The frequency of vortex shedding from a triangular cylinder in oscillatory flows in a pipe is measured. The flow oscillations are induced by means of a rotatable butterfly valve set downstream of the test section. It has been observed that the frequency of vortex shedding varies with both the oscillatory amplitude and Reynolds number under a constant oscillation frequency. These features are qualitatively the same as in the case of a circular cylinder. In order to control the vortex shedding, a triangular cylinder with a plate is used. The tendency of the Strouhal number to decrease with increasing oscillation amplitude is almost the same in the case of a circular cylinder, a triangular cylinder and a triangular cylinder with a plate. However, the triangular cylinder with a plate has the most stable characteristic of the three cylinder types in oscillatory flows. In a uniform flow, the shedding frequency varies with the distance s between the triangular cylinder and the plate, and shows a maximum value at a certain value of s.

View full abstract

The purpose of the present paper is to determine numerically the effect of the shape of the nozzle of the air-cushion pad on the flow behavior near the nozzle. In the present paper, a new technique applied to the vortex method is proposed for case of calculations of various nozzle shapes : the fundamental shape of the nozzle is mapped conformally and the panel method is applied to various shapes of the nozzle exit in the mapping plane. The numerical calculation is compared with previous experimental results. Numerical results show that the velocity distribution at the nozzle exit agrees fairly well with experimental one and the global feature of streamlines is similar to experimental ones. The pressure distribution on the surface of strips oscillates remarkably due to the existence of discrete vortices near their surface, but it is pointed out that the flow is steady, in a sense of time average, except a few early stages and the pressure gap between the cushion chamber and the outside of the jet is shown.

View full abstract

Recentry, high-pressure injection system have been used in many fuel injection pumps. This system is effectieve in improving fuel consumption, exhaust performance and so on. When the injection pressure is very high, however, cavitation erosions take place. Cavitation erosions in the fuel injection system of the diesel engine are very harmful to the performance and reliability of the engine. Hence it is difficult to adopt high-pressure injection. Cavitation phenomena depend on negative pressure. They have been studied from various up to now. However, there have been no reports on measuring negative pressure. So we trial manufactured a device to measure negative pressure exactly. We measure negative pressure at the same time as cavities by using a system to observe cavitation and a device to measure negative pressure. Then this report determines the relationship between negative pressure and cavities.

View full abstract

Transient shock wave phenomena in a liquid containing noncondensable gas bubbles are investigated experimentally and numerically. In the experiment, time evolution of the shock wave is measured using two vertical glass shock tubes with different diameters : 18 mm and 52 mm. In the numerical simulation, thermal processes inside each bubble are directly calculated by using full equations for mass, momentum and energy conservation, and the results are combined with the averaged conservation equations of the bubbly mixture to simulate the propagation of the shock wave. Initial spatial distribution of the bubbles is also taken into account. The numerical results reveal that the spatial distribution greatly affects the structure of the shock wave, such as the pressure peak of the shock front and the period of the relaxation oscillation. The present numerical results agree very well with the experimental results, when the bubbles are not distributed uniformly and are relatively concentrated along the axis of the shock tube. However, the experimental data are much different from the numerical results obtained using a uniform spatial distribution of the bubbles.

View full abstract

This paper reports numerical and analytical studies of the interaction problem between a gas interface of different densities and shock waves. The interface is formed between air and SF6 (light/heavy interface, i. e., the incident shock wave propagates from air to SF6), and there is a sinusoidal perturbation on the initial interface. The interface deformation process and the growth rate of a sinusoidal perturbation on the interface are investigated after shock wave interaction. Numerical simulation is carried out using the second-order TVD finite difference scheme, and the Euler equations are solved. Theoretical impulsive growth rate obtained using the stable velocities on the interface after shock interaction is compared with results of the numerical simulation. In the case of incident shock Mach numbers Ms < 2 and amplitudes of perturbed interface η₀k < 1, good agreement is obtained.

View full abstract

In order to obtain design data for the determination of a silencer configuration which can effectively suppress weak shock waves generated in an automobile exhaust pipe, a shock tube experiment and numerical simulation were carried out. Characteristics of the silencer were investigated by measuring pressure at the inlet and outlet of the silencer. It was found that a two-room silencer with an insertion offset pipe gives the most optimal efficiency of attenuation of shock waves. Numerical simulation indicated that the rate of change of the silencer cross-sectional area was directly proportional to the efficiency of attenuation of shock wave.

View full abstract

The overset grid method has been successfully applied to a wide range of steady-flow computations with sufficient accuracy. For unsteady compressible flows, however, there exist serious problems at the grid interface of the overset grid due to nonconservation of flow variables caused by an inherent interpolation procedure. The errors defined by the difference between the exact shock speed and the numerical one are investigated in detail in this paper when the one-dimensional shock-wave moves across the grid interface of the overset grid with various shock speeds. Numerical results show that the shock is slowed down or speeded up at the grid interface nearly in proportion to the ratio of the subgrid and main grid spacings. When relative shock speed to grid is close to zero, the shock wave cannot proceed past the grid interface. An important conclusion is that the overset grid method for unsteady compressible flows needs careful treatment at the grid interface.

View full abstract

Behavior of the turbulent boundary layer on a flat plate in a relaminarizing process with flow acceleration due to a contraction is experimentally investigated. First of all, it was confirmed that the characteristic properties in the boundary layer and the variation of the velocity distribution qualitatively coincide with the tendency of previous studies. Then, some statistical properties were measured and the turbulent structure was examined in detail. In the relaminarizing process, low amplitude fluctuating portions appear intermittently in the velocity oscillogram. In the retransitional process an irregular velocity fluctuation appears intermittently ; then it grows and always becomes turbulent. The turbulent energy dissipation decreases in the relaminarizing process and increases in the retransitional process. In the distributions of probability density function and skewness factor, motions in which the fluctuating velocity largely takes negative values are observed. In the relaminarizing process, vorticity increases in the large eddy and decreases in the small eddy. The small eddies become coarse, and the contribution from the high wave number decreases.

View full abstract

Regeneration of quasi-streamwise eddies is investigated using a data base obtained by LES of high resolution for a channel flow. It has been revealed that quasi-streamwise eddies are regenerated both in the upstream of the upstream end of a mother eddy and also in the downstream of the tail of a mother eddy. In both cases, parent and young eddies have vorticities of opposite sign and constitute a chain of eddies in the flow direction which is one of the typical coherent structures in the near-wall region. It has also been revealed that the main contribution to the generation of young eddies at the infant stage is the tilting of wall-normal vortices, and that to maintaining matured eddies is stretching.

View full abstract

A direct numerical simulation of three-dimensional isotropic turbulence of incompressible fluid was performed in order to examine an upstream-shifted interpolation (USI) method and a modified upstream-biased finite difference (UBD) method. These upstream techniques, which have been proposed by the present author, are based on an appropriate central finite difference method. In comparison with the 3rd- or 5th-order UBD method, the USI method agrees reasonably with a reliable central difference method and a spectral method. USI shows better consistency, especially for higher-order statistics of turbulence than the UBD method.

View full abstract

The effect of molecular diffusion on the dispersion of a substance in a homogeneous isotropic turbulence is examined by means of direct numerical simulation of the Navier-Stokes equation. Lagrangian statistics for a fluid particle and a substance particle with molecular diffusivity κ which was initially at the same position with the fluid particle are extracted from the Eulerian velocity field on three-dimensional grid points. It is found that a short-time behavior of the "substance velocity auto-correlation" agrees well with Saffman's theory [Saffman, P. G., J. Fluid Mech. 8, 273 (1960)]which states the existence of the interaction between the molecular diffusion and the turbulent motion. For long-time diffusion, however, the Peclet number dependence of the substance velocity auto-correlation is shown to be different from his theory probably due to the small Reynlods number used in this calculation. A modified formula suggested in this paper agrees well with our calculated result in the range of 0.556≤Pe≤55.6.

View full abstract

An investigation of turbulent structure accompanying secondary flow of the second kind in a compound open channel flow is very important in basic hydraulics and river engineering. In this study, a numerical analysis has been performed on fully developed turbulent flow in a compound open channel that is formed by the main channel and flood plain. In calculation, an algebraic stress model was adopted in order to predict Reynolds stresses precisely. Calculated results were compared with the available experimental data. It is pointed out from the experiment that the maximum velocity appears not at the surface, but rather just below it, and the secondary flow, which proceeds toward the free surface from the junction between the main channel and the flood plain, is observed. Moreover, the secondary currents in the main channel proceed downward from the free surface to the bottom wall. As a result of this characteristic phenomenon, the wall shear stress along the bottom wall of the main channel shows the maximum value at the location where the secondary flow arrived at the bottom wall. The present method can predict these characteristic features without great discrepancy. Furthermore, five of the six Reynolds stress components were compared with that of experiment.

View full abstract

A numerical analysis has been performed on developing turbulent flow in a square-sectioned 180 deg bend. The ratio of bend mean radius of curvature to hydraulic diameter is 3.35, and straight ducts 31 hydraulic diameters long extend from the inlet and outlet planes of the bend. In calculation, an algebraic stress model was adopted in order to predict Reynolds stresses precisely, and a boundary-fitted coordinate system was introduced to briefly set boundary conditions. calculated results were compared with the experimental data available. Special attention is paid to the distributions of Reynolds stresses in the square-sectioned duct and bend. The present method can predict well the measured sharp spike in streamwise velocity fluctuations near symmetry at 90 deg. Contrary to this agreement, rapid decrease of the radial velocity fluctuations at the same location in the experimental results is not predicted well by the present method. Although there are a few discrepancies, calculated results are in good agreement with the experimental data. Judging from these comparisons it is confirmed that the present method including the turbulent model is applicable for this complicated flow.

View full abstract

The development of disturbances of arbitrary shape propagating in a pipe was formulated. The disturbances were divided into two types and discussed independently ; one exhibits normal dissipation, and the other, an anomalous one. Governing equations for the two types of disturbances were derived and solved analytically. The equations include Weyl and Riemann-Liouville fractional derivatives, respectively, for the disturbance with normal dissipation and that with an anomalous one. It was found that the two types of disturbance both have a global nature which is quite different from each other especially near the wavefront.

View full abstract

We experimentally investigated the velocity characteristics of liquid lumps, such as a liquid slug, a disturbance wave and a base film wave in a vertical capillary tube. The tube diameter ranged from 6.0 mm to 0.5 mm. If the gas flow rate becomes smaller than that in the case of annular flow, the flow patterns of froth and slug flows are observed where the small waves and the large liquid lump such as liquid slug flow together with different velocities. As a definition of the mean velocity of waves or liquid slugs, we proposed a mass-weighted mean velocity which was defined by using the momentum of each wave. Based on the characteristics of velocity of liquid lumps, we proposed a new flow pattern map classifying bubble, slug, froth and annular flows, and we deduced the correlations for the two kinds of wave velocities of annular flow. They agreed well with the experimental results of wave velocity.

View full abstract

A method is proposed to practically formulate Strouhal numbers as a function of non-dimensional tube pitches combining Rosenhead's theory of Karman vortex street in finite breadth channels and the well-known characteristics of the actual free Karman vortex street. Two examples of its actual application are shown. The first is the formulation of fundamental Strouhal numbers which are considered to be those of Karman streets controlling the drag in the steady flow of tube bundles. As the numerical values of the formula were set tentatively using limited data from the literature allowing some uncertainties, the values and range of validity should be further verified reflecting future experiments. The second is the formulation of the well-known Chen's chart of Strouhal numbers measured by the resonance method.

View full abstract

In order to separate gas from gas-liquid two-phase flow, a cyclone separator is widely used, but its performance and mechanism are not yet well known. In the persent study, the performance Of a cyclone separator having a spiral casing attached upstream of a horizontal pipe is measured and its mechanism is elucidated. Using the results, several new methods are proposed to improve a performance, and are confirmed experimentally to give good gas separation efficiency.

View full abstract

Velocity vectors and concentration profiles of a coaxial swirling jet have been measured using flow visualization and image processing. It has been found that for X/D<2, where D is nozzle diameter and X is distance from the nozzle, an inner nonswirling jet and an outer swirling jet flow separately with a slight inclination caused by the cross flow, but for X/D>2 they rapidly mix with each other and start to incline. The diffusion width of a coaxial jet is nearly equal to that of a simple jet, but the inclination of a coaxial jet is smaller. These results can be applied to the design of burners or air ports in utility boiler furnaces.

View full abstract

A computational model of a neutrally buoyant stenosis deforming in fluid is presented. The fluid is assumed to be viscous and incompressible and the stenosis to be a massless and elastic boundary immersed in the fluid. Fluid motion is described on a fixed mesh (Eulerian description), and the immersed boundary is represented by a discrete collection of moving material points connected by springs (Lagrangian description). Communication between Eulerian and Lagrangian variables is required in order to apply elastic force to the fluid and to move the elastic material with the local fluid velocity. Such communications are mediated by an approximation to the Dirac delta function (immersed boundary technique). The computed results are compared with a rigid-model analysis by boundary-fitted coordinates (BFC). The comparison reveals essential differences in the high pressure gradients around the stenosis. It is explained that the elastic stenosis is deformed in the direction favorable for diminishing pressure gradients.

View full abstract

This paper presents the flow visualization of unsteady flow in a wall-attachment fluid amplifier. Velocity profiles in an enlarged model of the device during the switching process were obtained from hot-wire measurements for the various types of geometrical configurations of the device according to the three typical switching modes. Mean velocity profiles, which vary with time, revealed the transition of the relation of the jet behavior to the vortices during switching motion. Analysis of the unsteady velocity profiles clarifies the details of the mechanism of jet switching.

View full abstract

A computational method for a compressible Navier-Stokes equation is developed to study high-speed gas flow in a narrow gap. The computational method is composed of an implicit time marching scheme and 3rd-order accurate TVD scheme. The time marching scheme is based on the Beam-Warming delta form with approximate-factorization scheme and introduced in Pulliam and Chaussee's diagonal form. The TVD scheme is composed of Roe's flux splitting scheme and the MUSCL approach with van Leer's differentiable limiter. The reliability of the code is estimated on the plane Couette flow and shock reflection problem. Finally, the high-speed gas flow in the journal gas bearing is computed and the results are compared with those of the Reynolds equation in tribology.

View full abstract

This paper describes a methodology for the implementation of asynchronous parallel iterative algorithms, which is the domain decomposition method. This approach reduces the cost of synchronization that affects synchronous algorithms. As an application of the proposed approach, the algorithms based on the well-known point-SOR methods are presented in both synchronous and asynchronous forms. Results obtained using the MIMD (Multiple Instruction stream Multiple data stream) superparallel computer are given, for the model problem of the solution of a Poisson equation.

View full abstract

The flow inside straight and curved diffusers, both with inlets perpendicular or oblique to the diffuser axis, approaching channels (ducts) having positive or negative incidence effect, and as well as non-diverging exit channels (ducts), was studied by means of numerical calculations and air-flow experiments. It was clarified that the efficiencies of straight and curved diffusers, defined as the ratios of experimental pressurerise to the computed pressurerise, exhibited different characteristics. The efficiencies of straight diffusers with oblique inlets exhibited higher values in the cases of zero to negative incidences than those with perpendicular inlets. On the contrary, the efficiencies of curved diffusers with oblique inlets exhibited lower values for all cases presently investigated than those with perpendicular inlets. These differences seem to be attributed to the complex effects of diffuser curvature, inclination of inlet, and incidence.

View full abstract

A transient laminar flow in a spool valve has been studied via numerical analysis. First, steady laminar flow fields were investigated for several values of pressure difference between upstream and downstream boundaries. For a pressure difference larger than a critical value ; two solutions with distinct jet angles were obtained. Then the time-dependent numerical analysis was performed. The results for a step change of pressure difference revealed that the Step response of the flow is modeled with two time constants which are essentially the same as discussed in the previous pipe orifice analysis.

View full abstract

An experimental investigation and numerical prediction were carried out to study the effect of aspect ratio, pitch ratio and stagger angle on the loss, turning angle and behavior of secondary flow of a low-speed, straight, low-aspect-ratio compressor cascade. A finite-element full 3-D incompressible Navier-Stokes solver was used, which gives accurate secondary flow prediction in a cascade. Experimental results are presented in the form of pitch-averaged spanwise distributions of losses and turning angles. This solver has been validated by considering the experimental results. A large difference in the spanwise turning angle distribution pattern in the vicinity of the sidewall was observed between the two cascades with different stagger angles.

View full abstract

A high-speed digital video camera with a relatively low shutter speed was used to capture the movement of shock waves in transonic or supersonic cascade oscillation in pitching motion up to high frequency. Pictures are taken with constant shutter speed during some periods of blade oscillation and then rearranged according to the phase angle relative to the blade displacement. To confirm the ability and utility of this system, experiments were carried out using a controlled-oscillated linear cascade. The cascade was composed of seven flat-plate-profile blades and only the center blade of the cascade was forced to vibrate mechanically. Using this system, the movement of shock waves was analyzed in the cases of 23, 100, 150 and 200 Hz of blade frequencies at inlet Mach number of 0.65.

View full abstract

A method of flow analysis is developed by combining the formerly proposed cavitating-line theory with the well-known lifting-line one for obtaining the three dimensional characteristics of the cavitating hydrofoils. of finite span. Its main theoretical features are enumerated as follows (1) Three dimensional effects due not only to the finite hydrofoil but also to the finite cavity are reasonably taken into account in the analysis. (2) A theoretical basis for spanwise determination of the two dimensional cavity model adopted is shown clearly under the mutual interferences between the cavitating-line and lifting-line. Some notable effects of aspect ratio on the hydrofoil characteristics (lift, drag, induced drag) and cavity characteristics (effective cavitation number, cavity length, cavity termination thickness, cavity drag, induced drag) are clarified in some details through numerical examples of wide range for the super-cavitating hydrofoils of elliptic plan form.

View full abstract

In this report, it is shown that a ring which has the same axis as that of the casing of an axial fan has remarkable effects in reducing the noise level as well as improving the characteristics of the fan, such as the pressure rise and the efficiency. The effect of reducing the clearance between the wing tips and the casing is also investigated. The noise reducing effect of the ring is shown to be explained from the view point of acoustic impedance difference between the duct area and the gap between the ring and the casing. The optimum sizes in diameter and length of the ring are also clarified.

View full abstract

Detailed analysis of flows in centrifugal fans is required for improvement of efficiency and for noise reduction. However, both experimental and numerical analysis of the flow are very difficult, because the geometric configuration of the fan causes 3D flow. Even though a 2D model is applied for the analysis in regard to the flow through a part of the impeller in the rotational co-ordinate, it is not easy to determine the boundary condition at the position of the side wall of the scroll. Therefore totally 3D analysis of the whole flow field is required, but it is very time-consuming. This paper describes a simple numerical method of 3D analysis of the flow in a model of a flat-blade centrifugal fan with a scroll. In this method, the 3D flow through the partly divided portion of the fan at an arbitrary scroll angle is analyzed in rotational co-ordinates by assuming that the side wall of the scroll is replaced by the velocity boundary determined by the volume-increasing ratio at the scroll angle.

View full abstract

This paper describes the unique rotating pressure patterns which appear around an impeller under the rotating stall condition. The above pressure patterns rotate at higher velocities than the impeller rotation. Discussions are made on the source of these patterns based on the analysis of the experimental data. It becomes obvious that the interference occurs between the impeller channel flow and rotating stall cells, and waves propagating around the impeller are generated. One of the waves propagates at higher velocity than the impeller. In addition to that, it is made clear that various types of rotating stalls occur in the impeller/vaneless diffuser system.

View full abstract

Computer simulation of flow in packed wire screens of a regenerator is successfully attempted using a technique similar to the boundary element method. One-dimensional potential flow is assumed and the local helium volume increase by heating and pressure change is replaced by source or sink. Finally, fictious sources are attached to both ends, satisfying the given pressure conditions. Darcy's law is applied to obtain flow velocity from potential. This method requires no iterative processes. This fairly efficient calculation is installed in overall crycooler performance calculations, and the change of velocity distribution in the regenerator is successfully investigated in the course of the refrigeration cycle.

View full abstract

Absorption chillers are becoming attractive since they do not use Freon gases which destroy ozone layer, and can be driven by heat. Improvement of the absorber is necessary to make absorption chillers compact. In this paper, heat transfer surfaces which show maximum absorption performance are sought. Grooved pipes are known to be effective for promoting gas absorption into falling water. They are tested for water vapor absorption with four different groove depths as well as two different knurled surfaces. The pipe with the deepest groove shows enhancement of about 1.5-1.9 in the Reynolds number range of 200-500. The knurled surfaces are not effective for the absorption, but they have better wettability at a lower flow rate and furthermore for higher knurl height, absorption increases at larger flow rate. Film thickness is measured by the contact probe method. The average film thickness curve has large undulation for deeper grooves but it is flat for the shallow grooves.

View full abstract

Considering expected matter in calculation of thermodynamic properties of real fluids such as natural gas with molecular simulation, the accuracy of calculated values of thermodynamic properties and choice of potential model are discussed in this paper. The thermodynamic properties of argon and methane were calculated by the Monte Carlo method with Lennard-Jones 12-6 potential, and the calculated values were compared with the values tabulated by Rabinovich et al. and by Setzmann and Wagner, respectively. The accuracy of the density values is estimated to be better than ±1% in a wide range of temperatures and pressures. The behaviors of isothermal compressibility, vapor pressure, saturated liquid density and saturated vapor density are estimated to be reasonable, but that of isobaric expansivity is limited in a range of low densities up to the critical point.

View full abstract

Recently, Noise reduction for indoor air conditioning units has become necessary. This paper discusses : (1) Exploration at the noise source of capillary tube and outlet plenum. (2) the reason for high-and low-frequency noise of fluid flow based on R22 fluid flow pattern. (3) the method of noise reduction dependent on supercooling of refrigerant R22 by setting an additional cooler up-stream of the capillary tube. According to the above items, it is necessary for the noise reduction to control bubble contents at an outlet of the capillary tube.

View full abstract

A new method is proposed for measuring both solid-liquid contact angle and liquid surface tension. When a circular cilinder is horizontally immersed into a liquid bath and is lifted up gradually, a pair of two-dimensional liquid meniscus is formed under the side wall of the cylinder. At a certain critical height of the cylinder, the waists of the two meniscus curves contact with each other and the break-down of the meniscus occurs. Since the relation between the critical height of the cylinder H_BCr and the contact angle θ can be calculated theoretically, θ can be obtained by a simple measurement of H_BCr . The liquid surface tension also can be measured by using a grass cylinder to which most liquids attach at θ=0. The validity and usefulness of the method was expeimentally confirmed for various solid and liquid samples.

View full abstract

A low-Reynolds-number two-equation k_t-ε_t model has been constructed with the aid of direct numericai simulation (DNS) data bases on turbulent flows with different Prandtl numbers. The model adopts new length and time scales to represent various sizes of eddies in a thermal field. The predicted results for thermal fields at different Prandtl numbers were compared with the relevant DNS and experimental data. These comparisons indicate that the proposed model has sufficient capability to reproduce the DNS data, to perform highly stable computations and to analyze the heat transfer in various Prandtl number fluids.

View full abstract

A method of numerical simulation of the solidification of castings, particularly of complicated shapes, is developed. The proposed method is based on the finite volume method (FVM) and uses an irregular grid system. The irregular grid system is obtained by the boundary fit method. The coordinates lines of the grid are formed to coincide with the complicated boundaries of the castings. Nodal points are placed inside the each cell of the grid. Using the local oblique Cartesian Coordinates which are formed by connecting neighboring nodal points, the energy balance equation is directly solved by the vector analysis technique in the generalized coordinates. Some calculation results of two-dimensional sections are shown and the accuracy of the method is checked. The proposed method has the following advantages : (1) The algorithm is simple and calculation time is short comparable to the conventional finite difference method (FDM), and (2) It can handle castings of complicated shapes which could not be handled by the conventional FDM.

View full abstract

Accuracy and problems of a multiflux method are investigated. Radiation heat transfer is numerically analyzed in a simplified two-dimensional boiler furnace. From the analysis, it is shown that there are anomalous peaks in the wall heat flux distribution. Furthermore, it is found that the furnace exit gas bulk temperature predicted by the multiflux method is lower than that obtained by the Monte Carlo method. This temperature difference increases as the optical thickness increases.

View full abstract

Experiments using strip fins of actual size (0.1 mm in thickness, 1.5∼2.0 mm in width) were conducted to clarify the characteristics of forced convection heat transfer on strips in a cross flow, varying arrangement and the transverse and longitudinal pitches of strips. By measuring of the velocity and temperature distributions in a wake of the strip, the equivalent velocity and temperature were introduced for fitting their profile near the surface of the leading edge of the relevant strip. The empirical formulas for Nusselt number and Reynolds number defined by the equivalent velocity and temperature for multiple strips can be expressed by the same form as the equation for a single strip.

View full abstract

Heat transfer correlation for a rotating vertical cylinder heated with uniform heat flux in water was evaluated experimentally. The experiments covered the range of modified Grashof numbers Gγ^*_x from 2×10⁵ to 8×10¹¹ and the range of rotating Reynolds numbers Re²_ωD from 0 to 2×10⁴. The vortex motion caused by the rotation of cylinder enhanced the heat transfer and heat transfer coefficient was related to Re²_ωD (X/D)3. The experiments indicate that the rotation of a vertical cylinder has remarkable effects on the local heat transfer coefficient when the nondimensional parameter Re²_ωD(X/D)³/Gγ^*3/5_x is over 100.

View full abstract

Development of a heat sink for future integrated circuit chips has been achieved on the basis of boiling heat transfer in a vertical microchannel. The microchannel was manufactured on a 1.5 mm-thick copper plate by electric discharge machining. The performance of the three kinds of micro-channel heat sinks of which cross sections were 2, 4, and 16 mm wide and 0.5 mm high was determined by using distilled water as the working fluid. The effect of multichannels on the performance for heat sink shown experimentally. As a result, the maximum heat flux was found to be 300 W/cm² for the four-channels heat sink. Hydrodynamic instability on the vapor-liquid interface in a channel was considered for the theoretical evaluation of critical heat flux of the microchannel. Flow instability in a closed circulation loop of the heat sink was also analyzed.

View full abstract

This paper is one of the serial reports on an experimental study of heat transfer from a simulated module (31×31×7 mm³) mounted on the floor of a parallel-plate channel (20 mm high, 320 mm wide, and 650 mm long) to forced convective air flow (1-7m/s). In this report, particular attention is directed to the heat transfer from the floor which is under fluid mechanical and thermal influences of the module. Two combinations of the module and the floor are employed : a heated module on an adiabatic floor (HM/AF) and an adiabatic module on a heated floor (AM/HF). Liquid-crystal thermography is used to determine the adiabatic floor temperature (T_ad) on the HM/AF floor, and the heat transfer coefficient based on the difference between the surface temperature and T_ad on the AM/HF floor. Additional information is obtained through measurements of velocity and temperature of air in the channel. Plots of Tad and h_w show marked effects of flow development from the module and dispersion of thermal wake near the module. The distributions of air temperature and velocity defect from the undisturbed channel flow velocity show similar features to those of T_ad-isotherms.

View full abstract

We have developed a fundamental model of the human thermal system (AVA model) for the prediction of thermal comfort. The distinguishing feature of this model is a more precise description of heat transfer by blood flow (i. e., it includes arteriovenous anastomoses (AVA) of the extremities and a dual vascular network) than conventional models of human thermal systems. The following results were obtained : (1) The experimental verification under three different steady thermal conditions (22°C, 28°C and 34°C) and an unsteady thermal condition (28. 1°C+ 47. 1°C+ 28. 3°C) suggests that the AVA model can simulate body temperature profiles well. (2) The visualized results of the model predictions demonstrate that the calculated tissue temperature and blood temperature distributions are physiologically plausible.

View full abstract

Due to the change of life style and operation of industries, the gap of electric power demands between day time and night time is increasing year by year, and some of the thermal power plants are compelled to shut down or partial load operation, with lower thermal efficiency. The pumped storage power plant is the only commercialized system to mitigate such power demand gap. However, the locations suitable for a pumped storage power plant are limited. It would not be, therefore, easy to construct new pumped storage power plants in substaintial scale. The hot water storage power plant, added on the existing thermal power station, is proposed as one of the most effective technology to solve such severe situations. This paper is a brief discussion on the feasibility of the hot water storage power plant application to the utility power generation industry.

View full abstract

Performance of a forced-convection-cooling fin array in practical use should be evaluated taking its pressure-loss characteristics into account. On the other hand, the influence of the pressure loss on the cooling performance of a fin array is strongly dependent on the pressure-rise characteristics of fans/pumps driving the cooling fluid. From this poing of view, total performance of a forced-convection fin-array is analyzed in the present report considering its pressure loss and the performance of applied fan simultaneously. Based on this result, the optimum design method for forced-convection-cooling fin arrays is presented.

View full abstract

Fuel-lean premixed combustion is one of the effective methods to reduce NO_x emission from combustors. However, high-power lean combustion is difficult to realize in premixed flames, because their stability is worse compared with non-premixed flames. In the present study, a swirl flow combustor, which consists of an upright cylindrical chamber and two fuel nozzles installed tangentially on the side wall. near the bottom, was selected as a means to obtain such a stable premixed combustion. Experiments were conducted using propane as fuel. Then, it was found that this combustor has excellent performance for the stable combustion of the fuel-lean mixture, and that the stability is due to a toroidal flame with a forced recirculation flow, formed at the bottom along the inner wall of the combustor.

View full abstract

Effects of fuel mixing were investigated on flame height and flame temperature of coflow diffusion flames which were closely related to soot formation inside the flame. Four kinds of fuels with different sooting tendencies and transport properties (CH₄, C₃H₈, C₂H₄, C₂H₂) were diluted by nitrogen and used as component fuel gases. Main results obtained are as follows : (1) Flame tip temperature is a good measure to qualitatively estimate the intensity of soot formation inside the flame. (2) Addition of nitrogen to the fuel has both effects of dilution and soot suppression, and at a certain addition rate the flame tip temperature attains a maximum value due to their competitive action. (3) There exists a significant correlation between soot formation and adiabatic flame temperature and when soot formation is vigorous, factors increasing this temperature promote soot formation. (4) Effects of Lewis number and preferential diffusion should be taken into account for effective suppression of soot formation by mixing of fuels with different transport properties.

View full abstract

We discuss in detail the various thermal diffusion mechanisms on the basis of spectral analyses of turbulent diffusion flames. Spatially and time-resolved structures of turbulent diffusion flames were obtained by two-point laser Rayleigh spectroscopy (LRS) which did not directly interfere with the combustion media during measurement. The turbulent diffusion flame structures were divided into 4 regions based on the characteristics of their spectral analysis of time-dependent temperature signals. In particular, in order to discuss the macroscopic thermal diffusion mechanism, the coherent function from cross and power spectral functions at Regions I∼IV were analyzed. From these analyses, the following diffusion characteristics at each region were revealed. Region I : three-dimensional diffusion mechanism (x-, γ-, z- axes), Region II : one-dimensional diffusion mechanism (z-axis), Region III : two-dimensional diffusion mechanism (x-, z- axes), Region IV : no diffusion mechanism.

View full abstract

An experimental study was conducted on the behavior of the fuel spray and the nozzle needle in a multihole valve cut orifice (VCO) nozzle for diesel engines. The measurements of time lag for fuel discharge among individual holes and the photographic observation of fuel sprays at an arbitrary instant after the start of fuel discharge were carried out using a video camera and a microflash. The magnified image of the top portion of the nozzle needle illuminated by an argon laser and two He-Ne lasers was observed using a high-speed video camera through a window of diameter 0. 6 mm installed at the nozzle tip. The results showed that the fuel spray was classified into solid cone, hollow cone and mixed types, and that the radial displacement of the nozzle needle was straight and toward the opposite side of the hole from which the earliest fuel spray appeared among the five holes.

View full abstract

The atomization characteristics and the mechanism of spray flashing were experimentally and theoretically inivestigated. Detailed experiments have been conducted with regard to the flashing phenomena in a decompressed capsule, using glass nozzles with eight values of length and inner diameter. The inception condition of flashing, and the spray characteristics, such as droplet temperature, diameter, number density, and mass flow rate, were clarified in relation to superheat degree and nozzle dimensions. A new model for spray flashing evaporation is proposed, and analyses were performed on the basis of the proposed model. By comparison with analytical predictions and experimental results, the mechanism of spray flashing is clarified.

View full abstract