Using very thin steel plate specimens, the effect of the plate thickness on the fatigue crack propagation was studied and the following results were obtained. In very fine grained specimens, the crack propagation rate becomes slow with an increase of plate thickness. Based on the fatigue crack propagation resistance in the plane stress (in case of extremely thin plate specimen) and plane strain (in case of relatively thick specimen) conditions and the crack tip plastic zone sizes under maximum stress and cyclic stress conditions, the effect of plate thickness on fatigue crack propagation was estimated. The experimental results coincide with this estimation.

View full abstract

Transient response of an inhomogeneous elastic solid to an impulsive SH-source is considered. Its inhomogeneity is such that μ=μ₀ (r/a)^λ ρ=ρ₀(r/a)^λ-2 Due to this inhomogeneity, SH-wave velocity vanishes at a coordinate origin and the wave rotates around the origin as if it were a diffracted wave. Deducing a summation formula of a Fourier series of Schlomilch type, the response is given by a finite sum of waves rotated around the origin. It is found that an increase of inhomogeneity parameter |λ| causes rapid fluctuation in the response.

View full abstract

The viscoelastic behaviors of two kinds of high damping alloys, Mn-Cu alloy and Fe-Cr-Al alloy, are studied with a view to assessing the damping characteristics. Under the assumption that the material is a linear viscoelastic body, the complex compliance is obtained from the wave propagation data on a long rod analyzed with Fourier transform technique. Copper is also examined for comparison. Experimental results reveal the frequency and amplitude dependences of damping capacity in each material. It is also found that the 4-element model should be applied for both high damping alloys, and the Maxwell model for copper, respctively.

View full abstract

To calculate efficiently elasto-impact contact stresses and their distributions by the finite element method (FEM), the static, dynamic and mixed condensation methods are investigated. These methods are introduced in the formulation of the FEM and applied to a two-dimensional analysis of longitudinal impact of two prismatic rods with a uniform section and equal cross section. The following results are mainly obtained in this paper : (1) The contact-impact time and contact stresses by the FEM for two elasto-impact bodies with the static condensation method agree well with the normal results by the FEM without the condensation methods. (2) The contact-impact time by the FEM with the dynamic condensation method becomes shorter than that by the FEM without the condensation methods, because of the decrease in the element values of the mass matrix and the increase in those of the stiffness matrix. From the results described above, the static condensation method is found the best one of them. Moreover, by selecting any condensed points among the nodes apart from the contact surface between two rods, the accurate elasto- impact contact stresses can be calculated efficiently by the FEM with the static condensation method.

View full abstract

A numerical method for solving the unsteady two-dimensional incompressible flows through a cascade of the diagonal turbomachine is developed by using a discrete vortex method, based on the previous paper. In the discrete vortex method, "point vortices" are not singularities in the existing method but ones introduced only as calculating points on the Lagrangian coordinates. By the method a reasonably smooth distribution of vorticity on the Eulerian mesh can be obtained using relatively a few point vortices. The present method is applicable to an unsteady cascade flow in the water turbine runner subjected to the wakes from the guide vanes in which the distribution of vorticities is complicated.

View full abstract

A rigorous solution of complex velocity of a compressible potential flow past two-dimensional cascades of airfoils is obtained by introducing a pseudoanalytical function with respect to complex coordinates. A basic differential equation about complex velocity is derived from a continuous relation and an irrotational flow condition and then reduced to an inhomogeneous complex singular integral equation. Using successive substitution which depends on theorems of quasi-conformal mappings a solution of the equation can be numerically given with the help of results of an incompressible potential flow, since the complex velocities of compressible and incompressible flows are in one-to-one correspondence with each other. Using this function-theoretic means the solution is applicable both to axial and to radial flows and also to inverse problems. Numerical illustrations of the method are given for high subsonic potential flows through A₃K₇ turbine blades in cascade.

View full abstract

Experimental investigations on the characteristics of a flow around two circular cylinders of different diameters with the ratio d₂/d₁ = 0.68 arranged in tandem were carried out. Reynolds number defined by the diameter of the first cylinder was varied in the range of 1.3×10⁴ ≤ Re ≤ 5.8×10⁴, and the longitudinal spacing between the axes of the cylinders in the interval of 0.9 ≤ L/d₁ ≤ 4.0. The reattachment of a separated shear layer from the first cylinder, the jump phenomenon and the bistable flow at the critical region were confirmed in the same manner as the case of two cylinders of equal diameters. The differences between the two cases were discussed. Flow patterns were divided according to the spacing and Reynolds number. Characteristics of those flow patterns and effects of the Reynolds number were clarified.

View full abstract

This paper presents an experimental study of the perpendicular impingement of two turbulent plane free jets. The effects of velocity ratio and impingement location on the characteristics of a jet after impingement were studied. As a result it was found that the deflection angle increases with an increasing velocity ratio, and the mean velocity and turbulent intensity profiles in a similar profile region are almost independent of the velocity ratio and the impingement location.

View full abstract

Experiments on transition to turbulence in an oscillating pipe flow were made by using a hot wire anemometer over ranges of Reynolds numbers R_{e os} = |μ_{m, os, l}| D/ν(|μ_{m, os, l}|is the cross-sectional mean velocity amplitude, D = 2R the pipe diameter, ν the kinematic viscosity) from 600 to 65000 and of dimensionless frequencies √(ω') = R√(ω/ν) (ω is the angular frequency) from 2.6 to 41. Critical Reynolds numbers from laminar to transitional and from transitional to turbulent flows agreed well with published experimental results. In the turbulent regime turbulent bursts follow by relaminarization in the same cycle. It is observed that the instantaneous velocity profile in every phase when a turbulence with higher frequency appears is represented by the well-known Blasius 1/7 power law, but in the laminar-like phase it does not follow the laminar theoretical solutions for a steady oscillating pipe flow.

View full abstract

For the purpose of clarifying the transient characteristics of a diffuser pump during start up, a series of tests was carried out in which three factors affecting the characteristics were varied. These factors were mass of water in the pipeline, valve opening and starting time. The results of the tests indicate that the locus of operating points on the head-capacity plane during pump start up deviates from the system resistance curve to a great degree as the mass of water in the pipeline becomes large and the starting time becomes short, and the trail of the operating points approaches the shut-off point. The results also reveal that the operating points follow generally the steady characteristics curve within the region where φ≥0.05 on the dimensionless head capacity plane. Analysis of the total pump system was made on the basis of one dimensional equations of motion using an analogue computer. The results of the analysis were almost the same as those obtained from the experiments. Axial thrust during start up was measured and is discussed also in this study.

View full abstract

A combined use of a vibratory element and a servovalve is available for the application of the hydraulic oscillator utilizing the self-excited vibration phenomenon. The vibratory characteristics of the hydraulic oscillator with the servovalve mechanism depend on the dynamic characteristies of the actuator connected with it. In this study, we derived the nondimensionalized simultaneous equations of motions regarding an oscillator with servovalve mechanism and an actuator connected with it, and by using analogue computer, we obtained the solutions to the equations of which coefficients were systematically varied. By this analogue simulation analysis, the effects of the nondimensionalized factors representing actuator characteristics on the hydraulic oscillator characteristics were clarified, and a guide to designing of a hydraulic oscillator with the servovalve mechanism was presented.

View full abstract

Shock phenomena caused by a rapid valve closure in a slug flow region were investigated. The experiment was conducted in a horizontal acrylic tube of 20.7 mm ID, 4.85 mm in thickness, and 18.5 m in length. The profiles of the transient pressure caused by a rapid valve closure in slug flow are affected by the flow configuration adjacent to the valve, and these are classified into two types according to the existence of a gas slug or a liquid slug at the valve at the instant of valve closure. The characteristics of the transient pressure in the former were analyzed by an oscillation system model composed of a mass (liquid slug) and a compressible capacity (gas slug). Those in the latter were also analyzed for a homogeneous two-phase flow model by a similar method to that in a waterhammer analysis. The experimental results were well explained by these analyses.

View full abstract

As an application of the usual hot wire method, a method of determining the thermal conductivity of the specimen is discussed theoretically, in which the hot wire is sandwiched between a plate-type specimen and a reference material of the known thermal conductivity. It is found that when the thermal conductivity of the reference material is known the thermal conductivity of the specimen can be determined theoretically just as in the case of applying the usual hot wire method to a homogeneous material. The error and the reproducibility of the experimental values by this method are within ±4 % and ±1.5 % of the thermal conductivities of standard materials respectively and this method is found very good.

View full abstract

A power generating system utilizing natural circulation is proposed in order to use a large quantity of low-temperature heat effectively. First, the dimensionless performance characteristics of the output were obtained by a simple approximated theoretical analysis on the assumption of no slip between liquid and vapor, and constant friction losses over the tube length. Next, a more detailed theoretical analysis was made, taking the above assumption into account, and was applied to a system of actual size, using R-113 as a working medium. The effects of each parameter on the performance were shown. Furthermore, the influence of local heating on the output was discussed.

View full abstract

One method of improving the cold starting characteristics in a direct injection diesel engine, is that named TSCE (Two Stage Combustion EGR). It prevents the induction of a fresh air to half of the cylinders of the engine, and uses the exhaust gas from the already operating cylinders. This quickens the rise in intake gas temperature of the second stage cylinders, shortens the starting time, and guarantees a stable combustion at the earliest possible time. As the result, Hydrocarbon mass emission and Aldehyde concentration, i.e., the main factors of exhaust pollution, were reduced to one third the amount produced by the usual method of operation, when air, oil and water temperature were -10 to -12 °C.

View full abstract

Uneven distribution of combustion gas in the cylinder occurs in some open chamber type diesel engines. It is considered that this phenomenon is generated by an interaction of fuel spray with the combustion chamber wall. This study aims at revealing the influerace of uneven distribution of combustion gas on the combustion performance of the engine. In this report, the correlation of concentration distribution of combustion gas along the cylinder axis measured by the exhaust gas sampling method with the operating characteristics of fuel injection system, is examined under various operating conditions by changing the load, injection timing, and opening pressure of an injector. As the result it is found that mean fuel velocity through nozzle holes in the injection period, amount of fuel supplied in a cycle, and mean piston position in the injection period are the controlling factors for the formation of concentration distribution of gas.

View full abstract

The heat flux through the combustion chamber wall varies periodically with thermodynamical cycle of engine and locally according to various places of the wall. In this research project the heat flux was measured at the walls of piston crown, cylinder head, suction valve, exhaust valve, pre-combustion chamber and exhaust port, and the result was compared with the calculated value of heat loss based on cooling water. The heat flux at any point and at any period during the cycle depends on gas temperature, the surface temperature of wall and gas flow. Especially, it is affected by a jet flow from the pre-chamber and the arrival of a burning flame at these points. The percentage of heat loss from piston crown is about 7%, from cylinder head 4%, from prechamber wall 4%, from suction valve 1%, from exhaust valve 0.5% and from exhaust port 5% of the heat energy supplied by fuel at full load.

View full abstract

The effectiveness of a modal-least squares balancing method is investigated with regard to a balancing of flexible rotors. The method combines features of both modal balancing and least squares balancing methods. Numerical simulations are carried out with influence coefficients which are assumed to have phase angle errors. The good balancing conditions that were obtained in the second balancing trial are also described.

View full abstract

In designing a bolted joint, it is important to know the force ratio. This force ratio is a factor in the case where a load is applied only in the axial direction. it is not considered when an external bending moment is applied on the assembly. In this paper, it is pointed out that influences of the external bending moment on the increment of the bolt axial force should be taken into consideration. For this purpose, the force ratio for the external bending moment, which is the ratio of the increment of the bolt axial force to the external bending moment, is introduced. Then, in the case where clamped parts are T-flanges the method to find the force ratio for the external bending moment is shown. In order to verify these analytical results, experiments are carried out. Comparing these results, they are in a fairly good agreement.

View full abstract

In order to elucidate the effect of surface hardness on tooth strength and surface durability and on failure modes of gears, fatigue tests were conducted. The results obtained are summarized as follows : There are hardness ranges in which pitting, tooth breakage and spalling will break out. The degree of tooth profile deterioration can be used as a measure to evaluate the life for pitting. At the fatigue limit load, the dynamic load increase which is induced by tooth profile deterioration is not so large as to be taken into consideration. The failure mode and fatigue life can be affected by the performance of lubricant.

View full abstract

In this report, the meshing states of belt and pulley teeth at the initial tension and in the process when the transmission force occurs from the condition of the initial tension are theoretically discussed. From this study the following conclusions have been obtained, (1) the direction of belt slip at the time of starting varies with initial belt tension, and the meshing state of belt and pulley teeth converges to a certain condition, (2) the stiffness of the belt system and the coefficient of change of tension are affected by the meshing state of belt and pulley, and (3) the relation between tight and slack side tensions and the relation between the angle of rotation of the pulley and the belt tension are analyzed, the calculated results agreeing approximetely with the experimental values.

View full abstract

An open die forging to form opposed projections at the center of both surfaces of a specimen is adopted as one of the basic types in forging, and the similarity of metal flow patterns between the plane strain deformation and the axi-symmetric one is examined. Influences of lubricants, working speed and die geometry are discussed and the heights of upper and lower projections of the products are compared with each other. In this research, a new flow pattern, which has a centripetal flow running only towards the lower projection, is observed in the plane strain deformation. This pattern occurs due to the side wall friction and causes the un-symmetry of the upper and the lower projections of the products. It is, therefore, not correct to assume simply the existence of similarity of metal flow patterns between a plane strain deformation and an axi-symmetric one.

View full abstract