To clarify more precisely the mechanism of a multitudinous constitution of cavitation, the processes whereby a flowing-in nucleous grows and shifts into an arbitrary type of cavitation bubble are investigated carefully in a typical external flow around a Clark-Y 11.7 hydrofoil by using the technique of high speed photographing simultaneously from two sides, under a prescribed cavitation nuclei as well as various hydrodynamic conditions. The processes depend strongly on the place where a nucleous flows in and attaches to the hydrofoil surface. in practice, except for the so-called attached cavitation, there are deviations from sphericality due to the flowing-in conditions and each nucleous develops simultaneously into each individual type of bubble, e.g., spherical-, hemispherical-, concaved-, patched- and torus bubble. The torus bubble has a dent which seem to generate a microjet.

抄録全体を表示

The falling drag test of circular cylinders was carried out in a low Reynolds number region in order to study the effects of cylinder length and plane side-walls on the drag of cylinders. From the data analysis based on the dimensional analysis method, it is shown that the drag coefficient of finite length circular cylinders is expressed by an empirical formula composed of the Tachibana-Iemoto expression in two dimensional theory and the correlation function of three variables (L/D, H/D and B/H). In the useful region of our formula, the drag of cylinders increases with an increase of the ratio of length to diameter (L/D) and a decrease of plane side-wall variables (H/D and B/H).

抄録全体を表示

The velocities and drag forces experienced by two equal-sized spheres freely-falling along their line of centers are measured using the 'falling-sphere' technique at low Reynolds numbers. The width of the test tank with square cross-section used was 100 times as large as the diameter of the used sphere. The measurements, with an estimated uncertainty of ±1.2%, show that the lower sphere experiences the Oseen drag and that the upper sphere experiences the Stokes drag, which is predicted by Stimson & Jeffery. The flow visualization using a naturally occurring phenomenon in glycerol illustrates the difference of the hydrodynamic drag force between the lower sphere and the upper sphere. The numerical calculations of Stimson-Jeffery's solution reveal in detail the occurrence of vortex flows within a closed separating region between the two spheres.

抄録全体を表示

A two or three-dimensional turbulent jet along a curved wall, by the Coanda effect, has been investigated by many researchers. In spite of many studies for this Coanda jet flow, the flow characteristics have not been completely understood. This paper presents a theoretical analysis using an integral momentum equation and an experimental analysis on velocity and pressure distributions for a two-dimensional jet along a circular cylinder. From the theoretical analysis, the equation of velocity distribution, pressure distribution, decay of jet centerline velocity, changes of jet width and half width etc, are made clear, and it is shown that they are in good agreement with experimental ones.

抄録全体を表示

The self-excited oscillation of a rectangular jet impinging on an edge is investigated experimentally. Normalized power spectra of the pressure fluctuation on the edge surface and those of the sound pressure level are obtained. Since a fundamental oscillating frequency of pressure fluctuation at the proximity of the edge corresponds to that of the nozzle exit, it seems to be transmitted directly to the nozzle exit. So, the feedback path of pressure fluctuation was cut off by setting plates between the edge and the nozzle exit. As a result, the essential oscillating mode disappeared. Next, the jet was excited laterally at the proximity of the nozzle exit. When the frequency of excitation corresponds to that of the essential oscillating mode, similar oscillating phenomenon without cut-off plates appear. Hence, it is found that the oscillating phenomenon is generated by the feedback of pressure fluctuation at the proximity of the edge to that of the nozzle exit.

抄録全体を表示

When condensation occurs in a blade passage in a steam turbine, it affects an oblique shock wave originating from the trailing of the blade. In the present study, a condensing flow was produced by an expansion of moist air in a supersonic circular nozzle, and, by inserting a wedge-type shock generator placed in the supersonic parts of the nozzle, the effects of condensation on an oblique shock wave were investigated experimentally. As the result, the position of the oblique shock wave relative to the condensation zone and the change of shape of the oblique shock wave due to condensation are discussed. Furthermore, the load working on the nozzle wall, obtained from the measurement of pressure distribution, and the effect of condensation on the load are discussed.

抄録全体を表示

This paper dials with the effect of a diffuser radius ratio (exit radius/inlet radius) on the predicting equations, that is, the equations of a critical inlet flow angle for the onset of a diffuser rotating stall and of rotational speeds of stall cells. The main results in this study are as follows : A measured critical inlet flow angle for the diffuser with the largest radius ratio is the largest in this experiment and it is close to a predicted value. A non-dimensional form of a rotational speed of cells used in this paper proves to be valid for the diffusers with radius ratio smaller than 2.35. The predicting equations of rotational speeds of cells are available over wide ranges of a non-dimensional width and of a diffuser radius ratio. The predicting method of rotational speeds of cells presented in this paper is very simple ; nevertheless, predicted values agree well with measured values.

抄録全体を表示

To separate substances from solutions, membranes have been used. However in these cases, the phenomena of concentration polarization on the surface of the membrane and the growing of gel layers can not be avoided. These phenomena slow down the filtration. Creating a turbulence on the surface of the filter is one of the effective ways so prevent these phenomena on the filter surface. However, there are few reports which provide details about them. This report is a theoretical and experimental analysis of a flow patterns occuring around a single flat plate, the simplest type of turbulence promoter. The relationships between the Reynolds Number Re and the coefficient of loss, the coefficient of pressure, and the coefficient of pressure fluctuation were examined. The new non-dimensional number M which indicates the effect of the duct wall is proposed and the coefficient of loss ξ can be expressed as a function of the number M. And the relationships between the coefficient of loss in experiments and the number M are coincident with the theoretical values.

抄録全体を表示

In order to clarify the measuring error on differential-pressure-type flowmetering under the conditions of compressible flow with pulsation, the effects of fluid compressibility, beta ratio, upstream pipe length, pulsation intensity and Strouhal number on the error have been investigated respectively, by experiment and unsteady compressible flow analysis. The characteristics of the error E with the Strouhal number, St, are plotted indirectly and compactly, by introducing the variation of MR (ratio of mass flow amplitude in an unsteady flow case to that of quasi steady flow case) with St. This curve makes it possible to estimate the error curves qualitatively.

抄録全体を表示

This paper describes an experimental study of mean flow properties in the three-dimensional turbulent boundary layer over a ridge installed on a flat plate with a yaw angle of 30°. The initially two-dimensional turbulent boundary layer on a flat plate is subjected to the three-dimensional perturbation by the yawed ridge, and then the three-dimensional boundary layer relaxes to the two-dimensional one on a flat plate downstream of the ridge. Mean velocities have been measured using a V-type hot-wire prove. The changes in the velocity profiles are investigated by utilizing a polar diagram and logarithmic velocity distributions. Wall shear stress is estimated by the same methods as in the two-dimensional turbulent boundary layer. The extent of the relaxation region, a region where the variations of the boundary layer parameters such as the wall shear stress and integral thicknesses are remarkable, is made clear.

抄録全体を表示

An experimental study is made of turbulent shear flows over a concave surface influenced both by a streamline curvature and a positive pressure gradient in order to clarify the feature of a mean flow field. The momentum thickness Reynolds number is 3200 and the effective curvature parameter kδ₂ deduced from the streamline curvature is -0.028 at the onset of curvature. The acceleration parameter K is set to an order of -1.0×10^-6 near the onset of curvature. The mean velocity profile is mainly affected by the pressure gradient. A rapid decrease in the skin friction coefficient is observed. The turbulence intensity shows a second peak in the outer layer of the boundary layer.

抄録全体を表示

Turbulent shear flows over a concave surface in an actual turbomachine are subjected both to streamline curvature and to a pressure gradient. The purpose of the present paper is to investigate experimentally the boundary layer flow over a concave surface with a positive pressure gradient at the onset of curvature. Measurements of the Reynolds stresses and the triple products of the velocity fluctuations are made. The acceleration parameter K is set to the order of -1.0×10^-6 near the onset of curvature. A large change in the Reynolds stresses and the triple products due to the positive pressure gradient is found. The Reynolds stress (u²)^^^- is mostly influenced by the pressure gradient, and has a peak apart from the surface.

抄録全体を表示

An efficiency of a Multi-grid method solving two-point boundary value problem is tested. First, Poisson equation is solved by a point - and a line - SOR methods with and without the Multi-grid procedure. Work Unit, indicating iteration number, and CPU time are compared. The effect of the application of the Multi-grid procedure is remarkable; e.g. for the case of point SOR method with 192x192 grid points, the equation can be solved about thirty times faster than the usual single grid method. Second, the present method is applied to a Partially Parabolized Navier-Stokes (PPNS) equations solver. It is confirmed that the method effectively works without any difficulty caused by a non-linear effect of the equations. Finally, an effective Multi-grid PPNS solver is proposed.

抄録全体を表示

Differential pressure fluctuations are used to characterize various flow patterns in inclined two-phase flows as well as in vertical and horizontal flows. They are measured at the short observation scale equal to the pipe diameter and the long observation scale equal to ten times the diameter, to detect local and instantaneous flow changes corresponding to variations of the gas phase at the measuring interval or location. The experimental results show that nitrogen gas or air-water mixture flow in an inclined pipe of 20 mm i.d. exhibits peculiar features of statistical properties (probability density function and power spectral density function), and the flow patterns can be classified by the statistical characteristics of differential pressure fluctuations. Thus, it appears that an objective judgement of flow pattern can be made using statistical parameters such as standard deviation and coefficients of skewness and excess.

抄録全体を表示

Pulp suspension flow is characterized by a complicated relationship between fiber flocculation and turbulence intensity. The object of this study is to clarify the influences of duct construction on the flow, and to develop a high-consistency paper-forming headbox. Optical measurements were carried out to investigate the extent of dispersion in the pulp suspension flow in a duct having various turbulence generating elements. The results are as follows : (1) Both dispersion and reflocculation were remarkable in the duct flow with an orifice which is a strong turbulence generator. (2) The saw-blade-type duct flow indicated a higher dispersion rate and mild reflocculation, and therefore is suitable as the element of a paper machine headbox. (3) The power spectra of the flocculation signal suggest that the dispersion and reflocculation of a suspension is related to the turbulence scale characterized by the flowing duct dimension.

抄録全体を表示

This paper presents the results of steady air flow tests carried out on valves and ports of a diesel engine. One of the objects of the present investigation was to determine the boundary conditions for intake and exhaust valves or ports for use in the computer program of unsteady flow research. The second object was to establish suitable data for the design of valves and ports. The experimental investigation was carried out on a cylinder head with four valves of an actual 4-cycle diesel engine (250 mm bore×320 mm stroke, Semisiamese ports, L_max=22 mm). The flow coefficients for the intake and exhaust valves or ports were obtained and plotted against pressure ratios at given lift values. The flow through valves, ports and cylinder were made visible by the aid of an oil film method and the injection of visible tracers. The attempts to improve flow characteristics were conducted by alterations in the exhaust port based on the results of flow visualization.

抄録全体を表示

A linearized analysis has been made of a two-dimensional aerodynamic characteristics of infinitely thin, perfectly flexible, possibly porous airfoils. Jacobi polynominals have been used to express pressure distributions and airfoil shapes, considering well over the leading edge singularity. It is found that airfoil shapes become stable with increasing porosity. Furthermore, the effects of porosity on lift and center of pressure are determined as functions of angle of attack and excess length of material over the chord length.

抄録全体を表示

The effect of the tip clearance between the blade tip and the fan casing on the noise generated by a diagonal flow fan is examined over a wide range of the flow rate of a fan and discussed in relation to the characteristics of flow measured at a cross section behind the fan rotor. Wake width, which controls the fan noise level, increases with decreasing the flow rate of the fan, which results in the increase in the noise level. The increase in the tip clearance causes the increase in the wake width especially in the tip region of a fan blade. A method to estimate the wake width by the known quantity of the deviation angle of the flow is proposed and verified experimentally to be useful, which enable us to estimate the noise level even in the low fan flow rate region.

抄録全体を表示

Fan noise data are categorized into source characteristic terms and a frequency response term expressing attenuation in the propagation passage. These terms are determined by sweeping the fan speed from 1000 to 4500 rpm. The fan specific speed and the impeller diameter are 42 and 460 mm. The blade number is 12. By setting the exponent in a power law relationship of sound pressure level versus frequency as 3, two factors describing the source level are determined with simple relationships to the scroll cut-off location. By selecting the distance, the noise level is possibly reduced by more than 10 dB without influencing fan performance. The frequency response if satisfactorily presented by introducing a one-dimensional linear wave model, where the key passage is blade passages selectively determined by the noise wavelength. By choosing the blade number or the inlet duct length so that a frequency of the maximum attenuation is achieved, a noise reduction of 10 dB is again expected.

抄録全体を表示

Turbine step characteristics have been found out in the course of new model development of low specific speed reversible pump-turbines. The step characteristic shows a fairly large decrease in performance curves of efficiency, discharge and power increasing in vibration and noise. Then, the effects of runner exit diameter, entrance angle and sweep back of runner vane upon this characteristic are studied, and flows near the exit of a runner are investigated by flow visualization and velocity measurement. As a result, it is clarified that, (1) the characteristic occurs in the model which has a small exit diameter and a large entrance angle, and is independent of sweep back, and (2) it is caused by the growth of reverse flow at the band side near the exit of the runner.

抄録全体を表示

The acoustic characteristics of multiple supersonic jets discharged through a multihole plate have been studied over a range of pressure ratio up to 7, for the purpose of understanding the noise suppression mechanisms and developing acoustically improved multihole valve trim. The effects of multihole configuration (hole size, number of jets, and lateral spacing between holes), are investigated in close relation to the flow behavior by far-and near-field noise measurements and schlieren photography. The acoustic efficiency η, is found to decrease in a power law correlation to the diameter of small hole, and with an increase in the number of jets at high pressure ratios. These trends are discussed from the viewpoint of the noise spectrum modification and an acoustic shielding concept. The lateral spacing mainly affects the relatively low-frequency noise radiating from the downstream merged jet.

抄録全体を表示

The authors have been studying the Pendulor type energy converter with some excellent merits : -simple, highly efficient, safe and so on, for eight years. This paper deals with the absorbing efficiency of the wave converter affected by the system's elasticity which depends on the deformation of the power members and the compressibility of oil in the power cylinder. The authors investigated the behavior of the elasticity during the operation and explicated the relationship between the efficiency and the elasticity. The deformation by the elasticity and the compressibility reduces the efficiency, as the load ratio increases. To improve this insufficiency effect, appropriate measures are needed. According to these results, several methods for improving the efficiency, including the Pendulor's hydraulic circuit and its components, were introduced.

抄録全体を表示

The present paper describes the characteristics of heat transfer and the flow situation around several spacers (a single row of several cylindrical rods) in cross flow on a heated surface in a parallel plate duct. Temperature distribution was obtained by using a thermosensitive liquid crystal film and a narrow band optical filter method without human color sensation. The local Nusselt number between the two cylindrical rods is expressed in the function of the Reynolds number, the local position and pitch of the cylindrical rods. The effects of pitch and Reynolds number affect the wake flow patterns, which are classified into three domains.

抄録全体を表示

In the present paper, the effects of heating surface roughness on characteristics of mist cooling were experimentally studied with the objective of application in a controlled cooling method in the steel industry. To do this, artificial and systematic roughnesses were inscribed on the surfaces of test plates which were made of silver and stainless steel (SUS 303). The height of surface roughness and the pitch to height ratio were varied from 3 to 91 μm and from 2 to 40 respectively. The experimental conditions of mist flow were as follows : the air velocity was V_a=20 m/s, the temperature of the water droplet T₁=21°C, and the volumetric-droplet-flow-rate was D=0.0003-0.005 m³/m²s. The experimental results show that surface roughness of some specific profile is effective in stabilizing cooling rates by mist flow.

抄録全体を表示

A three-dimensional Galerkin scheme using scalar product to obtain residual equations is developed and compared with that using vector product ; the latter is proved unsuitable. In addition to the reported two-dimensional multiple solutions (1), three-dimensional solutions are obtained corresponding to different initial conditions. It is found that the three-dimensional flow cannot be obtained unless the initial perturbation is strong enough. The structure of the three-dimensional spiral flow is clarified. At the top of the annulus, there exist secondary flow cells with closed streakines of co-axial double helix similar to that observed in inclined rectangular box. This increases maximum local and overall heat transfer rates compared with those for the simple two-dimensional unicellular flow.

抄録全体を表示

The characteristics of the steady single-phase natural circulation flow of water were experimentally studied in a short rectangular loop from the point of view of residual heat removal in a reactor core. The loop is made of Pyrex glass tubes and has a height of 2.03m and a width of 0.81m, and is equipped with a heater and three coolers. For each type of cooler arrangement, the experimental values of the circulation mass flow rate are proportional to the square root of the heat input and agree well with the calculated values from the momentum and energy balance assuming one-dimensional flow. The circulation flow rate increases with the equivalent height defined as the distance between the heater and the coolers except for the case that the cooler is set in the riser. In this case, the water also circulates at a little lower mass flow rate than others.

抄録全体を表示

The annular transition in the rod bundle is interesting because the small gaps between rods exist in the flow area. This is a very important phenomenon in the boiloff accident of nuclear reactor core. As a first attempt, the effect of small gaps in the flow area was studied by using the vertical rectangular ducts with different narrow gaps (2×100, 5×100, 10×100 mm). Based on the experimental results, the transition void fraction was defined and the transition model was proposed. The model gives a good prediction of the wide range of previous experiments including the data taken in the channels with small gaps.

抄録全体を表示

In order to examine thermophysical properties and cycle analysis for a nonazeotropic mixture of Refrigerant 22 (CHCIF₂) +Refrigerant 114 (CCIF₂CCIF₂), and equation of state for Refrigerant 22 was formulated based on the available PVT measurements taking the thermodynamic consistency into consideration . The developed equation of state, whose functional form is the same as that for Refrigerant 114 reported previously, is expressed by the reduce Helmholtz function as a function of temperature and density. The derived properties of isochoric specific heat capacity, isobaric specific heat capacity, speed of sound and vapor pressure were critically examined. The developed equation is effective for a range of temperatures from 200 to 550 K, and of pressures up to 20 MPa, which corresponds to the density range up to 1500 kg/m³. With the aid of the proposed equation, the cycle analysis on the heat pump and refrigeration systems was theoretically carried out and it became clear that Refrigerant 22 is more suitable for low temperature heat pump systems than Refrigerant 114.

抄録全体を表示

The objective of the present study is to clarify the characteristics of metal condensation. In the present study, highly accurate measurements of condensation rate, condenser surface temperature and vapour pressure are rendered possible. In addition to the techniques for these measurements, the vaccum distillation method is adapted to ensure that the results are not affected by the presence of non-condensable gases. The conformity in the vapour flow field of the present apparatus to previous analyses is also assured. Measurements are made by using potassium for a vapour temperature rang from 553 K to 633 K. It is found that the present result if in agreement with the analysis of Labuntsov and Kryukov and the condensation coefficient is very close to unity. It is also concluded that Schrage's theory predicts higher condensation rates than those in actual cases.

抄録全体を表示

The stability of vapor film in pool film boiling on a horizontal surface is considered by taking account of the effect of change in the local heat transfer rate due to the perturbation of vapor film thickness. It is shown that a vapor film below a certain critical thickness becomes stable over all disturbances with any wavelength. Assuming that the average film thickness during the actual film boiling is proportional to the foregoing critical film thickness, the heat transfer coefficient can be derived. The predicted heat transfer coefficient agrees well with the recent film boiling data obtained in a range of unprecedentedly low surface superheat by Shoji using a very large surface.

抄録全体を表示

A Liquid-fillet is formed in front of agitator-blades in the cylinder of a horizontal agitated thin-film evaporator. The thickness of the liquid-fillet varies periodically due to the gravity while the blade moves around inside the cylinder. The authors' theory reveals that usually it is in the sub-critical condition, and the maximum thickness of the fillet is observed while the blade is going up. As it approaches the critical condition the amplitude of the oscilation is increased to infinity and the phase suddenly advances 180 deg at the critical condition. Experimental data agreed well with the prediction when it was far from the critical condition. Near the critical condition the amplitude was increased and the phase was advanced about 60 deg ; that is, the transition from subcritical to super critical occurred continuously.

抄録全体を表示

The formation of the ice growth that occurs in 45° curved and straight pipes containing water flow was investigated experimentally. The experiments were carried out under various experimental conditions over the range of Reynolds numbers (based on pipe i.d.) Re=310-6760 and cooling temperature ratios of the freezing parameter. T^*(=(T_f-T_w)/(T_∞-T_f), T_f=0°C, T_w=cooling pipe temperature, T_∞=temperature of flowing water) =1.25 - 12.9. From the visualization of ice formation in the pipes, the 45° curved pipe with flowing water was first blocked by ice formation after the curved portion of the pipe, subsequently the upstream side of the curved portion, which had already had a narrow flow passage by ice formation, was frozen off and eventually freezing of the remaining water in the curved portion resulted in a pipe rupture at around the curved portion of the pipe. The transient behavior of ice formation, according to time proceeding, was expressed in the relationship between the pressure loss coefficient f and the Reynolds number for various cooling temperature ratios T^*. In addition, the critical condition of the complete ice blockage in the 45° curved and straight pipes was obtained with experimental correlation equations in terms of T^* and Re.

抄録全体を表示

Confined turbulent premixed flames stabilized by a pilot flame were investigated to elucidate the influences of the inlet condition of unburned mixture on the flame spread and mixing processes in the combustor. Detailed distributions of local velocities and its fluctuating components were measured by a laser Doppler velocimeter, and fluctuating temperatures by a compensated 25 μm thermocouple. Velocity difference across the pilot stream enhances the development of the non-combusting mixing layer, but does not augment the flame spread when the mixture has a higher velocity than the pilot burner stream and when the mixture strength is weak. The turbulence intensity in the approach streams significantly promotes the mixing and flame spread in the combustor.

抄録全体を表示

Combustion pressure vibration is generated just after the ignition in high-pressure-hydrogen-injection, hot-surface-ignition engines under almost all engine operating conditions. Some experimental investigations have been carried out to find the true nature of this interesting phenomenon. As a result, the followings are mainly found ; (1) This phenomenon probably originates in the extremely high burning velocity of hydrogen-air mixture. (2) The longer ignition delay, the stronger vibration is generated because of the steeper rise of pressure. (3) The vibration is accompanied by the engine vibration and noise, and some increase of NO_x emission and piston surface temperature. (4) It has almost no bad effect on the engine performance. (5) It can be prevented by means of premixing a certain amount of hydrogen with the intake air, which shortens the ignition delay.

抄録全体を表示

A numerical calculation supplemented by experimental work was conducted of the behavior of fuel droplets subjected to pulsating air flow in the T-branched duct of a square cross section for a better understanding of fuel supply characteristics in an intake manifold of an automotive spark ignition engine. The Lagrangian approach was adopted to derive simultaneous equations of momentum, heat and mass transfer for computing the locations, temperatures and diameters of moving droplets as a function of spatial coordinates The results were expressed primarily in terms of the location at the impact deposition of droplets relevant to the formation of liquid film on the bottom wall of the duct. It was revealed that the calculated results agreed well with the empirical data. It was also suggested that the undesirable aspects of liquid film might possibly by relieved if small droplets with high volatility and low density are infused at optimum timing into a warm air flow with a short period of pulsation.

抄録全体を表示

A performance analysis of the two-component, two-phase Stirling engine has been carried out by using a two-space model, which involves the expansion space, the compression space and the two dead spaces, based upon the assumption that the thermodynamic state point of the condensable working fluid remains within the vapor dome throughout the entire cycle. Experimental studies have also been made using two engines which operate under different system-pressure conditions : the first one is a single acting, two-piston, parallel cylinder type engine using teflon bellows as a cylinder and operates at approximately 0.1 MPa ; and the other is an opposed piston engine with piston-cylinder mechanisms which can be operated at pressures greater than 1 MPa. As a result, it is shown that a performance estimate and optimum design of the two-component, two-phase, regenerative Stirling engine by the present engine model appears feasible. In addition, the experimental results indicate that in order to realize a self-operating engine, the condensable working fluid must go through a phase change(i.e., must boil) in the engine heater. This phase change then qualifies the engine cycle as a two-component, two-phase Stirling cycle.

抄録全体を表示