The purposes of this study are to create a mathematical model which can simulate vehicle occupant dynamic interaction, and to perform a parametric study of influences on human vertical and lateral behavior. With a cross-sectional model, vehicle-occupant movements are simulated by numerical calculations. The mathematical model has nine degrees of freedom : the vehicle model with three masses has four degrees of freedom, the occupant model with three masses and the hinged beam model has five degrees of freedom. These motions are considered with respect to the vertical, lateral and roll axes. A fairly good agreement is obtained between the results of mathematical simulation and the other data. Therefore, it can be concluded that the mathematical model is valid for practical purposes. With use of this simulation system, the effects of seat characteristics of occupants on human dynamic behavior are investigated. As a result, it is found that seat characteristics exert a significant influence upon both vertical and lateral vibration.

View full abstract

An exact method using a power series expansion is presented for solving free vibration of rotating thin circular cylindrical shells having two bonded elastic layers. Equations of motion, which are differential equations with variable coefficients, are solved exactly by using a power series expansion. Effects of various parameters such as material properties of the shell layers, rotating angular velocity upon natural frequencies and mode shapes are investigated.

View full abstract

This paper describes research on periodic shock pulses occurring in a roller chain driving system, which is used as the bed-sliding mechanism of X-ray diagnostic tables. The shock pulses, which had the periodicity of a mesh frequency between the chain and the sprocket, were observed only when the bed was moved downward with a loaded mass in a certain weight range. This roller chain system consists basically of a driving motor, a reduction gear, and two-stage chains and sprockets. Using an experimental device modeling the roller chain driving system, the shock-pulse phenomenon was investigated by changing the weight of the mass, the tension of the chain and the number of sprocket teeth. It was found that the phenomenon occurred due to periodic hits between the gear teeth at the reduction gear, which were caused by torque oscillation between the roller chain and the sprockets of the second stage. Theoretical analysis on torque oscillation was performed, which led to the elimination of the periodic shock pulses from the roller chain driving system.

View full abstract

In this paper, we propose a new method of detecting regenerative chatter of turning workpiece by using spectrum of vibration of the workpiece. Characteristics of vibrations of the workpiece during cutting was investigated first, and then algorithm of detection was thought out from the investigated results. In this algorithm, it was assumed that regenerative chatter had occurred when the following conditions were satisfied in progress of cutting. (1) Frequency at the peak of spectrum obtained every revolution was almost constant while the workpiece made three revolutions.(2) The amplitude of vibration in thrust force direction was larger than surface undulation of the workiece when the condition (1) was satisfied. Validity of this method was found by several cutting tests on which cutting speed and width of cut were varied in orthogonal cutting. Furthermore, it was shown that regenerative chatter of turning workpiece could be detected also by using spectrum of dynamic cutting force or acceleration of cutting tool.

View full abstract

A revised report and detailed summary of the previous papers are presented, in which exact solutions of the free oscillation in a system with a Duffing spring and a single degree of freedom are dealt with. It is well known that a theoretically exact solution to the homogeneous Duffing equation family (hard, soft and snap-through springs) is solved in terms of the Jacobian elliptic function family (sn u, cn u and dn u). However, their precise numeration is rather difficult using the constant-digit (regular) computer. An excellent algorithm and program are essential for avoiding such unfortunate cases in which almost all digits of the computed results are of poor accuracy for some complicated problems. A comprehensive discussion for relevant algorithms, and the listing of a final program for the hard spring system, are shown. Numerical results of the dynamics of a free oscillator with the hard spring are also discussed and demonstrated.

View full abstract

This paper describes the vibration characteristics of turbine blades with compound periodic structures. The structure includes both grouped and one-ringed blade structures in a turbine wheel to form a circumferential periodicity with a packet as a unit. An effective transfer matrix method for calculating the natural frequencies and modes of this type of blade structure is presented by utilizing the feature that the total transfer matrix forms a cyclic matrix. Natural mode analysis reveals that a natural mode is composed of several nodal diameter modes. Consequently, the resonance condition when the blade structure is excited by engine-order excitation is derived in a general form. Natural frequencies and modes of a simple blade model are measured and compared with calculated ones. As a result, the calculated results are found to agree fairly well with measured ones.

View full abstract

An engine cylinder block usually has complex mode shapes. When designing an engine, engineers try to achieve what they think is the best on the basis of their judgement. However they must consider a great variety of modes. In this research, a very simplified 4-cylinder block model has been developed. (1) to classify basic modes both qualitatively and quantitatively, (2) to predict the change in natural frequencies in neighborhood of the target with the optimization method, and (3) to optimize the natural frequency of this model using the techniques of the pseudo-inverse method. It has been found that the similarity between the vectors of natural frequency sensitivity has significant effects on mode-to-mode variations in natural frequency.

View full abstract

The computer-simulated results of many interesting phenomena occur in a magnetically levitated dynamical system with a single degree of freedom have been previously reported by the authors. For the experimental study, an apparatus is constructed, composed of an arm swinging in the vertical plane and installed on a vertical shaker. The arm is levitated at one end by means of a repulsing pair of permanent magnets and the other end is supported by a bearing. When the shaker is harmonically excited, the arm exhibits ordinary harmonic vibration, many kinds of subharmonic vibrations, aperiodic motions and bifurcations. Some aperiodic motions are identified as chaos by means of the Poincare map and FFT analysis. The results satisfactorily agree with the simulated results, thus proving that the computer-simulated results are realistic and that a new dynamic model in which chaotic motion occurs has been discovered.

View full abstract

To characterize chaotic oscillation arising in nonlinear systems, a method of statistical mechanics is applied. The chaotic oscillation is characterized in terms of the dynamic structure functions : the fluctuation spectrum, the q-weighted average and the q-weighted variance. In the previous paper, in order to obtain the local expansion rates that lead to the dynamic structure functions, the variational equations of motion were used. However the expansion rates cannot be obtained from the variational equations if the equations of motion are not known. Therefore, in the present paper, under the assumption that the equations of motion are not known, it is shown that the method of estimating matrices based on Poincare maps of ocsillation is applicable. As examples, the parametrically forced pendulum system and the Duffing system are taken up. The results of the method of estimating matrices are similar to those of the variational method.

View full abstract

Generally, it is difficult to carry out seismic tests of tall pillar-type structures using a vibration testing table, because the table must resist a huge rotating moment, on the structures are too big to mount on the table. In this paper, we discuss an equivalent test using a top excitation method. At resonant or nearly resonant conditions, it is easy to realize equivalent conditions between top and base excitations. At off-resonant conditions, only one among displacement, moment, or shear force can be equivalent, but other components cannot. From the analysis of transient response using seismic wave, impulse wave and sinusoidal wave of three periods at resonant frequency, it is found that nearly equivalent conditions can be realized, because normal modes are excited and are dominant in the response. The amplitude magnification factor at the top of pillar-type structures is usually expressed as m/2ζ, where m≥1. For a uniform beam, m is 1.56. The value of m changes according to the profiles of the structures, such as diameter or cross-sectional width and length at the top and bottom of the structure.

View full abstract

This paper describes the vibration control of a rotating disk to which three kinds of dynamic absorbers are attached for the case where the rotating disk is subjected to an axial external force fixed in space at a point on the disk. The effect of the rotation of disk and the coupling among modes are taken into account in the theoretical analysis. The vibration reduction capabilities of three kinds of dynamic absorbers are compared among each other. From the numerical computational results, it was made clear that the dynamic absorber of ring type with a distributed support spring between disk and ring is the most effective one for the vibration control of a rotating continuous body. The analytical results for this type of dynamic absorber were qualitatively verified through experiments.

View full abstract

In the previous paper, the effectiveness of an active electric torque control method in reducing vehicle idling vibration reduction was proposed. In this paper, the influence of the control torque waveform on this control is analyzed theoretically and experimentally with the goal of improving control considering limitations on the torgue control devices. Torque harmonic analysis shows that the control torque waveform can be used to maximize the amplitude of the firing frequency component which cancels engine torque variation and to minimize the average torque which increases the engine load and torque variation. Sine and rectangular waves are investigated as possible control torque waveforms. When the ratio of the device's torque generation and absorption is small, the influecne of the load torque increase due to control is significant, and the sine wave produces better results. As the ratio becomes larger, the load torque due to control decreases and the amplitude of the firing frequency component becomes more significant. In this case, the rectangular wave achieves beteer control.

View full abstract

Static and dynamic behavior of a rotational motion system composed of a primemover and its load depends upon the relationship between torque-speed characteristics of both components. With respect to the behavior, it is possible to realize the optional characteristic by feedback control of the torque and speed signals of the system with an appropriate ratio of the signals. Recently, almost all the equipment of feedback control of a rotational system is composed by a digital compuler, and there are several cases in which the period of control is not adequately small compared with the time constant of the relevant rotational system. In this paper, we discuss those characteristics including control period as well as signal ratio. The designing policy for such a feedback system was made clear as a result.

View full abstract

This paper presents numerical simulations for balancing a rigid rotor. The influence coefficient and least squares balancing method are compared assuming that the influence coefficients used to calculate correction masses contain errors. The errors in the influence coefficients occur in (1) amplitude, (2) phase angle and (3) the combination of the two. In the three cases, a number of numerical simulations are performed and comparison of balancing conditions using two balancing method is discussed. The results show that the least squares method produces better balancing conditions than the influence coefficient method.

View full abstract

In this paper, we propose a new parameter identification method for motion systems with unknown Coulomb friction. The proposed method utilizes a disturbance observer to identify parameters such as the momentum of inertia, the viscous coefficient, Coulomb friction and the constant disturbance torque. Estimated disturbance torque is composed of four torque components ; inertia variation torque, viscous torque, Coulomb friction torque and constant disturbance torque. Parameters are identified by deconstructing the estimated disturbance torque into the four torque components mentioned above. To reduce the amount of calculation required for parameter identification, orthogonality relationships among the estimated disturbance torque components mentioned above are used. The experimental results verify the exact parameter identification.

View full abstract

The rubber surface of a smoother roll in a paper-making machine is deformed into a certain convex polygon at a high nipforce and a an operating speed at which the rotating rolls are in contact with each other. This paper describes the theoretical analysis and the numerical computational results taking the viscoelastical deformation of the rubber surfaces of a pair of rolls into account, regarding this vibration phenomenon as instability due to the time lag. The authors found from the numerical calculation that (1) the influence of the rotating speed of rolls and the rigidity of the roll supports on the stability is the same as that for the case treated in the previous report, in which one of two rolls does not deform ; (2) this instability phenomenon tends to occur as the hardness of the opposite roll increases ; and (3) the unstable regions are classified into two types according to differences between roll diameters.

View full abstract

One of the authors proposed a numerical approach for analyzing bending deformation and vertical displacement of a ski, vertical deformation of snow, and reacting force acting on the ski from the snow when vertical force and bending moment act on the shoe center of the ski lying on the snow. In the present study, the justification of the approach has been discussed by comparing numerical results with experimental ones. Since it is very difficult to measure the vertical displacement of the ski lying on the snow, the displacement of the ski lying on a urethan sheet was measured when only the vertical force acted on the shoe center of the ski or when vertical force and bending moment acted synchronously on the shoe center of the ski. Furthermore, it is concluded that the correction of camber height is necessary for the numerical analysis of dynamic bending deformation of the ski lying on soft snow.

View full abstract

Motion of a powder bed is analyzed by assuming a simplified model to study the sifting mechanism of a gyratory sifter. Analysis shows that the relative motion of the bottom of the power bed with respect to the sieve diminishes as the bed depth increases. The sifting rate, or the particle flow through a sieve, is treated in a manner analogous to that for the fluid flow through an orifice. The proposed model shows general agreement with the experiment though the experimental results are more dependent on revolving speed than the analytical ones.

View full abstract

An ultra small actuator for the moving magnetic head in the DTF (Dynamic Track Following) system in VCRs has been developed. This actuator is the moving-coil type, utilizing cylindrical permanent magnet. The DTF system consists of an actuator and a servo control system. In order to achieve DTF system with high performance, it is necessary to improve the mechanical vibration characteristics of the actuator. This paper deals with the design method of the active vibration control loop using the velocities and displacements of a moving magnetic head. These physical quantities are measured by means of a moving sensor coil installed near the driving coil. In this construction, the alternating current caused by mutual inductance effect led to the sensor coil plays an important role in the active vibration control loop, and it is shown experimentally that a velocity and displacement feedback control loop which accounts for these influences can reduce mechanical vibration.

View full abstract

Even the potential method, which seems to be the most effective method for automatic planning of the collision-free manipulator path among obstacles, uses transformation of real space coodinates of the obstacle to joint angle space, so the analyses become difficult or impossible when the manipulator degrees of freedom become large. In our previous study, we decreased this difficulty through the use of fuzzy reasoning which also enabled the consideration of much useful different dimensional information, for example, the deviation from the target, the proximity to the obstacles and the rate of change of these quantities. However, an effective membership function or rule for good efficient planning is as yet unestablished. This is a general problem in fuzzy reasoning. Furthermore, the size of the lattice gap has a large effect on search efficiency, which is poor when the target point and obstacles are far from the manipulator under small and uniform lattice gaps. This paper proposes a plan to improve the above two points. We applied a Boltzmann machine of neural network to optimize the membership function, and the size of lattice gaps was made flexible according to the magnitude of deviation and proximity using fuzzy reasoning. As a result, planning in all ranges of manipulator action was achieved, and the search efficiency was better than in the previous study.

View full abstract

In this paper, we propose a new design method for a learning control system. The objectives of this paper are to enhance the tracking performance of learning controllers and to suppress disturbances acting on a plant. The structure of the learning controller is based on the parametrization of all stabilizing controllers, which is called Youla parametrization. The design procedure is divided into two steps. First, we design a basic controller using the LQG (linear quadratic Gaussian) method and consider the parametrization based on the LQG controller. Then, we adjust the free parameter to obtain good tracking performance by the learning algorithm. A numerical example is given to illustrate the effectiveness of the proposed method.

View full abstract

In extreme environments, e. g., in space or in a vacuum pan, there are increasing numbers of cases where it is necessary to move and manipulate objects quickly with a large moment of inertia. Hence, it would be desirable to develop more powerful and lightweight actuators. At present, electric motors are used for space manipulators ; however, their capability is not sufficient in view of output torque-inertia ratio. An actuator proposed and developed here for hydraulic control in space is named a flexible hydraulic actuator (FHA) . This actuator consists of an oil-proof rubber core tube and tenacious string cords (aramid fiber) braided to cover it. During an increase in inner pressure, the FHA contracts in an axial direction about 30% of the full effective length, and its output force to axial direction is 1500 N max (3 MPa of inner pressure) whene the weight is 0.3kgf. The main features of the FHA are that there is no need for lubrication and that it is susceptible to leakage. The precise mathematical model of the FHA in static and dynamic behavior is derived and compared with experiments. It is also ascertained that the FHA responds sufficiently quickly. In addition, the dynamics of a single-armed manipulator model composed by the two FHAs are investigated experimentally. An angle position control of the system, including minor control of the internal pressure defference between the two FHAs, is discussed and is well-established. Moreover, we try to eliminate the effects of the system parameter change and disturbance employing the disturbance estimation observer. The results with the robust control are in good agreement with the response of the desired system model.

View full abstract

This paper describes robot motion planning in a very dynamic environment, that is, a trajectory generation to attain a working purpose with redundancy freedom to use under the restriction of time and space. In particular, we focus on "juggling" as the typical example of such a task. We will describe a parametric representation of the robot motion which makes it easy to generate the appropriate trajectory of the end-effector for the "juggling" task. Also, we will describe a hierarchical learning control strategy which is effective in generating a desired motion pattern for juggling which has not been previously taught to the robot. For an actual robtot juggling several balls, we predict the location and time at which a ball is caught using sensory feedback signals and modify the end-tip trajectory. Experimental results obtained using a 2-d. o. f. direct-drive manipulator are also presented.

View full abstract

Neural networks can be used as controllers in close-loop systems. In the design stage, the network must learn the desired features from a teacher. The backpropagation algorithm is a key point in learning for feedforward multilayer neural networks. In the backpropagation algorithm, teachers are required for the neural networks to learn desirable characteristics of the input-output mapping. In this report, a method for learning of networks has been proposed which does not require any teacher : in other words, the neural networks learn autonomously in the trials. A cart-pole balancing example is used to illustrate the effectiveness of the proposed method.

View full abstract

The object of this study is to develop an "Active Human Interface (AHI)" that realizes communication of emotion between machines (computer and/or robot) and human beings. In establishing the new paradigm for the realization of AHI, as a first step, we have investigated the method of machine recognition of human emotion. This paper investigates the method of measuring the strength of six basic emotions rom facial expressions by means of a neural network. From continuous videotape recordings of subject's facial images, we select a weak facial expression for each of the 6 basic emotions from 19 subjects. We obtain the strength for each of these weak facial expressions by interviewing these clients under the condition that the strength of the basic facial expressions used above is equal to l and normal facial expression is equal to 0. The results are used for training the neural network for each of the basic facial expressions. The trained neural network is tested with facial image data other than those used in training, and the neural network method is found to be effective in measuring the strength of facial expressions since the method shows a rather high agreement with the recognition results of human beings.

View full abstract

Since the facial expressions of human emotions are usually not distinct, the development of a technique for machine recognition of mixed facial expressions is very crucial for the true realization of AHI (active human interface) between machine and human beings. This paper deals with a neural network method for machine recognition by decomposing mixed facial expressions into 2 or 3 components of 6 basic facial expressions. We obtain the images of mixed facial expressions from video tape which recorded continuous change in facial expressions, and using the information of the (x, y) coordinates of facial characteristic points (FCP) for 19 subjects, the neural network is trained. The recognition test is conducted by inputting facial information not used in training the neural network to the trained neural network. The neural network method is found to give a rather high agreement rate of about 70% compared with recognition results obtained by human beings.

View full abstract

This paper presents an active suspension design for vehicles using the concept of neural networks. The dynamic model is described by a system with six degrees of freedom including nonlinearities of the spring and damping forces. The application of a neural network to nonlinear systems affords some abilities to process and to learn the information in parallel, and to recognize some optional patterns. In the proposed design, the system neural network and the neural network controller are used, and the active control forces are determined by learning the vertical velocities of the vehicle body and by minimizing the output of the system neural network. The effectiveness of the proposed method is confirmed by the improved control performance, and by comparison with the method of the passive control.

View full abstract

A non-iterative method for the identification of the heat source in a solid body by a neural network approach is proposed. In this method, the location of the heat source is identified by the neural network through pattern matching with the surface temperature distribution. Then the intensity of the heat source is determined by a boundary integral method utilizing the information of heat-source location predetermined by the neural network and introducing an appropriate harmonic function. Without measuring the surface heat flux, which was required in the previous boundary integral method, this combined approach enables us to estimate both the location and the intensity of heat source more rapidly than by ordinary iterative optimization methods. The simulated results show that the proposed method is effective and has the potential to deal with noisy measurement data. Moreover, it is shown that the robustness of the method is improved by training the neural network using noisy learning data.

View full abstract

Howling occurs because the noise control signal from a secondary source is detected. A 2-Microphone system utilizing a time delay corresponding to the distance between the two microphones is often used as a howling canceller in long ducts where unidirectional acoustic wave is the main component. In this system, the delayed signal fed into a microphone is subtracted from another microphone signal. In short ducts, many reflected waves exist due to the difference in acoustical impedance at the ends of the duct. The authors show that in such cases it is necessary for the reflected waves opposite the opening end to be absorbed. For this purpose, the authors propose a system which they call the 2-Microphone delay system. In this paper, the noise reduction effect radiated from the noise speaker of the duct is examined. A noise control digital filter characteristic of the 2-Microphone delay system is examined theoretically and experimentally from the viewpoint of the digital filter causality. The noise control digital filter characteristic is compared with the normal 1-Microphone system without a howling conceller.

View full abstract

This paper considers the problem of sensing structural vibration to provide an error signal to adaptive feedforward control systems implemented to globally attenuate both structural and acoustic disturbances. It is shown that designing a control system which simply aims to minimize structural vibration is not necessarily the best approach, particularly in structural/acoustic problems. Rather, by considering the governing equation of the global error criterion of interest, it is possible to derive orthogonal groups of structural modes which contribute to the error criterion as a set. It is these orthogonal groupings which should be sensed as an error signal by an adaptive feedforward control system (and minimized in amplitude), a task which is easily accomplished using shaped piezoelectric polymer film sensors. Specific examples relating to the minimization of radiated acoustic power of a simply supported plate are given, showing that the use of a vibration-based error signal is a practical alternative to the use of an acoustic signal.

View full abstract

This report describes the design and the performance of a newly developed interferometer system. The interferometer enables us to measure the motion of a diffuse surface. Velocity ranging from l0^-3 m/s to 10⁰ m/s can be detected by mixing Doppler-shifted light reflected from the specimen with original (reference) light. The direction of motion is determined using two 90°out-of-phase reference beams. Uncertainty of the measurement is estimated to be ±5nm in displacement and ±5×10^-4 m/s in velocity at most. Consequently, this interferometer can be used for absolute measurement of the transient motion induced by low-velocity impact, and is applicable to determination of the input of displacement-related sensors such as accelerometers.

View full abstract

Tactile sensors are normally covered with an elastic cover to protect the sensor from being destroyed by shock or chemical contamination. However the spatial resolution of the sensor with an elastic cover decreases, because the cover exhibits low-pass spatial filtering characteristics. This effect is serious when fabricating a high-spatial-resolution sensor which spatial resolution is under 1mm, even if the cover thickness 200μm. In this paper, this low-pass spatial filtering characteristic is analyzed and filter gain for various kind of elastic cover materials is calculated using the finite element method. Elementary experimental results are shown.

View full abstract

Recently, much effort has been directed towards developing a joint time and frequency representation for nonstationary signal analysis. Above all, we have been interested in Wigner-ville distribution (WVD) and Wavelet transform (WT) because time-frequency analyses can be obtained more exactly with those methods. The purpose of this study is to provide a subjective time-frequency analysis of an acoustic signal, making use of a level-weighted WVD (LWD) and WT. It is made clear that the LWD is most suitable for subjective assessment of acoustic signals.

View full abstract

An increasing number of automotive direct-current motors, which emit high-frequency noise (20∼10kHz), are being incorporated into automobiles. This paper presents an experimental analyis of factors effecting the noise and how to control the factors to reduce the noise. An experimental device is contrived to drive the detached motor set by rare-earch magnetic coupling and to investigate the factors separately. Magnetization of the permanent magnet and commutation of the commutator are causes of the noise. Increase of clearance between the brush and the brush holder, surface roughness of the commutator, and brush pressure force, and decrease of the brush contact area on the commutator increase the motor noise. A countermeasure which reduces the noise by 5∼10dB(A)in 4∼8kHz is presented, wherein the overall noise reduction of 5dB(A) is achieved.

View full abstract

Fuzzy programing is applied to determining optimum shapes of beams subjected to a moving load with constant velocity. The multi-objective optimization has two objective functions and a constraint. This problem is transformed to the unconstrained optimization problem by an augmented objective function. The membership functions of these objective functions are defined. Then, fuzzy mathematical programing is carried out. Fuzzy programing with three kinds of membership functions is solved interactively. The Lagrange multiplier method was used for unconstrained optimization, and dynamic programing was applied for the minimization of the augmented objective function.

View full abstract

Various factors affecting the occurrence of numerous shallow radial cracks observed frequently on the rubbing surfaces of cemented carbides composing mechanical face seals are discussed based on experimental results. The experiments were conducted using internal mechanical face seals made from a pair of three kinds of cobalt-based cemented carbide rings. Residual stress in the ring surfaces before and after the experiments was also measured by the X-ray diffraction method. It has been found that the frequency of cracking is markedly influenced by the circumferential tensile residual stress, the value of which depends on the cobalt content in the cemented carbide.

View full abstract

In general, when a slider is about to be moved, friction is always at the maximum. This phenomenon has made it difficult to locate a slider at a definite position with submicron accuracy. The authors attempted to challenge this widely accepted concept in this paper. The basic idea is to strike a guideway from the side with a small steel ball, so that stress waves are generated in the body of the guideway. The stress waves induce slight movement of the asperities on the surface of the guideway. The movement will lead to the breakage of metal-to-metal cohesion and will cause considerable decrease in friction. The idea was tested using a steel block with a hardened surface for a guideway and another smaller block for a slider. The results obtained are as follows. (1)Even a small ball can induce a considerable change in the relative slding speed of metal-to-metal contact regions between the guideway and the slider. The relative movement is strongly dependent on the kinetic energy of the ball. (2) The above movement causes a great reduction in the friction coefficient. When a ball of 25.4mm diameter is dropped from a height of 80mm, the coefficient of friction drops from 0.37 to 0.05 when the slider speed is 1 × 10^-2mm/sec. (3) Though the low friction level is sustained for only 0.02sec, this period is long enough for a slider to start moving. (4) The lower the slider speed, the greater the reduction in friction.

View full abstract

Static load-deflection (P-δ) and tangential surface stress-deflection (σ_t-δ) relationships of coned disc springs having three spring characteristics (h/t=0.7, 1.4, 2.0, h : free height, t : thickness) were elastically and elastoplastically investigated using the finite-element method (FEM). The results of P-δ and σ_t-δ relationships determined by elastic FEM agreed with Curti-Orland-Podda's equations better than with Almen-Laszlo's equations. After setting δ=h, higher tensile residual stresses were produced in the vicinity of the inner surface periphery of springs.

View full abstract

This report describes an analysis of the jounal bearing under dynamic load taking into account bearing deformation and journal tilt in comparison with measurement of a marine diesel engine. The bearing characteristic is calculated using the finite difference method on 2-D Reynolds equation. The bearing load and journal tilt are calculated using the load of all cylinders, and the bearing deformation is derived considering not only the bearing housing but also the construction of engine body. The stiffness of crankshaft and engine body is calculated using the finite-element method. Generally good agreement is obtained for the bearing deformation, the journal tilt and minimum film thickness, between predicted and measured results.

View full abstract

The run-out characteristics of a rotor with out-of-roundness error in its cross section, supported by an externally pressurized gas journal bearing, are investigated theoretically under the influence of some related factors, such as displacement of rotor due to an external load, form deviation (i. e., out-of-roundness error) of bearing bore, and circumferential pitch error in arrangement of supply holes. The features and magnitudes of the run-out of the rotor are discussed in relation to the number of lobes of form deviation of the rotor cross section and the number of supply holes, as well as other factors of machining errors.

View full abstract

A method for the detection of the errors in hypoid gear cutting is proposed. The coordinates of the points on the gear tooth surface are measured using a coordinate measuring machine and a geometrical gear tooth surface is estimated by the method of least squares so that the surface can fit the data of coordinate measurements. This geometrical surface is implicitly defined from machine settings on a Gleason hypoid generator and the surface usually differs from the designed surface because of machine setting errors. The purpose of this method is to detect the machine setting errors. The coordinates of the points on the gear tooth surface were measured and the errors were detected using this method. Furthermore, the detected errors were compensated and corrective cutting was carried out. As a result, a gear tooth surface close to the designed one could be obtained. The validity was confirmed by observing the contact pattern for a pair of gears.

View full abstract

This paper presents a study on the bending fatigue strength of thin-rimmed spur gears under various loading conditions. The bending fatigue test of a thin-rimmed spur gear with a fan-shaped hole was carried out under various loading conditions, and the test results were examined on the basis of the root stress computed by the finite element method (FEM). A reverse bending fatigue test of the thin-rimmed spur gear was carried out, and the relation between the rim thickness and the fatigue damage on the tensile and compressive sides of the gear tooth was also examined. On the basis of these results, the characteristics of the bending fatigue strength of the thin-rimmed spur gear were determined.

View full abstract

This paper deals with the stress of a cup-type flexible spline in strain wave gearing (commonly known as harmonic drive gearing) with an oval wave generator. Nominal stresses at four positions on the tooth and at three positions on the cylinder were measured using three-element rectangular rosette strain gages. Then the distribution states of circumferential stress, axial stress and shear stress on each position of the flexible spline were analyzed, giving the following results. The amplitude of the varying stress on the center of the tooth is the largest in the flexible spline. The amplitude of varying stress is also large, and the maximal and minimal stresses are nearly the same as the maximum and minimum ones respectively at the open-side end of the tooth part.

View full abstract

This paper presents a study on the vibration and noise characteristics of austempered ductile iron (ADI) gears. Running tests for ADI, SCM 415 case-hardened, and SCM 435 normalized gears were carried out under different running conditions by means of a power-absorbing-type gear testing machine, and the circumferential, radial and axial vibration accelerations of the gears, the gear noise, and the root stress were measured. The vibration and noise of the ADI gear were found to be much lower than those of the SCM435 normalized gear due to the larger damping ratio of the ADI. Furthermore, the effects of the finishing method of a gear tooth and the isothermal transformation temperature on the vibration and noise characteristics of ADI gears were determined to a considerable extent.

View full abstract

An investigation of sound generated from a two-step reduction gearbox was carried out in order to obtain basic information for the reduction of sound. Test gears having an accuracy of Class JIS 1 were prepared, and the gearbox was driven with and without load. The experiments were carried out by the following three processes. In the first, the vibration transmission from the bearing housing to the top plane of the gearbox was examined. Second, the vibration and sound at nine positions on the top plane were measured simultaneously, and the radiation ratio at the positions were determined. Third, the distance characteristics of the sound at each position on the top plane were examined. From the experimental results and considerations, the vibration transmission and sound radiation of a gearbox were clarified, and some basic information for reduction of sound generated from a gearbox was obtained.

View full abstract

The mechanism of planetary gears is so complicated that an unbalanced force due to various errors occurs on the planetary gears. However, the causes of the unbalanced force of solar-type planetary gears have not yet been adequately clarified. Therefore, in the present study, using solar-type planetary gears for a bicycle, in order to elucidate the factors of unbalanced force, and to improve the performance of planetary gears for a bicycle, the bending vibrating displacement of the sun gear shaft was measured. Simultaneously, a simple measuring method of bending vibrating displacement at the end of the sun gear shaft from outside the planetary gears was developed. Furthermore, in order to investigate the effects of various machining errors and the direction of runout error of the planet gear and those of loaded torque and change in speed of the drive shaft, the bending vibrating displacement at the end of the sun gear shaft for two types of planetary gear sets was measured. The above investigations provided several useful results.

View full abstract

This paper addresses a modeling method of quantitative relationships between grinding parameters and grinding results. With consideration of reasons why conventional empirical formulae are not accurate enough for practical use, a new model named the "adaptive causality model" is developed, which has adaptivity to time-varying conditions and robustness for unexpected change of conditions. As a case study, the developed model was applied to the relationship between certain dressing and grinding parameters and surface roughness of a ground workpiece, and an experimental analysis was carried out. In the experiment, wear of the edge of a single-point diamond dresser was considered as a time-varying condition. Moreover, workpieces of different material were often ground to simulate unexpected change of conditions. The experimental result proves that the "adaptive causality model" determines the quantitative relationship with sufficient accuracy in spite of change of the dresser edge and existence of some deviating data.

View full abstract

In the cutting process, chatter vibration markedly restricts accuracy and efficiency. Establishment of an objective and practical method which can estimate the chatter commencement is necessary for automatic manufacturing systems and dynamic perfomance tests of machine tools. In this study, therefore, the estimation of chatter vibration is investigated for the turning process with statistical methods, and the generating mechanism and behavior of transient vibration on chatter commencement are clarified experimentally. The results obtained from this study are as follows : (1)The mechanism of the mingling of regenerative and parametric vibration is clarified qualitatively. (2) The commencing point of chatter vibration can be estimated practically using the auto power spectrum density of the workpiece amplitude before the occurrence of marked vibration.

View full abstract

In this study, error in the surface profile resulting from free-form surface machining using the side surface of a cylindrical tool is considered. The error is induced as a result of cutting marks on the surface of the rotating workpiece when the tool is fed continuously in the z direction. A calculation method for computer simulation is proposed for generating the surface profile from a tool path, taking account of the feed per revolution. The basic feature of the profile error is discussed using a circular cylindrical workpiece ; (a) a sectional shape of the workpiece and (b) relationships between machining conditions and the error in the surface profile are given. Numerical examples are given first for a uniform cylinder, and then a golf club represented by a physical model as an example of free-form surfaces is used.

View full abstract

Effects of the clamping conditions of disc-type tools, such as circular saws for wood cutting, on bending deformation and damping characteristics are examined in a model experiment. The clamped part at the center is regarded as a thin disc on an elastic foundation. The differential equation of bending deformation is solved, considering the elastic deformation effect at the clamped part, and the circumferential deflection in the thickness direction is clarified as a function of disc size and clamping conditions. Combining this result with the measured value of disc circumferential deflection, the experimental equation for estimating contact rigidity at the clamped part is obtained, and the parameters which exert an influence on the contact rigidity are clarified. Furthermore, the damping capacity for the fundamental vibration mode is clarified as a function of clamping conditions.

View full abstract