioned hybrid streamline upwind finite-element method, an unsteady one-dimensional Burgers' equation, which is well known as a typical example of convection-dominated flows, is numerically calculated at the Reynolds numbers Re(=1/ν)=10, 10², 10⁵. Furthermore, the Neumann's stability condition of the present scheme is also obtained for the one-dimensional advection-diffusion problem.

抄録全体を表示

The efficiency of the Lennard-Jones (6-12) -potential-type isotropic scattering model for the direct-simulation Monte Carlo (DSMC) method is studied by the simulation of normal shock wave structure in helium (He) and argon (Ar) binary gas mixture. The Lennard-Jones (LJ)-type isotropic scattering model has a varying diameter (collision cross section) with relative collision energy (velocity) which is consistent with the diffusion or viscosity coefficient of LJ potential. Calculation results of this model, such as number density, velocity, and translational temperature, are in reasonable agreement with those of LJ potential, although profiles of this model are slightly thinner than those of LJ potential.

抄録全体を表示

The effects of mean shear enforced on decaying isotropic turbulence have been investigated by numerical experiments with the direct numerical simulation (DNS) of Navier-Stokes equations. In flows rapidly distorted from isotropy, it is important to elucidate the behavior of turbulent quantities, in particular, the pressure-strain correlations contributing to the origin of the anisotropic state. The pressure-strain term in the Reynolds stress equations becomes positive only in the spanwise velocity fluctuations, and hence turbulence energy is extracted from components in both streamwise and shear directions. The redistribution process as well as the anisotropic behavior of turbulence statistics are accurately evaluated with a linear analysis based on the rapid distortion theory. Furthermore, from the assessment of turbulence models for pressure-strain correlations using the present DNS data, it is found that none of the existing models represent the energy sink from the shear direction except for a model using the Langevin equation, and the destruction of the Reynolds shear stress is described rather well using a classical linear model.

抄録全体を表示

The three-dimensional Navier-Stokes method with the pseudo-compressibility, the implicit formulation of finite difference, and the turbulence model has been developed for the Francis water runner. The viscous flow in the rotating field can be simulated well with the k-ε two-equation turbulence model in the design condition and several off-design conditions. The results agree with the experimental data under the design flow condition. The strong vortex is predicted near the leading edge and developed in the radial direction under the off-design conditions. Because this program employs the efficient approximate factored method as a time marching as well as the wall function near the wall for the k-ε model, it does not require large amounts of CPU time and it can be used even on a downsized computer such as a work station.

抄録全体を表示

An oscillating jet nozzle has been developed to improve the efficiency of water jet cutting. In this nozzle several piezo-electric transducers attached to an oscillation plate give high-frequency oscillation to inner high-pressure water. These transducers were driven by the electric circuit composed of a function generator, an amplifier and a matching box. Thus the amplitude of oscillation can be controlled arbitrarily. As a pre-experiment, the amplitude of oscillation was measured by means of a laser vibrometer. The cutting performance of the present nozzle has been examined. Furthermore, mechanisms for improving water jet cutting efficiency are discussed.

抄録全体を表示

Through-vane compressors are used for automotive air conditioners and performance of the compressor is affected greatly by leakage loss occurring in a cylinder. The leakage loss at a vane tip is closely related to such behavior of the vane that the tip cannot maintain contact with a cylinder wall during a half-revolution of the compressor. This paper examined leakage flow through the clearance between the vane tip and the cylinder wall, and investigated its influence on the compressor performance theoretically and experimentally. It was clarified that the leakage flow increases compression torque but hardly affects volumetric efficiency of the compressor because of the specific behavior of the vane in the cylinder. Experimental results were predicted well by a theoretical analysis employing hydrodynamic lubricating theory and using the dissociation factor of refrigerant from the leaked oil.

抄録全体を表示

The molecular dynamics method was applied to a gas in order to clarify the mechanism of the Joule-Thomson effect, which is one of the features of real gases having intermolecular potential. In the present study, molecular motions of a monatomic gas with the Lennard-Jones potential were simulated, and the molecular behavior in equilibrium states was obtained. Comparing thermodynamic states under various conditions, the changes of temperature caused by expansion were estimated. It was found that, for the process at constant internal energy, the temperature change was attributed to the redistribution of kinetic and potential energy, while for the isenthalpic process, the temperature change occurred due to the work done by the intermolecular forces, in addition to the energy redistribution.

抄録全体を表示

The absorption cross section of metal clusters having a simple electronic configuration, such as K-Xe clusters, K-K clusters, is studied with the quantum mechanics and molecular dynamics method by solving the time-dependent Schrodinger equation for electronic states and the classical Newtonian equations for ionic and atomic motion. Results show the difference of the absorption spectra depending on the atomic configuration and motion. The larger the cluster size is, the more peaks are observed in the absorption spectra. Such peaks are caused by the influence of other electrons or atoms on the unlocalized electron density.

抄録全体を表示

The field accompanied by evaporation and condensation of alkali metals is not correctly described by the analysis using the continuum equations, because discontinuity of physical quantities such as temperature at the vapor-liquid interface occurs in the phase change process. We apply the direct-simulation Monte Carlo method, which is based on molecular kinetic theory, to the condensation problem of potassium. The simulation results are found to be in excellent agreement with experimental results which were obtained by two of the authors, when the condensation and evaporation coefficients are set as unity. In addition to this, we investigate the vapor molecule behavior in the vicinity of the vapor-liquid interface from the standpoint of the velocity distribution function of vapor molecules, and clarify the physical necessity of the occurrence of negative temperature gradients at the interface.

抄録全体を表示

A further experiment was carried out to confirm EHD (electrohydrodynamic) effect on enhancement of evaporation of a liquid droplet, A steady-state heat flux during evaporation of a single droplet was measured by controlling the flow rates of liquid and heat supplied to the droplet. The liquids used in this experiment were refrigerant R113 and ethanol. DC electric voltage was applied between the heat transfer surface and a needle nozzle which was used to supply the test liquid. The results indicated that in the case of R113 droplet, a higher electric voltage was needed to enhance evaporation. On the other hand, in the case of ethanol droplet, evaporation enhancement was observed at a lower voltage, in particular at the region of low surface superheat. This tendency agreed with the result from a preceding report of this study in which the evaporation time of an ethanol droplet was measured. In the present experiment, the enhancement ratios of evaporation of R113 and ethanol droplets were 1∼2.8 and 1∼7.6, respectively.

抄録全体を表示

A theoretical study has been performed to optimize the fin shape and fin dimensions of a horizontal low-finned tube used in shell and tube condensers with shell-side condensation. A new class of high-performance fin and a rectangular fin are examined for theoretical consideration. Extensive numerical calculations with systematically changed values of fin dimensions and bundle depth are performed to optimize the fin geometry with HCFC-123 as a condensing fluid. Under typical operating conditions, the mean overall heat transfer coefficient of a tube bundle with a vertical bundle depth of 30 is improved by 460 and 420 percent over the smooth tube value, using the low-finned tubes with the new class of high-performance fin and the rectangular fin, respectively.

抄録全体を表示

Measurements of the time-varying cross-sectional mean liquid holdup for vertical upward huge wave and annular flows were made simultaneously at 269 axial locations with a spacing of 5 mm. Typical time-spatial maps of wave behavior are presented. Close inspection of these results reveals that a disturbance wave has usually a single peak, except during the absorption and discharge of waves, whereas a huge wave always has several peaks. Fully developed disturbance waves move not only with a nearly constant velocity, but with nearly the same shape. In contrast to this, the shapes of huge waves change considerably with time. Four patterns for ware behavior are discussed. Short lived huge waves and disturbance waves are found.

抄録全体を表示

Transient Marangoni convection in a liquid droplet suddenly subjected to a parallel radiative heating is analyzed. The effects of the Marangoni number and the Prandtl number with respect to the developments of flow and temperature fields within the droplet are examined. The following results are obtained : (1) The flow pattern in the droplet changes little with the Marangoni number over a wide range (Ma=1∼10⁴) ; (2) the dominant mechanism of heat transfer at early stages is conduction, but it is replaced by convection at later stages provided that the Marangoni number is over 100 ; (3) the developing time for the temperature field is almost independent of the Marangoni number ; and (4) the Prandtl number has little effect on the developing time for the temperature field.

抄録全体を表示

Boiling heat transfer from a small-scale heating element on a glass plate has been investigated using various sizes of elements (0.02 mm sq. to 5 mm sq.) in R 113 and FX 3250. It was shown that, with the decrease of the element size, the heat flux increases, with only one nucleation site for smaller sized elements, in which case significant temperature fluctuations occurred due to bubble growth and departure. Heat conduction in the substrate was analyzed numerically. The increasing effect of the heat conduction for smaller heating elements, and the effective surface heat-transfer coefficient and heat flux were estimated. The surface heat-transfer coefficient deviates from a conventional correlation when the size of the element is smaller than the bubble diameter.

抄録全体を表示

Heat transfer and Fluid flow of turbulent natural convection with a backward-step were investigated experimentally, and the effect of step size on heat transfer was discussed. The wall temperature and the fluid flow were visualized using a liquid crystal and fluorescent paint to clarify the behavior of heat transfer coefficients. It was realized from a series of experiments that low-temperature streaks appear on the backward-heat-transfer plate and that they a significant role in heat-transfer enhancement. These patterns are generated by low-temperature fluid lumps, which possess three-dimensional and unstable structures. Moreover, the maximum heat-transfer point is not consistent with the reattachment point of the separation flow, and these points shift downstream with increasing step size H.

抄録全体を表示

We propose a new two-equation heat-transfer model which is modified from the model developed by Nagano, Tagawa and Tsuji (Proc. ASME/JSME Thermal Eng. Joint Conf., 3 (1991), 233). The main improvement is made by introducing the Kolmogorov velocity scale, u_ε≡(νε)^1/4, instead of the friction velocity u_τ, to account for the wall-limiting behavior. After investigating the characteristics of various time scales for the heat-transfer model, we employed a composite time scale which gives weight to a shorter scale among the velocity- and temperature-field time scales. With these modifications, it is shown that the present model quite successfully predicts the turbulent heat transfer in a boundary layer subjected to sudden change of heat flux, and in a separating and reattaching flow downstream of a backward-facing step.

抄録全体を表示

The present paper describes the numerical results of the fluid flow and heat-transfer characteristics of the impinging turbulent plane jet with the confined wall. Numerical results were obtained in the range of the nondimensional height from H=1.0 to 3.0. The Reynolds number based on the nozzle diameter and mean velocity was set to 8000. According to the experimental results, it had been shown that two peaks of the distribution of local heat-transfer coefficient exist along the impingement wall. To the best of our knowledge, no numerical solution has been obtained for the heat-transfer characteristics of the impinging jet in a very narrow channel. We applied a low Reynolds number version of the anisontropic k-ε model of turbulence, i.e., the Myong-Kasagi model, and a WET model for the turbulent heat fluxes to simulate this turbulent impinging flow and heat transfer. The simulation results were in a very good agreement with the experimental one. We could predict the secondary peak of the local heat-transfer distribution on the impingement wall. The reason why the secondary peak appeared was determined. We pointed out that both treatments of low Reynolds number modeling and this anisotropic k-ε model of turbulence were very important for this very strong anisotropic turbulent shear flow.

抄録全体を表示

A thermal detection technique using an infrared radiometer was discussed and applied to the detection of the internal heterogeneous spatial position in materials. The internal heterogeneous space represents a cavity, an inclusion and a separation of materials. To maintain and manage various structures, their positions must be analyzed. Therefore, in this study, the transient temperature distribution above internal flaws was measured. Solar energy and lump energy were used to realize nonsteady temperature field on the surface of materials within an internal heterogeneous space and those temperature fields were estimated experimentally. Moreover, we certified the temperature field using the proposed model based on experimental results. The detection limit of the thermal detection technique was also discussed simultaneously.

抄録全体を表示

The characteristics of a heat exchanger for the melting of falling snow were investigated theoretically and experimentally. The perfect melting condition and the melting efficiency were related to the number of transfer units of the heat exchanger and two dimensionless parameters. The calculated results for the perfact melting condition and the melting efficiency were in agreement with the results obtained from the field experiments of the melting of falling snow. The melting efficiency of piled snow was compared with that of falling snow. It was found that the melting efficiency of the former was higher than that of the latter because the piled snow acts as an insulator during melting.

抄録全体を表示

In this report, we investigate the characteristics of an absorption refrigerator driven by two different heat sources, hot water and steam, to make use of waste heat from fuel cell. This type of refrigerator consists of the cyclic combination of the single-effect absorption refrigerator and the double-effect one. For absorption refrigerators, there are many types of cycles which depend on solution circulation and generator combinations. First, the optimum cycle from the viewpoint of recovering waste heat is clarified. Then the simulation model is constructed and the static characteristics are investigated. Furthermore, the effects of low-level heat source on the total COP is clarified. There is a lower limit of temperature of the low-level heat source because the temperature becomes lower than solution temperature. Furthermore, we discuss the temperature control when low-level heat source conditions vary and the flow rate of steam is varied.

抄録全体を表示

There are some types of absorption refrigerators which are driven by two different heat sources. In the last paper, the refrigerator driven by hot water and steam which were the waste heat of fuel cell was simulated. In this paper, another type of refrigerator driven by hot water and combustion gas which makes up for the shortage of hot water as the waste heat is simulated and its characteristics are clarified. Furthermore, the optimum control of this machine to use hot water effectively is suggested.

抄録全体を表示

Formation and sustaining conditions and the structure of high-voltage ultrashort pulse plasma of CH₄ diluted with H₂ and Ar gases have been studied for the use of active control of CVD processing. Using square-wave pulsed high voltage with 50∼1000 ns pulse width, voltage-current characteristics, timelag of current rise from that of voltage and emission intensity profiles of the plasma have been investigated. Very high voltages up to several kilovolts can be successfully applied without any plasma nonuniformity on the electrode surface because the pulsed high voltage drops before the transition to arcing. During the pulse period, very large current of more than 10²∼10³ times as that of usual DC plasma can enter into the plasma field. For the formation of stable pulse plasma, the pulse width should be longer than the timelag, which has been comprehensively expressed by the parameters of pressure, applied voltage and distance between anode and cathode. Through the measurement of potential profiles in the positive column of the plasma, it has been found that the stratified plasma structure can be controlled simply through the parameters of current and pressure.

抄録全体を表示

Effects of physical properties of cathode materials on the transition of discharge mode in the boundary layer of combustion gas plasma have been investigated experimentally. At the transition to big-arcs, micro-arc currents fluctuate and some micro-arcs with arc current beyond a critical value grow up to big-arcs, breaking through the thermal boundary layer, whereas most of the micro-arcs with arc current below a critical value remain as micro-arcs adjacent to the electrode wall surface. The micro-arc currents at the transition to the big-arc mode become smaller for the copper electrode than for the stainless-steel and the titanium electrodes. In the case of a copper electrode with lower boiling point and latent heat of evaporation, the electrode material will evaporate rapidly and ionize at the high-temperature spot in spite of its high thermal conductivity, which easily results in micro-arc growth to big-arcs. It is concluded that the electrode materials having both infusible properties and high thermal conductivity can suppress the transition to big-arcs.

抄録全体を表示

The purpose of this work is to investigate design and operation methods for a municipal solid waste incinerator for complete combustion of flue gas, which is essential for reducing the generation of trace hazardous substances such as dioxins. The combustion conditions of an existing incinerator, therefore, are examined by measuring carbon monoxide concentration as an indicator of complete combustion. The effects of oxygen concentration and combustion gas temperature on gaseous combustion are evaluated. Furthermore, the combustion condition of waste on the fire grate is estimated from the distributions of combustion gas components and temperature measured on the section of the furnace outlet. On the other hand, computer simulation is conducted in terms of gas flow distribution inside the incinerator. The results agree with the measured combustion data to a great extent, indicating that the simulation technique is fairly effective to predict gas flow in an incinerator in the case of combustion modification.

抄録全体を表示

The purpose of this work is to establish design and operation methods for a municipal solid waste incinerator for complete combustion of flue gas, which is essential for reducing the generation of trace hazardous substances such as dioxins. The combustion conditions of an existing incinerator, therefore, are examined, and the results demonstrate that sufficient gas mixing and oxygen concentration coupled with combustion gas temperature in the secondary combustion chamber substantially lower carbon monoxide (CO) concentration in flue gas, an indicator of complete combustion. Furthermore, from the results of investigation in conjunction with the computer simulation of gas flow distribution inside the incinerator, some modification plans are suggested and actually implemented : a nose is installed at the furnace outlet and secondary air is supplied to the secondary combustion chamber. The performance tests after the modification indicate that the gas flow distribution in the secondary combustion chamber is equalized and the concentrations of CO and dioxins are significantly lowered.

抄録全体を表示

Two staged combustion tests of coal-water mixture (CWM) fuel were conducted under relatively high-intensity conditions. The influences of spray and air flow on flame stability were tested using a newly designed slit-type atomizer for various air flow conditions, and NO_x reduction effects were investigated. Main results are as follows. (1) Suitable combination of strong primary air swirl with hollow-cone shaped fine spray assures stable ignition and flame stability even under the low primary air ratios, because hot recirculating gas surrounding the spray effectively promotes the volatilization of CWM fuel, (2) The primary combustion of high intensity assures high combustion efficiency and low NO_x emission, because a wide and hot deoxidation zone can be maintained between the primary and secondary combustion zones.

抄録全体を表示

Many pepers have been published on the measurements of in-cylinder flow of an engine, because flow velocity affects to heat transfer to the cylinder walls, and the turbulence intensity also affects the combustibility of the fuel air mixture. Backward scattered LDV is considered to be suitable for measuring the in-cylinder flow on account of its flexibility in application to the engine. However, the demerit of this method is the weak scattered light from the scattering particles. Hence a high powered laser and normal optics are needed and the system is to be so complexed. One of the revisions for the backward scatterd LDV is to use optical fiber to simplify the optics, but this method has demerits of slight scattered light and poor spatial resolution. If it is possible to measure using backward scattered LDV of a low power laser, the measuring system can be more simplified and useful. Therefore, in this paper, the capability of using a low power laser for LDV is studied with revised optics, a signal processor, and scattering particles, and finally the system is applied to an engine and reasonable velocity was measured. Thus the capability of backward scattered LDV using a low power laser is confirmed.

抄録全体を表示

The local values of turbulent burning velocity of propane-air homogeneous mixture in closed combustion chambers were measured from the difference between the flame propagation speed and the unburned gas velocity ahead of the average surface of the turbulent flame front. In turbulent flames produced by the flame passing through a grid and those propagating in turbulent swirl flow, the effects of turbulence characteristics on the burning velocity were investigated. The unburned gas velocity during flame propagation was measured with a laser Doppler velocimeter (LDV). The mean velocity and the turbulence intensity of flow during the short time interval immediately before flame arrival were determined by means of the digital filter of Butterworth type.

抄録全体を表示

Knock is one of the most significant problems that limit the efficiency of an internal combustion engine. It is caused by autoignition of the unburned gas ahead of the flame. In order to understand the knock phenomenon, it is important to measure the temperature of unberned gas. In this study, with polarization maintaining optical fibers, the laser interference measurement of unburned gas temprature was performed in a constant volume vessel and a specially designed engine which could be ingited only once. The engine fueled with n-butane, oxygen and argon, was operated under knocking conditions. When the density of the gas changes, the change of the optical path length of test beam corresponds to the change of refractive index. The temperature history of the unburned gas was determined by measuring the pressure and the change of interference signal. The optical fiber interference system had the advantage of resisting mechanical vibration because test and reference beams were transmitted in the same optical fiber and were seperated only in the test section.

抄録全体を表示

The effect of water injection on NO_x reduction in a high-speed IDI diesel engine was investigated. The results showed that water injection directly into the divided chamber just prior to fuel injection could reduce NO_x emission significantly, without causing deterioration in BSFC and other exhaust emissions including the soluble organic fraction and total particulate matter. The NO_x decreased with an increase in the amount of water injected, and stable combustion was maintained even above a 0.8 water to fuel weight ratio. However, excessive water injection resulted in saturated NO_x reduction and a slight increase in THC. The combustion characteristics with water injection were also examined by means of high-speed photography and in-chamber gas sampling.

抄録全体を表示

The proposed two-zone model consists of a burned zone and an unburned zone, in which the changing processes are thermodynamically independent except that the air is entrained from the unburned zone to the burned zone in accordance with the specified excess air ratio pattern during combustion. The excess air ratio of the diffusion combustion period is assumed to be constant in the present model. The calculated histories of the burned zone gas temperature and soot formation were compared with the measured time histories of the flame temperature and the KL value based on the infrared two-color method. The analytical results agree quantitatively or qualitatively with the experimental results, except for the prediction of the swirl effect. As a result, it was shown that the present two-zone model analysis is very useful and effective in evaluating the combustion process in a DI diesel engine.

抄録全体を表示

An experimental study on an optically accessible research engine used to investigate the effects of supercharging on both characteristics of fuel spray and combustion was carried out. By means of visualization and image processing, the phenomena of change in fuel spray, ignition and combustion have been revealed, and the behavior of flame motion and temperature has been measured under different supercharging pressures from 0 to 0.067 MPa (gage) when other parameters were held constant. Based on the results of combustion diagnosis, the effects of supercharging on the combustion and the methods related to combustion improvement have been discussed. Also, the changes of combustion characteristics agreed well with those of engine performance.

抄録全体を表示

The previous report in this series proposed the conceptual design of a prevaporization-premixing lean combustor for application to an automotive ceramic gas turbine. In the present report, two aspects of reliability assurance are discussed. First, this paper deals with the reliability design of the emissions under transient conditions. Optimization was carried out using the simulation results of the relationship between the response of the variable combustor geometry to follow load changes and the resulting air ratio changes for low-NO_x level. The load-variation pattern used in this investigation was that of the Japanese l0-15 mode regulation. Second, this paper describes the validity of the reliability design prepared for the ceramic liner of the combustor. Service life of the liner was predicted on the basis of stress analysis results and fatigue parameters.

抄録全体を表示