For a sufficiently wider range in low cavitation numbers of technical interest, we mainly observed "pseudo-supercavitation" around a typical supercavitating hydrofoil, whose cavity was filled with a bubble cloud. As the first step to find good performance hydrofoils, in this paper, supercavitation performances were systematically tested for three typical high-speed hydrofoils with pseudo-Kirchhoff noses in the pseudo-supercavitating region. The hydrofoils have no weak point in the leading edge strength, and one of then shows fairly high lift-drag-ratio compared with other existing hydrofoils.

View full abstract

The vibratory-erosion-tests on a typical material of 304 stainless steel were carried out. We carefully observed the erosion patterns and the surface-roughness-aspects with respect to test times, under a specified condition of a uniform nuclei-size-distribution, by a scanning electron microscope and by a profilometer. Both surface topography and surface profile divided the eroded surfaces into three stages. In the first one, the surface was plastically deformed and the roughness rapidly increased. In the second stage, the surface became more hardened, so the crack initiated and propagated, and after that, local removal of material occured. Also, the roughness doubled and slightly increased. In the third stage, almost all the virgin surface was removed and the surface roughness doubled, again. Both the skewness and the kurtosis was calculated. It was found that the skewness has a positive and a negative value before and after the material removal, respectively. The kurtosis had a positive value for all test times.

View full abstract

The cavitation bubble collapse pressures (impact loads) in a venturi facility were measured with a piezoelectric detector and were analyzed statistically. The generating number of impact pulses is very small, but larger impact pulses occur compared with the results obtained by a vibratory facility. When the distributions of impact loads are compared to the erosion pit sizes on the surface of mild steel and copper, a lower critical impact load which forms impulsively a pit is almost the same regardless of cavitation conditions. The impact energy that contributes to the pit initiation, which is calculated by integrating the pulse heights above the critical impact pulse, shows maximum value at a given cavitation number (σ=0.68) and is proportional to the 4.3 and 6.3 th power of the fluid velocity for copper and mild steel, respectively.

View full abstract

Fluctuation of the boundary layer grown on the end plate of a duct is studied experimentally for two inlet conditions: at the inlet of the curved duct, one is uniform flow which fluctuates in its magnitude sinusoidally and the other is the flow which always has a fully developed duct flow profile but which fluctuates in its magnitude. For both cases, the cross sectional mean velocity. U_m, fluctuates with the Womersley number of 44.3, and the ratio of the amplitude of U_m to its time mean value, U^^-_m, is 0.38 and the Dean number based on U^^-_m is 1.28×10⁴. It is found that (1) at the cross section of the duct turning angle of 60°, the fluctuation of the momentum thickness of main flow velocity, U_φ shows fairly good agreement with the 1/n power law but flow fluctuation has a great effect on the distribution of the cross-flow velocity component, U_r, very close to the wall.

View full abstract

The effect of the fourth-order truncation error produced by discretizing the convection term was studied on nonlinear characteristics of the solution for the instability of laminar flow and for fully developed turbulent flow. (1) Bifurcation was calculated by the third-order upwind difference method for the Kuramoto-Sivashinsky equation representing the instability of laminar flow. The fourth-order truncation error, however, affected the bifurcation characteristics, and error was introduced for the fourth-mode solution by the Kawamura scheme. (2) Energy spectrum calculated by the third-order upwind difference method for fully developed homogeneous turbulent flow in two dimensions decayed rapidly in a high wave number region in effect of the fourth-order truncation error.

View full abstract

A second-moment closure applicable right up to a wall, which was proposed and successfully applied to the boundary layers in pressure gradients in previous papers by the author, is employed here to calculate the boundary layer on a convex wall followed by a flat wall. It is shown that the numerical solutions are generally in good agreement with the experimental data. In particular, the model reproduces a slow recovery of the shear stress on the flay wall downstream, a feature not captured by most of the previous predictions. The improvement is traced to a particular generation term in the ε transport equation that augments the dissipation rate εin the recovery flow.

View full abstract

A unique method for an experimental investigation on coherent structures in a turbulent boundary layer is presented. A large-scale coherent vortex was artificially induced by injecting a pair of small pulsative swirling jets into the bottom of a fully-developed turbulent boundary layer. The quantitative inspection for the flow field around the vortex was made on the basis of a simplified conditional sampling method through which the artificial vortex could be easily detected at any fixed location and time. The artificial vortex is proven to have the sculpture of a horse-shoe composed of a pair of vortex tubes inclined downstream. The vortex tubes are joined with each other at the top at about Y/δ=1.2. Its structure resembles that of the natural large-scale coherent 'bulge'. The streamwise, lateral, and spanwise scales of the artificial horse-shoe vortex are about 5δ, 1.4δ, and 2δ, respectively.

View full abstract

LES (large eddy simulation) has been verified in simple flow fields. There are few examples in engineering applications, partly because the treatment of the wall boundary condition in complex flow has not been developed. In this paper, a new approach to imposing the wall boundary condition by a two-layer model is developed in LES. This method proved to be valid for application to turbulent channel flow.

View full abstract

A few factors which influence the Taylor-Gortler vortex flow are examined in a spherical Couette system with only the inner sphere rotating. The effects of eccentricity, through-flow, acceleration and history of rotation are considered. When the rotational Reynolds number increases quasistatically, the eccentricity does not remarkable influence flow modes in the range of small eccentricity. Through-flow makes the rotational flow stable on Taylor instability. Although the differences between flow modes in the absence and the presence of through-flow can be seen, the flow modes can be uniquely determined by the similarity parameters, independent of the direction of the through-flow (positive or negative through-flow), In the absence of through-flow, however, the flow modes and the patterns of travelling azimuthal waves depend on the acceleration rate or history of rotation, and can not be uniquely determined by the similarity parameters.

View full abstract

A novel method to measure the conditional turbulent zone averages in a fully developed jet is demonstrated. In conventional methods to study the turbulent intermittency, hot wire signals are processed to detect the turbulent zone and then the conditional average velocities are obtained. But this signal processing is cumbersome and difficult to set the threshold of turbulence without ambiguity. In a jet flow, the surrounding nonturbulent fluid is entrained into the turbulent zone, but the reverse process never occurs. Therefore small particles seeded in the nozzle flow remain only in the turbulent zone. By means of a laser doppler anemometer (LDA), the velocities of these particle are measured. Therefore, the turbulent zone averages are obtained directly. This paper describes and experimental comparison of this method with the conventional one. A circular air jet seeded with DOP particles was blown into a seedless coflowing stream. Then the mean velocities and the turbulent intensities were measured by LDA processed with a frequency counter, and they proved to coincide with the values secured by means of conditional averaging of HWA signals.

View full abstract

Mean flow properties of a d-type rough wall boundary layer in a transitionally and a fully rough regime have been examined experimentally. The mean velocity, the turbulent intensities of three components and the Reynolds shear stress were measured over the momentum thickness Reynolds number R_θ=700∼5140. Even in a transitionally rough regime, the logarithmic layer is confirmed and the value of the karman constant is the same value as that of the smooth wall flow. However, the velocity defect law has been found to be dependent on R_θ. The variation of the wake parameter Π with R_θ is significantly different from that of the smooth wall flow. The u_rms/u_τ profiles normalized with the inner variables indicate reasonable similarity for R_θ≥3200 in the inner layer, while these profiles normalized with the outer variables are similar in the outer region. The v_rms/u_τ and the w_rms/u_τ profiles do not exhibit similarity in their dependence on R_θ. The Reynolds shear stress -<uv>^^-/u²_τ profiles show the constant stress layer over the complete Reynolds number range, but these constant values depend on the Reynolds number.

View full abstract

The turbulent structure of a liquid-solids two-phase jet has been experimentally investigated to examine a model application for the particle kinetic eddy viscosity in a turbulent two-phase flow. The confined jet flowed in the gravitational direction at the Reynolds number of the inner jet 1.1×10⁴. Spherical glass particles of 486 μm were loaded into the inner tube at a volume fraction of 1 %. The flow fields were measured by a laser Doppler velocimeter, with a particle size discrimination. In the developing region of the shear layer, measurements revealed a a decrease in the turbulent kinetic energy and the turbulent shear stress of liquid phase involving solid particles. The model based on the characteristic time scale ratio of turbulent two-phase flow confirmed a good agreement with experimental results.

View full abstract

The purpose of this study is to investigate unsteady characteristics of the annular impinging jet. Water and air were employed as working fluids. Two pressure transducers made of a semiconductor were installed, one being at the center of the disk placed downstream end of a inner tube and the other on an impinged flat plate. Cross-spectrum and cross-correlation functions of pressures were measured by varying a distance H between the annular nozzle and the impinged flat plate. It was found that there were four types of distinguishable frequencies for pressure, according to a value of H. It was ascertained that fluctuations with the large scale of the annular impinging jet had essentially two modes and the frequencies of these changed linearly with an issuing velocity alone of the annular jet. Some discussion of a cause of their generation took place.

View full abstract

A two- or three-dimensional jet along a curved wall, by the Coanda effect, has been investigated by many researchers. In spite of many studies on this Coanda jet flow, the flow characteristics have not been completely understood. The authors clarified the velocity and pressure distributions for a two-dimensional jet along a circular cylinder by a theoretical analysis using an integral momentum equation and an experimental analysis. This paper clarifies the effects of nozzle width experimentally on the flow characteristics, velocity and pressure distributions, position of separation, decay of jet centerline velocity, change of jet width, etc.

View full abstract

This paper deals with the acceleration of a steel ball in an annular jet and the impingement on the solid wall. An experiment is carried out using a pressurized water jet from a nozzle with an inner pipe. A steel ball is introduced into the nozzle with air through the inner pipe, and then accelerated by the annular water jet. The results show that the annular water jet with air supply from the inner pipe has a similar velocity distribution along the axial direction as that of normal jet without the inner pipe, which increases the impact velocity of the steel ball against the target plate. The velocity of the steel ball calculated by the equation of motion, taking into account the Basset's force, agrees well with the experimental data.

View full abstract

The flow around a circular cylinder oscillating transversely in a uniform flow is studied to clarify the influence of the near wake of the cylinder on the stagnation flow by flow visualization techniques. Under a constant amplitude of the oscillating velocity of the cylinder, as the frequency of the oscillating cylinder increases the maximum displacement angle of the stagnation point exceeds the maximum angle of the incidence of the main flow to the cylinder. At the frequency of the cylinder synchronized with the shedding frequency of the vortex the displacement of the stagnation point is closely related to the structure of the vortices near the cylinder. At the other frequency of the oscillating cylinder it is rather closely related to the configuration of the separated shear layers. The displacements of the stagnation point can be estimated accurately with the potential flow model with a group of vortex filaments corresponding to the real flow in the near wake of the cylinder.

View full abstract

Experimental investigations of the effect of the relative roughness on the transition of flow were examined in the Reynolds numbers range of 5×10⁴≤Re≤10⁷, using 8 cylinders with different relative surface roughnesses. Flow phenomena on the cylindrical surface are classified in four types irrespective of surface roughness. They are: subcritical, transitional, supercritical and transcritical regions. These regions are classified from the variations in the drag coefficient and Strouhal number with Reynolds number. The Reynolds number range at the interface of these flow regions, the variation procedure and the degree of variation, depend on the relative roughness. For example, the Reynolds number at the interface between the subcritical and transitional ranges can be properly determined as a transitional point of relation between the drag coefficient and the base pressure coefficient. With an increase in the surface roughness, the interface points shift into the lower Reynolds number range. The effect of the relative roughness on these flow phenomena is classified and discussions are also made.

View full abstract

In the previous paper related to the H-II rocket, aerodynamic forces acting on three circular cylinders having different diameters closely arranged in line were clarified. In this paper, the fluid flow characteristics around the three cylinders were investigated as follows: Flow visualizations were carried out by using a smoke tunnel and the oil-film method, and the rms fluctuating pressure distribution and velocity of jet from the gap of the cylinders were measured. For an attack angle α=60°70°, the shear layer separated from the main cylinder reattaches and adheres to the rear surface of the cylinder beyond 210° by a coanda effect. From the characteristics of the wake of the cylinders at α≥60°, the three cylinders can be assumed to be a rectangular cylinder with round corners and two slits.

View full abstract

Fundamental studies on the control of the flow around a circular cylinder were conducted by a new method. It was found that a forced reattachment of the separated shear layer from the cylinder was realized by setting up a small cylinder in the shear layer near the main cylinder. The diameters of the cylinders were 40 and 2 mm, respectively, and the Reynolds number ranged from 5.1×10³ to 5.1×10⁴. At the gap of the two cylinders from 3 to 6 mm, the shear layer reattached onto the rear surface of the cylinder and the flow adhered beyond the rear stagnation point. The results obtained were as follows: The drag of the main cylinder decreased about 20 to 30% and the lift was generated by the forced reattachment with a value of C_L=1.0.

View full abstract

In the first report, it was made clear that a small cylinder controlled the shear layer separated from that main cylinder ant that two flow patterns occurred. One the pattern is a forced reattachment and the other is an elongation of the shear layer. This report is concerned with the characteristics of the wake flow of the patterns, namely, the Strouhal number, the time-averaged velocity near the rear stagnation point and the base pressure coefficient which are closely correlated. The rms value of the fluctuating pressure around the main cylinder and the pressure distribution around the small cylinder were measured. Thus, the decrease of the total drag force of the main cylinder including the small one was clarified. It was found that the direction of the fluid force acting on the main cylinder is related to the deflection of the from stagnation point on the main cylinder.

View full abstract

The heat generation in an unsteady flow between an oscillating plane and a fixed plane is studied theoretically. It is assumed that one of the two planes is oscillating in the horizontal and vertical directions. The flow is two dimensional and the fluid is viscous and incompressible. The equations governing the flow and the heat are analyzed taking the viscous dissipation into account. The coordinate in the direction along the surfaces of the planes is found to be separable in these equations. The governing equations depending on only two independent variables, the coordinate in the direction perpendicular to the plane surfaces and the time, are solved numerically using a finite difference method. It is found that, under the condition that the Prandtl number is very large, the oscillation combined in the horizontal and vertical directions strongly enhances the heat generation in the viscous flow. If the small vertical oscillation is superposed on the horizontal one, the temperature depends on the coordinate along the plane surface in the antisymmetrical form through the convection term in the energy equation. This dependency plays an important role in the heat generation. The mechanism on the heat generation due to the combined oscillation is physically clarified using an order-of-magnitude analysis.

View full abstract

The stability of two viscous fluid layers which are separated by a thin elastic plate under a vertical oscillation is investigated theoretically using a linear perturbation theory. Two fluid layers are horizontally bounded by two sinusoidally oscillating walls. The upper fluid is lighter than the lower one. The general stability boundaries for the given wave numbers are found analytically in the form of an infinite band determinant. And the neutral stability curves are composed of these stability boundaries in the plane of the frequency and the amplitude of the imposed vertical oscillation for the cases of first-, second- and fourth-order solution. When the ratio of thickness of the upper fluid layer to the total one is small or near unity, the system considered force of the elastic plate, their increase makes the system more stable, but it makes the system more unstable in the region of a small value of them. When the ratio of density of the upper fluid layer to the lower one increases, the system is stabilized.

View full abstract

The present paper reports a visualization study and a numerical simulation of the shock wave discharged from the open end of a shock tube. The three-dimensional shock wave flow is quantitatively visualized using double exposure holographic interferometry. In order to obtain the three-dimensional isopycnics behind the shock wave, a data evaluation process was described. The TVD finite difference scheme was applied to this problem. The experimental result was compared with the numerical result. A good agreement was obtained between them. The micro point explosions were conducted to verify the assumption of the present method that the collimated light was scarcely deviated at spherical shock fronts.

View full abstract

An experimental study is made to clarify the effect of magnetic field on the two-phase flow characteristics of a temperature-sensitive magnetic fluid in connection with the development of new energy conversion systems. It is confirmed that the sir injection in the throat and heating in the downstream side is effective to the pressure rise in the convergent-divergent nozzle under a nonuniform magnetic field. The experimental result of the pressure rise are compared with analysis of the one dimensional two-phase flow and reasonable agreement between them is obtained.

View full abstract

Basic equations for micropolar fluids in an electromagnetic field have not been established yet. In this paper, relativistic electromagnetic interactions on micropolar fluids are investigated by electromagnetic exergy. We obtained the results that electromagnetic momentum for micropolar fluid is well defined as g=D×B and an isotropic part of Maxwellian stress tensor can be expressed by electromagnetic exergy. Then as a result of thermodynamical consideration for micropolar fluids, The Gibbs's equation for micropolar fluids equals the equation for nonpolar fluids. Using the Onsagar's reciprocal theorem, constitutive equations of the dissipation parts have been derived more generally and easily for micropolar fluids than theories previous presented.

View full abstract

Low-solidity circular cascades, conformally transformed from a high-staggered linear cascade of 0.69 in solidity using double circular-arc vane sections, were applied as a part of the diffuser system of a transonic centrifugal compressor. Performance test results were compared with vaneless diffuser data. Good compressor performances, a wide flow range as well as a high pressure ratio and a high efficiency, superior to those with a vaneless diffuser, were demonstrated. Furthermore, the test circular cascade diffusers demonstrated a good pressure recovery in a wide range of flow angles, even when the inflow Mach number to the cascade was over unity.

View full abstract

Boiling heat transfer was investigated by experiment for grooving surfaces and some unwetted heating surfaces which were coated on the small copper surface with Teflon and some silicon materials. For a Teflon coated surface, the boiling curve showed the same characteristics as that of the copper surface, but the coefficient of heat transfer was considerable low in the nucleate boiling region. For some silicon coated surfaces, the boiling curve was much different from the usual heat transfer in pool boiling. The heat flux increased monotonously with the increase of the super heat and no critical heat flux and minimum heat flux were observed. For grooving surfaces, the critical heat flux was considerable low and observed at lower superheat than that in conventional boiling heat transfer.

View full abstract

The effect of forced convection on boiling heat transfer was investigated by experiment using water jet for a small unwetted heating surface which was coated with a silicon resin, AY49-208, on the copper surface under atmospheric pressure. The water jet velocity and the subcooling of water gave no effect on the heat transfer for a wetted heating surface in the nucleate boiling region, but the heat flux increased remarkably with the increase of jet velocity and the subcooling of water at the same superheat of heating surface for the unwetted heating surface. And the heat flux increased monotonously with the increase of the superheat of the heating surface. This unusual heat transfer was the same one for the pool boiling in the same kind of unwetted heating surface.

View full abstract

In order to clarify the effects of frost formation on the performance of an air-cooled heat exchanger, analyses by a uniform frost formation model and experiments were performed. Using this model, frost thickness and frost density can be estimated from the increase of the pressure drop of the heat exchangers. It is found that the heat transfer characteristics under frost formation can be grasped in the same manner as the case of no frost formation and that the analogy of heat and mass transfer holds in the practical application.

View full abstract

An approximate solution is proposed for the heat transfer problems of a latent heat thermal energy storage unit with a finned tube through which a coolant flows. In the actual storage unit, the temperature of the coolant changes along the axis of the tube so that the problems have been very difficult to solve. This analysis uses a dimensionless heat extraction rate or an apparent Biot number as a function of a dimensionless time, which has been obtained by analysis or experiment for the finned tube at a constant wall temperature. For this purpose, and analysis was made on the apparent the performance of a thermal energy storage unit with the finned tube and n-Octadecane as the phase change material agreed well with the experimental results.

View full abstract

This report a new method of thermoelectric power generation by use of the large temperature difference due to the fin effect and the amorphous semiconductor thin film. Instead of utilizing the small temperature drop in a thermoelectric device as adopted in a conventional device, the method makes use of the fin effect which causes a large temperature difference in a thin metallic plate cooled at one end and exposed to a high heat flux or a hot gas flow. The large temperature difference is available for the thermoelectric film. The amorphous film is deposited on the plate by an ion-cluster method and has a thin electric insulating film between the plate. Temperature distributions in the plate are theoretically and experimentally investigated and thermoelectric power generation performances are experimentally studied.

View full abstract

Ice formation phenomena around isothermally cooled cylinders arranged in a parallel line with water flow were examine experimentally The onset condition, which describe whether the cooled cylinders are linked by ice or not, were obtained by introducing the analytical model of ice formation around the cylinders. It was found that when the cylinders are linked by ice, the value of the nondimensional parameter, which represents how efficiently the cylinders can work as a thermal energy storage system, is not affected by cylinder pitch. A correlation equation which represents the effect of cylinder pitch on the amount of ice was proposed.

View full abstract

A theoretical study has been conducted on air-atomized fog cooling in the region equivalent to film boiling with low mass velocity of water. In the analysis, total heat flux is expressed by the sum of three components, i.e. radiation, forced convection by air flow and evaporation of liquid droplets. The air flow is approximated by an axisymmetric laminar stagnation flow. It is assumed that spherical droplets float on the vapor film formed along the hot surface and move in the radial direction with a droplet is estimated by conduction through the vapor film. It has been derived that heat flux by evaporation increases linearly with the mass velocity of water and inversely with the square root of diameter and impinging velocity of the droplet. The prediction performance of the proposed analysis if fairly good.

View full abstract

This paper presents the numerical analysis of the deformation of a droplet impinging upon a cold flat surface. The calculation was based on the basic equation of the SMAC method in the cylindrical coordinate, under the consideration of the surface tension. When Re, Fr and We numbers are the same, the experimental results are in good agreement with the calculated ones at each non-dimensional time. The heat transfer model was constructed for the case of impingement upon a hot surface, and the basic equation were solved by the energy method, one-dimensional heat condition, and two-dimensional heat condition. The experimental results of the life time of a droplet agree well with the calculated ones, as shown by the latter two methods. It is quite possible to simulate the temperature profile in the droplet, although the convection in the droplet and the evaporation process are disregarded in the model.

View full abstract

Thermal CVD reactors of a horizontal rectangular duct are widely used in the production of semiconductor layers, but in the process strong vortexes having axes in the flow direction prevent uniform growth of the layer. This research aims at a fundamental study of controlling strong to provide horizontally uniform fluid temperature. Experiments reveal three typical flow patterns caused by heating the side walls. Those are: a flow of a single-pair Benard cell, and unsteady flow with time-averaged uniform temperature and a pair of vortexes with circulation opposite to that of a Benard cell. Numerical calculations assuming a laminar, steady 2-dimensional flow were made by a SOR method. Experimental and predicted results were compared and indicated agreement in strong vortex cases, but little agreement otherwise.

View full abstract

The primary coolant in a PWR Plant contains boric acid to control the excess reactivity and Lithium-7 to reduce the induced reactivity of corrosion products. Li-7 which is produced by B-10(η, α) Li-7 reaction is necessary to be removed by cation demineralizer in order to avoid the accumulation of it. Recently, to reduce the solubility of corrosion products and the radiation level from radioactive corrosion products, it is recommended that the concentrations of boron and Li-7 in reactor coolant are controlled within the appropriate ranges. In this paper, the simulation analysis to determine Li-7, B and Cs-137 concentrations and the operation procedure to control Li-7 and B concentrations are studied, and an optimum operation procedure is proposed.

View full abstract

The present paper studies processes of the failure of detonations propagating in a divergent nozzle by conducting numerical simulations. A finite-difference method, which combines the Random Choice Method with a chemical kinetics simulation code and a computational step for effects of nozzle cross section, is applied to simulate gasdynamic phenomena of one-dimensional detonations moving along a divergent nozzle in a hydrogen and oxygen diluted in argon mixture. Also, two-dimensional simulations are carried out and compared with the one-dimensional results. It is shown that the present one-dimensional calculations simulate averaged flow fields of the two-dimensional results over the cross section of the nozzle.

View full abstract

The interferometric tomography method has been developed to obtain a three-dimensional temperature profiles in a flame. A multidirectional holographic interferometer is described, with which the interference fringe pattern can be obtained from 24 different directions in an instant. This technique is used to measure a section of three-dimensional temperature profiles in a asymmetric alcohol wick flame. A thermocouple probe measurement is used simultaneously for comparison with the interferometric method. Preliminary results indicated that the interferometric tomography technique is feasible and has great potential for combustion research such as flame structure studies.

View full abstract

Turbulence properties and the balance of turbulence energy in a two-dimensional impinging jet on a wedge are investigated. The angles between the wedge surface to the jet axis φ are 30° and 45°. The profiles of u'/U_m, v'/U_m, q^^-2/2U²_m in the fully-developed flow region are larger than those of the two-dimensional wall jet, but the profile of w'/U_m is almost the same as those, and these profiles are not similar. Each term of the production, the dissipation and the diffusion in the turbulence energy equation is larger than that of the two-dimensional wall jet except for the advection term. The production and the diffusion terms in the case of φ=45° are larger than in the case of φ=30°. The pattern of the isopleths of <uv>^^-/U²_j in the impinging region differs remarkably with each case of φ=30° and 45°.

View full abstract

An experimental study was made of the structure of the wake flame formed behind a porous sphere from which methane was discharged into a uniform stream of dust-free dry air. The laser Raman scattering technique was applied to measure the temperature and the number density of gaseous molecules in the flame. Flame temperature was determined from the measured results of the Stokes-Raman spectrum of nitrogen molecule which was irradiated by the Argon laser (488 nm). measurements were also made of the flame emission spectroscopy. The results showed that the maximum temperature was observed near the periphery of the visible flame. The ratio of the number densities of oxygen and nitrogen molecules in the central portion of the flame was much lower than the value in the room air. Strong OH emission was observed near the maximum temperature region.

View full abstract

An observation method for the distributed reaction zone in turbulent premixed flames was proposed. A mixing layer between a combustible mixture and a hot burned gas, which existed in a uniform shear layer, was studied to observe the interaction between the turbulence and reaction. A typical "wrinkled laminar flame" appeared when the equivalence ratio of the mixture was 0.65, and for an equivalence ratio much lower than the flammability limit, a structure which looked like a "distributed reaction zone" was observed. In the latter, and intermediate "reactedness" and a thick reaction zone were observed by the detection of fluctuating temperature, ion current and chemiluminescent emission of the zone.

View full abstract

Isothermal annular swirling flows were investigated to clarify the effects of a bluff body and a nonswirled central air jet upon the characteristics of swirl-induced recirculation zones. The velocity measurement using an LDA system revealed that the use of a bluff body increased the rate of recirculating mass flow and also the mass flow brought back into the quarl (or expansion) section. For flows with a primary air jet, two distinct flow patterns were observed: (I) ordinary IRZ (Internal Recirculation Zone) and (II) IRZ with primary air jet penetration. The IRZ strength could be determined by the combined parameter, Sa(D1/Do)², where the swirl number is Sa and the vortex diameter expansion ratio is D1/Do.

View full abstract

The possibility of removing the pollutants by arc plasma injection from dirty gas containing higher concentration of NO_x, SO_x, HCl and dust emitted from a municipal refuse incinerator has been studied in this paper. A low-power arc plasma generator is designed and operated continuously on a argon gas with high stability. The air is injected from a capillary into a argon plasma jet and the formation of NO_x is measured in order to assess the generation of specific radicals in a plasma wake. The optimum injection point for the generating radicals are discussed and the formation of NO_x in the presence of oxygen for a practical system are also evaluated.

View full abstract

The formation and growth process of flame kernels in the ignition process by short-duration electric sparks has been investigated by using a numerical simulation and schlieren photographic observations. In the simulation, unsteady two-dimensional cylindrical coordinate system was employed and the governing partial differential equations, such as mass, momentum and energy equations were solved by the finite differential method. The calculated results show that the ellipsoidal hot gas kernel produced by a spark discharge develops into a torus and that there are two high-temperature regions, namely within the rings of torus and just between the spark electrodes. It is also found that a narrow groove is formed inside the torus. These calculated features of the hot kernel structures agree with those in schlieren photographs taken by the high speed camera. Effects of diameter of spark electrodes and shaped of electrode tips on the formation process of flame kernels are also studied.

View full abstract

To gain insight into the micro-structure of turbulent flame, and analysis of the schlieren-images of propagating flame has been made by a turbulent combustion zone model. Combustion experiments were made in a constant volume bomb, where the characteristics of the mixture (laminar flame thickness, laminar burning velocity and oxygen concentration) and the turbulence intensity were varied independently. The obtained schlieren-images showed that the wrinkling of flame front increased and the scale of the smooth area decreased as the turbulence intensity increased. The increased wrinkling has been explained by a superimposition of the schlieren-images of turbulent flamelets existing along the combustion zone thickness. In this analytical discussion, the turbulent combustion zone has been considered to be mixed with the wrinkled laminar flame zone and the distribnuted-islands flame zone, where each reaction rate has been determined by a Damkohler number and each characteristic reaction time.

View full abstract

The swirl flow at a compression stroke in a cylinder of the two-stroke-cycle uniflow scavenging diesel engine plays a very important role in the combustion performance. In this report, the numerical analysis method of the gas flow in a cylinder at the compression stroke is developed. This method is proceeded by and approximation of a 2-D axisymmetrical flow to save computation time, but uses, for the initial condition, the calculated results of a 3-D scavenging flow simulation to ensure the accuracy of the starting swirl flow pattern which is deeply affected by the port arrangement. Some calculated results are compared with the experimental data of a model engine, and it is concluded that this method can simulate the flow pattern in s cylinder at the compression stroke with good accuracy.

View full abstract

This study is for establishing the spark ignition method for injected high pressure hydrogen to replace the hot surface ignition because the spark ignition saves electrical power, improves durability and prevents abnormal combustion. The experiments were conducted both in a constant volume combustion chamber with Schlieren method and in an actual piston engine in order to optimize the spark gap position and timing which could realize stable ignition with a short delay. Ignition for an actual engine took place in a much larger region than for a constant volume chamber due to high pressure, high temperature and gas flow, although a constant volume chamber can be used for understanding the fundamentals. As the optimized spark ignition method, the following was learned from the actual engine experiments; full-transistor spark generation at 3 to 5 mm from an injection hole and almost simultaneous fuel injection with the spark.

View full abstract

The airflow was measured by a LDV in a motored four-stroke direct injection diesel engine with a troidal cavity combustion chamber, under three different swirl intensity condition. The turbulence scale was calculated by a random sample data analysis utilizing multicycle velocity data. It has been clearly demonstrated that even though the swirl intensity was varied, the integral time scale increased once at the beginning of the compression stroke and decreased toward compression top dead center. On the other hand, the micro time scale hardly changed in suction and compression strokes. Furthermore, it has been revealed that the axial micro time scale was almost uniform in the cylinder under the high swirl condition, but the integral time scale remained large near the piston cavity center.

View full abstract

The authors have developed a method to analyze the turbulence scale accurately from the descrete data, which are usually obtained by LDV measurements, at low sampling rate with non-uniform time intervals. In this method, the turbulence correlation is calculated from the data of many engine cycles, and successively the turbulence spectrum is calculated from the inverse Fourier-transform of the turbulence correlation. The reliability of this method is confirmed by the model data sampled randomly on a sinusoidal process, and by the data measured in the cylinder of a motored diesel engine having a direct injection system. Furthermore, it is clearly shown that the spectrum in the frequency range higher than the Nyquist frequency defined by the mean data rate can be analyzed by this method, and that it is not necessary to change the range of frequency analysis despite a little variation of mean sampling rate.

View full abstract