An approximate solution of the standard k-ε model is found for two-dimensional turbulent channel flow. Here, k is the turbulent energy, and ε the dissipation rate of kinetic turbulent energy. Using this solution, five constants used in the model are all determined reasonably with the aid of experimental data. If an approximated solution with the logarithmic law as the leading term is sought, the standard k-ε model is not applicable since the second-order solution has divergent terms.

View full abstract

A revised k-ε model is proposed. Here, k is the turbulent energy, and ε the dissipation rate of kinetic turbulent energy. The proposed model is valid and is transferable to other two-equation models. This model has a diffusion term of ε in the k equation, which does not exist in the standard k-ε model. Similarly, the diffusion term of k is newly introduced into the ε equation. The constants used in the model are estimated using approximate solutions. A numerical example, which shows good agreement with the experimental results, including eddy viscosity distribution, is shown in the paper.

View full abstract

Recently, longitudinal ventilation systems tend to be adopted even in long road tunnels because of low system costs. In this system, ventilation air flows along the tunnel axis. When a fire breaks out, the smoke diffuses into the tunnel by the ventilating flow. Thus, analyses of the air flow and the smoke behavior in a tunnel on fire are important for ventilation design. Unsteady and three dimensional flow were solved by the SIMPLE algorithm based on the finite difference method. The boundary condition was extended to the curvilinear tunnel wall. As a first step, the fire was simulated by a heated plate. Experiments and calculations shown above were carried out for a model tunnel with the heated plate. It was found that these calculation results showed good agreement with the experimental data.

View full abstract

In this paper, smoke movement was considered in addition to the air flow calculation in longitudinal tunel flow. Furthermore, fire was simulated by an oil-burned plate which is nearer to an actual fire. Based on the verified calculation for the air flow in our previous paper, air flow and smoke concentration were calculated simultaneonsly. The numerical calculations are carried out for some heat release and smoke evacuation rates. It is found that smoke movements have two modes according to the heat release rates. Also, we show that a conventional 1-D smoke simulation with an isothermal assumption cannot predict the smoke behavior exactly, especially in the case of a large heat release rate.

View full abstract

In this paper, a finite volume method for solving the three-dimensional incompressible time-averaged Navier-Stokes equations in a general curvilinear coordinate system has been developed and evaluated. The main solution algorithm is the extension of the SIMPLE algorithm to make a general curvilinear formulation of the governing equations, leaving the dependent variables in their original orthogonal coordinates. The standard k-ε turbulence model is used for the closure of the Reynolds equations. This numerical procedure is applied to the calculations of turbulent flows over a surface-mounted cube with symmetry orientations (θ=90°, 45°) and a Bell-shaped hill. The results are compared with the available experimental data. The predicted flow properties show good correspondence with the experimental ones, but there are some discrepancies.

View full abstract

To understand the transport process over a rough plate, an experimental investigation was carried out concerning a single circular cylinder which was set upright on a flat plate. For the visualization of the near wake of the cylinder, dye was injected into the uniform water flow from the front stagnation point of the standing circular cylinder on the plate. The diffusion of the dye was visually observed. From this experiment horseshoe-shaped 3 dimensional vortices were observed in a fairly long range downstream of the cylinder. To determine the 3 dimensional velocity distribution around the vortices, 3 simultaneous velocity components were measured using the hydrogen bubble technique. From these experimental results, a structure of the near wake flow was obtained.

View full abstract

To understand the transport process on a rough flat plate, an experimental investigation was carried out concerning a single circular cylinder which was set upright on a flat plate. From a flow visualization experiment, horseshoeshaped three-dimensional vortices were observed over a fairly long range downstream of the cylinder. Using the experimental results of the three-dimensional velocity distribution around the vortices, the velocity was expressed as the sum of the time-averaged and the fluctuation components. The structure of the fluctuating flow was obtained and it was clarified that the fluctuating component is composed of a row of boomerang-shaped or hairpin-shaped vortices which are much different from the horseshoe-shaped vortices. This vortices were clarified theoretically to induce Reynolds stress effectively in all directions.

View full abstract

Momentum integral equations for the turbulent flow through curved pipes have been set up in consideration of a viscoelasticity of solutions and solved approximately. The resulting coefficient of pipe friction has been compared with experimental results and presented in a form suitable for engineering use. In addition, the boundary thickness, the velocity distribution of the secondary flow and the pressure difference between the inner and the outer side have been given respectively.

View full abstract

The generation of a disturbance wave on a liquid film flowing concurrently with an up-or a down-wards gas flow in a vertical circular pipe is investigated both theoretically and experimentally and compared with that in a horizontal duct. It is made clear that the disturbance wave is generated when the velocity of liquid particles exceeds that of interfacial waves. This is similar to the case of a horizontal duct. The boundary of the existence of the disturbance wave, however, depends strongly on the gravitational force, i.e., the orientation of the pipe line and the direction of the flow in a low gas flow rate region. This is because the gravitational force affects the wave velocity but not the mean liquid flow in a horizontal line, and vice versa in a vertical line ; and also because an annular film cannot be formed in the vertical upward flow if the gas velocity is low, i.e., the interfacial shear force is small.

View full abstract

The conservation equations based on the drift flux model are derived for one-dimensional gas-liquid-solid three-phase flow. It is clearly shown that the solid particles added to gas-liquid two-phase flow make the flow situation more complex. Their effects on the three-phase flow situation are expressed in terms of the drift velocities of the solid in the gas and the liquid phase.

View full abstract

In order to elucidate the atomizing mechanism of Coal Water Mixture (CWM), several experiments on the atomization of CWM will be carried out with twin fluid atomizers. The effect of coal sizes on the formation of CWM spray droplets is investigated in detail. It is also found that large droplets of CWM spray consist of numerous fine coal particles and that large coal particles do not form "Nucli" of large spray droplets. From these experimental results, the breakup mechanism of CWM uneer high shear stress by gas-blasting at the exitport of the atomizer is proposed.

View full abstract

In order to find the optimum atomizing condition for combustion efficiency, a combustion evaluation concept is introduced on the basis of the atomizing characteristics, and the dimensionless parameter Stk (Stokes Number) is used to analyze the trajectories of CWM spray droplets. Based on the these concepts, the Combustion Index CI is defined. The larger the CI values, the better the combustion. It is also found that the atomizing condition at a W_a/W_l of 0.26 is optimum for combustion efficiency. When W_a/W_l>0.26 blow-off occurs due to the increased injecting speed. On the other hand, the burn-out times become much longer because of worse atomization (in the condition of W_a/W_l<0.26). This is a very useful prediction method to evaluate the performance of an atomizer and the combustibility of CWM with various properties. The prediction results above are verified in a small-scale test furnace.

View full abstract

As a first step to explain non-ideal behaviour of magnetic fluids in a magnetic field, an analysis of particle aggregation of the magnetic fluid in a uniform magnetic field has been carried out. The average number of particles per cluster is obtained numerically by evaluation of the partition function using the Monte Carlo method. It is shown that there is no aggregation for water-based magnetic fluid in the range of magnetic field, H≤10⁵[A/m] if only the magnetic force between particles is taken into account and no interaction between clusters is assumed.

View full abstract

Basic theoretical and experimental studies are made on the behaviour of a magnetic fluid plug oscillating in a straight pipe with an exciting periodic magnetic field as a simple example of the magnetic fluid plug actuators. Assuming that a pipe friction coefficient in a magnetic field with oscillating motion is constant, the theoretical analysis shows good agreement with the experimental results. The maximum amplitude of vibration increases almost linearly with fluid plug length within the experimental range. However, the maximum fluid length is restricted by the effective region of the magnetic field, which also restricts the maximum amplitude of vibration.

View full abstract

An experimental investigation has been carried out on the sound resonance associated with the circular pipe of finite length, which is positioned coaxially downstream of a round free jet. Self-sustained oscillations arises due to the impingement of the free shear layer on the pipe's leading edge, and this results in the periodic production of discrete ring vortices in the jet. The predominant frequency of oscillation is found to vary intrinsically with the position of the pipe's leading edge as in the case of the circular ring, in which several modal stages of oscillation have been observed previously. However, when this intrinsic predominant frequency approaches the first or higher eigen-frequencies of the pipe, sound resonance between these two frequency components occurs. Thus, the predominant frequency and the distance between the nozzle exit and the pipe's leading edge, when non-dimensionalized by the initial momentum thickness of the shear layer and the nozzle-exit velocity, should be the only parameters describing these sound resonance phenomena. In addition, the sound power level of the resonance increases nearly in proportion to the sixth power of the jet velocity.

View full abstract

Using the airfoil NACA 64 A 006 which shows the flow pattern of typical thin airfoil stall, the relations between airfoil oscillation and the flow behavior around the airfoil are made clear for pitching oscillation around the midchord in the range where thee mean angle of attack is around the stall angle. Flutter characteristics are also made clear for this range of attack angle.

View full abstract

It is believed that a helmholtz resonator is a good device for reducing the noise level of a discrete frequency noise, and that the frequency of the resonator can be estimated by a simple theoretical formula. Actually, however, we cannot deny the possibility of an unexpected increase in the noise level due to the resonance between a traveling tone and the resonator. Also, there are few papers which treat of the actual design method of a resonator, especially the dimension of a resonator in relation to the size of a main duct conveying a tone to be reduced. In the present experiment, the theoretically estimated frequency is examined ad to whether its noise level is reduced at that frequency by the resonator or not. The experimental data which are useful for designing a resonator with an optimum geometry are also presented.

View full abstract

An iterative solution method has been developed for calculating flows in two-dimensional centrifugal pump impellers. The method was formed by combining an integral calculation method for a three-dimensional boundary layer on the whole channel wall with a surface singularity method for a two-dimensional inviscid core-flow. In the inviscid analysis, the blade shape and the area of flow passage were modified by calculated displacement thickness of boundary layer, and then a finite number of vortices were distributed on the modified blade surfaces with the Kutta condition at the outlet edge on the suction surface. This method was applied to flows in two impellers with different blade thicknesses. Comparing the results of calculated and measured flows, both were qualitatively in good agreement, including an effect of the outlet blade surface with constant radius, which showed that relative velocity near the pressure surface was faster than one near the suction surface.

View full abstract

The speed of a water turbine is regulated by a governor and a typical feed back control system is composed of the governor, the turbine and its load. The paper deals with the characteristics of a tachometric governor with temporary speed droop. First the dynamic properties of the governing system and the controlled system are expressed in the form of a block diagram, and a few considerations such as harmonic response, step response and self-excited oscillations of the system are conducted using an electronic analogue computer. Secondly, concerning the stability criteria of a surge tank system in a hydraulic power station, theoretical studies are pursued and compared with the classical Thoma formula which is based on the governing equation of ηQH=constant. The temporary speed droop equipped in the governor has an effective function to stabilize the surge tank system, and the paper presents a new formula which gives a smaller net area of the surge tank than that obtained from the Thoma formula.

View full abstract

The atmospheric Ar plasma freejets, generated by an electric arc discharge, are visualized and measured by the Moire-Schlieren method to obtain the details of their high temperature fields. Electron temperatures and electron number densities are also deduced from a combined method of the absolute intensity measurements of Ar spectral lines and the collisional-radiative process theory. From the experiments performed under the conditions of arc currents of 100-200 A and nozzle diameters of 3-4 mm, the following are found : (i) The gas temperatures are 1000-5000 K and the electron temperatures are about 8500 K. (ii) The distributions of gas temperature for the conditions show a similar form. (iii) The freejets are thermally in nonequilibrium, though the plasmas are fairly dense.

View full abstract

Numerical computations were carried out to determine the mass transfer for the absorption of low pressure water vapor into an aqueous lithium bromide film flowing down a constant temperature vertical flat plate. The film is assumed to be laminar and the velocity profile fully developed. In the current study, the selected operating parameters are expected to be present in the absorbent of an absorption refrigerating machine. The results are summarized as follows : (1) The total absorption mass flow rate over a certain length of the plate takes a maximum value at a certain entrance film thickness. (2) With an increase in the system pressure, entrance concentration and Lewis number, the total absorption mass flow rate is augmented. (3) The total absorption mass flow rate under conditions where variations of the solution properties occur is slightly lower than that of the constant properties.

View full abstract

The heat flux from the human skin to the environment was measured with heat flux transducers placed all over the skin surface of a male subject .The total heat flux from the unclothed subject was 113 W at the room temperature of 20 C, and 84.9 W from the subject clothed in a loosely fitting training shirt. The heat flux distribution was found to be approximately symmetrical in terms of the centerline of a trunk without clothes. Its correspondence with the skin surface temperature was obtained, approximately. A constant temperature bath made of a stainless steel cylinder was constructed in the shape of a human body. The total heat flux measured by the previous transducers agreed fairly well with the total electric energy supplied to a constant temperature bath of this model.

View full abstract

A fast numerical simulation procedure is proposed to simulate transient natural convection heat transfer in a spherical cavity, where the fluid is initially at rest and then the wall is subject to a step change of temperature. By decomposing temperature, vorticity multiplied by a radius, and a function related to an axisymmetric stream function into Fourier series of latitude, the equation of vorticity transportation, the equation of energy, and the equation of continuity can be separated into Fourier components of the same kind, so that the system of equations constructed can numerically be integrated independently of each other for different components, using implicit schemes. As a result, this procedure can predict not only the time required to reduce the temperature difference at the center to one half of the initial temperature difference but also the oscillation of the temperature at the center in the case of a high Grashof number flow.

View full abstract

This paper dealt with the numerical calculations of the heat transfer of a tube partially heated on its circumference, considering two-dimensional heat conduction within the wall. The contribution of the unheated region of the tube wall to heat transfer of the heated region was explained by the term of "fin efficiency of psuedo-fin", it was clarified that the fin efficiency of the unheated region was little affected by the temperature difference between the inner and outer surfaces of the wall, and could be approximated by the fin efficiency of a rectangular fin. Both the circumferential and radial heat conductions within the wall affected the temperature difference between the inner and outer surfaces of the heated region ; however, the effect of the temperature difference on the circumferentially average Nusselt number could be obtained by using the analytical solution of radially one-dimensional heat conduction. Using these results, a diagram showing the effect of wall conduction on heat transfer, which is useful for designing the circumferentially nonuniformly heated coolant passages, was obtained.

View full abstract

The tubes in this study have internal ribs and porous structures to enhance boiling of refrigerants on external surface. The ribs are provided by the application of a roller on the external surface, so that a part of the external surface is recessed with a certain pitch. Consequently, the tube-side enhancement reduces the enhanced area on the shell-side surface due to the inability to machine the porous structure on the recesses. Based on the experimental data, an analysis is performed so as to find the optimum pitch for the provision of internal ribs, where the enhancements on the tube-side and the shell-side are best utilized.

View full abstract

On heated horizontal tubes (18mm O.D.) having the porous surface, a falling film of R-11 (flow rate 0.008-0.091 kg/(m.s)) was evaporated at atmospheric pressure. The heat transfer coefficient of liquid film evaporation is higher than that of pool boiling when the wall superheat is small ; however, the enhancement ratio reduces on the surface structure where the pool boiling heat transfer coefficient is already high. The appearance of dry spots on the porous surface was observed only at a small flow rate close to the threshold value for complete dry-out. Another advantage brought by the porous surface is the virtual elimination of hysteresis phenomena.

View full abstract

The pressure drop of subcooled flow boiling in a narrow tube was investigated experimentally using water as a coolant. Experiments were conducted at nearly ambient pressure under the conditions : tube inside diameter : 1 and 3 mm, tube length : 10∼100 mm, and water mass velocity : 7000∼20000 kg/m²s. The friction pressure drop ratio of subcooled flow boiling to non-heating water flow was examined by increasing the heat flux. The ratio begins to increase at the heat flux proposed by the Saha-Zuber correlation that the bubble begins to detach for 3 mm inside diameter tube, though the heat flux is higher than the Saha-Zuber heat flux for 1 mm tube. The ratio was further compared with the Bergles-Dormer correlation. The two phase friction multiplier of subcooled flow boiling was examined assuming the Ahmad void fraction and applied to the Lockhart-Martinelli (L-M) correlation. The abnormarity of the subcooled flow boiling in the case of 1mm inside diameter tube was confirmed in these discussions.

View full abstract

The measured results of dropwise condensation curves for three organic vapors on a trilauryl trithiophosphite adsorbed copper surface were shown in the former report. Two different types of drop to film transition, continuous and discontinuous types, were observed. The variations in the condensation situations measured simultaneously by high speed and video cameras are shown for both transition types. Then, the contact angles of three substances during condensation are compared. The variations in the condensing situations in all cases measured are generalized and tabulated. Furthermore, the mechanisms of the transition are discussed. It is certified that the contact angle and the viscosity of the condensate are the controlling factors in the case of continuous type transition, and the shape of the discontinuous type curve is similar to the continuous type curve, except for a discontinuous part between the jumping and film points.

View full abstract

Different modes of transition from dropwise to film condensation were investigated analytically and experimentally. Two types of shift modes, the continuous mode and the discontinuous mode, were measured and the jumping phenomena were observed. The surfaces for dropwise condensation used were of two kinds : a ptfe-coated and a trilauryl trithiophosphite adsorbed surface. The dominant heat transfer mechanism was discussed and the factors determining the shift modes were analyzed. Close agreement between the measured and the analytical values were obtained. It was clarified that the phenomena determining the shift modes were controlled by the heat flux gradient in the logarithmic scale and by the conductance ratio of the vapor-side heat transfer and the overall cooling-side conductance.

View full abstract

An apparatus for measuring local humidity with optical fibers is applied to the measurement of local humidity change near a wetted surface, which is suddenly turned downward at the upper side of a dry air chamber. Measured values of water-vapor concentration and its change agree approximately with analytical ones. For more accurate measurement and a more compact construction, a double beams system was constructed and calibration tests and local humidity measurement were carried out. It was clarified that the calibration results are the same as with the one beam system, which has the same optical path length, and that this system is also applicable to measurement of local humidity change with double accuracy.

View full abstract

The effects of inclined angle, number of layers and the evaporator length on heat transfer limit were determined by using a planar heat pipe. The working fluid was distilled water and the wick was a 200 mesh screen. An analysis was performed to predict the heat transfer limit of the heat pipe, assuming a linear relationship between the receding of the liquid-vapor interface and the capillary pressure. The prediction based on such a simple model shows good agreement with the experimental results.

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 DI diesel engine with an M-combustion system. In this report, the experiment was performed to investigate the influences of fuel injection direction, which is cosidered to affect the quantity of deposited fuel on the combustion chamber surface and the influences of combustion chamber surface temperature on the diesel combustion. It is found that the combustion chamber surface temperature greatly affects the combustion of deposited fuel on the surface, and the diesel combustion is controled with the injection direction and the surface temperature of the combustion chamber.

View full abstract

Heat insulated pistons were applied to a small size DI diesel engine with the M-combustion system. In this study, two heat insulated pistons were prepared ; one of them was insulated by medium heat, and another by high heat. Experiments on measurements of the combustion chamber surface terperature of each piston, the engine performances and combustion analysis were performed. The test results of each piston are compared and the influences of the piston surface temperature on the engine performances and the combustion characteristics are discussed.

View full abstract

An adaptive identifier using a least squares method was tested as the first step foward realizing an adaptive control of internal combustion engines, which demand such a method because of their unique characteristics of non-linearity and time-variability. The selected parameters of the identification are the static characteristic between the value of the throttle angle and the value of indicated work, and its time constant. As a result of testing, the following conclutions were obtained. (1) The precision of the identification seems to be quite good. (2) Time length of the identification for one linearized region of the engine operating condition is less than 50 s for even the worst case of idling operation. (3) Therefore, there are good possibilities for the next step of adaptive engine control.

View full abstract