Experimental modal analysis has been widely used to identify the modal parameters of structures and reduce vibration. Mainly, it is based on frequency domain identification. Recently, a time domain technique has been proposed. This technique usually utilizes the FFT analyzer to condense the data, which is then inversely transformed to obtain the impulse responses. The main defect is the conversion error caused by the FFT calculation. This paper introduces a pure time domain technique. Impulse responses are calculated from the consecutive hammering force and response, and are then averaged to reduce experimental noise and error. The characteristic poles in the z-domain are calculated from these impulse responses and converted into the s-domain. The vibrating modes are determined so as to fit the response with the linear summation of the normalized responses of the identified poles. The proposed technique is applied and its accuracy is tested.

View full abstract

Many methods which estimate dynamic characteristics have been developed and are now widely used in many fields. Most of these methods use the Frequency Response Function (FRF) as input data. Recently, some new methods based on the Maximum Likelihood Principle which also utilize FRF as input data have been proposed by one of the authors. According to those methods, system parameters are estimated with a least squares principle using the reciprocal of FRF's variance as a weighting function. Therefore, it is necessary to estimate not only FRF itself, but also the variance of the FRF. In this paper, estimation methods of the FRF's variance are presented. Primarily, two kinds of methods are used to estimate the FRF from measured input and output signals : they are the "H₁" and "H_v" estimators. In this paper, both estimators are used and a variance is formulated for each estimator.

View full abstract

The objects of the present paper are to propose a variance evaluation method of modal parameters identified by the least square method with an experimental transfer function, and to show that the evaluated variances can be used to optimize the accuracy of the modal parameters at selection of proper excitation type, weighting function and so forth. The variances of the modal paramenters are evaluated by the first order approximation with the discrete values of experimental transfer function as stochastic variables, where the variance of the transfer function are evaluated with the coherence function based on a theory of statistical error. The results on a simple beam made of acrylic resin show that the variances can used to optimize in accuracy of the modal parameters, and the results on the cover of an outboard motor show that the proposed variance evaluation method is applicable to actual structures.

View full abstract

This paper presents a method to synthesize the structural modification required to realize the compliance shift on the Bode plot in problems of vibration with Rayleigh damping. The method is based on the notion that the objective design is searched near the baseline design under the equality constraint conditions derived from the first-order approximation of the compliance change with respect to the design variables. An iterative algorithm is formulated to determine the design variables and Lagrange multipliers on the basis of the sensitivities of the finite element solution and the given shift. The numerical examples are concerned with the compliance of a cantilever beam and a simply-supported beam. The moment of inertia and cross-sectional area at the element nodes are taken as the design variables. It is shown that the structural modification can be attained by several iterations of the algorithm for the given sets of the compliance shift.

View full abstract

A new identification method of modal parameters by using an exponential window is proposed as a single-degree-of-freedom method. The method is superior in simplicity to any other method in the case of a discrete transfer function such a when using a FFT analyzer. Using the method, the modal parameters of natural frequency and modal damping ratio are identified with a complex ratio of the transfer function obtained from the unit impulse response multiplied by an exponential window to the normal transfer function at an only point near resonance. The present method and former method are compared through the results identified from experimental results on a simple bar.

View full abstract

In order to minimize the height of Czochralski growing equipment for silicon single crystal. a wire rope winding method or a ball-chain winding method is used instead of a stiff shaft pulling a single crystal. When the wire rope or the ball-chain rotates, amplitudes of the whirling vibration gradually increase with time. and a self-excited vibration occurs. In this paper, the cause for this self-excited vibration is studied experimentally using a model of the wire-winding-type growing equipment for silicon single crystal. The self-excited vibration of a wire rope occurs when the rotation is above a resonant speed of the wire rope system, and then only the forward whirl of a wire rope becomes unstable independent of the direction of an initially added whirling. On the basis of the experimental results, it is clarified that self-excited vibrations of silicon single crystal growing equipment are due to an internal friction among element wires of the wire rope.

View full abstract

In order to develop a method applicable for identifying nonlinear vibratory systems without assuming the characteristics of nonlinearity, a new technique has been presented. The basic procedures of this technique are : (1) obtaining data of a periodic force applied to a system and of the periodic steady-state oscillations induced by the force (2) dividing the regions of displacements or velocities in small subdomains and expressing the restoring force or damping force in a piecewise manner over each subdomain and (3) applying the principle of harmonic balance. The validity of the technique is demonstrated by applying it to some single-degree-of-freedom vibratory systems.

View full abstract

In the present report, the authors propose a practical and useful method for identifying the characteristic matrices (general name of the mass, damping and stiffness matrices) of mechanical structures. The characteristic matrices of mechanical structures can be identified in a much shorter CPU time by the proposed method than by previous methods also proposed by the authors. In this paper, the theory of the proposed method is explained. Three examples are demonstrated in order to prove that the method is valid. practical and useful. The first example shows that the method can identify physically valid characteristic matrices. The second shows that modal parameters from identified characteristic matrices correspond well to the result from the Poly-Reference method. In the third example, a cylinder-head of an actual engine is identified by the proposed method.

View full abstract

An analytical study is presented on leakage-flow-induced vibrations in the case of a one-dimensional, narrow, tapered passage. The moment of fluid-dynamic forces acting on the walls of the passage when one of them is vibrating in the rotational mode is determined. It is found that the area increment ratio, on which condition negative fluid-dynamic damping and negative fluid-dynamic stiffness can occur, changes remarkably with the location of the pivot. Negative fluid-dynamic stiffness can occur in the case of divergent and convergent passage. Negative fluid-dynamic damping is apt to occur in the case of divergent passage, but if the pivot is situated slightly behind the center of the passage, it can also occur in the case of convergent passage. If the pivot is situated at the inlet of the passage, negative fluid-dynamic stiffness and negative fluid-dynamic damping rarely occur.

View full abstract

A new method of vibration analysis of continuous bodies is presented. In the method, the domain to be analyzed is divided into several subdomains. Analytical solutions are used for subdomains which have regular shapes and the FEM is used for irregularly shaped subdomains. The analytical solutions are transformed into relations between generalized forces and generalized displacements at nodal points located on the boundaries of the regular subdomains. This transformation is based on a kind of the energy method and the concentration of stresses. The relations are formally the same as those of the FEM. Therefore, the solutions for the regular subdomains can be joined directly to the FEM solutions for the irregular subdomains. As simple examples, some two-dimensional acoustic problems are analyzed by this method. It is confirmed that the method brings more accurate results and shorter computational time in comparison with the FEM only.

View full abstract

The rolling piston of the rolling-piston-type rotary compressors plays an important role in achieving distinction in mechanical efficiency. Before examining the mechanism by which the rolling piston effectively decreases the power losses due to the mechanical friction, this study presents the experimental data on the high mechanical efficiency and the rolling-piston dynamic behavior of a comparatively small colling-capacity rolling-piston rotary compressor. Furthermore, this study makes observations of the crankshaft, which was subjected to a long endurance test of about 2 000 hours, to examine the wearing conditions. It is concluded from this study that the mechanical efficiency is 90.3 %, the mean velocity of the rolling piston is 32 rpm and an excellent fluid lubrication is formed between the rolling piston and the crankpin.

View full abstract

In order to make vibration analysis an eary and automated task for designers, an interactive vibration analysis system using a personal computer or an engineering workstation was developed. The system has the following advantages: (1) The system has a user-friendly interface for graphic input of vibration models which enables users to make figures of vibration models on the display by choosing the vibration model elements listed in the menus by a pointing device (mouse). (2) The system can perform automatic generation of equations of motion from the graphic input, and automatic numerical analysis of them including eigen-value analysis and time history analysis. (3) The system is applicable to extensive models of two-dimensional and three-dimensional vibrations of multi-body systems which can contain various constraints and nonlinear elements including a friction element, an elasto-plastic hysteresis element and nonlinear springs/dampers defined by users. The main part of the system was written in Smalltalk-80 and the part for the numerical analysis was written in C.

View full abstract

A numerical analysis is developed for the nonlinear bending vibration of a flexible arm which is accelerated to high speed in a range of bending vibration period. Strict equation of motion of the flexible arm is derived, including the nonlinear effect of a center shaft motion. The time responses of the bending vibration of the flexible arm in a rapid access process are numerically calculated and the critical top velocity which produce nonlinear effect is discussed. From the experimental study it is found that the theoretical results agree well with the experimental ones if three nonlinear terms and the external damping term of the surrounding air are considered.

View full abstract

Free vibrations of a string with time-varying length and having a linear damping are analyzed. An equation of motion cannot be solved by the method of separation of the variables because the length of the string varies with time. New variables of the position and time are introduced and then the equation of motion which can be solved by the Laplace transformation is obtained. In order to determine the coefficients of the general solution. complicated difference equations and recurrence formulae must be solved. Approximate solutions of free vibrations induced by a distributed initial displacement are obtained.

View full abstract

This paper presents an algorithm for finding the optimal vibration support of continuous systems and its application to elastic structural elements such as plates and shells. The optimization here implies maximization of lowest natural frequencies, and the gradient method is used for this purpose. In numerical examples, an analytical method previously developed is used to obtain eigenvalues of four-point supported shallow shells, and the optimal design algorithm is demonstrated for this structural model.

View full abstract

The steady-state response of a rotating disk to a concentrated harmonic force at a space-fixed point is analyzed. The eigenfunctions are proved to be orthonormal, and formulation of the response is derived as an eigenfunction series. It is proposed that the eigenfunctions are given in the form of rotating waves, and the resonant frequencies and vibratory modes are discussed. Resonance occurs when. the exciting frequency coincides with one of the natural frequencies observed in the space-fixed coordinates. If two of the natural frequencies coincides with each other, the resonant mode that may be called "a quasi-standing wave" is formed in the space-fixed coordinates.

View full abstract

Rubber bearings of a 980-kN rated load were tested to investigate the breaking conditions of rubber bearings whose top and bottom planes were bonded to thick steel plates for fixing by bolts. Breaking tests by simple tension were carried out for the full-sized rubber bearings, showing that the breaking tensile forces were larger than the rated load (1.2 to 1.9 times). Biaxial breaking tests by tension and shear were performed using 0.4 scale models of the bearing. The tests showed that, through the breaking tensile forces decreased with increased shear deflexion, the breaking forces were still large (0.80 to 0.94 times of the rated load) even under a shear deflection of 0.3 m in full size. Biaxial breaking tests by compression and shear for the 0.4 scale models revealed that breaking shear strains were almost constant under various vertical loads up to 2 times the rated load and they were about 500% corresponding to a shear deflexion of 0.8 m in full size, which was larger than the diameter of the bearing. The breaking shear forces were 1.6 or 1.7 times that of the rated load.

View full abstract

Mechanical snubbers have often been used as supports for the piping in nuclear plants, but, recently the elasto-plastic-type piping support devices (named energy absorbers), which are low-cost and maintenance-free, are planned for use instead of snubbers. Under these circumstances, to establish the concept of the piping design method using the elasto-plastic-type piping support devices, device test and vibration test of the piping system supported by elasto-plastic-devices were performed. By the results of these tests, we confirmed the following items: (1) The evaluation formulas for the force-displacement relation and the dissipated energy of devices were established and considered to be reasonable in comparison with the test results. (2) In the case of the piping system investigated, the maximum damping ratio is about 10%.

View full abstract

Cylinders with various aspect-ratios and spheres, attached axially to a vertical spring, were vibrated longitudinally in still water by a vibrator which provided the top of the spring with a sinusoidal displacement. Then, the drag and added-mass coefficients were analysed by introducing the experimental values of amplitudes and frequency in the resonance into the solution of a spring-mass-damper system. First, to examine the validity of this method, the results on the spheres were compared with those by Sarpkaya, and both results proved to be in fairly good agreement. The results on the cylinders led to the following conclusions. There is a good correlation between those coefficients and Keulegan-Carpenter number K_c(=U_mT/D). A_S K_c. increases, the added-mass coefficient increases linearly and the drag coefficient decreases exponentially. When K_c exceeds about 10, the drag coefficient approaches the constant value. However, there is no clear correlation between these coefficients and Reynolds number.

View full abstract

A theoretical investigation is presented for the hydroelastic vibration of finite length, rotating cylindrical shells partially filled with incompressible and nonviscous liquid. Gravity and surface tension is neglected, while the initial tension due to the rotation is considered. On the basis of the Flugge-type equation, the shell with both edges clamped is analyzed by the variation of parameters. It is found that the magnitude of the initial tension due to the centrifugal force depends on the rotating speed and the quantity and density of liquid. The frequency increases with the initial tension, while it decreases with the inertia of liquid.

View full abstract

This paper deals with equations of motion of a flexible multibody system whose bodies are connected by rotational hinges. The equations of motion are derived by taking degrees of freedom of one body properly set in the system, hinge rotational angles, and modal coordinates of flexible bodies as independent variables. Kane's equations of motion are used in the derivation and each term in the equations is expressed in a convenient form for the efficient recursive calculation. The system with loop structure is also treated by cutting loops at proper hinges. Although arbitrary modes can be used to express the elastic motion of a flexible body, if the modes of the static deformation at hinges are used, the simulation accuracy is improved and the loop configuration is easily treated. An example of a simulation program based upon the formulation algorithm is also shown in the paper.

View full abstract

The damping force characteristics of a semiactive electro-magnetic absober which is composed of a ball screw, an aluminium disk and electro-magnets are discussed experimentally.The effects of vibration isolation of this absorber to a simply supported beam are also discussed experimentally and compared to the results of digital simulation. The results may be summarized as follows : (1) The damping force of the absorber is nearly proportional to relative velocity. (2) The viscous damping coefficient increases as the voltage of the electro-magnets increases, and is nearly linear for the voltage in some voltage range. (3) The viscous damping coefficient depends on the physical properties of the electro-magnets and the conductor disk as well as the lead of the ball screw.

View full abstract

This paper deals with the vibrations of linear and nonlinear dynamic dampers as examples of systems with Coulomb friction. A direct numerical integral method is presented as the one to analyze systems with discontinuities such as Coulomb friction. The analytical method to obtain the solution of vibration with two sticks in a period is also presented. Comparing solutions derived from these two methods clarifies that the solution of the latter is exact and that of the former is high approximate. The solution using equivalent viscous damping is also compared with the solutions mentioned above. As a result, it turns out that this is a good approximate solution only in small Coulomb friction.

View full abstract

This paper is on the subject of active vibration control of flexible rotors supported by magnetic bearings. Using a finite element method for the mathematical model of flexible rotors, we have formulated the eigenvalue problem taking into account the interaction between a mechanical system of a flexible rotor and an electrical system of magnetic bearings and a controller. However, gyroscopic effects are disregarded in this paper for simplicity. We will be able to adapt this formulation to a general flexible rotor-magnetic bearing system. Controllability with and without collocation is discussed for the higher-order flexible modes of the test rig. In conclusion, we have proposed that it is necessary to add new active control loops for the higher-order flexible modes even in the case of collocation. Then, it is possible to stabilize for the case of uncollocation by means of this method.

View full abstract

A theoretical study of the natural vibrations of rotating floppy disks with a R/W head system is carried out. A solution is obtained by applying both the methods of a one-mode approximation and a multi-modal expansion. It is shown that the one-mode approximation brings almost the same results as the multi-modal analysis when the values of the mass and the stiffness of the head are relatively small. It is also shown that the natural frequency of the system in general decreases with an increase of the disk rotation speed and finally comes to zero at a certain speed. After that, a divergence type of instability appears in the system. When the mass and the stiffness of the R/W head become greater, some flutter type of instability appears in the system before the divergence instability. It is found that the air film within the disk case plays a very important role in stabilizing the behaviour of the rotaing disk.

View full abstract

For a better understanding of hunting, it is desirable to be able to obtain the amplitude and frequency of hunting analytically. In the author's last two reports as a first step toward nonlinear approximate analysis, the effects of individual parameters of a closed-engine-governor system on the instability of equilibrium state are linearly estimated, especially on the frequency and increment of amplitude. In the present report, the effects of three parameters that have a marked influence on the limit cycle characteristic are estimated as a step to analytically explaining the mechanism of limit cycle evolution. Results show that the limit cycle disappears over all speed ranges with increasing moment of inertia of crankshaft system 1.8 times that in the actual engine as with minimizing subventuri pressure lag. Results also show that the amplitude decreases except at extremely low speeds with decreasing mass of the governing system, but it is difficult to decrease the mass in reality. Further, it is found that the limit cycle does not disappear no matter what value the damping of the governor may take.

View full abstract

Large Space Structures (LSS) such as space stations are expected to have the dynamic characteristics of high flexibility and low. inherent damping owing to their large dimensions and lightweight construction materials. Therefore, the needs for active control of vibrating LSS arise and the concept of actively controlled large space structures has received a great deal of recent attention. A tendon control system is proposed by the authors for the vibration control of a cantilever beam-like flexible space structure. It consists of a torque generator, placed at the root of a structure and linked with a couple of moment arms through tensile wires (tendons), thereby applying vibration control torques to the structure at the arm position. This paper discusses the dynamics of the tendon control system for the ground test beam hung in a vertical direction to investigate the feasibility of this control method. First, the tendon actuator is modeled as an added one-degree-of-freedom vibrating system and the equation of motion is derived for the flexible beam considering the effect of gravitational load and axial force generated by the tension of the tendons. Then, the mathematical model of the whole system is formulated, combining both sub-systems. Finally, a discretized dynamical model is constructed through FEM. By using this model, an open-loop transfer function from the applied moment at the upper link to the rotation angle of the moment arm is investigated. It is shown that only one open-loop zero vanishes at the frequency point determined by the natural frequency of the tendon actuator system and that its phase lags by as much as 360 degrees. Further, this fact is verified experimentally. Finally it is concluded that the dynamics of the tendon actuator system plays an important role in the design of the tendon control system for the flexible beam.

View full abstract

This paper experimentally studies modeling and control of a class of flexible spacecraft having elastic appendages. First. an equation of motion of the flexible spacecraft is derived from experimental data of flexible appendages. and modal analysis using unconstrained modes is applied to the equation in order to obtain the reduced order model. Subsequently, the characteristic transfer function method based on the design freedom is adopted to design the controller. The effectiveness of the controller derived by this method is demonstrated in control experiments by using a flexible spacecraft model on a single-axis air-table. The evaluation for the results of the control experiments is made concerning with vibration control, suppression of spillover instability, and step response of attitude angle.

View full abstract

A new method of designing a digital controller for a multivariable system is proposed in this paper. We consider the design problem such that the closed-loop system with a cascade compensator approximates a specified reference model. The method is based on an approximation of the open-loop transfer function matrix. A square error criterion is introduced as a measure of the approximation. By minimizing the criterion, a linear equation is obtained for parameters of an FIR (finite impulse response) controller. This equation can be solved recursively by increasing the order of the controller by a type of Levinson's algorithm. Since the order of the controller can be arbitrarily specified in this method. the lowest order which realizes the desirable accuracy of the approximation can be determined without regard to the orders of the plant or the reference model. The extension of this technique to an IFIR (integrated finite impulse response) controller is also provided.

View full abstract

Robotic manipulators working under underwater environments have influences of fluid forces from liquid, unlike in the air. The fluid force gives problems to the manipulators, such as oscilations in the liquid, which cannot allow the manipulator to move at the higher speed. Thus, the working efficiency cannot be improved. In this paper, a method of adaptive fluid force compensation is shown by using the model reference adaptive control approach in a serial-parallel connection. The proposed adaptive compensation method employs only the position and velocity information. The nonlinear effects of fluid forces are compensated. Some comparisons were carried out with the simple fixed gain feedback control with/without the output feedback loop of strain gauges.

View full abstract

Actively controlled air suspensions are expected to improve the' ride quality and stability of high-speed ground vehicles. In this paper, feedback control systems are discussed for active air suspensions by designing both a proportional-integral controller and an optimal regulator with a state observer. These feedback controllers are designed and realized by a microcomputer, to which the acceleration of a suspended mass is inputted, and which operates a pneumatic actuator by the pulse width modulation. An experimental set-up is constructed and the vibration isolation is examined by a test stand. The results are compared with those from theoretical calculations. According to this comparison, it is found that the model is verified and the vibration isolation is greathy improved.

View full abstract

A semi-active suspensions (SAS) is superior to passive suspensions, while requiring only a small amount of control energy. It uses a semi-active damper in which the force-velocity constant can be varied. In general, the control strategy of SAS requires both sprung-mass velocity and relative velocity across the suspension. This paper introduce a new control strategy of SAS which uses only directly measurable variables in vehicle applications. Simulations and experimental studies show the proposed system is superior to passive suspensions and comparable to ordinary semi-active suspensions.

View full abstract

The traditional measuring equipment called "Tympanometer", which measures the dynamical characteristics of the middle ear, is widely used, but its reliability is insufficient. Therefore, an attempt is made here to develop new measuring equipment which will give us more detailed information, and the dynamical characteristics of the middle ear of normal subjects and patients are measured with this equipment. The results show a significant dependence on the dynamical characteristics of the middle ear upon external auditory meatus pressure and the large individual difference of the middle ear dynamical characteristics, in spite of normal subjects.The possibility exists of the clinical application of this newly developed measuring equipment to the diagnosis of middle ear diseases.

View full abstract

The gyroscopic effect is proportional to the rotational speed and polar moment of the inertia of a rotor. This effect appears in a rotating machine borne by electromagnetic bearings due to high speed rotation and compactness of the rotor. Natural frequencies of the rotor system change depending upon the gyroscopic effect, which is proportional to the rotational speed. The natural frequency of a rotor at rest is separated into forward and backward natural frequencies as the rotational speed increases. The former increases with the speed and the latter decreases. The difference between them depends upon the gyroscopic effect. From the viewpoint of control engineering, this effect means the change of a control objector. Then, the phase margin of the controller must cover a frequency range where the rotor natural frequencies change. In this paper, some techniques are shown to improve the phase margin throgh by passing the actuator current, notch filter for rotor bending mode vibration, cross network based on the rotor whirl motion and so on, These techniques have been experimentally proven.

View full abstract

This paper describes the relationship between broken plastic gear teeth and rotary bending fatigue tests of plastic materials for a temperature rise in the gear teeth. When the temperature rises, plastic gears have a tendency to break. The equilibrium temperature under the condition of thermal balance was calculated, an experiment carried out and the temperature rise relationship was explained. The broken teeth were observed using an electron microscope. Two kinds of fatigue were shown by the test : fatigue due to the development of microscopic cracks and thermal fatigue. These two types of fatigue were compared with the rapid rise in bending for plastic materials has something in common with plastic gear fatigue. The results provided good design data., enabling the production of better and more durable plastic gears.

View full abstract

The gear dynamics is described by a time-varying nonlinear differential equation due to its time variable tooth stiffness and its backlash. Therefore, it is an important and interesting problem from the viewpoint of estimating the dynamic load or gear noise and contribution to the theory of nonlinear mechanics to discuss whether or not distinctive new phenomena occur in a gear system which can not in principle occur in a linear system. In this study, the bifurcation sets of periodic solutions under some gear parameters are obtained numerically by the variational equation, and chaotically transitional phenomena are investigated by the Poincare6 map.

View full abstract

In hobbing. gears are usually cut with axial feed of the hob. However, for cutting of helical gears or large gears. the path is too long from the point that the hob teeth initially come in to contact with the gear blank to the point from which the tooth profile begins to he generated. If radial feed is used in the initial cut-in process, the hobbing time will be substantially shortened since the distance of the feed is similer to the tooth depth which is shorter than that of axial feed. However, the load of the hob teeth is largo in radial feed and the possibilities of large wear of the hob and an increasing vibration of cutting exist. In this research. cutting tests were performed comparing the hob wear in radial feed, , with that in axial feed hobbing. The result are as follows. (1) The wear of the hob by radial feed cutting was nat as great for tests in which the feed rate was about 2/3 times the axial feed. (2) In helical gear hobbing, the wear of the hob engaged in radial-axial feed hobbing was less than that for a whole gear being cut by axial feed only, since the extent of teeth under the large cutting load in radial feed differs from axial feed cutting.

View full abstract

The relief wear of a noncoated hob grows in the direction of hob rotation. With the increasing of a side relief angle, the width of relief wear geometrically calculated becomes less. In case of a TiN-coated hob, however, a TiN-coating layer restrains the wear from growing in the direction of rotation, and the angle contained by the relief surface and wear surface is large, so a TiN-coated hob is expected to be insensitive to the measuring of a relief angle. As a result of the experiments, (1) a TiN-coated hob is found to be insensitive in measuring the relief angle at a low cutting speed : (2) at a high cutting speed, the smaller relief angle is effective for lengthening the hob life : and therefore, (3) in high-speed and feed hobbing of automobile gears, the relief angle of TiN-coated hob must be as small and as possible.

View full abstract

In order to apply a new u, material to the power transmission gears. the load-carrying capacity of the gear pairs with a combination of through-hardened steel (JIS SCM435) and as-cast spheroidal graphite cast iron ( FCD80 ) was s investigated by using a power circulating-type gear testing machine. The test results reveal that this combination of materials greatly influences the surface durability. The pinion (SCM35 ground gear) mating with a FCD80 hobbed gear operates better than that with a SCM435 hobbed gear. In the driven gears, the pitting fatigue life of FCD80 hobbed gears is longer than that of SCM435 hobbed gears or finish-hobbed gears. In the case of ground gears with a smooth tooth surface. the value of the endurance limit for the surface durability becomes greater in the FCD80 gears as well as in the SCM435 gears. From these considerations, it is found that gear pairs with a combination of SCM435 and FCD80 can be used for industrial gear units.

View full abstract

To investigate the mechanics of the staking process. the fastening of tee-joint with sheared tab and slot is carried out for different snaps. The staking modes of the joint unions at different press-fitting force are obtained, and results are discussed by using pull-apart force and appearance of the joint. The fastening strength in the staking, is found to be dependent on snap shape, combining materials, length of tab and edge shape of slot hole.

View full abstract

In order to develop deep wells in corrosive conditions, premium con-nections with high joint tensile strength and high seal integrity have been required. As for the high joint tensile strength, jump-out tensile failure, by which tensile strength and the elongation of connections are decreased, should be prevented. The mechanical behavior and preventive conditions of jump-out tensile failure are investigated through the failure analysis of connections with varied box-end profiles and thread length. Incomplete pin threads are necessary to prevent jump-out tensile failure, in addition to the negative load flank angle thread for main-taining the mating of the threads and the complete thread lenIn order to develop deep wells in corrosive condition5, premium connections with high joint tensile strength and high seal integrity have been required. As for the high joint tensile strength, jump-out tensile failure, by which tensile strength and the elongation of connections are decreased, should be prevented. The mechanical behavior and preventive conditions of jump-out tensile failure are investigated through the failure analysis of connections with varied box-end profiles and thread length. Incomplete pin threads are necessary to prevent jump-out tensile failure, in addition to the negative load flank angle thread for maintaining the mating of the threads and the complete thread length for ensuring the shearing resistance. The rigidity of the box-end increased by a longer incomplete thread length and an unthreaded ring added to a box-end can help prevent jump-out tensile failure by decreasing the deformation of the threads and the expansion of the box.

View full abstract

The stiffness and damping coefficients of oil film in porous oil journal bearings with nonuniform permeability, and the vibrational stability of a shaft supported by these are theoretically investigated. Three types of nonuniform distribution of permeability are analyzed. For the first type, the permeability of the inside part is lower than that of the outside part. For the second type, the permeability of the loaded part is lower than that of the unloaded part. For the third type, the permeability of part of the bearing ends is lower than that of the middle part. All nonuniform types have larger oil film coefficients, in which the cross-coupled stiffness term K_XY is larger in magnitude, together with a greater stability threshold speed than a uniform type in which the permeability is equal to the higher one of nonuniform types. Moreover, the first type has a greater stability threshold speed within a high bearing modulus λ(λ=So/ν, So : Sommerfeld number. ν : nondimensional shaft rotational speed), and the second type has a higher one within a low bearing modulus.

View full abstract

A new type herringbone grooved journal bearing, which produces an oil film bearing pressure with a shaft or bearing rotation in either direction, is proposed in this paper. A numerical analysis of the bearing parameters using the Narrow Groove Theory and the condition, in which the negative pressure is replaced by zero in the iterative pressure calculation, confirmed that the load capacity of this bearing and the radial load component (related to stability) compare favorably with that of a conventional bearing. The values of the bearing parameters which give optimum load conditions, in the case of either a shaft or bearing rotation, are also determined numerically.

View full abstract

The effect of a decreased content of impurities in steel on machinability was studied by meam of turning tests using cemented carbide tools. Low-carbon boron steels and medium-carbon steels were used as the work materials. The machined surfaces were examined by the use of a scanning electron microscope and an electron probe micro analyzer. The results of the investigation show that imperfections such as voids on the surface layer are produced even when workpieces are machined at high speeds without built-up edges. It is shown that the void formation is associated with inclusions and the occurrence of voids is reduced with a decreased content of inclusions. It is found that the surface damage decreased during the machining of calcium deoxidized free-machining steels owing to the formation of a nonmetallic layer on the tool flank.

View full abstract

Recently, many kinds of aluminum alloys have come to be used in the mechanical industry, but these materials are difficult to cut. Therefore, it is one of the most important subjects in machine shop to find out an effectual cutting method of these aluminum alloys. CO₂ laser cutting is one of the hopeful methods which meets such requirements. However, there have been few studies hitherto which have dealt with the CO₂ laser cutting of aluminum alloys. Under such a situation, CO₂ laser cutting of the thin aluminum alloy plates (A1050, A2017, A5052, A7075) was chosen as a subject of this study, that is, the influence of cutting conditions (laser power, assist gas pressure, focusing, cutting speed) upon cut quality (height of dross, flatness, surface roughness) was examined. The following conclusions were obtained : (1) The Effect of cutting speed on cut quality could be clarified. (2) The height of dross of each material was found to be influenced by surface tension and degree of superheat.

View full abstract

This paper presents a method to analyze the production efficiency of automatic assembling lines with branches. The method is based upon an analysis of the fundaments of assembling lines which are composed of two assembling stations with an intermediate buffer, as mentioned in the previouse reports. An assembling station (ST.M+1) is branched to M-stations (ST.i;i=1, 2…, M) with each intermediate buffer. An equation for the interrupted time is introduced by considerations for the interferences with ST.M + I and ST.i. The equation of those stoppage times for ST.M+1 and ST.i organizes simultaneous equations for stowage ratios of intermediate buffers. The production efficiency of branched lines is given by solutions of those equations. Some examples are calculated numerically and simulations are performed by a degital computer for those. From the results, it is shown that the method is valuable for estimating the production efficiency of automatic assembling lines with branches.

View full abstract

A lot-sizing problem is considered for a group of items with a common manufacturing setup. These items are aggregated in a group by a technique of group technology. The common setup cost for the group is generally less than the sum of each item's setup costs. The cost reduction is expected when lot-sizing is carried out for the group of items rather than for each item independently. The determination of the lot-size for a group of items sharing a common setup is called "group lot-sizing". It is shown by simulation that the percentage of the total cost reduction (PCR) is different due to the dispersion of time between orders (TBO) and the group setup ratio (GSR), the ratio of the group to the sum of each item's setup costs. A small dispersion of TBO or small GSR brings about an increase of PCR. Lastly, the optimal grouping method using dynamic programming and a numerical example is shown.

View full abstract

Many investigations of internal friction of metals have been made so far. But, few well-ordered data of frequency dependence of loss factors have been obtained. Furthermore, it seems that incorrect constants of the three element model for visco-elasticity of metals are used sometimes. Therefore engineers who deal with structural vibrations can hardly know damping characteristics of materials. Then measurements of internal friction of mild steel, brass and alminum are made by use of bending, longitudinal and torsional vibrations. The obtained data of loss factors are arranged in relation to frequency. These metals show similar tendencies of damping to one another.

View full abstract