Cavitation damage tests in High Water Base Fluids (HWBF) are performed with a vibratory cavitation test apparatus. The mass loss and surface roughness of the test specimens (aluminum and S 20 C carbon steel) in HWBF are measured, and are compared with those in water. It is found that the mass loss for the aluminum specimen in 5% HWBF is one half of that in water. The damage pits on the specimen surface are shallow and small in diameter. The mass loss M_L of aluminum at t=60 min is smallest in 1% HWBF, which is one third in water. For S 20 C carbon steel, the mass loss M_L at t=120 min in 5% HWBF is less than one fifth of that in water. The number of emulsion particles with smaller diameter decreases by cavitation exposure, and the number with larger diameter increases.

View full abstract

The effects of seeding and mass-separation in molecular-beams of He-Ar mixtures are experimentally studied in order to quantitatively handle composition distortions which inevitably occur during the molecular-beam mass-spectrometer sampling process. The experiment, with the aid of the time of flight mass-spectrometer, is carried out in the range of the source conditions of the argon molal fraction χ_o=0∼100% and the mixture total pressure p_o=25∼75 Torr under constant room temperature. The hydrodynamic velocity of heavier argon molecules remarkably increases approaching that of lighter helium ones, especially for χ_o<lsim>1%, and the translational kinetic energy enhancement reaches isentropically 10, i.e., the molecular weight ratio. The Mach number of Ar, however slightly reduces. The mass-separation effect is also prominent, especially for χ_o<lsim>1%, and the total species enrichment factor attains the molecular weight ratio. The effect is essentially due to the Mach number focusing, which is also confirmed from the spatial spread behavior of helium and argon beams.

View full abstract

This paper describes an experimental study on the flow of non-elastic and viscoelastic liquids in thin films held between a flat plate and a spherical surface which begins rolling movement. Properties of these liquids are Newtonian in viscosity. Empirical formulae have been derived for the maxima of liquid film thickness, cavitation points and thickening rates of liquid films. The viscoelastic liquids give about 1.5 times greater values than the non-elastic liquids with the non-dimensional thickness and thickening rate. The cavitation point is a function of the liquid film thickness and the formula is the same both for non-elastic liquids and for viscoelastic ones.

View full abstract

Excess pressure drops in a flow through small apertures are experimentally investigated for various kinds of liquid and the following three types are found according to the kind of liquid : non-dimensionalized excess pressure drops at Reynolds numbers of magnitudes about 10 have (1) nearly the same values as the theoretical ones of Stokes flow (liquid paraffin and butanol), (2) higher values than the theoretical ones of Stokes flow, but they are correlated with Reynolds numbers regardless of the dimension of orifices used (ethanol and carbon tetrachloride), and (3) higher values than the theoretical ones of Stoke flow, and they do not correlate with Reynolds numbers (water), increasing with increase in orifice diameter at the same Reynolds number. It is shown that the non-dimensionalized excess pressure drops of water have some correlation with a dimensionless number V(λ/ν)^1/2, where V is the mean velocity through an aperture, λ is the relaxation time of water and ν is the kinematic viscosity.

View full abstract

The velocity field of an inlet and outlet flow through small orifices was experimentally examined. The velocity along the center line near the orifices was measured with a laser doppler anemometer, stream lines in the whole flow region were photographed, and the following points were clarified : (1) The center line velocities of liquid paraffin agree with the theoretical value of Stokes flow in the region of Reynolds numbers below 10. (2) With distilled water, a diverging angle of the issuing flow from a larger orifice is smaller than that of the issuing flow from a smaller orifice. Furthermore, a characteristic isolated vortex is generated on the downstream side of the orifice at the Reynolds number of from 10 to 30. (3) With ethanol used as a test liquid, the diverging angle of the issuing flow is small even for the Reynolds number of less than 10, and the flow resembles a jet. (4) The magnitude of the diverging angle of the issuing flow have a decisive effect on the value of the excess pressure drop at the same Reynolds number ; the larger the diverging angles is, the smaller the excess pressure drop is. (5) The dimension of the isolated vortex generated with the outlet flow of water is, when non-dimensionalized with the orifice diameter, correlated with a dimensionless number V(λ/ν)^1/2, where V is the mean orifice velocity, λ is a time proper to water and ν is the kinematic viscosity.

View full abstract

This paper presents the results of an experimental study on the control of the deflection and reattachment of a radial wall jet through comparatively wide spaced two parallel disks, with an additional control flow. The flow field may be considered as having properties including those of the deflected jet, reattached jet and wall jet. The effects of opposite wall height (distance of opposite wall surface from the nozzle center plane) and control flow rate (ratio of control flow rate to main jet flow rate) on the flow pattern near the nozzle outlet section were investigated. The velocity profile, shape and length of a jet potential core, and the deflection of a jet center axis depend mainly on the combination of the opposite wall height and the control flow rate.

View full abstract

An experimental investigation was conducted on relations between a sound and vortices generated by a plane jet impinging on a circular cylinder ; the vortices consisted of those formed in the impinging jet and those shed from the cylinder. A thin splitter plate was attached to the rear side of the cylinder to modify the shed vortices. The nature of the shed vortices is shown to have a significant influence on the intensity and frequency of the sound. The intensity of the sound is closely related to the streamwise extent of distributed vorticity within the vortices ; the more compact the vortices, the stronger is the sound. The r.m.s. pressure fluctuation on the surface of the cylinder has no systematic relation to the intensity of the sound.

View full abstract

An experimental investigation was made of a shock wave induced flow past a circular cylinder in a dusty-gas shock tube. Joint research was conducted between the Muroran Institute of Technology and the Institute of High Speed Mechanics, Tohoku University. Shock tubes used for the present research have test sections of identical geometry. For a frozen shock Mach number of 1.3, flow visualization studies were conducted by a schlieren method and an Ime-Con High Speed Camera in Muroran and a pulsed laser holographic interferometer in Sendai. The behaviour of shock waves past a circular cylinder in a dusty-gas, the development of dust free regions and the formation of vortices behind a circular cylinder were observed in detail by those flow visualization methods.

View full abstract

Pulsatile turbulent inlet flow in a circular pipe was experimentally investigated. The wall pressure, the axial velocity and the turbulence intensity were measured at low pulsation frequencies under which fully-developed flow was quasi-steady. The measured data were compared with those of steady pipe flows. For the range of parameters investigated, the wall pressure of the inlet flow could be predicted from the distribution for a fully-developed flow. The inlet lengths of the velocity and the turbulence intensity were in agreement with a previously reported value (x<thkap>50D) for steady pipe flows. The effect of unsteadiness on the time-averaged value of the velocity and the turbulence intensity was very weak. An approximated analytical solution for the velocity of inlet flows was presented.

View full abstract

Hot wire measurements of three fluctuating velocity components u', v', and w' have been carried out in a pulsatile pipe flow accompanied by relaminarization. Conditional sampling was made in addition to the evaluation of turbulence intensities and Reynolds shear stresses. The onset of relaminarization and the following retransition to turbulent flow were judged from these experimental results. Turbulent motions were classified into four distinct categories. The contributions of sweep and outward interaction to the Reynolds shear stress and the turbulence energy are dominant in the earliest stage of the accelerating period where the ordered motions called bursting gradually cease due to a combined effect of viscosity and acceleration. Meanwhile, the ejection and the wallward interaction play an important role when the production of turbulence occurs again in the last stage of the accelerating period. After the turbulence generated near the wall reaches the pipe center, the ordered motions become the same as those in steady pipe flow.

View full abstract

A shock tunnel facility is fabricated and used for the study of a supersonic flow CO chemical laser. A high temperature mixture of CS₂, CS, S_Z, S and Ar is produced in a shock tube, where the thermal dissociation of CS₂ diluted in Ar is accomplished by a reflected shock wave. The shock heated mixture is exhausted through supersonic nozzles mounted at the end of the tube and mixed with the supersonic streams of O₂. Then, the vibrationally excited CO molecules are produced in the mixed streams by a chain reaction between CS and O₂ initiated by S atoms. The present paper shows that the shock tunnel facility is successfully used to study the dependence of the small signal gain coefficient on the plenum temperature, plenum pressure and the transition branch.

View full abstract

The effects of blade angle of attack, camber and thickness on unsteady aerodynamic forces caused by vibrating blade motion have been studied for two-dimensional subsonic cascades on the basis of an improved double linearization theory. It is found that an unsteady mass source proportional to the steady density difference times the fluid particle velocity exists at the mean position of a vibrating blade. Sample results illustrating the effects of blade angle of attack, camber, thickness, Mach number, reduced frequency, pitch-chord ratio and stagger angle on the unsteady aerodynamic response are presented.

View full abstract

An extension of the improved double linearization theory to two-dimensional supersonic cascades is presented to predict the mean loading effects on unsteady aerodynamic forces on blades in vibrating motion. In the case of supersonic cascades, the aerodynamic instability of the blade motion is significantly affected by the effect of the unsteady displacement of shock reflection points which appear as a fluctuating concentrated force acting at a mean reflection point. Sample results illustrating the effects of blade angle of attack, camber, thickness, Mach number and pitch-chord ratio on the unsteady aerodynamic response are presented.

View full abstract

A three-dimensional theory is developed for a linear supersonic cascade moving between parallel walls, vibrating or interacting with an inlet distortion to predict the effect of sound absorbing wall liners on the unsteady aerodynamic forces. A side wall of the model is partially composed of a sound absorbing liner with uniform acoustic admittance, which is assumed to depend on the frequency sensed by the wall liner. The lined section of the wall is confined to the portion swept by blades. The change of the aerodynamic work of the vibrating rotor blades with changes of the cavity depth of the wall liner becomes conspicuous when the wall admittance for the acoustic mode of the least circumferential order is in a certain effective range. The wall liner effect on the intensity of sound generated by interaction with an inlet distortion is larger for downstream propagating waves than for upstream propagating waves.

View full abstract

A method was developed for calculating unsteady aerodynamic forces acting on vibrating cascade blades in a three-dimensional flow field. Blades were allowed to have nonuniform spanwise steady loading, and to be vibrating in three-dimensional modes. The effects of wake deformation behind blade trailing edges were also considered. It was shown that nonuniform steady loading along the span controlled the unsteady aerodynamic characteristics of vibrating blades at lower reduced frequencies, and the effects of wake deformation became relatively significant at higher reduced frequencies. Compared to the results by a two-dimensional strip theory, there were considerable differences at lower reduced frequencies, especially in the case of highly loaded and staggered cascades. A high-order vibrating mode was also investigated in which an airfoil section might change its chordwise shape significantly. The results indicated that this mode could be critical in determining the flutter boundary.

View full abstract

It is important to study the nonlinear interaction of forces on foils and of wake patterns in the flow around cascade of fluid machinery, helicopter rotors, and flaps. In this paper, we tried to solve the unsteady incompressible flow around two two-dimensional oscillating foils by using both the discrete vortex approximation and the surface-singularity method. A numerical analysis has been developed for the wake configuration, the lift and drag on the aerofoils. The effect of the distance between two foils, the stagger angle, and the phase difference of the two oscillating foils is discussed by the numerical results.

View full abstract

The coherent structure of a turbulent boundary layer in the vicinity of the trailing edge region of an airfoil is investigated experimentally. The airfoil (NACA 65-010) is located in a wind tunnel at an angle of attack of 0° and 8°. Conditional sampling with the VITA technique, and quadrant analysis are applied to make clear the coherent structure at the near wall region, which is called the bursting phenomenon. The main results obtained in this paper are summarized as follows : The contribution from the sweep event of the bursting phenomenon to the mean Reynolds stress -(uv)^^^→ increases at the suction side of an airfoil with an attack angle. However, at the pressure side, the contribution from the sweep event decreases, and that from the wallward interaction event increases. This is the reason that mean Reynolds stress increases at the suction side and decreases at the pressure side.

View full abstract

An experimental investigation of a turbulent boundary layer in the vicinity of the trailing edge of an airfoil is reported in this paper. The airfoil (NACA65-010) is located in a wind tunnel at an angle of attack of 0° and 8°. The mean-velocity, turbulent energy, Reynolds stress, auto-correlation function and energy spectrum are measured. Further, the structure of the outer intermittent region is investigated using conditional sampling. The main results are summarized as follows: The turbulent energy at the pressure side increase compared with the suction side, but a reversed characteristic is observed for Reynolds stress. The center of intermittency moves wallward and the width of the intermittent zone increases at the pressure side. The turbulence structure of the turbulent boundary layer at the pressure side is different from that of a fully developed turbulent boundary layer.

View full abstract

A numerical technique for the analysis of three-dimensional compressible turbulent flows in a turbine stage is presented .To calculate the steady interaction flow fields in a nozzle and bucket simultaneously, the nozzle outlet elements and the bucket inlet elements are overlapped in the axial direction and are used for connecting boundary elements. To calculated the flows in arbitrarily shaped geometries, a control volume method combined with a body-fitted curvilinear coordinate system is used to obtain spatially discretized governing equations. In the present analysis, a two-equation model of turbulence is introduced to estimate the turbulence effect. In order to assure the effectiveness of the present method, a computation is carried out for the flow in a model turbine stage. Experimental data is also obtained by the use of a 5-hole pitot tube for the purpose of comparison with computational results. It is shown that three-dimensional flow phenomena due to the viscous effect are well predicted and the comparisons with experimental data give encouraging results for three-dimensional flow prediction in a turbine stage.

View full abstract

Revised k-ε models are proposed. Here, k is the turbulent energy, and ε the dissipation rate of kinetic turbulent energy. The proposed models are both widely applicable and are transferable to other two-equation models. These are so modeled that the diffusion term of the ε equation does not exert significant influence in the vicinity of the wall compared with production and dissipation terms. The model constants are estimated by using approximated solutions. Numerical examples, which give good agreement with the experimental results, including eddy viscosity distribution, are shown in the paper. It is also seen that the distribution of k has a small peak near the wall. This validiates that C_ν is nearly constant.

View full abstract

By making use of Laser-Doppler-Velocimeter, Reynolds shear stress and three components of Reynolds normal stresses have been measured in an axi-symmetrical confined jet. Additionally, numerical computation have been made with k-ε two-equation model of turbulence, full Reynolds stress model and algebraic Reynolds stress model. The turbulence near the wall has been confirmed to be more intensive in the recirculating flow region than around the flow reattachment point. The result gives a partial proof for the previously reported conjecture on the physical background for the spatial separation between the maximum Nusselt number position and the flow reattachment point. The measured turbulence quantities have also been discussed from other view points. The Reynolds stress model shows rather poor applicability to the presently studied type of partially recirculating turbulent flow. The k-ε two-equation model gives reasonable results.

View full abstract

The efficient method to solve two-dimensional, incompressible, steady, partially parabolized Navier-Stokes equations has been presented. The method has been extended from Keller's BOX method for boundary-layer equations. The Israeli's source term, a streamwise direction staggered grid and a multi-grid procedure have been applied in order to obtain numerical stability and fast convergence. Quantitative comparisons with Briley's full Navier-Stokes solutions of a separation bubble problem and Nishioka's experimental data of a flat plate trailing edge problem have been made. In the former case, the agreement is excellent, and the pressure distribution at the wall also agrees well with Carter's inverse boundary-layer solution. In the latter case, the agreement is reasonable. It is also found that the pressure gradients in both the streamwise and normal directions are almost of the same order. This suggests that the omission of the normal momentum equation, i.e., the boundary-layer equations, is supposed to be a rough approximation for the trailing edge flow.

View full abstract

The rate of flow in a pipeline can be measured by a bend flowmeter which utilizes the differential pressure heat between the inside and outside walls of the curved passage. This investigation aims to determine the most suitable measuring section at which the pressure difference has the largest value, and at which the discharge coefficient of the bend flowmeter for that section has a nearly constant value. Flowmeters of a 180°-bend having a radius ratio of bend R/a=2.50, 2.89, 4.00 and 5.00 were employed for this investigation. It is found that the most suitable measuring section is found at an angle of θ_i=22.5°∼45° and also the most suitable radius ratio R/a is in the range of 2.50∼2.89.

View full abstract

The fluid-dynamic forces acting on the float at various heights of a rotameter were measured by using a beam balance. The forces are found to depend on the flow speed through the gap beween the float and the tube and are composed of the form drag of the float and the frictional drag of the tube wall. The drag coefficient of the former is constant all through the range of the rotameter, while that of the latter decreases inversely proportional to the flow Reynolds number. The results can be used to scale the rotameter only from its geomtrical configurations such as diameter and taper-ratio of the bore if the float is specified.

View full abstract

A velocimeter was developed which measured the frequency of the Karman vortex downstream of a wire of small diameter which was set perpendicular to the water flow in a channel. The frequency of the Karman vortex was measured by the focused light of a laser diode which scanned downstream along the wire. The wire was heated by applying alow-voltage electrical charge which strengthened the refraction of the light resulting in a density change of the fluid. The velocity distribution could then be measured. THE following results were obtained : 1) the frequency and the velocity was linear ; 2) influence to the fluid was negligible though it was not noninvasive ; and 3) drift due to the terperature change was negligibly small.

View full abstract

For the purpose of improving performance of a mixed flow pump, it is necessary to investigate not only an impeller but also a diffuser. The latter, however, has been investigated scarcely. Thus, in this report, the performance and internal flow of the mixed flow diffuser with vanes were clarified. The following conclusions were mainly derived. (1) The outlet flow angle ins nearly constant under the range of the incidence angle-20°<i^^-_n<0°. (2) The pressure recovery coefficient becomes the maximum at i^^-_n=2°∼3°. (3) The flow loss, however, takes the minimum value at the negative incidence angle i^^-_n=-20°∼-10°, as the flow separation on the pressure surface is suppressed well by the secondary flow which runs toward its surface. (4) It is possible to predict the main flow using the presented quasi-three-dimensional analysis, when a large separation does not occur on the vane surface. (5) The separated region on the suction vane surface is wide near the hub wall at the negative incidence angle, but the separated region near the casing wall becomes wide with the increase of the incidence angle.

View full abstract

The change in spectrums of velocity and casing wall pressure fluctuations due to cavitation was investigated, along with the change in the flow pattern through an axial-flow impeller, analysis being carried out for a range under 10 kHz. The spectrums of pressure fluctuation measured on the impeller casing depend mainly on the tip clearance cavitation development. Levels of high frequency components over 1 kHz increase remarkably as cavitation grows, and there are no discrete components in the high frequency range. The energy of pressure fluctuation is much greater than that of velocity fluctuation under the cavitating condition. Thus, cavitation noise is concluded to be brought about by pressure fluctuations.

View full abstract

A Wells turbine operating under irregular wave conditions has been analyzed theoretically. It is shown that the turbine performance depends on a unique parameter, including characteristic parameters of the irregular wave, turbine speed and dimensions of the turbine and air chamber. In order to obtain the optimum value of this parameter, a model test has been performed using a computer-controlled wind tunnel, which can simulate arbitrary oscillating flows based on spectra of irregular waves. From this value a set of optimum dimensions of a Wells turbine system can be determined. Furthermore, starting and running characteristics of the Wells turbine have been obtained by a computer simulation. It is possible to predict the optimum value of the parameter and behavior of the output coefficient by computer simulation.

View full abstract

The Wells turbine for a wave power generator has hysteretic characteristics in a reciprocating flow, which results in the inaccurate prediction of performance in a quasi-steady analysis. In order to clarify the hysteretic characteristics, an experimental investigation has been made by use of newly devised turbine equipment in which a sinusoidal flow condition is simulated. The results have shown that hysteresis becomes more remarkable as solidity and blade thickness become larger, but it is insensitive to the Reynolds number and blade surface roughness. By means of flow visualization, it is found that the hysteresis is due to the different behavior of wakes between an accelerating flow and a decelerating flow.

View full abstract

An oil system is the equipment used to control the flow of hydraulic oil to the bearings and seals in process compressors and steam turbines. The conventional oil system, in many cases, has been controlled by pneumatically operated process control valves. It is generally difficult, however, to control the relatively rapid response of an oil hydraulic system by a pneumatically operated control system. A new-type of control valve for header pressure control was developed. The dynamic characteristics of the oil system using the new-type of valve was examined theoretically and experimentally. The effect of parameter variations on the pressure response and stability of the control valve were clarified. As a result, the stable region of this system increases with the increase of valve flow force.

View full abstract

A model of a ship's propulsion mechanism is proposed, in which two-series flat wings are arranged in a square channel. The motion of each wing is the same as that of the single-stage model previously proposed by the same authors. The analysis is performed by assuming that the flow is a potential flow and approximating the wings by bounded vortices. The unsteady forces and moment are estimated by using the multi-valuedness of the complex potential. The effects of the configurations of the two wings, such as the phase difference of their motions and the distance between them, on the thrust, the drag, and the moment coefficients are examined for various flow velocities. However, it has become clear that a more detailed study will be needed on the Weis-Fogh effect between the blades of the rotor and stator suggested by Furber and Ffowcs Williams.

View full abstract

Unsteady flow fields around a Savonius rotor and its power mechanism are studied by a flow visualization experiment for various rotor angles and tip-speed ratios. For small tip-speed ratios, the rotor receives a corotating torque mainly due to the difference of the drag on both the blades. As the tip-speed ratio increases, a Coanda-like flow pattern is formed on the convex side of an advancing blade, and the corotaing torque is expected by the induced lift, though the torque due to the difference of the drag decreases. The power performance of a Savonius rotor can be explained reasonably by the flow visualization results.

View full abstract

In this paper, two kinds of horizontal axis wind turbines, namely, upwind and downwind, were investigated and compared. In this experiment, the strain distribution and the rate of strain fluctuation on the surface of the rotating blade, and the strain and bending moment on the blade root, etc. were measured. As a result, the difference between upwind and downwind types was solved. The result is very important for the design of a horizontal axis wind turbine.

View full abstract

The purpose of this report is to investigate the fliud flow and heat transfer in an opposed plane jet in a uniform stream. An air jet heated up to a temperature of thirty degrees above a uniform air stream issues from a nozzle 3.0 mm in width. time histories, power-spectra and cross-correlation functions of velocity and temperature are measured by a hot-wire anemometer, a resistance thermometer and a FFT-spectrum analyzer. Moreover, unsteady motions are visualized by a Mach-Zender interferometer and a high-speed camera. These indicate that there are measurable flappings of the jet, having two modes. The frequencies of the flappings are varied with the ratio of jet velocity and uniform stream velocity.

View full abstract

The accuracy of the Monte Carlo method is checked by comparing the results with the methods of two-flux, spherical harmonics and Fn in one-dimensional absorbing-anisotropically scattering media. The results from the Monte Carlo method are shown to fit very well with the results of the Fn (exact) method. From a two-dimensional radiative-convective heat transfer analysis in a gas enclosure with anisotropic scattering, the existence of backward scattering raises the flame temperature and reduces the heat flux at the walls away from the flame.

View full abstract

Free convection heat transfer form the inner surface of a sphere was studied by using integrated boundary layer equations. Analysis was made for the condition that the sphere surface was cooled ; and the top of the sphere was a stagnation point. Approximated solutions agreed well with numerical solutions for a Prandtl number range of P_γ≥1. The analytical results were compared with the experimental results of free convection made in a hemishere. The Nusselt number of both results agreed within about 15%.

View full abstract

An experiment was performed on hot surface rewetting caused by the collapse of film boiling of R-113 flow in a 10 mm. I. D. circular passage through a copper block. Temperatures in the copper block during the cooling process were measured by thermocouples embedded in it. By solving the heat conduction equation in the block, boiling curves were obtained. In the flow boiling system, while heat transfer characteristics and rewetting mechanism changed according as the flow pattern changed with the flow quality, the wall superheat at the onset of surface rewetting as well as that at quenching was almost constant.

View full abstract

A cold energy storage method having excellent heat transfer characteristics was proposed, utilizing natural circulation of a refrigerant. Three layers, i. e., a refrigerant vapor(R 12 or R 114) layer, a water layer and a refrigerant liquid layer stratified in this order from the top were kept in an enclosure composed of a condenser, an evaporator and a condensate receiver-and-return tube. Direct contact heat transfers of a refrigerant with phase change were applied both for the cold energy storage and release processes. In the storage process, the refrigerant liquid vaporizes at the refrigerant liquid-water interface with the formation of ice, the vapor condenses at the condensing surface and the condensate returns to the refrigerant liquid layer simultaneously. In the release process, the refrigerant liquid vaporizes at the vaporizing surface, the vapor condenses on the ice with melting, and the condensate and melted water returns to the original layers simultaneously. It was confirmed by the measurements that rapid ice formation and melting were realized.

View full abstract

Heat transfer and visualization experiments were performed for a single cooled tube immersed horizontally in a fluidized bed under frosting conditions. Measurements were made of local and average heat transfer coefficients around the cooled tube surface. Glass beads having nominal diameters of 0.43 mm, 0.89 mm and 1.6 mm were employed as bed particles. The 30 mm diameter tube was kept 100 mm above the distributor. All of the results obtained under frosting conditions were for an air temperature of about 5°C and an air relative humidity of about 80%. Heat transfer coefficient with frosting evaluated in the present investigation includes the heat transfer coefficient from the frost surface to the bed and the thermal resistance of the frost layer. Comparisons were made of heat transfer data without frosting. The obtained data were found to be larger than those without frosting under the state of fluidization.

View full abstract

Two different types of the multiple effect solar still have been designed and constructed. The difference between them exists in the different methods of collecting solar insolation. The first still (7 stages with a 1.05 m² evaporating area) is associated with the fixed and concentrated type reverse flat plate collector, whereas the second still (5 stages with a 0.41 m² evaporating area) is equipped with a heat-pipe collecting system. The results we have observed throughout the outdoor experiments show that the possible maximum total amount of daily productivity would reach about 14 kg/(m²d) for the multiple effect solar still, and we can expect an improved distillation performance over the ordinary single effect solar still by three times, while another type of multiple effect still may improve by about two times.

View full abstract

To utilize solar energy for heat engineering purposes, the data of insolations and sunshine hours in Kitami and Sapporo during the eight years from 1978 to 1985 were processed and analyzed as daily values. The data of insolations of Kitami are the values measured by the authors at the Kitami Institute of Technology, and those in Sapporo are the values observed at the Sapporo Meteorological Observatory. The number of sunshine hours for both cities were the official data published by the Meteorological Agency. Consequently, the correlation equations to estimate the horizontal-total and normal-direct insolations from the sunshine hours, the rate of clear days to total days in a month and the frequency distribution of the continuous cloudy days were obtained.

View full abstract

A technique is developed to measure burning velocity using soap bubbles and a hot-wire anemometer. This method is a modified version of that mode by Andrews and Bradley. The principle of this method is the simultaneous measurement of flame speed and unburned gas velocity ahead of the flame in a soap bubble. The flame propagation is recorded using a schlieren system with chopped light source. The method has been used to measure the burning velocity of methane-air mixtures over a wide range of equivalence ratios. This method gives a maximum burning velocity of 45 cm/s, at an equivalence ratio of 1.08. Values of burning velocity determined in this way are in good agreement with those of Andrews and Bradley and those of other accurate methods, but in moderately rich mixtures, the present data are slightly higher than those obtained by other workers using burner methods. The influence of flame curvature upon the burning velocity is discussed.

View full abstract

The possibility of realizing a wide range, low noise valveless pulse combustor is examined by charging a rich propane-air mixture having an equivalence ratio of 3.0 with a relatively high supply pressure of 0.5-35 kPa under the condition of forced ventilation. The utilization of the backer plate as a baffle plate is found to act effectively, not only on the strong mixing of fresh inflows and residual hot combustion products, but also on the prevention of back flow during an explosion period, provided that the clearance between the nozzle exit and the baffle plate is adjusted to the optimum. The results show that the present valveless pulse combustor operates more silently than the flapper-valved pulse combustor by 5-8 dB(A), having a large turn down ratio of 9.5-11.5. Also, the volumetric efficiency of the pulse combustor varies from 3 to 8 %, the range being nearly the same as that of the flapper-valved pulse combustor.

View full abstract

In order to measure the transient temperature of gas, a laser interferometry was applied. When the density of the gas changes, the effective optical path length of the test beam changes with a corresponding change of the refractive index. So the change of temperature of gas can be determined by measuring the pressure and the shift of interference fringes with a pair of photo-transistors. In this study, examples of the temperature change measurement of the gas due to only the expansion, to the compression and expansion with a piston motion, and to the compression with flame propagation were described. It is recognized that interferometry is a useful method for the measurement of the transient temperature of the gas in the field of combustion research.

View full abstract

A method of infrared-rays computed tomography (CT) for measuring the rigorous temperature distribution of gases such as a flame was developed. This CT algorithm takes account of the effect of absorption by the gas itself which has not been counted in the past works. The importance of the effect was investigated by the simulation of the measured temperatures about model distributions. This technique was used to measure the temperature distributions of four types of stationary flames of premixed gases, and the results agreed fairly well with the ones measured by a thermocouple. At the same time, the possibility of a density distribution measurement of gas components such as H₂O and CO₂ is shown. It is proved that this CT method provides a new tool for combustion research.

View full abstract

The behavior of a single-shot gas jet injected into a high pressure gas chamber was investigated with schlieren photography. The configuration of the gas jet was similar to a diesel spray and a water jet injected into water, and the penetrating distance X_S was expressed by the following equation. X_S²=a√(α·P_f/P_g)d_n·U_o·t/tan θ_S, where d_n=nozzle diameter, U_o=injected gas velocity at nozzle, α=discharge coefficient of nozzle, t=time from infection beginning, θ_S=average half cone angle of jet, P_f=density of injected gas, P_g=density of surrounding gas and a=empirical constant. The value of "a" was 2 for the gas jet and the auto-motive diesel spray, and 1 for the high pressure spray (85-180 MPa) of marine diesel.

View full abstract

It is widely known that NO_x emissions from an engine play an important role in the formation of varnish and sludge in engine oil. Therefore, the authors measured the concentration of nitric and nitrous acid ions in the lubricating oil and analyzed the process through which NO_x mixes into the lubricating oil. It was found that NO_x mixes into the lubricating oil through oil film on the cylinder wall and through crankcase gas ; and that the amount of NO_x mixing into the lubricating oil was proportional to the NO_x concentration in the exhaust and crankcase gas. It was also found that a rapid increase in the concentration of nitrous acid ions when the engine temperature drops results in the condensation of water vapour contained in the crankcase gas.

View full abstract

In this paper, we investigated the effect of water mixed in methanol on the performance of 4-cycle spark ignition engine. The water mixing ratio in methanol was varied from 0 to 30% by weight. Experiments at various excess air ratios were caried out under a constant charging efficiency and engine speed. The mean burning gas temperature in a cylinder was measured with a platinum wire resistor thermometer fixed near the cylinder head. When the excess air ratio was decreased or the amount of water mixed in methanol was increased, the mean temperature in the cylinder decreased and the ratio of burned fuel to supplied fuel decreased. In order to facilitate fuel vaporization, the air fuel mixture was heated, which resulted in an increase in thermal efficiency.

View full abstract

The previous reports have been determined that the problems of cold startability and blue and white smoke (cold smoke) in a small direct injection diesel engine are improved by the shallow type combustion chamber and weaker swirl ratio. In this report, the lip type combustion chamber, which provides good performance and good economy, was tested in order to know the influence on the cold startability and the cold smoke of various lip diameters, lip thicknesses and shapes. The following results were obtained. There is an optimum lip diameter for cold startability. A thick lip is inferior to a thinner lip in so far as cold startability and the cold smoke is concerned. The lip shape effects cold startability, and a round shape, which diminishes the contraction flow, shows a good result. Cold startability and the cold smoke deteriorate with the swirl strength in the lip type as well as the bowl and square cavity combustion chambers.

View full abstract