This paper discusses a structural optimization method that achieves optimum shape and topology based on the phase field method. The proposed method has the same functional capabilities as a structural optimization method based on the level set method. Since the proposed method does not require extra operations such as re-initialization of the level set function or smoothing of sensitivities, the computational cost is lower than that of typical level set methods. Structural shapes are represented by the phase field function defined in the design domain and optimization of this function is performed by solving a time-dependent reaction diffusion equation. The artificial double-well potential function used in the equation is derived from sensitivity analysis. The proposed method is applied toa two-dimensional linear elastic problem. The provided numerical examples illustrate the convergence of the compliance minimization problem.

View full abstract

During car frontal crash, crash energy is absorbed by the parts of front bumper, front side member, front panel member and so on. Previous research has indicated that front side member plays major role in energy absorption. For protecting the passengers, the front side member is expected to absorb crash energy as much as possible. In this study, we adopt cylindrical thin-walled structure using origami engineering as front side member instead of structure with box-shaped cross section which is generally used. We develop an optimization system of the cylindrical thin-walled structure using origami engineering, in which the objective function is to maximize the energy absorption of origami structure; the design variables are structural parameter, number of divisional sections along axis, number of edges of polygonal cross section and number of subdivision levels; the mass and initial peak load of optimal structure must be less than those of structure with box-shaped cross section. We then discuss the optimization results that the optimal structure is capable of absorbing energy 91% more than that of original box-shaped cross sectional structure which is usually bended on the way of being crashed, 37% more than that of original structure which is ideally crashed to bottom without bending.

View full abstract

In the present paper, some experiments for press molding and numerical simulation about micro press molding of glass devices using finite element method is investigated. Thermo-viscoelastic properties of the glass materials were estimated using unidirectional compression creep test based on traditional thermo viscoelastic theory. In this study, Pyrex and D263 were used as glass materials. Glass micro press molding was carried out with Glass-like Carbon (GC) mold given to Line & Space patterns machined by dicing. The adaptive condition of the molding temperature which given appropriate transcription profile of the glass was investigated. Moreover, numerical simulation for micro press molding of the glass was carried out by finite element method using universal FEM code (ANSYS ver. 11.0). As a result of comparing experimental results with numerical ones, the cross section shape and the height of groove profile obtained by FEM simulation approximately agree with experimental value.

View full abstract

In this paper, the collapse behaviors of rectangular tube subjected to pure bending are studied by using the finite element method. Such bending collapse has been studied for a long time, including the landmark study by Kecman. According to these studies, there are two types of collapses. The first type is a collapse due to buckling at the compression flange. The second type is a collapse due to plastic yielding at the flanges. However, there may be another collapse. For a rectangular tube in which the web is wider than the flange, it is found that collapse due to buckling at the compression web may occur. Further, an approximation prediction methods is proposed for estimating the maximum bending moment of rectangular tubes in which the web buckling is also taken into account. Its validity is verified by comparing with the numerical results by FEM under various conditions.

View full abstract

Alternative three-dimensional finite element for piezoelectric problems is formulated with three restrictions: no volumetric charge, equipotential electrodes, and no surface charge except for the electrodes. In the conventional formulation, independent variables are mechanical displacement and scalar electric potential so-called voltage but in the present formulation they are mechanical displacement and vector potential that derives electric displacement. From thermodynamic viewpoint, the present formulation is better for nonlinear ferroelectric problems and the resulting global stiffness matrix is at lest semi-positive definite. For descritization, nodal-edge finite element is used to keep the arbitrainess of vector potential in null-space of the matrix. Numerical examples of three-dimensional piezoelectric problem are demonstrated. For solving the linear systems, direct method with tree-cotree decomposition gauge and conjugate gradient iterative method are used.

View full abstract

Head Finite Element (FE) models, which represent potential Japanese male head shape, were created using Multi Dimensional Scaling (MDS) and Free Form Deformation (FFD). The models were then used in simulations of translational impacts, rotational impacts and combinations of both in order to determine the effects of individual differences on the pressure and shear stress responses of the brain. Individual differences of head shape have significant effects on the pressure response in translational impact simulations, and on the shear stress response in rotational impact simulations. Differences of maximum and minimum values of the pressure and shear stress responses cannot be neglected in comparison to the tolerance levels of brain injury previously reported. In combined impact simulations, the differences in pressure response increase significantly, and the substantial influence of rotational impact on pressure response was also confirmed.

View full abstract

Diffusion theories of hydrogen subjected to a constant force are studied and entablished. First, the Einstein relationship, which connects ensemble averages of mean displacement and mean square displacement to the diffusion coefficient, is derived via formulation of dynamics theory based on Langevin equation. Then, the jump theory based on the stochastic process is applied to the jumps of hydrogen in face-centered-cubic crystal and the diffusion coefficient is expressed in terms of the average of expectation value of the number of jumps. We clarify that both theories agree well with each other under some appropriate assumptions. Secondly, we show that these theories are validated by using a molecular dynamics simulation. The result of microscopic information such as displacement and the number of jumps obtained by the simulation is universally represented and a unique value of diffusion coefficient can be determined with comparing the theories and simulation.

View full abstract

Electro deposition technology has been utilized widely in various industries. However, the internal stress generated in the plated film during deposition often results in peeling and crack of the thin film. The investigation of generation mechanism of the internal stress is indispensable for improving the plating film. Measurements of internal stress are required during the early stages of film growth. In this study, TV holographic interferometry, which can capture holographic images at TV frame rates, was used to sensitively measure with the accuracy of λ/2 the deflection of the cantilever beam during the deposition on the substrate. Internal stress is calculated in-situ by substituting the deformation date in Stoney's equation. In this experiment, to examine the influence of the substrate on internal stress of electroplating, the epitaxial growth of plating was compared with the plating growth on an amorphous substrate. Next, internal stress was measured in various plating (Zn, Cu and Ni) without the additive. The generation mechanism of the internal stress was considered by the surface morphology (SEM) and the cross-sectional (TEM) observation of the plating film. The main results are as follows. 1) Internal stress during the early stage of film growth changes greatly on the surface of amorphous substrate compared with the epitaxial growth. But, the difference of both becomes small as the film grows up. 2) In general, internal stress rapidly changes in the initial stage of film growth, and changes gradually afterwards. 3) The compression stress is generated in Zn plating, the tensile stress is generated in Ni plating and in Cu plating, a tensile stress is generated in the initial stage of film growth, which immediately changes to compressive stress.

View full abstract

Higher frictional force between skin and air bag influences greatly the scratch damage of skin when an air bag was inflated and slid against skin. Reduction of the coefficient of friction is therefore desired. In this research, we proposed a new method to reduce frictional force by means of producing air lubrication between air bag made of non-coated fabric and human skin. The air could be generated, and the device which could measure frictional force was produced experimentally. Frictional force of the air bag with air was measured, and the effectiveness and efficiency was confirmed. When there are the air, the friction disk materials and fabric materials and fabric structure have nothing to do with frictional force and a coefficient of friction. The coefficient of friction is influenced by air mass flow passing fabrics.

View full abstract

The manufacturing method of resist film insert metal powder injection molding (MIM) namely Resist/μ-SpiMIM process has been proposed. The resist film made of PMMA polymer with numerous micro-holes was used as micro sacrificial plastic mold (SP-mold). A novel molding machine which could achieve to mold without pelletizing was also used to produce the green compacts with a high efficiency in experiment by using a small amount of feedstock. The feedstock prepared by pure copper powders with various particle sizes was injection-molded into the SP-mold and the sintered parts with micro-pillar structure were produced after debinding and sintering process. The effects of particle size on density, shape-transcription and surface roughness of those sintered parts were discussed. The decreasing of particle size results in a marked improvements of surface roughness, transcription and dimensional variation of sintered parts.

View full abstract

Suction tests are carried out under the various molten metal temperature, suction pressure and cavity thickness conditions; and the effects of these parameters upon the fluidity length and the time taken for flow stoppage are investigated. The time taken for flow stoppage is determined by the pressure change comparison method that is proposed in the present study. The results are as follows. In a high suction pressure condition, where the inertial force is relatively greater than surface tension, a projection is formed at the front of molten metal flow. The fluidity length from the bottom to the tip is increased monotonously with an increasing in the suction pressure; however, an increase in the length between the bottom and the root of the projection becomes gradual in the high suction pressure condition. The time taken for flow stoppage is decreased as the suction pressure increases. The fluidity length and the time taken for flow stoppage are expressed as non-dimensional correlation equations.

View full abstract

AZ91D plates of 5mm in thickness are made by semi-solid molding process, and its relative density and bending strength are investigated under various barrel temperature and gate velocity conditions. The results are as follows. In cases that the plates are made by semi-solid molding process, the relative density can be expressed as a function of the apparent Reynolds number that considers the influence of the solid fraction. At a given apparent Reynolds number, semi-solid molding process can achieve a larger relative density than conventional completely melted process; and the difference of the relative density between these two processes increases as the apparent Reynolds number increases. The bending strength increases as the apparent Reynolds number increases, however it decreases when the decrease of the relative density becomes remarkable. The difference between the maximum and the minimum bending strength of five trials at a given molding condition becomes remarkable when the apparent Reynolds number exceeds a certain value. Therefore, there exists an optimum apparent Reynolds number that can achieve a high strength stably. It is also found that a lower barrel temperature condition is required to obtain both higher relative density and greater bending strength.

View full abstract

Low-density porous materials like sponge have various abilities as functional solid in mechanical design. Especially the remarkable nonlinearity in forming process causes difficulty in numerical analysis of the mechanical behavior of the porous materials. In this paper, the nonlinear deformation process of the low-density porous materials is represented by using the constitutive equation incorporating structure transformation. The porous material has the complicated cellular structure mesoscopically, and then two structures can be defined with and without the cellular structure. The structure transformation which loses the cellular causes the volume change. The adopted constitutive equation includes the valuable of volume fraction, but large change of the volume needs the idea of nominal and true volume fraction even if only nominal one is used in the analysis of conventional transformation problems. Two types of low-density porous materials are inspected to verify the ability of the equation. One has reversible character like typical elastic sponge, and the other has irreversible character like expanded polystyrene. The equation is applied to represent the experimental results of compression testing on two materials. Then it is shown that the adopted constitutive equation has the ability to represent the mesoscopical change of the low-density porous materials macroscopically.

View full abstract

Spherical indentation testing is used to measure the mechanical properties of materials because of its convenience and low-invasiveness. A method based on the Hertz theory is one of the wellknown methods used to verify the reliability of this testing technique, but it cannot be applied in the case of soft materials with a finite thickness and large deformation. In this paper, Young's modulus measurement using equivalent indentation strain in spherical indentation testing is carried out to evaluate the applicability of this measurement method for various specimens such as soft materials, biological tissues, and so on. In particular, the applicability of this method is validated by using some type of indenters and materials under different experimental conditions. The effects of the types on the relationship between the specimen thickness, indenter diameter, and specimen hardness are studied to formulate a function of Young's modulus and the diameter. The formulated results are evaluated by comparing between experimental and simulation results, and it is shown that the Young's modulus of soft materials with a finite thickness can be measured by the spherical indentation testing technique with the measurement method using equivalent indentation strain.

View full abstract

In this study, evaluation method for fatigue strength of cast steel under variable stress amplitude conditions in addition to constant stress amplitude condition was investigated. Fatigue tests were carried out using round bar specimen of SCW480 cast steel. As the experimental results, fatigue strength property was greatly sensitive to the intrinsic flaws of cast steel and stress patterns and it was difficult to determine fatigue evaluation curve using conventional S-N curve. Therefore, the predicted fatigue limit of flawed material, which was obtained based on threshold condition of small crack growth, was used as the parameter considering the intrinsic flaw size in order to eliminate the effect of the flaws to fatigue strength property. From S-N curve modified by the predicted fatigue limit, it was found that there was little variability in fatigue strength property. In addition, variable stress amplitude test results showed shorter fatigue life and constant and variable stress amplitude test results were determinately separate. Their plots could be represented by different two curves and fatigue evaluation curves considering variable stress amplitude conditions could be obtained without depending on not only the intrinsic flaw size but also variable stress patterns.

View full abstract

This paper shows phase characteristics of crack indications obtained with a 48-channel ECT system, and proposes a method to reduce noise and extract crack indications from detection signals of this ECT system based on these phase characteristics. Because this ECT system performs two kinds of scanning patterns: U-scan and T-scan, and obtains two sets of signal distributions for one scanning area simultaneously, the proposed method uses a two-stage process that includes so-called a main filter and a sub filter. While the main filter uses the phase information of the same scanning pattern by which the target signals are obtained, the sub filter uses the phase information of the other scanning pattern for the target signals. Using these two filters imposes more strict conditions on extracting crack indications so as to reduce more noise. The results of a round robin test shows that this filtering method is effective to extract crack indications from the background of complicated noises.

View full abstract