Sachs and Mesnager described measurements of three principal residual stresses of an isotropic cylinder in the directions of radial, circumferential and longitudinal axes. They are not applicable to a curvilinear-orthotropic or polar-anisotropic body because of their large errors. For the purpose of study about "Ate"of a log (reaction wood), one must know the order of the residual stress values of it. Though a log is composed of biologic fibers, it is proper to consider it approximately as a polar-anisotropic body in mechanical handling such as deformation or force in case of log sawing. The authors derive the measurement of three principal residual stresses in a polar-anisotropic cylinder. The authors'formulas involve Sachs'and Mesnager's formulas as special case of elastic compliances. Then the authors measured and calculated three principal residual stresses in a log by their method.

View full abstract

For measurement of two principal residual stresses existing in an isotropic plate such as metallic one, Treuting-Read's or Kuno-Doi's Method by deflection, and Doi's Method by strain gauge have been proposed. But they are not applicable to an orthotropic plate because of their large errors. The authors proposed in this paper the measurements of two principal residual stresses in an orthotropic plate when their directions were known. The experimental technique is as follows : thin layers are successively removed from one surface of the plate and the resulting change of curvature (Curvature Method) or strain (Strain Method) of the remaining plate is measured at any time, then their stresses are calculated. The authors'methods treat the measurements of an isotropic plate as a special case of elastic compliances. Finally, the authors showed two principal residual stresses in a wooden plate.

View full abstract

In this paper, the author has developed a new method for solving the two-dimensional stress problems using a digital computer. As examples, he calculated the stress distribution of a rectangular plate subjected to symmetrical loading and the stress concentration factors of a semi-infinite plate having a circular notch or a V-notch under tension. The stress distributions of an infinite plate subjected to a point load are used as a fundamental solution in this method. To satisfy the boundary conditions, the body force distributed on the boundary curve imagined in an infinite plate is used.

View full abstract

A numerical method which can be applicable to the nonlinear axisymmetric bending of arbitrary shells of revolution is proposed. It is based on the iterative scheme and the finite difference approximation, and is found to show good convergence and to be practical. Using this method, the large deflection problems of the corrugated diaphragm and bellows are solved. The results show the unnegligible nonlinear effects and the limitations of the linear theory and the approximate nonlinear solution now used.

View full abstract

This paper presents an analysis of the uniform bending of an anisotropic plate by using an anisotropic plasticity based on the maximum shear stress yield condition. When the stress points remain in two special sides of the yield surface and the plastic deformation proceeds, this problem can be analyzed as a plane strain problem. But, in the other cases, it is found that the constant strain e₃=α in the direction of the plate width becomes positive or negative. In the latter cases the stress distributions plotted on the π-plane show that the stress points are not only on the sides, but also at the corners of the yield surface. It is found that there is the region in which the total strain theory can be applied for the plastic bending of plates, but this fact is neglected. Therefore, the imperfect results of analysis still remain mainly near the neutral surface.

View full abstract

The transverse oscillation of a visco-elastic cantilever beam is analysed, using a three-element model. And the coefficients of solid viscosities for metals are determined by comparing the theoretical results with the experimental ones. The effects of solid viscosities on dynamic load factors are also investigated for the cases of a beam and a bar when transverse and longitudinal step impulsive loads are applied. The results show that the solid viscosity can not be neglected for the stress behaviour under dynamical loads.

View full abstract

Various formulas and methods for predicting the ultimate pressure of the cylindrical pressure vessels have been investigated. However, concerning the ultimate pressure it seems that there exist some discrepancies between the experimental data and the predicted values for the thick walled cylinder though they coincide for the thin walled cylinder. In this paper, at first the experimeatal data on the ultimate pressure for various diameter ratios of mild steel and aluminium cylinders are presented, then comparisons made between the experimental data and some predicted values, and it is pointed out that the predicted value obtained from the stress-strain relation under torsion coincides well with the experiments when a new formula proposed by Nakanishi is used in our calculations as the relation between pure shear and simple shear.

View full abstract

Transmission electron microscopy was used to study the structure developed in pure and very fined polycrystalline aluminium, several microns in grain size. Observations were made on the specimens fatigued at stress amplitudes, 8.5, 5.8, 4.8 and 3.8 kg/mm². At the highest stress tested, a cell structure was formed even with such small grains and very dense parts of dislocation were observed near the grain boundary. The transition from dislocation tangle to cell-boundary was clearly seen. Dislocation loops which were likely to be formed by the condensation of vacancies were rarely observed. Main feature of the structure at the medium stress levels was the so-called dislocation patch which consists of the elongated loops arrayed in the direction <211>. Dislocation loops distributed in wide area were observed at the lowest stress. Surface observations were also done with a light microscope. Speculation was made on the propagation of the fatigue crack.

View full abstract

If the topography of fracture surfaces represents the peculiar pattern to various reasons of fracture, the reason could be known by observing them in detail, and further the observation might bring out some information on the fracture mechanism. From the above point of view, we conducted a fractographic investigation on fatigue fracture surface under rotating bending and repeated tension tests. The appearance of the surfaces is apparently different due to their location observed and the stress level in rotating bending test. But the surface pattern from the repeated tension test seemed to be same in the range from the stress at the fatigue limit to rather large stress under which plastic deformation occurs and it closely depends upon the heat-treated structure. The impact fracture surface consists of austenite grains, dimple pattern that shows ductile fracture, and other complex patterns.

View full abstract

The fatigue property of low carbon steel under the composite stress which was made by superposing the ordinary rotating bending stress and the vibrating stress of higher frequency was examined in this study. The experimental results were discussed by focusing on the effect of the ratios of the magnitudes and the frequencies of component stresses. It was ascertained that in some ranges of these parameters the fatigue life under the composite stress could be estimated by the criterion that the composite stress was equivalently replaced by the stress of constant amplitude whose magnitude was the sum of those of the component stresses. For the values of the parameters outside the ranges, the criterion did not hold and the fatigue life was reduced appreciably from the value given by the criterion. Another method for estimating the fatigue life by calculation was also presented, which was available for the latter case.

View full abstract

The author describes empirical formulae of bending and torsional stress concentration factors for a crankshaft of marine engine and analytical methods for calculation of bending and torsional moment as well. He also describes a method of dynamical stress measurement and cites some examples of the results of analysing the stresses on crank fillet and additional stresses due to vibrations in crankshaft. Furthermore, bending and torsional fatigue tests of model crankshafts and defective materials making use of large specimens are summarized. The author concludes that the stresses of crankshaft can be approximately calculated by his investigations and strength design of crankshaft may become more precise when the additional stresses due to vibration are exactly calculated.

View full abstract

In vibratory systems having multiple degrees-of-freedom and under parametric excitation of frequency ω, oscillations of "summed and differential types"with frequencies ω_i (≒p_i), ω_j (≒p_j) take place when ω becomes nearly equal to sum of and difference between two natural frequencies p_i±p_j=p_ij, i.e., ω≒p_i±p_j, even if there is a damping force in the system. In the present paper, the characteristics of these kinds of oscillations are studied in detail, and it is shown that the magnitude of damping ratio has considerable influences on the characteristics of unstable oscillations, especially on unstable regions and negative damping coefficients, and that unstable oscillations can occur only in summed type and not in differential type.

View full abstract

Although theoretical analysis and experiments on this problem have been attempted by others, the stability analysis for the jumping granular solid has not been carried out. In this paper an exact solution for the periodic motion of the jumping granular solid is derived, and its stable regions are determined. The following conclusions can be drawn from this paper. (1) When the coefficient of restitution R&nedot;0, the stability boundaries K is given by [numerical formula] (2) When R=0 and the relative acceleration of the granular solid just after an impact y₀>0, K is given by [numerical formula] (3) When R=0 and y₀<0, K is given in the table, in which K=bω²/(gcosα), b : amplitude of the track surface. ω : angular frequency. g : acceleration of gravity. α : inclination of the track surface. n : 1, 2, 3, …….

View full abstract

Experiments were carried out on velocity distributions of the three-dimensional boundary layer on a cylindrical half body with a hemispherical nose rotating in axial stream. When the body remaind stationary, there was a separation bubble just behind the hemisperical nose. This separation bubble disappeared when the ratio of peripheral velocity of the cylinder to the reference main stream increased up to about 0.44, because of the additional acceleration produced by centrifugal force at the nose. This had also an effect on the polar plots of laminar boundary layer to modify the velocity profiles in the meridian direction. Momentum thicknesses obtained from the experiment agreed well with results of two theories. As in case of two-dimensional flow, boundary layer separation occurs far up-stream from the point which is predicted by the theory of the 4-th order polynominal velocity profiles.

View full abstract

Propagation behavior of initial shock front in a shock tube is investigated at moderate shock strength. Experimental result on the shock speed suggests the following flow model. At early stage of the diaphragm opening, a compression wave races into the driven gas, and it generates and intensifies a shock front. Once the maximum speed is achieved, the shock front gradually diminishes in intensity, overtaken by a rarefaction wave. A platinum wire heat gage is also used to confirm the validity of this flow model. The formation distance is more than 20 times the tube diameter, and becomes longer at larger diaphragm pressure ratio and lower driven gas pressure. Deviation of the maximum shock speed from the theoretically calculated speed depends mainly on the formation distance. The formation distance and the maximum Mach number are better correlated to diaphragm pressure ratio when parametrized by driven gas pressure. As a general remark the formation distance significantly affects the shock tube experiments as well as the shock Mach number does.

View full abstract

The two-dimensional potential flow on the blade surface of a centrifugal impeller can be calculated exactly. But no effective method for practical use to calculate the flow at any point in a centrifugal impeller passage seems to have been developed yet. In this paper, the authors show an approximate, but comparatively simple method applying the velocity function method which was developed previously by one of the authors to facilitate the calculation of the flow on the blade surface. Using a free impeller type apparatus, the authors measured the relative flow behind a centrifugal impeller and compared that result with theoretical one. It has been made clear that at a flow rate larger than about half of shockless entry flow rate, the theory expresses the practical flow pretty accurately.

View full abstract

In this paper, a theory of a three-dimensional compressible flow in axial turbomachines with varying boundary shapes is presented, in which the assumption is made that the axial velocity profiles and the specific mass flow profiles are respectively expressible by a single parameter. For these two parameters the flow equations reduce to non-linear ordinary differential equations and can be solved numerically very easily. Comparisons with two more accurate calculations are made to investigate the accuracy of the present theory.

View full abstract

An existence of a back-flow on the suction side of impeller blades of turbomachines and its remarkable effects on the characteristics have been admitted as well known facts. But its cause and mechanism are scarcely clarified. In this paper, such problems were treated by measuring the pressure distributions and flow directions in the passage between two blades, on the outside of three different axial-flow impellers, which had a big scale back-flow. As the result, it was discovered that the back-flow was regarded as the separation of flow from the surface of blades, due to the increase of discharge around the leading edge, which was divided by the stagnation stream line, on the outside of impeller blades. Furthermore, the unstable performance of head-discharge curve and the conspicuous increase of shut-off horsepower were also discussed in this paper.

View full abstract

This paper deals with the problem of combustion in the NSU-Wankel type rotary piston engine, and presents a summary of the test results. The authors conducted many tests using both 1-rotor engine and 2-rotor engines and cleared up the influences of variation of induction system, number of spark plugs, ignition timing and air-fuel ratio on combustion of fuel-air mixture in this new engine by measuring the gas pressure and flame speed by means of pressure indicators and ionization gaps, respectively.

View full abstract

This paper presents an approximate method of calculating the mean roll pressure in the hot rolling of rods and flat plates with finite width. The equetion is proposed as follows, [numerical formula] This is checked against measurements made when rolling some rods and flat plates. There is fairly good agreement between the calculated and the measured values. Moreover, the results calculated from this equation nearly agree with the values obtained from Sims'formulae where the ratio width/thickness for the material is infinite.

View full abstract

In this study the stability of externally pressurized air journal bearing which has a peripheral row of feed holes is experimentally investigated to make clear how the onset speed of selfexcited whirl is affected by the three factors of supply pressure, number of feed holes and bearing clearance. As the result, the following facts are revealed. (1) Under a certain supply pressure and number of feed holes exists an optimum clearance which gives a maximum whirl onset speed. (2) Under a certain supply pressure and number of feed holes exists a clearance which gives a maximum whirl frequency, and this clearance is slightly smaller than the optimum clearance. (3) The whirl onset speed is directly affected not by the supply pressure and the number of feed holes but by the whirl frequency and the bearing clearance. The item of (1) can be explained by those of (2) and (3).

View full abstract

This paper presents some methods to obtain easily the rotational conditions, i.e., the geometrical conditions under which the driving link can make a full rotation. The condition is expressed in terms of the limits of the lengths of the couplers for the spatial four-bar mechanisms whose driving links and couplers do not vary their lengths. The calculating and constructive methods to get the rotational conditions and their examples are presented for three mechanisms : A₁·2·1 mechanism having two revolute and two spheric pairs, A₁·2·2 mechanism having one revolute, one prismatic and two spheric pairs, and A₀·2·1 mechanism having two revolute, one cylindric and one spheric pairs.

View full abstract