The flow behavior of viscoelastic non-Newtonian fluid in circular and non-circular ducts is of special engineering interest. Therefore a numerical analysis has been performed for viscoelastic non-Newtonian fluid in elliptical duct. Special attention is paid for the generation of secondary flow for laminar flow by using two kinds of constitutive equation, i.e., Maxwell and Reiner-Rivlin models. As for Maxwell model, body force caused by the elastic stress is approximated by linear source term. In calculation, viscosity was represented by adopting power-law fluid and boundary-fitted coordinate system was introduced as the method of coordinate transformation. The calculated results of two models show the secondary flow in elliptical duct as the same as theoretically analyzed by Green and Rivlin. Adding to the prediction of secondary flow, the generation mechanism of secondary flow has been argued by evaluating the production terms of the transport equation for streamwise vorticity. As a result of this examination, it was found that the term of viscous diffusion and the term containing second normal stress difference played an important role in producing the secondary flow near the wall. At the same time, it is interested phenomenon that the circular direction of secondary flow for viscoelastic fluid is opposite sigh to that of secondary flow for Newtonian turbulent flow. As its cause, the present study clarified that the term containing second normal stress difference of viscoelastic fluid is the same type equation for that of turbulence, while the sign of its term for viscoelastic fluid is opposite to that for turbulent flow.

View full abstract

This study aims to develop a partitioning algorithm for data parallel computation. The greedy partitioning algorithm for the message passing developed by Farhat was modified for the data parallel computation of an element-by-element finite element method. The influence of the numbering of nodes and elements onto the parallel efficiency is examined through the parallel computations using CRAY-T 3 D. It was shown that the proposed partitioning algorithm considering the ordering of nodes and elements for data parallel computation can save time of solving problems and will improve the efficiency of parallelism. In the case of CRAY-T 3 D with 128 PEs, the maximum number of communications of the cyclic ordering version in the flow inside a subway station model with about 1 410 000 elements is 89% smaller than the original one. The rate of CPU usage of the cyclic ordering version is 92%.

View full abstract

Direct numerical simulations of MHD homogeneous turbulence in a uniform magnetic field are conducted to investigate the mechanism of anisotropy and decay of MHD turbulence. By applying a uniform magnetic field, fluctuations of velocity and vorticity are dumped significantly. Intensity of the velocity and vorticity fluctuations show anisotropic behaviors in which the intensities of velocity and vorticity parallel to the uniform magnetic field are greater than those perpendicular to the magnetic field. In this process, integral length scale and Taylor micro scale parallel to the uniform magnetic field become longer. The magnitude of fluctuating magnetic field and current density also show directional dependence. The fluctuation of the magnetic field and current density in the perpendicular direction is larger than those in the parallel direction.

View full abstract

Fine scale structure in MHD homogeneous turbulence is investigated to clarify the relationship between anisotropy and coherent fine scale structure in turbulence. In MHD turbulence, mean diameter of the coherent fine scale eddy is about 10 times of Kolmogorov micro scale and maximum azimuthal velocity is about 0.7 times of u_mrs, which coincide with those in homogeneous isotropic turbulence, turbulent mixing layer and turbulent channel flows without magnetic field. Effect of magnetic field appears in angle distribution of axis of the coherent fine scale eddy. The coherent fine scale eddy perpendicular to the mean magnetic field decays fast due to strong Lorentz force, which results in anisotropic behaviors of MHD homogeneous turbulence.

View full abstract

In this work, we propose a new unified-solution algorithm for solids and fluids by GSMAC-FEM. We use the Lagrangian description for displacement vectors field in solids, on the other hand, the Eulerian description for velocity vector field in fluids. And we carried out the analysis of static beam bending problem and dynamic visco-elastic response problem in order to verify validity of the present scheme for solid phase. Then the analysis of self-excited oscillation of a cylinder in uniform flow is performed as coupled vibration problem between fluid and solid.

View full abstract

The artificial compressibility algorithm for incompressible flows is applied to an unstructed grid method. It is based on a finite volume cell-vertex upwind technique and uses a time-marching procedure to efficiently solve the continuity equation as well as the momentum equations. The Lower-Upper Symmetric Gauss-Seidel (LU-SGS) approximate factorization scheme is implemented with the Navier-Stokes solver on arbitrary three-dimensional unstructured grid. The LU-SGS algorithm provides a significant enhancement of the convergence rate. The accuracy of the method is validated by applying the method to incompressible flow problems of a 2-D cavity and 3-D duct.

View full abstract

Experimental studies show that the growth rate of compressible mixing layers is reduced with the increasing Mach number. The reduced growth rate is believed to be due to the effect of compressibility on turbulence. DNS databases of compressible mixing layers show that reduced pressure fluctuations are responsible for the changes in growth rate via the pressure-strain correlation term. In this parer, a turbulence model for the pressure-strain correlation term in which compressibility effect is included is derived. The derived model is used to simulate compressible mixing layers, showing that the model predicts the reduced growth rate which is often observed in experimental studies.

View full abstract

Fluid mixing and pressure drop in a grooved channel are studied numerically for pulsatile flow at an intermediate Reynolds number. In particular, the imposed oscillatory frequency effect is received attention for small and large oscillatory fractions of the flow rate. Hydrodynamic resonance takes place at the imposed frequency close to the self-sustained oscillatory frequency even for a large oscillatory fraction, and the time-averaged vortex strength has a peak at its frequency. However, the pressure drop tends to increase as the imposed frequency becomes large, in the range considered here. Fluid particle ejection from the groove and Lyapunov exponent near the groove wall are examined to represent the fluid mixing characteristics. The best fluid mixing takes place near the frequency for hyrodynamic resonance. Also, flow visualization studies support the numerical simulation results.

View full abstract

A direct numerical simulation (DNS) of turbulent round jet was carried out in the cylindrical coordinate system to provide a tool for a numerical experiment for jet control. The spatial development from the formation of vortex rings to the transition to turbulence was accurately obtained by a finite difference method. The characteristics of simulated flow field, such as vortex rings and self-similar profile of mean flow in turbulent region, is in good agreement with experimental findings. Insufficiency of the computational domain in the radial direction leads the flow field to turbulence earlier. The influence of superposition of coherent disturbances to random perturbation at the nozzle exit, e.g. axial and helical ones, has also been investigated. The perturbation mode strongly affected the configuration, strength and location of vortices and their turbulence transition. The delay in turbulence transition, especially by axial excitation, was due to the maintenance of the strong vortices. Such a structure was clearly detected by the proper orthogonal decomposition (POD) method.

View full abstract

In order to clarify the draft-tube performance of cross flow-turbines or the submerged water-jet-performance through the air, two-dimensional aerated-bubbly-flow-behavior within a narrow duct, especially on the energy recovery rate of the jet, have been systematically experimentally studied, when the jet is vertically injected into cahm water. The recovery rate η is closely related to the jet velocity V and the volume rate μ_aw of air-water-mixture. Such a flow pattern is also visualized.

View full abstract

Many studies about flow induced vibration of a single cylinder have been conducted and a lot of problems were cleared. However, most of these studies are concerning to the velocity, at which the vortex resonant vibration will occur. Recently, cylinders in a high-density fluid have vibrated in a very low velocity range. Several studies have been conducted but many unknown points are remained. Vibration tests of a single cylinder in water were conducted to make the vibrations clear. These tests revealed that the vibration starts at the reduced velocity of about 1.4 and ceases at the reduced velocity of about 4.1. Most of the vibrations are in drag direction but some of them vibrated in locus of a character 8 of Arabic mumerals. Unsteady fluid dynamic forces acting on the cylinder at the low velocity were also measured and ensured the existence of the vibrations in a low flow velocity.

View full abstract

Experimental investigations on the flow characteristics of the forces acting on and the flow patterns around square cylinders at different angles of attack (0°≤α≤45°) submerged in oscillatory water flow were carried out. Both in-line and transverse forces were simultaneously measured in a relatively wide range of Keulegan-Carpenter number (KC) from 1 to 90. The force signals were spectrally analyzed to get dominant frequencies, and flow visualization technique was also employed to understand the correspondence between flow patterns and force coefficients. We have discussed the influence of the angle of attack of the cylinder on the flow pattern and fluid dynamic characteristics of a square cylinder. Further, force coefficients and Strouhal numbers compared with the results obtained in steady flow.

View full abstract

A new vortex scheme for simulating flows involving the natural convection and the interaction between the temperature and the vorticity is presented. The creation of vorticity from the temperature is modeled either by creating a vortex pair from a temperature particle or by changing the strength of vortices according to the vorticity equation. The diffusion velocity method is used for simulating the diffusion of the vorticity and the temperature. The vortices of negative and positive strength are separately treated in the diffusion process to avoid an unreasonably large diffusion velocity. Our results indicate that these techniques successfully simulate creation of the vorticity from the heat, the diffusion and the convection of the temperature and the vorticity, and the interaction of them.

View full abstract

Fast summation algorithms for a particle simulation are presented. The diffusion velocity which appears in the vortex methods is treated as an example of rapidly decaying potential and the convection velocity as far field potential. Computational time is reduced from O(N²) to O(N), where N is the particle number, for the diffusion velocity and to O(N log N) for the convection velocity by the use of blocks with various widths which are arranged in a computational domain based on a simple strategy. Our method is applied to a simulation of heat-vortex interaction and it is shown that the theoretical efficiency is achieved in this practical use and that our code is well parallelized.

View full abstract

This paper presents a potential for a passive control of a horseshoe vortex at the root of the wing. NACA 0024 wing is established on a turbulent boundary layer. A pair of vortex generators of half delta wing is installed upstream of the wing. The controlled horseshoe vortex is tested qualitatively by flow visualization technique. Also, the potential for controlling is quantitatively investigated by wall static pressure and total pressure. The horseshoe vortex is remarkably controlled in Common Flow Up Configuration (CFUC) of vortex generators. The distortion of the total pressure contours is diminished by 49% and the vortex is located closer to the wing. In case of Common Flow Down Configuration (CFDC), the mass flow averaged pressure loss is decreased by 29% compared with the case without a pair of vortex generators.

View full abstract

This paper presents the behavior of a passively controlled horseshoe vortex at the root of NACA 0024 wing which is established on a turbulent boundary layer. A pair of vortex generators of half delta wing is installed upstream of the wing. The flow field of the optimally controlled horseshoe vortex both in case of Common Flow Up (CFUC) and Common Flow Down Configuration (CFDC) is carefully investigated by an X-array hot-wire. In case of CFUC, the horseshoe vortex is not shifted from the wing, because the longitudinal vortex is restrained. The interacted vortex presents a circular profile, in a optimally controlled case. In case of CFDC, the interacted vortex that has strong vorticity by the pairing process is shifted away from the wing. Then, the high momentum fluid flow penetrates between the wing and the vortex.

View full abstract

Reynolds number dependence of the structure of energy dissipation is investigated analytically in a weakly non-axisymmetric Burgers vortex tube. An asymptotic solution representing the viscous dissipation rate as well as the axial vorticity is obtained in the weak non-axisymmetry limit λ=(α-β)/(α+β)«1, α and β being strain rates of background flow normal to the tube, at finite Reynolds number Γ/(2πν) ranging from 0 to 100, where Γ is the circulation of the vortex and ν the kinematic viscosity. The asymptotic solution demonstrates that a Burgers vortex tube is deformed into an elliptical shape by the non-axisymmetry of the ambient strain and that the major axis of the ellipse. turns from the principal axis of the strain at a 45° angle, as the Reynolds number is increased. The variation of the rotation angle against the Reynolds number is in excellent agreement with that in the numerical solution obtained by Robinson and Saffman (1984). In contrast to the axial vorticity, i.e. the vortex itself, the high dissipation region stays almost independent of the Reynolds number. At relatively low Reynolds number Γ/(2πν)=O(1), the high dissipation region deviates from the high vorticity (and so the high enstrophy) region. The reason for this deviation is discussed.

View full abstract

A numerical simulation by a three dimensional vortex blob method has shown that an axisymmetric sinusoidal forcing generates a vortex ring with circulation which is greater than that for the unforced flow, under the condition that the same volume of fluid issues from a circular nozzle during a fixed time interval. The time interval is chosen as 4.0(2R/U) based on the experiment of Gharib et al. (1998), where R is the radius of the nozzle and U is the average velocity of the issuing fluid. At a fixed amplitude of forcing, this maximum circulation is realized at a forcing frequency of 0.125(U/2R). This is because most of the circulation shed from the nozzle is absorbed into the vortex ring at this optimum frequency. The maximum circulation is greater than that of the unforced flow by an amount proportional to the amplitudes squared. This is discussed in terms of the circulation shed from the nozzle during the issuing time.

View full abstract

The cavitation performance in a cylindrical choke with a sharp edged-corner was investigated in previous paper. Many differences were detected in cavitation performance by changing the diameter and the length of a choke. The purpose of this paper is to find out in more detail the reason which causes the differences in the cavitation performance. Experiments are carried out using two kinds of cylindrical chokes with a sharp edged-corner and chamfered length at the choke inlet. The pressure distributions at three different locations within the cylindrical choke and just at the back of the choke outlet are measured in the same way as previous paper. As a result, the reason which cavitation performances indicate the phenomena according to shapes and sizes of cylindrical choke has been clarified in more detail as compared with the study before now.

View full abstract

This paper deals with an experimental study on the behavior of the Cavitation bubbles by using a flow visualization for the steady and unsteady flow. In the previous paper, experiments were carried out for various pressure changes of the upstream while keeping the downstream pressure constant. In this study, experiments were performed for various pressure changes which take the steady and unsteady flows at the downstream of a choke keeping the upstream pressure constant. The behavior of the cavitation bubbles is clarified with a camera and a high-speed video camera under three parameters: four kinds of the diameters d, two kinds of the lengths l, two kinds of the chamfered lengths s of the choke. As a result, three types of the sheet, bubble and cloud cavitations are able to observe in a long choke. The possessive region of each cavitation feature depends on d and s for a long choke even though the cavitation number is same value for the different d and s. The bubbles which occur in a choke repeat the growth, the division, the collapse at very short intervals.

View full abstract

We calculated the flow-field and surface response condition of a planetary probe entering the Jupiter atmosphere, Within the region behind a detached shock wave, the blunt nose is heated to high temperature. So that, the numerical results were obtained under a viscous-shock-layer analysis. To protect the probe against high temperature condition and to reduce the heat flux, which includes convectional and radiative heat transfer, we used the ablation of C/C composite materials. This simulation included mass injection, radiative transfer, diffusion, and viscous effects. The radiations included were molecular bands, atomic lines, and continuum processes. It was found that carbon ablation materials were effective in reducing the radiative heat flux.

View full abstract

A model for predicting noise of propeller fans has been developed experimentally. First, the frequency response of turbulence on the blade of a rotating propeller fan was experimentally measured. Then, propeller fan noise was predicted by this model, which calculates lift fluctuations by Sears's equation and sound pressure in the far field by Curie's equation. The model assumes that the normalized intensity of turbulence (i.e., velocity fluctuation) is small in the high-flow-coefficient range but increases sharply in the low-flow-coefficient range. And the model showed that propeller fan noise depends on averaged relative velocity, intensity of turbulence, and length scale of turbulence. These results agree with the experimental measurements. Accordingly, the model predicts propeller fan noise and noise spectrum over a wide range of flow coefficients with reasonable accuracy.

View full abstract

Evolution of multiple stall cells with short length-scale in an axial compressor rotor was investigated experimentally. In a low-speed research compressor rotor tested, a short length-scale stall cell appeared at first, but did not grow rapidly in size unlike a so-called "spike-type stall inception" observed in many multi-stage compressors. Alternatively, the number of cells increased to a certain stable state (a mild stall state) under a fixed throttle condition. In the mild stall state the multiple stall cells, size of which was on the same order of the inception cell (of a few blade spacings), were rotating at 72% of rotor speed and at intervals of 4.8 blade spacings. With further throttling, a long length-scale wave appeared overlapping the multiple short length-scale waves, then developed to a deep stall state with a big cell. The ensemble averages of the axial velocity components upstream and downstream of the rotor, which were obtained phase-locked to the cell rotation, showed that the stall cells were of a part-span type.

View full abstract

Pump efficiency drops rapidly with a decrease in specific speed n_s in the very low n_s range. The cause for low efficiency of a very low specific speed pump is mainly due to large disk friction. In the present study, two methods of reducing disk friction is put forward and is confirmed experimentally. One method is to decrease the amount of leakage flow, which resulted in a rapid decrease of disk friction and raised the pump efficiency by about 7%. The other is to increase the surface roughness of impeller shroud, which raised the pump efficiency by about 3%.

View full abstract

Flow and heat transfer of natural convection over heated horizontal plate was investigated both experimentally and analytically. The main concerns are the turbulent transition mechanisms and the local heat transfer characteristics of transition region. In order to investigate the mechanisms of turbulent transition, an unsteady, three-dimensional analysis has been performed on the water flow over 150 mm wide plate heated with uniform heat flux. The results showed that longitudinal vortices appear over the plate when the local modified Rayleigh numbers exceed 3.5×10⁶. These vortices develop toward downstream, then, lose their regularity and, finally, a turbulent transition occurs over the plate. By comparing these results with visualizations on the flow and temperature fields and also with the local heat transfer coefficients botained from the experiment, it was confirmed that the present analysis well predicts actual transition process.

View full abstract

An enhancement technique was developed for natural convection heat transfer from tall, vertical heated plate to water. Rectangular grid fins attached to the base plate were utilized as a heat transfer promoter. These grid fins redirect the high-temperature fluid ascending along the base plate toward the outside of the boundary layer and introduce the low-temperature ambient fluid toward the base plate instead. The heat transfer coefficients of thus-treated surfaces were measured and compared with a non-treated, fiat surface and a conventional surface with vertical straight fins. The highest performance was achieved for the experimental surfaces. In particular, the experimental surfaces with 5 mm-high, non-conducting grid fins and with 10 mm-high, conducting grid fins show 27% and 80% higher heat transfer coefficients compared to the turbulent heat transfer coefficients of the non-treated, fiat surface, respectively.

View full abstract

In the present study, heat transfer coefficients affected with a projection in the turbulent region of a vertical flat plate were measured experimentally for various heights of projections in the range of 0∼20 mm. The wall temperature and fluid flow fields were also visualized using liquid crystal sheet and nylon 12 powder. Average and local Nusselt numbers reach 1.07∼1.22 and 1.2∼1.7 times of those for pure turbulent natural convection, respectively. The maximum enhancement rates of heat transfer are attained in the location of 2.3∼3.3 times of projection height from upper surface of projection towards downstream, and these locations are in good agreement with those of reattachment of fluid flow and center of dark red region in the liquid crystal. On the other hand, the heat transfer coefficients reduce in the just upstream and downstream regions of the projection when compared with those for the no projection. By introducing a nondimensional parameter Ra^*_H, the present experimental results are rearranged quantitatively and effectively.

View full abstract

Slush hydrogen is a mixture of liquid hydrogen and solid hydrogen particles, and is being considered as a spaceplane fuel or as a means of transport for hydrogen used as a source of clean energy. This paper describes nucleate boiling heat transfer characteristics to slush hydrogen and slush nitrogen. For visual observation of heat transfer states, a heat transfer unit was placed in a glass dewar designed to minimize the heat leak from atmospheric environment. The heat transfer unit used was a circular fiat plate 0.025 m diameter made of electrolytic tough pitch copper. For orientations of the heat transfer surface during testing, three different orientations were used; horizontal facing up, vertical, and horizontal facing down. In addition to that for slush hydrogen and nitrogen, heat transfer data for normal boiling point (NBP) liquid hydrogen, triple point (TP) liquid hydrogen, NBP liquid nitrogen and TP liquid nitrogen was obtained up to the critical heat flux (burnout). Comparison of these data with the slush hydrogen and nitrogen data, including the results of observation of the heat transfer surface, clarified nuclear boiling heat transfer characteristics to slush hydrogen and slush nitrogen, which have hardly ever been investigated.

View full abstract

The effects of inner rotating plate with horizontal axis on the heat transfer between vertical two wails in a rectangular enclosure with stratified fluid layers were investigated experimentally. The aspect ratio of the enclosure height/width was 1 throughout the experiments. An acrylic plate with small thermal conductivity was installed horizontally at the center of the square enclosure, and was rotated at various speeds for normal and reverse rotations by using the motor attached outside of the enclosure. Purified water and silicon oil were used for the working fluid and were stratified in the enclosure. It is clarified here that the heat transfer rate of the enclosure with stratified fluid layers differs largely from that of the enclosure with single fluid layer. Namely, the heat transfer rate increases exceedingly at low rotating speed range, and keeps almost constant value at high rotating speed range.

View full abstract

In the prior paper, a loop-type osmotic heat pipe was investigated on the heat transport rate from the upper position to the lower. The heat transport rate of this osmotic heat pipe, however, was very small because of small driving force for the solution circulation. Therefore, the osmotic heat pipe with the higher driving force is devised. The larger driving force is produced by a two-phase flow taking place in the solution riser. For the sake of that, a vopor-liquid separator is installed at the top of the apparatus and a heated section is attached as some distance below the separator, so that buoyant force caused by density difference between the fluids in the solution riser and downcomer increases. This osmotic heat pipe with the two-phase solution circulation loop indicates the higher heat transport rate than that of the osmotic heat pipe shown in the prior paper.

View full abstract

In the present paper, heat transport rates and fluid flow patterns of a bubble-driven heat-transport device (BD-HTD) made of glass were experimentally investigated by using water, soapsuds, ethanol, and R141b as an operating liquid. In this type of HTD, heating and cooling sections are connected with each other by a closed loop of tube meandering between them, and operating liquid of a volume fraction is enclosed in the meandering loop. The BD-HTD was set vertically, and it was heated at the bottom by heating water and cooled at the top by cooling water. Experimental parameters were the inner diameter of the tube (D=1.8, 2.4, 5.0 mm), the total temperature difference of the heating and cooling water (ΔT=20-60 K), and the liquid volume fraction (α=18-98%). The main results are summarized as follows. The heat transfer coefficient of the operating liquid at the heating and cooling sections, h_fi, is not strongly dependent on α and ΔT. Among the present test liquids, the effective thermal conductivity, k_ef, is highest for R141b but the heat transfer coefficient, h_fi, is highest for water. As k_ef is sufficiently high even for water, the heat transport rate, Q, is highest for water. Q of the BD-HTD using water can exceed the maximum heat transport rate of the conventional heat pipes of the same geometry with the present HTD. For R141b, the BD-HTD operated for D/λ₀=1.5 to 4.2 (λ₀; the capillary length) and Q is not srongly dependent on the tube diameter. This result indicates that the BD-HTD is suittable for micro HTDs, but for water the BD-HTD did not operate for D/λ₀=0.65.

View full abstract

An approximate analytical solution for the initial transient process of close-contact melting occurring between a phase change material kept at its melting temperature and an isothermally heated fiat surface is derived. The model is so developed that it can cover both rectangular and circular cross-sectional solid blocks. Normalization of simplified model equations in reference to the steady solution enables the solution to be expressed in a generalized form that depends on the liquid-to-solid density ratio only. A selected result shows an excellent agreement with the previously reported numerical data, which justifies the present approach. The solution appears to be capable of describing all the fundamental characteristics of the transient process. In particular, dependence of the solid descending velocity on the density ratio at the early stage of melting is successfully resolved. The effects of other parameters except the density ratio on the transient behaviors are efficiently represented via the steady solution implied in the normalized result

View full abstract

Thermal comfort in an automobile is predicted with numerical simulation. The flow field and temperature distribution are solved with a grid system based on many small cubic elements which are generated automatically with cabin and passenger configuration. Simulation of temperature is combined with simulation of cooling cycle and calculation of heat transfer at the wall including solar radiation to treat transient and actual driving conditions of the vehicle. In order to evaluate thermal comfort, transitional effective temperature is calculated from simulated thermal conditions and physiologic values which are calculated by a simple model of human thermal system. This system can well predict thermal sensation of passengers in a short period of time.

View full abstract

A thermal efficiency of heat engine systems is generally evaluated as the ratio of the available net work to be taken out from the system to the fuel energy into it. In usual systems such as steam cycle or combined cycle, the evaluation stated above is reasonable one. However, in Waste-to-Energy systems to be combined with gas turbine and steam turbine, which are named as Super-WTE in Japan, the evaluations to be usually used are not always reasonable one. Author has newly defined the ratio of the power increase to the inputted gas turbine fuel energy as repowering efficiency. In this paper, the reason that the usually used thermal efficiency is not reasolable on evaluation of combined WTE system has mentioned and the relationships between the repowering efficiency versus gas turbine capacity and its pressure ratio have been studied.

View full abstract

In previous multi-dimensional modeling on spray dynamics and vapor formation, single component fuel with pure substrance has been analyzed to assess the mixture formation process. Then it shold be expected that the evaporation process could be performed for the multi-component fuel such as actual Gasoline and Diesel gas oil. In this study, vapor liquid equilibrium prediction was conducted for multi component fuels such as 3 and 10 components mixed solution with ideal solution analysis and non ideal solution analysis. And the computation of distillation characteristics was conducted for the steady state fuel conditions to understand the evaporation process. As a result, calculated distillation characteristics is consistent well with experiment results. Further the vapor concentration was measured for the analysis of the binary fuel spray using IR absorption measurement method.

View full abstract

Behavior of gasoline hollow cone spray at pressurized condition was studied. Swirl type fuel injector was used. Surrounding pressure was varied 0.1∼0.5 (MPa) which covers the surrounding pressure of compression stroke. The spray behavior was observed by LLS (Laser Light Scattering) method and LDV (Laser Doppler Velocimeter). Spray angle was narrower when fuel injection period was longer and surrounding pressure was higher. The pressure difference between spray inside and outside effects the spray angle change. The Spray pattern at pressurized surrounding condition was canged by the initial center distributed spray.

View full abstract

Jet ignition method by hydrogen oxygen flame is proposed to control the combustion of lean mixtures. Hydrogen oxygen flame jet is supplied from the cavity whose volume is less than one percent of main chamber volume through a nozzle. The burning velocity of the cavity gas is very fast and the combustion duration can be drastically shortened. The pressure oscillation often takes place. The heat flux at the wall of main chamber becomes bigger than the case of hydrogen air mixture. Combustion processes for various conditions are investigated. The effects of cavity configuration on pressure and heat flux histories are shown. The bigger the cavity volume becomes, the larger the pressure oscillation and the value of heat flux become.

View full abstract

An experimental and numerical study on soot formation in a counterflow methane/oxygen diffusion flame was made. In the experiment, PAH concentration distributions in the flame was studied by using laser induced fluorescence (LIF), and soot volume fraction was studied by using the light scattering technique. In the numerical study, the flame structure and the gas-phase PAH concentration distributions were calculated with detailed reaction mechanisms for combustion and PAH reactions. The predicted distributions of temperature and main species concentrations correlated satisfactory with those observed in the experiment, suggesting that the numerical calculation is reliable. The comparison of the observed LIF intensity distributions with the predicted PAH distributions has shown that the LIF possibly comes from Benzene. The comparison of the light scattering intensity and the predicted PAH distributions has suggested that Naphthalene should play a key role in the soot nucleation process.

View full abstract

Distributions of fuel vapor concentration, fuel vapor velocity and ignition region in a transient fuel free spray in a rapid compression machine were visualized two-dimensionally by a laser sheet scattering technique called as Silicone oil droplets Scattering Imaging (SSI) method. The residual silicone oil droplets suspended in fuel vapor are illuminated by a sheet of light from copper vapor laser flushing at 20 kHz, and the blight scattering from silicone oil droplets was imaged by a high speed camera synchronized to the laser. The distribution of fuel vapor velocity was analyzed from sequential SSI images by a cross-correlation PIV technique. The ignition region in a spray could be detected as dark regions where the silicone oil droplets were vaporized by local heat release due to ignition. Distibutions of fuel vapor concentration, velocity and ignition region in a transient spray obtained revealed that ignition starts behind the head vortices in a spray where surrounding air is entrained into spray strongly.

View full abstract

In our previous work, the preferential diffusion effect was found to play an important role in premixed turbulent combustion characteristics, where nitrogen was added as an inert gas to several fuel/oxygen mixtures. In this study, the inert gas is changed to Ar, He and CO₂, and the relationship between the turbulent burning velocity and the diffusion coefficient of each inert gas is examined experimentally. As a result, Ar and CO₂ added mixtures show slightly larger turbulent burning velocities than that of nitrogen added mixtures. On the contrary, He added mixtures show very smaller ones at the same equivalence ratio. These characteristics are discussed in connection with each diffusion coefficient of fuel, oxygen and inert gas in multi-component system.

View full abstract

ATAC (Active Thermo-Atmosphere Combustion) is autoignition combustion in two stroke engines, which occurs by diluting fresh gas with residual gas to keep high temperature at low load operation. In this study, a two-stroke ATAC engine test was carried out to obtain fundamental knowledge for controlling the autoignition and combustion characteristics in the premixed charged compression-ignition lean combustion engine. The influences of delivery ratio, equivalence ratio and adiabatic compression speed on autoignition timing, autoignition temperature and combustion duration were investigated. ATAC autoignition temperature and combustion duration do not depend on delivery ratio and equivalence ratio, but is determined by the individual fuel characteristics. Increasing adiabatic compression speed, ATAC autoignition temperature was lowered. ATAC autoignition occurs at the crank angle when the mass-averaged temperature of the mixture reached a constant temperature that belongs to the individual fuel. Then autoignition timing depends on mass-averaged temperature at port closing timing. But over a certain limitation of delivery ratio, autoignition does not occur nevertheless the temperature at this timing is same, we have to analyze more about this result.

View full abstract