A numerical method for analysing the steady incompressible potential flow through an impeller and its upstream and downstream ducts in a turbomachine is presented. The 3-D flow region containing the impeller and the 2-D axisymmetric flow regions both upstream and downstream are calculated iteratively. Since the 3-D flow calculation has been described in the previous papers, a computational method of the axisymmetric flow and patching conditions are presented here. In this computation, the equation of Strokes' stream function is solved by the variational FEM using bilinear quadrilateral elements, and a vorticity component corresponding to the trailing vortices is considered in the downstream region. The flows in an impeller having different shaped upstream and downstream ducts of a single suction centrifugal pump are calculated, and the effects of these ducts are shown.

View full abstract

A numerical method for analysing the steady incompressible potential flow through an impeller and its upstream and downstream ducts in a turbomachine is presented. The 3-D flow region containing the impeller and the 2-D axisymmetric flow regions both upstream and downstream are calculated iteratively. Since the 3-D flow calculation has been described in the previous papers, a computational method of the axisymmetric flow and patching conditions are presented here. In this computation, the equation of Stokes' stream function is solved by the variational FEM using bilinear quadrilateral elements, and a vorticity component corresponding to the trailing vortices is considered in the downstream region. The flows in an impeller having different shaped upstream and downstream ducts of a single suction centrifugal pump are calculated, and the effects of these ducts are shown.

View full abstract

The equation for the growth of a bubble in a polymer solution is derived, and the equation for an equilibrium condition of the bubble is obtained. In PAM (polyacrylamide) aqueous solutions, the equilibrium radius of the bubble, the critical bubble radii, the maximum bubble radius and the maximum growth velocity of the bubble are found, and are compared with those in water. The growth critical radius, Rc, defined by the condition where the bubble grows infinitely, increases with polymer concentration. The equilibrium critical radius increases and the growth velocity of the bubble decreases with increasing polymer concentration. These results may give an explanation about the mechanism of cavitation suppression by a polymer addition.

View full abstract

In order to investigate the effect of yaw angle of Karman vortex shedding from a circular cylinder in a uniform flow, the flow pattern was observed by the smoke wire method. The pressure distribution, secondary flow in the wake along the cylinder and the velocity fluctuation, u, at a point where Karman vortex shedding was clearly observed in the u-signal, were measured. The yaw angle produced a pressure gradient along the cylinder in the dead water region behind it. Secondary flow due to this pressure gradient made the formation of Karman vortex 3-dimensional. The frequency and the regularity of the vortex shedding reduced with increasing yaw angle. When a pair of parallel thin plates was set in the wake to interrupt the secondary flow, the above effects of yaw angle were effectively reduced in the area between them.

View full abstract

Fluid forces on a centrifugal impeller rotating and whirling in a volute were analysed. It is assumed that the number of impeller vanes is so large that the flow in the impeller is perfectly guided by the vanes. Calculations are made for various flow rates and whirling velocities and the results are compared with experimental values. This model predicts the unsteady forces fairly well and simulates most of the important effects found in the experiments. It has been made clear that the unsteady volute/impeller interaction enlarges the unsteady forces and brings about a destabilizing effect on whirling motion.

View full abstract

The flow ratio and loss coefficient for a 90° branched tube are experimentally investigated for steady and pulsatile flow conditions. The Reynolds number defined at the upstream point is varied within the range of R_e=3×10⁴9×10⁴, and the frequency of the pulsating flow is f=0.171.5Hz. The main results obtained here can be summarized as follows: (1) The flow ratio of the branch tube does not show clear dependence on the Reynolds number R_e, frequency f or pulsating compliance of flow. (2) The loss coefficient of a branch tube for pulsatile flow is almost the same as the steady flow in the branch side but decreases slightly in the main branch with increasing amplitude of flow pulsation β.

View full abstract

The behavior of shock waves in a two-dimensional transonic circular diffuser has been experimentally investigated by a schlieren optical method and by pressure measurements. When the overall pressure ratio of the diffuser, φ, increases, a few weak shock waves appear near the throat section before the flow is completely choked at the throat. These weak shocks are not stable but fluctuate in the channel, and the pressure at the throat p* increases with an increasing φ, although it is not in proportion to φ. when the flow is completely choked, these weak shocks coalesce into a single normal shock, and p* increases in proportion to φ. The behavior of the shock waves in diffusers with different geometries can be plotted on a single curve in an x/√(h*R)-φ plane if the position of the shock from the diffuser throat x is non-dimensionalized by the characteristic length √(h*R), where h* and R are the throat height and the radius of curvature of the diffuser, respectively.

View full abstract

In a Ludwieg tube, the opening time of a quick-opening value which separates the high pressure section from the low pressure section is considerably longer than that of the diaphragm. In cases where the valve is located downstream of the test section, the disturbances generated by the opening of the value do not arrive at the test section after the flow at the exit of the test section becomes supersonic. On the other hand, in cases where the valve is located upstream of the test section, all of the disturbances at the valve propagate through the test section. In the present paper, the effects of the opening valve upstream of the test section wee clarified by the results of the pressure measurements for the different valves of the distance between the valve and the throat of the nozzle, the throat area of the nozzle and the plenum flow rate.

View full abstract

In this paper, equations of the one-dimensional, unsteady and compressible flow are transformed into characteristic equations to analyze the flow in a long pipe, and the characteristics of the pressure reduction in a long pipe is investigated numerically by the method of characteristics. The calculations are performed for pipes which have four kinds of bore, three lengths and five orifice ratios, and are compared with the experimental results. It was shown that the calculated results agree with the experimental results qualitatively well and that the characteristic method is available to analyze the flow in a long pipe.

View full abstract

Fully developed flows in a rectangular curved duct having a breadth/curvature-radius ratio of 0.25 are investigated experimentally. More than a 150° flow turning angle is required for the flow to reach a fully developed state in any duct which has an aspect ratio (height/breadth) of the cross-section ranging from 0.4 to 16. In case of a low aspect ratio, the main flow has a forced vortex type velocity distribution, and has free vortex type ones in a high aspect ratio. Similarity based on a vorticity defect law exists in the boundary layer on the inner wall, and an angular momentum defect law on the outer wall.

View full abstract

Similarity in the velocity distributions of the secondary flow component is examined experimentally. It is found that (1) the profile of the velocity distribution is the same one as of two-dimensional wall jets, and (2) the thickness of the secondary flow boundary layer should be considered separately from that of the main flow.

View full abstract

Developing turbulent flows in the entrance region of a rectangular curved duct are experimentally studied, with emphasis being placed on the secondary flow velocity distributions in the skewed boundary layer on the duct end wall. the results obtained are as follows: (1) Secondary flow is at its maximum at a flow turning angle of about 60°; And a turning angle of more than 150 degrees is required for the main flow to reach a fully developed equilibrium state; (2) Similarity of secondary flow velocity distributions is found at any flow-developing stage. The thickness of the secondary flow boundary layer can be defined as 2.5 times the half-width of the maximum secondary flow velocity; (3) Secondary flow velocity distribution and skewed angle distribution can be represented by a modified Mager's model using the thickness of the secondary flow layer.

View full abstract

This paper presents the results of an experimental study on the control of the reattachment of a radial turbulent jet flow (main jet flow) discharged from a cylindrical nozzle onto an adjacent offset disc plate, with the addition of another control flow. distributions of mean velocity, turbulent velocity and wall static pressure were measured. The effects of step height and control flow rate (suction and blowing) on the flow properties at and after the reattachment point, i.e., deflection of jet center axis, reattachment point, changes of jet center velocity, turbulence at velocity maxima and jet half width were investigated. Velocity and turbulence profiles were compared with those of a two-dimensional wall jet flow and radial wall jet flow.

View full abstract

This paper describes the laminarization phenomena of the flow in an axially rotating pipe when a fully developed turbulent flow was introduced into it. The rotating boundary layer is strongly stabilized by the centrifugal force due to the rotating velocity component, resulting in a laminarization of flow. This flow laminarization has been found more remarkable in the inlet region where the nonrotating inner core does exist in the pipe section. The laminarized flow tends to be unstable as the flow goes downstream, and bursts of turbulence appear, which was confirmed by the flow visualization.

View full abstract

Measurements have been made to study the detection criterion the VITA technique which was widely utilized for detecting the bursting phenomena. The mean bursting frequency and the conditional averages of u-, v- and uv- patterns were obtained from three kinds of VITA techniques, associated with an increase or a decrease in velocity at the detection time, but not with the velocity slope criterion. It was found that the most probable threshold value K was about 0.8 to 1.2, and that an appropriate range of the non-dimensionalized averaging time T* existed between 20 and 24. For a large number of events beyond N=90, the mean bursting frequency was almost constant. Compared with the results of other researchers, the present one obtained from the VITA technique with the most probable values for K, T*, and N was found to be reasonable.

View full abstract

A turbulent model applicable to turbulent swirling flow in a straight pipe has been developed. Two models, the standard k-ε model with higher order terms in Reynolds stress equations are applied, and a modified k-ε model with the hypothesis of anisotropy of turbulence is proposed. The comparisons between the predicted flow distributions and the experimental data show that the modified k-ε model predicts the complex flow fields successfully.

View full abstract

This experimental study treated the characteristics tics of pressure surge and swirl flow in the elbow type draft tube of a Francis turbine at partial load. It was demonstrated that QS type surge or violent pressure fluctuation observed in an original elbow draft tube was the resonant condition entrained by the oscillation of pressure recovery. Thus, a modified elbow draft tube was made to reduce the surge and to improve the turbine stability. Its shape was decided following the concept that an elbow draft tube should be designed so as to secure sufficient pressure rise in the inlet conical-diffuser portion. From the experiments, it is confirmed that QS type surge did not occur in the modified tube. Further, it is found that there is a flow regime where double spiral vortex cores appear intermittently, depending on the cavitation parameter as well as the swirl parameter.

View full abstract

Conditional averages of longitudinal, normal velocity and of their products have been obtained in a turbulent boundary layer over a d-type rough surface at a momentum thickness Reynolds number of R_θ&elDot;800. Averaged properties, like probability density distribution, higher order moments etc. are also provided in order to yield information related to coherent structures. It is shown that the ensemble averaged patterns reflect the well known ejection and sweep events, even in the present low Reynolds number flow. Those ejections and sweeps both made strong, intermittent contributions to the Reynolds stress and turbulence production. Near the wall, the ejection phase plays a predominant role. The normalized averaged bursting frequency is larger for the present rough wall flow than for the smooth wall flow at almost the same Reynolds number. It is inferred from the increase of both the width of intermittent region and the entrainment rate that the width of highly irregular interface between turbulent and non-turbulent flow increases with a decreasing Reynolds number.

View full abstract

A loss coefficient for a sudden expansion, based on the velocity v₁ in the smaller pipe, can be well approximated by Borda-Carnot's equation under the present experimental conditions, m≦0.4295 and R_el&elDot;(3)×10⁴, where m=(D₁/D₂)², the area ratio, and R_el=v₁D₁/v, the Reynolds number, and where D₁ and D₂ are the inner diameters of the smaller and the larger pipes respectively. Measurements of a loss coefficient ζ_c for a sudden contraction and a loss coefficient ζ for a sudden expansion and a sudden contraction closely placed with a distance L were made under the same conditions. The measured value of ζ_c was found to depend on R_el as well m. Meanwhile that of ζ increases linearly with L when L/D₂lE:2, regardless of m and R_el. The gradient dζ/dL gradually decreases with further increases in L, and ζ agrees well with the value calculated by neglecting the interaction between a sudden expansion and a sudden contraction when L_D2≧. Empirical formulae for ζ_c and ζ are presented.

View full abstract

The experiments were conducted to explore the effects of the mutual interaction between two closely spaced Savonius rotors. The effects of the distance X between two rotor-axes, the relative phase angle a of the buckets of two rotors, the directions of the rotation and the wind direction on the power and torque coefficients were investigated using the wind tunnel. The optimum set of the parameters was counterrotation (inside direction), X=1.3R(R: rotor diameter) and a=90°, and in this case the mean power coefficient of the rotors was about 12% larger than that of the single rotor. This advantage could be expected if the change of the wind direction was less than ±35°.

View full abstract

With regard to high pressure cryogenic pumps, the adiabatic efficiency should be used instead of the ordinary pump efficiency, because the comparatively great compressibility of the cryogenic fluids influences the pumps efficiency. In the present study, a quantitative estimation was made by both a brief calculation and an experiment in order to know the difference between the adiabatic and ordinary pump efficiency. It was made clear that regarding high pressure liquid hidrogen pumps, a considerable correction was necessary if the ordinary pump efficiency was used as the pump efficiency. It was also confirmed that the adiabatic efficiency could be experimentally obtained with pretty small uncertainties.

View full abstract

Oil-hammer or water-hammer caused in various fluid pipelines usually needs to be investigated in relation to valve stroking over a specified time. This paper deals with the pressure response solutions of oil-hammer caused by finite-time valve stroking. A fundamental equation used in this analysis is that for the fluid line model in which the frequency-dependent effect of viscosity is taken into consideration. It is confirmed that the theoretical estimation agrees well with experimental results obtained under various system parameters.

View full abstract

Concerning the complicated peculiarity of flashing flow, the paucity of reports relating its flow peculiarity, except for bubble flow and annular flow, prompted us to investigate it. In this paper we describe an experimental study of the behavior analysis of a stream-water single slug in an adiabatic vertical pipe. We measured slug growth rates, liquid temperatures and steam temperatures in the slug, etc. We experimentally clarified the relationship between slug growth rates and the degree of super heat, steam temperature in the slug and saturation pressure. Furthermore, we have made measurements of velocity profiles of liquids at the nose of the slug by the use of LDV.

View full abstract

A theoretical analysis is presented for the conjugate conduction-convection problems of a free convection fin with an infinite length, embedded in a porous medium of infinite extent. In view of practical applications, the analysis takes account of a possible thermal stratification within the surrounding porous medium. An integral procedure is exploited to solve the boundary layer equations for the Darcian fluids, while the thin-fin approximation is evoked for the quasi-one-dimensional heat conduction equation for the fin. Similarity solutions re obtained for a class of the fins in which the conduction-thickness product varies according to the power function of the distrance form the origin. Some of the results are compared with available exact solutions. It will also be shown that the presence of thermal stratification within a porous medium can significantly displace the isotherms around the fin.

View full abstract

A plate-like phase change material (PCM) is initially subcooled below its fusion temperature. Melting is initiated by a step-wise temperature increase at one of the vertical boundaries of the PCM. Other boundaries are insulated. The melting process with the natural Convection in the developing melt region is analyzed by a two-dimensional finite difference method and a one-dimensional approximate analysis. The shapes of the melting front and the temperature distributions in the melt region, obtained by the two-dimensional analysis, agreed well with the results of the experiments carried out by using n-octadecane as a PCM. The one-dimensional approximate analysis predicted the melting rate and the amount of the thermal energy storage with a sufficient accuracy. Though the initial subcooling significantly decreases the melting rate, the subcooling and the aspect ratio of the PCM barely affects the rate of the total thermal energy storage including the latent heat and the sensible heat.

View full abstract

The purpose of this investigation is to implement a feedforward control system for controlling the heating surface temperature of heater blocks in all three boiling regions, particularly in the transition boiling region. This technique features an electrical heater output which is determined in advance according to the estimated steady state heat flux at the heating surface and which corrects the temperature deviation by feedback control according to the temperature signal. Test results show that this control technique can control the heating surface temperature fairly well in the transition boiling region.

View full abstract

For tube bundle of flooded evaporators, boiling heat transfer coefficients are improved by upward streaming vapor bubbes. The authors tried to clarify the relation between boiling heat transfer characteristics of single tubes and those of tube bundles. In this report, boiling heat transfer characteristics of low-finned-tubes are investigated, using bundles of 4 horizontal tubes in a vertical row with side walls. Experimental results are summarized as follows: (1) when the clearance between tube and side wall is larger than 5 [mm], the effects of side walls on boiling heat transfer of tube bundles cannot be recognized; (2) bundle effects are affected both by the heat flux of the tube itself and by the total heat flux of the lower tubes; (3) the heat flux increase by upward streaming vapor bubbles has maximum value at wall superheat 5 [K] for 1590 [kW/m²] of total heat flux of the lower tubes.

View full abstract

An analytical study of the transient process on flattening the temperature distribution of a high temperature steel slab is carried out by using the fundamental model, which is composed of the laminar slot jet and the porous medium. The momentum and energy equations are solved using an upwind finite difference representation of the equations cast into thier vorticity-stream function form in an unsteady circumstance. The flow situations are found to become almost steady in dimensionless time τ=61.54. The thermal energy, which is moved from thee center of the steel slab to the impinged surface by heat conduction, is transported by jet flow, and attributes to the temperature increase of the slab edge.

View full abstract

Heat transfer of the shell-side has been experimentally examined for air in a shell-and-tube type heat exchanger composed of a closely spaced tube bank (tube arrangement is triangular style). It is found that average heat transfer behaviors around each tube in the heat exchanger can be explained by the heat transfer in the vicinity of the front stagnation point of the tubes in the cross flow region; and the total heat transfer rate of the baffle compartment of the narrow tube spacing is not lower, compared with previous data of the general tube spacing.

View full abstract

The optimized design conditions of pebble bed regenerative heat exchangers are discussed for minimizing their pressure loss with minimum deterioration of their thermal performance. Blow-down experiments for heating argon by hot air or combustion gas are carried out to measure the pressure loss of argon in a pebble bed. Appropriate correlations for predicting the pressure loss are selected based on the measured results of the pressure loss. The optimum values of the pebble diameter and the mass flow of a working gas are determined by calculating the pressure loss and the thermal performance of a pebble bed regenerative heat exchanger under cyclic continuous operation. The use of helium as a working gas of closed cycle MHD instead of argon is advantageous from the view point of both the pressure loss and the thermal performance.

View full abstract

A combined type thermoelectric power generation system is proposed to improve the conversion efficiency of thermal energy. In that system, thermoelectric elements are directly attached to the outer wall of the channel of a working fluid without a heat transfer block in order to reduce the thermal resistance; but on the contrary, high performance heat transfer pipes can not be used. Therefore, the effective heat transfer rate of a working fluid is increased by using a larger diameter channel to increase the contact area of the fluid to the channel. The performance of the combined type power generator is experimentally and theoretically investigated and compared with that of the conventional separated type generator. The optimum condition for power generation is also discussed. It is clarified that the maximum efficiency of the combined type generator becomes higher and its optimum R_e number becomes larger compared with those of the conventional type. It is also made clear that the effect of the turbulence promoter on the maximum power output is negligible in the case of a generator having a large channel diameter.

View full abstract

This paper presents an experimental study on the ebullition process of overexpanded water droplets on solid surfaces. In this study, the effect of wettability of solid surfaces on the disintegration of water droplets due to ebullition was considered for water droplets under transient overexpanded conditions during the test runs. The solid surfaces used in the experiments were Pylex glass, nylon, PMMA, PVC and Teflon. It was clarified from this study that the nucleation rate of overexpanded water droplets on the solid surfaces increased with a decrease of wettability.

View full abstract

The cooling characteristics of a ;molten carbonate full cell were studied both experimentally and theoretically using a stack model which can simulate the heat transfer in a cell. In this model. electrically heated plates perform the role of the electrolyte/electrode, and air is used instead of both fuel and oxidant. From this study we obtained the following results: (1) The temperature distribution depends on the flow direction of the fuel and the oxidant. In co flow, temperature in the cell increases along the stream pass, while in counter and cross flow arrangements, peak temperature appears near the fuel inlet portion. (2) The temperature distribution and cooling characteristics of a molten carbonate fuel cell depend mainly on the convective heat transfer of oxidant.

View full abstract

Experimental studies are carried out on the heat transfer of open circular thermosyphons (the radius of length to radius are 6 and 10) using liquid sodium. The Nusselt numbers (N_ur) measured for liquid sodium are plotted against the Rayleigh number (R_ar) and the product of the R_ar and Prandtl number (P_r). These results cover a range of R_ar from 4×10² to 1.7×10⁵ or R_ar·P_r from 10 to 9.5×10³ and are compared to the previous results of mercury. It is shown that the arrangement of N_ur against R_ar·P_r is more appropriate than that against R_ar. The heat transfer of open thermosyphons for liquid sodium and mercury shows similar characteristics. The effect of convection flow on N_ur becomes significant over about 60 R_ar·P_r.

View full abstract

This paper describes an experiment in the analysis of the heat transport characteristics of a downward heat transport device utilizing the vapor pressure change as the liquid pumping force. The condensate stored in the accumulator returns to the evaporator because of the pressure difference between the accumulator and the condenser produced by unsteady evaporation due to the change of the liquid volume in the evaporator. The vapor pressure in the evaporator was restricted by the accumulator pressure which was almost constant because of the effect of the noncondensable gas, and good temperature control capability was obtained through the passively changing effective condenser length just as in a variable conductance heat pipe. The temperature control capability and the effective condenser length were predicted by the analysis of the simple operation model and the theoretical results agreed well with the experimental data.

View full abstract

An experimental study is made for the separated an reattached flow behind a backward facing step to clarify time-mean and instantaneous heat transfer performances near the reattaching point. The reattaching point is found to fluctuate almost in accordance with the normal distribution. By using several time-mean heat transfer experimental results of different fluctuation amplitude, the instantaneous heat transfer performance near the reattaching point is obtained. Based on this instantaneous performance, the time-mean heat transfer characteristic is calculated and found to be in good agreement with the experimental results.

View full abstract

In order to measure equivalence ratio (φ_R2) instantaneously, we developed a photoelectric photometer which could measure immediately the luminous intensity of OH, CH, C₂ radical in a flame. By using this apparatus, we investigated the effects of (1) the focusing position of the optical system to the flame front; and (2) the angle between the normal of flame front and the optical axis, with a premixed laminar burner flame. As a result, it was clarified that φ_R2 could be measured by the ratio of C₂/CH with 2.2% maximum error. (3) so, we applied this method a Burke-Schumann type diffusion flame, and made it clear that mean <φ_R2>^^- at the flame front was about 1.1 nearly equal to the value assumed until now. (4) and, for the open type diffusion flame, we made it clear that the reason for large <φ_R2>^^- at down-stream was because of insufficient diffusion and the obstruction of burned gas.

View full abstract

The phenomenon of a small droplet impinging upon a hot surface is applied in various industries. Such applications are divided into those employing atomization by the impingement of the droplet and those employing the heat transfer from surface to the droplet. The purpose of this paper is to obtain fundamental information concerning the heat transfer process and breakup behavior characteristics of individual small droplets impinging upon a hot surface. A uniform sized water droplets array at room temperature under atmospheric pressure was produced by the vibratory method to impinge upon a heated flat copper surface. And then, the deformation and the breakup behavior owing to the impingement of the droplet in observed by means of a drum camera recording high-speed microscopic photographs. The transient change in the diameter of the radial film which is formed after the droplet impinges on a surface is adjusted with the Weber number, and it is revealed that contact resistance in solid-liquid interface varies with surface temperature. The breakup form of the impinged droplet is classified into 7 types: R, RB, B, N, H, V, and F. The changes in the Sauter mean diameter of breakup droplets and the volume distribution of breakup droplets are examined.

View full abstract

Spontaneous ignition was studied on fuel sprays injected into a steady hot air flow by a hole-nozzle with a single hole. Usually, the ignition delay is the so-called Arrhenius type and makes a straight line in an Arrhenius plot. In the present study, the effect of physical factors on ignition was investigated, and special attention was paid as to how the gradient of the straight line is decided. five kinds of experiments were carried out: Where pure fuels of iso-octane and n-heptane were used in the beginning, then the mixing condition was changed with a cylinder inserted into fuel jets or a wall facing a fuel nozzle, and the spray condition was changed with a change of fuel injection pressure or with a change of water content in n-propanol/water solution fuel or diesel-oil/water emulsion fuel. As a result, it was found that physical factors change ignition delay greatly, but have almost no effect on the gradient of the straight line in an Arrhenius plot. Then, the temperature dependency of the ignition delay is considered to be controlled almost completely by the chemical factor.

View full abstract

NO_x reduction effects by two-stage spray combustion in a high intensity swirl combustor have been investigated using kerosene with a pyridine content of 0-1.7%. The secondary air was radially injected into the combustion chamber through a multiholed tube, which was axially inserted from the rear side of a cylindrical combustor. Experiments have been carried out varying a secondary air injection point, a primary to secondary air ratio, pyridine content and a fuel flow rate. The main experimental results are as follows: (1) by the effective mixing of a fine fuel spray with an air flow of a large swirl degree in the primary combustion zone and by suitable supply of the secondary air, combustion of low NO_x emission can be realized even in a high intensity combustor. (2) The injection of the secondary air at a point where the primary combustion is almost completed is effective for reducing NO_x emission. (3) A primary to secondary air ratio for low NO_x emission has a tendency to increase with a decrease of combustion intensity.

View full abstract

In order to realize the real time optimalization control of the engine combustion parameter, it is essential to compensate rapidly a deviation of the indicated work which is caused during the action of optimalization. From this point of view, the compensator of the indicated work was brought into focus as a concrete application of the LQ control theory. the fuel control of a gasoline engine was selected as the object. Therefore, the concrete structure of the compensator becomes a closed loop engine computer system which has a fuel injector for a disturbance signal, a throttlemotor as an actuator, and a calculating device of the indicated work for a feedback signal. As a result, it became clear that the considerable improvement was attained by the LQI control compensator in comparison with the usual PI control one of the best tuning.

View full abstract

The combustion process and the exhaust emissions of a diesel engine with a dual -fuel (ethanol-diesel) injection system have been investigated. The combustion process of ethanol fuel can be explained by the flame propagation. The soot formed by thee combustion of diesel fuel can be reduced by the combustion of ethanol. The maximum substitution rate of ethanol for diesel fuel reaches 96 vol %. Though the indicated thermal efficiency of the dual-fuel injection engine is about 4% lower than that of the original engine operated with diesel fuel at low load, it is about 1% higher than that of the original engine at high load. Smoke emissions are extremely low resulting in smoke-free operation. CO emissions are the same. Though HC emissions are about 2 times higher than that of the original engine, NO_x emissions decrease to about a 50% lower value.

View full abstract

Film lubrication equations were solved using the boundary Element Method. The problem is associated with lubrication of the piston skirt, which might provide a clue for reducing piston-slap-induced noise and vibrations. The results show that the upper part of the piston skirt is apt to lack an oil film, particularly when the piston is near Firing TDC, at which time the severest piston slap usually occurs. The area in which oil film is formed was calculated throughout the engine cycle. The variation of the area is qualitatively well compared with that experimentally observed with a reciprocating compressor.

View full abstract

Improvements of cold startability and blue and white smoke (cold smoke) during warm-up are important subjects for the development of a small, direct-injection diesel engine. This paper describes a tendency of cold startability and cold smoke when the distance between the center of the combustion chamber and the injection nozzle are changed and when combustion chamber diameter is altered. It is found that good startability and low cold smoke are obtained when the center of combustion chamber is eccentric from the center of cylinder and approaches the injection nozzle, and when the ratio of combustion chamber diameter to cylinder diameter becomes large.

View full abstract