Cavitation erosion is studied for oil in water emulsions, aqueous solutions of surfactants and polymers, and polymer-added emulsions. A magnetostrictive vibrator with an open beaker is used as the test apparatus. Weight loss of an aluminium alloy specimen is measured after exposure to cavitation. The surface of the test specimen after exposure to cavitation is observed with SEM. Cavitation erosion characteristics of high-water-content fluids depend strongly on the constituent materials. Emulsified liquid paraffin with higher density suppresses cavitation erosion more effectively, and no saturation is observed up to 11 percent volumetric density. Polymer solutions can suppress erosion when the molecular weight of the polymer exceeds a certain value. Erosion suppression of mixtures of emulsion and polymer solution becomes lower or higher than those of the original constituents. The apperance of the eroded surface also varies widely with the fluids.

抄録全体を表示

In order to establish a useful monitoring method for predicting the erosion induced by high-speed submerged water jets, we attempt to clarify the relation between the jet-structure and the cavitation noise arising around the free jets, by using two hydrophones, and also analyze the AE characteristics of the impulsive pressures induced by cavitation on the specimen of the impinging jets. It is found that the energies of high-frequency components in the jets change markedly with the injection pressure and the standoff distance. It is found that the AE characteristics are closely related to the impinging erosion.

抄録全体を表示

In order to clarify curvature effects upon the peening action on a working surface resuling from impinging high-speed submerged waterjets, we systematically investigate the spatial distributions of the impulsive pressures induced by cavitation for several convex- and concave-cylindrical surfaces with various curvatures, by means of a pressure, sensitive film technique which can detect the spatial distributions easily. The distributions do not appreciably change with the curvature, except for the case when the cylinder diameter is less than the jetwidth. Expectedly, the impulsive pressures develop almost uniformly as far as the rear convex surface, even for a slender cylinder.

抄録全体を表示

A three-dimensional detailed numerical simulation for flows with bubbles was conducted in the present study. The simulation was based on the local instantaneous field equations of the gas liquid two-phase flow and on the interface tracking method of the volume of fluid method. The velocity and pressure fields were calculated using the modified SOLA method. The validity of the numerical method was confirmed by the comparison between measured and calculated terminal velocities and shapes of a single bubble in an infinite stagnant liquid for a wide range of the Morton number and the Eotvos number. Then, laminar bubbly flow in a vertical square duct was analyzed by making use of a periodic boundary condition. As a result, two typical void distributions, the wall peak and core peak, were obtained by varying the Eotvos number. This result agreed well with existing experimental data.

抄録全体を表示

A numerical method for calculating an equilibrium state of discrete-velocity gas is presented. Seven nonlinear equations for seven unknowns are derived. These equations are solved using Newton's method, and typical numerical results are shown. It is found that the equilibrium distribution of the discrete-velocity gas composed of more than 7³ velocities within a reasonable region agrees extremely well with the Maxwellian distribution of continuous velocity gas.

抄録全体を表示

Experiments for a cluster of equal spheres falling freely in quiescent glycerol at low Reynolds numbers are carried out using a large test tank. The number of spheres tested is from 3 to 8. When the number of spheres is below 6, several equal spheres, set adjacent at the initial time, arrange themselves to take regular polygonal positions on the same horizontal plane. On the other hand, in the case of numbers above 6, equal spheres, set adjacent at the initial time, do not take regular polygonal positions on the same horizontal plane. The numerical results obtained using Stokes' flow field do not predict these experimental results. However the numerical results obtained using Oseen's flow field agree well with the experimental results.

抄録全体を表示

Low-mass-velocity two-phase flow in a vertical pipe shows lower void fraction than high-mass-velocity two-phase flow even though their qualities are the same. In order to clarify the flow characteristics of the low-mass-velocity two-phase flow, air, water two phase flow experiments were conducted under the froth or annular flow conditions. Experimental results show that wall shear stress is positive even though both gas and liquid superficial velocities are positive. Measured water film average velocity is negative under this condition. These results indicate that local flow reversal should exist along the channel wall. This local flow reversal gives rise to the low void fraction in low-mass-velocity two-phase flow. It is also clarified that the drift flux model can be applied to the low-mass-velocity two-phase flow with local flow reversal.

抄録全体を表示

Budgets of turbulent heat flux and temperature variance were calculated by means of direct numerical simulations of the return-to-isotropy process of decaying homogeneous turbulence on 96³grid points. The process was simulated for different sets of anisotropic initial conditions. In each case, a constant mean temperature gradient was imposed in one direction. The mechanism of isotropization and its effects on the turbulent statistics were investigated. It was found that the rate of isotropization should be affected by the sign of the third invariant of the stress anisotropy tensor, and is faster when the third invariant is negative. The isotropization of vorticity and dissipation tensors proceeds more rapidly than that of the Reynolds stress tensor. When the turbulence Reynolds number is relatively small, as in the present study, the anisotropy of Reynolds stress has a vigorous effect on the directional dissipation components of temperature fluctuation and turbulent heat flux ; the temperature gradient becomes larger in the direction in which the velocity fluctuation is smaller than in the other directions. The ratio between the pressure scalar gradient correlation and the dissipation of turbulent heat flux also should be affected.

抄録全体を表示

To clarify the effect of a collimator in the collimation sputtering method, the film fabrication process in this apparatus is numerically examined by the test particle Monte Carlo method. Distribution of film thickness and coverage of contact hole are analyzed under various conditions given the following parameters : collimator thickness, deposition probability for collimator, collimator location, and background gas pressure. The increase in collimator thickness remarkably improves the overall uniformity of film and the bottom wall coverage of the contact hole, while the film growth rate decreases. The increase in deposition probability, approach of the collimator to wafer, and rarefaction of background gas also enhance the efficiency of the collimator. The enhancement, however, is not as drastic as that by the increase in collimator thickness.

抄録全体を表示

We simulate the grand-canonical partition funtion for intersecting and nonintersecting one-dimensional interfaces by a Monte Carlo technique. The canonical partition function for nonintersecting interfaces is also simulated. We show that the total length of the interface grows longer as the chemical potential increases, which is one of the free two parameters in the system, while another parameter, rigidity, remains fixed. We find that there are no drastic changes in the total length, the internal energy and their fluctuations when the two parameters vary gradually. In the canonical MC simulations for nonintersecting interfaces, we find that there are two distinct phases, ordered and disordered, that there is no phase transition between them, like the case of the intersecting interfaces, and also that the disordered phase is characterized by the fractal dimension H=4/3. Moreover, we find, in the grand-canonical MC for nonintersecting and intersecting interfaces, that there are no phase transitions between the ordered and disordered phases.

抄録全体を表示

A numerical scheme has been developed for solving tho two-dimensional incompressible Navier -Stokes equations on a domain that is infinite in the vertical (y) direction and finite in the streamwise (x) direction. The fourth-order equation for the streamwise velocity (u) is advanced in time explicitly using a compact third-order Runge-Kutta scheme. A standard Fourier method is used in the x direction and a mapped spectral method in the y direction. The various parts of the code are tested by solving two problems with analytical or independently-established known solutions. The nonlinear Stuart solution is employed to test the convective parts, the Poisson part of the code and the time advancement. Finally, we find that the disturbance obtained from linear stability theory grows in the linear regime of a free shear flow and that the nonlinear growth of the amplitude corresponds to the vortex roll-up.

抄録全体を表示

Secondary flow, caused by the Coriolis force, takes place when viscous fluid flows through a straight pipe rotating about an axis perpendicular to the pipe axis. This is analogous to secondary flow, caused by centrifugal force, in a flow through a stationary curved pipe. The quantitative analogy between these two fully developed turbulent flows is demonstrated through similarity arguments, use of experimental data and computational studies using the k-ε turbulence model. It is demonstrated that there are strong similarities between the two turbulent flows regarding friction factors, heat-transfer rates and flow patterns.

抄録全体を表示

In a previous paper, it was reported that the unsteady forces and moments obtained by the proposed "experimental method of the small-scale vibrating cascade" having the same splitter blade profile as main blade were in good agreement with those obtained by the conventional method of vibrating all blades. The splitter blade profile is generally different from the main one in the circular cascade, so that it is necessary to verify the usefulness of applying this methot to the circular cascade equipped with a conventional splitter blade. Using this method, in this report, the unsteady forces and moments on the blades of two small-scale cascades, one of which is equipped with main blades and others which are fitted with splitter blades having about two-thirds the chord length of the main blade, are compared with the results obtained by the conventional method. The former is in fairly good agreement with the latter. Also, the fluid dynamic interference effects between the two small-scale cascades were elucidated in detail.

抄録全体を表示

We have experimentally investigated the relationship between the periodic velocity fluctuation in the near wake of a rotating flat-plate blade and the generation of broadband noise in the case of different setting angles and types of blades. A typical periodic fluctuation of the near-wake flow is caused by Karman vortex shedding from the trailing edge of the blade. Periodic fluctuation due to Karman vortex shedding generates a discrete frequency noise even in a rotating flow field. The frequency of the periodic fluctuation in the rotating flow field increases with the radius of the blade. Discrete frequency noises due to different frequencies generated from a near-wake How along the blade span result in the occurrence of broadband noise from the rotating flat plate blade.

抄録全体を表示

A numerical technique for the computation of three-dimensional compressible turbulent flows through a turbine stage which has a shrouded bucket with tip seal clearance, is presented. In the present analysis, a two-equation model of turbulence is introduced to estimate the turbulence effect. To calculate nozzle and bucket flow regions simultaneously, a steady interaction is assumed on the interface. To compute the flow through the tip seal clearance, a computational mesh system is generated in that flow region. A finite volume method is used to obtain the spatially discretized governing equations, while the damping surface technique is employed for the time integration. In order to verify the effectiveness of the present method, a computation for the flow through a turbine stage is carried out. The present method yields circulating flows between tip seal fins, and a mixing flow phenomenon in the downstream region of the shroud, and the computational result shows good qualitative agreement with the experimental data.

抄録全体を表示

An experimental investigation of a counter-rotating fan was conducted with special attention to the effects of the supporter shape of the electric motor and the distance between two rotors on the fluid dynamic and fan noise characteristics. A comparison of the fan noise and the fluid dynamic characteristics between the counter-rotating fan and a two-stage rotor fan was also made. As a result it was concluded that the noise generated from a symmetrically supported fan was lower by 3 to 6 dB than that from an asymmetrically supported fan, and the fluid dynamic characteristics of the former were superior to those of the latter. The larger the distance between two rotors is, the lower the fan efficiency and the fan noise are. Although the fluid dynamic characteristics of the counter-rotating fan are superior to those of the two-stage rotor fan, the counter-rotating fan can't be recommended yet because it generates from much more noise than the latter.

抄録全体を表示

The effects of three parameters, such as the combination of the number of blades of a front rotor and a rear rotor, the solidity and the blade tip clearance, on both the fluid dynamic and noise characteristics were experimentally investigated. As a result it was concluded that the noise due to the interaction between the front rotor and the rear rotor blades can be reduced if the combination of the number of two rotor blades is well selected. When the solidity of the two rotors was the same, the total pressure coefficient became high and the specific noise level low. The smaller the tip clearance was, the higher the fan efficiency and the total pressure coefficient were.

抄録全体を表示

The propagation of acoustic waves in a circular duct with a cavity has been numerically studied. The governing equation for acoustic disturbances is a linearized unsteady Euler equation. This is solved numerically using a MacCormack-type scheme with spatial fourth-order accuracy to obtain the multidimensional structures of the entire sound field. The present result shows that the sound pressure level of transmission waves is markedly reduced through the cavity region at the specific values of cavity radius. However, the level of transmission waves is not influenced significantly by the length of the cavity and mean flows.

抄録全体を表示

Complete characteristics of a screw-type centrifugal pump were obtained experimentally, and the performance of this pump as a reversible pump-turbine was estimated. Moreover, effects of a net positive suction head on pump characteristics were investigated. As a result, it was found that this pump has a good head-discharge curve and a good limit-load characteristic, and the pump efficiency is superior to that of conventional non clogging pumps. However, this pump is not suitable as a reversible pump-turbine, because the efficiency is inferior to the pump efficiency when it is operated as a turbine. We observe a phenomenon wherein the head and load suddenly increase just before the break-off point of pump efficiency with decrease of the available net positive suction head. Based on the result of flow visualization, it seems that this phenomenon is caused by an increase in the theoretical head due to the occurrence of cavitation in the impeller passage. The required net positive suction head of this pump is smaller than that of conventional centrifugal pumps, and it is found that the suction performance of this pump is superior to that of pumps with the same specific speed.

抄録全体を表示

A series of experiments was carried out on the suction characteristics of a water-ring vacuum pump pumping saturated humid air at 30°C, while the temperature of ring water was varied in the range of 10∼30°C. The steam flow induced by the temperature difference between the water in the suction air tank and the seal ring water drives air against an adverse pressure gradient. Thus, the partial pressure of air at the suction port of the pump was increased. The internal characteristics of the pump under such a condition is related to the standard pump characteristics pumping saturated humid air at an isothermal water-ring temperature. The combined performance is improved markedly by reducing the water-ring temperature, and it is predictable.

抄録全体を表示

Taking into account the equilibrium characteristics of nonideal mixtures, numerical analysis has been conducted for an evaporating annular two-phase flow of nonazeotropic binary mixtures flowing upward in a vertical tube. The equations of continuity, momentum, energy and mass conservation for each phase have been solved numerically by a finite volume method. The predicted streamwise distributions of the heat transfer coefficient, wall temperature and quality were examined for three different mixtures, R11-R113, R22-R114 and R22-R11. When the concentration of the more volatile component of a mixture is high, the heat transfer coefficient was found to decrease towards the downstream in some cases as evaporation proceeds. The assumption of ideality of the mixture can affect the heat transfer characteristics by twenty percent. Nonideality of the mixture must be taken into account, especially if all the local values of heat transfer coefficient, temperature and quality are to be predicted. The significance of the nonideality of the mixture cannot be discussed based on only the difference in the boiling temperature of each component.

抄録全体を表示

Observation of the boundary layer is important to examine phenomena of boiling when a non uniform electric field is applied. Accordingly, temperature changes near the boundary layer were observed by the Schlieren method using a nickel-chrome fine wire as the heater and a rod as the electrode to which alternate current up to 1000 volts was applied. The result shows that the temperature boundary layer is disturbed by the electric field and the stronger the turbulence, the higher the applied electric field. The bubbles generated on the heater are released through the boundary layer and expand outside of the boundary layer. The turbulence in the boundary layer was observed as stripes arranged relatively regularly. The interval of the stripes in the temperature boundary layer decreased as the applied electric field increased.

抄録全体を表示

Detailed measurements are performed on time-averaged pressure and heat transfer distributions around a blunt body which is affected by periodically incoming wakes from the upstream moving bars. The blunt body is a test model of a front portion of a turbine blade in gas turbines and consists of a semicircular cylindrical leading edge and a flat plate afterbody. A wide range of steady and unsteady flow conditions, such as Reynolds numbers based on the diameter of the leading edge and the bar-passing Strouhal numbers, are investigated. The measurement of the surface pressure distribution, in conjunction with the corresponding numerical analysis by BEM (Boundary Element Method), reveals the existence of the separation bubble around the junction between the semicircular cylinder and the flat plate, which is emphasized especially in higher Reynolds number cases. The wake-affected heat transfer distributions indicated that the wake passing over the leading edge caused significant increase in heat transfer in that area, and such phenomena become more pronounced in the higher Strouhal number cases.

抄録全体を表示

Flow visualization and heat transfer measurements were made for a turbulent channel flow between two parallel plates obstructed by an insertion of a square rod. The space between the square rod and the duct wall was changed in five steps. The intermittent dye injection method was introduced to visualize the Karman vortex shed from the rod, and the locations of the Karman vortices were visualized as they move downstream. The results reveal that the Karman vortex shows crisscross motion. In accordance with the motion of the Karman vortex, a high-vorticity wall layer is intermittently lifted from the wall. The second peak in the streamwise distribution of local Nusselt number is caused by the action of vorticity island, which is produced periodically from the tip of the lifted wall layer.

抄録全体を表示

Experiments with a copper-water wickless heat pipe were carried out. The influence of working fluid fill charge and inclination angle on the heat transfer coefficients of the evaporator and the condenser was studied. The following results are summarized. (1) The heat transfer of the evaporator does not depend on the inclination angle. (2) As the heat transfer coefficient in the case of the liquid film is larger than that in the case of the liquid pool, the more the liquid fill ratio decreases, the more the evaporation heat transfer coefficient increases. (3) In the region of nucleate boiling, if dry spots do not exist on the evaporator, the evaporation heat transfer characteristics agree with Kutateladze's correlation in the cases of V⁺=0.283 and 0.455. (4) The condensation heat transfer agrees with Nusselt's equation at the inclination angle of 90 degrees. However, the equation does not agree with the condensation heat transfers at the other inclination angles.

抄録全体を表示

Instantaneous spanwise heat transfer distributions around the reattachment region of a separated shear layer were measured using constant temperature dynamic heat flux meters arrayed in one or two spanwise lines. The spanwise scale of the heat transfer distribution and the traveling velocity of the characteristic patterns of the heat transfer distribution were derived from these data through various time and/or space correlations. Consequently the results show that the instantaneous average of the spanwise heat transfer rate increases when and/or where the spanwise irregularity of the instantaneous heat flux distribution is high.

抄録全体を表示

Experimental and numerical studies of heat transfer for turbulent flow in noncircular duct with rough wall are few and far between. In this study, numerical analysis has been carried out on heat transfer for a developing turbulent flow in a square duct with a rough wall using Reynolds stress model and turbulent heat flux model. Calculated results are compared with the experimental data available. The present method can predict well the mean flow distribution and temperature distribution distorted by the secondary flow of the second kind. From calculations, the similarity which appeared between the mean flow distribution and temperature distribution is not observed near the smooth wall facing the rough wall, while it is observed from the experiment. Although the local Nusselt number of smooth walls are understimated by the present method, the production terms of the normal stress u²₃, shear stress u₁u₃ and turbulent heat flux u₃ T' are in good agreement with the experimental data.

抄録全体を表示

Temperature response in a functionally gradient material (FGM) which is subjected to transient heating is investigated in order to estimate the profile of the thermal diffusivity from the temperature response on the rear surface of the FGM. Focus is placed on a distribution parameter which yields the profiles of the thermophysical properties when there exists an exact analytical solution for the temperature response in the FGM. An explicit relationship is obtained between the distribution parameter and thermophysical properties on the rear surface. This explicit expression can represent the dependence of the temperature response on the thermophysical properties to ±4%. It is expected that this identification can provide useful insight into the estimation of thermophysical properties in FGMs. Usefulness of this relationship is also examined by comparing given and estimated profiles for the thermal diffusivity. Fair agreement is demonstrated, as far as the trend and the approximate magnitude are concerned.

抄録全体を表示

Development of the three-dimensional structure of planar mixing layers under a pressure gradient was studied using a visualization system which enabled time-resolved two-dimensional concentration measurement. The unsteady concentration field in the transverse plane was visualized and analyzed to obtain statistical quantities of concentration. The data were also used to visualize instantaneous distribution of dissipation rate in the transverse plane and to reconstruct the three-dimensional image of the mixing layers based on Taylor's hypothesis. It was found that development of the spanwise instability was enhanced by a positive pressure gradient, as was the development of the Kelvin-Helmholtz instability, and that the development of spanwise instability was suppressed by the negative pressure gradient. The spanwise structure occurred with the same periodicity as the initially given spanwise perturbation even under the pressure gradient. Streamwise vortices caused by the spanwise instability resulted in the strong dissipation of concentration at the peak and in the valley of the spanwise structure. The spatial development of the spanwise instability could be easily understood by the three-dimensional image of the concentration field.

抄録全体を表示

The gas velocity in the circulating fluidized bed of aluminum ore calciners, petroleum crackers, boilers and so on is higher than the terminal velocity of the particles. This paper presents the growing chain model, wherein the particles form vertically growing chains in the fluidization, to discriminate theoretically the regimes of the fast fluidization. In referring to the fluidization condition empirically recognized as the satisfactory circulating fluidized bed boiler's operation, the upper limit of the fast fluidization gas velocity is found to coincide with the terminal velocity of the infinitely long chain, a few times as high as that of a single particle.

抄録全体を表示

An investigation was conducted into the generation of a disturbance wave on a liquid film flowing down a heated inclined plane. Theoretical analysis shows that the Marangoni effect due to the surface tension gradient contributes to reduction of the critical flow rate of the liquid on the generation of a disturbance wave. Theoretical results also suggest the existence of two boundaries for generation of a disturbance wave, at a fixed inclined angle. The lower boundary, however, could not be observed for thin water film because liquid film breakdown occurred before the lower critical flow rate was achieved. In this paper, silicone oil with high viscosity was used as the working fluid instead of water. Small waves were observed under the lower critical flow rate. The surface temperature gradient along a wave is one order of magnitude larger than that of a disturbance wave under the higher critical flow condition. We defined a wave under lower critical condition as a Marangoni wave. The measured velocity of a Marangoni wave is almost the same as the liquid surface velocity.

抄録全体を表示

The change in refractive index with temperature has been determined at atmospheric pressure for water, methanol, ethanol, 1-propanol, 1-butanol, n-amyl alcohol, pentane, hexane, heptane, octane, nonane, decane and four different kinds of dimethyl silicone oil from about 2 to 80°C. All measurements were made using the D spectrum of the natrium line (wavelength : 589.3 nm). Liquid sample was boxed in a small Pyrex container in order to prevent evaporation of test liquid during measurement. The Pyrex container was set into an Abbe refractometer. The measured refractive index for the length of 589.3 nm was transferred to that to the wavelength of 632.8 nm using dispersive power. The data of refractive index vs. temperature were fitted to a second-degree poloynomial. By differentiating tese equations with temperature, thermooptic coefficients dn/dT were obtained over the range of temperature. Estimated maximum error in the thermooptic coefficient is within 3%.

抄録全体を表示

The two new developments in the analysis are reported of melting of a snow layer with the combined effect of temperature and concentration, which are a moving diffusion layer model (MDL- Model) and a fixed diffusion layer model (FDL-Model). The conservation of mass is satisfied in the mass balance integral of both models. The diffusion layer thickness in the MDL-Model increases, while that in the FDL-Model is fixed during the melting process. The difference between a classical Stefan problem controlled by temperature diffusion and the present melting problem controlled by concentration diffusion is clearly observed. It is also found that the optimum initial thickness of the diffusion layer is about 0.04mm in the MDL-Model, and that the optimum fixed thickness of the diffusion layer in the FDL-Model is about 0.03 mm fitting the tickness by Nernst's concept for dissolution.

抄録全体を表示

This study shows quantitatively that existing experimental data on countercurrent flow limitation (CCFL) or flooding in vertical annular channels can be predicted well by applying a criterion that the CCFL condition be given by maximizing falling water mass velocity with respect to falling water film thickness for the entire flow channel, which is the same criterion as that which the author successfully applied to the CCFL conditions for both vertical circular tubes and vertical rectangular channels. Quantitative analyses were carried out for the previous experiments of both air-water and steam-water countercurrent flow systems under atmospheric pressure in the vertical annular channels of 0.46 to 1.22m in length, 0.91 to 1.85m in circumference and 12.7 to 50.8mm in gap size in both cases where water is injected into the upper plenum and through nozzles, simulating a vertical channel in PWR pressure vessel.

抄録全体を表示

The shell-and-tube heat exchangers mentioned in this paper are the typical types of heat exchangers. We divide them into four classes, on the basis of the fluid condition. That is, (1) one fluid condenses and the other vaporizes, (2) one fluid condenses and the other does not change its phase upon heating, (3) one fluid vaporizes and the other does not change its phase upon cooling, and (4) neither fluid changes its phase upon heat-exchange. In this paper, the dynamic characteristics of the four types of shell-and-tube heat exchangers mentioned above are investigated experimentally and theoretically. In the theoretical study, the dynamic characteristics were investigated from the viewpoint of the network theory. As a result, a very simple network model was derived. It is shown that this model is adaptable for all types of shell-and-tube heat exchangers regardless of the operational condition and the kind of fluids in the heat exchanger. The validity of the model was proven by comparison with the experimental results.

抄録全体を表示

The electric power generated by the photovoltaic array depends on the insolation and cell temperature, and the cell temperature depends on the insolation, ambient temperature and wind speed. In this study, it is shown that the cell temperature and output power can be calculated by considering the heat balance and the voltage-current characteristics of solar cells. A photovoltaic power generation system connected with a commercial power source was constructed, and a long term test of the system under natural weather conditions was performed. From the experimental results, it was confirmed that the method of calculation yielded a close approximation.

抄録全体を表示

Recentry, much attention is paid to the co-generation system which consists of both a power generator, such as combustion gas engine, and an absorption refrigerator. In this system, it is necessary to clarify the dynamic behavior of the absorption refrigerator for disturbances, such as flow rate change and inlet temperature change of the ho)t water. Moreover, downsizing of the absorption refrigerator is now desired. However, the operation becomes more sensitive to disturbances as the size of the refrigerator decreases. Is this study, the authors establish the dynamical model of the absorption refrigerator driven by hot water. This model is to be used as the tool for the design of a control system and refrigerator with good controllability. For this purpose, the model should be as simple as possible. Lumped, parameter models are adapted for the elements where this approximation is adequate. The validity of this model is confirmed by comparing the calculation results with the experimental results.

抄録全体を表示

To study the characteristics of flow and particle diameters in large-angle conical spray proposed for Diesel engines of premixed combustion, unsteady conical spray in the air has been analyzed using a phase Doppler anemometer (PDA). In experiments, the frequency of injction is 20.8 Hz and the amount of injection fuel is 10.3 mg in each cycle. Axial and radial distributions of bulk velocity and mean diameter of particles are obtained. As a result, it is found that the profile of mean velocity distribution is close to the Goertler distribution, and the Sauter mean diameter of droplets in the conical spray is as small as 30μm. Correlation between particle diameter and velocity in the entire spray is negative and unusual, but is found to be normal if the spray is divided into parts and studied. We also find that particles with small diameter (20μm) which are emitted at high speed (60m/s) at the earliest stage lose velocity rapidly, merge into bigger particles, and are overtaken by bigger but slower particles which are emitted at the later stage. Particles with large diameter (50μm) gather at the tip of the spray as they travel farther from the nozzle exit.

抄録全体を表示

The present authors have been investigating the spontaneous ignition of gaseous and liquid fuels injected perpendicular to a steady hot air stream at atmospheric pressure. In this report, which concludes this series of studies, the ignition characteristics of gas jets and spray jets of hydrocarbon fuels were compared using results newly obtained during this study and reported before, and results of other investigators. The following results were obtained : (1) Ignition delay curves are straight lines with large gradient in the Arrhenius coordinate. (2) Apparent activation energies are almost the same in all gas and spray jets. Additionally, activation energies of fuel gas jets nearly coincide with those of premixed gas ignition. (3) Although ignition delay data measured with open combustors correspond well to each other, the data for closed combustors show large variations.

抄録全体を表示

In order to develop methods of plasma injection for removing pollutants such as NO_x and SO_x contained in combustion exhaust gas, the authors have investigated radical formation in Ar and N₂ arc plasma. It has already been concluded that the NO_x removal efficiency has a maximum value in the low range of plasma power. In this paper for the purpose of further savings of the electrical energy in practical use, N₂ gas is excited by passing through a corona discharge tube in a high-voltage field, and produced nitrogen radicals are mixed with simulated exhaust gas (N₂+O₂+CO₂+H₂O+NO) to reduce the NO_x contained in gas. The O₂ and H₂O contents have no effect on the NO_x reduction, and CO gas from CO₂ is not generated. The NO_x reduction ratio of about 10% is obtained by N₂ corona discharge under the input power of 30 watts.

抄録全体を表示

Combustion and exhaust emissions in a DI stratified-charge engine with a large cylinder diameter (135 mm) were examined. In order to obtain a flexible stratified mixture distribution, the fuel was injected in two stages : the first stage was set before the compression stroke to make a uniform premixed mixture and the second stage was at the end of the compression stroke to maintain stable ignition and combustion. The results showed that, compared with homogeneous stoichiometric combustion under rated power operation, 30% SFC and 50% NO_x reduction could be simultaneously achieved with lean stratified-charge combustion. SFC improvement with lean stratified-charge combustion resulted mainly from faster combustion and decreased cooling loss. Optimization for spark timings, secondary fuel injection timings, the proportions of two-stage injection, and the number of nozzle holes were all essential factors for sufficient combustion.

抄録全体を表示