[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

In machine designing with CAD systems, Boolean union operations of 2-dimensional figures are used for obtaining a figure representing the machine part shape. Designers often place such figures to be combined in a certain configuration so that they have some degenerate intersections, for example an edge of a figure going through a vertex of another figure. In order to securely judge such intersections, tolerances are given to edges and vertices of the figures so that the effect of numerical errors caused by the floating-point computation can be ignored. Determination of a proper tolerance value is important because geometric computations become unstable if the tolerance is too large, and judgements of critical degenerate intersections become uncertain if the tolerance is too tight. In this paper, the authors propose a new method for determining a proper tolerance value which guarantees stable computations of Boolean union of polygons with degenerate intersections. Topological conditions for computing Boolean union are derived based on a classification of intersection points of polygon boundaries. Proposed algorithm determines the proper tolerance value by iteratively decreasing the tolerance until the topological conditions are satisfied. The algorithm is implemented and computational experiments are demonstrated.

View full abstract

Geometrical errors by manufacturing processes have been evaluated by the maximum width of tolerance zones which envelop the actual features. Details of error within such zones should however be investigated when the designer studies relation between errors and functionality of a product and compares and examines manufacturing methods to realize high accuracy for machined surfaces. A new method is proposed for the evaluation of form errors using a series of points such as measuring points on the product surfaces. This evaluation consists of two parts: 1. Form error features which consist of error models (Bézier Template) and error conditions, 2. Segmentation by fitting of form error features using the Simulated Annealing Method. Form error models are defined as a series of quadric or cubic Bézier curves. A set of Bézier control points can be used for classifying or characterizing typical geometrical errors. Segmentation and fitting of these form error features are realized by the Simulated Annealing Method, a statistical optimization method which can be used to search for the global minimum solution of a combinatrial problem. Our method allows the designer to estimate the effects of errors on functionality by using fitting Form Error Features as a substitute. This leads to functional tolerancing.

View full abstract

Three-dimensional rigid-plastic FEM (Finite Element Method) simulation of plain milling process is performed. The down-cut process in plain milling cutting is simulated by analyzing three-dimensional plastic deformation field due to the cutter rotation from the beginning of the cutting until the formation of the chip. The chip formation process of a carbon steel with showing spiral shaped curl is three-dimensionally simulated. The variations of the chip formation process including chip shape and cutting forces with cutting conditions such as tool helix angle and feed rate are calculated. It is obtained that the chip shape varies with simulation conditions of tool helix angle and feed rate, and the chip curl pitch increases as the tool helix angle increases. The cutting forces increase as the feed rate increases, and thrust component of cutting force becomes large as the tool helix angle increases. This simulation method is considered to be effective to simulate the actual three-dimensional cutting process.

View full abstract

As LSIs are scaled down, dielectric degradation of a gate oxide film due to metal contamination during both wafer and device fabrication processes becomes a serious problem because of the severe requirements of device yield and electrical performance. In conventional polishing processes of both wafers and devices, colloidal silica with an aqueous solution of KOH is used. Thorough rinse, however, is needed to avoid potassium contamination which influences greatly the electrical characteristics of devices. This paper deals with a novel polishing method using tetra-methyl-ammonium-hydroxide (TMAH), which can etch silicon without any metal contamination. After a soft touch test in a TMAH solution is conducted using the cantilever of atomic force microscopy (AFM), silicon wafer polishing is successfully carried out and the stock /removal ratio is found to be almost the same as that for KOH etching.The soft touch test using AFM has proved to be a powerful tool for exploring a new polishing solution.

View full abstract

An existing three-dimensional finite element simulation system for metal machining has been revised to cover a wide range of tool geometry and cutting conditions. When an 18%Mn-18%Cr manganese steel is turned by a P 20 carbide tool, the effects of tool geometry on chip flow and tool wear have been investigated. Cutting performance of square, diamond-shaped and button inserts is affected by the side cutting angle, rake angle and tool corner radius. The selection of a large side cutting angle and large tool corner increases the chip flow angle, and decreases the real undeformed chip thickness in the chip flow direction at the corner. These cause the cutting temperature to decrease and the crater wear to reduce as well. From this point of view, the use of a button insert is most favorable for rough turning of the difficult-to-cut material.

View full abstract

This paper deals with the possibility of electric field assisted biomachining using chemolithotrophic bacteria Thiobachillus ferrooxidans that usually lives in waste fluid of mine. It is pointed out that the removal rate of anode workpiece increases steeply in this method when stainless steel bar is used as a cathode tool electrode. The etch factor for copper in this method is approximately 2, which indicates that there is an anisotropic characteristic in the removal amount. The grain boundary of workpiece is likely to remove preferentially for copper and iron. After all, it was made clear that the removal rate of workpiece was controlled by the machining time and the applied voltage between workpiece and tool electrode. It is considered that some genetic technology will be needed in the future for controlling the activity of individual bacteria.

View full abstract

The deposition of diamond films by conventional arc discharge plasma jet CVD is limited to control in terms of dimensions and quality. An arc discharge plasma jet CVD apparatus that utilizes a plasma jet regulated its properties by a magnetic field has been designed and developed in order to synthesize homogeneous diamond films in thickness and quality. As a result of the assistance of a magnetic field, the diameter of a plasma jet increases and the distribution of substrate temperature is improved, and uniformity of thickness, grain size and orientation of crystal faces of diamond films occurs when the magnetic flux density of above 1.6 × 10 ^-2 T is applied.

View full abstract

The deformation process of a specimen surface during measurement of hardness under 10 mN load is examined by FEM computer simulation. In calculation model a pure aluminum surface is indented by indenters (Vickers type and triangular pyramid), with and without finite tip radius using an ultra-microhardness tester that directly measures the depth of concavity. R-correction at the indenter tip and shallowing correction of concavity, which were proposed in the previous report, are investigated. The results obtained are as follows. (1) Correction for the radius of the indenter tip is effective in evaluating hardness both under test load and after unloading for any indenter. (2) Correction of errors due to "shallowing" of concavity is also effective under 10 mN load.

View full abstract

This study deals with an automatic scraping machine, namely a bearing recognition mechanism of scraped surfaces and a specially designed scraping mechanism. A scraping mechanism with the performance as skilled workers was tried to develop. The research is as follows; first the manual scraping operations were investigated, and secondly, a scraping unit was newly designed and made after manual actions. Many experiments were made to explain the performance of this mechanism, such as cutting force and surface form after cutting under various cutting condition. Physical phenomena in cutting were explained using a theoretical model. This mechanism performs smoother surface than skilled workers. For predicting the cutting force and shear angle the "maximum shear stress theory" was useful. These will be helpful to develop a tool movement unit and predict scraped depth quantitatively.

View full abstract

The purpose of this investigation is to clarify the edge-chipping generation-mechanism of ferrite in slot grinding. By releasing the grinding wheel quickly from the contact zone during the grinding operation, changes of edge-chipping size and of kerf width are measured. Edge-chipping size tends to decrease greatly in the grinding zone, while kerf width tends to increase greatly in the same zone. The changes of edge-chipping size and of kerf width are opposite. The sum of edge-chipping size and kerf width in the grinding zone shows a nearly constant value. It is seemed that the initial chipping which occurred at the tip of ground groove is continuously ground with small depth of cut by the side of the grinding wheel and that the initial chipping tends to be decreased by an increase of kerf width. Especially, in down cut and low work speed the apparent function of increasing of kerf width is shown.

View full abstract

In grinding of axisymmetrical surfaces, residual removal increases in peripheral region. Therefore, the surface profile becomes concave and the form accuracy deteriorates. In this paper, the mechanism of generating axisymmetrical form error is made clear by analyzing the change of grinding force. The following facts are revealed by experiments. (1) In case of traversing of a wheel from outer to inner, enormous normal grinding force in the periphery results in large residual removal. (2) Increase in the normal grinding force is considered to be the result from direct contact of bond and workpiece in grinding process. (3) By traversing of a wheel from inner to outer, the form error reduces less than 120 nm p-p.

View full abstract

This paper describes the development of a high-frequency wheel spindle for an ultra-high speed surface grinding machine. This project aims to achieve a high efficiency grinding of steel with cBN wheels, in which a stock removal rate is over 300 mm³/(mm·s) at a wheel peripheral speed of 400m/s. At such a speed, the power loss caused by air and coolant friction is larger than grinding power. At first, a theoretical analysis of grinding forces was carried out to estimate the grinding power. From this analysis the wheel diameter of φ250mm and motor power of 22kW at 30000min^-1 were decided. Also, an eigenvalue analysis by FEM was carried out to decide the diameter of the spindle shaft that avoids the resonance caused by a heavy wheel. In the spindle unit, a newly developed lubrication method called the "retainer lubrication" was adopted to achieve dn value of 195 × 10⁴. The retainer lubrication reduces the no-load power and the temperature rise of bearings. Furthermore, constant pre-load mechanism was adopted to achieve a high stiffness without burning out the bearings. Finally, the advantage of a small diameter wheel is confirmed experimentally by measurement of air friction loss, which conformed to calculated results using the logarithmic law.

View full abstract

Generally, XYΓ positioning device is fabricated by piling up X, Y and Γ table separately, however this construction makes the positioning device complicated and not compact. This paper describes a new method of XYΓ positioning device consists of four piezo elements with a weight and a center slider. Through controlling the deformation of piezoelectric actuators, step displacement in any directions can be achieved by using friction force and generated inertial force. This device has simple and compact structure. There is also no limitation of the traverse length with fine resolution in any directions. The motion in XYΓ directions was demonstrated through application of controlled voltage waveforms. Minimum step sizes were obtained of 0.05 μm in either X and Y directions and 0.65'' in Γ direction. The XYΓ motion can be controlled by adjusting applied voltage waveforms of four actuators. As the result, motion errors decreased from 1.0 μm to 0.03 μm in shift and from 20'' to 0.62'' in yaw with 25 μm traverse length in X direction. The positioning device moved in desired direction by step displacement in X and Y displacement were repeated sequentially.

View full abstract

A non-destructive testing technique for ceramics cracks has been developed. The method utilizes the phenomenon of slow evaporation of penetrated solvent inside the crack. The experimental apparatus consisted of a sealed chamber and a specimen locating device. The FID (Frame Ionization Detector) was fixed on the chamber for measuring the evaporated solvent. Three kinds of experiments were carried out. The 1st experiment was to determine the best condition for FID operation and to measure the FID characteristics. From this experiment, the optimum FID operating conditions were established; and the dynamic characteristics, e.g. FID response time of 0.17 s, were determined. The 2nd experiment was to measure the sensitivity of the crack detection. For this purpose, a jig, holding two lapped square ferrite blocks with a spacer in between, was used. The thickness of the spacer was from 1 to 10 μm. The experimental results show that the FID output varies linearly with the gap width for measurement carried out approximately 70 to 90 s after solvent introduction. The 3rd experiment was to detect the cracks on actual ferrite magnet specimens. The result shows that cracks with width 1-3 μm could be detected, with SN ratio 3.

View full abstract

The equipment which measures a diameter of small and deep hole whose aspect ratio is about 80 has been developed. The light projected to a specimen makes a small spot in a small hole. The position of wall in a hole is recognized by detecting an image of small spot reflected at the wall. Taking a ferrule (the hole diameter=125.0-126.0μm, the length=10mm) which is used for optical connector as a specimen, the hole diameter at a height of 5 mm from the end is measured using this equipment. The measured diameters are in the region of 125.8-126.1μm. The true value of diameters are in the region of 125.9-126.0μm. Therefore, the measuring errors are within ±0.2μm.

View full abstract

This paper describes a new automatic calibration system for the polygon mirror. A new differential method which can calibrate simultaneously and autonomously the dividing errors of a polygon mirror and angular profile of surface is proposed. The basic principle of the system depends on the self-check-method of two autocollimators. This system consists of two angle sensors with 0.02 s (arc) resolution as a differential probe, a rotary table for giving angular displacement and an XZ-axis stage for scanning surfaces of the polygon mirror. As an example for the experiment, a 24-sided polygon mirror has been calibrated on the system and its effectiveness has been confirmed. It was found that the repeatability error of calibration was about 0.2 s (arc).

View full abstract

The surface roughness parameters in ISO are obtained according to the mean-line system based on the high-pass digital Gaussian filter called the phase-corrected-filter. As the discrete data are used, the transmission characteristics of filter have the error caused by the inadequate range of convolution integral and space of data. This paper deals with an optimum condition of the high-pass digital filter for the roughness profile. The new knowledges of this paper are as follows. (1)The allowable error of transmission characteristics is determined as 2% at the point where the fluctuation of roughness parameters increase radically. (2)The range of convolution integral is given by 0.9 times the cut-off length. (3)The condition applied to the filter by this paper is compared with that recommended by ISO. (4)The roughness parameters by the mean line system differ little from those by the center line system which is a traditional method except S_k and S_m.

View full abstract

A miniature friction drive 3-D. O. F. actuator has been developed for downsizing of mechanical systems. This actuator consists of a spherical rotor and several drive units. Each drive unit has a truss structure composed of three piezoelectric actuators. The rotor is driven by frictional force generated between the rotor and the drive units. In this paper, the spherical actuator design is referred. And then, its performance is verified by some experimental results: the maximum torque is 2.5 N·cm, the maximum rotational speed is 560°/s. Moreover, the rotational angles are controlled using a position sensing detector experimentally and these experimental results almost coincide with computational simulation results.

View full abstract