The artificial compressibility algorithm for incompressible flows is applied to an unstructured grid method. It is based on a finite volume cell-vertex upwind technique and uses a time-marching procedure to efficiently solve the continuity equations 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. Numerical simulation is performed for steady and unsteady incompressible flow problems, 2-D wake of cylinder, 3-D stenosed tube and branched tube.

View full abstract

In the 1st report, numerical results were presented for the swirling flow field in a rotating hollow turbine shaft. The existence of a rotating spiral vortex at the place where the swirling flow turns radially outward was shown. The first non-axisymmetric mode of a single spiral vortex was transformed into the second mode of a double sprial vortex at a specific rotating speed of the shaft. In this report, the downstream region of the computational domain is extended to the wheel space, the cavity between the corotating turbine disks, to solve the swirling flow field in the internal cooling air system of a gas turbine. The data on precessing frequencies of the rotating spiral vortex in this numerical analysis are compared with experimental results in a companion paper(3rd report). In addition, attention is paid to the three-dimensional swirling flow field in the rotating cavity with the rotating spiral vortex in the straigth tube.

View full abstract

In our previous research, the modified Galerkin method was proposed as one of the most efficient methods for the convection-diffusion problems and the two-dimensional viscous fluid flows. In the modified Galerkin method, since the inertia term is considered explicitly, only the symmetrical matrixes appear. Then an artificial viscosity is introduced through an error analysis approach to improve its accuracy and stability. In this paper, this modified Galerkin method is applied to the three-dimensional incompressible viscous fluid flow analysis coupled with the Semi-Implicit Method for Pressure-Linked Equations formulation. The cubic cavity flow problem sare investigated for the Reynolds number of 400, 1000, 2000 and 3200. Finally, we confirmed the effectiveness of our proposed method.

View full abstract

In this paper, three-dimensional FEM analysis of visco-elastic fluids in a slit is carried out. Here the GSMAC-method is employed for the analysis of the momentum equation and the MUSCL-TVD method for the analysis of the constitutive equation. And we use Larson model as a constitutive equation of visco-elastic fluids. In order to investigate the validity of this present analysis, the present solutions are compared with the experiment data of the HDPE-melts in the former literature. The present solutions of principal stress difference are in good agreement with the experiment data. We also investigate effects of flow rate in the flow field.

View full abstract

In the previous work, we investigated the transmission behavior of deformation, caused by a lateral impulsive hit, on a viscoelastic fluid jet falling vertically in air driven by the gravity. In this work, the jet is subjected to a lateral periodic deformation at a point. A standing wave is formed on the jet at a suitable frequency and position of the deformation. The wave length of the standing wave is strongly affected by the frequency of the deformation, the deformation position and concentration of the polymer solutions. Since the flow rate does not affect the shape of the standing wave, it is found that the transmission velocity of the periodic deformation is independent of the falling velocity of the jet and the deformation transmits with a constant velocity along the jet. The normal stress acting on the jet can be calculated from the transmission velocity of the standing wave with fluid density. The elongation viscosity is estimated by the normal stress and elongation rate evaluted by the jet shape.

View full abstract

Particle-in-cell simulation is carried out for the expansion of a multi-charged dense plasma for X-ray laser systems. All Coulomb collisions among charged particles are taken into consideration by using Nanbu's theory of Coulomb collisions[Phys, Rev.E55(1997)4642], together with our recent extension to the case when the weight assigned to a simulated particle varies from particle to particle [Trans.Jpn.Soc.Mech.Eng.64, B(1998), 2512]. The effect of Coulomb collisions on the expanding plasma is found to very large, e.g.the splitting of temperature components during the expansion is large reduced. The elecvtron temperature is nearly spatially uniform, and it decreases with time. The charge neutrality condition is broken near the ion front, and the electric field reaches its peak at the ion front. In the expanding region, the ion density decreases exponentially with radius γ, and the ion flow velocity increases linearly. These features of ions agree well with the self-similar hydrbyonamic solution for the planar expansion.

View full abstract

The measurements of the velocities of water and abrasive particles in abrasive waterjet are carred out. Double-exposured photographs of the abrasive waterjet are taken using a red flash and a blue flash. The water velocity is obtained by the PIV method that evaluates the cross-correlations between the red omages and the blue image extracted from the photographs. For the measurement of the abrasive particle velocities, PTV method, that traces movement of each particle in the photograph, is employed. The velocities of water and abrasive particles are successfully measured by this system. The relation between the cutting capability and the kinetic energy of water and abrasive particles are examined. It is also shown that the condition of abrasive waterjet can be classified into three regions in respect to the abrasive flow rate.

View full abstract

Three-dimensional Eulerian air velocities and Lagrangian particle trajectories are numrically simulated to describe the effect of particle existences on a high Re number(Re=10⁴)gas-particle turbulent jet using a two-way coupling and Large Eddy Simulation in which the effects of particle existences on subgrid-scale flows have been taken into account. The calculated rasults of air and particle turbulent characteristics(mean velocity distributions and turbulent intensity distributions)are in good agreement with experimental data obtained using a laser Doppler anemometer. Comparison of the instantaneous air vorticity iso-contours of gas-particle and clean air jets shows the production of vortices and eddies in the initial and the transitional regions and the reduction of air turbulence in the developed region by the presence of particles. Based on the model for the effects of particle existences on subgrid-scale flows, the flow states which reduce or enhance the air turbulence in high Re number gas-particle flows have been discussed.

View full abstract

To give appropriate assesment of propellant management systems for space application, the behavior of two-phase flows under low gravity conditions was both experimentally and numerically studied. By use of a numerical analysis method with C-CUP(CIP Combined Unified Procedure)scheme coupled with Level Set method and CSF(continuum Surface Force)model, the flow fields in a liquid rocket propellant tank were simulated. Main concern was placed on the generation of a deep dip of the gas phase. The numerical results were compared with corresponding experimental data obtained in a drop tower, and they showed good agreement. The generation of the dip was clearly revealed, and the phenomenon of gas suction was properly reproduced in the simulation. The effect of gravity force on the dip generation was studied, and it was found that the dip generation was alleviated with increase in the gravity force.

View full abstract

The flow around a rotating circular cylinder placed at various hights above a moving plate has been investigated numerically and experimentally for the high Reynolds number(Re=3.3×10⁴∼6.6×10⁴). The flow was affected considerably by the boundary layer on the moveing plate and the initial conditions of the experiment. Starting from different initial conditions, two different flow patterns were observed for smaal gaps(0.03&inE;H/D&inE;0.12)under the same experimental conditions. The transition of the flow pattern occeurs when the jet through the gap swiches from the moving plate to the cylinder wall. This phenomenon is known as some kind of the Coanda effect. Computation of the two-dimensional unsteady Navier-Stokes equations was also conducted. The numerical results qualitatively agreed with the expermental ones. The standing vortex was found at 2.5 diameters behind the cylinder in the case of Coanda effect.

View full abstract

The flow around rotating circular cylinder that is close to a moving plate has been investigated numerically and experimentally for the high Reynolds number(Re=3.3×10⁴∼5×10⁴). Two different flow patterns were observed under the same conditions changing peripheral velocity of the circular cylinder. The transition of the flow pattern occurs when the jet through a small gap(H/D=0.1)switches from the moving plate to the cylinder wall. This phenomenon is known as some kind of the Coanda effect. Furthermore the Coanda effect was accompanied by hysteretic effect. Computation of the two-dimensional unsteady Navier-Stokes equations was also conducted, The numerical results qualitatively agreed with the experimental ones. It was founs that the Coanda effect and the hysteretic phenomenon were greatly related to the standing vortex that exists behind the cylinder.

View full abstract

A cyclone separator is widely used in many industrial fields. However, it is difficult to separate particles, which have sub micron sizes, from many different size ones using traditional cyclone separators. Therefore, we produced new-type cyclone separator in order to accomplish high accuracy separation of particles. The purpose of this study is to clarify relationship between turbulent characteristics of swirling flows in the cyclone separator and motions of particles. In this paper, we measured swirling flows in the cyclone separators, which have different heights of outlet pipe, using a slanted hot-wire sensor. Based on the mean velocity data, the calculation of particle trajectories with various diameters in the swirling flow fields is carried out in order to estimate the classification efficiency of the cyclone separator. As results of these investigations, it made clear the influences of heights of outlet pipe on the swirling flows in the cyclone and the classification efficiency.

View full abstract

Turbulence measurements have been performed by using an LDV in a fully developed squareduct flow at low Reynolds number Re〓7000. A local wall shear stress, turbulence intensities and Reynolds shear stresses, -(u'v')^^^ and -(u'w')^^^, are presented to discuss the consistency of recently reported low-Reynolds-number DNS data(Gavrilakis 1992;Huser and Biringen 1993)with the experimental results. In contrast to high-Reynolds-number cases, the spanwise (z-directional)profile of a wall shear stress has two peaks in 0≤z/H≤0.5, one of which is located at a wall bisector z/H=0.5 and another of which is around z/H=0.1, and the maximum appears at the wall bisector. Such low-Reynolds-number behaviour has also been observed for the wall stress in the DNS. However, some of turbulence statistics, in particular turbulence intensities, reported by the DNS study are found to be inconsistent with those of the present experiment. This could suggest the possibility that DNS fails in predicting proper statistics if the axial extent of a computational domain is not taken to be long enough to capture a large-scale structure in the axial direction.

View full abstract

In this study, the particle-laden flow in a wall turbulence is studied by the Large Eddy Simulation(LES). The results are compared with the Direct Numerical Simulation(DNS)conducted by Rouson et al.(1997). The wall Reynolds number(Re_τ)is 180 and the particle diameters are assumed to be smaller than the Kolmogorov length scale. The one-way coupling method is employed for the interactin between the fluid and the particle. The Sub-Grid Scale(SGS)model dependencies on the numerical results are analyzed. Smagorinsky, dynamic SGS and scale similarity modeles are used for the comparison. Although the results simulated by the LES show reasonably good agreement with the DNS results, the difference between the DNS and the LES is larger than that among different SGS models. Further, the stochastic effects of the particle dispersion due to the SGS fluid motion are modeled and investigated. The results show that the stochastic effect is small enough in the present conditions.

View full abstract

The performance of an active feedback control with wall deformation was tested by direct numerical simulation of a fully developed turbulent channel flow. The local wall movement is determined based on the flow condition detected by virtual sensors distributed in the vicinity of the wall;i.e., the wall velocity is made opposite to the wall-normal velocity at y/δ=0.1. The turbulent friction drag is reduced by 10% on average. The gain in the pumping power reaches about 28 times of the energy consumption for actuating the wall. The primary mode of the resultant wall velocity distribution, which should be most effective in the present drag reduction scheme, is found to have wavelengths of about 300 and 50 viscous wall units in the streamwise and spanwise directions, respectively, while the time period is on the order of the time scale of the quasi-coherent vortical structure of near-wall turbulence. Effects of active wall deformation on quasi-coherent structures are investigated by a conditional averaging technique. With the present control scheme, the Q2 vortex is displaced away from the wall and the wall-normal and spanwise velocity fluctuations associated with the Q2 event are decreased. On the other hand, the location of Q4 vortex from the wall remains unchanged while the vorticity of Q4 vortex are substatially decreased. In the region downstream of the Q2 cvent, the wall is deformed as a shallow groove, which stabilizes the near-wall streaky structures along it.

View full abstract

Passive control method of pseudo-shock wave with bypass system is experimentally investigated. The principle of the method is to feed back the high pressure at downstream of the pseudo-shock wave to upstream through porous wall and a bypass channel that is set parallel with the main channel. The result shows the total pressure loss is reduced, but the pressure recovery is slightly decreased and the length of the pseudo-shock wave is increased by 3∼5 times of the hydraulic diameter. The position of the first shock wave is fixed at the leading edge of the porous wall against back pressure ratio as large as 0.02. Thus the pseudo-shock wave is considerably stabilized both statically and dynamically.

View full abstract

Experiments and 2D linear stability analysis of the behavior of flexible thin sheets cantilevered at the trailing edge in a uniform flow are presented. In the experiment, the sheet displacament is measured temporally with a laser displacement sensor. In the stability analysis, the problem is reduced to an eigenvalue problem by representing the sheet flexural vibration equation for the sheet motion by the finite difference method and the flow field by the vortex method. Reasonable agreement is obtained between experimental and analytical results. Flutter occurs at a lower flow speed. As we increase the flow speed, the flutter ceases once and divergence occurs at a higher flow speed. The reduced frequency increases as the mass ratio increases. The first order bending mode is observed for flutter and divergence. The frictional force makes the flutter onset region smaller and the divergence onset region larger.

View full abstract

This paper describes fundamental experiments on active noise control applied to a ducted model fan. An acoustic modal cancellation is adopted for dealing with spinning modes in fan noise. Two experiments were carried out to verify the feasibility of reducing spinning modes. At first, the modal cancellation of single spinning mode was demonstrated by use of a modal fan as a noise source and acoustic driver units as anti-phased secondary sound sources. The spinning mode in a first blade passage frequency was attenuated by as much as 15 dB with little spillover of the secondary sound. Next, modal cancellation of multimodes was assessed by two approaches of simultaneously suppressing different spinnng modes. One approach is to separate secondary sound sources so that each sound source relaases corresponding spinning mode. The other is to superpose control signals for spinning modes and to release the superposed signal through single sound source. As a result, spinning modes in both a fundamental blade passage frequency and the higher harmonics could be suppressed with little spillover of secondary sound.

View full abstract

In this study we investigated ways to reduce the noise caused by refrigerant flow during the cycleheating dehumidification operation. In this operation, the indoor heat exchanger is divided into a condensing part and an evaporating part by a dehumidification valve which is located between these two heat exchangers. The noise source is from the two phase flow passing through this valve. To reduce electric power consumption it is effective to reduce the compressor input power, that is lower the rotational speed of the compressor. At first, we examined the relationship between the kinetic energy of refrigerant flow and noise, and found a linear relationship between these values. It means that reducing the flow rate is important for reducing the noise. Then, we developed a new valve that has a smaller opening with a notch type throttle. We achieved electric power and noise reduction for the dehumidification operation by using this new vaelve under a reduced flow rate.

View full abstract

The effect of operating methods of wind turbine-generator system on net power extraction under wind speed fluctuations is discussed in relation to the dynamic behavior of the system. The system is composed of a Darrieus-Savonius hybrid wind turbine and a load generator. In this paper, two types of operating method are examined; constant tip speed ratio operation for stand-alone power systems(Scheme 1)and synchronous operation for utility power systems(Scheme 2). The computed results of the net extracting power using our dynamic simulation model show that the dominant factor of power extraction in Scheme 1 is the dynamic characteristics of rotational components and that it is important to select the appropriate rated wind speed in Scheme 2. Thus, it is concluded that a conformable operating method and rated power output of the system exist for each wind condition. In particular, small-scale systems, which are smaller than approximately 10 kW-system range, are desirable to be operated under a constant tip speed ratio as stand-alone power systems.

View full abstract

Detailed observations have been conducted on the interfacial deformation of the silicone oil surface and a water drop fallen on it. Eleven kinds of silicone oils with wide variations of kinematic viscosity, ν_T=1∼10⁵mm²/s, have been tested. The oil surface is disturbed by a water drop with a diameter d_L=3.1mm, which falls freely on it from on height of 100mm∼1000mm. Special attention has been directed to the maximum depth of the cavity formed on the oil surface D_M, and to the maximum diameter of the water drop spreading on the oil surface d_M. We have categorized the configurations of the oil cavity into seven patterns, and those of the water drop at the oil-water interface into five patterns. The maximum cavity depth D_M/d_L can be well correlated by a dimensionaless group Re_TLWe_TLwhere Re_TL is Reynolds number based on d_L and ν_T and We_TL is Weber number with the water density and surface tension of oil. The maximum diameter of the impacting water drop d_M/d_L can be correlated by the Reynolds number with a viscosity of water(Re_L)and the Ohnesorge number(Oh). Moreover, the condition under which the impacting water drop is smashed into pieces has been also examined based on Re_L and Oh.

View full abstract

Directly applying power generation technlogy for large-scale hydraulic turbines to a small hydraulic power generation system inevitably results in high costs. Consequently, technology specific to small hydraulic power generation must be developed. For example, a technology to use reverse running pump turbines, wherein mass-produced centrifugal pumps are operated in reverse rotation to obtain power, has been gaining popularity. The 1 st report gave the detailed test results on grasping the turbine characteristics and clarifying the internal flow. On this research, the radial thrust in the turbine mode and at the runaway speed condition was measured and this was evaluated comparing with the measuring data of distributions of circumferential velocity and static pressure at runner inlet and outlet, then the radial thrust tactor Kr was obtained. In addition to this, detailed experiments on the internal flow at the runaway speed condition were carried out to clarify the runaway speed characteristics, further the possibility of predicting the runaway speed characteristics was studied.

View full abstract

There have been quite few experimental investigations in the area of condensing heat transfer to cryogenic fluids. The investigations with nitrogen and exygen have shown reasonable agreement between experimental data and those predicted by Nusselt's theory. On the other hand, in the previous investigations with much colder fluids, such as hydrogen, deuterium and helium, the experimental condensing heat transfer coefficients are smaller than those predicted by Nusselt's theory and those differences become much larger when film Reynolds number or decreasing temperature difference across the condensate film is decreased. In the present investigation, hydrogen and nitrogen were condensed inside a vertical tube(d=15mm, L=30mm)under steady-state conditions respectively and condensing heat transfer coefficients were precisely measured. From the experimental resuls, the condensing heat transfer coefficients for saturated hydrogen and nitrogen vapors agree with those predicted by the Nusselt's theory within ±20%. The results of the present study suggest that deuterium and helium might also behave as predicted by the Nusselt's theory.

View full abstract

An ice storage is excellent in the cold thermal storage systems. We had been studying on a suspension made of an emulsion as a new thermal storage material. The emulsion was formed by an effect of static electricity generated by stirring an oil-water mixture. According to cooling of the emulsion with stirring, it changes into the suspension which was a slush ice. The purpose of this study is to prove formation of high IPF suspension without adhesion of ice to a cooling wall. In the previous report the result that an additive was necessary to make the suspension consist of dispersed granular ice was obtained. So the suitable additive was selected and was used in experiments. From the experimental results, it was clarified that formation of high IPF suspension without adhesion to a wall depended on a material of vessel. Besides, the formed suspension IPF was over 70%, and preservation of the suspension in a refrigerator for a long time was possible.

View full abstract

This paper presents the experimental and simulation results for thin-layer electrochemical depositions. The effect of interaction between electrodeposits on morphology is investigated around the two circular deposition plates. The experiment is carried out on the two circular plates in Hele-Shaw cell. For comparing with the experiment, a Monte Carlo simulation is performed for the two-dimensional electrochemical deposits around the two circular plates. The mass transfer(diffusion and electric migration), the electric field and the surface reaction are taken into account in this simulation. The morphology obtained from the simulation agrees with experimental result.

View full abstract

The authors proposed the snow cooling system of direct heat exchange between hot air and snow block with a vertical hole. In this snow cooling system, the characteristics of heat exchange is little affected by the amount of remaining snow, and the outlet air temperature is almost constant, and be enough low to use for the actual air conditioning system. So this system has become to be adopted widely. In this study, the suthors investigated the effect of initial snow height on the characteristics of heat exchange by the experiment with models of practical scale. Experimental results showed that heat transfer coefficients on the snow surfaces of top and bottom plane were not affected by initial snow height, but the heat transfer coefficient of snow hole increased with decreasing snow height being affected by the entrance region. It was found that heat transfer coefficient of snow hole approached to a constant value on condition that snow height was over 2 meters. A empirical equation compensated by the initial height of snow block was obtained from the experimental results.

View full abstract

Heat and mass transfer characteristics within a reforming catalyst bed have been analytically investigated. A numerical analysis was carried out in a two-dimensional steady-state model of reforming catalyst layer. Reforming tube was filled with catalyst and the tube wall was uniformly heated, a mixture of steam and methane was reformed through the catalyst bed. Predicted temperature, formed gas composition, methane conversion rate, and heat transfer coefficient distributions in the catalyst layer showed good agreement with experimental data. The effects of space velocity, steam carbon molar ratio, and wall temperature on the heat transfer coefficient were analytically presented. From temperature and composition distributions simulated by two-dimensional analysis, the effects of these factors above mentioned and diffusion on the transport phenomena were qualitatively predicted.

View full abstract

This paper describes the transient process of the impingement heat transfer and flow in a narrow space with confined wall from a single circular nozzle. Three dimensional governing equations were solved in unsteady condition for water by the SIMPLE method and QUICK scheme. At first, the generation process of a recirculated region near the upper insulated wall was clarified. Three heat transfer regions, namely, two-dimensional forced convection region, three-dimensional mixed convection rgion and natural convection region move along raedial direction with time. Local heat transfer corresponding to the flow process takes the meximum and minimum values along the radial and peripheral direction.

View full abstract

We have experimentally studied the influence of buoyancy force and inlet flow conditions on the laminar-turbulent transition process of fully-developed air flow in a heated horizontal tube with uniform wall heat flux at modified Rayleigh number Ra_q=3.1×10⁶. Eight time-series of the air temperature were simulaneously obtained using eight thermocouples positioned within the tube along a vertical line passing through the tube axis. We have determined the nature of the transition behavior in space and time by analyzing these instantaneous time-series.

View full abstract

This paper presents a numerical analysis method for shape determination problems of unsteady heat-conduction fields in which time-histories of temperature distributions on prescribed subboundaries or time-histories of gradient distributions of temperature in prescribed subdomains have prescribed distributions. The square error integrals between the actual distributions and the prescribed distributions on the prescribed subboundaries or in the prescribed subdomains during the specified period of time are used as objective functionals. Reshaping is accomplished by the traction method that was proposed as a solution to shape optimization problems of domains in which boundary velue problems are defined. The shape gradient functions of these shape determination problems are derived theoretically using the Lagrange multiplier method and the formulation of material derivative. The time-histories of temperature distributions are evaluated using the finite element method for space integral and the Crank-Nicolson method for time integral. Numerical analyses of nozzle and coolant flow passage in wing are demonstrated to confirm the validity of this presented method.

View full abstract

A new concept of gasification power generation systems fueled by coal, RDF and various kinds of wastes using high temperature air is proposed. Cycle analyses of these IGCC and boiler systems are done. The calculation shows that heating value of syngas almost doubles when we increase the gasification air temperature from 25°C to 1000°C. More than 45% thermal efficiency is obtainable both for coal and RDF by employing the new high temperature air blown IGCC system. This IGCC system is applicable to various wastes including low grade wastes such as sludge.

View full abstract

A numerical study was made to identify the most appropriate criterion to describe the spontaneous ignition of an externally heated solid fuel. Several ignition criteria were tested if they can describe correctly the effects of the gravity and ambient oexygen concentration on the ignition characteristics. The gravity affects significantly the ignition behavior since the buoyancy induced flow plays a key role in the transient mass and heat transfer processes, leading to the two distinct types of ignition^<(1)∼(3)>. Seven ignition criteria, which were often used in 1-D ignition process, were evaluated in this study. It was found that some of the criteria based on temperature profile lead to predict the occurrence of false ignition. This is because that complicate transport phenomena caused by buoyancy induced flow affect the transient temperature profile strongly, to produce the situation where the criteria were satisfied instantaneously. The study has shown that the criterion based on the local reaction rate was the universal and most appropriate one.

View full abstract

In Direct Internal Reforming Molten Carbonate Fuel Cells(DIR-MCFC)deterioration of catalytic activity takes place in the anode channel due to both the liquid-phase pollution and vapor-phase pollution. The liquid-phase pollution meant that catalytic activity is deteriorated by the molten salt's(62 Li₂CO₃/38K₂CO₃)adhering to the catalyst. It can be solved by installing the protective barrier in the pollutant pathway. On the other hand, the vapor-phase pollution meant that that catalytic activity is deteriorated by KOH adhering to the catalyst. Because the vapor-phase pollution is caused in the entire electrode, an effective defense method has not established yet. Moreover, a reactivation method of vapor-phase polluted catalyst has not been developed yet. In order to study the reactivation method, the adhesion form of potassium compounds in the polluted catalyst under the various gas conditions was evaluated by using a thermogravimertic analyzer in which water vapor can feed. Additionally, the activity of catalyst treated demonstratively was also tested by a differential reactor. As a result, KOH changes to K₂CO₃ under the condition which CO₂ concentration is larger than 25%. The catalyst with K₂CO₃ cannot reactive. However, the activity of polluted catalyst is revived until 80% of initial activity by controlling the gas species concentration, especially for CO₂. Based on the results obtained by these fundamental experiments, the reactivation methods of polluted catalyst are proposed as follows; i)Catalyst should load more in the upstream in the anode. ii)In order to reactive the polluted catalyst, the ratio of H₂O to CH₄ in the fuel should increase, when DIR-MCFC is under operation. iii)Gas compositions under cell maintenance mode should be applied in the case that power generation quits.

View full abstract

This paper reports the effects of applying methods of electric field on a sooting diffusion flame. Essentially, two types of electric fields were tested in this experiment. One is a uniform electric field formed between parallel plate electrodes. The other is a non-uniform field formed between needle and plate electrodes. In the uniform field, the flame shape was changed by positive ions moving in the fields. Soot emission was suppressed with increasing the applied voltage, because an increase of the flame surface temperature promoted the soot oxidation. In the non-uniform field, on the contrary, ionic wind was formed by corona discharge. The ionic wind blew from a needle electrode toward a plate electrode, irrespectively of the polarity of the needle. The flame was inclined toward the plate electrode by this ionic wind. As the result, the surrounding air was entrained in the flame, and the suface temperature of flame also increased with increasing applied voltage. The soot suppression efficiency of the non-uniform electric field was higher than that of the uniform one.

View full abstract

CH₄-O₂-N₂ laminar inverse diffusion flames are investigated experimentally on the flame shape, stability and thermal structure. Experimental parameters are the fuel velocity, the fuel mole concentration, the oxidizer velocity, and the oxidizer mole concentration. The results obtained are as follows. In laminar flow region, inverse diffusion flames are categorized into eight types by changing the four parameters. High temperature region moves from the flame tip to the flame base by increasing of the fuel flow rate in the case of inverse diffusion flames, therefore the flames formed in the high fuel concentration are more stable. The inverse diffusion flames have less stable feature compared with the normal diffusion flames formed on the same burner. This is because the flow rate at the flame base of inverse diffusion flames is faster than of normal diffusion flames.

View full abstract

A fiber Laser Doppler Velocimeter(LDV)was used to measure a flow field behind a bluff body at high fuel flow rate condition which was considered to be close to the real operating circumstances in reacting flow. These results were compared to that of low fuel flow rate condition in smooth combustion in order to understand the reason of the onset of rough combustion characterised by flame stretch, noise generation and oscillating combustion. It was found that as the fuel flow rate increased, the reverse flow regions started to show very complex shapes because of the fuel jet penetration. At the end, the recirculation zones combined surrounding the jet. This recirculation zone combination hindered the growth of the recirculation zone itself. Therefore, the air entrainment flow rate into the flame holding regions was not proportional to the increase of bulk air velocity and fuel flow rate. Oscillating frequency was observed in velocity power spectrum at high fule flow rate. The very sharp peak at high frequency was due to the vortex shedding. The main reason of rough combustion at high fuel flow rate were found to be the lack of air entrainment caused by the combination of the recirculation zones together with the vortex shedding behind the bluff body.

View full abstract

The purpose of this paper is to study the feasibility to utilize methane, which is the main component of natural gas and biogas, as a fuel for lean-burn engines. To realize this, however, there are mainly two problems to solve; the substantial decrease in the burning velocity and the large increase in the misfire probability in the lean mixture region. It has been shown that such problems can be solved by adding hydrogen to the lean methane mixture. By adding only a small amount of hydrogen to the lean methane mixture, the turbulent burning velocity is substantially increased and the lean limit is greatly extended, at the same time. Furthermore the effect of hydrogen addition on the turbulent burning velocity is quantified to search for the most suitable running condition.

View full abstract

By injecting fuel at the very early stage of compression stroke and thus creating homogeneous lean mixture before ignition, (PREDIC;PREmixed lean DIesel Combustion), simultaneous reduction of NO_x and smoke was obtained. However, since increasing the mixture equivalence ratio cause knocking, it was difficult to operate at higher load conditions. In this study, in order to reduce combustion rate at high load conditions in a premixed compression ignition engine, multiple stage injection method and EGR were combined, and heterogeneous mixture was made before ignition. The engine test results showed that NO_x emissions could be reduced to less than 50 ppm, without knocking even at full load conditions. In addition, smoke emissions were also maintained below invisible level. It can be understood that the premixing of fuel was advanced, smoke was reduced, and EGR rate was increased, resulting lower heat release rate and NO_x emissions.

View full abstract

We experimentally evaluate a performance and exhaust emissions of a single-cylinder, fourstroke direct injection diesel engine operating on diesel fuel containing 50 and 60% volume waste food-oil(blended oil)and also a transesterified vegetable oil(TVO). The combustion characteristics and emissions such as NO_x, CO, HC, HCHO and smoke are compared with data for the case of JIS #2 diesel fuel. We found that(i)the NO_x emissions are similar for blended oil and TVO, though the NO_x with diesel fuel is lower compared to blended oil. The HC and HCHO emissions are similar for all fuels at all load conditions, (ii)smoke emissions are lower with the waste food-oil and much lower with TVO compared to diesel fuel in the whole engine operating range, and (iii)carbon deposits on cylinder head are observed after 400 hours running. Also discussed are the usability of the blended oil and TVO as alternative fuel, as well as availability for a treatment process of the waste food-oil.

View full abstract

Transient characteristics of the exhaust gas emissions of THC, NO_x and so on during starting operation in a DI diesel engine and the influence of engine operating parameters and fuel properties on them were investigated. The results indicated that THC concentration just after onset of starting increased until about 50 th to 200 th cycle after the starting when the fuel deposited on the combustion chamber showed the maximum, and then decreated to reach a steady state value after about 1000 cycles when the piston wall temperature became almost constant. NO_x showed an initial higher peak at about 6th cycle after the starting, and increased to reach the steady state value after about 1000cycles. Exhaust odor during starting had a strong correlation with the THC while the odor decreased much after reaching the maximum. Except just after the onset of starting, an increase in piston wall temperature resulted in a decrease in the THC concentration and an increase in the NO_x. The THC increased significantly with increased fuel injection amount and with decreased cranking speeds while there was little effect of fuel injection timings. The THC during the starting increased with fuels with higher viscosity, higher 90% distillation temperature, and lower ignitability.

View full abstract

Many papers concerning air quantity of 4-stroke cycle gasoline engine have been published. It has been reported in these papers that heat transfer in surface of cylinder and inlet port gives big influence to the air quantity. But it has not been clear which influence of heat transfer in cylinder and inlet port is strong. So the autors derived a function of the air quantity thermodynamically considering heat transfer, and examined which of influence of heat transfer was strong. The results show;(1)The influence of heat transfer in cylinder is small(about1%)at full load, and is also small(about 5%)at light load.(2)Heat transfer in cylinder almost increases the air quantity.(3)The influence of heat transfer in inlet port decrases the air quantity with around 30% greatly.

View full abstract