Journal of Structural Engineering

The inelastic load carrying capacity of cold‐formed steel beams with stiffened compression flanges has been investigated. A failure criterion, based on the ultimate compressive strain, is developed experimentally for beams under uniform moment and moment gradient. An extensive testing program was carried out to study the influence of the flat width‐to‐thickness ratio, the moment gradient, and the support width on the strain capacity of stiffened compression flanges. On the basis of the proposed failure criterion, a semiempirical expression for determining the inelastic rotation component of partial hinges is also developed.

Fatigue tests of 62 specimens determined the effect of 3‐year, alternate, and 8‐year weathering on the number of cycles to failure of a transverse stiffener detail fabricated from ASTM A588 weathering steel. These three weathering conditions reduced the mean losses in fatigue life by 42%, 42% and 54%, respectively, as compared to the life of the nonweathered control specimens. The loss in life came at the expense of the number of cycles needed to initiate fatigue cracks and is attributed to rust pitting. The progressive nature of the loss in life with length of exposure time is similar to that reported in the literature for corrosion penetration and pit depth.

The spacing and width of cracks in a parallel crack system is approximately analyzed using the energy criterion of fracture mechanics as well as the strength criterion. The energy criterion indicates that the crack spacing is a function of the axial strain of the bars and also depends on bar spacing, bar diameter, fracture energy of concrete, and its elastic modulus. Both the energy and strength criteria yield a minimum strain necessary to produce any cracks. The energy criterion and the bond slip conditions further yield a lower bound on possible spacing of continuous cracks. The rules for the formation of shorter, partial length, cracks are also set up. Approximate expressions for the crack width at, and away, from the bars are derived. Numerical comparisons indicate satisfactory agreement with existing test data, and lend theoretical support to one aspect of the empirical Gergely‐Lutz formula obtained by statistical regression analysis of test data. Finally, formation of skew cracks in a biaxially stressed and biaxially reinforced plate is analyzed and a crack spacing formula is derived for one typical case.

Historic highlights and milestones are presented to cover the development and application of the Euler formula for buckling strength of a geometrically perfect column together with modifications for inelastic behavior. The latter (nonlinear) behavior may be due either to inherent material properties, such as for a structural aluminum alloy, or due to the presence of residual stresses induced by cooling of hot‐rolled shapes. The story of the column formula has had continuity over a 239 year period and can well form the starring point toward a better understanding of the column buckling theory and design.

Determination of the feasibility of utilizing prefabricated, curved steel panels in the construction of tall cylindrical shells. A unique method of load testing curved panels was developed which allowed for the inclusion of actual elastic boundary conditions. A theoretical development is presented that includes both the effects of initial imperfections and eccentricity. The specific purpose of the research was to determine the buckling capacity of cold‐rolled, steel, curved panels and to investigate the effect stiffeners had on the buckling capacity. The research involved seventeen full scale tests on curved panels of 16, 14 and 12 gage thicknesses.

The global buckling problem of slender tubulaj columns subject to distributed load is studied. The load consists of the column weight, tension/compression exerted at the top of the column, and internal and external variable static pressure forces due to fluids in gravity field. Nonmovable supports at the upper and lower columns ends define the boundary conditions. The resulting nondimensionalized eigenvalue problem is general, and applicable to columns, tubes, pipelines, marine drilling and production risers, mining risers and legs of tension leg platforms. The eigenvalue problem consists of a linear, fourth order differential equation with variable coefficients and four homogeneous boundary conditions. The exact solution is derived in terms of Airy functions of the first and second kind. A power series solution is presented as well. Both methods are numerically implemented and become unstable only for high values of the loading variables. The combined results of the two methods cover the entire domain of practical interest. The critical curves are plotted in the space of the two loading variables for the first six modes. Their asymptotic behavior and its dependence on the boundary conditions is revealed.

Geometric and physical nonlinearities are considered in the investigation of the post‐critical ultimate load conditions of space trusses. The inelastic behavior of struts is accounted for by tracing the complete stress‐strain relationships in the elastoplastic range. The unloading characteristics of struts in the post‐critical elastic‐perfectly plastic stage are determined and the irreversibility of plastic strains are considered through an incremental procedure. The dynamic relaxation method is used for the solution of the nonlinear algebraic equilibrium equations. The method does not require the evaluation of any stiffness matrix, but only direct use of the relatively simple force‐displacement relations of the individual discrete elements is needed. Numerical examples are presented for two test problems and comparative results are obtained for different stress‐strain laws.

This paper describes the design and testing of holding down bolts of the type used at the base of large industrial chimney stacks. The bolts provide a ductile mechanism whereby large earthquake induced vibrations, displacements and forces can be absorbed and so isolate the rest of the structure from the bulk of the input energy. Results from static tensile tests on 50 mm diameter holding down bolts show that satisfactory behavior can be achieved from bolts made from structural grade mild steel bars with coarse pitch ISO metric screw threads. The 800‐mm long bolts stretched 50 mm‐100 mm before fracture. The performance of holding down bolts is found to be in good agreement with that predicted by conventional bolt theory. Yield and ultimate loads can be determined by multiplying yield or ultimate stresses by the tensile stress area as given in standards or codes.

Historical and economic influences on bridge design, including the effects of the Great Depression and World War II, are reviewed. The conditions under which a bridge is built and will serve are also examined in regard to the esthetics of bridge design. Particular attention is given to arches, retaining walls, pier faces, workmanship and design procedure. The paper focuses on steps taken by bridge engineers to improve the appearance of California's bridges.

Extensive research conducted during the past several decades has led to significant design advancements related to the use of wood as a highway bridge construction material. Factors which have contributed to the renewed use of wood in the field of bridge construction include the development of new wood composite products and manufacturing techniques, improvements in preservative treatment methods for resisting decay, developments of new wood bridge system concepts, advancements in timber bridge design and analysis techniques and the consequences of the need to conserve natural resources. Woods' high strength to weight ratio, ease of fabrication and installation and natural renewability as a resource make it a highly viable material for consideration in bridge construction. This paper summarizes recent timber bridge research activities and traces the evolution of the modern timber bridge to its current form.

Welded T‐joints in steel circular hollow sections (CHS) have been tested under combined axial load and in‐plane moments in order to obtain interaction relationships at failure. In all, 105 tests have been carried out—seventeen with in‐plane moment only, seventeen with axial load only and seventyone tests with combined axial load and in‐plane moment. The effect of geometric properties on the ultimate strength of the joints has been studied by varying joint parameters $d_1/d_0\text{,}$ $d_0/t_0$ and $t_1/t_0\text{,}$ where $d_0$ and $d_1$ and $t_0$ and $t_1\text{,}$ respectively, are the diameter and the thickness of the chord and the branch. Interaction curves have been plotted in terms of non‐dimensional ratios $P/P_u$ and $M/M_u\text{.}$ Research has been carried out at Kingston Polytechnic, in conjuction with the Tubes Division of the British Steel Corporation, who fabricated and supplied the specimens and provided the authors with most valuable help and collaboration.

In a companion paper by Bazant and Oh, a simplified theory for predicting the spacing of cracks in a parallel crack system in reinforced concrete was presented. In this paper a different approximate analysis yields about the same result. For this purpose, a solution based on the formulas for the stress caused by a concentrated force in a halfspace is used.

The effect of web distortion on the lateral buckling of I-beams whose top flanges are unrestrained at seat supports is studied by using a finite element method. The theoretical predictions are shown to agree well with available experimental results. The results of the study are used to evaluate present design code approximations, which are found to be unconservative for short length beams, and over-safe for long length beams. An improved approximation is suggested for beams in uniform bending which buckle in a symmetric mode.

Formulation of stiffness matrix, consistent geometric stiffness matrix and consistent mass matrix for a beam element, including shear effect by coupling transverse and shear angular displacement field is presented.