Unsteady numerical simulations are conducted for flows in axial rotors in rotating stall regime. Although numerical viscosity is inevitably introduced to stabilize computation to this highly complicated flow, the obtained flow reasonably explains the observations in experiments. In a certain rotor geometry characterized by solidity, aspect ratio of a blade, hub ratio, etc., stable rotating stall cells appear, as observed in the corresponding experiments, but in another rotor, they do not. It turned out that the rotating stall cell is a unified strong vortex tube which emerges from a couple of inter-blade flow passages. The body of vortex has different moving velocity depending on its portion. Its uppermost part is a stable, nearly two-dimensional vortex and distortion of whole flow field having rotating stall cell is attributed to the flow induced by the vortex. The nature of the stall such as whether the obtained stall is representing light stall or heavy stall has not been identified.

抄録全体を表示

This paper describes a new numerical method to compute the motion of a toroidal bubble that is formed after the liquid microjet threads the bubble. We use the boundary element method (BEM) combined with the finite volume method to calculate the toroidal bubble dynamics considering the heat transfer inside the bubble. The unstructured triangle grids are generated inside the bubble. We investigate the bubble collapse near a plane rigid wall. The results show that the initial radius affects the motion of the toroidal bubble, the temperature fields inside the bubble and the pressure fields outside the bubble. When the initial radius is small, the internal thermal boundary layer becomes thick and the bubble collapse is accelerated. This violent collapse induces the extremely high pressure region near the point of liquid-jet-impact. The wall pressure, therefore, becomes locally much high when the tiny bubble collapses near the wall.

抄録全体を表示

In this paper, we propose a new lattice Boltzmann model for two-phase flows pertinent to thermal nonideal fluids. The LBM introduces a fictitious forcing term and recovers thermohydrodynamic equations for nonideal fluid through the Chapman-Enskog expansion procedure. Numerical measurements of thermal conductivity and of surface tension agree well with theoretical predictions. Simulation of phase transition with heat transfer is carried out. It shows that the phases spontaneously separate into domains occupied by the coexisting phases, and demonstrates that the model is applicable to two-phase flows with thermal effects. Using finite difference Lattice Boltzmann method ensures numerical stability of the scheme.

抄録全体を表示

It takes very long time to exhaust gases from a narrow gap between parallel plates by a vacuum pump, especially in free molecular regime. Pumping characteristic of a narrow gap in free molecular regime is investigated by use of the DSMC method. An expression for the time constant, which governs the exponential decay of gas pressure, is proposed. Two kinds of gaps are considered; the gap between square plates, the gap between circular plates. The size of those plates is 10, 20, 50, 100, 200 or 500 mm. The gap δ is set 0.1 mm. The time constant is a function of only one factor L/δ or D/δ (L: size of square plate, D: diameter of circular plate), and increases exponentially with increase of L/δ or D/δ.

抄録全体を表示

A Vibrating Elastic Plate Flow Generator (VEPFG) is an application of a vibrating plate to a flow generator. It became recently to be utilized for flow generation, in addition to propulsion purposes such as boat thrusters. The fluid-dynamic characteristics of the VEPFG as a fluid-machine, however, have not been fully elucidated. This paper aimed to make clear the pumping action, i.e. flow energy input, flow rate and induced flow field, in addition to the thrust. The performance parameters were made dimensionless by normalizing on the basis of the driving conditions at the leading-edge. The parameters thus defined were proposed in place of those of the thrusters. They could be useful for planning the VEPFG. The induced flow fields were analyzed on the assumptions of small-perturbation and potential flow. The results will serve as a referential base for comparing with actual VEPFG flows and experimental results.

抄録全体を表示

The distribution of flow in a right angle branch during oscillatory flow was studied to clarify the flow phenomena in avian lung. Temporal variation of flow rate was measured by a laser-Doppler velocimeter, and differential pressure between the main and the side branch by a pressure transducer. Flow patterns at the bifurcation were visualized with a tracer method. The results show that the direction-dependent flow separation at the bifurcation affects the flow distribution and causes a net flow from the side branch to the main branch. The net flow magnitudes increase with an increase in the Reynolds number, defined using the cross-sectional average velocity amplitude and the tube diameter in the primary tube. The results also show that the net flow is enhanced by the constriction installed at the entrance of the bifurcation.

抄録全体を表示

This paper describes an attempt to clarify effects of stationary bar wakes upon blade boundary layer with separation bubble. This study aims at exploration of a possibility for reducing the aerodynamic loss due to the boundary layer around blades in turbomachines. The test model used in this study consists of semi-circular leading edge and two parallel flat plates. Detailed flow measurements over the test model are conducted using a single hot-wire probe. Emphasis in this study in placed on the effect of bar shifting or bar clocking across the inlet flow in order to see how the bar-wake position against the test model affects the separation bubble as well as aerodynamic loss generated within the boundary layer. The present study reveals a loss reduction through the separation bubble control using a properly clocked bar wake.

抄録全体を表示

Check valves for checking reverse water flows in drainage systems sometimes make large impacts through their sudden closure caused by accidental cutoff of drain pumps. In order to characterize such phenomena and to find design criteria for impactless check valves, we observed the motion of a check valve and measured its impact acceleration and pressures in the drainage pipe, using a subscale drainage system. Two peaks of impact acceleration were observed; The first one seemed due to collision of the check valve and the valve seat, whereas the second one did due to water column separation and recombination caused by the low pressure during the closure state of the valve. The former can be reduced by adding mass to the check valve and by setting an appropriate valve-seat angle. The latter possibly can be reduced by relaxing the low pressure during the closure. Subsequently, the check valve opened through the water-column recombination impacts and sometimes showed opening-closing oscillation.

抄録全体を表示

Transitions and pressure drop characteristics of flow in periodically grooved channel are numerically investigated by using finite difference method. The channels consist of fundamental channel elements which have concave or convex rectangular grooves on the smooth plate. The flow fields are assumed to be two-dimensional and to be periodically fully developed. We have examined the effect of combination of the fundamental channel elements on transitions and pressure drop characteristics. The critical Reynolds numbers where a steady state flow bifurcates to a self-sustained oscillatory flow and the corresponding frequencies are evaluated for the combined channels. It is found that the self-sustained oscillatory flow occurs as a result of Hopf bifurcation. As for the pressure drop characteristics, it is found that the pressure drop for the combined channel which consists of elements with concave and convex grooves takes a middle value between the channel of concave elements and that of convex ones.

抄録全体を表示

An experimental study on the effects of geometry and aspect ratio of obstacle on the flow characteristics around the obstacle in a vertical upward bubbly two-phase cross-flow is described. It showed that Karman vortex was shed periodically under most experimental flow conditions for rectangular cylinders with blockage ratio β of 1/3, while it was shed irregularly when void fraction was α≳10% for circular cylinder. Its effects on the vortex shedding frequency ƒ, pressure difference ΔP between the front and rear sides of cylinder etc. were also discussed. Two experimental equations were found to express the relations between the drag force exerted on the cylinder and ΔP for square cylinder and circular cylinder, respectively.

抄録全体を表示

In order to better understand the effect of gravity on two-phase flow, the microgravity experimental tests were conducted aboard MU-300 aircraft capable of parabolic trajectory flying. They are carried out in an air-water two-phase flow through 10 mm diameter adiabatic test section, which is 600 mm lengths of transparent acrylic resin horizontal tube. The obtained experimental data covered a range of liquid and gas flow rates with the liquid superficial velocity ranging from 0.095 m/s to 2.56 m/s, and the gas superficial velocity ranging from 0.032 m/s to 21.08 m/s. Flow pattern results obtained under μg, 1 g and 2 g conditions were compared with void fraction based flow pattern transition models developed by Bousman (1995). The effects of the gravity change on flow patterns were significantly large at low gas and liquid superficial velocities. In the comparisons of the microgravity void fraction results with the correlation proposed by Inoue-Aoki (1970), it was found that the micorgravity void fraction was greater than the model as 42% for bubbly flow region. Two-phase frictional pressure drops obtained under three g-levels were compared with each other, and it was also found that frictional pressure drop were well fitted with the Lockhart-Martinelli-Chisholm model. The effect of gravity change on the pressure drops was insignificant for the turbulent flow reginos.

抄録全体を表示

Our main objective of this study is to clarify the energy transfer near the isothermal and adiabatic walls in the compressible turbulent flows. To this end we investigate the turbulent kinetic, mean kinetic and internal energy budgets using two cases of direct numerical simulation (DNS), Case 1 and Case 2. Case 1 is the compressible turbulent flow between isothermal walls and Case 2 is one between adiabatic and isothermal walls. In Case 1 and Case 2, the Mach number based on the bulk velocity and sound speed at the isothermal wall is 1.5 and the Reynolds number based on the bulk density, bulk velocity, channel half-width, and viscosity at the isothermal wall is 3 000. Moreover, two cases of DNS of the incompressible turbulent channel flow with the passive scalar transport, Case A and Case B, are carried out for comparing with the compressible one. The wall boundary conditions of Case A and Case B correspond to those of Case 1 and Case 2, respectively. In Case A and Case B, the Reynolds number based on the friction velocity, the channel half-width, and the kinematic viscosity is 150. The main results are as follows: (1) The turbulent and mean kinetic energy budgets of compressible turbulent channel flow agree well with the data of the incompressible one, when these profiles are scaled by the local mean variables. (2) The compressibility term of turbulent energy equation is very small and that of mean kinetic energy equation is small near the wall. The direction of energy transfer of these terms near the isothermal wall is opposite to that near the adiabatic wall. (3) The compressible part of the turbulent energy dissipation is not negligible near the isothermal wall, while it is negligible near the adiabatic wall. (4) The compressible part of the mean kinetic energy dissipation is negligible near the isothermal and adiabatic walls.

抄録全体を表示

Turbulence statistics near the wall with almost zero wall shear stress are investigated using DNS data obtained from a series of direct numerical simulations (DNS) of Couette-Poiseuille type turbulent flow. Several significant differences are found in the near wall turbulence statistics between Couette-type and Poiseuille-type flows. It is shown that the Couette-type flow has the different production and redistribution mechanisms of turbulence from those of the Poiseuille-type flow. Further investigations of turbulence structures are carried out by using structure dimensionality and circulicity tensors. Turbulent velocity and vorticity fields near the wall with strong shear stress are two-dimensional and two-component respectively, while those with almost zero shear stress are three-dimensinal and three-component respectively.

抄録全体を表示

In high-Reynolds number flow problems, very fine and stretched grids are generally required to resolve accurately thin boundary layers developed along the body surface and so the application of the tetrahedral unstructured grids has several difficulties in computational efficiency and solution accuracy. One promising approach is to use a hybrid unstructured grid technique for the boundary layers; prism, pyramid, tent and tetrahedron. In this paper, the hybrid unstructured grid technique is incorporated into a CFD-based design system, CASPER. To validate the present CFD codes, we carried out the Reynolds-Averaged Navier-Stokes simulations of a military aircraft configuration such as an F-16A in both subsonic and supersonic regions. On wing surface pressure coefficient distributions, longitudinal, lateral and directional forces and moments, the present computed results are quantitatively compared to the wind-tunnel testing data, the Euler computed and the multi-block structured N-S computed results.

抄録全体を表示

In authors' previous works, the new one-dimensional model was proposed to simulate the scroll discharge pressure pulsation, and the estimation of the frequency spectrum for the pulsation noise was made possible. In this study, the Gabor wavelet transform was applied to the pressure pulsation signal simulated by the one-dimensional model in order to specify the time and position of the large noise generation for the scroll compressor. It was found that the absolute value of wavelet coefficient shows the peek around 800 Hz-1.5 kHz at the rotation angle Θ=150°, which corresponds to the phase that the discharge port just starts opening. The experimental measurements of the acceleration of vibration of the scroll compressor have been also performed. It is shown that the measurement result is consistent with the wavelet analysis for the simulated pressure pulsation signal. So it is confirmed that the present model and wavelet analysis are useful to the prediction of the pressure pulsation noise and its generation mechanism in scroll compressor.

抄録全体を表示

The objective of the study is to establish the measuring method of the thin condensate thickness of liquid mixtures by using a laser light absorption method in the unsteady process of water-ethanol Marangoni dropwise condensation. First, the extinction property of test material, whose property against the infrared laser light having wavelength of 3.39 μm have not yet clarified, was measured. Next, the variations of the condensate film thickness after the sweeping of the heat transfer surface by departing drops in Marangoni dropwise condensation cycle were measured. The precision of this method was investigated on the basis of the extinction coefficient of the test material and the thickness of the liquid film measured. As a result, it was shown that the method of this study has good precision satisfactorily and was applicable to the measurement of such kinds.

抄録全体を表示

The objective of the study is to investigate the condensate film thickness for water-ethanol solutal Marangoni dropwise condensation. The laser light absorption method for measuring the unsteady behaviors of the condensate thickness was adopted. The relation between the condensate behavior and the measuring position was specified by the observation using the high-speed digital camera. The variations of the film thickness were measured by changing the surface subcooling for various vapor concentrations. The unsteady variations of the thickness of condensate film, which was appeared after the sweeping of the heat transfer surface by the departing drop and existed among the various drops, and also the initial drop distances in Marangoni dropwise condensation cycle were measured. The dependencies of the minimum film thickness and non-dimensional condensate thickness among the drops on the initial drop distance were shown. It was clarified that the thinnest thickness of the condensate film was around 1 μm. The relation among the dominant factors determining the phenomena was discussed.

抄録全体を表示

Heat transfer mechanism of a swirling impinging jet in the stagnation region has been experimentally examined, using a combined technique of particle image velocimetry (PIV) and laser induced fluorescence (LIF) for the simultaneous measurements of velocity and temperature fields. The present study indicated that the radial width of the jet was stretched with incresing swirl number Sw, and that this stretching phenomena contributes to radially the maximum local heat transfer coefficient. At the stagnation region, the flow near the heated surface was mixed intermittently by reverse flows toward upstream, and spatial distributions of temperature were well correlated with instantaneous velocity vector maps. The dynamic behavior of recirculation zones, which were attributed to swirl number Sw and impinging distance, mainly determined the turbulent heat transfer at the stagnation region.

抄録全体を表示

An analytical method has been developed for inverse problem of two-dimensional heat conduction by using Laplace transform technique. The inverse problem is solved only for two unknown surface conditions and the other surfaces are insulated in a finite rectangular body. In actual measurement, number of measuring points in a solid is usually limited so that the number of the measured temperatures required to approximate the temperature change in it is short to obtain an approximate function with a half polynomial power series of time and the Fourier series of eigen function. In order to clear this shortage, the third order Spline method is recommended to interpolate unknown temperature at a different point from the measured temperatures and a minimal requirement of the temperature measurements is also recommended to be 8 points enough. The results calculated thereby for some representative cases show that the surface temperature and the surface heat flux can be predicted at good agreement compared with the given surface condition.

抄録全体を表示

An experimental study on an effect of jet exhaust gas in a simulated ozone layer was conducted. A stratospheric condition was simulated in use of a vacuum chamber, where its pressure was sustained 2.67 kPa, this being equal to the one at the altitude 25 km. NO gas, which was diluted by N₂, was fired from a nozzle as an exhaust gas in the simulated ozone layer. NO and NO₂ concentration in jet streams were measured by LIF methods, and then reaction amounts between NO and O₃ were derived from these LIF images. The reaction rate derived from NO-LIF was much bigger than the actual one, because O₃ absorbed laser beam near about 226.1 nm to simulate NO molecular and NO-LIF signal. It was not proper to use NO-LIF method to determine the reaction amount between NO and O₃. However, it was useful to visualize the flow in this condition because of its strong intensity. The reaction amount derived from NO₂-LIF was valid compared with a result of simple calculations. NO₂-LIF should be used to measure the reaction rate qualitatively between NO and O₃ in the simulated ozone layer.

抄録全体を表示

It has been believed that exposure to hypertonic solution during freezing is the major cause of the post-thaw cellular injury after freezing at low cooling rates. To develop a fundamental understanding of cell damage due to the hypertonic solution, we have developed a new perfusion technique that isolates the effect of exposure to hypertonic solutions from other effects associated with ice crystals. Using human prostatic adenocarcinoma cells, we characterized the post-hypertonic survival as a function of hypertonic solute concentration, the time of exposure, and the rates of increase and decrease in concentration. Also measured were the membrane permeability and the osmotically inactive volume of cells that are required for the description of osmotic change in cell volume.

抄録全体を表示

Experiments were conducted to study the effect of the size of micro-pin-fin on the boiling heat transfer from a silicon chip immersed in a pool of degassed or gas-dissolved FC-72. Four kinds of micro-pin-fins with the dimensions of 10 × 10 × 60 to 50 × 50 × 60 μm³ (width × thickness × height) were fabricated on the surface of a square silicon chip with the dimensions of 10 × 10 × 0.5 mm³ by use of the dry etching technique. Experiments were conducted at the liquid subcoolings Δ T_sub of 0, 3, 25 and 45 K under the atmospheric condition. The results were compared with those for a smooth chip and previously developed enhanced surfaces. The micro-pin-finned chips showed a considerable heat transfer enhancement as compared to a smooth chip in the nucleate boiling region. The boiling curve showed a steep increase in the heat flux with increasing wall superheat. For the micro-pin-finned chips, the critical heat flux was 1.9 to 2.3 times as large as that for the smooth chip and the wall temperature at the critical-heat-flux point was less than the upper limit for the reliable operation of LSI chips (=85°C). While the wall superheat at boiling incipience was strongly dependent on the dissolved gas content, it was little affected by the liquid subcooling.

抄録全体を表示

The purpose of this study is to clarify the boiling in a horizontal channel and the one of the inclination angle θ of 5° with rectangular section and the behavior of bubble generated from the heated upper wall experimentally. This channel is submerged into the saturated pure water and both sides of the channel are opened. The height H of the channel is between 3 and 14.5 mm. In the case of θ of 0°, the whole heated wall is always covered with coalescent bubble and burnout occurs easily. In the case of θ of 5°, the condition of the heated wall surface depends on the channel height. For H ≤ 4 mm, nucleate boiling continues owing to the periodical inflow of water to the whole heated wall. For H ≥ 5 mm, the heat transfer is the nucleate boiling at a part of the heated wall and the natural convection at the other part. The slight inclination of the channel improves remarkably the characteristics of the heat transfer.

抄録全体を表示

By analyzing numerically a buoyant flow in stably stratified air in an enclosure, a thermal cylinder was found and the cross-over phenomenon (CO) was elucidated as follows: A plume occurred time-averagedly from a line heat source and its height was suppressed by stratification. To transport heat upwards from the plume front, an isothermal line was divided near the front and a thermal cylinder was generated. The time-averaged temperature decreased upwards in the plume, and the temperature at the front became lower than that in environment. Thus, the time-averaged cross-over occurred. In the region where thermal cylinders were rising intermittently above the front, the convective heat was transported by thermal cylinders and the conductive heat were transported from above and both side. As a result, the time-averaged temperature of the buoyant flow increased upwards and approached the environment temperature.

抄録全体を表示

Fluid flow and mass transfer characteristics in an axisymmetric sinusoidal wavy-walled tube are investigated experimentally in the Reynolds number range from 50 to 1 000. The attention is paid to the transitional flow, which is observed in the Reynolds number range of 160 to 200. In the laminar flow regime wall shear stress and mass transfer rate increase with the slope of 1 and 1/3, respectively, whereas in the turbulent flow regime they increase with the slope of 3/2 and 3/5, respectively. In the transitional flow regime they increase dramatically with a sharp slope. It is found that lamiar-like motion and turbulent-like motion alternatively take place with different time intervals in this flow regime. This is quite different from the flow instability for the wavy-walled channel, in which the Tollmien-Schlichting waves are observed. The flow instability in the wavy-walled tube is considered to be responsible for a significant increment of the wall shear stress and mass transfer rate in the transitional flow regime.

抄録全体を表示

The mutual diffusion coefficients of HFC 32 (CH₂F₂), HCFC 124 (CHCIFCF₃), HFC 125 (CHF₂CF₃), HFC 143 a (CF₃CH₃) and HFC 43-10 mee (CF₃CHFCHFCF₂CF₃) into air have been measured in the temperature range 0∼180°C and at atmospheric pressure by the use of the Taylor dispersion method. The mutual diffusion coefficient decreases as HFC 32 > HFC 143 a > HFC 125 > HCFC 124 > HFC 43-10 mee. The present results can also be generalized by our generalization method for the mutual diffusion coefficients of the halogenated-hydrocarbon refrigerants into air based on the principle of corresponding states.

抄録全体を表示

An innovated heat storage system with metal hydride (MH) applies for air conditioning of buildings is applied. In case of the heat storage systems using water and ice, the temperature of making cold water is almost is almost same as that of heat storage process. However, the temperature of heat storage process is about 35°C and that of making cold water is about 5°C, in case of the heat storage system using MH. The part of the absorbed hydrogen is used for cooling the heat storage vessel, MH and absorbed hydrogen, i.e., the absorbed hydrogen is not used for making cold water, through this process. The new concept, which is called the ineffective hydrogen quantity, is proposed for the purpose of estimating this hydrogen loss, i.e., the energy loss. Under this concept, the operation for reducing the energy loss and increasing the absorbed hydrogen per unit mass of MH is studied in this paper. As the main result, the ineffective hydrogen quantity is improved to 10% in our operation, while it remains to 29% in basic operation.

抄録全体を表示

In the present study, effects of iso-osmolar alginate and trehalose solutions on electroporated or unpulsed control murine myeloma cells (Sp2) during the freezing-thawing process are discussed. To investigate the effects of each cryoprotectant, the freezing-thawing process of electroporated or intact cells were observed, which were suspended in alginate-NaCl solution, alginate-trehalose solution or trehalose solution. The microscopic observation took place also at various cooling rates (-1°C/min, -90°C/min, -5°C/min). It was found that extracellular ice grains in the solution containing trehalose did not mechanically suppress cells even at lowest cooling rate. Further, only a seldom intracellular ice nucleation was observed at all cooling rate for the pulsed or unpulsed cells which were suspended in trehalose solution.

抄録全体を表示

This paper presents a method to determine the optimal operational parameters of a combined cycle power plant, using a dynamic simulation program and nonlinear optimization technique. This proposal method combines the optimization calculation based on the nonlinear programming method with the dynamic simulation. The operational design is optimized in consideration with the plant dynamic characteristics, i.e. machines thermal stress, start-up time and energy loss, etc. are predicted by the dynamic simulation. An application for the optimization of plant start-up scheduling is able to reduce the start-up time and energy loss with keeping the machines thermal stress under the allowable values. The result shows the start-up time and energy loss are reduced by 21.7% and 19.5% during the gas turbine loading process, compared with a conventional heuristic method.

抄録全体を表示

Air cooling blade is usually applied to gas turbines as a basic specification. This blade cooling air is almost 20% of compressor suction air flow and it means that a great deal of compression load is not converted effectively to turbine power generation. This paper proposes the cascade cooling method of turbine blade and the consequent improvement of power generation, which is achieved by the reduction of cooling air consumption with effective use of recovered heat. By this technology, current gas turbines (turbine inlet temperature: 1 350°C) can be up-rated to have a potential high efficiency, equivalent to that of conventional high temperature gas turbine (turbine inlet temperature: 1 500°C).

抄録全体を表示

Closed circuit steam cooled gas turbines are drawing attention because of their efficiency compared with that of a conventional air cooled gas turbine. In the closed circuit steam cooled gas turbine, coolant steam is not discharged into the gas path, so dilution of the hot gas stream, rotor blade pumping loss and pressure loss due to mixing of coolant steam are drastically reduced. In this paper, combined cycle performance using closed circuit steam cooled gas turbine are analyzed. The evaluated parameters are bleed and return point of the coolant steam in the bottoming cycle and leakage, pressure loss, mass flow of the coolant steam. The result of this analysis is that the combined cycle efficiency reaches about 53.6% on the HHV bases.

抄録全体を表示

Economic feasibility of microturbines as cogeneration application is investigated. For this purpose, the relationships between the installation number of microturbine cogeneration units and maximum energy demands are surveyed based on an optimization approach. A method to obtain a maximum energy demand for which the optimal installation number changes is proposed by combining an optimal unit sizing problem and a non-linear equation with this problem as a lower level one. In the optimal unit sizing problem, equipment capacities except for microturbine cogeneration unit and utility maximum contract demands are determined together with the systems operational strategies so as to minimize the annual total cost subject to satisfaction of energy demand requirements. Based on this method, a map expressing the relationships between the optimal installation number and the maximum energy demands is illustrated. Through numerical studies for a hotel and an office building, the economic feasibility of the microturbine cogeneration system is clarified.

抄録全体を表示

A variety of kinds of ribbed channel flow is adopted for the internal cooling structure of 1 300°C class gas turbine first stage rotor blade. A characteristic of heat transfer of periodic ribbed channel flow was examined by the numerical analysis and the experiment. The Reynolds number of the flow was examined over 100 000 that is assumed with an actual machine. The numerical analysis was done by LES, and examined both of the Smagorinsky model and the Dynamic model as LES turbulence model. On the other hand, flow field was measured by hot wire anemometer in the experiment. The distributions of temperature and heat transfer coefficient were clarified by measuring by thermocouples and thermovision. The numerical analysis results were good agreement with the experiment results, and it was confirmed to such a complex field that LES was an effective means.

抄録全体を表示

Flamelet structures of flame bases of turbulent lifted diffusion flames have been investigated experimentally by using the Rayleigh scattering method, which can evaluate two-dimensional temperature fields. Methane/hydrogen mixture was issued from a center tube of 3.2 mm I.D. into a surrounding air flow. Two flames were investigated, of which the Reynolds numbers were 4 200 and 6 700, respectively. Temperature dissipation rates and scalar dissipation rates conditional on the maximum temperature were evaluated from the two-dimensional temperature distributions. It found from the results that probability density functions (PDF's) of flame temperatures of flamelets are close to normal and PDF's of the temperature dissipation rate and scalar dissipation rate are to lognormal. It also found that variances of PDF's of the scalar dissipation rate at the flame base are smaller than values for fully burning turbulent diffusion flames. Moreover, the model for turbulent lifted diffusion flames, which uses the scalar field variable in the flamelet model, has been verified as adequate to express the flamelet structure at the flame base except for treatments of the variance of the scalar dissipation rate and a coefficient C_G related to the brush width.

抄録全体を表示

Useful insight into the flame propagation characteristics of sprays may be derived from the study of flame spreading phenomena along a linear array of liquid fuel droplets under micro-gravity condition. Depending on the kind of fuel, droplet size, inter-droplet distance and ambient gas condition, various flame propagation modes will be excited. In the present report, a map of flame propagation modes is proposed through simple physical consideration, thus preparing for the detailed analyses of flame propagation characteristics which will be conducted in the following two reports. It is found that there are three modes in which the expansion of diffusion flame formed around the frontal burning droplet plays an important role for the ignition of next droplet.

抄録全体を表示

Analytical expressions for the mean speed of flame propagation along a linear droplet array under micro-gravity condition is derived on the basis of physical model developed in the present paper. They correctly predict the inter-droplet spacing dependence measured in micro-gravity experiments. It is found that the inter-droplet spacing nearly equal to the maximum diffusion flame diameter which isolated droplets have in the same ambient condition yields the maximum flame propagation speed. For inter-droplet spacing shorter than this critical value, the time which is required for the ignited droplet to heat up to the boiling surface temperature dominates the flame propagation speed so that the flame propagation speed linearly increases with inter-droplet spacing. On the other hand, for inter-droplet spacing greater than the critical value, the auto-ignition delay time of the droplet in the influence of heat flow from the burning droplet dominate the flame propagation speed, yielding an decreasing function of inter-droplet spacing.

抄録全体を表示

The downward flame spread over solid fuel has been studied. Paper sheets are used as samples, and experiments are conduced in a combustion chamber with a vertical duct to obtain uniform ambient atmosphere, keeping the inflow velocity constant. The gas composition and temperature can be changed to examine the effects of ambient atmosphere. The downward flame spread rate has been measured at lower oxygen concentration and preheated temperature (∼150°C). Results show that, flame spread rate is increased with an increase of ambient temperature. As the oxygen concentration is decreased, the flame spread rate is decreased monotonically. Then, the flame can not propagate and the flame extinction occurs at some oxygen concentration. We define this condition as the limiting oxygen concentration for downward flame spread. It can be reduced as the ambient temperate is increased. To discuss these experimental facts, we estimate the net heat flow into the preheat region, which is an important factor to sustain the flame spread. The net heat flow into the preheat region is decreased as the oxygen concentration is smaller. It is interesting to note that this heat flow near the flame extinction is smaller with an increase of the ambient temperature. These experimental data are useful to make clear the controlling mechanism of flame spread.

抄録全体を表示

Influences of fuel spray characteristics on mixture and combustion behavior in a direct injection Gasoline engine were studied. Fuel-Air mixture and combustion gas behavior was calculated by numerical simulation. From this study we obtained following results. By using skewed type spray, amount of attached fuel on a piston at stratified charge operating mode is less than the case of non-skewed type spray. This leads to faster flame propagation near a piston surface.

抄録全体を表示