An investigation has been carried out concerning the effects of water absorption on the mechanical properties and fatigue strength of two types of unidirectionally reinforced carbon/epoxy composites, i. e., 130°C cure-type composite T-1/347 and 180°C cure-type MM-1/982 X. T-1/347 composites absorb more water than MM-1/982X composites. After long immersion of about seven months, however, MM-1/982X composites absorb water rapidly, and the fatigue strength is considerably decrease. In the case of wet specimens of T-/347 composites preconditioned in water for two months, high amount of water absorption degrades the matrix and decreases the strength of the fiber/matrix interface, thereby decreasing the mechanical properties. For MM-1/982X composites, small amount of water absorption moderately decreases the interfacial strength and increases the ductility of the matrix, thereby increasing the mechanical properties. However, fatigue life of wet specimens of both T-1/347 and MM-1/982X composites is decreased in water, but the one of dry specimens proconditioned in laboratory air is increased in water. In these tests, the synthesized evaluation of acoustic emission (AE) signals using several AE parameters and AE signal location have been done, and the fracture mechanisms are discussed in terms of the observations of internal damages by a scanning acoustic microscope (SAM) and of fracture surfaces by a scanning electron microscope (SEM).

View full abstract

As there are very few studies on fracture mechanisms of short fiber reinforced composite materials, this study is conducted for the purpose of clarifying the fracture mechanisms of short fiber reinforced polyethylene terephthalate (FRPET) by the acoustic emission method in the cases in which the loading direction to the injection direction and the fiber content are different. It is found that in the test specimens in which the angles between the loading direction and the injection direction are 0°(parallel) and 90°(perpendicular), the resin cracking is dominant damage mode, and that in the test specimen of 45°, fiber breaking is dominant. It is also found that in the test specimens of lower fiber content by weight of 15% and higher content by weight of 45%, fiber debonding and pulling-out is dominant.

View full abstract

Effects of the loading direction and the fiber content on the fatigue crack growth behavior of FRPET are studied by the acoustic emission method. From the behavior of AE event count rate, the fatigue crack growth process is divided into three zones similar to the three stages of the fatigue crack growth curve. Many AE generations under the mean load phase suggest nearness to the final fracture. Three kinds of damage modes occur at all zones in a specimen for which the angle between the directions of loading and injection is 0°(parallel), but not in the specimens of 90°(perpendicular) and 45°(oblique) angles. Fiber breakage is a dominant damage mode in all zones of the fatigue crack growth process in specimens of fiber contents of 15 and 45 % by weight.

View full abstract

In this parer, rotating bending fatigue tests were carried out on injection moulded carbon-fiber-reinforced polyamid 6.6 composites. The fatigue mechanism in the composites was clarified through successive surface observations using the plastic replica method. Moreover, strain concentration near the end of the fiber was measured.

View full abstract

Impact Fatigue and usual fatigue experimenls were carried out by using carbon steel and high-manganese steel specimens. Impact fatigue lives for crack initiation at the notch roots and for fracture were unfavorable compared with usual fatigue. Although crack propagating periods under pulsated and repeated loadings were comparable in the case of usual fatigue, difference of such a loading condition influenced on both crack initiation and propagation period in the case of impact fatigue. According to the result of measuring of the dislocation density in terms of the Xray method, plastic zone size near a crack was estimated to be small in the case of impact fatigue compared with that of usual fatigue. Similar results were obtained for high manganese steel, and the influence of rolling direction was not observed.

View full abstract

Fundamental investigations were made as to grain growth phenomenon in copper electroplating foil under a variable amplitude stress which consists of overstress and understress. The applicability of the foil to fatigue gauge that detects fatigue damage was discussed. It was found that the modified Miner's rule does not hold for the initiation of grown grains under such conditons as mentioned above. The cumulative cycle ratio is dependent on load histories. It was also found that fatigue damage under two-level multiple loading increases as the cycle ratio of overstress to understress increases. This indicates that fatigue damage during understress applied after overstress is greater in the early period of the cycles. Furthermore, it was found that the density of grown grains initiated under variable stress is dependent on the equivalent stress. This enables the evaluation of the equivalent stress based on the density of grown grains.

View full abstract

In this paper, rotating bending fatigue tests were carried out on plain specimens of austenitic stainless steel SUS304 under water cooling. The processes of nucleation and propagation of cracks were clarified by successive surface observations using the plastic replica method. The main results obtained are as follows. (1) Fatigue cracks are nucleated in very small regions. (2) The fatigue crack growth rate is proportional to the crack length in a limited range. (3) The fatigue limit corresponds to the crack initiation limit.

View full abstract

In order to investigate the effects of small defects and nonmetallic inclusions on the fatigue strength of high-strength steels under tensile or compressive mean stress, fatigue tests for a maraging steel with the static strength level of 2500MPa (H_v=740), were carried out. The experimental data on a 0.13 % C steel (H_v=105) as well as the maraging steel are analyzed from the viewpoint of the previous study in which the prediction equation for the fatigue strength under zero mean stress (R=-1) was proposed on the basis of the Vickers hardness H_v and the square root of the projection area (√(area)) of a defect or nonmetallic inclusion. The new equation is proposed to predict the fatigue strength under tensile or compressive mean stress. The proposed equation gives not only accurate prediction of the fatigue strength of a high-speed cutting steel SKH-51 (H_v=615, 654) which fractured from nonmetallic inclusions, but also can be used to predict the lower limit of the wide scatter of the fatigue strength of many specimens.

View full abstract

This paper deals with a method to estimate numerically the reliability of fatigue sensitive structures with respect to the fatigue crack growth. A method is proposed to experimentally determine the Probability distribution functions of material parameters of the Paris law, da/dN=C(ΔK/K₀)^m, using stress-intensity-factor-controlled fatigue tests. The auto-correlation function of the resistance to fatigue crack growth, 1/C, is also estimated from the experimental data. The results of a high tensile strength steel show that the distribution of the parameter, m, is approximately normal and that of 1/C is a 3-parameter Weibull. The merit of the proposed method is that only a small number of tests are required to determine these functions. The probability distribution of the fatigue crack length after a given number of load cycles or the number of load cycles for a crack to reach a given length can be estimated by simulation of non-Gaussian random processes with these functions.

View full abstract

The diminution of fracture ductility of smooth specimens and small cracked specimens on surfaces subjected to torsional prestrains was investigated. For the smooth specimens, the transition of tensile-fracture ductility after a critical torsional prestrain is independent of the existence of microcracks and attributed to the embrittlement of the surface thin layers of the specimens due to the formation of anisotropic helical structures. For the cracked specimens, the degree of fracture ductility depends on the area of the crack and the magnitude of torsional prestrain. In the case where the torsional prestrain is lower than a critical value, the surface crack starts in the direction close to the maximum shear stress, and ductility remains within the specimen. If the torsional prestrain exceeds a critical value, the specimen is broken in a brittle manner, and the crack extension shows the helical pattern. This behavior is due to the embrittlement of the specimen surface layer.

View full abstract

An elastic-plastic dynamic line-spring model is proposed for the simplified analysis of impact response of a part-through surface crack in an elastic-perfectly plastic plate. Bending of the plate is treated based on Mindlin's theory. Stiffness of the dynamic line-spring is evaluated by considering the growth of the plastic zone in the remaining regament and the unloading. The crack tip opening displacement, CTOD, for a semielliptical surface crack in a plate subjected to impact loading at its ends is calculated by using the present model. It is found that CTOD takes its maximum value at the deepest penetration point of the surface crack and the maximum value is larger than that in the static case.

View full abstract

Pulsed holography is applied to take an instantaneous photograph of a crack front propagating through a PMMA plate specimen at a speed of several hundred m/s. The cracks are in the opening mode and not interacting with any stress waves. A pulsed holographic optical system records and reconstructs a crack front observed through a side boundary of a specimen, and the reconstructed image is photographed. A method is developed to decrease an aberration introduced by inhomogeneity of refractive index in PMMA, where the inhomogenity results from manufacturing process of the PMMA. Also discussed are effects of changes of specimen thickness and refractive index in the strong stress field near a crack tip. From an obtained photograph one can know the shape of a crack front, which is not always symmetric. One can moreover measure the two angles between the crack front and the two specimen surfaces. The angles are important quantities to understand the three-dimensional structure of the stress field around the crack tip. The measured angles are similar to the numerical results given by Bazant and Estenssoro.

View full abstract

In order to examine the effect of environment and polyaxial stress on the subcritical crack growth parameters of glass ceramics, constant stressing rate tests were conducted by four-point bending of plates (uniaxial stressing) and by Concentric ring loading of disks (equibiaxial stressing) under different environments of vacuum, air, and ion-exchanged water. The crack growth parameters were determined from a straight line on a logarithmic plot of fracture stress as a function of the stressing rate. In the air environment the parameter n was obviously lower for the equibiaxial tension than for the uniaxial tension, whereas in the water environment no difference was observed between the two modes of stressing. In addition, the n values in water were higher than those in air for both modes of stressing (especially for equibiaxial stressing mode). The higher values of n in water can be attributed to the fact that the resistance to fracture was increased by a rise in the extent of microcracking in the microstructure at the crack tip as the crack grew longer at the low stressing rate. The predictions of lifetime were made for this glass ceramic using the crack-growth parameters obtained by the stressing rate technique.

View full abstract

The critical stress σ_wi for crack initiation from a small crack and the effective threshold stress intensity factor range ΔK<eff.th> of 70/30 brass were determined by the annealing method. The specimens containing small cracks were annealed in order to relieve residual stress. Subsequent fatigue testing was performed again at different stress levels. The length of the nonpropagating crack Δl from the initial crack was measured, and by extrapolation to Δl=0, the critical stress σ_wi for fatigue crack initiation from an ideal crack and ΔK<eff.th> were determined. The experimental results are compared with those of a medium carbon steel (0.46%C) which were obtained in the previous study. The values of ΔK<eff.th> for 70/30 brass are dependent on crack size in the range from 150μm to 2600μm, although those for 0.46%C steel were independent of crack size in the range from 100μm to 1100μm. The relationships between ΔK<eff.th> and the crack length l are expressed as follows: ΔK<eff.th> ∝ l^1/3 for 70/30 brass, and ΔK<eff.th> = const. for 0.46%C steel. The difference in the dependence of ΔK<eff.th> on crack size in 70/30 brass and a medium carbon steel may be attributed to the coaxing effect at the crack tip in carbon steels caused by strain aging. The dependence of ΔK<eff.th> on crack size in 70/30 brass must be considered a crucial factor which complicatedly influences the variation or scatter of the cumulative fatigue damage D under variable amplitude loading.

View full abstract

In this paper, the fracture surface of epoxy resin is investigated precisely with an interference microscope. In order to explain the characteristics of the fracture surface, the maximum roughness was measured by a technique of pattern analysis on interference fringes. The relation between crack propagation behavior and fracture surface roughness was also discussed. The results obtainedc are briefly summarized as follows; 1) Fracture surface roughness is precisely and quantitatively obtained by analysis of the interference fringe pattern. 2) The maximum roughness of the fracture surface shows dependence on temperature. A characteristic feature of roughness variation is seen just after the onset of crack growth. 3) The value of resistance for crack propagation J'₁ measured at 10 μm propagation just after the onset of crack growth, proved to be constant under any temperature condition. Thus, it is implied that the micromechanism of the local fracture at the propagating crack tip is influenced by molecular mobility which reflects time-and temperature-dependent behavior of microheterogeineties of molecular entanglement.

View full abstract

Long-term reliability againt thermal fatigue is required for solder joints of surface mount IC packages. This study is carried out in order to estimate the thermal fatigue strength of the solder joints of three types of leads, namely, batt, gull-wing and J-bend leads. Strain on the solder joint induced by thermal expansion mismatch between the package and substrate has been analyzed by considering elastoplastic behavior of the solder and by treating leads as rigid frames. It is found that the fatigue lives of these types of solder joints can be estimated based on the equivalent plastic strain range calculated by the present analysis. The maximum fatigue lives of the solder joints agree approximately with the fatigue lives of the bulk solder speciments. The fatigue strength of the solder joints depends on the lead and solder size, and the comparative meris of these leads for fatigue strength cannot be discussed unconditionally.

View full abstract

We have developed a single-edge notched beam technique using a very sharp V-notch for evaluating fracture toughness of structural ceramics. The technique has several advantages such as simple testing procedures, applicability to a wide variety of materials and suitability to high-temperature testing. Results obtained by the technique agree quite well with the values of SEPB (Single Edge Precracked Beam) technique for several ceramics by using the modification proposed by Takahashi et al. except for alumina.

View full abstract

This paper presents a methodology of direct experimental measurement of the T^* integral which has great potential as a nonlinear (elastoplastic) fracture mechanics parameter. A hybrid numerical-experimental method was developed to measure the T^* integral by the maximum size of reflected caustic patterns. To this end, the formation process of the caustic pattern for an elastoplastic crack-tip in a compact tension specimen was simulated by the previously developed finite element simulation technique aided by computerized symbolic manipulation. Experimental measurement of·the caustic pattern in the compact tension specimen was also carried out. Both simulated and actual caustic patterns agreed very well. The relations between the T^* integral and the size of causticpattern were obtained f or various optical setups.

View full abstract

Nonlinear finite element analyses of semi-elliptical surface cracks are performed under the fully plastic condition. The power-law hardening materials and the deformation theory of plasticity are assumed. Either the penalty function method or the Uzawa's algorithm is utilized to treat the incompressibility of plastic strains. The local and global J-integral values are obtained using a virtual crack extension technique for plates and cylinders with semi-elliptical surface cracks subjected to uniform tensions. The fully plastic solutions for surface cracked plates are given in the form of polynominals with geometric parameters a/t, a/c and the strain hardening exponent (n). In addition, the effects of curvature on fully plastic solutions are discussed through the comparison between the results of plates and cylinders.

View full abstract

The applicability of the latest CEGB R6 approach (the two-parameter approach) to fine kinds of metallic materials having various strength levels from high to low was examined. The construction of a new failure assessment diagram (FAD) was also attempted. The results obtained are summarized as follows: (1) Based on the FAD of the latest CEGB R6 approach, it was shown that various types of fractures from the linear-elastic fracture to the plastic collapse, including the stable crack growth of the elastic-plastic fracture, were predicted precisely for the five materials. (2) The applicability of the failure assessment curves (FAC) in the R6 approach was quite different for each material. Option 1 FAC did not give a conservative prediction for all the materials. Option 3 FAC did not have a high accuracy for some of the materials. (3) Based on the results obtained, Option 1 FAC was modified and a new FAC was proposed. (4) It was shown that the applicable range of each fracture criterion could be evaluated exactly on the new FAD incorporating the ASME screening criteria.

View full abstract

A microcracking fracture process zone is known to accompany tensile macrocrack propagation in rocks. The fracture process zone with nonnegligible length is suggested to be largely responsible for the dependence of fracture mechanics data on the specimen size and geometry. In this paper, an ultrasonic timing method is employed to detect the extent of the fracture process zone (FPZ) during fracture toughness tests of granite, where the travel time of longitudinal waves propagating through fractures is monitored using 500-kHz transducers with commercially available equipment. It is demonstrated that the travel time technique can locate the tip of the FPZ continuously during fracture experiments. The growth behavior of the FPZ observed by the ultrasonic method is shown to agree with the previously published results of numerical analysis that is based on the tension-softening model. Finally, a method is developed for determining the critical J-integral value, J_C on the basis of the length of the FPZ, and the measured J_c values are compared with those obtained by Li's proposed multiple-specimen technique. It is shown that the present method permits a valid fracture-toughness value to be evaluated by use of a single fracture specimen.

View full abstract

Applicability of the tension-softening model to characterize the fracture behavior of rock is examined. Tension-softening-curve measurements were made on three kinds of granite by employing Li's proposed J-integral-based technique, where the specimen-size effect on measured tension-softening curves was examined by testing CT specimens of several sizes (1.5inch∼6inchCT). The J-based measuring method is shown to be able to determine the size-independent tension-softening curve by use of laboratory-sized fracture-toughness specimens, providing that the specimen satisfies a minimum size requirement. A series of analyses of fracture-toughness tests on the granites are performed using the tension-softening law determined above, where the boundary-element method is used to simulate the overall fracture behaviors and fracture process zone growth. It is demonstrated that the load-displacement curves simulated for various sized CT and three-point bend specimens compare well with the experimental results, indicating the usefulness of the tension-softening model. These calculations are also used to propose a minimum size requirement for determining a valid tension-softening curve.

View full abstract

In this paper, the stresses of band adhesive butt joints, in which adhesive bonds were used partially at the interfaces, were analyzed using a two-dimensional theory of elasticity in order to determine the usefulness of the joints. In the analysis, similar adherends and adhesive bonds which were set at two or three regions were replaced with finite strips. In numerical computations, the effects of the ratio of Yound's modulus of adherends to that of adhesives, the thickness of the adhesives, the bonding area and position, and the load distribution on the stress distributions at the interface were shown. It was seen that band adhesive joints were effective when the bonding area and positions were varied with external load distributions. Photoelastic experiments and experiments concerning the strains of adherends were performed. The analytical results were fairly consistent with the experimental results.

View full abstract

In the ordinary boundary element method for steady three-dimensional thermoelastic problems, the Galerkin vector is used to express the displacements, and Green's second identity is used to transform volume integrals into area integrals. In this paper the thermoelastic displacement potential is introduced for analysis. The displacements and stresses caused by temperature distribution are given by the derivatives of thermoelastic displacement potential. The method of superposition for solution is used to satisfy the complementary mechanical boundary conditions. The discontinuities of stress on the boundary are investigated by the analytic formulation. In order to investigate the accuracy of this method, steady thermal stress distribution in a cylinder is obtained. As a numerical example, the thermal displacement and thermal stress in a cup-shaped object are obtained.

View full abstract

This paper deals with the steady thermal stresses in a finite composite circular cylinder, in which two finite circular cylinders of different materials are joined together at the ends, under uniform temperature distribution. The thermoelastic problem is formulated in terms of a thermoelastic displacement potential and two harmonic stress functions. Numerical calculations and measurements using resistance strain gages are carried out for the case of the composite cylinder which consists of stainless steel and carbon steel. The experimental results are compared with the theoretical ones.

View full abstract

Fundamental equations based on the linear-thermoviscoelastic theory were established to obtain the temperature and residual stress by cooling in a 3-layer laminated beam. The transient thermal stress and deformation in the laminated beam made of composite materials constructed LSI plastic packages were calculated using these fundamental equations. It was found that the residual stress is strongly related to glass transition temperature and Young's modulus of epoxy resins, and the material design method necessary in order to decrease the residual stress was clarified.

View full abstract

An experimental study on the thermal ratcheting of a cylinder subjected to axially moving temperature distribution was performed in this paper in order to verify both the ratcheting mechanism and the predictive equations for the ratcheting strain which the authors had proposed. Hollow cylinders, which were made of SUS304 steel, were subjected to axisymmetric temperature distribution with the maximum temperature of 650°C. Three kinds of movement of this temperature distribution were considered, that is, stationary cycling without moving, hot-front cycling and cold-front cycling with opposite moving directions. As a result of the experiment, little ratcheting occurred in stationary cycling. On the contrary, remarkable ratcheting occurred in the cases of hot-and cold-front cycling. An expansion of the diameter of the cylinder was observed in the case of cold-front cycling, and a contraction was observed in the case of hot-front cycling. These phenomena coincided well with the prediction, and the validity of the proposed ratcheting mechanism was verified. Conservative prediction of the ratcheting strain by using the predictive equations was possible at the large number of cycles where the work hardening and/or the cyclic hardening was involved in the experiment.

View full abstract

The use of interference-fitted assemblies is very common in engineering practice, for example, in bolted connections and fitting of sleeves into cylindrical shafts. For practical-purposes, it is important to prevent stress concentrations which are generated at contact surfaces of the components assembled with interference fits. In this paper, a method for determining the interferences to uniform stress distribution at the contact boundary of the components is proposed. The finite-element method is used to solve stress and displacements in two component systems in contact with each other in which slip does not occur. The difference between maximum and minimum stresses at the contact boundary is minimized under a constraint of contact pressures, as nodal interferences are taken as design variables. The proposed method is applied to two basic examples of a rectangular plate fitted with a circular plate and a column fitted with a sleeve, to show its effectiveness.

View full abstract

Recently, to clarify the mechanism for material deformation during plastic working and to outline the guidelines for plastic design, it is necessary to develop the methods of stress-strain analyses in the elastoplastic and plastic states. In the previous paper, the authors attempted to find a possible strain freezing method in photoplastic experiments, and the results showed that a very fine linear relationship is maintained between the isochromatic fringe order N and principal strain difference (ε₁-ε₂) over a very wide range under the uniaxial tensile stress state. In this paper, we discuss the type of relationship between the fringe order and principal strain in the biaxial stress state. The discussions on the results obtained are based on the Moire method and photoplastic method in the use of the strip sheet and cross-type specimens of polycarbonate.

View full abstract

The accuracy of a solution obtained from a finite element analysis is deeply dependent on a mesh used in computation. Hence, it is desirable to understand how much error exists in the finite element solutions. The present paper discusses a posteriori error estimation based on finite element solutions. Since a higher order term of one degree higher than the original finite element analysis dominates the error field, an 8-node element is utilized to estimate error in the 4-node finite element model. A large memory or long computing time is not necessary because element by element computation is utilized. After estimating the error in the finite element solutions, we can use the error in two ways. One is that adaptive meshes for the subsequent stress analysis are decided based on the error distribution. The other is that the finite element solutions are improved by adding the estimated errors to the original solutions Some numerical investigations are performed in order to examine the validity of the method.

View full abstract

This paper describes the verification of the accuracy of the residual-stress measurement by the hole drilling method by the use of the indirect fictitious-boundary integral method. As an example, a finite rectangular plate with the initial stress is treated. Also, the residual stress is calculated using the relieved strain during hole drilling, and the accuracy of residual stress measurement is verified by comparing the residual stress, i. e., given initial stress, and the measured (calculated) residual stress. The results are shown for various distances of strain gauges from the hole and for various angles of strain gauges.

View full abstract

For the contact problems in the theory of two-dimemsional elasticity, in which the rigid punch is pressed on the elastic half plane, we usually use the contact condition between the vertical displacement and the punch profile, and ignore the effect of the tangential displacement. However, the surface deformation on the contact area is different from the punch profile because the deformation is given by the vector of both displacements. In the present paper, we analyze the same problem under the more rigorous contact condition in consideration of both displacements, as follows: (v)_y=0+kf'(x)·(u)_y=0=δ₀-kf(x), |x|≤a where (v)_y=0 and (u)_y=0 are the vertical and the tangential displacements on the contact area, respectively, δ₀ the indentation displacement and kf(x) the punch profile. We calculate numerically the two cases of the parabolic and the wedge punches, and discuss the effect of the tangential displacement on the stress state. In the results, we show that all of the contact stresses, the indentation forses and the indentation displacements are smaller than those in the conventional cases.

View full abstract

This paper presents a series of new boundary integral formulations to obtain the shape design sensitivities of two-dimensional potential problems. The so-called direct differentiation method is used in the derivation process. The integral equations for potential and also its potential gradient are both differentiated explicitly with respect to the shape design variable. An integral expression concerning the potential gradient is derived from the weighted residual form in which the gradient of the fundamental solution is used as the weighting function. A computational scheme based on the boundary element method is proposed for the solution of the newly derived set of integral equations for design sensitivities.

View full abstract

We investigate here the load transfer between femur and prosthesis following total hip replacement. First, compression tests using a human cadaveric femur are conducted. Then plaster models for the femur are made from the molds taken from the femur. We believe thet the plaster model femur can allow us to distinguish the minute differences affected from experimental setups. Utilizing the five kinds of cementing materials and the two kinds of prosthesis stems, the effects of the modulus of the cementing materials to the strain distribution on the model femur are examined. Further, a numerical analysis method based on the beam theory is proposed, and the load transfer problems between femur and prosthesis are solved by the Runge-Kutta method. The numerical results obtained are in good agreement with the experimental results, which indicates that the distal load transfer between femur and prosthesis takes place in a manner similar to that between two beams. Both the experimental and analytical results predict that the strain distribution induced on the femur with implanted prosthesis differs from the physiological one considerably; i. e. the strain at the calcar is maximum before implantation whereas the strain at the distal point become dominant after implantation. It may be concluded that the stress shielding can't be avoided as long as the presently available prostheses are utilized.

View full abstract

A particular formulation of the finite-element method relevant to the analysis of the Brinell hardness is established. The Brinell hardness of 0.35%C steel and 70/30 brass is analyzed by the proposed method using the stress-strain curve of the tensile test. The numerical results are in agreement with the experimental results within an error of less than 8.9% for 0.35%C steel and 5.8% for 70/30 brass. The extension of the plastic zone induced by the indentation of the indentor is investigated by the analysis. The residual stresses in the specimens after removal of compressive load are discussed. Further applications of the proposed method are also suggested.

View full abstract

This paper presents a boundary element application to determine the optimal impressed current densities in a cathodic protection system. In this system, enough current must be impressed to lower the potential distribution on the metal surface to the critical values. The potential within the electrolyte is described by the Laplace's equation with nonlinear boundary conditions which are enforced based on experimentally determined electrochemical polarization curves. The optimal impressed current densities are determined in order to minimize the power supply for protection. The solution is obtained by using the conjugate gradient method in which the governing equations and the protecting conditions are taken into account by the penalty function method. The boundary element method is employed to descretize the governing equations. Several numerical examples are presented to demonstrate the practical applicability of the proposed method.

View full abstract

Three-dimensional steady-state deformation in shape rolling with grooved rolls is simulated by the rigid-plastic finite element method using the isoparametric hexahedral elements with eight nodes. To simulate complex steady-state deformation in shape rolling, a method featuring both the steady-state and the nonsteady-state schemes is proposed. For the three-dimensional element mesh, the boundary conditions of metal flow along the curved roll surface and of friction between the roll and the workpiece are dealt with at each node over the interface. Contact of the nodal points with the roll surface and detachment from the surface are judged at each step of calculation. As examples of three-dimensional simulation, rolling processes of bars and beams with grooved rolls are chosen. The deforming shape and the distributions of stress and strain in the workpiece are computed.

View full abstract

An extended model is proposed for lung respiration. This model is made to cope with the gas diffusion by convection in the airway network, gas exchange at the alveolus and gas transport by blood flow, as well as the gas flow phenomenon in the airway network induced by diaphragm motion. The submodel for each phenomenon is combined into a total respiratory system, and it is formulated in the context of discretization based on the adjoint variational principle. The model is examined first in terms of normal respiration and then high-frequency ventilation (HFV) is studied. The effects of HFV are evaluated through several case studies and it is quantitatively shown that the influence of the convective diffusion is dominant in the gas transport in the case of HFV. The proposed model is expected to be used to determine proper parameters of HFV.

View full abstract

Mechanical structures are designed to obtain minimal weight shapes by transforming the contours of the external boundaries and by shaping the vacant spaces of the internal regions. The given initial structural shapes of two-dimensional plates are transformed using the successive approximation in the steepest descent of the sensitivity function for the total strain energy of the structures constrained by the maximum equivalent stresses. The weight of the structural shape is minimized by the modification of the external contours, as shown by shape 4. The elements with very low values of the sensitivity function exist in the central region of shape 4. The structure of shape 4 is minimized by the shaping of the vacant spaces of the internal region, as shown by shape 8, which had about 14% less weight than shape 4.

View full abstract

The state-dependent variable a was introduced to the formula of the rate of heat evolution deduced from the thermal analysis of the molding compounds. The a was determined by the inverse solution of the FEM for temperatures measured. The variation of a is expressed as a function of time after the end of the filling process on the basis of the conversion of resin. The temperature profiles in the mold cavity after the end of filling were calculated using the FEM coupled with the a. Making use of the a in the calculations, the solidification front is estimated by the analysis of non-stationary heat conduction. The relationship between molding temperature and warping of the moldings was evaluated on the basis of the variation mode of a. The fundamental data used to estimate the relation between molding temperature, soldification front and warping of the moldings were obtained.

View full abstract

The software which enables us to classify the mixed mode fracture surfaces without human assistance has been developed based upon computer image processing and pattern recognition techniques, computing co-occurrence matrices of small divided rectangles. Two types of software have been developed; one is the mapping method, and the other the clustering method. Applying this software to the mixed mode fracture surfaces of cyclic stress corrosion cracking of an austenitic stainless steel, SUS 304, the fracture surfaces were quickly and accurately classified into intergranular, brittle transgranular, and ductile transgranular cracking. The merits and the demerits of two methods and how to improve the accuracy of classification are also discussed.

View full abstract