In the present report, we have theoretically and experimentally studied the whirling vibration of an asymmetrically rotating elastic shaft supported by asymmetrically elastic pedestals. The steady-state response for this multi-degree-of-freedom system having the terms of variable elasticity was obtained by means of the asymptotic method and the Newton-Raphson method. The whirling vibration was also described using the superposition of the main resonance component for unbalance and the secondary resonance component for the effect of gravity. The numerically analyzed results of the constructed simulation system agreed well with the experimental ones.

View full abstract

Theoretical analyses have been carried out on the nonlinear hydroelastic vibration of a cylindrical tank with an elastic bottom. In this report, a linear axisymmetric free vibration analysis of the bottom plate coupled with the liquid contained is presented. In the analysis, the effect of the static deflection of the bottom plate due to the static liquid pressure was taking into account. Numerical calculations were carried out for kind kinds of bottom plates with different thickness and made of different kinds of materials. The effect of the liquid contained on the bulging-type natural frequency, as well as the influence of the bottom plate motion on the sloshing-type natural frequency, was investigated.

View full abstract

The finite-element method is used to study the effects of rotating the panel principal axes with respect to the geometric axes, orthotropicity parameter, E₁/E₂, and shear deformation on the vibration and flutter characteristics of anisotropic plates. The vibration and flutter of anisotropic square plates, with a variety of boundary conditions, are analyzed by using the triangular finite element used in the previous paper. This element incorporated shear deformation according to Mindlin's theory.

View full abstract

The self-excited vibration of a rotor system induced by joint friction is investigated both theoretically and experimentally. The system is such that two flexibly supported rigid shafts are connected to each other and to the laterally fixed driving and loading shafts by two Kardan joints and one cross-groove (CG) joint. The initial intersecting angle is given to the CG joint. The joint friction of analysis is modeled as Coulomb friction. As the results of investigation, it is found that the analytical stability boundary characteristics correspond to those of the experiments, and the initial intersecting angle has the linear stabilizing effect on the critical friction force. The center joint friction has a stronger effect on the stability than does that of the end joints.

View full abstract

The mechanical snubber is an earthquakeproof device for a piping system under high temperature. It follows a slow displacement of the piping due to thermal expansion, but restrains the piping by a strong resistance force during an earthquake. This strong force can cause elastic failure of grooves on a brake disk, where steel balls are placed. In this report, a new semiactive snubber using giant magnetostrictive actuators and a ball screw is proposed in order to obtain an earthquakeproof device which has less resistance force in comparison with the mechanical snubber. Earthquake responses of a simple beam supported by the semiactive snubber are simulated using a continuous system simulation language, and the restraint effect and the effective controlling condition of the snubber are discussed. The semiactive snubber is built using Terfenol-D (Tb_0.3 Dy_0.7 Fe_1.92-1.95) rods, and its resistance force characteristic is investigated by means of a vibration table.

View full abstract

A simplified estimation method for first excursion probability of mechanical systems subjected to earthquake excitations is proposed. First, by a simulation method, first excursion probability of a single-degree-of-freedom system is estimated using stationary random noise as input. When the tolerance level is normalized by the expected value of the maximum response, which is well known as the average response spectrum, the variation of first excursion probability for various values of damping ratio and natural period is very small. These characteristics can be seen for the system subjected to nonstationary random noise. Second, the above mentioned result is examined by a theoretical method. Finally, the obtained result is applied to a secondary system. In this case, the variation of first excursion probability for various values of damping ratio and natural period is very small when the tolerance level is normalized by the expected Value of the maximum response of the secondary system.

View full abstract

The stability of a cantilever beam subjected to one-dimensional leakage flow is investigated. The motion of such a beam is expressed as the sum of the first few eigenfunctions of a cantilever beam. The critical flow velocities and the natural frequencies on the neutral stability are determined as a function of gap width. Experimental results are in agreement with analytical ones. The complex frequency of the four lowest modes of the system is calculated in several representative cased as a function of flow velocity. In the case that the beam is clamped at the upstream end, the system is found to lose stability by coupled-mode flutter. On the other hand, in the case that the beam is clamped at the downstream end, the system is found to lose stability by divergence first, and successively lose stability by flutter with increasing flow velocity.

View full abstract

This parer deals with the active vibration control of a flexible beam using the active sink method proposed in the previous papers. It is the purpose of this paper to present control laws of the active sink method with feedback control. First, by using a transfer matrix method, this paper describes the active sink system with displacement feedback. Then, two kinds of control laws are derived, one is to eliminate the reflection waves at the active sink point and the other to make the transmitted waves null. From a viewpoint of numerical analysis, the fundamental caracteristics of the active sink system are clarified. If the final goal of the vibration control is to make all the vibration modes inactive, the corresponding optimal control law is obtained by the active sink method. Therefore, by comparing the conventional control method such as velocity feedback with the active sink method, the causes of the incapability of the velocity feedback are clarified. Finally, the stability problem of the active sink system is discussed.

View full abstract

A method of evaluating simultaneous structural, control and shape optimization of flexible structures is presented in this paper. Two different and mutually conflicting criteria, the structural weight and the control energy to reduce vibration, are taken as objective functions. The qualitative aspect of the optimum solutions are discussed with transient response and control effort. Effect of geometrical shape of structure is also investigated for a two-bar and a twelve-bar truss problem.

View full abstract

Recent advances in experimental modal analysis have afforded powerful tools for practical mechanical designs. The machines or those idealized models are subjected to impulsive or sinusoidal loads in the test and the obtained experimental data combined with the finite-element analysis can provide modal properties or property matrices. The optimum design methods have been of interest among practical designers since the optimizers for optimum designs were added to new versions of general computer programs for structural analysis and many programs were released. However, most of the optimum design methods were based on analytical models, for example, the finite-element models, and there are few papers by which to study an optimum design method utilizing experimental data. Thus, this report aims at presenting a new method of direct sensitivity identification of property matrices by utilizing experimental data, and a new approximate optimum design method based on the sensitivities. Fundamental examinations of the proposed methods were made through simple examples.

View full abstract

This parer considers the robustness against spillover in the reduced-order control system of an active dynamic vibration absorber for high-rise buildings by computational and experimental research. Three methods are investigated: a LQ control using a reduced-order model, a spillover-considered suboptimal control and a spillover-considered suboptimal control using a low-pass filter. In the numerical calculations, it is shown that the control obtained from the displacement criterion possesses the stability limit in the control performance due to the spillover, and the highest performance is achieved by the spillover-considered suboptimal control using a low-pass filter. The optimal threshold frequency of the filter is also discussed. On the other hand, there is little spillover in the suboptimal control from the velocity criterion.

View full abstract

For many years a high-powered damping material has been requested for use in various industrial products. Viscoelastic, composite or sandwich materials have been developed and used only to be revealed as insufficient below 200 Hz. This paper introduces a lightweight, low-priced, and high-powered viscoelastic damping layer, in which varisized masses are distributed to compose dynamic dampers. The dynamic spring constant of the damper depends on d/h : d is the diameter of the mass and h is the height of the layer underneath the mass. The experiment manifests remarkable effects of the damping layer : only four or six masses are enough for the layer of 0.5 m×0.5 m size, the masses have exclusive effects for their respective natural frequencies, and the masses distributed regularly can be sufficient. The analysis of the FEM model with damping masses proves approximately the same effect as the experiment.

View full abstract

This paper discusses the robust stability against spillover for various reduced-order control systems of an active dynamic vibration absorber for high-rise buildings. Four methods are investigated ; a LQ control using a reduced-order model, a spillover-considered suboptimal control, a spillover-considered suboptimal control using a low-pass filter, and an H-infinity control using a reduced-order model. In numerical calculations, it was shown that the spillover-considered suboptimal control using a low-pass filter can achieve higher performance than the H-infinity control under the condition of the limited order of the filter.

View full abstract

In this report, an application of the fuzzy control algorithm to an active vibration control problem is considered. The method was evaluated through experiment with laboratory-scale dynamic vibration absorber-type active vibration control experimental equipment. As is well known, the fuzzy control algorithm does not depend on the information concerning the mathematical model of the controlled plant. Hence, it is not necessary to take into consideration the system order, effect of friction, values of plant parameters, and so on when we construct the vibration control system. Further, in spite of quite simple structure of the algorithm high performance concerning the vibration suppression is attained as compared to other active vibration control techniques.

View full abstract

Modal control of magnetic bearings is applied to a shell-structured rotor to reduce the structural vibration. It is intended that the rotor is light and has a wide supporting region of magnetic flux, hence it is made of a flexible shell sructure. These structures are apt to produce shell vibrations, in addition to bending vibrations of the rotating shaft. To improve the damping property of the structure, a new control scheme for magnetic bearing is introduced. Each magnet is controlled individually, leading to favorable structural damping, which contrasts with the traditional controller where a pair of magnets is controlled in a push-pull operation. The modal control technique is applied to the proposed magnetic bearings. The results obtained indicates a good damping property.

View full abstract

This paper proposes the theoretical design method of dynamic absorbers for controlling the noise radiation of the structure, which is expanded from "Vibration Control Method of Multi-Degree-of-Freedom Systems using Dynamic Absorbers". The main point of this method is combined with vibration analysis (finite-element method) and sound analysis (boundary element method). We demonstrate theoretically and experimentally that this method is effective for reducing the radiation noise of the plate structure model, which is a basic element of the structure, by setting up optimum-designed four dual dynamic absorbers.

View full abstract

Using a linear preview control model of a driver in lane-change maneuver, the stability criteria and performance index for the driver-vehicle system are derived analytically. In the stable area on the u-T_p : velocity, T_p : preview time) plane, the vehicle design parameters, rear wheel steering control algorithm and driver's characteristics can be discussed in terms of the measure for the good drive-feeling or possible safety criteria.

View full abstract

In measuring the roughness of machined surfaces by STM, it is necessary that the STM have a large scanning area. In this report, a newly developed STM system which meets this requirement is described. It consists mainly of an X-Y stepping motor stage and three tunneling tips in a line, each of which is driven by a piezoactuator. The tip positioned at the midpoint is used as a measuring probe; the right and the left ones are used as smoothing probes which are not sensitive to surface roughness. The STM is designed to cancel the relative shift in the z-direction due to thermal drift and/or vibration between the measuring tip and the specimen by using two smoothing tips. The basic characteristics of the STM system are investigated and measurements with millimeter-scale scanning are tried. The STM makes it possible to perform a large-scale scanning without reducing the resolution apparently. A new system combining an optical sensor with the STM is also constructed. Using the system, the results of the STM and those of the optical sensor can be compared closely.

View full abstract

Delamination in plastic-encapsulated integrated circuit packaging can lead to premature failure and also decrease heat resistivity. Therefore, development of nondestructive inspection of the bondind surface in a plastic package is an important subject for reliability improvement of integrated circuits. In this paper, delamination at the resin-chip interface which is measured using ultrasonic testing is discussed. At the delamination surface, the phase of reflected ultrasonic waves is reversed and at the bonding surface it is not. By using this phenomena, bonding and delamination can be discriminated. Delamination filled with water is also discussed. In this case, characteristic variation with frequency of ultrasound used is available for measurement of the gap thickness.

View full abstract

This paper describes an optical detection method of the edge position of the object with a curved sharp edge for developing an optical profile measurement system. The measuring principle is based on the straightness of a laser beam and the diffraction property at the edge of various shapes. Assuming a laser beam with Gaussian distribution parallel to the beam axis, the transmitted beam power is reduced with an increase in the beam interception area by the sharp edge which is positioned at a right angle to the beam. The effects of optical parameters and edge curvature, including straight and circular edges, on the diffraction pattern and the transmitted beam power were experimentally evaluated. The obtained results are as follows: (1) The transmitted beam power is directly related to the diffraction pattern and the attenuation of beam power is a function of the edge position. (2) The detecting sensitivity is influenced by the radius of a photodetector much more than by the distance of the detector. (3) The transmitted beam bower decreases and the secondary wavelet of the diffraction pattern becomes small with an increase in the edge curvature. (4) A maximum exists in the difference of detected positions between circular and straight edges.

View full abstract

"Zero-spacing error (ZSE)", which occurs in measuring flying height by light interference, is caused by phase shift of reflecting light from a magnetic head slider surface. The measured flying height is larger than the true height. The ZSE is measured by two methods. The first measures ZSE by calculating from complex indices of refraction of three kinds of slider material measured by ellipsometer. In the 400∼800 nm wavelength range, ZSEs were confirmed with ±7% accuracy (average). The second method is a direct measuring method using a "standard wedge spacer". The spacer makes a wedge-shaped space between the ceramic samples and the spacer. This method was also used to reconfirm ZSE.

View full abstract

For the measurement of straightness or transverse displacement using a laser beam, pointing instability of the laser beam caused by thermal deformation of the laser instrument was a primary factor of error. In order to cancel the effect of the instability, the optical method was conceived. In this paper, the principle of the new measuring method of the transverse displacement, which has no effect on the instability, is described and the experimental result of measurement using the new method is reported.

View full abstract

One of the methods of improving the accuracy of the three-dimensional coordinate measuring machine (cmm) is numerical error correction. For the method, calibration of cmm must be carried out periodically. However, the long term stability of the accuracy cannot be predicted, and it is often difficult to determine the interval time of calibration. For the purpose of determining the periodic time of calibration, the secular change of a cmm has been investigated for over seven years. The same calibration procedure was performed with the cmm under the same environmental conditions. Results of tests show that the change in the accuracy occurs in a short period of one or two months. The accuracy must be frequently measured for the judgment of the time to practice the calibration procedure.

View full abstract

This paper deals with the synthesis of a single-input/single-output ripple-free deadbeat tracking control system by using a two-degrees-of-freedom compensator scheme. The proposed system can simultaneously follow reference input and eliminate the effect of disturbance with no ripple in finite time. Both the reference input and the disturbance are general ones which are generated by linear time-invariant systems. Moreover, the optimal ripple-free deadbeat tracking control system, which can make as moderate as possible the transient responses for the reference input and the disturbance, is proposed. The algorithm is based on a simple iterative operation with no matrix inversion.

View full abstract

Suitable operation planning is essential for multifingered hands. A new algorithm is proposed for the planning, i.e., to compute the admissible region of fingertip position, to select the best fingertip position and to determine the optimal finger force. The approximate area of fingertip position is computed by working out a series of optimization problems solvable by the simplex method.

View full abstract

This paper proposes an integration method of multiple sensors for applications to a curved metal surface cutting system. For precise cutting, the work surface must be measured by high accurate sensors. In general, these sensors do not have a wide measurement range. For the automation of the sensing process, we set the sensor array at the tip of the manipulator. Then the robot must move the array to the work surface so that the sensors are active. In order to shorten the approaching time to the work, the system should use wide measurement range sensors. For precise cutting and fast approach, the system should use both high accurate sensors and wide measurement range sensors. In order to use different types of sensors effectively, we propose the multisensor integration system based on the neural network and fuzzy inference techniques. In the experiment, we used 3 different types of eddy current sensors. This sensor suffers measurement error from the inclination angle between sensors and an object. With the consideration of the inclination angle in this integration system, the approaching time is improved. The proposed system is shown to be effective through extensive experiments.

View full abstract

This paper describes a torque control method for a geared DC motor. The torque sensory feedback provides a benefit for the fine torque generation by geared DC motors. However, it reduces the structural stiffness and causes a low response in both torque and position controls. To improve this, a control method based on an acceleration control is proposed. The proposed method uses two control loops. At the inner loop, a high-performance control system for the acceleration of the motor axis is constructed. At the outer loop, a control system with a positive feedback of the load position is introduced. With this structure in mind, the paper suggests that a high-stiffness torque control system can be achieved when the acceleration control system at the inner loop is carefully designed using the disturbance observer.

View full abstract

This paper describes a directional control method and optimum feedback gains for directional correction of a small diameter tunnelling robot which constructs conduits for telecomunication cables. The directional control of a tunnelling robot conventionally depends on both the experience and intuition of the operator, and there have been no studies with regard to its automation. There-fore, in order to establish automatic control technology for a small-diameter tunnelling robot, we proposed the feedback directional control method and evaluated the control laws by a dynamic simulator. Results clarify that both a directional error feedback loop and a position error feedback loop are necessary for stable control. Moreover, we obtained optimum feedback gains to minimize position error by using a dynamic simulator. Simulation results clarified the validity of the proposed directional control method.

View full abstract

This paper describes the autotuning of feedback gain for a small tunnelling robot. We have already proposed a directional control method in that the head angle of the control input is the sum of the deviation multiplied by feedback gain K_p and the angular deviation multiplied by feedback gain K_a. In this paper, we use a neural network to obtain feedback gains K_p and K_a. The input of the neural network is the initial deviation and initial angular deviation. The output of the neural network is the feedback gains K_p and K_a. This neural network learns from deviation error. The optimum gains obtained by the proposed method agreed with the optimum gain obtained by trial and error. The neural network which can apply to any initial deviation was formed by using plural deviations. Moreover, this method can tune the optimum gains to any design line. These results showed the validity of this method.

View full abstract

Even the potential method, which seems to be most effective for automatic planning of the collision-free manipulator path among obstacles, uses transformation of real space coordinates of the obstacle to joint angle space, so the analyses become difficult or impossible when the manipulator degree of freedom becomes large. This paper proposes the application of fuzzy reasoning to enable handling of the coordinate of the obstacle in the real space coordinate without transformation to joint angle space in order to solve the above-mentioned difficulties, while basically adopting the potential method. Moreover, this proposal adds a great advantage in that much useful different dimensional information can be used for planning, offering a valuable method to further develop this problem. For example, in addition to the information about the deviation from the target and the proximity to the obstacles, the rate of change of these quantities can be included in the planning. This usefulness was confirmed by simulation.

View full abstract

The cellular robotic system (CEBOT), which has been studied by the authors, can carry out a variety of tasks by an optimal structurization. CEBOT consists of various kinds of components called "cells". A cell, which has a basic function and its own knowledge base and database, forms a group called a "cellular robotic society". In order to organize many kinds of structures for task execution and the cellular robotic society, each cell should genetically have algorithms similar to a living creature. This paper deals with a self-organizing method based on the genetic algorithm proposed by the authors. We report system organization methods (horizontal-and vertical-organization), and discuss how to organize the system effectively by using the proposed numerical analysis. As one example of the self-organizing system, a self-organizing distributed intelligence system based on the concept of CEBOT is reported. Some simulations of the self-organization and numerical analyzing results are also reported.

View full abstract

Impact sound occurs when a collision between objects occurs. Using this impact sound, we propose a robotic control method to reproduce the desired impact sound, namely, the audio-feedback control for impact sound. It is rare to control the manipulator using sound information, but is very useful in our life. Therefore, we hope it has the possibility to control the manipulator in wider tasks. Impact sound occurs in very fast phenomena, so it is effective to use a learning control. In this paper, we propose a learning control method to optimize the least-squares criterion of learning error, and carry out experiments on audio-feedback control for the impact sound, showing the effectiveness of the proposed method. This method is applied to obtain a unique control gain based on the system's impulse response model. Therefore, if the impulse response model becomes available, the learning control system can be constructed without knowing the system's parameters.

View full abstract

The application of dynamics in the living organism to the field of engineering is very instructive. To make an artificial bending wave propulsion mechanism, a simulation study of propulsion mechanisms imitating the bending motion of organisms in water was carried out. As the bending motion, a swimming snake's motion was employed. The bending wave of the organism was changed into a construction of the multilink mechanism. This simulation model was defined as the "multilink model". In addition, the simulation model based on the approximate theory of the organism's propulsion was defined as the "approximate organism model". This model is a prototype of the "multilink model". The thrust force and the swimming speed were calculated. The characteristics of the thrust force and the swimming speed with change in the frequency, the wave amplitude and the number of links were discussed.

View full abstract

In order to study the biological information system in plant, a measuring system for bioelectric potential distribution on the plant surface was newly designed and developed. The system includes a tissue culturing subsystem to measure bioelectric potential changes of callus continuously for more than 30 days in the process of redifferentiation. By using our sytem, biological potential distribution on the surface of the callus and electric field formed around the primary root were measured. Consequently. we found that bioelectric potential on the callus surface fluctuates in the vicinity of the position where a shoot will generate. And once shoot generated, the fluctuation disappears. Furthermore, we showed that patterns of the electric field formed around the primary root fluctuate and change prior to its growth, and the patterns are changed by other closely settled root. Accordingly, it is probable that the fluctuation, and rhythm of bioelectric potential are related to generating information in differentiation and growth, and intercommunication between plants.

View full abstract

Recently, the fine positioning system has been used in the mechanical industry, and has required the improvement of positioning accuracy, reduction of positioning time, increase of carrying capacity and so on. However, the existing control method for the positioning stage may cause delay of phase response or residual vibration when disturbance or overweight acts on the system or stage drives at high speed. In order to obtain the optimum control inputs, it is necessary to grasp the dynamic behaviour of the system precisely. First, this paper deals with the transient response of the positioning stage using the linear graph theory reported in the previous papers. The simulated results of transient response of the positioning stage are compared with the experimental ones, and they are found to be in good agreement. Next, a procedure which can reduce the settling time of positioning is proposed by utilizing this method. As a result, this procedure was able to reduce the settling time by 55%, as compared with only proportional control.

View full abstract

One of the effective ways for reducing the weight of a helical spring is to make a hole in the cross section of the wire of the spring. This paper discusses the shapes of outer and inner boundaries of the wire cross section of the helical spring that has minimum weight. Stresses are calculated by FEM based on the stress function, and the optimization problem is solved by the steepest descent method used in the previous paper. In consequence, the converged solutions obtained from various initial shapes greatly differ in shape, but are almost equal in weight. The stresses around the outer boundary of the converged cross sections can nearly be made uniform. It is also found that there are many solutions whose weights are almost equal to those of the converged solutions. In comparison with a helical round wire spring, the weight has been reduced by about 50%.

View full abstract

Kinematic and static characteristics of robotic parallel crank mechanisms have been analyzed and represented in the form of charts which support rational syntheses of the mechanisms as robot and legs. Namely the value of torque of motive shafts to hold attitudes of mechanisms and the static compliance of the work points when the mechanisms take arbitrary shapes have been formulated by the total link lengths and by introducing four dimensionless parameters. Characteristic charts of statics have been made by plotting lines of positions of the output point where both of the factors are constant in dimensionless working spaces. Making use of several characteristic charts and simple transformation formulas, the values of the two factors can be analyzed, respectively, when the mechanisms take arbitrary positions and attitudes.

View full abstract

A method of determining the parameters of the rheological properties which express the model of the limiting shear stress of lubricants is proposed. The relations betwen the traction coefficients of the roller tests and the mean film temperature rises of lubricants are calculated. The maximum traction coefficients are revised to those of the isothermal condition by the traction coefficient-temperature relations and analyzed by using an approximate formula which estimates a maximum traction coefficient based on the limiting shear stress. Zero pressure value of limiting shear stress τ₀ and slope of the limiting shear stress-pressure relation m are presented for two traction fluids Santotrac50 and Daphne 7074 at temperature 313 K. Maximum traction coefficient and limiting shear stress at high temperature are predicted by using the traction coefficient-temperature relations.

View full abstract

A calculation method of rotor loads and some numerical examples are presented. Gas loads acting on the male and female rotors are calculated from sealing lines which are formed as border lines between working spaces. The procedure is simplified by taking advantage of the property that the sealing lines do not change in shape when the meshing rotors rotate. Axial loads as well as radial loads are obtained from sealing line shapes projected to two planes which are parallel to the rotor axis and intersect perpendicularly. Especially, the axial loads are calculated using the principle of virtual work. The calculation results show that a high operating pressure itself does not necessarily increase the amplitude of the rotor load, but the mismatching between the operating pressure ratio and the built-in pressure ratio makes the amplitude larger.

View full abstract

A 3K-type planetary gears CVT (Continuously Variable Transmission) can be manufactured, which provides from zero to a high-speed ratio and output shaft characteristics with constant power. However, it is said that this type of mechanism is not as efficient as other types of transmissions. This report proposes that the mounting position of the loading cam affects the efficiency of this CVT. Three examples of position; of loading cams were supposed, and their efficiencies were numerically and experimentally obtained. The results of both experiments and equations were compared. From this comparison, it was found that the most appropriate position of the loading cam is between the track disk and the casing.

View full abstract

We have developed an hourglass worm gearing for valve actuating devices, or limitorques. The basic theory is the secondary action theory of the skew gears, and the modifying cutting method of the Hindley-type hourglass worm gearing is applied. The new worm cutting machine was made for this worm gearing, and the completed worm gearing was tested. The worm gearing showed a load capacity of two to three times that of the traditional cylindrical worm gearing, and also the efficiency showed a higher value than that of the cylindrical worm gearing. This hourglass worm gearing has run over 5 000 cycles of rotation without trouble, while a sufficient life cycle for this device is about 3 000 cycles of rotation.

View full abstract

This paper presents a study on the effect of supporting conditions on root stresses of thin-rimmed internal spur gears used for planetary gear unit. Stress analyses by the two-dimensional finite element method (FEM) for thin-rimmed internal spur gears fixed with bolts or supported with pins were carried out. The effects of the supporting condition, the position of the bolt (pin) and the number of bolts (pins) on the root stress, the stress at the hole for the bolt (pin) and the deformation of the rim were clarified.

View full abstract

In order to improve the chucking performance of a wedge-type power chuck with three top jaws, the top jaws with a tapered gripping surface were ued in this study. Gripping accuracy tests of the chuck, bending tests of a chuck-work system and cutting tests were performed. The investigation showed that the use of the top jaws with the taper angle of 0° makes the gripping accuracy superior, while increase of the taper angle causes a lowering of it. The greatest bending stiffness can only be achieved when the gripping force is adjusted to the taper of the top jaw gripping surface used. The method of tapering the gripping surface of the top jaws is also effective for preventing chatter vibration in practical use.

View full abstract

It is not always easy to verify workpiece accuracy determined by the minimum zone method even with a long calculation time. Therefore, the author clarified the geometrical condition of roundness in another report. According to the report, if there are two points A, B on the inscribed circle determined by the minimum zone method, and another two points C, D on the circumscribed circle, segments AB and CD intersect each other. Thus, this paper shows a calculation method based on the geometrical condition and the computer simulation of the method. The calculation method has the advantage that the true solution is definitely obtained, the algorithm of it is very simple and the calculation time is much shorter than the standard method.

View full abstract

To achieve a damage-free grinding surface of brittle materials, the characteristics of brittle-ductile transition must be realized and the surface texture of the grinding wheel must be adequately controlled. In the study, a grinding wheel is truncated by grinding cast iron after dressing. The peak-valley value of the cross-sectional profile of the truncated wheel is much better than the dressed one. Moreover, after analyzing the surface texture results from various experiments, it is found that the statistic mean value of the critical depth of cut d_c is approximately 40 nm by the truncated wheel with up-cut grinding. The percentage of fracture surface is decreased with the decrease in the depth of cut in each case. The data points on the skewnesskurtosis plot of ductile mode grinding surface are relatively concentrated compared to the fracture mode surface.

View full abstract

The thermal expansion and wear of the grinding wheel are measured during cylindrical grinding by a new experimental method called the "noncontact scanning-type radial difference comparison system". This method is developed from the previous automated difference comparison method, and a new noncontact scanning system is introduced instead of the blade method. A laser optical displacement sensor supported on the positioning table is scanned on the grinding surface of the wheel by means of computer control, and a surface profile of the grinding surface is constructed from the output of the laser displacement meter. By using this method, the measuring procedure is eliminated and time interval between measurements can easily be shortened.

View full abstract

The influence of cutting conditions (cutting speed and feed), shank rigidity, shape of cutting edge and tool wear on axial hole deviation was examined in the special machining device using 16 kinds of gun drills of 4 different shapes, each in 4 different lengths. As a result, the following are clarified. The axial hole deviation increases with the increase in shank rigidity. It decreases reversely with the increase in cutting speed. It becomes minimal at a feed of 0.06 mm/rev. It increases with the increase in self-piloting of the tool if the tool inclines at the beginning of drilling. The axial hole deviation of the worn tool is almost the same as that of a new tool as long as the wear is not severe.

View full abstract

Metallurgical orientation generated by the plastic sliding on the cutting edge produces material anisotropy in both the cutting direction and the feeding direction on a machined surface. This paper deals with the experimental research to evaluate the material anisotropy on the machined surface using a scanning acoustic microscope. A point-focus anisotropic lens, that can be applied for observing an acoustic image, and evaluating an acoustic material signature, was newly developed. Then, material characterization on the end milled surface was carried out. It was made obvious that the Rayleigh wave velocity, measured while rotating the anisotropic lens, showed a periodical change having peak values at the cutting direction and the feeding direction respectively. At the same time, it was made clear that the degree of material anisotropy, the angle of metallurgical orientation and the mean Young's modulus can be evaluated separately through analysing the measured Rayleigh wave velocity.

View full abstract

An automatic image recognition system that comprises a CCD camera, frame memory and computer, has been built to recognize the bearing of scraped surfaces. Elimination of the influence due to uneven lighting by computer software, and reduction of the noise component contained in image data by introducing the mathematical morphology method were attempted. The results were successful. Furthermore, comparison of unevenness of the scraped surface measured by coordinate measuring machine with the results obtained by the image processing system, has been made. Introduction of image processing techniques enables us to count the number of bearing spots per unit area on the scraped surface automatically, and the possibility of quantitative evaluation of flatness has also been shown.

View full abstract

A magnetic force exhibits a strongly nonlinear property and is usually modelled by a simple function proportional to (pole-gap distance)^-P where P > 0. If the function is to be rigorously calculated mathematically however, a great deal of effort may be required. This paper discusses an algorithm and presents program to identify numerically the magnetic force in the form : F(x)=A/(B+x)^P+C, in which constants A, B, C and P are determined by processing a set of discrete data points to satisfy the least square error condition. The listing of the sub-program is given together with an example of the main program which drives the conversational computation. An example of the conversation is recorded and numerical example is demonstrated, in which the case of a remarkably nonlinear spring (P=7∼8) is found experimentally in a pair of super-permanent magnets whose material is composed of Nd₂Fe₁₄B. The identification error of the formula is less than 3% of that of the conventional formula when B=C=0 formally.

View full abstract