In order to establish the necessary condition for such cavitation-erosion-tests in vibratory-erosion test-apparatus, the cavitation-nuclei size-distribution and the relative air content rate of test water were carefully measured with respect to test time, by means of a Coulter-counter and a van Slyke apparatus respectively. It is worth noticing in the vibratory cavitation apparatus that the exponent and the volume concentration characterizing the nuclei-distributions do not fluctuate with test time in the equilibrium period, which reaches after 60 min. And that the air-content-rate first decreases monotonously with the time, then it reaches a definite equilibrium value after 40 min, which is much longer than the ASTM recommended value.

View full abstract

In order to distinguish the impulsive pressures of a microjet and a shock wave more precisely than in the previous study, another local pressure sensor is manufactured. The sensor is composed four local pressure sensor elements. The size of each sensitive area is 1 mm×1 mm, which is almost the same as the main cross section of the microjet. By the sensor and flash photographs of the cavitation bubble, the impulsive pressures of the microjet and shock wave are evaluated more precisely, and the cavetation mechanism is revealed more clearly at the first collapse-reexpansion stage of a spark-induced caviation bubble near the sold boundary.

View full abstract

The velocity fields around a circular cylinder which was started impulsively in a water tank were measured by flow visualization techniques. The pressure on the surface of the cylinder was estimated by the line integral of the Temples equation using the measured velocities on a closed circuit which is selected to avoid the cumbersome estimation of viscous terms. The estimated distributions of the pressure coefficient C_pθ at Re=1 200, which are in disagreement with those of the numerical analysis for Re= 1 000 by Bar-Lev, are in agreement with those of the numerical analysis for Re=500 by Collins. The variation of the local values of the pressure coefficient C_pθ (θ=90° and 180°, θ: azimuthal angle measured from the frontal stagnation point of the cylinder) with elapsed time was found to be similar as that of C_pθ obtained by both the conventional numerical analysis and the experimental study, though their quantitative agreement is insufficient.

View full abstract

Numerical simulation was made for flows around two square cylinders set tandem in the middle of a channel between two parallel walls. The finite difference method was applied to the two-dimensional Navier-Stokes equations given in the stream function-vorticity formulation. For flows in a multiply-connected region like this, the value of the stream function on each cylinder cannot be given a priori. A method proposed previously by the authors was adopted to determine this value based on the circulation flowing out across the control surface enclosing the two cylinders An additional procedure was introduced to divide the circulation between the two cylinders, based on the cross-produced features of experimental results; as the distance between the two cylinders exceeds some value, the flow pattern changes markedly.

View full abstract

Experiments were carried out to study the transient lift produced on a circular cylinder, which started to rotate impulsively about its axis in a uniform flow. Since the lift was influenced by the strength and position of the shedding vortex, special care was taken to select the initial condition of rotation which is expressed by the lift coefficient and its derivative at t=0. Tests were performed for various acceleration rates of rotation and final rotational speeds in two representative initial conditions. As a result of this study, it has become clear that the higher acceleration results in a bigger shift of the transient lift from the steady state characteristics.

View full abstract

The mean velocity and root-mean-square value of turbulence component in pulsating round jets of low pulsation frequency were measured in the axial range of z/d≧30 by means of a hot wire anemometer, where z is the distance from the nozzle and d is the nozzle diameter. The measured values were approximated by finite Fourier series. The time-averaged velocity on the jet axis decayed faster than the steady value. The time-averaged width of the pulsating jet, represented by the half-value-radius of time-averaged velocity, was slightly greater than the steady value. However, the time-averaged flow rate did not differ from that of the steady jet. The profiles of the time-averaged velocity and the amplitude of the fundamental velocity component became similar downstream at about z/d=5 and 10, respectively. A quasi-steady model for predicting the instantaneous velocity and turbulence on the jet axis is proposed.

View full abstract

A simplified method for the estimation of whirling forces on a centrifugal impeller in a vaned diffuser is proposed. It is shown that the pressure increase in a whirling impeller flow passage is represented by three components: Euler's head, inertial pressure increase, and that caused by centrifugal force due to whirling motion. Euler's head is estimated by quasi-steady use of experimentally determined impeller steady performance. In the same way, the pressure increase in the diffuser is estimated from its steady performance and inertial effects. Based on these ideas and some other assumptions including linearization, the pressure distribution around the whirling impeller is determined. Experimental verifications at zero whirl velocity were made for three different diffusers. At larger flow rates, reasonable agreements between estimated and measured pressure forces were obtained, and it was shown that the forces promote inpeller whirl in the same direction as impeller rotation. Only qualitative agreements were obtained at smaller flow rates.

View full abstract

A method for solving three-dimensional incompressible viscous flows is presented. A complete set of Navier-Stokes equations is transformed and expressed in terms of vorticity, scalar and vector potential. In this formulation, the velocity field satisfies the equation of continuity automatically. The vorticity transport equations discretized by the central finite difference are solved by the rational Runge-Kutta (RRK) time integration scheme, and the Poisson equations are solved by the successive-over-relaxation (SOR) method. The numerical solutions of flows in a cubic cavity, in a square duct and in a rectangular duct with the aspect ratio of 2 are presented. Comparison of the computed results with other calculations confirms the acurracy and reliability of the present approach.

View full abstract

Noncircular jets issuing from square, triangular and rectangular orifices are investigated using a flow visualization technique employing laser-fluorescent dye, and controlling vortex evolutions under excitation. In square, triangular, and 2:1 rectangular jets, the corner regions of noncircular vortical structures undergo severe deformation due to self-induced velocity, resulting in the switchover of the geometric axes. The deformation of vortices in a 4:1 rectangular jet is so severe that the major-axis sides come close to each other, collide, and crosslink, and the original single vortex ring bifurcates to two small vortices via the so-called cut-and-connect process. Interactions of two noncircular vortices are affected by self- and mutually-induced velocities, and produce partial merging, vortex stretching, and splitting to small vortices. These results suggest that noncircular jets are very effective in the enhancement of mixing.

View full abstract

Velocity distributions and fluctuations in a Taylor vortex flow between concentric cylinders in which the inner cylinder rotates (a circular Coutette system) are investigated by means of a flow-visualization technique and hot-wire measurements. The experiments are performed up to Re/Re_c=10 in a system where the Reynolds number Re is slowly increased at a constant rate. It is found that both the dimensionless wavelength λ/δ and the number of travelling azimuthal waves become large as the radius ratio γ_i/γ₀ approaches 1. Wave speed decreases rapidly with an increase in Re and becomes constant: it is smaller than the fluid velocity at the mean radius. The amplitude of waves at the source is larger than that at the sink. The highest peak of the velocity power spectrum does not always correspond to the wave frequency.

View full abstract

Statistical characteristics of a three-dimensional jet issuing from a cruciform nozzle have been examined experimentally to clarify the feature of the present flow field due to the secondary flows. In the extent of x/d≧50 where the mean velocity profile U/U_ox shows the similarity, the ratio of the integral length scales L_u/L_v on the x axis takes a constant value. In the relatively upstream extent, both the skewness factors S(u) and S(v) show a negative value, while both the flatness factors F(u) and F(v) take a positive one. The turbulent large eddy structure imagined from the space-time correlation is significantly different at each location in the whole cross section.

View full abstract

In the usual boundary element method, some special numerical techniques have been developed in order to get accurate solutions at internal points near a boundary where the singularity of the integral equation becomes very strong. However, such superior devices have not succeeded in completely removing the singularity except the analytical integral scheme which tends to contain bigger errors due to discretization. From this point of view, a boundary integral equation i newly formulated. In the implementation, a relative quantity of potential or displacement function is sued. In this report, we confirm the excellence of the present device, and equivalence of the present formulation with relative quantity to the rigid body condition and to the constant potential condition is verified.

View full abstract

One of the 3-D velocity components, which is perpendicular to the main and the radical flows in a 270° curved rectangular channel, was measured in detail by using a Laser-Doppler velocimeter. The developing steady and laminar flow was dealt with under the entry condition of a fully developed laminar flow. The curved rectangular channel having the aspect ratio of 2 and the curvature ratio of 8 was used. The Dean number was 220 and the corresponding Reynolds number was 622. From the LDV measurements at the various angular sections along the main flow direction, the following results were obtained. The secondary flow is directed towards the inner curved wall at the entrance of the curved channel. The counter-rotating vortices in addition to the main secondary flow vortices appeared near the inner wall of the central portion in the 75° angular section. Flow fields obtained by visualization may be quantitatively clarified more accurately by plotting the vectors.

View full abstract

The effect of flow direction on the pressure drop in a two-dimensional channel consisting of a periodically serrated wall and a plain one is studied numerically and experimentally under the steady state condition. The serrated wall is composed of lines inclined and lines perpendicular to the direction of the mean flow. The difference in the pressure drop due to the flow direction depends on the aspect ratio, the Reynolds number and the angle of inclination of the serrated wall. It is found that, in the range of the laminar flow, the pressure drop in the case that the fluid flows in the direction such that the cross-sectional area of the channel expands suddenly is large compared with that in the opposite case. This result is experimentally confirmed by using a water flow. The difference in the pressure drop due to the flow direction is found to take a maximum value in the comparatively small angle of inclination.

View full abstract

The authors have shown that the dynamic behaviour of a centrifugal pump (pump impedance) plays a very important role on the low cycle system instabilities. The phenomena becomes remarkable especially when the following conditions are fulfilled: (1) multistage centrifugal pump, (2) partial flow operation, (3) high rotational speed. This paper describes the development of the advanced measuring system for the pump impedance and the application of it to a three stage model centrifugal pump. the influence of the pump flow rate, rotational speed and number of stages are experimentally clarified.

View full abstract

The authors have reported a new measuring technique using a closed loop feedback system for the pump dynamic behavior. Also, the pump flow rate, pump rotational speed and number of stages have proved to affect the pump impedance. This paper reports the experimental result of the three-stage high-speed model pump. The result shows a very high dependence of pump resistance on the clearance between the pump impeller and volute inlet.

View full abstract

Pump performance depends on the outlet flow of the impeller. A method of surface singularities for core flow in the centrifugal impeller, combined with an integral method for a boundary layer, would explain the mechanism of the performance change with cutting the outlet edge of the impeller blades down. This method is applied to flows in the impellers with various cut downs of the blade edge, and then the calculated results are compared with the experimental ones. Both results are shown to be in quantitatively good agreement. On the influence of cutting the blade edge on the outlet flow, it is indicated that the cut of the pressure surface results in the decrease of relative flow angles with the decrease of radial velocity in the core flow, while that of the suction surface results in only decrease of the radial velocity. The change of the flow separation region due to the cut on the suction surface, however, combines to cause the pump performance to deteriorate.

View full abstract

The hydrodynamic force of a supercavitating cascade is obtained by two methods: the integration of the pressure around the foil surface and the momentum balance due to the velocity diagrams for up- and downstream. since the results by the above two methods are not the same in the linear theory, two kinds of wake models are proposed: Nishiyama's and Oba's models. the purpose of the present paper is to discuss the wake model in the nonlinear theory. First, Pellone and Rowe's cavity model is used and it is shown that this cavity model is not reasonable. Second, a new corrected cavity model is proposed and the discussion of wake models is carried out by the numerical results.

View full abstract

Breakup and interaction of two droplets exposed in a shock wave induced air flow of 195 m/s in velocity were observed using holographic interferometry. Experiments were conducted in a 60 mm × 150 mm shock tube for shock Mach number 1.4 in air. The 1.4 mm diameter water droplets were examined whose Weber number and Reynolds number were 730 and 26 000, respectively. The interval between the two droplets was 8.8 mm. It is found that displacement of a droplet in the first droplet was larger than that in the second one, however, this trend was reversed at the later stage. At the final stage, the mists of successive droplets were coupled into almost one group of mist. As a conclusion, it is revealed that there is a strong interference between the two droplets.

View full abstract

Pressure rise performance of a jet fan is measured by a model experiment. A 1/35 scale model is constructed and experiments under various combinations of parameters is executed. The measured value of pressure rise is compensated with regard to friction loss and resistance of the perpendicular pipe in order to obtain the true value for comparison with theoretical results according to momentum theory. When the jet fan is at the center of the pipe, a good agreement is observed; on the other hand, 15% of the momentum is lost by friction when the jet fan is in contact with the wall. The result will play a useful part in planning longitudinal ventilation systems for tunnels.

View full abstract

In order to extract a wave power efficiently in a reciprocating flow, a new version of an impulse air turbine with a self-pitch-controlled guide vane is developed. The basic data of the turbine are obtained by a model testing of a turbine rotor with fixed guide vanes under steady and sinusoidal flow conditions. A series of experiments has been executed to clarify the effects of the rotor pitch to chord ratio, guide vane geometry, rotor geometry, and guide vane configuration on the performance. The results are compared with that of a Well turbine. It is clarified that the impulse turbine with self-pitch-controlled guide vanes is superior to a Wells turbine in efficiency and starting characteristics. Furthermore, a suitable choice of design factors is suggested.

View full abstract

A new model for predicting turbulent natural convection boundary layers has been developed. In this model, a velocity fluctuation is assumed to be composed of two components, i.e., one represents a forced convection component and the other a buoyancy-influenced component. The turbulence quantities are modelled with reference to the improved two-equation models for forced convection flows. Owing to the strong reaction of the temperature upon the velocity field, the modelled transport equations for the temperature field are solved simultaneously. the validity of the present model has been tested by application to a typical natural convection from a vertical heated plate. The numerical solutions showed that the present predictions fit the experimental results very well, while the existing turbulence models ended in failure.

View full abstract

An experimental study has been performed to investigate the influence of density inversion of water on the free convection heat transfer of air-water layers within a cooled circular tube. The experiments were carried out under the condition that the tube-wall temperature was successfully decreased at a uniform rate. A holographic interferometry technique was adopted to determine the time-dependent distributions of thee temperature in the tube. It was observed that the temperature distribution characteristics were to a great extent changed based on the cooling rate of tube wall, along with the time-dependent sequential flow patterns. The heat-transfer characteristics along the tube wall were also extensively determined. Furthermore, it was found that there were three kinds from the water-air interface of the dendritic ice growths and also a pure ice growth along the tube wall.

View full abstract

The unsteady flow of a liquid film and the surrounding gas on a rotating disk, which is related to a spin-coating technique to form a thin uniform film on a plate, is simulated numerically with emphasis on the heat and mass transfer at the interface. The full governing equations and boundary conditions are formulated and a newly developed calculation method is applied. The numerical results reveal that the liquid film thickness decreases mainly by convection in the middle stage. The radial and azimuthal velocities are proportional to the radius, and other values are independent of it. According to these results, one-dimensional calculation models for both phases are developed. The results show that the increased viscosity due to the evaporation of the solvent has considerable effects on the film formation process when the film is sufficiently thin. The liquid film thins mainly because of evaporation in the last stage.

View full abstract

The surface tension of two alkanes, namely n-hexane(C₆H₁₄), are investigated for temperatures from 273 K to 463 K in the present study. Under the coexistence of the sample liquid with its saturated vapor in equilibrium, the surface tension has been measured by the capillary rise method. At the same experimental condition, two sets of surface tension data have been independently obtained with two different Pyrex glass capillaries I and II whose inner radii are 0.1516±0.0004 and 0.1703±0.0005 mm, respectively. The purities of samples are not less than 97 % for n-hexane and 99 % for n-octane. The accuracy of surface tension measurements is estimated better than ±0.20 mN/m. The temperature dependence of the present results has been represented by van der Waals-type correlation with a standard deviation of 0.14 mN/m for n-hexane and 0.12 mN/m for n-octane, respectively. The uncertainty of the present correlations is estimated to be ±0.2 mN/m.

View full abstract

The problem of the thermofluid mechanics of the steady squeezing of a viscous fluid between a hot wall and a surface of the phase-change material is studied theoretically. The partial differential system consisting of the Navier-Stokes equation, the energy equation and the boundary conditions is found to be reduced to an ordinary differential system depending on only one coordinate in the direction normal to the surface of the hot wall. This ordinary differential system is solved by using the approximate and numerical methods. The problem is governed by two nondimensional parameters, that is, the Stefan number and the Prandtl number. It is found that the surface of the phase-change material melted by a flat wall with a higher temperature than the melting temperature is always left flat. The relationship between the thickness of the liquid layer, the melting rate and the force acting on the interface is obtained, as well as the distributions of the velocity, the pressure and the temperature in the liquid layer. If the inertia term in the Navier-Stokes equation is disregarded, the ordinary differential system is solved analytically. Its solution is obtained in the closed form. It is found to be very useful in many practical cases.

View full abstract

A new method was developed to solve the nonsteady-state heat conduction problems by superposing solutions of steady state and simultaneous transient heat conduction problems. These steady-state heat conduction problems and eigenvalue problems equivalent to the simultaneous transient heat conduction problem were formulated by the finite element method. Especially the eigenvalue problem equivalent to the simultaneous transient heat conduction problem was discretized in space by newly derived variational plinciple. This method was found to give a more accurate solution than that given by the ordinary finite element method using Crank Nicholson method.

View full abstract

In the present study, analytical and experimental investigations were performed in order to clarify the characteristics of freezing heat transfer in a porous media saturated with water in a vertical rectangular cavity. In order to establish the momentum equation, the law of conservation of momentum was adapted for the fluid in our control volume and the equation took into account Forchheimer's extension as the resistance to flow in the porous media. Three different size glass, iron and copper beads were used as the porous medium in this study. The temperature of the cold wall was maintained at -10°C, while one of the hot wall was varied from 2°C to 22°C. comparisons of the analytical results with the experimental ones show good agreement with the exception of copper bead.

View full abstract

The effect of interal convection on heat transfer to superfluid helium (He II) is measured in the rectangular channel (1 mm×10 mm×120 mm in length) with an open upper end and closed lower end. The heater at the lower end induces upward normal fluid flow and downward superfluid flow. The test section is located at a distance of 50 mm from the open end of the channel. The results show that normal fluid flow does not influence the heat transfer. Heat transfer coefficients in the mode of single phase flow are equal to the Kapitza conductance and those in the mode of two phase flow are about 0.6 to 1.0 times as large as the Kapitza conductance.

View full abstract

An observational study has been preformed to investigate the melting behavior of a horizontal ice cylinder immered in quiescent sea water. The experiments were carried out in 3.5 % sea water for the ambient temperatures ranging from 2.8 to 20.3°C. It was observed that the flow consisted of a laminar bidirectional flow for the lower part of the ice cylinder and of a turbulent upward flow for the higher part of the cylinder, and that the ice surface was roughened by a secondary flow caused by an instability based on the buoyancy force in the boundary layer flow. It was also found that the heat-transfer characteristics were markedly affected by the equality of the flow in the boundary layer.

View full abstract

The fluid motion within the mold cavity during the packing and the cooling stages is studied. To evaluate the pressure history in the postfilling process of injection molding, an analytical approach is chosen. The fluid is considered as Newtonian, and its compressibility is explained by the Spencer-Gilmore equation. Also, the cooking effect during the whole process is considered. The results of the analysis are compared with the experimental pressure profile in the model die. The analytical pressure profile is in fairly good agreement with the experiment for different mold thickness parameters. The pressure drop by the cooling effect is also explained quantitatively.

View full abstract

Numerical and experimental studies were conducted to clarify the thermal mixing characteristics of a coolant in the core bottom structure (CBS) of the High Temperature engineering Test Reactor (HTTR) being developed in JAERI. The experimental Study was carried out using a one-seventh scale model of the CBS, and the numerical study was made using a three dimensional time-dependent flow and heat transfer code named STREAM. The numerical results were in good agreement with the experimental ones and it was confirmed that the STREAM code was able to use the prediction of the thermal mixing in the CBS of HTTR.

View full abstract

Heat rating of a diffusive catalytic combustor is limited by the diffusion of oxygen. Therefore, corona wind made by electric discharge was applied in order to supply air to the burning surface effectively. The temperature and concentration profiles were measured, and it was shown experimentally that the corona wind enhances combustion, especially in the region of high heat rating. The mechanism of this enhancement was identified as a forced convective effect by comparison with the equivalent air jet. From the measurement of the electric field by an electrostatic probe, the effect of ions in the combustion field was also clarified.

View full abstract

The combustion process of linear fuel droplet streams is analyzed by the method of matched asymptotic expansions based on an appropriate mathematical model. The interdroplet interaction enhances the unsteadiness involved, through which different properties from the single droplet case appear. The droplet lifetime, which lengthens with decreased spacing between droplets, may depend significantly on thee ambient gas condition. when droplets move faster than the ambient gas, the droplet lifetime increases in proportion to the velocity difference in the lowest order of approximation.

View full abstract

A numerical analysis has been made for the transient behavior of a steam reformer for a fuel cell, the fundamental concept of which is a high-performance radiant heat exchanger composed of a couple of porous media energy converters. A numerical model is developed describing a one-dimensional model of the heat exchanger involving a mixture of methane and steam flowing through a porous medium catalyst, on which a reforming reaction takes place. Some illustrative examples of control patterns are shown in this paper as well as the basic transient behaviors of the reformer. The results show that the time constant of the reformer is much shorter than that of conventional heating systems. Further, the time lag of the output is easily reduced using a simple control pattern. Therefore, the steam reformer can well respond to the variation of load of the fuel cell.

View full abstract

Measurements were made of the turbulent burning velocity of C₃H₈-air mixtures at high temperature and pressure using a spherical bomb. The effects of temperature, pressure, turbulence intensity, and stoichiometric ratio on the burning velocities were examined. It has been found that the turbulent burning velocity and the ratio of the turbulent velocity to the laminar velocity is increased with the progress of combustion. The increase rates of these become larger with increases in the initial and instantaneous turbulence velocity, temperature, and stoichiometric ratio and the relationship gradually deviates from the Damkoehler's burning velocity with an increase in temperature.

View full abstract

A new method to solve three-dimensional combined radiative and convective heat transfer problems is developed by utilizing the radiosity concept as applied to the accurate Radiative Heat Ray Method for systems surrounded by nonblack (gray) walls. This method is applied to determine the heat transfer within a cylindrical gas reformer to obtain the profiles of gas and wall temperature and tube-wall heat flux. The calculated results can represent the shading effects of tube row on radiative heat transfer.

View full abstract

Confined turbulent premixed flames stabilized by a pilot flame were investigated to elucidate the local fine structure of flames and its influence on mixing processes in the combustor. The fine flame structure was classified on the basis of the measured turbulence characteristics, and thee probability density functions (PDF's) of local fluctuating velocities and temperature were examined in detail. A plateaulike distribution in the temperature PDF, which may correspond to a "distributed reaction zone", was first observed in a limited portion of the flame, in addition to a typical bimodal distribution corresponding to the structure of a "wrinkled laminar flame" or "laminar flamelets". The influence of the fine flame structure on the mixing processes was examined using an inert gas tracer.

View full abstract

Dynamic characteristics of fuel flow in an intake manifold of an automobile fuel supplier are analyzed by a three-dimensional computer program for a gas-liquid two-phase flow. The mechanism of the air-fuel ratio increase is clarified for the cases where the throttle valve opening varies from 20 degrees to wide open throttle (WOT). when the throttle valve is choked, air velocity is large not only at the throttle valve but also in the whole intake manifold and fuel is transported to an engine cylinder by this high velocity air. On the other hand, air velocity decreases when the throttle valve is opened, and fuel is slowly transported to the engine cylinder and some portion of injected fuel remains in the intake manifold after the suction stroke. Therefore the fuel amount which flows into the cylinder decreases just after opening the throttle valve.

View full abstract

Mixtures of gaseous methanol, oxygen, and nitrogen were ignited on a hot surface in an enclosed combustion chamber without convective flow. It has been found from thee measurements of ignition temperature of these mixtures that a transition occurs in the ignition temperature under certain conditions. The same transition was also observed fro the gaseous methanol-iso-octane blends. Further, the intensity of OH emission at ignition was measured near the heated plate. The intensity change of OH emission and the gradient of its rise were different in the high and low-ignition temperature regions. Next, the ignition mechanism of methanol by a hot surface was investigated theoretically. The theoretical result suggested that the transition occurs if the activation energy for the low ignition temperature region is lower than the generally recognized value.

View full abstract

The effects of radiant heat fluxes on the stability range and burning velocity of a laminar premixed flame were examined by inserting a water-cooled Bunsen burner into a radiant heat recirculation-type furnace. The intensity of the radiant heat flux in the furnace was altered by regulating the surface temperatures of the furnace wall and downstream permeable plate. It was found that the flame blowoff and flashback limits were hardly affected by a radiation field as intense as the one surrounded by a refractory whose temperature was between 1250K and 1450K, and that no perceptible changes appeared either in the laminar burning velocity or flame thickness.

View full abstract

The heat release rates of an engine with divided combustion chambers have been calculated up to now from the indicated pressures of both chambers (called the gas-flow method here). The enthalpy transfer is estimated by the gas flow, the error of which increases if the pressure ratio between the chambers becomes close to 1. Such a pressure ratio often occurs when the passageway is thick. Because the gas-flow method cannot be applied in such cases, one of the authors has proposed a new method (called the distribution method), suitable when the pressure ratio lies in the vicinity of unity. the relation of the boundary between these methods are, however, not yet clear. The purpose of this paper is to investigate the relations in the boundary region and obtain knowledge about the application limits of both methods by computation and experiments.

View full abstract

This paper gives the experimental results obtained with a two-stroke cycle single-cylinder research engine using an electrically heated combustion chamber to simulate the combustion chamber of a low-heat-rejection (LHR) DI diesel engine with an M-combustion system. In this report, the experiment was performed to investigate the influences of combustion chamber surface temperature and swirl ratio on the combustion of an LHR DI diesel engine with an M-combustion system. It is found that the combustion of deposited fuel on the surface is greatly affected by the combustion chamber surface temperature, and the diesel combustion is controlled by the swirl and the surface temperature of the combustion chamber.

View full abstract

To understand the effect of a torch jet on the flow characteristics in a prechamber spark ignition engine, a laser Doppler anemometer (LDA) is used to measure the velocity in a main chamber under the motored condition. The engine is operated at a constant speed of 1000 rpm with a wide open throttle (WOT). The mean velocity, turbulence intensity and turbulence scales are calculated by a cycle-resolved analysis with a cut-off frequency of 650 Hz. The reasonable selection of the cut-off frequency is based on the ensemble-averaged power spectra. During the expansion stroke, the mean velocity, turbulence intensity and integral length scale are found to be strongly affected by the torch jet flow, which is dominated by the torch nozzle area size.

View full abstract