In order to investigate dynamic behavior of robot manipulators to the Coriolis forces and the centrifugal forces, a vibration analysis of 6-degree of freedom manipulators with flexibilities of transmission systems has been developed. A set of governing equations of motion for robots was derived and analysed by the direct integration of Runge-Kutta procedure on the personal computer. The simulation results showed the influences of the Coriolis forces and the centrifugal forces on the dynamic behavior of robot manipulators controlled by the position and the velocity feedback.

View full abstract

A method for FEM eigenvalue problems of symmetrically laminated composite shells is developed. A 15 d. o. f. flat triangular shell element is formulated considering the anisotropic behavior of composite materials with the assumption that the transverse normal strain is linear in symmetrical lamination. The present method enables us to analyze large scale problems by only a desktop microcomputer, using a subspace iteration algorithm and a skyline solver scheme. In order to estimate the accuracy of the program, eigenvalue problems of Donnel-type circular cylinders are calculated and compared with exact solutions.

View full abstract

When the new Shinkansen train set, the Series 100 EC train, came into service, lasting longitudinal vibration with a constant frequency arose frequently near the maximum operation speed. The vibration was of a sort not experienced before. This paper describes the calculation analysis for the longitudinal vibration if the Shinkansen bogie cars. The results are summarized as follows. The lasting vibration is the 1st modal one caused by vertical track irregularities through the pitching motion of bogie trucks. It becomes marked in the speed range where the pitch phase of the front truck is close to that of the rear one. The bolster-anchor height has an optimum value against the 1st mode, but it is so narrow that the vibration grows rapidly as the real height differs from the optimum one.

View full abstract

The steady-state response of damped disks rotating at a constant angular velocity was analyzed subjected to the external harmonic force at one point fixed in space. The distribution of the internal stress was obtained analytically from the equilibrium equation of radial force which was expressed by the space coordinate. The equation of motion of the rotating disk with the internal damping expressed by the complex Young's modulus was derived by use of Galerkin's method. The driving-point impedance, the mode shape, and the relationship among the angular velocity, the harmonic exciting frequency and the maximum amplitude have been calculated numerically.

View full abstract

In this paper, the global stability and nonlinear behavior of a poppet valve circuit with a pipeline are studied by the digital simulation method. First, the local stability, or the stability of the steady state, is examined analytically, assuming that the pipeline is a lossless distributed-parameter element, and it is clarified that the instability of small valve lift is due to the effect of the pipeline. Second, it is shown that "hard" self-excited vibrations occur when disturbances beyond a certain critical value are imposed on the valve, and they can occur even for supply pessures lower than the setting pressure of the valve.

View full abstract

This paper presents a three-dimensional finite-element model of a human middle ear, which includes the ossicular chain and reaction force from the inner ear. Physical constants such as eardrum Young's modulus and the reaction force from the inner ear and boundary condition of the eardrum are determined from a comparison between the frequency responses of the eardrum volume displacement obtained from FEM analysis and those of the temporal bone experiments. Modal vibration patterns of the eardrum and the ossicular chain obtained from FEM analysis are in good agreement with the experimental results using holography and the stroboscope.

View full abstract

Vibration response analysis is applied to a classification of cast iron products. We have studied the correlation between the mass and the natural frequency of the products. The test pieces used here are some types of water pump impellers for automobiles. It is difficult to manually classify many kinds of impellers which have similar configurations and dimensions. We have found that the frequency characteristic of test pieces, which depends on the rigidity of a disk part, can be used for the classification. As a result, we have made a characteristic map on the mass-natural frequency plane of the products, which facilitates classification.

View full abstract

An air damper has a great advantage in that it is independent of temperature change. This paper proposes an analytical approach for designing an air damper with a piston and a cylinder. The characteristics of the air damper depend on the properties of air: viscosity and compressibility. Thus, the air damper shows the so-called visco-elastic property: the applied force to the piston is resistant to the velocity and displacement of the piston. As a result, the air damper has two important factors: the damping coefficient and the spring constant. Two types of air damper are discussed. Type I is a viscous damper whose resisting force is proportional to the piston velocity, and type II is a nonviscous damper whose resisting force is proportional to the velocity-squared of the piston. The present analysis was well explained by the experimental evaluation of the damping coefficient and the spring constant of the air dampers. This paper leads to a simple and accurate method for designing an air damper.

View full abstract

The behaviores of a truck crane during pile pulling out, tipping and load hoisting are numerically analyzed, using FEM model with nonlinear elements. To verify the analysis, experimental tests are carried out using a scale model. The numerical results are in good agreement with the experimental data. The differences between the dynamic stress and rope tension for each motion are clarified.

View full abstract

This paper describes a study of dynamic fluid forces acting on a long seal where the rotating motions of a rotor were composed of conical and cylindrical whirl. Theoretical fluid forces were determined using a bulk-flow model to obtain the rotor-dynamic coefficients of the long seal. The experiment was conducted under changing phase difference between the seal inlet and outlet whirling movements. The tangential and restitutional forces were obtained from the measured dynamical fluid forces at various rotor speeds, whirling amplitude and with different pressure drops across the long seal. Then these measured values were compared with the theoretical finite-length solution and were found to be compatible. As a result, the dynamic fluid forces are largely affected by the whirling phase difference between the seal inlet and outlet. It was confirmed that the total fluid force acting on the rotor is given by the sum of two fluid forces due to the cylindrical whirl and due to the conical whirl, when the rotor and the outer cylinder are set in concentric alignment.

View full abstract

Stand experiments of hunting of a bolsterless experimental truck for high-speed electric cars using a compound circular wheel profile were conducted. Experimental results were compared with theoretical results. Stand experiments were carried out with and without yaw dampers while varying the tread surface condition (dry or oily) and direction of wheel rotation (front-half vehicle experiment and rear-half vehicle experiment). Nonlinear digital simulations for these stand experiments were conducted. Simulation results agreed well with experimental results, verifying the validity of the simulation method. Then, eigenvalue analyses for three linearized models (front half of vehicle, rear half of vehicle and whole vehicle) were performed. Results of linearized analyses coincided well with experimental results, especially in the case without yaw dampers. Critical speeds of these three models were nearly equal, implying that hunting characteristics of the whole vehicle could be estimated by half-vehicle stand experiments.

View full abstract

When a bar collides perpendicularly against a beam, an impact force deters the movement of the bar and produces a longitudinal wave in the bar. This impact force also pushes the contact point of the beam, and the beam begins to deform. The longitudinal wave produced by the impact travels backward, reflects at the back surface of the bar, and comes back to the impact point. At the time when the reflected wave reaches the impact surface of the bar, this wave pulls back the impact surface of the bar and separates the bar from the beam, and thus the impact terminates temporarily. As the center of gravity of the bar continues to move in the initial direction with a slightly reduced velocity, the bar collides again with the beam and the second impact begins. The second longitudinal wave is produced, and after a short period this second impact also terminates. This type of intermittent collision occurs repeatedly until the half-cycle of the beam's deformation motion is accomplished, at which time the collision finishes. These processes are calculated analytically and observed experimentally.

View full abstract

The squeal of a wheel brake with a single shoe is investigated by a beam-disk model. Resins attached to the center of a flexible beam in place of the brake shoe are pressed against the circumferential surface of a thin steel disk which rotates at a constant angular velocity. Two types of squeals were caused. One was caused by the coupling between two beam modes. Its frequency was nearly equal to the natural frequency of the beam, which was near the natural frequency of the disk. Then backward wave, forward wave components or standing wave components with stationary nodal lines are shown in the transverse disk vibration according to the relationship between the squeal frequency and the natural frequency of the disk. The other squeal was caused by the coupling of a single disk mode to a single beam mode. Its frequency was equal to the natural frequency of the disk, the damping coefficient of which is small.

View full abstract

A method of predicting sound absorbing capability for multi layer sound absorbing materials was studied by using the boundary element method. The sound absorbing material was considered as a medium transmitting sound waves, and it was characterized by the complex propagation speed and the complex effective density, which were measured using the two-microphone technique. The sound field containing the sound absorbing materials was analyzed using the boundary element method developed for multi domain problems. The normal sound absorption coefficient of the materials was calculated and compared with the experimental coefficients. The excellent agreement achieved suggests that the present method is sufficiently reliable to predict the normal sound absorption coefficient.

View full abstract

The present study deals with an energy-saving speed-control system which consists of a variable displacement axial piston motor receiving constant pressure oil and a digital servomechanism using two high-speed two-way on/off solenoid valves for adjusting the displacement of the motor. In this system, the motor speed is controlled by displacement adjustment. A command pulse, whose width is determined according to a PD algorithm, is sent to either solenoid valve. It is shown that although the swash plate of the motor is moved stepwise by the on/off action of the solenoid valves, the motor speed can be controlled smoothly because of a low-pass-filter characteristic of the motor and inertia load. This speed control system using a digital servomechanism has very good steady and dynamic characteristics in a wide speed range.

View full abstract

This paper describes an active control method to suppress liquid vibration while transferring a container. We propose a basic model of a mobile robot for the transportation of liquid using a container. This model considers the rotational motion of the liquid in a container in a case in which the liquid is accelerated horizontally. The model is a linearized lumped model with two degress of freedom, such as one rotation and one directional transferring metion. Based on this basic model, we propose the application of the optimal control mothod. The efficiency of this control method for suppressing the liquid vibration in a container is shown by simulation and experiment in comparison with the other control methods in this paper.

View full abstract

This paper discusses a method for the optimal active vibration control of a plate using piezoelectric actuators. The analytical results under the optimal control were obtained by combining the eigenfunction procedure and the optimal active control theory. Numerical calculations were carried out for a clamped circular plate with concentrated masses which was controlled by bending moments generated by the piezoelectric actuators. It is confirmed that there is an optimal location for the actuator in each mode.

View full abstract

This paper is concerned with the active vibration control of a free-free beam. The beam is reduced to a finite-degree-of-freedom system by the modal analysis, in which the mode function is derived from the transfer matrix method. An active control force is produced by a pair of tendons and a DC servo motor attached to the beam. The state of the beam is presumed by the minimal order state observer and the control force is determined by the digital optimum regulator theory. It is found that the active tendon control method is effective for suppressing the vibration of the free-free beam.

View full abstract

This paper proposes physical parameters indentification of flexible structures handled by robots. A simplified physical model based on the constrained mode method is derived in this paper. Using this model, we present two kinds of identification algorithms: the first one uses the least squares method. In this algorithm, we analyze the effects of neglected residual modes on the identification results. In some cases, neglected residual modes in the algorithm causes degradation (bias) of the identification results. We reveal that this degradation (bias) can be removed to some extent by simply feeding back the velocity of the elastic displacement adequately. We also describe desirable sensor position. The second algorithm uses the extended kalman filter by which we can identify the physical parameters without using acceleration information. Some simulation examples of identification are shown to confirm the effectiveness of these two methods.

View full abstract

This paper deals with the mechanism and control of a micromanipulator with six degrees of freedom aimed at applications in biotechnology and microsurgery. Since operations in these fields develop rapidly and are complicated, a dexterous micromanipulation system has long been desired. This paper proposes a 6 D. O. F Micromanipulator to make such work easier. The manipulator has the following features: (1) The manipulator has 6 D. O. F. The tip of this manipulator can move along three axes, and also can rotate along each axis. (2) This micromanipulator is very small and light, since it is made of PZT actuators. (3) Since it has a position and force feedback system, it can control desired poses and forces. (4) A fuzzylike variable gain feedback system is employed here, and hence the system performs better than the fixed gain feedback system. These technical merits will make it possible to operate the micromanipulator in much more complicated environments under a microscope.

View full abstract

The present paper deals with precision insertion for chamferless parts with uncertain positional information by heuristic seach. The heuristic search is constructed of fuzzy pattern matching between fuzzy sets of search areas and fuzzy sets of the hole center position. Fuzzy pattern matching evaluates the value of the heuristic function. The search of the hole is started in the search area having the highest value of the heuristic function. RCC (remote center compliance) and a force sensor are used in the search. RCC modifies the position of the peg finely and the force sensor measures whether the insertion has succeeded. The usefulness of this method is shown by experimental results.

View full abstract

Sensors are important for recognizing the system state environmental status in the intelligent robotic system. Thus, the sensor integration system (SIS) has been studied in a wide range of applications. In this paper, it is shown that the SIS can expand the measurable region of sensors with higher accuracy by multiple sensors and that operators can use the system as easily as a single high-performance sensor system. Systems which have been reported so far do not have flexibility for changing/replacing sensors. Thus, this paper presents an approach to the SIS with the knowledge data base of sensors, so the proposed SIS has the flexibility for changing/replacing sensors. This system consists, of four subsystems: 1) sensors as hardware sensing devices, 2) knowledge data base of sensors (KBS), 3) fuzzy inference, and 4) neural network(NN). This system can estimate the error for the sensor's measured value by fuzzy logical inference with KBS. The measured values are integrated by NN. The inferred error and measured value are put into NN. Then, NN's output gives the integrated value of multiple sensors. Thus, the proposed system is shown to be effective through extensive experiments.

View full abstract

In this paper a new model is developed to explain complex biped locomotion. This model is based on a ballistic walking model which assumes no torque on any joints in the swing phase, and uses the penalty function to solve the constrained dynamics of the stance phase and the swing phase. The computer simulation using the model showed that successful walking requires; (1) equality and inequality constraints such as to lock the knee of the stance leg extended, (2) the turning point of the joint torque in order to move the center of gravity of the body then the swing leg forward, (3) the ankle torque of the swing leg in order to avoid stumbling.

View full abstract

Cellular robotics (CEBOT) has been previously reported by the authors as one realization of a dynamically reconfigurable robotic system (DRRS). CEBOT is considered to be a very flexible system and will be applicable to a robotic system which works in various environments. When CEBOT is required to perform tasks, many cells, which can be knowledge sources, communicate with each other and then carry out the tasks automatically. So CEBOT is also a decentralized coordinated reasoning system and is one of the distributed intelligent systems. In designing a distributed intelligence system, the distribution of the communication volume among the cells becomes a central issue. For the case of CEBOT, it is best that reasoning can be carried out among each knowledge sources with as little communication as possible. Therefore, each cell must have the ability to reallocate their knowledge automatically in order to reduce the amount of communication; this is called intelligent communication in this paper. In this paper, we propose one of the communication evaluation method based on the amount of communication information and also describe an optimal knowledge allocation method on CEBOT as one realization of a distributed coordinated reasoning system by introducing the sensitivity function of the knowledge allocation.

View full abstract

This paper deals With the path planning and sensing planning expert system with learning functions for the pipeline inspection and maintenance robot, Mark IV. The robot can carry out inspection tasks to autonomously detect malfunctions in a plant pipeline system. Furthermore, the robot becomes more intelligent by adding the following functions: (1) the robot, Mark IV, is capable of inspecting surfaces of storage tanks as well as pipeline outer surfaces; (2) in path planning, the robot has a learning function using information generated in the past such as a moving path, task level and control commands of the robot; (3) in inspecting a pipeline system with plant equipment such as valves, franges, T-and L-joints, the robot is capable of inspecting continuous surfaces on pipeline. Thus, together with the improved, path planning expert system (PPES) and the sensing planning expert system (SPES), the Mark IV robot becomes intelligent enough to automatically carry out given inspection tasks.

View full abstract

Diesel railcars in Japan have played a big role in transportation on mountainous lines. However, it can hardly be said that the efficiency of the current transmission is high enough when running up rising gradients. The authors built a new transmission based on an original design conception, aimed at simple construction and high efficiency. It is equipped with a pneumatically operated claw clutch. The clutch is electronically controlled to synchronize its output shaft speed with that of the imput shaft before the clutch is engaged. Built into a diesel railcar, the transmission was tested on a test stand and later in a car run on the line in the yard of RTRI, JR. The test result was satisfactory.

View full abstract

In order to obtain a basic guide for the optimization design of the flowmeter, the dynamic behavior of a measuring mechanism is analyzed by using the signal flow graph. New methods of estimating measurement performance are established. As a result, the mechanism of the occurrence of measurement errors is elucidated in a periodic changing flow rate. To improve the measuring accuracy, a guide to the design of the fluid and mechanical frictional resistance including both the location of vane wheel based on digital simulation techniques and experimental inspection is proposed.

View full abstract

This paper describes an allocation and distribution planning method for packing objects in a material flow system. We consider a problem of packing objects in an allowable space in this paper, for which the knowledge-based system is proposed with the space geometrical expression methods, assuming that objects are rectangular. In packing objects in the admissible space, the space left for the allocation is divided into several subspaces, according to the size of the objects. Then, these divided subspaces and objects are represented by the relation of connection and the division of the space. For the reasoning method, we employ human like heuristic knowledge: in cases in which the width of an object is larger then that of the allowable space left, we cannot pack the object in the space, but can pack the object by applying the force and pushing the object into the space. In cases of packing the object in unfavorable space, the system takes two steps backwards and reconsiders the packing process in a better way by the backtruck mechanism of the system. This paper shows some comparisons of packing efficiency for the evaluation of the knowledge-based system which the basic rules, so rules, so that the packing planning system can be improved by different additive rules, so as to decrease the dead space in the total area. Extensive simulations have been carried out to evaluate different system performances.

View full abstract

This paper presents a study on the stability of the 2" floppy disk drive. A design method of the disk stabilizer that makes the rotating disk stable is presented. The stabilizer and the read/write head are both modeled by translational springs with high stiffness. The air-film stiffness within the disk cover is determined from the flow rate of the air inside the cover, based on the Navier-Stokes equations. The solution is obtained by applying the Galerkin method to the governing equations, with the use of the multimodal expansion approximation. Numerical results show that the rotating 2" floppy disk is unstable at 3600 rpm without the stabilizer. It is also shown that the stability of rotating disk is strongly dependent on the shape of the stabilizer attached to the disk surface. A stabilizer that contacts the disk at four points is developed and shown to be quite effective in stabilizing the 2" disk rotating at 3600 rpm.

View full abstract

An investigation was carried out to clear the natural vibrations on the outer ring itself of the ball bearing. In this paper, the vibration of fifteen outer rings of the ordinary size ball bearing were studied. In the experiments, the outer ring itself was struck in the radial or the axial direction. The sound generated by the strike was measured by a condenser microphone. Natural frequencies of the outer ring itself were determined from the frequency spectra of the sound. Moreover, some theoretical approaches were done about the measured natural frequencies. From the experiments, analysis and considerations, the natural vibrations on the outer ring itself of the ordinary size ball bearing were clarified, and it was found that the frequencies of the natural vibrations could be calculated by Kirkhope's theories and Love's theory.

View full abstract

In this paper, a new type of hydrostatic journal bearing is proposed to improve the stiffness by using a self-controlled restrictor. This restrictor makes it possible to easily obtain very high stiffness (nearly infinite stiffness) in a hydrostatic journal bearing by using the balance of of pressure acting on a floating disk. Here, the static characteristics of the proposed journal bearing employing the self-controlled restrictor is theoretically and experimentally investigated. Consequently, it is found that the proposed bearing can have a very high stiffness over a wide range of the imposed load irrespective of the load direction.

View full abstract

This analysis aims to determine the severest engine operating condition for deterioration of oil film lubrication around the crankpin journal, and to clarify the reason why minimum oil film thickness (M. O. F. T.) occurs during engine operation. As a result, that operating condition is experimentally found to raise the surface temperature of the crankpin at 7400 rpm and no/load or deceleration. According to the narrow bearing theory for dynamic loading, a decrease in M. O. F. T. is caused by the following factors: the first is a deformation of the connecting rod and crankshaft, the second is an increase in oil temperature and the third is an increase in the weight of reciprocating moving parts.

View full abstract

In the case of cast gears, the dimensions of various parts of gear bodies are mostly designed on the basis of empirical data. However, for plastic gears, such empirical design data are few: accordingly, even though the indicated transmitting power is the same, large difference is observed in the thickness of gear bodies and others among different to manufacturers. Therefore, in this study, as one procedure of examining the effect of the difference in the thickness of gear bodies such as these on the operation of the plastic gears for transmitting power, operation tests were carried out by using various thicknesses of disks and of rims in disk type plastic gears, and the relationships of the change of these thickness with the life, wear, change of tooth profile and the other characteristics of plastic gears were examined.

View full abstract

This paper presents a study on root stresses of thin-rimmed. spur gears with spokes. The stress analysis by the 2.dimensional finite element method (FEM) and static loading test for thin-rimmed spur gears with a fan-shaped hole were carried out, and the effects of the rim thickness and the rim length on the root stress and the position of the maximum root stress were examined. Root stresses of thin-rimmed spur gears with spokes were computed by FEM, and the relation between the root stress and the position of the loaded tooth, the effect of number of spokes on the root stress, the position of the maximum root stress and the stress change during one revolution were investigated. Furthermore the tooth deflection of a spoked gear was also examined.

View full abstract

The first paper described the manufacture of injection-molded plastic magnetic gears and static capacity and dynamic tests. This paper provides information on the manufacture of plastic magnetic gears (module 2) with the establishment of a new method for magnetic york and the results of static and dynamic tests. Those gears show a great deal of wear when the gears are used with no lubrication, but little wear when oil was used as lubricant. The shrink-fit plastic and plastic magnetic gears were made as the nonlubricating type. Those types showed a slight degree of wear. It was possible to put the injection-molded plastic magnetic gears to practical use in this experiment. Those gears had a flex density of about 0.03T and it was possible to provide strong magnetism by improving the magnetic york for better practical nonbacklash gears.

View full abstract

In order to study the wearing manner of the plastic helical gear composed of a crossed helical gear and a steel pair, the somewhat worn tooth profile is investigated theoretically on the assumption that the steel gear is not worn at all and that the shaft angle is 90° The trajectory of contact point is given numerically on the plastic tooth flank when it is not worn, and pairs of boundary curves are also given for several small worn depths defined on the pitch circle in the plane in which the common perpendicular is contained. The profile inside such a pair is the envelope of steel tooth flank moving relatively to the fixed plastic gear for named worn depth, but the outside is the original plastic tooth flank. The worn profiles are considerably different according to different helix angles, and they are represented three-dimensionally.

View full abstract

In order to develop a better method of predicting gear noise of automotive power transmission as a function of gear accuracy, the correlation between the amplitude of the tooth mesh frequency component of measured noise and that of the simulated single flank rolling test of gears using the measured form accuracy of the tooth flank was investigated. Experimental and simulated results show, when the worst composite error surface of meshing tooth flanks is chosen as representative among many different composite error surfaces for possible combinations of meshing tooth pairs and the simulation is worked out under the hypotheses, that this worst composite error surface exists at every meshing of teeth, a very good correlation between simulated results of the single flank rolling test of the pair of gears and measured noise level of power transmission can be obtained. This result indicates that the development of the gear production method to minimize the scattering of deviation of tooth flank accuracy at each tooth of a gear is more effective in producing noiseless power transmission than in improving the average quality of tooth flank accuracy of gears.

View full abstract

Gear vibration is induced by the excitation due to manufacturing/alignment errors and periodical chasnge of meshing tooth stiffness with progress of gear rotation. As an index for evaluating the degree of gear vibration and noise as functions of gear dimensions, accuracy and driving conditions, transmission error of gears and total vibrational excitation are widely accepted. This report clarifies that the influence of transmission error and that of total vibrational excitation on gear vibration are equal in their first order frequency component, and the difference between them is due to the higher order frequency components of the periodical change of meshing tooth stiffness. The difference in the degree of influence of these two indices on the relative rotational vibration of gears is discussed using numerical examples. Propriety between both indices is also discussed to predict the actual gear noise of automotive power transmission of personal cars in production.

View full abstract

Main test gears were mirror-like finished using a CNC gear grinder with a cubic-boron-nitride (CBN) wheel. Running tests were conducted using a new gear load testing machine with traction elements for loading test gears. Effectiveness of mirror-like grinding was shown by a comparison test conducted using conventionally ground and mirror-like ground gears with almost the same accuracy (0 class of JIS). Reduction in running noise achieved in the case of mirror-like ground gears may be ascribed to a reduction in rubbing noise due to formation of nearly full EHL oil film between contacting teeth. Appreciable reductions in gear noise and vibration were a chieved when a proper amount of profile modification was given to the mirror-like ground gears. The load sharing ratios of a pair of teeth and the rotational errors of the pair of the gears were precisely calculated using a computer. It was found that the best modification is a little greater than the elastic deformation of contacting teeth of test gears.

View full abstract

This paper presents a study on the bending fatigue strength of austempered ductile iron (ADI) gears. The bending fatigue tests for ADI, S 45 C, FCD 80, and SCr 420 case-hardened gears were carried out, and the bending fatigue limit loads of ADI gears were compared with those of other gears. The bending fatigue limit loads of the ADI gears used were found to be almost equal to those of case-hardened gears. Furthermore, shot peening was performed on ADI gears, and the effect of shot penning on the bending fatigue limit loads of ADI gears was investigated by carrying out the bending fatigue tests. The bending fatigue limit loads of ADI gears can be raised considerably by shot peening.

View full abstract

In order to obtain all individual errors which are defined in JIS, a master gear which is composed of two parts, is used for a single flank meshing test. Also, each tooth of the master gear has an effectual flank with a modified face width, which is geometrically arranged at a position in the direction of the gear axis. The experiment with the master gear was attempted for some test gears demonstrating characteristic errors. The first one has lead error and the second shows the modified curved, such as sine curve, in the meshing direction. The rest of the gears is eccentrically ground for pitch error between neighbouring teeth. Individual errors computed from the record of single flank meshing test between the master and the test gears give a good coincidence with the results from a conventional method, except for lead error.

View full abstract

This paper discusses the hobbing of gears made of austempered spheroidal graphite cast iron, which has recently attracted attention as a new material for gears. This iron has many advantages, such as the properties of casting parts of complicated forms easily and a high strength equivalent to that of steel, and is also an economical material. However, this cast iron is hobbed before austempering heat treatment at present, because the hobbing of the teeth of a hardened gear blank is considered to be very difficult. Yet, if the cutting of a hardened gear blank is possible, the cost will be lower because the gear can be used without performing a finishing operation on the tooth surface to remove the distortion caused by heat treatment. The following conclusions are obtained. (1) The life of the hob appears to be largely affected by the feed of the hob, and the life is longer at small feed. The influence of cutting speed is also large, as in the case of hobbed alloy steel. (2) It is observed that the teeth with greatest wear are concentrated in one area of the hob. (3) The accuracy of hobbed ADI is very good.

View full abstract

Satisfactory handling of cutter interference in NC machining of free-form surfaces requires not only avoidance of cutter interference, but also detection of all the interference regions on the surfaces. This paper describes an algorithmic method for obtaining cutter interference regions due to self-intersections of an offset surface. In this method, the offset surface between a ball-endmill and a machining surface is composed of an offset surface of the machining surface and envelopes defined by a reversed tool shape with Z-direction along the circumference of the machining surface. Self-intersection curves are calculated efficiently by using the Powell-Zangwill method and the Runge-Kutta-Gill method, and cutter interference regions are represented on a machining surface. The numerical result is accurate and CPU memory is saved because all offset surface data including the envelopes are exactly generated from the original (machining) surface without offset approximation.

View full abstract

To achieve the mirror surface finishing of the outer surface of a pipe, a new idea was applied to the ironing process. An aluminum pipe was set into a steel container with a very smooth surface. The inside of the pipe was ironed and the outside was flattened at high pressure with a small slip of the metal on the container surface. The experiment was carried out with various reductions and different viscosities of oil. Higher reductions and lower oil viscosity produce more excellent surfaces, and the finest surface has nearly the same topography as the inner surface of the container has.

View full abstract

It has been noted that the deflection of the end mill and workpiece produced by the cutting force is a main cause of surface error in the end milling process. The deflection, therefore surface error, becomes even worse in the case of inner cornering cut. To improve machining accuracy, a predictive analysis on the behavior of the cutting force is performed indirectly by means of chip area. Based on the analysis, cutting methods with decreased axial depth of cut or improved shape of the finishing surface in the preceding cut are proposed. Also constraint adaptive control is considered for the rapid variance in cutting force. Through experiments, the effectiveness of the cutting methods and adaptive control in decreasing the error in the corner is acertained.

View full abstract

KANBAN is known as a practical tool to control the inventory of materials and the timing of production. This paper presents a probabilistic model for the KANBAN system, considering such parameters as shipping frequency per day, traveling lead-time of KANBAN, consumption variation of materials and overtime. Some statistical analyses concerning inventory levels and order quantities of materials are carried out.

View full abstract