This investigation treats the damped free vibrations of a double-cantilever beam with an adhesively bonded lap joint. The adhesive is considered to be a viscoelastic material, the modulus of which is assumed to be small compared is that of the elastic beam. The frequency equation is derived and the effects of the thickness, length and stiffness of a viscoelastic material on the damped natural frequency and logarithmic decrement of the system are discussed. Experimental data for damped natural frequencies and loss factors are presented for comparison with theoretical predictions.

View full abstract

In the previous papers, the authors proposed a numerical approach for analyzing the vibration problem of combined systems. For the free vibration analysis, the approach consists of two steps. One is estimating Green's function for a static bending problem of the same plate and the other is to solve a frequency equation. First, this approach is applied to a symmetrically laminated FRP (fiber reinforced plastic) composite plate. Numerical calculations are carried out for a cantilevered plate and a clamped plate. Green's function of a cantilevered plate is estimated by the Ritz method and one of the clamped plate is obtained by Galerkin's method. Accuracy and convergency of natural frequencies of these plates are discussed. Next, it is an attempt to reduce the number of dimensions of the frequency equation, by assuming that the deflections of an adjacent 5 points in a straight line can be approximated as a parabolic function. This approximation reduces the number of dimensions of the frequency equation and the number of Green's functions. The improved approach was applied to a cantilevered plate. Numerical results showed that the computing time was reduced extensively by the approximation.

View full abstract

Concerning the passage problem through a critical speed, it must be taken into account that the acceleration reduces the damping coefficient for a forward whirl whether the rotor is symmetric or not. This phenomenon may be in opposition to the conventional conclusion that the maximum amplitude decreases with the increase in acceleration. The interaction of a driving torque with a change in the amplitude is an occurrence in the second order of the shaft deformation. Thus, equations of motion for the rotating shaft are introduced by operations ensuring the second-order accuracy of the deflection, the angle of deflection, and the torsion. An asymptotic solution is derived from the equations of motion, provides some clues to the physical explanation for the transient behavior of the rotating shaft.

View full abstract

A self-induced manometer oscillation in a flow channel connecting two large tanks is observed, which is caused by the circulating flow in the upstream tank. Water flows into the thin rectangular tank horizontally and flows out vertically through the outlet duct located at the bottom center. The duct is connected to the large overflow tank directly. At a certain water level and flow rate, the free surface of the test section oscillates upward and downward periodically, causing the oscillation of the outlet flow rate. The period of the oscillation is found to be that of the manometer oscillation without flow. Consequently, the oscillation occurs in the multi-free-surface system, i.e., the test section surface and the overflow tank surface. The oscillation growth model based on the flow pattern transformation is proposed.

View full abstract

This paper investigates the effects of bolted joints on the vibration characteristics of the hydrostatic air bearing spindle rotor by using the Finite Element Method and experimental modal analysis. The test rotor is composed of a journal part and two thrust plates bound with bolts. First, it is shown that the measured natural frequencies increase and modal damping ratios decrease as the clamping torques of each bolt increase. The calculated natural frequencies of a modeled rotor without bolted joints are shown to be always larger than the measured ones. Next, it is found that this discrepancy can be eliminated if the joint layer between the journal part and thrust plates are modeled as distributed normal springs with stiffness K_z and tangenial springs with stiffness K_r, which are identified by the two experimental natural frequencies. The modal dampings mainly depend on the normal spring, and increase in inverse proportion to K_z.

View full abstract

In this paper, the heaving and pitching vibration of a rail vehicle with an air spring and a flexible body has been analyzed in order to obtain the effects of the air spring and the rigidity of the car body on the vehicle vibration. A harmonic balance method and a Newton-Raphson method are used to analyze the nonlinear system. The results show that the nonlinearity of the air spring makes the response-per-unit exciting force small in the case of the in-phase mode and large in the case of the out of phase mode, and the damping of the air spring appears only in a rigid mode.

View full abstract

The authors propose and develop a new experimental vibration analysis method named modal energy analysis method in this study. Is this first report, the modal energy analysis method is proposed and its basic theory is explained. By using the proposed method, the individual loss factors of all components can be identified practically at any vibratory frequency, in the vicinity of resonant frequences, from experimentally measured FRFs of objective total structures. Two simulative and an actual experimentally based applications are carried out and their results are shown in order to verify the theory.

View full abstract

This paper describes the stability analysis of a multistage centrifugal pump in boiler feed. The individual contribution of hydraulic impeller forces, noncontact seals and bearings to rotordynamic behavior are evaluated according to logarithmic decrements. The influence of bearing and seal shapes on the rotor stability is investigated. Also, the vibration of actual multistage pumps is measured corresponding to the analysis. Theoretical stability threshold speeds are compared with experiments and are found to be compatible. As a result, it was confirmed that the rotor stability is strongly affected by the characteristics of the seal type.

View full abstract

A new and robust scheme for the eigenvalue analysis of the Helmholtz differential equation by the boundary element method (BEM) is developed in this paper. Unlike the existing methods using highly complicated transcendental equation including unknown wavenumbers, the present method can reduce computational tasks greatly with the help of the Multiple Reciprocity Boundary Element formulation in terms of the fundamental solution for the Laplace equation and related simple calculation for polynominals. The Newton method is employed for determination of the desired eigenvalues. Two-dimensional problems with various homogenous boundary conditions are solved to show the versatility of the proposed scheme.

View full abstract

It is generally recognized that the gravicorder is less accurate when subjects are light. As the position of gravity center is calculated from a equilibrium among forces to supporting columns, the forces must be transformed to electrical signals. This process probably causes errors of measurement. In this paper, the relationship between given positions where a load was placed and calculated positions was mathematically analyzed first. The result showed that the given point can be presented by a linear model using calculated coordinates. Following this, weight ranging from 98 N to 321.4 N was placed on 4 different locations. Parameters corresponding to each weight were determined using a method of least squares. Then the calculated coordinates were corrected using these parameters. Finally, These correction algorithm made it possible to reduce the average absolute error from mm order to 1/10 mm order.

View full abstract

In the previous paper, we developed equations of motion for analyzing the dynamic behavior of crawler-type vehicles during straight propelling on soft ground and firm ground. In this paper, the experimental tests were carried out in order to verify the validity of the simulation model, The power shovel, a typical example of the crawler-type vehicles is used in the tests. We measured the dynamic characteristic of the soil for the characteristic of the settlement and the shear that considered hysteresis. The results of experimental tests are in good agreement with the theoretical results. The dynamic behavior of the vehicles, the dynamic ground pressure and so on, are determined.

View full abstract

The objective of this paper is to solve the minimum time control problem of a one-degree-of-freedom positioning mechanism whose actuator is restricted by the average heat generation. We found that if we use the conventional quadratic cost functions, i.e., total heat generation in one motion, we can't always get the solution when the gravity load is involved. To avoid this difficulty, we proposed a new modified cost function including a term of average heat generation. It is shown that the proposed method derives the minimum time solution even if the gravity affects the mechanism. A closed-form solution under average heat generation is derived, and its calculated examples are illustrated. In order to extensively apply this modified cost function to nonlinear multi-degrees-of-freedom mechanisms, a new numerical analysis method using a 5th-order spline function is proposed.

View full abstract

Small sliding in contact points of fingertips is investigated in this paper. The grasp-stabilizing method with internal stiffness is proposed and its stabilizing effect is shown for a planar three-fingered hand. The grasp-stabilizing conditions by the internal stiffness are shown, and an optimal grasp which is different from a conventional point of view is discussed. Finally the stabilizint effects by the internal stiffness are illustrated by numerical examples.

View full abstract

A new dynamic absorber system using lever and pendulum mechanisms was proposed for controlling the displacement of high-rise structures, and its optimum tuning condition was determined by the uncoupled frequency ratio of mass of the system to main mass and critical damping ratio. However, the decided frequency and damping ratio are a function of four parameters, such as three mass ratios and a lever ratio. In this report, the determination of these parameters is treated as a nonlinear minimization problem having constrained and is solved by using the SUMT method. Finally, the optimum combination of the parameters in this system is obtained, and the characteristics of this system are discussed in physical meaning.

View full abstract

This paper proposes the new dynamic absorber using viscous and elastic materials (GEL) to prevent numbness in the hands when batting with a metal bat. First, in order to obtain data for designing the dynamic absorber, the vibration characteristic of GEL is examined experimentally because there is no information on the spring factor or damping factor of GEL. This dynamic absorber, which we propose in this research, is formed from an absorber mass and a selected optimum GEL. Therefore, this dynamic absorber is compact, light, and easily made. It can be mounted in the grip end of a metal bat. The effectiveness and usefulness of this device in reducing the numbness in the hands have been demonstrated experimentally and theoretically

View full abstract

For vibration control of a flexible rotor suspended by active electromagnetic bearings, the PID control is widely used. But the capability of this controller for improving damping properties of the mechanical system is not so good because high order bending modes often become unstable increasing the derivative feedback gain. In this paper the feedback control scheme which uses the external forces estimated by an inverse system is presented. Simulating forced vibrations, it is found out that without the large derivative gain this controller enables to reduce resonance vibrations of a flexible rotor in case of random external forces being operated to the mechanical system. In the experimental system applied above feedback control, measured values of forced vibrations agree with numerical results.

View full abstract

This paper proposes a new speed control method for hydraulic elevators to save on operating energy and improve the smoothness of changes in car speed. The car is operated by varying the speed of a hydraulic pump driven by a speed controlled induction motor. The leakage flow of the pump depends on the number of passengers and oil temperature, and induces a difference between command and actual car speed. To improve the performance of speed control and suppress the vibration of the car, a noncolocated feedback control method is employed. The dynamics of a hydraulic elevator are investigated to design a controller of sufficient robustness. The controller designed is verified by numerical simulations and the results indicate the good performance of the developed hydraulic elevator system.

View full abstract

This paper is concerned with the optimal trajectory control of an intelligent truss structure composed of a truss-type mechanism. The intelligent adaptive structure can change its geometric configuration as well as physical properties in order to meet mission requirements. Many variable-length members are incorporated in the intelligent adaptive structure, resulting in a highly redundant system. Thus, the intelligent structure which is used as a manipulator arm can utilize the redundancy for the optimization of trajectory control. Efficient control methods are proposed for the trajectory control optimizing a performance index by using the redundancy. The validity of the proposed methods is demonstrated through numerical simulations. An adequate performance index enables the intelligent adaptive structure to be an appropriate configuration.

View full abstract

The extracting and building-block approach, or E-BBA, using mode synthesis was developed to predict the modified vibration characteristics of a global structure when substructures are modified. This prediction is performed by adding the difference in mass and stiffness of the modified substructures obtained by FEM to the global structure modal parameters obtained experimentally before modification. Also, in order to obtain the compatibility of DOF from the test (usually 3 DOF) and the EFM, 6 DOF of FEM is reduced to 3 DOF by Guyan reduction. Since this method reduces the effects by joint stiffness and residual term, and eliminates the measurement of the rotational DOF, the prediction is more precise than that by the conventional method. Additionally, this minimizes the manpower required for the test and analysis.

View full abstract

A new type of mass damper was developed to implement active vibration control for tall buildings at cheaper costs than normal active mass dampers. The mass damper has two modes of operation, an active mode and a passive mode. It works as an active mass damper to control relatively small vibrations of a building caused by winds and weak earthquakes. At the same time, it works as a passive mass damper against strong earthquake excitations. For the mass damper, a new type of hydraulic actuator was developed which could change over from a normal actuator to an oil damper for the passive mode, and change over in the opposite direction for the active mode also by switching a flow control valve of the hydraulic circuit of the actuator. Excitation tests were carried out by using a experimental model of the mass damper with a 72kg moving mass and a 5-story building model with a total mass of 5 500kg and a height of 3.4m. The tests showed that the mass damper was effective not only to save the costs, but also to achieve high performance of vibration control by the actuator of a small capacity.

View full abstract

This paper describes finite element analysis of electromagnetic fields and optimization of eddy current densities. First, we derive the state equations by using finite element method. Eddy current densities are represented by derivatives of the state variable x(t). A linear quadratic performance index is introduced so that eddy current densities will track desired functions. We obtain an optimal control law from the stationary conditions of the performance function. Optimal inputs are expressed by the solutions of two differential equations. Finally, we show numerical examples.

View full abstract

This paper proposes an actuator consisting of a piezoelectric element, two electromagnets and a spring for controlling displacement of flexible beam with large deflections. The voltage pattern for the piezoelectric element and the electric current patterns for the magnets and the relationship between the actuator displacement and the deflection of a beam are presented. The experimental tests are also caried out. It is ascertained that the proposed actuator has some advantages as compared with the general deflection control apparatus using a motor.

View full abstract

The positioning mechanism of flexible disk drive (FDD) generally consists of a step motor and a lead screw. The rotational fluctuation of the step motor excites the vibration with impact at the clearance of the lead screw, and the transfer force of the impact to the frame gives rise to the noise. Firstly, we analytically and experimentally investigate the carriage motion in the positioning mechanism to clarify that there are two types of impact vibration, in which the worse type can be eliminated by choosing the appropriate configuration of the carriage. Secondly, we compensate the excitation timing in the start-up process of the step motor in order to optimize the torque for acceleration. The optimized torque enables one to reduce the rotational fluctuation by lowering the supply voltage to the motor. The effectiveness of these methods for noise reduction is demonstrated by the experiment of the actual FDD in consideration of operating conditions.

View full abstract

In this paper, the control problem of the landing attitude of a somersault is considered by applying the optimal control theory and a cascade neural network model. A physical model for a human being is fundamentally modelled with a two-link model. First, by solving the nonlinear optimal control problem, the optimal torque to control the landing attitude was obtained. Second, in the application of neural network, the motion of a somersault was evaluated by performing computer simulations and then a control torque was obtained by the back propagation method. As a result, it was demonstrated that in both methods, the landing attitude can be controlled, and the control torque obtained from the neural network is similar to the one from the optimal control theory.

View full abstract

Skiers control their stability by pumping their bodies up and down. The dynamic model of a skier traversing an undulant snowy surface was formulated in order to investigate the mechanism of a pumping strategy to minimize the vertical acceleration. A mechanical model of legs was built with a link joint. The muscle was modeled by a contractile element, a series of elastic elements and a damping element. By numerical calculations, the strategies to avoid the separation of the feet from the snowy surface were investigated. The stiffness of the leg is softened by bending the knee ; consequently, the shock can be absorbed passively by deep bending. Up-pumping first increases the pressure against the snowy surface and then decreases it, and down-pumping induces the opposite effect. The shock is also reduced actively by pumping.

View full abstract

The active control system is expected to improve the characteristics of the machine actively because there is an active force subsystem which is installed on the machine in this system. However, the interference (called forced pressure) between the machine and the active force subsystem existed in the system constructed by an electrohydraulic control type. In this paper, we explained the influence of the forced pressure on the static characteristics and the dynamic characteristics of the system by the results of the experiment and the simulation, and studied the effectiveness of the decoupling control method presented by us using the model reference adaptive control. It was shown that the decoupling control method is useful to reduce the forced pressure in all experiment conditions ; in particular, the reducing rate is over 50% for large amplitude of sinusoidal input signals, and is also expected to reach a higher level.

View full abstract

A flexible arm has been developed as one of the solutions to improvement of the work speed and energy efficiency of a robot arm. The control which considers the vibration of the elastic links is necessary for a flexible arm to complete the precise positioning of the joint angles. This paper treats two matters. One is the design of the optimal torque inputs to the joints which suppress the vibration of the elastic links. These inputs are designed with both the linearized model of a fiexible arm and the nonlinear one. And the optimal control theory is used for design. The other is the design of a tracking control system which is composed of the feedforward and feedback control. The optimal inputs are applied to the feedforward control, and the feedback control is dependent on the optimal regulator. The effect of the control system is shown by the experiments.

View full abstract

Use of a wall-climbing robot for such purposes as rescue, wall inspection and fire fighting in high-rise buildings has been anticipated for a long time. Three quite different types of models have been developed in our laboratory. The present one can move on a wall by using the thrust force of propellers, which are inclined a little to the wall side to produce the frictional force between wheels and wall surface. It has a long and lightweight body to climb up on an irregular wall surface, and its mechanism and the control system by which it can move in a calm or weak wind condition are discussed briefly.

View full abstract

In this paper, a methodology using the Lyapunov direct method is proposed to analyze the stability of an multiple degree-of-freedom linear manipulator system which is positioned on a flexible wall with collision phenomenon. A method-ology to determine the controllaw of the manipulator which guarantees the system stability is also discussed. As collision is a phenomenon with energy dissipation, it is used in the field of industry for such purposes as to suppress vibrations. On the other hand, in the field of robotics, the phenomenon is usually avoided, because the controllability of the system becomes poorer at contact. But as demands for fast tasks against environment increases, collision becomes a problem which must be solved. In our previous reports, some approaches using the Lyapunov method were discussed. In this report, we expanded the manipu lator complexity to n-D. O. F and proved the stability of the system, including even collision. Some numerical simulations were also executed to show the transient behavior of the system and to compare their results with the results of the theoretical analysis.

View full abstract

The automatic operation is important for the in-service inspection of the operating nuclear power station or the decommission of retired plants. The master & slave control will be introduced for work-robot control. It is desirable that the slave involves the capability of problem solving. This paper assumed that the slave involved the general problem solver algorithm. In view of having solved the puzzle of the "monkey & banana", the slave system is regarded as the reasonable alternative which incorporates the capability of problem solving. Basically, the GPS solved a problem by reducing the difference between an initial state and a goal state, and hence the performance of GPS depends on selecting the difference to be reduced. The usual GPS is given in advance the ordering which indicates the importance of the differences. In this paper, the GPS was improved by making use of the rules which decide the order. When several choices are found on the given difference, the fuzzified decision to determine the action is demonstrated in this paper.

View full abstract

This paper describes the stiffness and damping coefficient variation in bearing oil films induced by large dynamic loads. A rotor-journal bearing system was analyzed as a coupled problem of momentum equations of the rotor and the Reynolds equations for all the bearings because of the nonlinearity. By using the displacements and the velocities of the rotor center obtained by the analysis, the stiffness and damping coefficients have been numerically calculated as an unsteady state along the locus of the rotor center. The variations per rotor revolution for the coefficients are shown for journal bearings for a rotary compressor. The results clarified that most of the coefficients for all the bearings vary by more than one order of the value, and that oil film nonlinearity is largely concerned in the rotor action.

View full abstract

The thermal deformation of the machine tool spindle has recently become the most important problem to materialize the manufacturing system for a product with higher accuracy and performance. In order to solve such problems, a large number of investigations have so far been dealt with. Of these, the thermal deformation of the spindle, mainly caused by the internal heat sources such as a bearing, is a major subject. However, there is no effective model to express the dynamic and thermal behavior of the spindle bearing system in the actual rotational condition. In this paper, therefore, to establish the dynamic simulation model of thermal characteristics in machine tool spindle systems with the rolling bearing, the thermal behavior of the main spindle system has been investigated by using the thermal closed-loop concept. Indeed, the effects of the bearing surroundings, i.e., those of the quill and bearing housing structures, have been experimentally clarified. In addition, the cooling effects of air blown to such bearing surroundings have been investigated.

View full abstract

Because plastic gears possess sufficient self-lubricating ability, their operation without lubrication is feasible. Accordingly, they seem to be most suitable as the gears for use in a vacuum in which lubricating oil cannot be used. However, in the case of plastic gears there are also the problems of heat resistance, the occurrence of dimensional change as the moisture contained in the atmosphere dehydrates in vacuum, and so on. For the purpose of putting plastic gears into practical use in a vacuum, in this research the dimensional change in a vacuum of various plastic gears which were worked in the atmosphere was examined, and by carrying out an operational test, the results were investigated.

View full abstract

Recently, plastic gears have become to be used also in water, but according to the kinds of plastics, there are some which show a great deal of wear when they are used in water. However, it is considered also that in water or in the place where moisture exists, water performs lubricating action, and wear decreases as compared with the case without lubricathon. At present, the report which can answer this problem, and can become the design data for the plastic gears lubricated with water is scarcely found. Due to the situation like this, in this study, the nylon gears and polyacetal gears which have been used widely as plastic gears were operated in water and with small quantity of water drops, and the various characteristics such as the wear of the teeth, the change of the tooth profile, the roughness of the tooth surfaces and so on due to the difference of the bathing water quantity were examined.

View full abstract

Modified tooth profiles and their deflections cause a gear pair to contact away from the theoretical line of action on which contact is assumed to occur in most present theories, so that they result in transmitting inconstant rotational motion. Actual points of contact which are not on the theoretical line of action can be described with the tangential-polar coordinates. With these coordinates and measured single-flank errors, gear rotational motion will be analyzed in this series of reports. In this first report, the engagement between modified tooth surfaces of a helical gear pair is replaced with that of equivalent tooth profiles with the same single-flank error. The equivalent profile is expressed algebraically through the tangential-polar coordinates and results in an involute curve with a base circle whose radius varies. Through the equivalent profile, a path of contact of the gear pair which transmits inconstant rotational motion is analyzed.

View full abstract

The equivalent tooth pofile under load is analyzed and its algebraic statement is given with the tangential-polar coordinates. It consists of two different curves for one pitch, which means one curve for single meshing and the other for double, or one for double and the other for triple, and so on. The corresponding instantaneous base-circle radius varies and is discontinuous at the intersection of the two curves. The spring constant under multiple meshing is defined, which depends on the amount of modification. The fundamental requirement for gear-tooth contact is also given with the equivalent profile, which says that relative normal velocity at a point of contact equals zero by taking the deflection into account.

View full abstract

This paper presents a study on the deflection and bending moment of the WN (Haseg SymMarC) gear tooth. The tooth deflection and root stress of a Haseg SymMarC gear tooth due to a concentrated load were calculated by the three-dimensional finite element method (FEM) and compared with those measured. The validity of the gear model and constraint condition used in this FEM analysis was confirmed, and the critical section of the Haseg SymMarC gear tooth and the effects of the number of teeth and face width on the tooth deflection and bending moment were clarified. Furthermore, on the basis of these results, the approximate equations for the influence functions for deflection and bending moment of the Haseg SymMarC gear tooth were derived.

View full abstract

The accuracy of tooth profiles of gears finished by form grinding is strongly affected by positioning errors between the grinding wheel and the work gear, such as errors in the center distance, the inclination angle of the wheel axis and the position of the grinding wheel in the axial direction. In this paper, the effects of these positioning errors on the profile errors of the helical gears being ground are investigated numerically and two practical methods to compensate the profile errors of the gears ground with the positioning errors are proposed. One is a method of modifying the grinding-wheel profile based on the measured profile error of the ground gear. The other is a method of analyzing interactively and removing the positioning errors producing the profile error of the ground gear. To demonstrate the usefulness of the methods, error compensation for the involute helical gear ground with two or more kinds of positioning errors is conducted.

View full abstract

The greatest possibility of the conical involute gear is to use the straight conical involute gears instead of the bevel gears. However, the key technology to realize this usage is to establish a grinding method of the straight conical involute gears having a comparatively large cone angle. In this paper, three grinding methods to finish such a gear are developed. They are Niles-type infeed grinding, table sliding infeed grinding, and inclining work-arbor infeed, grinding. The principles and the available machines are described, the test results are discussed, and it is concluded that the most practical method is table sliding infeed grinding.

View full abstract

A numerical method of calculating the static transmission error of screw compressor rotors is presented. Factors such as profile errors, lead errors, bearing position errors, and misalignment of the rotors with respect to the rotational shafts are considered in the calculation. All compressor parts are assumed to be rigid. Clearance between the rotors in the direction of the rotor rotation is calculated from rotor surface displacements, composed of the above individual errors. By tracing the clearance along the theoretical contact line from the suction end to the discharge end, the minimum value is obtained and it is converted to the transmission error. Some numerical examples of the static transmission error, its Fourier-series coefficients, and movement of the rotor contact point are shown for a sample compressor, of which the shaft lines are not parallel to each other and (or) rotor axes are misaligned.

View full abstract

The sealing characteristics between a turned surface of soft material such as copper and a smooth seal face of a rigid steel ball in static contact were investigated. This seal face received not only normal force but also tangential force. The following results were obtained from calculations by using a model that considered a surface roughness on a turned seal face as a wedge-like projection, and gas leak tests. (1) The compressive force for sealing, P_c/l, decreased with a decrease in the conical half angle, θ of seal face. In the case of a surface roughness of 8 μm, the value of P_c/l at θ=15° was about 0.4 times that at θ=60°. (2) The recommended values of θ were in the range from 5° to 35°, and the P_c/l values corresponding to the range were able to be forecast.

View full abstract

The mean contact pressure of a machine tool slideway is relatively low, and its sliding velocity is also low. Many such contact surfaces are in a boundary lubrication state. In this report, the electric contact resistance between model contact surfaces under boundary lubrication is precisely measured and the boundary contact situation, especially the information on adsorbed lubricant film, is investigated. The result is compared with the static friction measurement on the real rough surfaces lubricated by various kinds of lubricants, and it is clarified that the information from electric contact resistance measurement has a strong correlation with the boundary lubrication in real rough surfaces.

View full abstract

A continuously variable power transmission has potential for the improvement of fuel economy and cruising comfort as an automotive propulsion system. A traction drive CVT changes its speed ratio by the control of side slip force on the Hertzian contact area. This mechanism has the feature of equal torque transmission of every rolling element and low wasting power for the speed ratio change, but sometimes falls into self oscillation at high rotating speed. The paper shows the principle of the speed ratio changing mechanism of the half-toroidal CVT and highlights a digital compensation method for stabilization of the electrohydraulically operated speed ratio control mechanism.

View full abstract

A new method for synthesizing the planar multilink mechanism is presented based on the high-speed comparison of two curves. The curve comparison is achieved by scaling up or down the coupler curve in order for its length to coincide with that of the desired curve, and by rotating and translating the coupler curve so that the maximum value of distances between two corresponding points on the coupler and desired curves may be minimized using the method of least squares. The formulas of coordinate transformation of the coupler curve are also presented for minimizing the maximum value of perpendicular distances from the points on the coupler curve to the tangents of the desired curve at the corresponding points. Numerical examples included here show that the method has attractive convergence properties in a search domain composed of kinematic constants.

View full abstract

An index of motion-transmission characteristics τ of the planar six-link mechanism with a prismatic pair is presented through investigations of the sensitivities of the angular and linear displacements of the links to the deviations of the kinematic constants and the forces acting at the pairs to the external forces on the links. Generally speaking, the average of the sensitivities S and the average of the forces acting at the pairs F decrease rapidly in the interval [0, 0.3] and slowly in [0.3, 1] as τ increases. The τ-S and τ-F relations are approximated by the regions bounded by two rectangular hyperbolas. Consequently, it is possible to estimate the values of S and F of any mechanism at the given crank angle by means of τ.

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 arms and legs. Namely, the joint forces and bending moments of links when the mechanisms take arbitrary positions and attitudes have been formulated by the total link lengths L and by introducing four dimensionless factors. Characteristic charts of statics have been made by plotting the lines of the positions of the output point where the values of the joint forces and bending moments of links are respectively constant in dimensionless working spaces. Making use of only several characteristic charts and simple transformation formulas, the joint forces and bending moments of links can be analyzed when the mechanisms take arbitrary positions and attitudes.

View full abstract

In this papers, we first clarified the definition of freedom of pair and mechanism. Next we determined two unit mechanisms of different kinds. One is a mechanism which has two links and one pair, and the other is a simple loop chain. Generally a mechanism consists of one or more unit mechanisms. We call a mechanism with more than one unit mechanism a complex mechanism. We reported the number synthesis of a mechanism in the 1st report, but this method can be used only when a mechanism has one or more loop chains which all have the same freedom space and does not apply to a general complex mechanism. In this report, we searched for a number synthesis method which can be applied to a general complex mechanism which has unit mechanisms with different freedom spaces.

View full abstract

An optical roughness sensor employing the critical angule method has two displacement sensitive ranges which are called the precision range and the wide range. In the present paper ; the smoothing characteristics of a large ortical spot were investgated using four kinds of objectives with different specifications. The large spot can smooth periodic curves such as a grating and a machined surface which has both periodic and random components. The smoothing effects in the experimental results were less than the theoretical ones obtained by averaging the profiles over the spot area. The large spot in the wide range can therefore be used as displacement sensor to measure straightness error in scanning motion, to make a datum line for in-process profile measurements and to make an optical skid in the optical roughness sensor.

View full abstract

Recently, the automated system for the polishing process of metal molds has become important. For polishing metal molds, the surface dimensions must be known by the control process before polishing. In order to increase the productivity in the polishing process of metal molds, a polishing robot which includes an on-line constant force control system is proposed in this study. A 6-axis force sensor is installed in the polishing tool which is itself mounted on the robot arm. The direction of the normal vector of the mold surface is determined from the three components of the measured polishing forces. Fuzzy theory is applied to control the movement of the polishing robot.

View full abstract

Resin concrete has been widely used in machine tool structures in order to improve the dynamic properties, to reduce the lead time of production and to reduce thermal deformation. It has been pointed out, however, that Yung's modulus of the concrete is about one-seventh that of cast iron and is inferior to conventional materials in static stiffness. Therefore, in this paper, a new material named ceramics resin concrete, has been developed for precision machine tool structures. This material consists of epoxy resin and ceramics which are alumina, silicon carbide, zirconia or a blend of these. It is concluded from the results that Young's modulus is three times and the compressive strength is two times that of conventional resin concrete. In addition, production methods are established for the ceramics resin concrete.

View full abstract