The authors conducted uniaxial and biaxial (Ratio of stress in warp and fill = 1 : 1) stress relaxation and creep testings on a PTFE-coated glass fiber and PVC-coated polyester plain-weave fabric, including iterative stress relaxation testings. The applied stress magnitudes for the testings were 3 to 7 kgf/cm, the durations of a unit stress relaxation were 10, 120 minutes (for the iterative relaxation) and 24 hours, the durations of creep testings were 12 and 24 hours, and the temperature conditions taken only for the testings on uniaxial stress conditions were 20, 40 and 60℃. Strain-, temperature-, iteration times-, stress ratio- and yarn direction-dependency on stress relaxation, and stress-, temperature-, stress ratio- and yarn direction-dependency on creep of the specimens were studied, and the useful knowledge of visco-elastic property was found out. One of the conclusion of the study is that the rate of convergence of relaxation curve highly depends on the number of times of iteration of relaxation without dependency on yarn direction, the duration of unit stress relaxation and stress ratio.

View full abstract

A load map for thermal degradation which is available to predict degradation of elastomeric roofing materials has been developed. The thermal degradation load is given by the formula Σexp (-B/T)- t, where B is a mate rial constant, T is absolute temperature (K) and t is time (hrs). The temperature of the roof membrane, ambient air temperature, solar radiation and wind velocity was measured for one year at Yokohama to make clear the relation between the thermal degradation load and the weather elements. Being based on the meteorological data from 1979 to 1988, the thermal degradation loads at 66 points in Japan were calculated, and a load map was drawned by dividing them into several zones. The field tests of two rubber sheets were carried out at Sapporo, Yokohama and Naha from May 1987 to April 1988 to evaluate the usefulness of the map. The degraded values, elongation rate at break of rubber, were compared with the predicted values. And it was concluded that this method by applying the thermal degradation map was useful for prediction of degradation of materials which were mainly affected by solar irradiation heat.

View full abstract

This paper theoretically evaluates the deformation capacity of R/C columns which fail due to the crush of concrete after flexural yielding. The following findings were drawn from this study. 1) In the process from yielding to failure, a stable limit appears from the point of view of energy absorption of columns. This stable limit corresponds to the limit of the extension of tensile longitudinal reinforcing bars. Before the stable limit, the columns show stable load-deflection curves. After that, the external energy associated with the axial force rapidly increases and the internal energy absorbed by concrete dramatically increases, then the lateral load carrying capacity drastically decreases. 2) The strain of the extreme compressive concrete fiber in the critical section at the stable limit is obtained from Eq. (29) expressed in terms of the stress versus strain relations for concrete and steel, and the axial stress level. Then, the strain in the tensile bars, the curvature, and the depth of the neutral axis are easily obtained. 3) The deformation mechanism of columns beyond flexural yielding can be represented by the truss model proposed in this paper. The tension strain distribution along the height in the hinge region obtained from this truss model was in excellent agreement with the strain distributions measured in the test. 4) The drift angle of columns at the stable limit can be calculated by Eq. (50), using the value of the strain in the tensile bars and the depth of the neutral axis in the critical section at the stable limit.

View full abstract

Creep strain of plain concrete subjected to varying load can be expressed by the sum of three components, i. e., delayed elastic strain, flow creep strain and creep recovery strain. However, the Authors indicated in the previous paper that the effect of creep recovery upon the result of creep stress analysis is so small as negligible in case of prestressed or reinforced concrete member. Therefore, in this study, creep strain analysis was carried out on the member section of prestressed concrete by neglecting the effect of creep recovery. Analytical results showed that axial and rotational creep strains at arbitrary section of member can be expressed as the sum of three components, i. e., strains due to delayed elastic strain, due to flow creep and due to drying shrinkage of concrete, respectively. And further, equivalent creep factors were defined for axial and rotational creep strain of member section. Also, based on these creep strains for member section, average axial creep strain throughout the whole length of member and creep rotating displacements at both ends of member were derived in simple ex- pressions with the average axial and end rotation creep factors.

View full abstract

This paper presents results of an analysis of ultimate behavior of tubular beam-columns based on statistical data for cold-formed circular tubes under combined bending and axial loads. The conclusions drawn are summarized as follows : (1) The hardening exponent included in a stress-strain model for tubular sections after cold-forming was found to be represented by a simple function of yield ratio. Statistics of the yield ratio were determined by using statistical data on compressive yield stresses and ultimate tensile strengths. (2) Mean values and coefficients of variation of the column buckling strengths were obtained by using a semi-probabilistic numerical approach. These values were found to agree well with exact values. (3) A simple theoretical approach based on a proposed stress-strain model was found to be capable of interpreting ultimate strengths of tubular beam-columns in the existing data base. Numerical analysis according to the first order probabilistic approximation was carried out to determine mean values and coefficients of variation of resistances. of tubular beam-columns. (4) On the basis of these numerical results, recommendations were given for the design of tubular beamcolumns.

View full abstract

One dimensional combined geometrically and materially nonlinear finite element method (PERT) has recently been refined to be FERT-P utilizing the incremental perturbation method (Refs. 5, 6). In this study, a general methodology and an associated computer system are developed for the dynamic analysis of nonlinear plane frames. A dynamic version of incremental perturbation method is developed to give a powerful tool for the solution of dynamic equations. It is observed that the proposed method for solving dynamic equations based on the incremental perturbation method has following advantages : (1) This method enables to assign the proper time increment automatically so that the calculation errors in each incremental step would fall within prescribed tolerance limits. (2) The elastic-plastic hysteresis path of each element can exactly be traced by this method, while it is inherently not possible by any other usual method. (3) The most significant part of computer operations is the product of inverse mass matrix and perturbation coefficient vector. It should be noted that the mass matrix is usually regular and constant so that the inverse operation need not be repeated. The above mentioned tool for dynamic analysis is implemented to FERT-P with slight modification to develop an accurate and effective dynamical analysis system FERT-PD. Through earthquake response analyses of portal frames, the computer system FERT-PD is verified against FERT-D which is the dynamic version of the FERT. The numerical integration method used in the FERT-D is the incremental averaged acceleration method. It can be concluded that the FERT-PD represents a reliable, highly accurate and efficient computer system for the dynamic analysis of combined nonlinear framed structures.

View full abstract

The present paper investigates both the elastic and elastic plastic dimple buckling behaviors of a rigidly jointed latticed dome of single layer on a hexagonal plan. The calculation is performed changing several parameters; the number n of grid along line through the apex, the angle θ_0 of the central member at the apex, and the type of boundary conditions. Those buckling loads are expressed in non-dimensionalized form by classical linear buckling load of continuous isotropic spherical shell. And the present study also has revealed that modified Dunkerley's formula is effective to predict the elastic plastic buckling load of the present latticed dome.

View full abstract

The forced vibration tests in horizontal and vertical directions are performed on pile foundation models. The purposes of these tests are to obtain the dynamic characteristics of pile groups foundation and to develop the dynamic analysis method for pile groups foundation. In these tests, single pile model and pile groups model are examined. As for the pile groups model, the effect of grouting between the ground surface and the concrete block bottom is found not to be negligible, and the embedment effect of the block is verified. Then, vibration mode of pile is examined. Comparing the single pile and the group piles, it is recognized that the group effect of piles appear remarkably in the real parts of the impedance functions, whereas in the imaginary parts, these effects can not be recognized. The correlation analyses are performed between test results and analyses, and it is found that the test results before backfilling are well simulated based on the 3-D thin layer method, and that the results after backfilling are proved also well simulated based on side spring of the block with the above mentioned method.

View full abstract

In order to establish the preventive method of big fire in urban area under strong wind , it is necessary to comprehend the general behavior of fire gas flow and fire-spread. This paper considers big fire area as a belt-shaped heat source, purposes to obtain temperature distribution of fire gas flow from this heat source under strong wind. Equations for temperature distributions are derived by determining several prameters appearing in this model, and shown as the function of dimensionless temperature (Φ), dimensionless velocity (Ω), Froude number (Fr) and ratio (x/D) of downwind distance (x) to wide (D) of the heat source. In order to examine the relation between theory and experiment, measurements of temperature and velocity are performed in a wind tunnel which is placed a belt-shaped vessel for burning alcohol, and the validity and limitation of equations are clarified by this wind tunnel test. If the steady state of fire source is assumed, it is possible to apply the proposed equation for estimating temperature distributions of fire gas flow under strong wind.

View full abstract