This paper is concerned with an analytical formulation and a numerical solution of the elasto/visco-plastic dynamic response of the multi-layered shells of revolution subjected to impulsive loads with application to a cylindrical shell. The equations of motion and the relations between the strains and displacements are derived by extending Sanders' theory for elastic thin shells. As the constitutive relation, Hooke's law is used in the linear elastic range, and the elasto/ visco-plastic equations by Perzyna are employed in the plastic range. As a numerical example, the elasto/visco-plastic equations by Perzyna are employed in the plastic range. As a numerical example, the elasto/visco-plastic dynamic response of a fixed supported two-layered cylindrical shell composed of mild steel and titanium subjected to impulsive load is analyzed. Numerical computations are carried out for three cases of the ratio of the thickness of the titanium layer to the shell thickness. It is found from the computations that stress distributions and deformation vary significantly depending on the thickness ratio.

View full abstract

This paper is concerned with development of an effective computational procedure for noise source identification using boundary element software and one of the algorithms available for the optimum problems. The noise source identification in this study is formulated as an optimum problem to determine an optimal set of parameters which define the shape and the location of the noise source by minimizing the square sum of the residuals between the computed and measured quantities at selected points in the acoustic field. In this study, the acoustic intensity is used as the measured reference data, since the acoustic intensity is a vectoral value which includes information on both the intensity and the direction of sound. Use is made of the usual computer code based on the steepest descent method for the optimum problem under consideration. The acoustic intensity is calculated by the boundary element software previously reported on by the authors. Numerical experiments are carried out for a few sample problems using the computational software developed, and the usefulness of the proposed procedure is thereby demonstrated.

View full abstract

The procedure to introduce J2G (or MG-constitutive equation), which was previously proposed by the author, into a FEM program is presented. Another constitutive equation, J2N, is newly proposed which is a kind of non-normality plasticity theory involving J2F and J2D as its special forms without the difficulty of discontinuous transition from plastic state to elastic state during plastic deformation unlike J2D and other non-normality theories commonly used. A few fundamental large elastic-plastic deformations are analyzed by FEM using J2G, J2G-II (or MG-II), J2N and J2F and the results are discussed. It is found that J2G-II gives results close to those by J2G, and that J2N is less sensitive to strain localization than J2Gs are.

View full abstract

The finite element method (FEM) is a very powerful device for solving many continuum mechanics problems which can be treated neither theoretically nor experimentally. In this paper FEM is developed for the analysis of semimicroscopic effects within inelastically deformed polycrystalline metals. The suggested numerical procedure is based on a model of the thermoactivated motion of dislocations which is responsible for inelastic time-dependent deformation. A comparison between numerical and experimental results for an activated slip system is presented.

View full abstract

The extended viscoplasticity theory is modified without the use of the equilibrium stress function and the modified model is shown to be applicable to multiaxial nonproportional loadings using the updating rule for the viscosity control function. Numerical computations are made for 2 1/4Cr-1Mo steel at elevated temperatures (450-600°C) under typical multiaxial nonproportional loadings such as stress or strain path bending, mechanical ratcheting, combined creep-plasticity interaction, cruciform strain path and circular stress or strain path, some of which are experimentally tested in this study. Computational results represent the inelastic deformation characteristics observed in the experiment fairly well except some cases such as mechanical ratcheting, which requires that the anisotropic hardening effect to be taken into consideration.

View full abstract

The multiaxial form of the extended viscoplasticity model is given under the condition of proportional monotonic loading and is incorporated into a nonlinear finite element code by means of an initial strain technique by replacing the creep strain increment with the inelastic strain increment. The accuracy and stability of the finite element scheme is examined by comparison with numerical analysis. A structural analysis of a thick-walled cylinder of 2 1/4Cr-1 Mo steel at 600°C under internal pressure is carried out as a practical implementation of the three-dimensional viscoplasticity theory, and cpu time and stress-strain redistribution behavior are compared with those of FEM analysis by the superposition model.

View full abstract

Creep buckling analyses under varying loads are performed on a circular cylindrical shell with an initial imperfection subjected to axial compression and on a partial spherical shell under uniform pressure. The finite element method is applied to the creep deformation analysis to obtain the critical time when creep buckling occurs. The results show that a linear cumulative damage rule for creep buckling can be applied to the creep buckling of the circular cylindrical shell but cannot be applied to that of the partial spherical shell.

View full abstract

A series of experiments were performed in order to clarify creep-fatigue crack propagation behavior and also to identify the parameter that characterizes the crack growth rate of 2 1/4 Cr-1 Mo steel at 500°C. The following results were obtained (1) Under cyclic loading with tensile load hold, the creep crack propagation during the tensile load hole was dominant. (2) While the crack growth rate was steady during the tensile load hold, the crack opening displacement increased radically right after the load hold started and afterwards increased steadily to the end of the tensile load hold. The radical increase of the crack opening displacement right after the tensile load hold was found not to be due to the crack growth. (3) The creep-fatigue crack propagation rate was characterized by the creep J-integral range calculated from the steady increase of the crack opening displacement during the tensile load hold, as is shown by Eq. (1) in the text.

View full abstract

Effect of grain and grain boundary strengths on the life of main creep crack initiation in a notched plate specimen was investigated, using an austenitic heat resisting steel. In a range of lower grain strength below about H_V 300, the life of main crack initiation increases with an increase in grain strength, independent of grain boundary strength. In a range above H_V 300, however, it decreases drastically with increasing grain strength when a specimen has a weaker grain boundary strength. On the contrary, if a specimen has a stronger grain boundary strength, it continues to increase with increasing grain strength. Further, a relationship between the main crack initiation life, t_i, and the mean notch opening displacement rate, V^^·_m, that is, t_iV^^·ⁿ_m=C (n and C are constant), was obtained which was analogous to the Monkman-Grant relationship. Finally, the effect of grain and grain boundary strengths on t_i was discussed in terms of the relationship t_iV^^·ⁿ_m=C, considering the effect of both strengths on V^^_m and a critical notch opening displacement at the time of main crack initiation.

View full abstract

The static fatigue strength of both smooth and notched bars of sintered Si₃N₄ was investigated experimentally under four-point bending at 1000°C. The static fatigue strength of the notched bars was greater than that of the smooth bars when they were evaluated for the maximum elastic stress value. It was shown that the static fatigue strength of the notched bars can be estimated from that of the smooth bars by the combined theories of the Weibull distribution of the strength and slow-crack-growth (SCG). But it should be noted that the results obtained by the theory are very sensitive to the material constants. The observations of fracture surfaces revealed that there is a very rough region in the fracture surface, which is thought to be formed by SCG accompanying creep deformation and crack initiations at grain boundaries, and is a characteristic feature in the static fatigue failure at high temperature.

View full abstract

It is generally known that the fatigue phenomenon of non-ferrous metals such as aluminum alloys are somewhat different from those of carbon steels. For example, it has been said that the knee point of S-N curve and non-propagating cracks do not appear in these metals. However, the exact meaning of these phenomenon still seems ambiguous growth behavior at the apparent fatigue limit (defined for N=2×10⁷) and fatigue strength was investigated using aluminum alloy 2017 (aged and annealed). In annealed and aged aluminum alloys, non-propagating cracks can exist after repetition of (25)×10⁷ stress cycles. This suggests the possibility of existence of fatigue limit even in aluminum alloys.

View full abstract

We previously reported that the experimental crack velocity in a center-cracked specimen of unsaturated polyester resin is well expressed by one of the three different velocity equations of a constant stress, a mixed condition and a constant strain due to the central crack length C₀₁ under tension. Here, we show that the three different velocity equations are well fitted to express the velocity of fast fracture in PMMA sheet specimens having different crack lengths. The experimental crack velocity generally decreases as the crack length increases, and its change with the relative crack length C₁/C₀₁. The velocity expressions, combined with the existing relation between crack velocity and a dynamic stress-intensity factor, give the variations of the dynamic stress-intensity factor and a dynamic strain-energy release rate with C₁/C₀₁ for the specimens, both of which are in qualitative agreement with the experimental variations of the number of fracture surface marks and their density in the specimens.

View full abstract

Recently, many studies as to the evaluation of fracture toughness for ceramics have been conducted. But the fracture toughness of ceramics as a brittle material is variable with flaw size. And the value of fracture strength and fracture toughness shows large scatter. This paper discribes the development of a failure assessment diagram for ceramics. The basic approach method was conducted under the two hypothesis (1) Critical COD is a material's constant. (2) Process zone is formed by a numerous microcracks or phase transformation. Developed failure assessment diagram for the various ceramics showed good agreement with the experimental values of fracture toughness and static fatigue limit.

View full abstract

Fracture toughness tests are carried out using three kinds of precracked specimen: (1) a specimen with a fatigue precrack introduced at an elevated temperature in a method proposed by the authors. (2) a specimen with a brittle precrack introduced by the bridge compression (BC) method proposed by Sadahiro et al., and (3) a specimen with a surface precrack introduced by Vickers hardness indentation. Fracture toughness evaluations by the chevron notch method and the indentation fracture method are also carried out. The precracks introduced by the three methods can not be assumed to be ideal cracks under the as-precracked condition. However, if the relevant treatments are made, the precracks are presumed to be ideal cracks, and almost the same fracture toughness values are obtained by using treated precracks. Based on these results, the authors recommend the method of precracking by cyclic fatigue at room temperature at the low ΔK_fmax where a precrack introduced by the BC method is used as the starter for fatigue cracking.

View full abstract

An ultrasonic technique for determining the r-value in cold-rolled steel sheets is developed. The theory is reviewed for the interrelation between the velocity anisotropy of guided modes along the sheets and the texture (or preferred orientation), which is defined by the orientation distribution coefficients. The electromagnetic acoustic transducers, fabricated with permanent magnets and meanderline coils, allow the quick and easy measurement of transit time of the lowest symmetric (S₀) mode. The in-plane average of transit times shows a close correlation with both the r^^- -value and {111} pole intensity obtained through X-ray diffraction experiments.

View full abstract

Tension-torsion biaxial creep-fatigue tests have been conducted with 304 stainless steel at 650°C. An overstress and back stress were analyzed under biaxial stress conditions. Under proportional biaxial loading, peak stress components were well arranged by using Von Mises equivalent stress. However, the overstress and the back stress showed a dominant anisotropy. A long-term stress relaxation behaviour was examined under proportional loading. The result suggests that a time-dependent inelastic deformation process is composed of at least two origins. One is caused from a relaxation of the overstress and the other is caused from that of the back stress. The former is much faster than the latter. The relaxation of the back stress indicates a recovery of the anisotropy.

View full abstract

This paper is concerned with a coupled thermoelastic problem in an infinite elastic solid containing a penny-shaped crack, considering the effect of coupling between the temperature and strain fields. It is assumed that transient thermal stress is set up by the application of negative heat generation to the crack surface. By use of the thermoelastic potential and harmonic functions in strain fields, and the finite difference method for the time variable only, the analytical solution for the spatial variables can be obtained. The numerical results are shown for the effect of coupling between the temperature and strain fields on the stress intensity factor.

View full abstract

Some thermal stress problems are solved for a semi-infinite plate with two circular holes. The axis connecting their centers is parallel to the straight edge. The following four thermal loadings are considered. (1) The two holes are heated by the same prescribed temperatures T₀, and the temperature is zero on the straight edge and at infinity. (2) The two holes are heated by prescribed temperatures T₀ and -T₀ respectively, and the temperature is zero on the straight edge and at infinity. (3) The two holes are insulated and heated by a uniform heat flow vertical to the straight edge which is kept at a constant temperature. (4) The tow holes are insulated and heated by a uniform heat flow parallel to the straight edge which is insulated. Numerical results are obtained for the plate with holes having radius/half pitch or radius/ distance from edge of 0.1-0.95 and are given in the form of curves. The effects of the presence of a hole or a straight edge on the magnitude of thermal stress concentration in a semi-infinite plate with a hole or in an infinite plate with two holes are clarified. The numerical results of the latter two problems are also given newly and show good agreement with the results available.

View full abstract

The magnitude and direction of uniaxial tensile stresses in thin plates were measured using SH₀ waves exited by an electromagnetic transducer with periodic permanent magnet array. Analyzing the measured velocity anisotropy of SH₀ waves, we can separate the difference of principal stresses from the texture-induced anisotropy. In this case, knowledge of the stress-free state is unnecessary. This technique also makes it possible to estimate the principal direction, even when it doesnot coincide with the rolling direction. We tested specimens made of alluminum alloy, stainless steel, and mild steel. We were successful in obtaining stress measurements with sufficient accuracy for alluminum-alloy and stainlesssteel specimens. However, stress measurement was difficult for mild steel in which residual magnetization was present.

View full abstract

Caustic interference fringes formed by the reflecting of light rays from any internal point of the objective plate are investigated to apply Caustic experimental method to problems like three-dimensional stress analysis. Half planes with concentrated or uniformly distributing loads are adopted and Caustic fundamental theory and formulae are reestablished. Experimental specimens made of acrylate resin plate are fabricated by combining various thicknesses of plates using adhesive. Caustic patterns of the bonding surface are observed and the characteristics of the reflecting positions are investigated in detail. The computational theoretical patterns ased on the obtained formulae are also plotted. The comparison of the results of the two methods shows good agreement. Therefore the application of this method to three-dimensional problems have tremendous possibilities.

View full abstract

The present paper describes the first step in building an interactive computer-aided system for preliminary structural design of plane frames. This system is such that if an initial shape which partially satisfies design specifications is inputted by a designer, the system will automatically modify the shape so as to fulfill all specifications, or tell the designer to input an alternative shape with reference to knowledge bases. To make remodeling feasible, the design specifications are resolved into shape design variables, geometric constraints and mechanical constraints. Free design variables are determined by the forward reasoning system to fulfill constraint relations. During remodeling, the system checks inconsistency between design variables with constraint relations and generates alternative plans satisfying the constraints. Some examples of remodeling processes are given.

View full abstract

The frame structure of a motor cycle is required to be as rigid and as light as possible. In this paper, the minimum compliance design of the frame structure of a motor cycle is treated numerically. Torsional rigidity between the head pipe and swing arm pivots, which is very important for the driving stability of a motor cycle, is maximized under constraints of the total volume. Two typical cases of design are implemented. In the first case, cross-sectional areas of the frame members are taken as design variables. Shape optimization of the frame structure subjected to geometrical constraints, which are caused by keeping the driving function and the space of some parts, is carried out in the second case. In the latter case, positions and cross-sectional areas of frame members are varied simultaneously. Rigid and simple frame structures of the motor cycle are obtained for each design problem.

View full abstract

Cavitation erosion generated in a rotating disk device was studied through observations of eroded surface and mass loss measurements using Cu, Al and mild steel. By comparing the number and size of erosion pits occurring at the initial stage with those in a vibratory device, we found that the generating frequency of large collapse pressure that forms a pit was very small in a rotating disk device. This means that erosion proceeds due to surface fatigue in such a way that the surface first deforms and the fractures by the repeated action of the small collapse pressures below the critical pressure which forms impulsively a pit. The variations of the mean depth penetration rate (MDPR) as a function of velocity and cavitation number are shown for Cu. Al and mild steel and their differences in the materials are discussed.

View full abstract