In order to investigate the relation between the critical stress amplitude of slip-initiation at the crack tip and the crack length, fatigue cracks of tensile mode grown up in an eutectoid steel are examined under various magnitudes of the mean stress. Taking note of the fact that slip-initiation phenomena are controlled by shearing stress amplitude, the relation between the mean stress and the critical stress amplitude of slip-initiation can reasonabaly be explained by the stress concentration factor at the microregion near the crack tip, resulting from the crack opening or closing, and the crack tip opening stress.

View full abstract

Push-pull fatigue tests were conducted on butt-bonded three layer composite plates composed of iron and copper, to study the effect of the difference in elastic moduli between the constituent materials on the behavior of the fatigue crack propagating in parallel with the interfaces. It was observed that, when a crack propagated in the core iron which was flanked with copper plates having a lower elastic modulus than iron, the growth rate was higher than that of a crack propagating in the monolithic iron plate, and that, for the crack propagating in the core copper which was flanked with iron plates, the growth rate was lower than that of the crack propagating in the monolithic copper plate. These observations agreed with the trends of k_I-value in each composite plate obtained by numerical computations based on F. Erdogan's analytical method and on the finite element method.

View full abstract

By using a split Hopkinson bar type apparatus, impact bending tests of 0.45% carbon steel were carried out on U-notched annealed specimens and quenched-and-tempered ones. The load-deflection curves were determine easily by the tests. The temperature dependences of the yield load, the maximum load, the fracture load and the absorbed energy were determined within the temperature range from room temperature to -195°C. The absorbed energies agree fairly well with the ordinary Charpy values at these temperatures for both specimens. Repeated impact bending tests with a short striker bar were also carried out and it has become clear that the crack initiates at or slightly beyond the maximum load.

View full abstract

In this investigation, we are concerned with the deformation and strength of an underground pipe subjected to the soil pressure unsymmetrically distributed in circumferential direction, by using the fundamental differential equation of a cylindrical shell introduced by one of the authors. The ground is treated as a semi-infinite elastic solid. Boussinesq formulas or modified equations by Frohlich are used for the analytical calculation of soil pressure on a buried pipe. In addition, the modulus of the soil foundation is introduced. The results for the case of two traveling concentrated loads maintaining a constant distance on the soil surface are easily obtained by super-position of the solutions. Numerical calculations are reported and demonstrated for various combinations of the positions and distances of point of application of loads.

View full abstract

In this paper the strength of a rotary kiln, which is widely used in the field of chemical industries, is discussed from both theoretical and experimental approaches. The analysis is made exactly by regarding the rotary kiln as a cylindrical shell and a full-sized model of kiln is used in the experiment. In the analysis the various factors of loads, such as the weight of kiln and deformation of tire, force of the loading block acting on tire, those due to the contents and driving equipment, etc., are taken into consideration. A detailed discussion on the results of analysis and experiment confirms the importance of an appropriate design for the blocks between kiln and tire. Furthermore the effect of thermal stress on the strength of the rotary kiln is examined.

View full abstract

In this paper a method of solution is developed for the two-dimensional problems on the compression of granular materials by applying the finite element method to the fundamental equations of the statics for granular materials derived by T.Nagao. This method has been applied to the compression of granular materials in a square box with one of the patterns in several kinds of dimensions set at its bottom. The results are compared with those obtained from the compression tests of silica sands, and the theoretical stress values are found to agree fairly well with the experimental ones. The stress state in the so-called"Elastic Zone"can be made clear by using this method; analysis of this problem has been considered intractable by the existing methods.

View full abstract

A new method of measuring a three dimensional unsteady flow is presented in this paper. The method was applied to the measurement of a flow just before the impeller of a centrifugal fan operating in the state of rotating stall, and complex flow patterns in that place were made clear. The method described in this paper is very useful to elucidate the complex flow in turbo-machines which operate in an unstable state and the flow which varies its direction over a wide range periodically.

View full abstract

The stability of flows of two different fluids which are separated by a thin elastic plate is studied on the basis of hydrodynamical stability with respect to small disturbances. The governing equations are derived assuming that both fluids are inviscid and incompressible, and compressive or tensile in-plane forces act on the elastic plate. The growth rate as a function of the wave number, the critical wave number and the critical in-plane force are found with the aid of an eigen relation. For long waves the stability of the system depends only on the density defference of both fluids and does not depend on the properties of the elastic plate at all. In the case in which the lower fluid is heavier than the upper one, instability does not arise, unless magnitude of compressive in-plane force exceeds the critical value.

View full abstract

This paper presents an experimental investigation of the deflection angle and the characteristics of a two-dimentional turbulent free jet issued into a still air after deflected by a one-side flap. The flap was fitted up at the lower edge of a rectangular nozzle of aspect ratio10. The center flow of the jet was confired to be two-dimensional. The velocity and static pressure distributions and the lines of the maximum velocity were measured for various flap widths and flap angles. It was found that the deflection angle of a jet is equal to the flap angle when the flap width exceeds 2.5 times the nozzle width and the decay of the maximum velocity and the spread of the deflected jet are different from those of a free jet in the absence of the flap.

View full abstract

In order to examine the influence of an annular protrusion, which is set in a pipe in Poiseuille's flow through the pipe, two kinds of experiments were carried out; that is, head loss of the flow downstream of the protrusion was measured at several positions of the flow, and the aspect of the flow was observed by the aid of a dye streak. The flow is possibly disturbed by the protrusion even when Reynolds number (R_e) is less than its critical one, and then the friction factor becomes greater than 64/R_e. And it also becomes clear that the friction factor is given by such a function of the distance of the flow from the protrusion that the friction factor decreases as the flow goes down, and finally the factor becomes 64/R_e. From the results of a series of experiments the damping-up distance the local turbulence due to the protrusion is given as a product of power function of internal diameter of the pipe and Reynolds number of the flow. By observing the dye streak through the pipe the damping-up distance is proved to coincide with the result obtained by the experiments of the pipe friction.

View full abstract

Water issues downward from a straight vertical tube into the atmosphere. Velocity profiles near the tube outlet were measured by using a Laser-Doppler velocimeter specially designed. It was revealed that the gravity force considerably affects the flow pattern. The following results were obtained: (1) Inside of the tube, although velocity profiles at a position by one tube-radius upstream from the tube end are the same as that of the Poiseuille flow, the velocity near the tube axis increases as the flow comes close to the tube outlet. It was found that the flow pattern at the tube end is no longer that of the Poiseuille flow. (2) Outside of the tube, since the axial velocity component in the free jet increases as it goes downstream and the rate of the increase of the axial velocity near the free surface is larger than that near the jet axis, the flow pattern approaches a uniform one at a position by one tube-diameter downstream from the tube end. (3) The axial velocity near the flow axis both inside and outside of the tube increases with a decrease of the Froude number, that is, with an increase of the effect of the gravity.

View full abstract

A theoretical study was carried out to investigate such a flow that a liquid of low viscosity issues downward from a vertical tube into the atmosphere. Numerical calculations were performed by using both vorticity transfer and momentum theory. From the study the following results were obtained: (1) Inside of the tube, velocity profile gradually becomes different from that of the Poiseuille flow as the flow comes close to the tube outlet. For R_e=753, F_r=9.75 ×10^-2, the velocity on the tube axis at the tube end becomes 1.23 times as large as that of the Poiseuille flow. Accordingly, the flow pattern at the tube end is no longer that of the Poiseuille flow. Issuing from the tube outlet, downstream, and the flow at a position by one tube-diameter downstream from the tube end settles to uniformity. These tendencies becomes remarkable with an increase of the ratio of the Reynolds number to the Froude number. (2) The pressure on the tube wall takes a minimum value at a position by the tube-radius times 0.7 from the tube end. This value is lower than the atmospheric pressure.

View full abstract

When a heavy fluid is superposed over a light oil sheet adhering to a horizontal solid wall, the gravity will cause the lower fluid to rise to the upper fluid. In such a case, whether or not there is a stable stationary state is studied theoretically. It is assumed that the flow is two-dimensional, the fluids are incompressible, the upper fluid is ideal and the lower one is viscous. To the fourth order for wave number and wave amplitude, the height equation of the interface is derived from the long wave expansion. It is found from this equation that nonlinearities of the governing equation have a stabilizing effect and there is a stable equilibrium state with amplitude of about 0.4 times the height of the lower liquid layer.

View full abstract

Two and three stage amplifiers in which the main jets of the first and the second stages act upon the main jet of the next stage directly without transmission line were developed as analogue amplifiers with high gain and excellent frequency characteristics. The measurements of their characteristics were carried out by measuring the flow gains. The frequency characteristics were determined by measuring the flow gains. The frequency characteristics were measured by flow visualization of the jets employing a synchronized stroboscope. The gain of the two stage amplifier developed in this study was found approximately equal to that of the two stage beam deflection amplifier connected in cascade and the frequency characteristics of the amplifiers developed satisfactorily showed a constant amplitude up to 400Hz.

View full abstract

Two-dimensional free convective heat transfer within an enclosure having a parallelogram shaped cross section and defined by two vertical walls of different temperatures and two oblique parting walls has been investigated. Two thermal conditions on the parting walls are considered namely, (1) linear temperature variation and (2) a perfect thermal insulator. Numerical computations are carried out for conditions: Prandtle number 0.72, aspect ratio 1, oblique angles -π/3 to π/3 and Rayleigh number 1 to 2×10⁶. Heat transfer coefficients on the hot wall are measured for air on the condition that the temperatures of the parting walls vary linearly and the experimental results agree with the numerical ones.

View full abstract

Infrared emissivities of iron, mild steel and hard steel are measured at temperatures from 850 to 1350 K including the A₁, A₂, and A₃ transformation points and at wavelengths from 0.95 to 12.0 μm. The radiative properties are found to be continuous at the three transformation points, and the experimental results can be explained well by a two-electrons-type dispersion model of optical constants. Electronic constants of the equation are given and formulated in order to correlate the macroscopic properties of the radiative heat transfer at temperatures of 800 - 1600 K and at wavelengths longer than 0.7 μm. Spectral and total emissivities are calculated for the three metals as the functions of temperature and phase of crystals.

View full abstract

The analysis is carried out on the reflective properties of a painted layer of absorbing and anisotropically scattering medium. The cases of oblique and hemispherical incidences are treated together with a special case of axisymmetric incidence. The azimuth-dependent functions in the case of oblique incidence are expressed in a finite Fourier series whose argument is the azimuth difference between the direction of incidence to the layer and the direction of the diffuse radiation inside it. The effects of the optical thickness, the incidence angle, the optical properties of the pigment (TiO₂, Fe₂O₃ and carbon) and the reflection characteristics of the substrate (specular or diffuse) on the bidirectional, directional and hemispherical reflectances are examined in the wavelength range from 0.35 to 20 μm. The results for hemispherical incidence with uniform irradiation show a coincidence with those for axisymmetric incidence at a polar angle of 60°.

View full abstract

An annular mist flow using air and water at room temperature has been studied experimentally in a vertical pipe with a nozzle along the axis of the pipe for supplying liquid. Observations were made of flow patterns of liquid on the inner surface of the pipe, and measurements were made of pressure losses in pipe, profiles of radial distribution of liquid droplets and total flow rates of the liquid droplets. Changes of these four factors along the pipe were measured in the non-equilibrium region. It was found that the non-equilibrium length should be decided by a position where any changes in the four factors mentioned above could not be recognized in the axial direction. For relatively high velocities of air, i.e., for apparent gaseous Reynolds number R_ego ≥9.4×10⁴, it was ascertained that the annular mist flow reached equilibrium at a distance of 170-190 diameters from the nozzle outlet when apparent liquid Reynolds number R_elo = 62.1-183.6.

View full abstract

This paper deals with the response of a nonlinear system with collision subjected to a nonstationary Gaussian shot noise which is a simple model of earthquake ground accelerations. The system considered consists of an oscillator and two reflectors at both sides of it, and the oscillator collides with each reflector. The response of the system is analyzed based on Markov-vector approach. A distribution based on the Maxwell-Boltzmann distribution is proposed as an approximate solution for the unsteady Fokker-Plank equation which gives the time-dependent joint probability density function for the response displacement and the velocity. The approximate solution contains three unknown functions of time, which are determined by employing the method of weighted residuals (the method of moment). Various stochastic properties of the response are derived from the approximate joint probability density function, including the probability density functions and the second moments of the displacement and the velocity, the average number of collisions, and the probability density functions, the mean values and the probability density functions, the mean values and the standard deviations of impact velocity, impulse and impact acceleration due to collision. Experiments are also carried out to examine the analytical results, and good agreement between the analytical and the experimental results is obtained.

View full abstract

Free vibrations of a rectangular plate which is hinged along two opposite edges and has viscoelastic stiffeners transverse to the hinged edges are discussed in this paper. The governing equations are derived considering the inplane deformation of the plate and the effects of shear deformation, rotatory inertia and torsional rigidity of stiffeners. Conventional trigonometric series are used to reduce the governing equations to sets of matrix equations. Transfer matrix techniques are used to determine the frequency equation. The complex roots of the frequency equation are obtained and changes in natural frequencies, logarithmic decrements and normal modes are investigated. It is shown that an optimum number of stiffeners exists for which the value of logarithmic decrement becomes maximum, Solutions are compared with experimental observations of logarithmic decrements and the agreement is found to be reasonable.

View full abstract

An equation of motion for general rotating shaft systems with many disks and bearings is formulated by the finite element method and a consistent gyroscopic matrix is derived. In this report, undamped eigenvalue and critical speed analyses for a rotor supported by isotropic bearings are discussed. An algorithm for an eigenproblem solution is presented. Using the algorithm developed, approximate eigenvalues and eigen vectors converge iteratively to the exact values, based on the asymptotic method. The procedure leads to a perfect orthogonality of eigen vectors. Numerical results are demonstrated for typical rotor-bearing systems.

View full abstract

Vibrations of cylinder blocks of diesel engines with 4 and 6 bores are analyzed by the reduced impedance method proposed by the authors. A cylinder block is divided into some component structures, and the mechanical impedances of each component are calculated by the finite element method. These impedances are reduced and composed to get a unified equation of motion. This equation is solved to get the natural modes and the natural frequencies of the total structure. On the other hand, the natural frequencies are detected experimentally, using the Fourier translations of the measured impact waves of the force and the acceleration. The natural modes are observed by the holographic technique with an argon gas laser. The calculated results agree well with the experimental ones.

View full abstract

Feasibility studies have shown that a flywheel powered vehicle will increase fuel economy under the city driving. So, some designs of drive train system which gas high efficiency and can regenerate braking energy by a flywheel storage unit have been contrived. But, far from it, the most interesting subject to be taken up first is the analysis of the gyroidal effect on vehicle's motions and durability of equipment. The author has analysed theoretically how the gyroidal effect has influence on vehicle's motions under bouncing, braking and cornering. Next, the author has built a prototype flywheel powered vehicle on this analytical basis and check the dynamic behavior predicted by the theory and obtain some preliminaly road performance data.

View full abstract

We investigated a method of balancing a flexible rotating shaft, in which the magnitude of correction masses is constrained. Analytically, we used the least squares method, but we introduced a constraint coefficient into the performance function. The magnitude of correction masses is controlled by the constraint. Experimentally, we applied the above method to one rotor of a 5-span, 10-bearing rotor system and obtained good balancing results.

View full abstract

The asymptotic properties (the unbiasedness as time t→∞ and the consistency) of the measured value are theoretically investigated, introducing a noisy mathematical model of weighing system. The result shows that both the unbiasedness and the consistency are guaranteed in case of the system model of displacement pick-up type, but none of them are guaranteed in case of velocity pick-up type. From the result of computer simulation concerning the transient characteristics (transient behaviors of the mean value and the standard deviation), however, the velocity pick-up type is concluded to be better implementation of"Dynamic Weighing Method-B"than the displacement pick-up type, though its asymptotic properties are inferior to those of the latter. The reasonableness is also confirmed of setting the estimation point at about half the natural period of scale.

View full abstract

This paper deals with a theoretical study of the radial spring characteristics of a ball bearing, paying particular attention to their fluctuation due to the ball revolution. The following are clarified and demonstrated: 1) computation of the running accuracy problem by applying the (modified) Newton-Raphson method, a problem first examined by Perret-Meldau, 2) detailed analysis of radial motion of the inner ring and the fluctuation of the two-dimensional derivative stiffness caused by the ball revolution, shown by numerical examples and 3) effects of some system parameters (e.g., radial clearance, radial load, number and the revolution angle of balls, etc.) generally expressed and the relation to the extreme characteristics, presented in the former paper, which are derived from the theory of a case with an extremely large number of balls.

View full abstract

In order to improve the quality of roll-finished involute helical gears, a calculation program for roll-finished tooth form has been developed, which can deal with tooth form correction of die-wheel and tooth form error of work-gear before roll-finishing. In this calculation, gradual changing of tooth form of work-gear with progress of plastic deformation is taken in consideration. Some comparisons between calculated and measured tooth forms of helical and spur gears roll-finished by die-wheels with and without die-profile correction under some roll-finishing conditions showed fairly good coincidence. This program can therefore help to decide the optimum tooth form of die-wheel and the optimum working condition.

View full abstract

The effects of guide pads on burnishing action and accuracy of machined hole are investigated in deep-hole-drilling with BTA system solid boring tool using a specially designed tool experimentally. The cutting forces are balanced at the small forward regions of guide pads. The burnishing action takes place under a high contact pressure between the bore wall and those rigions. And the over-size mechanism of machined hole by the guide pads is discussed.

View full abstract

This is a study on the multi-project scheduling under restrictions of limited resources (manpower, equipment etc.). In the model proposed here, the duration of each job is regarded as a controllable variable, and a wide variety of job-speeds are given to each job. Then, 1) a 0-1 integer linear programming formulation, into which a coefficient-formulating power function is introduced, is presented to find optimal solution in the strict sense and 2) a heuristic algorithm, into which a normal job set is introduced, is developed in an algorithm which can be put to effective use for a large-scale system.

View full abstract