In this study, the influence of moisture content on the critical stress intensity factor KIC was examined experimentally at room temperature. The materials used for the experiment were Japanese cypress, Alaska cedar, agathis, red lauan and tamo. The critical stress intensity factor was maximum between 4% and 7% moisture contents. And avobe 7% moisture content, the critical stress intensity factor decreased with increasing moisture content.
To investigate the accuracy of present design formulas for mechanical joints of glued laminated timbers (GLTs), dowel-bearing tests with a drift-pin were conducted in this study. GLTs with four kinds of moduli of elasticity (5.9, 7.9, 9.8, and 11.8GPa) were prepared for the tests. These GLTs were composed of mechanically graded Japanese cedar and Japanese larch laminae with uniform modulus of elasticity. Four types of specimens were cut from these GLTs; “Parallel type” in which load was applied parallel to the grain, and “Perpendicular type” in which load was applied perpendicular to the grain. Five kinds of drift-pins with the diameter of 4, 8, 12, 16 and 20mm were embedded into GLTs. Bearing stress was calculated by the ratio of a bearing load to the diameter and the length of a drift-pin. We investigated the relation of the diameter of a drift-pin to “five percent offset values” specified in ASTM-D5764, “initial stiffness” calculated by the ratio of a bearing stress to a unit bearing deformation, and “effective elastic foundation depth” (the ratio of modulus of elasticity of wood to the initial stiffness) based on the theory of a beam on elastic foundation. The results obtained were as follows. (1) Five percent offset values were constant regardless of the diameter for the “Parallel type” whereas they showed a declining tendency with increasing of diameters for “Perpendicular type”. (2) Initial stiffness decreased with increasing of diameters for the “Perpendicular type”, on the other hand, this tendency was not clear for the “Parallel type”. (3) Effective elastic foundation depth increased with increasing of diameters for both types. Dr. Hirai's formula commonly used to estimate the relationship between the depth and diameter of fasteners was applied to these results, however the conformability was low. This discrepancy resulted from the fact that the formula was derived from experiments in which fasteners of small diameters were used.
The objective of this study was to determine the effect of polyethylene glycol (PEG) treatment on thefracture of wood under mixed mode loading. Specimens were cut from commercially dried western hemlock, Douglas fir, agath-is, spruce and Japanese cypress lumbers, and some of them were treated by PEG-4000. The experimental results were compared with various fracture criteria hitherto proposed in literatures. The fracture criteria experimentally obtained for solid and PEG treated woods do not follow any of these criteria. The fracture criterion obtained for PEG treated wood is different from one obtained for solid wood. The values of KII at Mode II fracture obtained experimentally for PEG treated wood are smaller than one obtained for solid wood.
Growth stresses occur in the stem of living trees. They frequently cause considerable degradation of timber based on end-splittings and warps of logs and boards after felling or during machining. This paper examined the influence of rate of growth in stem diameter on peripheral longitudinal and internal longitudinal growth strains in 22, 25 and 30 years old plantation grown trees of Cryptomeria japonica. The relationship between peripheral longitudinal growth strain and wood properties was also investigated. Strain due to growth stress relief was measured by using electronic strain gauges glued to the outer xylem and within stem xylem. In addition to, set recovery (that is, dimensional change) due to growth stress relief within logs was measured by using a length comparator after heat treatment. A negative correlation was found between level of growth strain and rate of growth in diameter in silviculturally treated trees. Lower average levels of growth strain occurred in trees with higher growth rate. The internal longitudinal strain gradient, along a radius from pith to outer xylem, was reduced in trees with a faster rate of growth. As such, less splitting and warp during conversion would occur in faster grown trees. Moreover, the fact that peripheral growth strain was lower in trees with higher wood density, longer tracheid length and smaller mean microfibril angle suggested that splitting and warp due to growth stress relief would be decreased by using silvicultural treatments.
To study the mechanism of strength degradation of wood-based panels, outdoor exposure tests and strength tests were conducted. Specimens used were commercial oriented strand boards (OSB) and particle boards (PB) which had been exposed in outdoor condition for 1, 3, and 5 years at a field testing site in Forestry and Forest Products Research Institute. To investigate probabilistic properties of modulus of elasticity (MOE), modulus of rupture (MOR), and internal bond strength (IB) of the boards, sixty or seventy-five specimens in one testing condition were tested. Transitions of the distributions for these strength properties were investigated with considering the degradation of adhesion and spring back of elements. The mechanism of strength degradation of the boards is explained with a spot adhesion degradation model. In the case of PB, the degradation of spot adhesion occurs mainly in surface layer after one-year exposure, and decreases the MOEs and MORs of the board. But spring back does not take place, because the amount of degradation is not enough to cause the spring back. After 2 or 3 years, the thickness of the board increases because a number of effective spots which are enough to restrict the spring back, do not remain at the surface layer. After 5 years, the effective spots at the surface layer decrease to tear off a part of particles. Though the rate of the degradation of OSB is lower than that of PB in general, spring back takes place even after one-year exposure because the surface is so rough that some water infiltrates into the board more easily. As the degradation occurs uniformly in the board, relationships between MOEs and MORs do not change. After 5 years, the thickness of the board increases because effective spots enough to restrict the spring back do not remain.
The longitudinal free vibration method is one of the practical methods of evaluating dynamic Young's modulus of commercial lumber. Young's modulus measured by this method, however, is the apparent value throughout the full length of the lumber regardless of whether it is clear or it contains knots or other defects. This fact is important in improving the reliability of this method as a nondestructive estimation of the strength of lumber. Because the strength of lumber with knots is mostly weaker than the strength of clear lumber even though both of them have the same Young's modulus. It is well known that damping characteristic of lumber is affected by knots. Then, if we can introduce some information of the damping characteristic of lumber with knots in addition to the dynamic Young's modulus, the estimation of the strength of lumber by the longitudinal free vibration method may be improved to some degree. In this study, dynamic Young's moduli and damping constants were calculated from the wave data of Douglas-fir specimens with or without knots obtained by the longitudinal free vibration tests, and the bending strengths of them were determined by static bending tests. Then an empirical equation for the bending strengths was derived using the dynamic Young's moduli and the damping constants in the following way. 1) An exponential regression curve was fitted to the relationship between the dynamic Young's moduli and the bending strengths. 2) A regression line was fitted to the relationship between the damping constants and the ratios of the actual bending strengths to the strengths estimated by the regression curve above. 3) The exponential regression curve was modified by introducing the damping constants. This modification gave slightly better estimation of the bending strengths by the longitudinal free vibration method.
Importance of the shear properties of wood-based panels has been increasing due to trends toward building code performance standardization of the building code in Japan. Shear-through-thickness properties of the panel materials, which are not defined in either the JIS and the JAS standards, were evaluated by the two-rail shear (TRS) and the edgewise shear (EWS) methods. Shear strength (τ) and modulus of rigidity (G) were determined for three types of panels used as structural materials in housing; particleboard (PB), softwood plywood (PW), and oriented strand board (OSB). Comparisons were made between the panels and between the two test methods. The results obtained were summarized as follows: From the samples tested, OSB had the highest shear properties (τ and G) and PW had the lowest, while PW had the highest bending properties. High shear properties in OSB could be due to its degree of strand alignment and the layer structure. Wire-strain gauge was used for measuring strain in the both methods; eight gauges were used for one test piece of the TRS and six for the EWS test. It was shown that G could be obtained in the EWS test. Although τs obtained from the TRS method were larger than those from EWS method for all panel types, Gs from the TRS method were almost comparable to those from the EWS. Linear relationships were found between shear properties from the TRS method and those from the EWS method. These results suggested the possibility of employing the EWS test as an alternative method for TRS.
The termite bites and chews the wood and this causes microfractures in the wood. The authors have detected the acoustic emission (AE) generated by the microfractures and confirmed the feasibility of AE monitoring as a non-destructive inspection method to detect termite attack in wooden houses. In this study, the feasibility of piezoelectric polymer, PVDF (polyvinylidene fluoride) as an AE sensor was investigated. Flexible PVDF film of 40μm inserted between the jointing surfaces of wooden posts of 105mm by 45mm in square could detect the artificial AEs generated by the breaking of pencil lead and the AEs generated by termite attack. AE amplitude increased in accordance with the pressure applied to posts up to about 1MPa. Due to the anisotropy of the AE attenuation in the wood, the monitoring area of a PVDF film inserted between the end-end surfaces of the posts was larger the film between the end-side surfaces. The higher AE amplitude was obtained by sandwiching the PVDF film by a pair of elastic Teflon sheet of 0.5mm thick to get an intimate contact among the film and the posts. The sensitivity of the one of the three-layered PVDF film about two times larger than the PVDF film of 40μm.
To obtain the new idea on the improved optimum composite material structures, the fore-wing structures of Allomyrina dichotoma and Prosopocoilus inclinatus have been investigated using scanning electron microscopy. As a result, the structural characteristics of the fore-wing of beetles are made clear and some new knowledge on the optimum light weight composite material structures is obtained. Particularly, the frame structure of the central void is observed in the fore-wing of the beetle, and it consists of the upper lamination, the central void and the lower lamination. The upper lamination and the lower lamination are joined through trabeculae between both of them. The cross section at the edges of the fore-wing is the rectangle and/or the triangle of the central void except the apex. The structure of the fore-wing of beetles possesses both high flexural rigidity and stable secondary moment of inertia and the frame structure of the central void is adopted ingeniously in it. Furthermore, a joint structure between the left and right fore-wings is found.
In this paper, the cause of corrosion of steel bars in concrete, particularly carbonation of concrete and the effect of repair work are discussed acording to comprehensive investigation of existing reinforced concrete structures. A coefficient of carbonation rate (=α) which is calculated from the examination exceeds the usual value. Deviations of the coefficient of carbonation rate (=α) have a tendency to be smaller in the past few years. A depassivated critical value of the difference between cover and carbonation depth is nearly 10mm or less. When the difference between those values is 0mm, the steel bars are perfectly depassivated. When the half cell potential is nearly -120mV vs Ag/AgCl, the depassivation begins and if the potentials is -230mV vs Ag/AgCl, the steel bar is also perfectly depassivated. There is little influences of chloride on the corrosion of steel bars in existing carbonated concrete. The difference between cover and carbonation depth has the most important role in the corrosion of steel bars.
Epoxy resins used for encapsulation of integrated circuit (IC) devices are filled with silica particles to reduce the thermal expansion coefficient and to improve the thermal conductivity. We investigated a method for estimating lifetime of cracking of epoxy resins filled with silica particles near the stress singular point. The flexural strength of the resins was measured by using the 4-point bending test on the notched specimens, and the stress distribution near the notch edges was analyzed by using a finite element method. The stress distribution curves obtained from both the particulate-filled epoxy resins and the unfilled epoxy resins with various stress singularity factor (λ) intersected one after another at a specific distance from the stress singular point. The specific distance of particulate-filled epoxy resins varied depending on the volume fraction of the added particle, the diameter of the particle and the adhesion strength at the particle-matrix interface. This specific distance was related to the plastic zone size of the matrix resin.
Load-controlled fatigue tests were carried out on PAN-based monofilament carbon fibers under cyclic tensile load conditions at a frequency of 10Hz. Fatigue strength data are obtained under pre-determined maximum load Pmax with three different stress ratios of R(=Pmix/Pmax)=0.1, 0.5 and 0.7. The maximum load Pmax is not an appropriate parameter to evaluate the fatigue strength behavior because the cross-sectional area of monofilament carbon fiber is not constant and varies along the longitudinal direction. This fact results in a large scatter of fatigue lives when the data are plotted on Pmax-Nf diagram. In order to evaluate the fatigue strength behavior more precisely, the fatigue strength data must be plotted on S-N diagram by using the maximum tensile stress σmax determined from the cross-sectional area of the fracture surface. It is found that S-N properties of monofilament carbon fibers clearly show the fatigue behavior, depending on the stress ratio R. It is also found that the fatigue strength of monofilament carbon fiber is governed by two parameters such as the maximum stress σmax and the stress amplitude σa, and that the combined stress parameter σmax(1-α)×σaα is useful to describe the fatigue strength behaviors of the different stress ratios.
In this study, Al2O3/Cu functionally graded materials were tried to fabricate based on the results obtained by numerical material design. Obtained results are summarized as follows. (1) Residual stress fields in the functionally graded materials were analyzed by finite element analysis, based on actual fabricating process. From some analysis results, it was found that residual stress is reduced by fabricating the fuctionally graded material with thinner ceramic layer than metal layer. (2) FGM fracture map, which includes crack initiation criterion obtained from bending strength of Al2O3/Cu composite materials, was proposed. It was shown that optimum functionally compositional profile can be found easily by using this map. (3) It was shown that good functionally graded materials can be fabricated by the FGM fracture map.
Fatigue crack propagation tests in vulcanized natural rubber sheets were conducted by using narrow, wide plate specimens under the constant displacement condition. The J integral was kept constant during fatigue crack propagation. The fatigue crack propagation rate was expressed as a power function of the J-integral range, and the power exponent was 1.81 under the ratio R=0 of the minimum to maximum displacement. When R is greater than 0.23, the crack propagation rate is lower than that predicted by the power function. This reduction may be caused by the fiber structure of molecules formed at the crack tip which is not relaxed even at the minimum load. Microcracks are formed ahead of the tip of a crack propagating at high rates. Fatigue fracture surfaces are rough and are accompanied by secondary cracks. No fatigue striation corresponding the crack propagation rate was observed on the fatigue fracture surface.
The effect of the specimen thickness on the fatigue crack propagation in polycarbonate plates was investigated. Fatigue cracks were propagated under axial load conditions with constant amplitude loading, a single peak overload and two-step loading. The crack propagation rates both on the surface and in the interior of the specimen were measured. The crack closure behavior and the appearance of fracture surface were also examined. The fatigue crack propagation rate, da/dN, under constant amplitude loading increased with increasing specimen thickness at low ΔK levels. The crack closure behavior was not appreciably affected by the specimen thickness. The measurements of shear lip width on the fracture surface indicate that the thickness dependence of da/dN is closely related with the formation of shear lips. The crack extension during a single peak overload was larger in the thick specimen. The fatigue crack retardation after the peak overload was more marked in the thin specimen, which was similar to the case of stress decrease under two-step loading. These results can also be understood in terms of the contribution of shear lips to fatigue crack propagation.
This paper presents the experimental results of the flow property of fresh polymer-modified mortars having cement/sand=1/3 (by mass) with varying water-cement ratio (W/C) and air content at the various polymer-cement ratios (P/C), using carboxylic styrene-acrylic ester co-polymer dispersions. The flow property of the fresh polymer-modified mortar was affected by the P/C, W/C and air content, especially, the P/C had a marked effect on the flow property. The experimental regression equation for calculation of the flow value is derived from P/C, W/C and the air content as the coefficient and constant. The physical meaning of the effect of each item in the equation on the flow property was explained by the author.
A simple analysis method was developed by making well use of finite element method (FEM) program for general purposes. The temperature dependence of mechanical properties for every material consisting of ceramic/metal jointing was considered in this method. Using this method, the residual stresses producing during the jointing process were analyzed for Si3N4/SUS304 joints with the insert layer of Cu made by an active metal brazing (Ti-Ag-Cu) method so as to clarify the residual stress behavior due to the creep of Cu. The equivalent normal stress calculated from the multi-axial stresses was used to evaluate the strength of fracture which resulted from initial flaws existing in the ceramic side. From the analytical results, it is found that the temperature at which the residual stress begins to generate is about 550°C, and it is clarified that the creep of Cu layer during cooling process may relax the residual stress in ceramics side and can not be neglected for the estimation of ceramic fracture due to residual stress.