The purpose of this paper is to derive a general theory of design from experimental results of design by the method of structural analysis of them. A specification is given to designers. They discuss it, analyse it, propose tentative design solutions, and finally get to a conclusion which satisfies the specification. The conversation is recorded and reviewed to extract rules governing the thought process by which they converge to the conclusion. This experimental method is found useful to derive the general theory of design. Four models of designing process are extracted : (1) function-attribute transition model, (2) Markovian model, (3) flow diagram model, and (4) logical structure model. These represent characteristic aspects of designer's designing activity respectively. They reveal the merits and demerits of a designer, from which some proposals for improving the design efficiency are obtained, concerning the design process, such as sequence of design steps, amount of information given at each step, distribution of evaluation steps among others, etc. These proposals are useful for practical construction of CAD systems. The models are also compared with deductive models constructed by the axiomatic approach. Good correspondences are found between them.

View full abstract

The fracture strengths of grinding wheel materials with vitrified bonds are theoretically analysed. The matrix is idealized as a system composed of spherical grains with bonds, an ellipsoidal force distribution around a grain is assumed, the relation between the microscoptic force acting on the adja-cent grains and the macroscopic stress is deduced and the inter- and trans-grain fracture conditions are derived. The following conclusions are obtained : (1) The intergrain tensile fracture strength is proportional to the strength of bonds and the square of the ratio of the radius of bond and the distance between the adjacent grains. (2) The intergrain shear fracture strength depends on the properties of bonds. If the bonds are subjected to the shear fracture according to the maximum shearing stress theory, the fracture conditions similar to those of von Mises hold, while if they are subjected to the shear fracture according to the Mohr failure envelope, so is the matrix. (3) The transgrain fracture strengths are given by the products of the grain strengths and the volumetric ratio of grains.

View full abstract

The surface topography produced by the electric discharge machine is analyzed by a stationary stochastic process and the power spectral moment approach. It is shown that the topography can be represented by the Gaussian isotropic surface and its statistical characteristics can be found from a fourth order continuous autoregressive model. The statistical properties such as the wear land, contact area, number of summits per unit area, number of peaks per unit length, surface gradient, spectrum width and the area under the peaks (excess area) are calculated. In particular, the mean number of peaks, expected number of zero crossing per unit length are shown to agree well with theory. The effect of experimental conditions on the surface roughness is also discussed in this paper.

View full abstract

As integrated circuit complexity increases, automatic lithographic mask inspection becomes necessary, because of the limitations of human inspectors. It is especially difficult to detect defects on opaque samples, such as X-ray masks and wafers. This paper describes experimental research on an automated inspection technique for opaque sample defects. The light from a flying spot tube is focused on and used to scan two neighboring chips of a sample. The reflected light from each chip is detected and the outputs of each are subtracted. When the difference between the two signals exceeds floating threshold level, a detect signal is generated. The results obtained from experimental investigation are as follows : (1) Pattern detecting resolution by direct scanning on the wafer is about 0.8 times the diameter of the scanning spot. (2) In case of weak light detection, signal to noise ratio to determine the detecting resolution can be estimated by the shot noise of the light source. (3) To improve the defect detection coefficient, it is effective to multiply logically the two-valid signals of the two scanning lines. (4) All defects larger than 1.5 μm on high contrast samples were inspected. It was necessary to make use of diffused reflection light detection in case of low contrast samples.

View full abstract

From the basic study on the rolling friction the main causes for rolling friction of linear ball bearing can be assumed as follows. (1) The variation of rolling friction which is due primarily to waviness and roughness of the rolling surfaces, and to the angle of relief of the bearing raceway. (2) The inherent bearing friction by the sliding friction which is due to the spinning between the rolling elements and the retainer except the variation in the friction force. In this study, the friction forces of the linear ball bearings in a low speed range by the factor (2) except the variation by the factor (1) were obtained from the recorded waves of reciprocate measurement of forward and reverse rolling. Moreover to clarify quantitatively the characteristics of the individual friction forming the rolling friction of this bearing, some experiments on the rolling friction behavior using a ball were performed. From the results of the above experiments and analyses, the friction characteristics of linear ball bearing can be explained. The sliding friction due to the spinning accounts for 70 to 75% of the total friction, and the pure rolling friction accounts for about 10 to 15%.

View full abstract

In this report, the piston-slap in the approximately balanced rotation-reciprocation device is compared with that in the known one. The device in which the explosion pressure in a reciprocating engine can be simulated qualitatively, and the photoelectric detector of the angular displacement of piston are newly developed for this study. And it is ascertained that their performances are sufficient for the purpose of this study. Then, by utilizing them, the piston-slap in this device and the known rotation-reciprocation device is measured, and their piston side thrust forces are calculated. From these results, it is ascertained that the piston-slap in this device is far smaller than that in the known device essentially. Especially, the application of this device to high pressure reciprocating compressors and engines is considered to be effective to reduce piston-slap.

View full abstract

In order to guard against thermally induced damage of a ground surface and improve on the grinding accuracy of workpiece dimension, some knowledges on the partition ratio theory in regard to the heat generated in grinding are necessary. In this report, a two dimensional high speed cutting model with the rake face bearing a large negative angle is used and the distribution of temperature around the cutting edge is calculated numerically. Consequently it is confirmed that most of the heat generated in the rake face conducts into the workpiece and only part of the heat generated conducts into the abrasive grains and chips and that the average temperature of the chips, rake face and shear plain do not always have the same value. The theory of the partition ratio is established in consideration of the above mentioned results and the numerically calculated results of heat conduction into the abrasive grains with two dimensional nonstationary model and the influence of temperature rise of wheel-work interference zone by preceded cutting edge. In the case of grinding hardened steel by alumina wheel, the partition ratio theory thus confirms that the ratio of the heat conducted into the workpiece is about 70%. This ratio decreases with increase in the depth of cut and work speed but is hardly influenced by the wheel speed and grain size.

View full abstract

In machining it is a basic theme to make a wide plane accurately. For this purpose a milling machine is used, however, it is usually difficult to obtain good plane because of presence of stair between neighboring cutter passes and of tendency of concavity even in one cutter pass. The study aims at evaluating such machining errors in face milling. Considering that the deformation of machine tool due to cutting force and the thermal expansion of tool cause machining error mainly, the analysis was made provided that by using a face milling machine with relatively good geometrical accuracy machining is done within a short time the thermal deformation of machine tool structure has nothing to do with. Predicted values of machining error from data about machine tool structure and cutting condition correspond to measured ones in cutting experiments quantitatively. As a result it becomes apparent that the obliquity of cutter axis owing to deformation of machine tool structure by cutting force causes the concavity of plane and that the change of cut of depth due to thermal expansion of tool shapes stairs between neighboring passes.

View full abstract

A new specular reflection measuring device by means of a concentric distribution optical fiber bundle is developed to measure automatically specular reflection of a small area from the polished free-form metal surface of a mold cavity and data processing system is also developed by using a microcomputer AIDACS-3000 E. The automatic specular reflection measuring device which enables us to estimate the degree of surface integrity of a mold cavity has the following properties : (1) The optical fiber bundle put into the scanning probe is rotated around the two axes. One is the main spindle axis of the NC milling machine and the other is perpendicular to the machine spindle axis. Rotation around the perpendicular axis is converted into reciprocating motion. The specular reflection of the free-form metal surface can be measured easily by these two axes rotation mechanism. (2) The degree of surface integrity of the free-form metal surface can be estimated in the three grades of surface roughness of under 0.1 μmR_max, about 0.4 μmR_max and over 1 μmR_max. (3) After the free-form metal surface is finished by NC milling machine, the specular reflection measuring device is put on the spindle of the same machine and the surface integrity can be estimated directly on the machine.

View full abstract

This paper reports the effect of surface asperities on the approach between a sphere and a plane surface in contact. The relation between the approach and the normal pressure was experimentally obtained by locating symmetrically three steel balls between two rough planes and by loading normally on rough plane surfaces. The results are as follows. (1) The experimental results of the approach are well approximated by the theoretical ones computed after the first report process. (2) The approach increases due to surface asperities in contact. (3) The approach depends scarcely on the hardness of contacting bodies. (4) The effect of surface asperities on the approach is considerable for lower normal load and smaller diameter of contacting ball.

View full abstract