A hypothesis for the mechanism of non-propagating fatigue crack related with the stress amplitude and the critical strength of a material in a region ahead of the fatigue crack tip below notch root was proposed. It was shown that the calculated results based on the hypothesis well explained the experimental results of a rotating bending fatigue test of a round bar specimen with a circumeferential deep hyperbolic notch (i.e.Neuber type notch) in each case of two sample (fine grained steel and coarse grained steel) which had been obtained by the heat-treatment of a mild steel.

View full abstract

Rotating bending fatigue tests were performed on three kinds of aluminum coated steels. The effect of three aluminum coating methods, namely electroplating, hot dip, and spraying on the fatigue strength of the steels were studied. The results obtained are as follows. (1) The effect of electroplating on the fatigue strength of the steel is not recognized at all. But a little effect of the spraying on the fatigue strength is observed. (2) In hot dip aluminum coating of the steels, a decrease of the fatigue strength is observed. This is considered to be due to the formation of cracks in a brittle alloy layer. Such cracks act as shape notches on the surface of steel base. Moreover, the formation of cracks decreases the effective area of specimen whic supports the alternating stress. (3) The addition of 1% beryllium to molten aluminum bath markedly improves the fatigue strength of hot dip materials.

View full abstract

Elastic deformation of polycrystalline metal was studied analytically with simplified models of polycrystals. The stress and the strain distributions in a plate having discontinuity of elastic constant were calculated adoping continuum elasticity theory. Both successive approximation method with Fourier series and finite element method were used. The deformation mode of a grain is represented with constraint ratio defined previously. It was found that the deformation mode of polycrystalline metal is affected markedly with the ratio of the width to the length along the stress axis of each grain. When the grain is long enough in the direction of stress axis, the strain becomes nearly constant across the grain. On the other hand, when the grain is flat in shape, the stress is nearly constant. The distribution of constraint ratio or strain across the grain boundary parallel to the stress axis is represented approximately with a sine curve.

View full abstract

There are many important problems in the field of the study of one-dimensional stress wave propagation. One of them is the stress wave propagation in a bar of variable cross-section. Several authors treated this problem, but the function which represents the variation of the cross-section along the axis is usually restricted to be in a simple or special form. When a stress wave encounters a discontinuity of area, the reflected wave and the transmitted wave are determined by the areas of the two cross-sections. In this paper, a bar having an arbitrary cross-section is assumed to be a bar having continuous multi-stepped diameters. Based on this assumption, an analysis of the stress state, especially at the wave front, is performed.

View full abstract

For the one-dimensional rate-independent longitudinal wave propagation problems, the method of analysis using the hodograph along with the characteristic field, which was proposed by Karman, is discussed and generalized. Many solutions using the hodograh are shown. The main features of this method are as follows : (i) On the hodographic plane the image of characteristics can be drawn with only two rules specially prepared. (ii) On the hodographic plane the initial and boundary conditions are introduced intuitively. (iii) The image of characteristics in the hodographic plane can be drawn in advance of characteristics in the physical plane. Use of hodograph has proved convenient to analyse the one-dimensional longitudunal impact.

View full abstract

This paper presents an analytical solution for floating-ring journal bearing and a theretical analysis of unbalance vibration of a rotor-bearing system. Theoretical solutions for fluid film force, friction coefficient and stifness and damping coefficients are derived. The effect of a floating-ring journal bearing on unbalance vibrations of a symmetrical rotor-bearing system and some design recommendations for optimum bearing dimensions are discussed.

View full abstract

For the analysis of the machine-tool dynamics, a mathematical model employing a signal flow diagram has been developed by the authors, through which a direct simulation and a precise analysis of the various nonlinear vibrations in the machine-tools, such as regenerative vibration, stick motion, etc, are made possible. The point of this method lies in direct introduction of the analysis in the design of a real machine, making it possible for the machine-tool designer to take account of the vibration problem in his design work prior to drawing the conclusion. In this paper the following problems are treated by the above method. (1) Vibration problem of machine-tools anchored to the ground. (2) Self-excited chattering taking place in the long ram of the milling head of horizontal boring machines.

View full abstract

Concerning double concentric jets, the authors performed experimental studies on influences of nozzle conditions such as thickness of nozzle wall or boundary layers on the inside and outside walls of a nozzle on the characteristics of jets in the main region. And the following results are obtained. (1) In the main region, the dimensionless radial velocity profile fits the cosine curve independently of values of δ^^-, λ and χ^^-. (2) The velocity decay on the central axis and the intensity of turbulence are hardly affected by the value of δ^^- for small values of λ. But for large values of λ, the tendency of velocity decay becomes remarkable and the intensity of turbulence is affected heavily by the wake behind the nozzle wall with an increase in the value of δ^^-. (3) Values of ι^^- and ε^^- vary in radial and axial directions and these values are different respectively in both cases of large and small values of λ.

View full abstract

Recently the blade row of axial flow machinery is designed using the two-dimensional cascade data, which depend on many experimental results. However, for the purpose to design a low pressure fan without the cascade data, the method using a modified isolated aerofoil theory with the interference coefficient method may be effective. But the reliable information about this analysis has been seldom published. The purposes of this paper are to provide the reliable information about the design method using the interference coefficient. New carpet diagrams for designing the axial blade are presented and they are compared with the experimental cascade data.

View full abstract

We studied experimentally the frictional moment (torque resistance) and the pressure drop (loss of head) of the flow in co-axial cylinders of which the outer cylinder rotates, using water and a glycerine-water solution. Comparing these experimental results with those of the inner rotaing cylinder, we clarified the behaviors of shearing stresses of the flow in the co-axial cylinders. When the clearance ratio β is large, the coefficient of the frictional moment C_M and the coefficient of the loss of head λ is extremely smaller than that of the inner rotating cylinder in a wide range of rotating Reynolds numbers R_ω and axial Reynolds numbers R_c for the turbulent flow. When R_ω is slightly larger than the critical value from laminar to turbulent flow for R_c=O, the values of C_M and λ approach the theoretical values of the laminar flow as R_c increases from zero to a certain value. In the case of β=0.03, λ shows a good agreement with the theoretical value for turbulent flow.

View full abstract

Concerning the model-prototype similitude of pressure surge due to rotating flow from exit of hydroulic-turbine runner, the experimental studies through model tests under various head conditions and based on a comparison of measured data of model and prototype under dynamically and geometrrically similar conditions were conducted with the following results. Frequency and amplitude of draft-tube suges were proportinal to the square root of head and direct proportional to head respectively, and also surge frequency was inversely proportional to the dimension. From these relations, similitudes of frequency and amplitude were elucidated with dimensionless parameters. These parameters between the model and the prototype coincide very well at definite values of gate opening, revolving speed per unit head and unit dimension, and cavitation coefficient, moreover at corresponding positions for amplitude. Using these parameters, the procedure to predict draft-tube surge characteristics of the prototype from the model test results has been established.

View full abstract

The motion of a vapor bubble in subcooled boiling was studied experimentally using slightly subcooled water, ethanol, and carbon-tetrachloride under the atmospheric pressure. The degree of subcooling ranged usually from about 2°C to 50°C, and in a few extreme cases it was extended to below 1°C. The experimental data for the collapse stage were found to be well correlated by the equations which were derived by assuming a laminar heat trasfrer between the spherical bubble and the surrounding liquid with uniform subcoolings.

View full abstract

A study was made on the ignition delays of n-butane-oxygen mixtures to clarify the high-temperature oxidation mechanism of gaseous higher-hydrocarbon fuels. The ignition delays were measured in a shock tube by detecting the visible emissoin from the reactants employed as driven gases. Experiments were carried out for various equivalence ratios and for a range of shocked-gas temperatures ancl pressures. Using n-butane-argon mixtures as driven gases of the shock tobe the cracking reaction, supposed to occur prior to ignition, were observed by means of the integrated schlieren method and the gas chromatography. A reaction model for the oxidation of n-butane was proposed, which was similar to the well-established reaction model for hydrogen-oxygen mixtures, and iginition delays were calculated on the basis of this reaction model. Calculated igniiotn delays and the dependence of the apparent activation energy on the equivalence ratios agreed fairly well with experimental results.

View full abstract

To reveal the mixture formation and combustion in a shallow-dish type engine with a multi-hole nozzle, a high-speed photographic study has been made. Comparisons of the photographs taken under various conditions have let to the following results. Under no swirl condition, various kinds of gas motion are produced in the flame in the piston cavity soon after ignition. They govern the subsequent state of combustion, finally affecting the air-utilization in the later stages of combustion, and they are dependent on the diameter of the nozzle-hole as well on the direction of injection. When a swirl is employed, the radial extension of the spray is considerably reduced with its intensity, thereby the subsequent spread of the flame being suppressed. Unless a sufficient penetration of the spray is assured, air-utilization is poor in the peripheral space, leading to the so-called overswirling.

View full abstract

The mixture formation and combustion in deep-bowl type direct-injection engines have been investigated by means of high-speed photography. This type of combustion chamber is often adopted in compact high-speed engines and may be characterized by squish motions created by piston movement. At first, the effects of the squishs on combustion have been evaluated in several chambers. It has been revealed that a flame rapidly penetrates into the clearance space soon after ignition, assisted by the reversed squishs motion, thereby the flame lasting for a longer period than in the combustion chamber. This implies that a too early outflow of fuel into the clearance space is to be prevented in order to ensure a favorable combustion. Indeed the swirl well removes such an excessive outflow, but at the swirl or a too high intensity the mixture formation in the periphery of the combustion chamber is hampered, thus inviting poorer air utilization in the clearance space.

View full abstract

The methods for eliminating secondary injection and cavitation have been investigated for highly rated diesel engines. Retraction characteristics of a delivery valve of a fuel injection pump have been particularly investigated theoretically and experimentally. The results of the above investigation are as follows. (1) The velocity and the duration of a delivery valve retraction should be suitably selected to avoid both secondary injection and cavitation. (2) Retraction velocity of a delivery valve should be as fast as possible at early stage of retraction and suitably shower at later stage to avoid elongation of injection duration and secondary injection respectively. (3) Two new type pumps have been developed with excellent injection performances free from secondary injection.

View full abstract

The transient phenomena of the driven shaft torque transmitted from the driving shaft through the dry friction clutch are studied theretically and experimentally using a model of clutch and the analog computer. Substantial agreements are found between the theoretical results and experimental observations. Results obtained are as follows ; The characteristics of friction on the slip surfaces of the clutch has a considerable effect on the time rate of increase of the driven shaft torque. An increasing friction force with an increasing relative speed between in-and output plates of clutch depresses the increase rate of the driven shaft torque and causes the retardation of the engaging time and an increase in the amplitude of the torsional vibration of driven shaft with an increase of the initial speed of the input plate of the clutch. With the same pressure on the plates of clutch, the excitation, the sustenance and the damping of the torsional vibration of the driven shaft depend on the initial speed of the input plate of the clutch.

View full abstract