Many small distributed cracks have been observed on the specimen during corrosion fatigue process, and the damage of corrosion fatigue is related to the behaviour of these distributed cracks. The distribution of crack lengths during corrosion fatigue was approximated well by the three parameter Weibull distribution under plane-bending fatigue tests of carbon steel in salt water. A method of estimation of corrosion fatigue lives was proposed. The crack initiation, crack growth behaviour and the variation of the distribution of crack lengths during corrosion fatigue process were taken into consideration in the method. Comparing the estimated results with experimental results, it was concluded that this method could estimate the corrosion fatigue lives with good accuracy.

View full abstract

In order to investigate the threshold condition for propagation of Mode I micro fatigue cracks in an aged cracked specimen made of carbon steel, both the stress concentration factors α_op and α_cl at the micro region of the annealed crack tip, resulting from crack opening and closing respectively, and α_cl* at the aged crack tip are adopted. Models for estimating the stress amplitude acting on the micro region near the crack tip are proposed applying the values mentioned above. The stress condition at the crack propagation threshold can be explained by a model using α_op and α_cl*.

View full abstract

Dynamic circumferential displacements and shear stresses are analyzed for a hollow disc which rotates at variable speeds, and the inner face of which is fixed on a rigid shaft. The problem is solved by using the Laplace transform and Cauchy's integral theorem. For a disc accelerating exponentially with respect to time, ω (t) = ω₀ [1 - exp (-ct) ], the shear stress changes cyclically as a sine function of time. The relation between a parameter c in the above equation, which specifies an increasing rate of rotations, and the amplitude |Δτ^^-_rθ| of the cyclic variations of the shear stress becomes linear, that is, [numerical formula]. The ratio of the maximum dynamic and quasi-static stresses, and the period of the cycliclically changing stresses are obtained.

View full abstract

As a step for clarification of the mechanism of cavitation erosion as well as noises at a jet-flow gate-valve, the cavitation-aspects and the associated sound-pressures were simultaneously observed systematically by means of high-speed-photography and sensitive pressure-detection through a hydrophone, in a 1/10-scale model-valve for various openings. We found the following three different types of cavitation, i.e. stringlike vortex cavitation (Type I), bubbly cavitation (Type II) in the jet shear layer, and induced cavitation (Type III). Each of them has its own desinent cavitation number as well as zone where cavities occur, and emits an impulsive pressure of special wave form. Especially for a narrower opening, the bubbly cavitation develops dominantly there emitting very high impulsive pressures, and the free jet is remarkably deflected downwards in the settling chamber, often touching the bottom, so that the valve will be heavily eroded.

View full abstract

The one-parameter integral method proposed in the last report, which is generally applicable to laminar boundary layers whether they are separated or not, is improved on the velocity profile assumption and computational scheme. The governing equation system being hyperbolic, the spontaneous generation of a singularity associated with unsteady separation is confirmed as a focusing of characteristics. The obtained results of the boundary layer quantities as well as the generation of a separation singularity are in good agreement with those of exact methods (e.g. van Dommelen and Shen) about the starting flow of a circular cylinder.

View full abstract

The grading up of LDV performance inevitably follows the increase of weight of the equipment and it requires fine techniques for setting up optic devices due to its complexity of the structure. For measuring the flow pattern a large traverser will have to be used. However, the probe will be miniaturized by use of fibre optics and will be easily employed in any attitude. This paper deals with the development of a fibre LDV with double Bragg cells to measure backward flows and presents the results measured in the boundary layer on a heating plate in an inclined flow to the gravitational direction.

View full abstract

Turbulence structure in the intermediate wake of a normal flat plate was studied experimentally by a conditional-sampling technique. The measurements were made at a section 8.0 heights downstream of the front face of the plate. Velocity fluctuations were decomposed into those produced by large-scale structures and those associated with small-scale structures : the former fluctuations were again decomposed into the phase-averaged fundamental component and the deviation form it (large-scale deviation). Velocity vector fields, contours of streamlines, vorticity of the fundamental component, together with contours of intensity and correlation of the large-scale deviation velocity, were presented and discussed. Contours of intensity, correlation and turbulence intermittency of the small-scale fluctuations were also presented in and around the large-scale structures. Large Reynolds shear stress associated with the large-scale deviation velocity was suggested to be the result of a stretching of vortex filaments in the saddle-point flow region.

View full abstract

Mean flow properties of air jet issuing from the cruciform nozzle with a sufficiently large aspect ratio into still air have been observed over a wide area of the flow field. From the experiment, it is clear that the streamwise variation of the both scales of velocity and length is slower than that of two-dimensional and axi-symmetric jets. The velocity contours show a symmetric, crisscross shape even at the most downstream section measured in this experiment, whilst in the upstream region the velocity profile exhibits a saddle-back shape for both symmetric slices of y = 0 and z = 0. In the region of x/d≥75 the velocity profiles can be expected to be geometrically similar in the entire flow field.

View full abstract

An experiment was carried out to understand effects of the slot shape at the cylinder-side-wall juncture and the angular location of a blowing slot on the spanwise distributions of various characteristic values near the side-wall of a circular cylinder with tangential blowing. The range of the side-wall effects and the characteristic values near the side-wall are much influenced by the slot shape and the location of the slot. When the slot shape like a knife edge, termed "Edge", is used, the range of the side-wall effects becomes narrower as the angular location of the blowing slot is farther downstream.

View full abstract

The influence coefficient matrix method is one of the rapid solvers of the Poison equation and it is well known as the error vector propagation method. However, the method is limited in practice owing to its fatal defects, which take place in the application to a large domain and to a Neumann boundary condition. The defects are removed by some simple matrix calculations in this paper. Further, the present method is applied to the cavity flow as an example and then it is extended to solve the simultaneous partial differential equations concerning to the stream function and the vorticity with the particular boundary conditions. Consequently, the rapid numerical calculation of the cavity flow becomes available by the present method in the wide range of Reynolds numbers. In particular, it is made clear that the method is effective for the cavity flow of very low Reynolds numbers.

View full abstract

Studies have been made to compare, and discuss, the analytical solutions obtained by the method of characteristics with the experimental results about the waterhammer. The measurements have been carried out using a 1/6 scale model of the primary coolant system of the experimental fast breeder reactor "JOYO" belonging to the Power Reactor and Nuclear Fuel Development Corporation. The waterhammer has taken place at the valve throat area ratio ranging from 1/4 to 4/4 with variations in circulating pump revolving speed in the angular acceleration rages 20 to 180 rad/s² and -20 to -180 rad/s². Good agreement is found between the analytical and experimental results. Therefore the calculation models and idea used in this paper are practically useful enough.

View full abstract

A Darrieus-type cross-flow water turbine has been proposed by the authors for utilizing undeveloped low-head water power resources, but available data for good hydrodynamic design of it are still scanty. Here, effects of geometrical attitude of runner blade o hydrodynamic performance of turbine are investigated experimentally and discussed in comparison with theoretical results obtained by the method proposed in the preceding paper. Experimental results supported the theoretical concept that the geometrical attitude of runner blade can be specified by a single parameter, and the best efficiency was obtained when the chord line of NACA0030 aerofoil was in contact with the pitch circle tangentially at 40% chord point. Comparing the calculated results with the measured data, the geometrical attitude of blade for the best runner efficiency is identical and the shape of time variation in runner torque is similar in both the calculation and the measurement, while quantitative agreement is still insufficient.

View full abstract

The transient behavior of oil flow under negative pressure is studied experimentally. To observe the negative pressure, three types of pipe models with different boundary conditions are used. The pipe used in experiment is 20 m in length, and 8 mm or 12 mm in diameter. Quick operation of a valve mounted at the end of pipe causes fluid transients. The negative pressure is measured mainly using strain gauges set on the outer surface of the pipe. By applying the principle of pressure difference meter, the relation between strain and negative pressure was tested equivalently. The negative pressure below absolute zero was observed experimentally. The numerical computation by means of characteristics method showed a good agreement with the experimental results. The condition of inner surface of the pipe has a great influence on the transient behavior under negative pressure.

View full abstract

A matter of consideration is pointed out for adequate use of the determination method of entrained the droplet flow rate in two-phase annular flow, which is based on the extraction technique of liquid film through a porous sinter tube. The extraction rate of liquid film is complicatedly related with various parameters such as the axial profile of droplet concentration and the effective length of a porous tube over which a couple of base-film and disturbance wave are completely extracted. An analytical model for the functional relationship between these parameters is proposed. Experiments were made by means of a special film extraction device, which is precisely and continuously variable in the extraction length. Detailed data on the effect of the extraction length upon the extraction rate were obtained ; hence, it was confirmed that the analytical model can reasonably correlate the main parameters for determining the extrainment flow rate.

View full abstract

Pressures, velocities of gas and solid phases and volume fraction of solid phase along the conveying pipe in vertical pneumatic transport of powdery material at high solids ratio are numerically analyzed by considering the compressibility of gas density and assuming an isothermal flow. The result is discussed by comparing with our experimental values. Possible values of initial volume fraction and friction factor of solid phase are limited to a finite region by attaching a condition to the analysis in order to get a realistic solution.

View full abstract

For the purpose of utilization of extremely low-head water power, a small model of Darrieus-type cross-flow water turbine was produced and examined for the effects of channel height and number of blades on turbine performance. Then, time variation in generated torque in a revolution of runner was investigated experimentally and compared with the results of theoretical calculation. Performance tests showed that the best efficiency was obtained by double-blade runner under the narrowest channel height condition. Time variations in measured torque showed strong dissimilarity in their wave pattern and quantitative disagreement in their peak values to the theoretical results obtained under the assumption of a uniform steady relative flow. Theoretical analysis suggests that this difference arises from disregard for the effects of circular motion on a flow around the blade and the unsteady character of flow on a Darrieus blade in the above simple calculation.

View full abstract

An expression was obtained for the fluid drag on a capsule having two end-plates. Capsule motion in the start region where it was accelerated was calculated numerically based on the equation of motion in consideration of the characteristics of a blower and the air discharge through a by-pass pipe. Calculation was also made for stopping the capsule in the closed end region of the pipe utilizing the reaction of the air compressed by the capsule. It was found that the calculations agreed well with experimental results.

View full abstract

As a fundamental study of transport of many capsules in the same pipeline, transport of two capsules was investigated both experimentally and theoretically. The experimental results show that the time interval between starts of two capsules has comparatively large effects on the whole system and that these effects are remarkable when the capsule to pipe diameter ratio is high. A theory for a single capsule was extended and applied to the transport of two capsules. It is found that motion of two capsules can be predicted satisfactorily by the present theory.

View full abstract

The phase change process of a melting solid pressed on a hot heat transfer surface was investigated experimentally, as a basic research of the heat storage process in a latent heat TES (thermal energy storage) capsule. Along with measurements for various conditions, the physical parameters deciding the phenomena were analyzed, normalizing the basic equations and boundary conditions. It was shown as the result that the nondimensional heat flux expressed as a function of the Stefan number and the nondimensional contact pressure. Based on the experimental results, the relation among the bove nondimensional quantities was shown by a simple formula.

View full abstract

By realizing dropwise condensation of propylene glycol vapor on PTFE coated surface, dropwise condensation curves were measured over a wide range of surface subcooling and the modes of condensate were observed using video recording and photography simultaneously. The parametric trends were presented by the experiments conducted for the four cases i.e., two values each for departing drop diameters and vapor pressures. Although the curve shapes of propylene glycol vapor were similar to those of steam on oleic acid promoted copper surface, the values of heat transfer coefficient and heat flux were much smaller and the transition mode from drop type to film type was quite different. Further, the effect of non-condesable gas on heat transfer was demonstrated to be negligible in the measurement.

View full abstract

Ethylene and acetylene jet diffusion flames have been studied by two photoelectron correlation techniques. One is laser Doppler anemometry used for determining the local average velocity, turbulence intensity, and velocity probability density. To obtain these, Fourier transform has been performed using the maximum entropy method. The other is the intensity fluctuation spectroscopy of a light scattered form soot clouds. Correlation signals, indicative of spatial distribution of soot-containing eddies, are processed to give the intergral scale of the eddies and the spectral density of the scattered-light. As the result, it has been suggested that in the flames tested, turbulence is intensified by a heat-release due to combustion and this augments the eddy structure of the jet, thus enhancing engulfment of the surrounding air and promoting its penetration into the jet core. A discussion is given of the effect of fuel on the size of the turbulent eddy.

View full abstract

The effect of fuel properties on the temperature distribution and emissions in a swirl type combustor was studied. The fuels used were gasoline, kerosene and heavy oil. Gasoline and heavy oil were used as the most and the least volatile fuels respectively. The momentum flux and the Sauter mean diameter of the issuing sprays of three fuels were controlled to have the same value. The difference of the temperature distributions between the flames was prominent in a non-swirling flow condition. Emissions of soot and NOx took high values in the heavy oil flame and low values in the kerosene flame. Under the heavy oil combustion, exhaust emission of NOx increased by decreasing the swirl strength of the flow. Exhaust soot emission increased by increasing the swirl strength of the flow for all three fuels.

View full abstract

In order to improve the performance of a two-stroke hydrogen injection engine, it is necessary to clarify the scavenging characteristics with the supercharging taken into account. In the present study, the gas in the cylinder before combustion was sampled and analyzed to obtain the contents of H₂ and O₂ molecules. On the other hand, by using the chemical reaction theory, a theoretical equation with H₂ and O₂ contents as variables was derived to obtain the scavenging efficiency and the trapping efficiency. Applying this theory to the experimental results with a practical operating hydrogen fueled engine, the scavenging characteristics have been evaluated. Also, the effects of mechanical supercharging and turbocharging on the two-stroke engines were investigated both theoretically and experimentally. As a result, is was found that supercharging was practically effective for increasing the output power.

View full abstract

In this paper, the dynamic response of a plane curved bar with varying cross-section is analyzed under a dynamic load working towards the center of curvature of the bar. As numerical examples, time responses of displacements and bending moment are found for a symmetric semi-elliptic arc bar with varying cross-section and clamped ends subjected to a step concentrated impact load. The time responses are compared with those for a static load to find out their characteristics. The behaviors of responses due to the variations of the cross-section and the load point are clarified.

View full abstract

A plucked string of musical instrument damps abnormally at specific frequencies. In order to investigate this phenomenon called "dead point", we measured the damping of a plucked string supported at two points on a simply supported beam. Consequently, when the natural frequencies of the system with strong coupling between the string and the beam approach a natural frequency of the beam, the logarithmic decrement of the string becomes large. However, with weak coupling, the string does not damp largely, even if the natural frequencies of the system approach a certain natural frequency of the beam. If they are closer to a natural frequency of higher vibrational modes of the beam, the logarithmic decrement of the string does not become large without stronger coupling.

View full abstract

To avoid resonance of steam turbine blades during operation, it is necessary to establish a prediction method for the natural frequencies of blades. Therefore, a calculation method of the natural frequencies is developed for structures with blades bound on the entire disk circumference. Five natural frequencies exist within the range below 500 Hz in calculations of the natural frequencies of a single blade with turbine. Vibration characteristics for these natural frequencies are shown systematically by taking the direction of the vibration into account when moment of inertia is small at the middle of the blade between blade root and blade tip. The relation between the natural frequencies of all bound blades and those of a bound group consisting of several blades, and variations of the natural frequencies according to variation in blade dimensions are also discussed.

View full abstract

The subharmonic oscillations of order 1/2 of a system with a Duffing type restoring characteristic are investigated by making use of a harmonic balance method, under the condition that both the periodic exciting force (e.g., an unbalancing force) and the constant force (e.g., gravity) are operative. Unlike in the preceding paper, the restoring characteristic of this system is assumed to be a soft spring. As a result, the phenomena of these vibrations are considerably complicated and become extremely different from the same kind of vibrations in the hard spring system. However, the occurrence of the 1/2-th subharmonic vibrations in this system can be similar to the hard spring system, even if the constant stimulating force approaches zero.

View full abstract

A new type impact damper consists of a bed of granular materials moving in a container fixed to the primary vibrating system. The problem is to determine the characteristics of the impact damper for reducing the vibration of the system to a prescribed value. This report deals with the indicial response of the damper, i.e., the damping characteristics of free oscillations from an initial displacement of the vertical and horizontal vibrating system. Experimental models were tested and numerical analyses were made to find the damping characteristics for several values of mass ratio and clearance. The results of numerical analyses agreed well with those of experimental tests.

View full abstract

Buckling and postbuckling behavior of a flexible clamped-hinged pipe which is handing vertically, due to an internal flow, are examined theoretically and experimentally. The equations governing the axial flow of the fluid and the motion of the pipe, which are coupled through the nonlinear terms, are derived. Assuming that the deflection of the flexible pipe is infinitesimal, the critical velocity of the fluid, above which the pipe buckles, is determined by linearizing the equations mentioned above. Under the supercritical condition, a stable equilibrium solution with finite deflection is obtained analytically. Also a solution, which is available for large deflection, is obtained numerically. The latter solution agrees well with the former solution in the case of small but finite deflection. Furthermore an experiment was undertaken by using a pipe model conveying water. The experimental results are shown to be in good agreement with the theoretical solutions.

View full abstract

Important mechanical systems installed in nuclear power plants and other industrial facilities must be designed not to lose their performance during and after destructive earthquake excitations. In this paper, reduction of the first excursion probability of such mechanical systems with inelastic restoring force-deformation relation is examined. First, the first excursion probability of inelastic mechanical system is estimated from experiment using stationary random noises as input. Next, relation between restoring force and deformation is identified from simulation technique using hysteresis loop obtained from experiment. Next, effectiveness of reduction of the first excursion probability for mechanical systems subjected to earthquake excitation is examined by using nonstationary artificial time histories. Finally, obtained results are examined by theoretical technique using equivalent linearization method. It is concluded that the first excursion probability of inelastic mechanical system is smaller than that of elastic system.

View full abstract

This study analyzes the role of human factors in the steering system of two-wheeled vehicles, using equivalent mechanical elements as the first step toward systems analysis of the man/vehicle relationship. This steering system, including human factors, has approximately one torsional degree of freedom. These factors can be obtained by the frequency response of a steering bench model with a rider. First, the repeatability and linearity of human factors are checked. Then, the human factor values are shown to change accordingly as the rider's handle grip and press forces vary. Finally, an equation for two-wheeled vehicle motion is derived, and the result of these calculations makes it clear that human factors play a major role in the behavior of two-wheeled vehicles.

View full abstract

The quasi-static vibratory analysis of a rotor supported by ball bearings is presented. The system is an ideal ball bearing in which the inner ring moves in the radial plane with two degrees of freedom under a constant radial load. The motion of the inner ring centre due to ball revolution, which has bee known as the Perret-Meldau problem since the 1950s, is analyzed in detail. The results show that the inner ring motion has complicated features and changes drastically with the design and operating conditions. All computed results are arranged in a numerical table and in charts. They represent a complete package for the estimation of the magnitudes and characteristics of the inner ring motion, and may be used in the design process to examine the rigidity, critical speed and vibratory nature of a rotor.

View full abstract

In view of troublesome nature of impact sound, a series of studies have been undertaken in order to find out a way to reduce this effect. This second report concerns the mechanism of impact sound generation when a ball collides against a plate. Simply, a ball attached with a force transducer was collided with a freely suspended square steel plate. The pulse-like contact force occurring between the ball and the plate, vibration of the plate, and impact sounds at positions near and far from the plate surface were measured simultaneously. These signals were stored in a wave memory instrument and analyzed by FFT analyzer. Frequency characteristics at four steps concerning the impact sound generation were obtained. These frequency characteristics were not affected by the size of the ball, the ball material or the impact velocity. From the results of analysis, the mechanism of generation of impact sound has been clarified.

View full abstract

A convenient photogrammetry system equipped with a mirror was developed for the measurement of the coordinates of a moving body. The accuracy of the present system is comparable to that of the ordinary multi-camera system. The errors involved in the present system are analyzed in detail and its theoretical limit is discussed. The applicability of the present system is verified by measuring the angular velocity of a spatial four-bar linkage, the angular momentum of a projectile and the motion of the upper limb of a human.

View full abstract

As a prevention method of tooth profile change of a nylon gear, which meshes with a steel gear, this paper proposes a profile modification of steel gear. And a reasonable amount of modification is obtained by experiments. From the results, it is clear that the expression of the amount of profile modification Δα (minutes) of steel gear is as follows. Δα = ζ (Tc·T/9.8b) + ξ ζ : Change value of pressure angle change (minutes)/(N·m/m)·s, Tc : Contact period of teeth of nylon gear (second), T : Torque acting on nylon gear (N·m), b : Face width (m), ξ : Basic value of pressure angle change (minutes).

View full abstract

The occurrence of jump phenomena in gear system to random excitation are investigated theoretically considering the transmission error as a stochastic process. First, the occurrence is examined from the standpoint of mean square value, and second, it is investigated about the probability density function in simplified form. As the result, it is made clear that the phenomena occur in case of a narrow band random excitation and the density function separates into independent parts, but the phenomena don't occur in case of a wide band excitation.

View full abstract

The effects of tip relief on the load carrying characteristics of spur gears were studied by means of FZG-Spur gear testing machine. Two kinds of tooth profiles were used for our studies. One was a nearly standard gear with almost no profile shifting, the other was a gear with a large profile shifting. In the case of the standard gear or the gear with profile of a small profile shifting, the load carrying characteristics were very good, if the tip relief was applied to both the driver gear and the follower gear, but it was unfavorable for the load carrying characteristics, if the tip relief was applied only to the driver gear. Furthermore the influences of tip relief in the case of profile shifted gears upon the load carrying characteristics were studied. The results with profile shifted gears were different from the results with standard gears.

View full abstract

The differential equation which expresses the temperature of rolling/sliding contacting bodies is so improved that the measured surface bulk temperature of gear teeth and of contacting rollers as a function of running time coincide quantitatively well with the simulating results. When the running time vs surface bulk temperature curve has a point of inflection, the temperature diverges and a scoring failure occurs, that is, the thermal stability expressed by this differential equation corresponds to the occurrence of a scoring failure. Because the thermal stability of the differential equation is decided by its coefficients, the scoring index is defined using these coefficients : Theoretical analysis shows, that when this scoring index is less than 0.9, a scoring failure does not occur, and when this scoring index is larger than 1.0, a scoring failure occurs. The scoring test using a power circulating gear test rig has confirmed the propriety of this scoring index.

View full abstract

The first report investigated some practical problems of a new cutting method which makes it possible to cut the bevel gears almost correctly by single traverse end milling. But, this cutting method gives a constant tooth depth instead of a variable tooth thickness which is surely better when we take the bending strength into consideration. This paper proposes a new cutting method which gives a tooth depth varying almost proportionally to the tooth thickness and retains the merits of the single traverse end milling. Using this cutting method, plastic gears are cut trially by a numerically controlled machine tool which was made in the authors' laboratory. The design procedure for the root angle and the pressure angle are described and the conjugation of the gear teeth is analyzed theoretically. This analysis shows that the deviation from the perfect conjugation increases as the tooth depth becomes large. This deviation can be reduced by means of profile modifications of cutter.

View full abstract

The normal tooth profile of a bevel gear changes its form along the tooth trace. When this profile is generated by an end milling cutter of straight axial profile, the number of cutter paths to generate the tooth will surely become large. If the axial profile is suitably designed and the generating point is properly selected in accordance with the cutter profile relative to the tooth profile, the number of the paths will be reduced greatly. This report proposes a new cutting method based on this idea. It is a procedure to avoid the interference between the profiles and to select the generating points, giving also the unremoved volume on the generated tooth surface. The ratio of the unremoved volume in the case of three cutter paths by the curved profile cutter compared to that by the straight profile cutter is estimated to be about 1/20 for the pitch angle 〓 =15°, and about 1/7 for 〓 =75°.

View full abstract

A new metal forming method in which h metal material (99.7%Al) is solidified with vibration and directly formed at high temperature is tried. By this method a material which consists of fine and uniform crystalline grains is obtained. The refinement and uniformity of crystalline grain in solidification process are most significant in this Direct Forming. Mechanical properties of formed material are improved to some extent. Therefore it is clear that high quality material can be obtained by this energy-saving and process-saving method.

View full abstract

It is very useful to know the cutting forces of spiral bevel gears for the decision of the stable cutting conditions and design of the spiral bevel gear generators. Therefore, the cutting forces during generation of a spiral bevel gear are measured by a special attachment. Then, the theoretical area of cross section of undeformed chips is calculated and the specific cutting forces are investigated. The result shows that the specific cutting forces vary with the position o gear generation.

View full abstract