Comparison was made on the static and fatigue strengths of one kind circular and three kinds of rectangular cross-sectional specimens under torsion. The rectangular specimens were made to have the same T/τ (torsional moment/maximum stress) as that for the circular specimen. The main results obtained are ; (1) the circular specimen has the weakest strengths in static and fatigue tests, (2) the sequence of the fatigue strengths is entirely the same as that of the static strength, and (3) for three rectangular specimens the fatigue cracks initiate at the same number of cycles, but the rate of crack propagation is different depending on the specimens.

View full abstract

It was successful to apply the X-ray microbeam technique for observing fatigue phenomena of annealed low-carbon steel. The following conclusions were obtained : (1) Misorientation and lattice strain increase quickly in the early period of fatigue process, and subsequently they continue to increase till initiation of a crack. Formation of subgrains proceeds relatively slowly in the early period, and it becomes thereafter a main process of fatigue. Formation of subgrains is not observed for fatigue below fatigue limit. (2) Density of excess dislocation of one sign in subboundaries for fatigue is much smaller than those for uniaxial tension and rolling. (3) It is considered that the reason why the half-value breadth increases during fatigue is not the formation of subgrains but is mainly the increase of lattice strain. (4) Schematic diagram of deformation in a grain for fatigue is suggested. The formation of subgrains is closely related to fatigue fracture.

View full abstract

Some approximate solutions to the problem of stress distribution at the neck of a tensile specimen have been developed by Bridgman, and Davidenkov and Spiridonova, assuming the following conditions : (1) The elastic strain can be neglected. (2) The strains are uniform over the cross section of the neck. (3) The yielding stress, therefore, is constant over the cross section of the neck. The strains of the surface of the minimum section were measured from the elongations of circumferential lines drawn finely by a diamond tool. By these measurements, it has been made clear that the axial strains over the minimum section are considerably different. It will be reasonable, therefore, to consider that the yielding stress is not constant over the minimum cross section. Using the obtained strains, we calculated the stress distribution over the minimum section, considering the difference of strain hardening over the minimum section.

View full abstract

An exact steady-state solution is obtained for the thermomechanical behavior of an elastic thick plate having a symmetrically located spherical cavity. In this report the stresses in a similar slab under such a steady thermal condition that the surface of the internal cavity and the circular regions of the bounding surfaces being exposed to any different constant temperature are investigated. The temperature function in question is constructed by using a series of periodic harmonic functions and a harmonic integral. So-called thermoelastic displacement potential corresponding to the temperature is introduced and finally the particular solution of this system is determined to satisfy the boundary conditions. The solution is illustrated by some numerical examples.

View full abstract

This paper deals with the stress in a helical corrugated shell subject to axial tension. In order to describe the form of such a shell, the corrdinate system (γ, θ, φ) is used, where coordinates γ, θ are those in cilindrical corrdinates (γ, θ, z), and φ is the length in z direction measured from the base helix. Because this is a skew system, an orthogonal system (λ, μ, n) is used together with it. The stress and strain are expressed referring to the system (λ, μ, n), and the position of the point is represented by the system (γ, θ, φ). Using these two systems, the problem treated now becomes a boundary value problem of ordinary differential equation. This equation is numerically integrated by electronic computer. The result shows that the weakest part of the helical corrugated tube is the trough, and the value of stress is close to that of the shell of rotational symmetry.

View full abstract

On the assumption that the cords are inextensible in filament winding structures, the plane stress problem of a rectangular strip subjected to normal concentrated loads acting on a pair of opposite sides is solved in this paper. The superposition of simple systems of stresses and displacements on this solution gives the solution for the problem of the extension of a rectangular strip, fixed along the sides perpendicular to the direction of extension. Moreover, on two-dimensional boundary-value problems of filament winding structures with inextensible cords, the following results are shown : First, the discontinuity of stresses and displacements on boundaries induces the corresponding discontinuity on characteristic curves passing through discontinuous points on boundaries ; second, if concentrated forces are applied to boundaries, concentrated tensile stresses are produced on characteristic curves passing through the points of application of the forces ; and third, if discontinuous points exist on boundaries, the assumption introduced to simplify the analysis is not necessarily effective.

View full abstract

The problem treated is that of two jointed semi-infinite plates subjected to a concentrated moment or doublet acting in x-direction or y-direction, and it is ascertained that the former results coincide to the results that are introduced from the later two results. The elastic properties of these two semi-infinite plates are different from each other, and a case of perfect bond and one of smooth joint are treated. Explicit formulas are given for the stress components in Cartesian corrdinates and some numerical calculations and photoelastic experiments are worked out and compared.

View full abstract

The building-machine structure system is simulated by an idealized two-degrees-of-freedom system with small mass ratio, the response spectrum of the system is obtained by the analog computation. It can be seen that the coincidence of the periods of both systems has to be evaded from the view point of the design of the machine structure, and it is proposed that the two-degrees-of-freedom system can be examined easily through T_b-T_m response spectrum. The width of the two-degrees-of-freedom system around T_b=T_m is discussed to know the shape approximately of the peak of the spectrum. The response spectrum of a single-degree-of-freedom system to the acceleration recorded upon a building is computed. A statistical computation in case of the two-degrees-of-freedom system is performed and the result is compared with the response spectrum by the analog computation.

View full abstract

This paper describes the equipment that generates and alternating torque by the action of unbalanced masses attached on planetary gears revolving around a sun gear. Each unbalanced mass that revolves around both the sun gear and the planetary gear axis produces a centrifugal force regarding the sun gear axis and another centrifugal force and a Corioli's force regarding the planetary gear axis. These forces in different ways give an alternating torque to the sun gear and the planetary gear carrier during the revolutions of the planetary gear. The author theoretically analyses the torque-generating mechanism of this equipment and confirms the analytical results by his experiments showing that using the sun gear as the output element is superior in easy control of generated torque to using the planetary gear carrier as the output element. And he also shows some applications of this equipment.

View full abstract

When a body of revolution rotates in an axial stream, skewed boundary layers develop on the body surface. Experiments are carried out on the skewed boundary layers over a body of revolution with streamlined nose. Measured velocity profiles differ a little from the quasi-collateral condition, that is, a velocity component perpendicular to the streamline appears in a layer near the wall with thickness of 1.5 mm. The magnitude of this velocity component is about 5% of the main stream velocity. Measured boundary layer thickness is compared with two theories which use the momentum integral equations. Although there is some difference in the assumed velocity profiles between the theoretical and experimental results, their agreement is fairly good.

View full abstract

This paper is concerned with investigations on rotating flow in a radial vaneless diffuse. In the vaneless diffuser, a skewed turbulent boundary layer develops under the influence of pressure rise. The author has assumed equations for the velocity distribution of a three-dimensional boundary layer and the integral relations obtained from equations of motion in the boundary layer. If the state of flow at the entrance is given, it becomes possible to calculate the pressuer distribution. Experiments are carried out to ascertain the validity of these assumptions. Comparison of the theoretical results with the experimental results shows good agreement.

View full abstract

An exact theoretical analysis is presented for the axi-symmetric flow of a viscous fluid through a porous wall into a narrow, uniform gap. In this analysis, the two equations of motion, one for the flow in the gap and the other in the porous medium, are solved under the generalized condition that the fluid may flow radially in the porous medium and as the result may slip on the porous wall surface. The important conclusions confirmed through this research are as follows. The flow characteristics are governed by the two non-dimensional parameters, [numerical formula] and [numerical formula] (k : permeability of the porous medium, h : gap width, ν₀ : constant normal velocity component at the porous wall surface). Except the limiting case λ→0, the flow generally has a tangential velocity component at the porous wall surface, and especially for the flow with large values of R, the no-slip assumption which has been made customarily to date, becomes invalid even when λ is very small. For the flow of boundary-layer type the slip of fluid may usually be ignored, but for the flow through a porous wall into a narrow gap, such as in the porous gas bearing, it is necessary to take account of the new phenomena found by the present research.

View full abstract

Dilute solutions of high molecular weight linear polymers with viscoelastic properties display significant reduction of frictional resistance in the turbulent pipe flow. This anomalous phenomenon is analytically treated with the concept that the turbulence intensity is partially surpressed by the elasticity of fluid. Combination of Weissenberg assumption and the power law for non-Newtonian viscosity is adopted as the flow model. The expression of the friction factor λ derived as the function of rheological constant n and two dimensionless numbers R₆ and N₆ includes Karman-Prandtl's formula for Newtonian fluids and Dodge-Metzner's formula for non-elastic, non-Newtonian fluids as special cases.

View full abstract

When a fluid enters a circular pipe with a rounded entrance from a large reservoir, the fully-developed velocity distribution can be attained only through the entrance region. Many non-Newtonian fluids have some elastic effects and the flow behavior in the entrance region is supposed to be affected by the elasticity of the fluid. In this report authors calculated the inlet length and the pressure loss of the viscoelastic fluid in the entrance region basing on the Weissenberg assumption and the momentum balance. As a result, the inlet length and the Hagenbach correction for the viscoelastic fluid flow are derived as functions of the rheological constant n and the dimensionless numbers R₆ and N₆.

View full abstract

This paper deals with an analysis to determine the heat transfer characteristics in the thermal entrance region for the established laminar flow of a heat generating pseudo-plastic or dilatant fluid between two parallel plates with wall heat transfer. The thermal conditions at wall treated in this paper are for both cases of uniform wall heat flux and uniform wall temperature. In this analysis, the numerical calculations were performed by means of a digital computer KDC-1. As the result of calculations it is shown that the relationship between the Nusselt number and the distance from inlet in the presence of internal heat generation is different from that in the absence of internal heat source.

View full abstract

This study has been undertaken to make clear the motion of condensate-drops falling down along a cooling surface in dropwise condensation. Since the condensate-drops in steam condensers have various weights, shapes and velocities and make the experimental results much complicated, an experiment has been carried out in the air by using a single water-drop of known weight on an inclined surface. Observation has been made with motion picture. From this experiment and theoretical analysis, the behavior of a water-drop with constant weight moving on the inclined surface has been made clear and formulas predicting the velocity, moving distance and length of the water-drop have been derived.

View full abstract

Subsequent to the 1st report of the theoretical study on the radiant heat transfer from the soot particle cloud and the experimental study on the luminous flame produced by inserting activated charcoal into the non-luminous flame of town gas, in this report the experimental study on the radiant heat transfer from the luminous flame in the case of spray combustion of A-heavy oil and light oil is described. Soot formation and its concentration in luminous flame, the relation between the effective emission thickness and the emissivity of the soot particle cloud and the influences of the differences of diameter and optical characteristic of soot particle on emissivity are discussed. And it is made clear that the emissivity of luminous flame may take various values according to fuel, spray mechanism, etc., namely according to not only soot concentration but also particle diameter and optical characteristic.

View full abstract

To examine the characters of the scavenging air flow through the scavenging ports in two-stroke engines, the inlet flow pattern through the model ports has been observed by a high speed motion camera, using a single cycle test apparatus. In general, the scavenging air flow is not coincident with the designed direction of the scavenging port at comparatively slight opening, and the efflux direction changes with the piston displacement. In a small crankcase-scavenged two-stroke engine cylinder, to keep the scavenging air flow close to the direction of the scavenging port it is effective to make the scavenge air guide long and to incline the scavenge air passage between the crankcase and the cylinder axially along the circular surface of the cylinder. In a large loop scavenged engine cylinder, to make the scavenging air flow coincide with the direction of the scavenging port in the early stage of scavenging, it is effective to set the guide plate inside the scavenging port.

View full abstract

The wear of steel is investigated under repeated rubbing in the lubricating oil containing hard solid particles. The rolling motion of particles between two surfaces gives plastic strains on the surfaces, and the surface is worn off due to fatigue failure under repeated plastic strains. By applying the knowledges on low cycle fatigue to the contact problems, the rate of stationary wear ω is formulated. The wear tests are carried out by rubbing the fixed test piece of carbon steel on the cylindrical surface of a rotating test piece of the same steel in machine oil containing SiC particles. The results are discussed in the formula. When the concentration C of SiC particles is large, the particles bear the whole load and ω is in proportion to C^1/2/δ²H or C^1/2E²/H³, where δ, H and E denote the fracture strain, the hardness and the modulus of elasticity of materials respectively. When C is small, a part of the whole load is borne by metal-to-metal contacts and ω caused by particles is found in proportion to C^3/2/δ² or C^3/2E²/H².

View full abstract

In this paper, the authors investigated the mechanisms of bending fatigue breakage of gears in one side and reverse loading conditions by examining the change in microstructure and in magnitude of residual stress with the number of load cycles and proposed N_a-N_p diagram, in which the number of load cycles after reverse N_a is plotted against the number of load cycles before reverse N_p, to indicate the bending fatigue properties of various gears such as annealed carbon steel gears, case-hardened gears etc. Moreover, the bending fatigue properties of shot-peened gears which had residual compressive stresses of considerable magnitude were investigated and it was made clear that there exists the optimum shot-peening conditions for the bending fatigue strength. The results obtained in this paper indicate some points which must be taken into consideration in the practical strength design of gears.

View full abstract

In the nose-suspended type traction gears, the angle between the projected lines of axes of pinion and gear on pitch plane often becomes equal to about 10' and end tooth bearing happens, leading to tooth failures. In JNR, as the countermeasure for these failures, heavily crowned gears are used for nose-suspended type traction gears. The calculation method for such heavily crowned gear was already introduced by one of the authors. In some types of traction gears, the positions of pinion and gear axes are comparatively stable, but even in these cases, due to distortions of gear case and bogie frame, error of gear alignment, etc, the axes of gears become non-parallel slightly. For this reason, such gears of JNR are relieved at the tooth end in order to avoid end tooth bearing. Up to this time, the gear tooth stress of such end-relieved gear has been calculated with uniform loading assumed over the total facewidth. But it has become clear through experiments that the stress of end-relieved gear grows considerably larger than that under uniform loading, and the authors proposed a calculation method for the stress of end-relieved gear. For practical application, it is sufficient that the factor of stress increase induced by end relief is considered for the tooth stress when uniform load is applied over the total facewidth, and they gave diagrams for deciding the stress increase factor. By experiments they demonstrated that introduction of this factor in stress calculation would be very reasonable.

View full abstract

In the present paper, the author has derived a method for calculating the stresses in herical gear teeth by applying the solution for the moments induced in a cantilever plate with infinite length subjected to a concentrated load. That is, the helical gear tooth was regarded as a cantilever plate with finite length and the distributed load on the contact line of tooth was considered as a group of concentrated loads per unit length. And then, from the further investigation for this calculating method, a simplified formula for the calculation of stresses in helical gear teeth has been developed. The stresses in gear teeth calculated by these methods are shown to be in good agreement with those measured by wire resistance strain gauges cemented in the actual gear teeth.

View full abstract

Few investigations are available concerning the profile modification of helical gear teeth and there are many important problems which should be clarified. The object of the present investigation is to find out the effect of the profile modification on the dynamic properties of helical gears, and furthermore to derive the method for determining the optimum amount of profile modification. The profile modification of both non-crowned and crowned helical gear teeth was dealt with in this investigation. As the result of the investigation, it was found that the vibration caused by a gear mesh was reduced by giving a suitable profile modification to the teeth, especially with regard to the crowned helical gears, and the optimum amount of profile modification to the teeth of both non-crowned and crowned helical gears could be determined from the tooth deflection under working load, similarly to the case of spur gears.

View full abstract