In order to investigate the threshold condition of micro fatigue crack propagation under mixed mode, tests of cyclic plane bending, cyclic torsion and their combinations are carried out with a stress ratio R=-l. Annealed cracked specimens made of carbon steel are used in the tests. Based on the assumption that a crack begins to propagate when the circumferential principal stress near its tip reaches a critical value, the threshold stress of the crack propagation under the mixed mode can be determined by making use of the test results of cyclic plane bending and cyclic torsion.

View full abstract

Rotating bending fatigue tests were carried out on S50C steel specimens having four drilled holes with random circumferential distances in the critical section to investigate statistically the propagation behavior of surface cracks and their coalescence leading to a failure of the specimen. The important findings in the present study are summarized as follows: It was observed that the first crack coalescence occludes between adjacent holes with the minimum or second minimum distance. There was little difference between the number of cycles to crack coalescence and that to a failure. The distribution of the numbers of cycles to crack coalescence has been theoretically derived on the basis of crack coalescence probability assuming that the first crack coalescence occurs between adjacent two holes of the minimum distance or the second minimum distance. The theoretical distribution thus derived shows good agreement with the experimental result.

View full abstract

For the osteotomy on hip joint diseases caused by abnormality of the shape of bones, an accurate 3-dimensional femoral shape must be recognized before operation. It was been reported by the present authors that spinal shape is sufficiently analyzed by a developed system based on bi-plane photogrammetry. This paper describe an application of the system to the femoral shape analysis. The shaft axis, the neck axis, the head center of femur and the radius of the head are reconstructed 3-dimensionally using the vector anlysis of plane and line, and the least square approximation method. The obtained axes and head are graphically displayed on the screen of a personal computer through the perspective transformation. The shape parameters usually used in clinic, such as the anteversion angle and the neck-shaft angle, are also calculated by the present method. Result obtained by this system is compared with that by photographical measurement of exposed femurs, then the present method is reduced to have higher accuracy than Kai's method currently used.

View full abstract

The authors have proposed a method to measure the residual stress by X-ray on both surface after removing thin layers of equal thicknesses from both surfaces of a plate. In this paper is derived a new method of X-ray after removing inequithickness from both surfaces of a plate. The theories of the formar Single Surface Removal Method and Both Surface Removal Method are derived from the present theory as special cases. The authors applied this method to a quenched and tempered alloy steel plate of SKS 51 and discussed the influence of neglecting the inequality of removal thickness on both surface.

View full abstract

To study the effect of penetration of pressure fluid on the bursting of brittle materials, the experiments were carried out with cast iron hollow cylinders under internal, lateral fluid and external pressures, and the leakage and the bursting pressures were measured quantitatively. Pressure fluid used in this experiment is turbine oil (#32), and the experiments were carried out with cylinders coated on the inner surface or the outer one or not coated. From these experiments it is made clear that the leakage and the bursting pressures of the cylinders are influenced by the stress distribution in the cylinder and the rate of increasing pressure of internal pressure.

View full abstract

This paper deals with the stress concentrations of an elastic strip with two pairs of semicircular notches under tension. The method of solution is based upon an application of Green's functions for body force problems of a strip. That is, a principle of the method of solution is to distribute body forces in the interior of a strip with no notches and to determine the intensities of the body forces distributed so as to satisfy the boundary conditions of the notches. Numerical results are obtained systematically for various kinds of notch sizes. An investigation is made for interference effects of stress concentrations due to two pairs of notches and the influences of radii of notches and distance between two pairs of notches on the stress concentration factors.

View full abstract

The general condition by which the rate of crack growth is controlled is discussed based on the concept of crack energy density ε, and especially the parameters of creep crack growth rate are proposed. That is, it is shown that the average values of ∂^^ε_t/∂t and ∂^^εx_l/∂X, concerning ε which correspond to the increase rate and the gradient of ε at the crack tip respectively are the parameters of creep crack growth rate based on the creep crack model in which a crack extends discontinuously with respect to time. Moreover, the crack growth of a center notched specimen in tensile creep under plane stress condition is simulated by means of finite element method. In the analysis, the condition of constant εc (critical value of ε) is used. The applicabilities of ∂^^ε_t/∂t and ∂^^εx_l/∂X as parameters of creep crack growth are numerically verified through this analysis.

View full abstract

Three-dimensional analyses are carried out by the finite element method for CCT specimens of four thicknesses (B=200, 18, 6, 4mm). In elastic integrals analyses, there is no significant difference in the distribution of J-integrals along the crack front between the straight and the curved crack front specimens. In elasto-plastic analyses, the distributions of J_x and COD along the curved crack front and the J_x versus displacement curves are obtained and compared. The same tendency is shown in the distribution of J_x along the crack front, which is independent of thickness of specimen. But, there is a significant difference between the 4mm thick specimen and the specimens thicker than 4mm. The thickness effects are discussed from the distributions of J_x along the crack front and the plastic zone profiles. Finally, the availability of J_r values is examined, which is evaluated from the load versus displacement curve by using Rice's formula.

View full abstract

In this work, the plastic behaviours of polycrystalline metals in two typical examples of complex loading conditions are computationally analyzed by using a simplified mechanical model of polycrystal in the manner of a crystal plasticity type. The first example is a deformations along a strain path with a corner, and the second one is a so-called mechanical ratcheting deformation. The variations of microscopic stress in some single crystal components in the polycrystal model are discussed in detail, and then the relation between the macroscopic stress and microscopic stress is clarified. Especially, it is an important mechanism related with some macroscopic features of polycrystalline metals under such complex loading conditions observed in the experiments.

View full abstract

Using commercial 70/30 brass, OFHC copper, and ll00-0 aluminum sheet alloys, the work hardening behavior and limit strain in plane-strain condition have been studied by means of the uniaxial and plane-strain tensile tests in relation to the effect of empirical work hardening law equation and yield function. The parameter m of Hill's non-polynomial yield function is confirmed to be the function of strain by applying Wagoner's rigorous method. The best-fit empirical work hardening law equation describing the relation between logarithmic work hardening rate and effective strain is Voce law equation (saturation law) for brass, Swift law equation (modified power law) for copper, and Hollomon law equation (power law) for aluminum. The results indicate that theoretical ε_lc of forming limit in plane-strain condition are almost independent of m-value and theoretical ε_lc based on best-fit work hardening law equation agree well with experimental ε_lc. The conditions which govern the m-value and the factors which influence the forming limit are discussed.

View full abstract

A new approach of describing the time-independent tensile stress-strain curve is presented by means of the concept of the field of resisting force against the movement of dislocations. The cyclic plasticity model proposed in the previous paper based on this concept indicates that a stress-strain relation in the steady state can be expressed by three internal variables associated with such field and microscopic structures. In this paper, a tensile stress-strain relation with different types of yielding can be described by reflecting the microscopic behaviors such as the catastrophic increase of movable dislocation density at the early stage of one of these three variables. This description has provided satisfactory agreement with each experimental curve for carbon steel and brass with a wide range of strains.

View full abstract

This paper presents that an impact contact force can be expressed as a product series of penalty functions and subsidiary contact conditions on a contact area between two bodies. Using the expression of the impact force, all the subsidiary conditions are included in a virtual work principle for two bodies, and penalty function method can be transformed from the method of a least square into that of a virtual work. This paper also describes a virtual work principle using penalty functions for various impact contact states of two bodies. A two-dimensional response for longitudinal elastic impact of two uniform rods is calculated using a finite element method (FEM) based on this principle, in order to examine the validity of the impact force. The mean value of the impact force by the FEM agrees with the value from the one-dimensional theory of propagation of elastic stress wave.

View full abstract

The impulsive response of a fluid-filled finite cylindrical shell with both ends clamped subjected to waterhammer waves is analyzed on the basis of three axisymmetric shell theories and the potential theory for perfect fluid. In the analysis Mirsky-Herrmann's theory considering the influence of the transverse shear deformation, Flugge's theory as an accurate classical theory and a simplified theory most frequently used are employed and it is assumed that the influence of the fluid-shell interaction is negligible. As a result, it is shown that if the accurate bending moments and the shear force are to be obtained, the use of Mirsky-Herrmann's theory is recommended. The numerical inverse Laplace transformation with use of the FFT algorithm contributes to the reduction of CPU time and the increased accuracy of results.

View full abstract

The influence of the poppet angle and the oil temperature on the flow characteristics and the cavitation phenomena of an oil hydraulic poppet valve are studied experimentally using a half cut model. The following results are obtained. A change of the poppet angle has great effect on the pressure distribution and the flow contraction degree in the restriction between the poppet surface and the valve seat. Hence a remarkable effect of it appears on the valve performances especially when the chamfer is on the seat. The effect of the oil temperature is explained generally by a change in viscosity and Reynolds number.

View full abstract

An experimental investigation was carried out on the characteristics of the flow around four circular cylinders arranged in line. concerning the behavior of the shear layers separated from the first cylinder to the downstream ones, the flow patterns were classified according to the longitudinal spacing between the axes of the cylinders and Reynolds numbers. Flow characteristics of these patterns were elucidated. Especially, the Reynolds numbers corresponding to the reattachment of the shear layers onto the second cylinder, were obtained. The flow characteristics around the downstream cylinders changed drastically in this transition region. Thereby, a bistable flow and a hysterisis phenomenon emerged.

View full abstract

In order to determine the flow characteristics of a transient flow in a circular pipe, an accelerated and a decelerated flow are studied, and effects of acceleration upon the formation of a velocity profile, transition and a friction coefficient are determined for a wide range of accelerations. The results of the accelerated flow show that there are two patterns in the formation of a sectional velocity profile and transition, one of which is observed when the acceleration is relatively large compared with the propagation of viscous effects, and the other when the acceleration is comparatively small. A friction coefficient of an accelerated flow tends to be larger than that of a quasi-steady flow in the laminar region, while it has reverse tendency in the turbulent region. For a decelerated flow a turbulent region is maintained almost all over the deceleration periods, and the friction coefficient becomes larger than that of a quasi-steady flow.

View full abstract

Detailed measurements of turbulence quantities in a three-dimensional jet issuing form the cruciform nozzle have been made in order to investigate the similarity of each turbulent profile and the streamwise development of the characteristics scales. Experimental results are summarized as follows: Up to far downstream, the decay rate of turbulent velocity scales is slower than that of mean velocity scale, and the pertinent properties such as three components of turbulent intensity and Reynolds shear stress are similar in the extent of x/d≧100. The Reynolds number based on the velocity scale Uox and the length scale b' is constant, so the present turbulent jet remains steady.

View full abstract

This paper examines the axisymmetric type vortex breakdown in a swirling pipe flow, in order to clarify the growing process of a bubble. The unsteady flow field is investigated based on the inviscid linear theory of the internal wave (inertia wave). An expression of the time-dependent flow field with the internal wave propagation, which satisfies the condition of arbitrary velocity change at the upstream boundary of the circular pipe, is derived. Using this expression, the transition of the flow field, the streamlines and the streaklines calculated by computer are examined. The unsteady flow field, as the result of a sudden azimuthal velocity change at the upstream boundary, clearly shows the birth of a bubble and its growing process into a stationary state. The theoretical results are compared with the experimental growing process of the bubble in an actual swirling pipe flow obtained by the visualization technique. both results agree with each other fairly well concerning the bubble structure during the growing period. The behaviour of the fluid elements which are entrained into the bubble is also examined from the results of calculated pathlines.

View full abstract

The present paper deals with an experimental review of orifice flows in oil hydraulic lines. Pressure differences were measured between two separated points on a pipe wall downstream of an orifice. Unexpectedly an abrupt vanishment of pressure drops and recoveries was observed at a specified rate of flow. Measurements of continuous pressure profiles revealed that the 'vanishment' results from an unknown phenomenon which characterizes orifice flows of low Reynolds numbers: when the Reynolds number falls below a critical value, the pressure recovery region starts to move downstream farther from the orifice as the Reynolds number decreases. The critical Reynolds number takes a smaller value for a greater enlargement ratio of an orifice to a pipe. Unexpectedly it also varies with the line pressure.

View full abstract

Nonlinear waves on a thin elastic plate horizontally separating an upper fluid stream and a lower one are analyzed theoretically. The lower stream is denser. The elastic plate is governed by the equation for large deflections including the in-plane forces due to the longitudinal deformations of the plate. Two fluid streams are assumed to be incompressible and inviscid, but on the elastic plate the balance of the normal component of force is expressed by taking the viscous stress into account. Some progressive waves of finite amplitude are found on the elastic thin plate. The nonlinear elevation of the elastic plate is obtained up to the third order approximation in the wave amplitude. The unsteady amplitude of waves tend to a finite limit after a long time under the effect of the viscosity of the fluid.

View full abstract

Although much progress has already been made in solving problems in step responses of pressure in a linear tapered pipe, new developments are still needed before the two-dimensional wave equation can be solved routinely. This paper describes one such development. A new method of solving a one-dimensional wave equation in linear tapered compound pipes with arbitrary cross-sections has been devised. Results obtained by this theory are compared with experiments in pipes with square cross-sections. And we obtained the following conclusions: (1) Damping coefficients in an arbitrary cross-section are defined and are expressed analytically for some representative cross-sections. (2) The viscous solution of a wave equation obtained by Laplace transform explains well distorted pressure histories in a linear tapered pipe with an arbitrary cross-section. (3) Pressure waves in a square cross-section include higher modes of frequency and damp more rapidly than those in a circular cross-section. but fundamental characteristics as waves are independent of the cross-section of the pipe except for the small aspect ratio.

View full abstract

An optical fiber probe was developed for measurements of velocities and concentrations of solid particles. This probe was applied to an air-solid two-phase flow in a horizontal pipe, and distributions of velocities and concentrations were obtained over the pipe cross section. It was found that the particle concentration increases exponentially in the vertically downward direction. Corresponding to the distributions of concentrations particle velocities have unsymmetrical distributions. The flow structure becomes more unsymmetrical at low air velocities and high loading ratios. Also, the gradient of concentration is well correlated to the ratio of particle to air velocity.

View full abstract

The problem of a steady laminar flow past an arbitrary body is formulated using the boundary element method. First, surface integral equation is derived for the Oseen flow, and then, extended to the Navier-Stokes equations. In the latter case, the repetition of solving the surface integral equation is needed. The nonlinear terms in the Navier-Stokes equations are regarded as the Oseenlets distributed in the flow field, and surface stresses are obtained numerically. As an example, the flow past a two-dimensional flat plate is solved. The numerical results are coincident with the experimental data for the Reynolds numbers smaller than 100. But the proposed method is not efficient for a high Reynolds number flow, because it is based on the Oseen solutions. This technique can be applied to a three dimensional flow past an arbitrary body, and expected to make up the Reynolds number region between the Oseen flow and the boundary layer approximation.

View full abstract

In this report, the factors that are responsible for the accuracy of numerical calculations of two-dimensional low Reynolds number flows around cylindrical bodies immersed in a uniform stream are discussed on the basis of thee Navier-Stokes equations. The finite difference calculations have been carried out for a circular cylinder and it has been confirmed that the difference of the outer boundary condition affects little the accuracy, if the outer boundary is sufficiently far from the body, but if a comparatively near outer boundary is assumed, Oseen flow is recommended as the outer flow on the boundary. The results can be extended to the other geometries of cylinders and Imai's asymptotic solution can be used in place of the Oseen flow. Some aspects of the flow, such as streamlines, length of twin-vortices, angle of flow separation, etc. are also discussed.

View full abstract

In the present study abnormalities in the exchange of substance from fluid to wall of a narrow rectangular stenosed channel have been studied. The transfer of substance such as oxygen is considered, and is numerically estimated by calculating the nonlinear diffusion equation. Furthermore, both the local and the mean Sherwood numbers on the wall downstream of the stenosis are measured by the electrochemical technique, in which a redox system comprises the electrolyte and the test nickel electrode. Measurements are carried out in the Reynolds number range of 100-300 and a Schmidt number of about 1000. consequently, the local Sherwood number is maximum near a reattaching point, and it asymptotically decreases at the wall downstream of this point and steeply decreases at the upstream wall. Above Re=130, the mean Sherwood number on the wall just behind the stenosis is lower than that away from the stenosis.

View full abstract

Detailed theoretical and numerical methods of a mass-weighted pitch-averaged meridional through flow introduced in our solution in the preceding report are described. The solution for a mean flow may consider secondary flow effects by means of conformal mapping of a twisted S1 surface onto an axisymmetric one and then deal with complicated flows such as those through twisted passages with inlet distortion. Basis relations for the assumed S2 relative surface are reduced to a singular integral equation which fully depends on pseudo-analytic function theories, and solved rigorously. Then itemized numerical method for the solution is shown just in the same way as that for S1 relative surfaces in the report. Owing to the existence and uniqueness of a solution of the equation successive iteration is used. A numerical example is presented. The computation procedure is clear and easy to accomplish.

View full abstract

When negentropy is applied to a system under a general situation, it is shown in the preceding report that negentropy should be defined by Kullback-Leibler information, for, if it is defined by the difference of entropies, or by the entropy with the inverse sign, it may become negative under the general situation. Negentropy represents the potential of work, and the present report also shows from this point of view that the negentropy defined by Kullback-Leibler information is more appropriate than that defined by the difference of situation that an isothermal system is given, the maximum work which can be taken out of the system under this situation is proportional to the negentropy defined by Kullback-Leibler information, and if the system is not isothermal, infinitesimal changes of these quantities are proportional.

View full abstract

Heat transfer characteristics of thermal energy storage apparatus having parallel plate fines were investigated through experiments and numerical analysis. Then, the optimum aspect ratio H/D, which gives the maximum nondimensional average heat flux q^^-, was found to exist for a given combination of three parameters: Grashof number Gr, Prandtl number Pr and the nondimensional fin pitch D/w. The optimum values of q^^- were summarized, for various combinations of Gr, Pr and D/w.

View full abstract

The heat flux across a fluid layer flowing between two sinusoidal walls is theoretically investigated. The surfaces of those walls are assumed to be maintained at constant and different temperatures. Specifically attention is given to the convective heat transfer due to a flow induced by the wavy walls. In the case of the sinusoidal walls with small amplitude, the heat flux integrated over one wavelength of the walls is proportional to the square of that amplitude and the proportional constant depends on the wave number, the phase difference, the Reynolds number, the Prandtl number and the Grashof number. When the wave numbers are from 2 to 3, the convective heat transfer becomes maximum. The heat flux in the case with no phase difference is larger than that in the case with a phase difference, except the case that the Reynolds number is large and the wave number is comparatively small.

View full abstract

This experimental study concerns the mixed, forced and natural convection heat transfers in rectangular cross-section cavities heated from bottom surface. Flow visualization, measurements of temperature profiles and heat transfer from heating surface are performed for various free-stream velocities, temperature differences between the free stream and the heating surface and dimensions of the cavity. It is found that flow pattern and the amount of convective heat transfer are influenced by an interaction between forced and natural convections, and there exists a range of Reynolds numbers in which the amount of convective heat transfer decreases with an increasing Reynolds number. this paper correlates the experimental data in terms of Nusselt number, dimension ratio of cavity (depth/width of cavity), Grashof number and Reynolds number with the range of Reynolds numbers.

View full abstract

Measurements of solar radiation in the past five years from 1978 to 1982 at Kitami Institute of Technology were processed and analyzed, and it was then proved that there was a high correlation between the total solar radiation on a horizontal surface and the direct solar radiation on a normal surface to incidence which were both normalized by dividing by extraterrestrial solar radiation. From the frequency distribution of a number of days which was obtained by grading the daily total radiation on a horizontal surface every 2 MJ/m²d of intensity, a limit value of the direct solar radiation on a normal was introduced in order to define a clear day on which it was possible to use solar energy and a cloudy day on which it was impossible to use solar energy. According to the results of data analysis, the limit value of the direct radiation on a normal, H^^-*_ND' on the borderline between such clear and cloudy days as above was about 3 MJ/m²d, and the limit value of the total radiation on a horizontal, H*_HT, which corresponded to H^^-*_ND and the ratio of clear days to total days could be obtained as a monthly average over five years.

View full abstract

For the application of solar energy to heat engineering, the data of solar radiation measured over five years from 1978 to 1982 at Kitami Institute of Technology were processed and analyzed. Based on all the measured data a correlation equation which could be mployed to estimate the normal-direct solar radiation from the horizontal-total solar radiation was introduced. The daily horizontal total radiation was classified into a scale rank by every 2 MJ/m²d of intensity, and total radiation on a tilted surface with any angle of inclination and direction in every scale rank was calculated by means of a precise calculation method using hourly values of the measurements. At the same time, the total solar radiation on a tilted surface, i.e., the inclined-total radiation, was also calculated from the averaged values of the horizontal-total radiation and normal-direct radiation. Subsequently, it was proved that in comparison with the accurate value obtained by the precise calculation method, the inclined-total radiation could be caluclated correctly enough by both the monthly average horizontal total radiation and the monthly average normal-direct radiation with any distribution weight.

View full abstract

Three kinds of component mode synthesis methods are compared with one another in this paper. The first method is "CMS" proposed by the authors. The second is a method using the residual compliance matrix. The third is one without the residual compliance matrix. Comparing the accuracies and the calculational speeds of these methods with one another, their characteristics are studied. In addition, as an example of the application to actual mechanical structures, a crank-shaft is analyzed by CMS and the method with the residual compliance matrix.

View full abstract

This report deals with fundamental study concerning various kinds of dynamic load combination problems. Up to the present, the square root of sum of squares (SRSS) law has been widely utilized for various load combination problems. However, considerable error is produced during the combination based upon SRSS law, particularly when correlation among loads to be combined becomes significant. In this study, a load combination technique introducing an evaluation of the correlation is proposed. It is applied to a seismic response analysis problem of a large-scale liquid storage tank inducing sloshing effect in order to evaluate the base shear coefficient and the overturning moment. It has been made clear that the proposed method is reasonable by comparing with the SRSS-based law and the absolute value summation. And, the merit becomes dominant when combination is carried out not based upon ordinary absolute acceleration response spectrum but based upon relative acceleration response spectrum as input load.

View full abstract

In the previous paper, an algorithm was presented to obtain the periodic solutions and stability of nonlinear multi-degree-of-freedom systems with high speed and high accuracy, based on the harmonic balance method and the infinitesimal stability criterion. A revised algorithm is presented to give only odd order solutions which are composed of odd order harmonics only and so reduce the dimensions of the amplitude vector and Jacobian matrix to about one-half of the previous ones. The Duffing system with hard spring is analysed by this algorithm and the detailed frequency responses are computed for odd order superharmonic resonances (order 3, 5, 7, 9) which are odd order solutions. The results are shown for each resonance region in terms of (a) maximum amplitudes and norms, (b) superharmonic amplitudes, (c) fundamental amplitudes, and (d) fundamental and superharmonic phase angles. Some of these are confirmed by numerical simulation.

View full abstract

Behaviors of a self-exciting system of van der Pol type subjected to a parametric excitation are investigated. In the present paper the parametric excitation is expressed by the product of a nonlinear function of deflection with an asymmetric characteristic and a periodic function of time. A resonance of order 1/2 and the solution in the neighborhood of the resonance are obtained by the averaging method and the effects of nonlinearity are investigated. since a squared nonlinearity merely makes the resonance have a constant component, it is found that a cubic nonlinearity plays a more important part in the occurrence of the resonance.

View full abstract

The forced nonlinear vibration and its stability of a single-degree-of-freedom system with a ball screw type damper which is composed of a ball screw, a flywheel and a flyball governor are discussed theoretically. The results are compared with a linear solution and the experimental results. The effect of vibration isolation of the damper is also discussed. Numerical examples are given for several ratios of the natural frequencies in the primary system to the governor. It is demonstrated that, when the natural frequency ratio is chosen suitable, the damper is effective in suppressing the resonance of the primary system. The amplitude response curve of the primary system is approximately equal to the linear solution except in the vicinities of the natural frequencies of the primary system and the governor, and is stable except in the vicinity of the natural frequency of the governor.

View full abstract

Vibrations of a driven shaft with asymmetrical stiffness are governed by simultaneous equations with parametric excitation. The equations of motion are analyzed by the asymptotic method, and the characteristics of vibration are explained. When the shaft rotates near the major critical speed, the fluctuation of angular velocity generates an unstable vibration whirling backward in addition to that whirling forward, and it expands the unstable region. When the shaft rotates up to nearly half of the major critical speed, the fluctuation causes an unstable region which separates into three regions. In the mid region the shaft undergoes an unstable vibration which has a complex change of amplitude and frequency; in the other two regions it undergoes unstable vibrations which have amplitude increasing monotonously and constant frequency.

View full abstract

This paper is concerned with theoretical and experimental investigations on the dynamic behavior and the stability of an automatic balancer using centrifugal pendulums. It is found that the unbalance of the rotor can be completely corrected only when the pivots of the pendulums are at the center of the shaft. A violent self-excited vibration can occur when the speed of the shaft is nearly equal to the sum of the critical speed of the rotor and the natural frequency of the pendulum. To verify the theoretical results experiments are carried out.

View full abstract

The effect of a lumped mass and damping on stability of tubular cantilevers conveying fluid is studied theoretically and experimentally. The cantilevers are nonconservative and become subject to flutter-type instability. comparisons are made between the theoretical flutter velocities and the corresponding experimental results. The experimental values were close to the theoretical predictions made by taking a measured damping coefficient into account. It is shown that the effect of damping in combination with a lumped mass is occasionally remarkable on stability of this kind of nonconservative systems.

View full abstract

For the purpose of developing a vibrationless forging hammer, this paper proposes a new method to realize the vibrationless forging hammer. The principle of the vibrationless forge system is based on the active vibration control method proposed in the previous papers. First, this paper presents the principle of the vibrationless forging hammer and discusses the characteristics of the system. Then, analyzing the vibrationless forge system from the viewpoint of feedforward control, this paper derives a state equation of the system. Next, with a view to simulating the system designed by dynamic compensation method, an analog computer with an input signal of an actual impact force is used. and then, the control effect to suppress the impulse response of the system is shown. finally, in order to obtain a perspective to realize the vibrationless forging hammer, an experiment is carried out and the effectiveness of the system is verified.

View full abstract

Porous materials find a wide range of applications as bearing metals because of their self-lubricating action. On the other hand, grease its widely used as a lubricant for porous metal bearings because of its simplicity of application. Unfortunately, since the governing law for flows of grease in porous matrices, that is Equivalent to Darcy's law for Newtonian fluids, has not been established, the hydrodynamic lubrication of porous metal bearings with grease has not been reported. Applying a modified Darcy's law for a Bingham solid postulated by the author to porous journal bearings lubricated with grease, this paper analyzes how the yield stress of a Bingham solid and the permeability coefficient of porous matrix affect the flows of lubricating film and porous metal.

View full abstract

To reveal basic dynamic features of a gearbox for four-wheel-drive automobile, a test gearing unit having a countershaft was manufactured. Gears can be placed at arbitrary positions on the shafts, and fixed with any phase shift between two meshing pairs. Through the vibration measurement of gears, it has become clear that the effect of countershaft on vibration depends upon both spacing and phase difference between the two pairs. And a resonant state with tooth separation is evident. These features agree well with simulated results which take the effect of the shaft into consideration. It has also become clear that the sound radiation is especially related to the bending vibration of the countershaft.

View full abstract

In order to elucidate the effects of lead and sulfur components on surface durability of case-hardened chromium molybdenum steel SCM415 rollers, case-hardened rollers made of commercial SCM415 steel, special SCM415 steels with 0.056% and 0.169% lead and special SCM415 steel with 0.056% sulfur were fatigue-tested under a rolling with sliding contact condition. The experimental results were discussed by the calculated results about amplitude of ratio of shear stress to hardness. It was clarified that at a higher Hertzian stress range the fatigue life was shortened by addition of lead and sulfur to the steel but that at a lower Hertzian stress range the fatigue life was hardly affected by the addition of them.

View full abstract

In order to improve the tooth form accuracy of roll-finished involute helical gears, a roll-finishing method whose process of applied loading decreases during the operation was carried out. It was found that a tooth form accuracy of gears roll-finished under this loading process is considerably improved. another roll-finishing method adjusting a die-wheel approach was investigated, which has a similar loading process to the former case. As a result, a fairly good tooth form was also obtained. A fact was revealed that there is an optimum amount of die-wheel approach independent of an applied loading value.

View full abstract

In order to improve the tooth form of roll-finished involute helical gears, roll-finishing by a die-wheel with longitudinal tooth form correction, i.e. crowning, was investigated. The simulated and the measured results indicate that the use of a crowned die-wheel enables roll-finishing under lower applied loading than the use of a die-wheel without correction, and brings a high tooth form accuracy. Even in case of roll-finishing of work-gears with large face width, a fairly good tooth form over a whole tooth flank can be also obtained by using the crowned die-wheel, if the condition of axial feed of work-gear is selected properly in consideration of the amount of crowning correction of die-wheel.

View full abstract

A calculating method of tooth contact marking of involute helical gears is shown with the following results: 1) When the shape of composite gear error surface of a pair of involute helical gears is mathematically similar and the ratio between amplitude and the transmitting load is kept constant, the pattern of tooth contact marking does not change and the state of tooth loading distribution is similar. 2) When a pattern of tooth contact marking is the same, the ratio of stress increase owing to gear error is almost the same. Some restricting conditions for these two theorems are indicated. Measured and calculated tooth contact markings and tooth fillet stresses of gears which have concave tooth form error and simultaneous longitudinal crowning of tooth trace correction verify the propriety of these two theorems.

View full abstract

Patterns of tooth contact marking of helical gears with pressure angle error, with pressure angle error plus alignment error, and with pressure angle error plus longitudinal crowning tooth trace correction were calculated and compared with measured patters. Measured and calculated tooth fillet stresses of these gears were also compared. The comparison showed good agreement and certified the reliability of the analyzing method for tooth contact marking and gear stress state. for tooth fillet stress, contact stress and flash temperature, the rate of increasing owing to gear error is discussed in relation with tooth bearing: When the amplitude of pressure angle error and longitudinal crowning changes, the stress increasing rate show two different characteristics, which correspond to the difference in the changing fashion of the tooth contact marking pattern. When a gear has concave tooth form error and pressure angle error, the stress increasing rate can be large in spite of a state of good tooth bearing.

View full abstract

To examine the scoring resistance of spur gears with various coatings under dry friction, scoring tests on gear pairs coated with copper, graphite, zinc, molybdenum disulfide and phosphate were carried out under dry friction using a power-circulating gear machine. Further, variation in the bulk-temperature of the gear tooth was measured by a thermocouple embedded in the side of the tooth. From the experimental results a significant effect of copper and graphite on the scoring resistance under dry friction was recognized. The bulk-temperature of the gear tooth dropped after the initial stage of scoring for all gear pairs.

View full abstract

In this paper is described a method to draw a self-loosening-limit diagram based on the theory of mechanism of the self-loosening. The method is presented taking account of the characteristics of impulsive friction on the interfaces of a fastener estimated from the results of an experiment in which the coefficient of impulsive friction on the test pieces is measured. The diagrams obtained by the method agree with the experimental results. And after the examination and comparison of those diagrams, the following are found effective to prevent the loosening: (1) the compressive force on the interfaces still remains, even when the tensile stress waves arising from an external impact come to the interface of fastened pieces and (2) the actual compressive stress on the pressure flank does not exceed the compressive yield point.

View full abstract