The finite difference method is applied to estimate the reliability of the single degree of freedom system subjected to white noise excitation. In this paper, the probability that the system response stays within a safe region is used as the system reliability. It is difficult to obtain the accurate reliability from the probability density of the system response. Then the boundary condition of the finite difference scheme which we presented in the previous paper for the probability density of the system response is modified to obtain reliability directly from the Fokker-Planck equation, which governs the probability density. The reliability of the Duffing oscillator is calculated in the illustrated examples. The validity of the present method is demonstrated by comparison with the results of the Monte Carlo simulation.

View full abstract

Chaotic responses are investigated for a post-buckled beam under the interaction between internal resonance and dynamic snap-through. The beam with variable cross section is clamped at both ends and the beam is axially compressed to a post-buckled configuration. The buckled beam is excited by a periodic acceleration. Applying the Galerkin method to the governing equation of the beam, nonlinear differential equations of a multi-degree-of-freedom system are reduced. Periodic solutions of steady-state responses are calculated by the harmonic balance method. In a typical frequency region, chaotic response bifurcates from the periodic response. Time progress of the chaotic motion is calculated by the numerical integration. The chaotic response is examined in detail by the Fourier spectrum, the Poincare section, the Lyapunov exponent and the Lyapunov dimension, respectively. Under the condition of the one-to-two internal resonance, the chaos is generated by a small amplitude of excitation. Two modes of the vibration induced in the chaos. Increasing the amplitude of excitation, the chaotic response transits to the complicated response coupled with the dynamic snap-through and the internal resonance. Induced modes in the chaos are counted as nearly four.

View full abstract

In this paper, an effective numerical method for the damping performance of an impact damper with granular materials is proposed. In this method, the effect of granularity on the dynamics of the impact damper with granular materials can be considered by using Discrete Element Method (DEM). It is shown that the damping performance of impact damper with granular materials depends on the following parameters: mass ratio, number of particles and clearance. The validity of this numerical method is examined by comparison with the experimental results.

View full abstract

In the previous paper, a global formulation method was proposed for representing the general response solution of piecewise linear oscillator system. In this paper, the method is expanded to further generalized systems such as those whose solutions are contracting with the boundary region or having trapping behaviour. First, after giving the definition of "degenerative" solution, it was made clear that only the acceleration can be discontinuous on the boundary region among the other differentials of those solutions. Secondly, by use of Filippov's definition for the expanded solutions, following specific characteristics regarding the solutions were obtained. (1) While "bounded leap" is keeping positive on the discontinuous point of restoring function, singular local trapping solutions appear which can not be given through ordinal linear solutions. (2) By regarding contact of the solution with the boundary region as non-transeverse crossing in wide sense, proposed global expressions are available even for the contact case. Finally, discussions are devoted to the development of this global formulation. And it leads to unified expression for both of trapping and non-trapping solutions whereby local trapping solutions within the crossing interval are replaced by local "constant" solutions.

View full abstract

Piston-type pneumatic vibrator is a device for generating mechanical oscillations by only pneumatic energy. This vibrator is used in parts feeder etc., because it has advantages of simple structure and high power. When applying a vibrator to the system, required frequency and amplitude must be realized. However the behavior of the vibrator is not formulated, so it is the difficult to decide the dimensions of vibrator for given conditions. The purpose of this study is to clarify the characteristics of the piston-type vibrator by modeling the nonlinear dynamics and by analyzing its lineralized model utilizing Fourier Series development. Consequently it is found that this phenomenon is a self-excited oscillation caused by the relay element. Equations relating frequency and amplitude to load mass, cylinder diameter and so on, are obtained. The calculated results from the model or its derived equations are in good agreement with experiments and are effective for the design of vibrator.

View full abstract

Nowadays, we have a lot of pipes underground for water service or gas supply. If leakage at these pipes occurs, it will not only come to huge economical loss, but also result in the cause of serious accident. Therefore, the leakage must be detected at fast as possible to be repaired. However, the identification of the leak location is difficult because the pipes are invisible. In the case of water leak, the detection has been relied upon the experts who have such skills that they can identify the leak location by hearing the noise of surface vibration on the ground. According to the lack of experts, however, more convenient and effective approach has been desired instead of the existing method. This paper describes the application of vibration intensity measurement for the detection of water leak location. The analytical model of the surface vibration on the ground is shown and the spatial distribution is estimated. Furthermore, the simulation of vibration intensity measurement is performed and the improvement in accuracy is proposed.

View full abstract

An FEM eigenvalue analysis program for such sheet winding products as CFRP golf club shafts is developed. A new cylindrical beam element is formulated considering the coupling effect between tension and torsion for anisotropic materials. In order to verify the coupling effect of the presented method, composite cylindrical cantilevers are analyzed. And the non dimensional torsion by the axial tension of the beam element model is compared with a shell element model. And the influence of coupling effect on the natural frequencies of CFRP golf club shafts is evaluated for various fiber angles by eigenvalue analyses.

View full abstract

The floating behavior of a flexible belt on a steel trough by an air flow is analyzed to investigate the stable operation for a belt conveyer of air suspended type. The deformation behavior of a belt element is described by the curved beam theory and air flow is described by the Bernoulli equation including flow loss effect. These equations form a coupled system of equations of the structure (flexible belt) and the fluid (air flow). These equations are numerically solved by an iteration method based on the Newton-Raphson method. Experimental investigations are simultaneously conducted using the test equipment specially prepared. A good agreement between analytical and experimental results is obtained.

View full abstract

A design method of hybrid piezoelectric damping for flexible structures is proposed in order to achieve the equal vibration suppression performance with less active control power requirement. Values of design parameters of hybrid damping: active control gain, number and location of piezo-elements to be used to passive and active dampings, are designed under equal vibration suppression performance condition. Vibration suppression performance is evaluated by using the gain of the accelerance transfer function of the closed loop system in modal space. Availability of the proposed design method is experimentally demonstrated by using a flexible cantilever beam. The experimental results indicate that hybrid damping is effective comparing with purely active damping from the viewpoint of active control power requirement.

View full abstract

Noise from rotating tire is studied by not only running on a road or drum test but also numerical simulation, however it is not clear to find the cause of noise generating mechanism and take prompt measures. In this research, a vibration deformation is predicted by sound radiation from a rotating tire under constant speed and the noise reduction is carried out by structural modification. First, the contribution of each vibration mode shape is estimated by using model analysis and boundary element analysis and verified by numerical simulation. Next. the proposed method is applied to a rotating tire to predict the vibration deformation at some frequencies of the noise problem. The advantage of this method is as non-contact measurement and to obtain the contribution of each vibration mode shape. Finally, the highest contribution of the vibration mode shape is focused for structural modification to reduce the noise of the rotating tire.

View full abstract

The wheel climb derailment sometimes occurs when a train passes a steep curvature with low speed. To investigate this phenomenon, an experimentation using a model track and model trucks was done by the authors. The "Nadal Limit" has been used to evaluate the possibility of derailments. However, this Limit is so much changed by the friction co-efficient between rails and wheels that the Limit does not necessarily indicate the possibility of the derailment. The real possibility of derailments is shown by the wheel lift-up value from the rail. To measure this value exactly, the authors proposed a new measuring device upon a new conception which adopts a bundled optical fibers. They tried to measure the adhesion co-efficient instead of friction co-efficient between rails and wheels introducing the Slipping-adhesion Model Truck. As a result, they found that a wheel climb phenomenon sometimes causes a wheel slip-down phenomenon and that a wheel climb derailment occurs when the wheel climb-up value excesses wheel slip-down value.

View full abstract

This paper examines the ability of the vehicle equipped with sensor steering mechanism (SSM) by utilizing a simulated trajectory in comparison to experimental results. Dynamical equations of the four-wheeled vehicle with front and all wheel steering are derived. The program developed the trajectory of the vehicle moving on the flat oval surface in order to conform these maneuvers. This experiment tests the vehicle's capacity to automatically following in guide way alignment. Numerical and experimental results show the effectiveness of SSM in programing these maneuvers.

View full abstract

We reported in the previous report about a three-dimensional simulation method for the vibration of bogie-track system of railways, and showed that the frictional vibrations of torsional mode are generated in the rail on curved track with 160 m radius. We simulated for the case of 84 m radius with same method, and the torsional vibrations are generated in the rear axle of the bogie in this case. Considering that several modes exist for the frictional vibrations of the system, we made a simplified dynamic model to check the stability of the vibration system for the given slip conditions at the tread. The calculation results for bogies and tracks of two types are in accord with the simulation and measured results.

View full abstract

According to the decrease of the three-axle bogie for locomotives, the need for diminution of the gauge widening in curved track has been pointed out by rail maintenance division. Meanwhile, bolsterless bogies which have no side bearer are employed in many railway vehicles recently, and it is expected that various steering bogies will be employed in the near future. For vehicles which have these bogies, the gauge widening keeps its effect on curve negotiation even now. In this study, the dynamical analysis is carried out about the relationship between the gauge widening and the track load, considering the bogie mechanism. Numerical simulation is used for the analysis, and the effectiveness of gauge widening is summarized from dynamical viewpoint of two-axle bogie vehicles. Further, wear indices are introduced, and the wear tendency of flange and tread is estimated quantitatively.

View full abstract

In this study, a method for the regularization of the acoustic-vibratory inverse analysis has been proposed. The method is that the satisfactory fundamental solution in BEM is selected according to the problem to be solved for the regularization. The adequate condition for solving the acoustic-vibratory inverse problem is found by using the objective function which are the condition number of the coefficient matrix and the error norm caused by the rank reduction of the matrix. In a numerical example, the adequate condition can be found to the identified problem of the vibration condition of sound source in enclosed space. It was seen that the adequate condition obtained by the proposed method is more adequate than the one obtained by the conventional method from the viewpoint of the objective function. The proposed method is, therefore, proved to be very effective for the acoustic inverse analysis.

View full abstract

In the present paper, it is described to apply electro-rheological fluid (ERF) to a weight gyrostabilizer (WGS). By using ERF to control the motion of weights on WGS rotor and imposing the electric field no them, it is expected to decrease the restoring force of the WGS instantaneously with on change of rotational speed of the gyro. Therefore this gyrostabilizer (ER-WGS) can protect the structure from its fracture due to the excessive impulsive forces acting on it. The restoring force of the ER-WGS for the impulsive external forces is investigated experimentally and theoretically when the electric field is imposed or not, and those results are compared.

View full abstract

This paper proposes a new method to identify concurrently unbalance and sensor runout on a rigid rotor supported by magnetic bearings. At first, the paper investigates the effectiveness of the incremental least square online method applying to identify unbalance and sensor runout by a numerical simulation. Secondly, the paper also presents results of the numerical simulation on identification accuracy at various speeds and effects of the rotor model error. Inconclusion, we are confirming that the new identification method is effective by comparing the simulation results on the various speed and influence of the motor model errors.

View full abstract

Measurement error of a hot wire air flow meter due to heat conduction loss through probe supports was analyzed and a method for error reduction was proposed. The air flow meter uses constant temperature difference principle, which is composed of a hot wire probe and an air temperature probe. When both of the probe support are heated, the air temperature probe and the hot wire probe are influenced by heat conduction through the supports. Temperature of the air temperature probe becomes higher than that of air flow and heat quantity which conducts from the hot wire probe to the wall decreases. These two points lead to the measurement error of the hot wire air flow meter. But the error can be reduced, when size and material of two probes and these supports are selected appropriately. In practical use, when the same probes are used for the hot wire and air temperature probes, the error due to heat conduction loss can be reduced. This effect is described by experimentally and analytically.

View full abstract

Pneumatic motors have not so far been well-studied in the robotics literature, despite their many advantages. In this paper, we develop a nonlinear, time-variant model of the dynamics of a vane-type pneumatic motor for purposes of controller design. We develop two models: one considers the pressure dynamic of supply ports only (Approximate model) and the other considers pressure dynamics of all chambers (Precise model). It is shown that unlike in the case of pneumatic cylinders-which have been studied extensively-the dynamics of both pressure variations and mechanical dynamics are complex. We apply sliding mode controller design using pressure difference in place of acceleration for motion control of the actuator. We compare control performance of the sliding mode controller using the two models. As a result, it is revealed that both the models are precise, and robust to payload variations, but the approximate model is simpler to implement. Results of experimental implementation illustrate the effectiveness of the proposed sliding mode controller.

View full abstract

To recognize an environment, the robot should have as many sensors as possible. When we use sensors, we must consider the characteristics of the sensors, such as range, processing time, error, and so on. In this paper, we pick up ultrasonic wave sensor that is today the most common technique employed on indoor mobile robotic systems, and we propose a new technique to estimate the smoothed value and the differential value of the distance measured by a ultrasonic wave sensor. When we propose this system, we take these restrictions above mentioned into consideration In spite of many methods are proposed, it is very difficult to eliminate the noise of sonar perfectly. Therefore, we smooth the distance value by supposing the continuity of the signal which is obtained by the sonar, and take advantage of this continuity, we compose an estimator. The estimator is based on the sliding mode system, we can realize the robust estimator.

View full abstract

The contraction and relaxation of muscle in the vicinity of the skin play an important role to keep the internal temperature constant. In this paper, we propose a new temperature control system modeled after the function of skin. The characteristic of this system is that the precise control of temperature is achieved by the control of flow rate distribution. Moreover, we used the digital adaptive control to improve the performance and flexibility of this system. To verify the validity of this system, the experiments were carried out and it was shown that this system can control the internal temperature within the range of ±0.02°C.

View full abstract

This study applies differential game theory to simplified four-wheeled vehicle models. We describe the motions of two four-wheeled vehicles on plane as nonlinear systems and formulate receding-horizon differential game problems. Performance indexes of the differential games have a moving performance interval. The game optimal solution (minimax solution) is calculated by applying a real-time algorithm without successive approximation. Numerical results of pursuit-evasion and overtaking are reported and discussed.

View full abstract

This paper deals with a flexible transfer system (FTS) in self-organizing manufacturing system (SOMS), SOMS is composed of modules that self-organize with information from other modules and environment. Each module makes outputs through the interaction with other modules. FTS is composed of a number of conveyer units and machining centers. This system contains 3 processes of tool locating, job scheduling, and path planning. In this paper, we focus on path planning of the FTS. There are two stages of decision making: preplanning by global control and replanning by local control. A steady-stated genetic algorithm and stochastic learning automaton are applied to the preplanning and to the replanning, respectively. Furthermore, we propose an automatic control method of the providing interval of materials using simplified fuzzy inference in order to improve the performance of this manufacturing system. Finally, this paper discusses the effectiveness through several computer simulations.

View full abstract

This paper presents a "Kansei" model for musical timbre using a neural network. The three-layered network, which is constructed based on the structure of auditory sense, consists of seven input nodes, four neurons in the hidden layer, and three neurons as the output nodes. The input is a pattern of tone pressure levels for harmonics and the output is a three-dimensional vector of the Kansei information that is characterized by "clearness-unclearness", "thickness-thinness" and "sharpness-softness". After training the network, the analysis of the synaptic weights of all the neurons gives the following results that agree well with the knowledge of psychology and physiology. 1) Each harmonic equally influences the "thickness-thinness". 2) The higher harmonics give a contrary effect to the "clearness-unclearness". 3) Two neurons with lateral inhibition appear in the hidden layer and extract the 6 th and 7 th harmonics. Another neuron in the hidden layer extracts the 5th, 6th and 7th harmonics. These three neurons determine the "sharpness-softness".

View full abstract

This paper presents a visual feedback control scheme for nonholonomic wheeled robots. The visual servo controller guides the robot to the goal by observing the feature points in the environment. The feature points are controlled directly in the image plane by using an image-based controller. Thus the possibility of missing the feature points is reduced considerably. Simulations are carried out to evaluate the validity of the proposed scheme and experiments on a radio controlled car with a CCD camera are also given.

View full abstract

We have developed an objective lens actuator for high-speed CD-ROM drives. It has high acceleration sensitivity and good vibration characteristics so that it can follow a rapidly rotating disc. We produced a novel actuator structure which contains an inner yoke with four planes. This increases the acceleration sensitivity and stiffness of the actuator. We measured the acceleration sensitivity to be 140 m/s²/V in the AF (auto focus) direction and 90 m/s²/V in the TR (tracking) direction. The higher-order resonant frequency is over 30 kHz in both directions. These values are high enough for 40 X or faster CD-ROM drives.

View full abstract

This paper proposes an algorithm to determine sub-optimal paths and trajectories for cooperative multiple manipulators with passive joint. The problem of operating and transferring the object from a given start point to a given goal point is considered. In this case, in general, degrees of freedom of the manipulators are larger than that of the object. Then this manipulator system involves redundancy. It does not make big effect for the mobility of the object not to consider several redundant degrees of freedom. So several joints can be changed to passive ones. We use remaining redundant degrees of freedom to refine trajectories. We apply this algorithm to a two-manipulator system. Simulation results show the effectiveness of the algorithm.

View full abstract

This paper proposes a system, which realizing a collective autonomous behavior such as an autonomous conveyance order formation in the AGV (Auto Guided Vehicle) transportation system. We attempt to deal with a large scale distributed autonomous system in dynamic environment feasibly. However, as far as we use a global evaluation function in order to control each agent, it is necessary that we must rewrite the global evaluation function of the system whenever the environment changes. If we use such a method, the system can not be called the real distributed autonomous system. In this paper, to realize the dynamically reconfigurable formation in the dynamic environment, we propose two ideas, the learning based on the agent's own action and the interaction with the other agents by the relative evaluation. By use of these ideas, it is shown that the dynamically reconfigurable formation emerges as the autonomous conveyance order formation of AGV transportation system in the dynamic environment.

View full abstract

In order to detect a license plate on moving vehicle, it becomes important, different from stopping vehicle to consider its size on the image. So, we have developed a new method to detect it utilizing information of character's row up to now. But in the pre-proposed method, there is a problem that reliability is low if the large figure on the license plate is only one, and there are some cases that we can't decide license plate position estimated from detected characters if detected figure is less than two. In this paper, we illustrate a stable method to detect it by improving the evaluate function.

View full abstract

A structural optimization method is proposed for designing energy supply systems in consideration of their multi-period operation. The structural optimization problem is formulated as a large-scale mixed-integer linear programming (MILP) problem with binary variables for selection and on/off status of equipment, and continuous variables for capacities and energy flow rates of equipment. The dependence of performance characteristics of equipment on their capacities and loads as well as that of capital costs of equipment on their capacities are incorporated into the optimization model. A decomposition method for solving MILP problems with the block angular structure is applied to carry out the optimization calculation efficiently. Through a numerical case study, the validity and effectiveness of the structural optimization method proposed here is clarified in terms of solution optimality and computation time.

View full abstract

A Predicting method of tooth root stress of hypoid gears is proposed. The deformation of tooth and rim of hypoid gears calculated by FEA is compared with that of cylindrical involute gears, and the approximation equation for the compliance of cylindrical involute gear tooth is modified to that of hypoid gear tooth. Then considering the FEA calculated tooth root stress of hypoid gears, the approximation equation of tooth root stress for hypoid gears is derived by modifying the approximation equation of cylindrical involute gears. The contact pattern, normal load and tooth root stress of a hypoid gear set for 3 different patterns of tooth flank contact marking are calculated by the prediction method proposed in this paper. It is shown that the tooth root stress calculated by the proposed method matches well with the value calculated by FEA.

View full abstract

The prediction method of the tooth root stress of hypoid gears proposed in the 1 st report is evaluated by the measured values. The state of tooth root stress of three different tooth contact patterns, i.e. for toe, center and heel contact pattern setting, under different loading are compared and the accuracy of the proposed approximating prediction method for tooth root stress of hypoid gears is discussed. It is shown that the tooth root stress calculated by this prediction method matches well with the value measured under heavy loading. Slight deviation under light loading is supposed to be caused by the insufficient accuracy of the measured tooth flank form deviation and of the load depending characteristics of contact compliance incorporated.

View full abstract

Fatigue failure of synchronous belts is investigated under fluctuating torque conditions, because the driving torque fluctuates in the automotive engine camshaft drive-systems. There are two types of fatigue-failure modes. The first mode is de-bonding along the interfaces between the stranded glass filaments in cords. The initiation of the de-bonding is explained by the cyclic bending deformation when the belt is wound around pulleys and transmits the power. A simplified mechanical model is proposed to explain the failure mechanism. The second mode is crack initiation between the cords and the facing fabrics. Further investigation is necessary to reveal the second failure mode.

View full abstract

It is an important subject that the tribological properties are investigated for the combination of the polymer composites from the standpoint of the lighter weight, the miniaturization and the low cost of the machine elements. In this study, the friction and wear properties are investigated for the combination state of pitch carbon fiber reinforced PPS (Polyphenylene Sulfide) composite materials. This tribological test under the rolling-sliding and the light load condition is carried out for the pure PPS resin and three types of PPS matrix composites, which short pitch carbon fibers are varied to 15, 30, 45 wt.%. It becomes clear that the transfer film is formed on various combination conditions of PPS composites and the decrease of frictional coefficient is caused by this transfer film. Also, this transfer film is formed on the surface of the test piece of higher fiber content for all test conditions except the condition of the no filler test piece. And, in the case of the combination conditions of CF/PPS composites, the friction and wear properties are not improved by increasing the higher carbon fiber content. As a result, the better tribological properties are showed on the case of the combination of 15% carbon fiber filled PPS composite and 45% one.

View full abstract

The effect of the deformation of connecting rod and the piezoviscosity of the lubricant on the performance of bearing are studied in a high speed loading condition of an actual engine, using an elastohydrodynamic lubrication (EHL) analysis. As the result, it was confirmed that the EHL analysis is important for investigating piezoviscous effect on friction loss in such condition. And It was found that, in such heavy load condition, the stiffness of the housing affects the friction loss because of the piezoviscosity, and the piezoviscosity is important for bearing performance as well as the viscosity.

View full abstract

In order to examine the behavior of planetary traction drive when the loading torque varies rapidly, experiments were performed. From the experiments and discussions, following conclusions were obtained. (1) Because of the higher contact pressure between rollers, slip ratio between ring roller and planetary roller increase first when the loading torque is over the transmissible torque of the planetary traction drive. (2) If the slip ratio between ring and planetary roller became high, traction coefficient between ring and planetary roller increased and slip occurred between sun roller and planetary roller. At last severe wear and siezure occurred. (3) If the temperature of the surface of roller reached about 160°C by frictional heating, traction coefficient started increasing. Similar phenomenon was found in the four ball test using hot traction oil.

View full abstract

A new tightening method for the rotary shouldered connection (RSC) has been developed. In the developed method, the coefficients of friction between mating thread surfaces and between mating shoulder surfaces can be accurately and separately detected by sinultaneous loading of torque and compressive force to RSC. The tightening experiments have been achieved to verify the serviceability of the developed method. The 1/4 scale model of NC50 are used as the test model of RSC. The scatter band of axial force on RSC has been reduced within the error of ±10% under lubricated conditions by grease with MoS₂. The reliability of the method is compared with Baryshnikov's method.

View full abstract

In injection molding of short fiber-polymer melt mixture, influence of the interaction between a fiber and a mold wall on fiber distribution and flow property cannot be neglected, when fiber length is not very short comparatively to the thickness of a mold cavity. In the flow through a cooled mold cavity, solidifying layer grows and it affects that relation further. In this paper, effiects of these causes on fiber orientation and velocity distribution have been made clear by using numerical analysis, in which each fiber motion is analyzed from equilibrium conditions and fiber orientation distribution is obtained by integrating frequency of orientation angles of each fiber over the whole angles, and the following results have been obtained. (i) Near a mold wall, mixture is cooled and no flow layer is generated. On the other hand, apparent viscosity increases in the region, apart from a mold wall by more than half of fiber length, because of high fiber concentration, and plug flow-like velocity distribution is generated. (ii) Pressure gradient increases toward downstream in low injection rate, because of cooling of mixture, whereas it decreases in high injection rate, because of viscosity decrease due to shear heating. (iii) As fiber length decreases, pressure gradient decreases. (iv) Fiber orientation distribution changes considerably with injection rate change.

View full abstract

This paper presents a study on effects of coil configuration on residual stresses and hardened layers of shafts due to the stationary induction hardening. Amounts of heat due to the eddy current induced in the shaft were calculated for various coil configurations by the FEM electromagnetic field analysis method, and the effects of the heating condition (electric power and frequency) and the coil configuration on the amount of heat were examined. An electromagnetic field analysis, a heat conduction analysis, and an elastic-plastic stress analysis during the induction hardening process of the shaft were carried out for various heating conditions and coil configurations by the FEM, considering changes of the magnetic permeability, the resistivity, the thermal expansion coefficient and the yield stress with the temperature. The effects of the heating condition and the coil configuration on the temperature and the stress changes during the induction hardening process, the residual stress and the hardness distribution were determined. On the basis of these calculated results, a selecting method of optimum heating condition and coil length for the residual stress and the hardened layer was indicated.

View full abstract

This paper presents a study on residual stress and hardened layer of shouldered shafts due to dual frequency induction hardening (using both high- and low-frequency heat sources). Amounts of heat due to eddy current induced in the shouldered shaft were calculated by the FEM electromagnetic field analysis method, and the effect of the frequency on the amounts of heat was examined. An electromagnetic field analysis, a heat conduction analysis, and an elastic-plastic stress analysis during the single and the dual frequency induction hardening process of the shouldered shaft were carried out by the FEM, considering changes of the magnetic permeability, the resistivity, the thermal expansion coefficient and the yield stress with the temperature. The effects of the induction hardening method on the temperature and the stress changes during the induction hardening process, the residual stress and the hardened layer were determined. The validity of the dual frequency induction hardening method for the shouldered shaft was confirmed. Furthermore, a method of determining optimum coil configuration and hardening condition was indicated.

View full abstract

Spin angle controlled ball lapping method developed recently, is to have an excellent performance to lap the ball, in which V-groove lapping plates are separated into two parts and three plates are able to rotate independently. Present paper focuses on an effect of degradation of grain size on finishing characteristics of a steel ball. It is found that degradation of grain size has yielded greater stock removal and has brought smaller sphericity of the ball. Experiments show that the stock removal is dependent upon spin angle θ which is closely related to slipping distance of the ball. Measurements of the sphericity of lapped ball indicate that geometry of lapping plate has small effect with respect to improvement of it, but causes larger stock removal rate and very smooth surface.

View full abstract

In this paper, the authors use the concepts of a 'virtual process'and a 'real process', and use a 'constraint blackboard' for CAPP for non-axisymmetric hot-forging products. A virtual process is a primitive and basic forming process, which may exist only in the mind of a process designer. A real process is a complex process havig virtual processes as its elements. When the designer designs a forging process, he guesses the transitional pre-form shape in each successive virtual process, and finally assembles the virtual processes into one real process. These virtual processes restrict each other in compressing direction or parting line, deformability, and ability of combination and works cooperatively when they are assembled. The authors researched such virtual processes and their constraints, entered them as 'processing cases data' into a database called the 'processing case base', and implemented an expert system for connecting rod products. A 'constraint blackboard' is a common working memory which can be accessed by the procedures described in every processing case. The system starts design from the final shape and works back toward the raw material. It establishes a manufacturing plan at each virtual process. Once a virtual process starts to be designed, some procedures on the processing case are executed and new constraint is written on the constraint blackboard. When the other virtual process is designed, the system examines the process by referring to the constraint blackboard. Thus, the design process of each virtual process is done cooperatively to others. The system continues to assemble virtual processes into one real process until it fails to find a virtual process which is consistent with past fixed virtual processes.

View full abstract