Journal of Structural Engineering

Although most flexural systems are designed as a series of elements, they generally behave as a system in which loads are transferred to the foundation through a combination of members which resist loads in proportion to their relative stiffness. These structures are said to possess multiple load paths. The redundant members may even be small bracing members not designed to resist primary loads. The desirability of multiple load paths in flexural systems is presented. The AASHTO Highway Bridge Specifications penalize nonredundant steel members. The engineer is assigned the responsibility of determining if the structure is sufficiently redundant. Presently available methods of determining the degree of structural redundancy are reviewed. A survey of bridges known to have suffered some distress is presented. A surprising amount of redundancy was observed in most of these structures. Several areas in this field which need considerably more work are identified.

The side‐sway effective length of a crane jib is investigated. For jibs with uniform cross section, restraints provided by the derrick, hoist cable and tower are also considered. When the jib end is completely fixed at the pivot, an improved approximate analysis for the effective length is presented. For a varying moment of inertia cross section, the stability equation is derived. Based on the computational results, the approximate formula for effective length is developed with sufficient accuracy.

A series of experimental investigations was performed to determine the stress distribution in basic two types of gusset plates, the maximum stress intensity and its location, and to propose a simple and acceptable method for rational gusset plates designs. Experiments were performed on eight specimens of Warren‐ and Pratt‐type truss joints with both spliced and monolithic gusset plates. Based on these test results, design formulas for determining the required gusset plates thickness are given. These design formulas are incorporated in. Tentative Design Guidelines of Gusseted Truss Joints by Honshu‐Shikoku Bridge Authority, which give the required gusset plate thickness from the following viewpoints: (1) Transmitting axial and bending forces from web members to gusset plate; (2) transmitting the shear stress resultant along the critical section; and (3) the maximum equivalent stress in the critical section.

For dynamic analysis of trusses, we usually consider only axial deformations in the members. However, it is also known that inertial and body forces occur along the members in their transverse directions, which cause flexure. Therefore, we have devised a component‐mode analytical model in which a member in bending constitutes a substructure. Treating the member as a simply‐ supported beam, we include a limited number of its vibrational mode shapes as displacement shape functions. For plane and space trusses with only a few members, we find by computer analysis that flexure plays an important role in their dynamic responses. However, in trusses with many members, this influence is not very significant, because the flexural deformations in members are localized.

Results of an experimental and analytical investigation on the strength and deformation of biaxially loaded short and tied columns with L‐shaped cross section are presented. The study explores the behavior of reinforced concrete columns under loads monotonically up to failure. A few tests loaded cyclically are also compared with those loaded monotonically. The strength interaction curves and load contours of L‐shaped columns based on analysis and some test results are shown in this paper to provide advice for design information. Two design examples are given to provide possible design procedures.

This paper describes the performance of buildings and other constructed facilities in the Houston‐Galveston area during Hurricane Alicia on August 18, 1983. Records obtained from 17 anemometer sites in the Houston‐Galveston area provide estimates of the fastest mile speed at 10 m above ground in open rrain. The wind speed estimates at various locations are compared with codes and standards for wind speed design values for constructed facilities. The analysis indicates that wind speed estimates exceeded the ANSI (American National Standards Institute) Standard A58.1982 recommended values in only the Baytown‐Laporte area and, possibly, near the west end of Galveston Island. For performance evaluation purposes, all structures are divided into four categories based on the level of engineering attention given during their design and construction phases: fully engineered, pre‐engineered, marginally engineered, and nonengineered structures. Inasmuch as the wind speeds equaled or slightly exceeded ANSI standard recommendations at only a few locations, little building and structural damage should have occurred during Alicia, yet this was not the case. In summary, most of the damage from Alicia in the Houston‐Galveston area was caused by lack of hurricane‐resistant construction rather than the severity of the storm. Provision of adequate fastenings and anchorage of houses in the Galveston area and control of the availability of windborne debris in the Houston area would have substantially reduced damage caused by Alicia.

The effect of stiffness or strength degradation, or both, under cyclic loading on the seismic response of single‐degree‐of‐freedom inelastic systems with 5% initial damping is investigated. Rules are proposed to construct inelastic design spectra for these systems (typical of reinforced concrete construction) as a function of the desired displacement (or ductility) ratio. These rules complement those used at present based on the response of simple elasto‐plastic systems without any degradation.

Following a brief review of the Discrete Fourier Transform (DFT) method of analyzing the dynamic response of linear structures, the limitations and principal sources of potential inaccuracies of this approach are identified, and an evaluation is made of the nature and magnitudes of the errors that may result. Two versions of a modification are then presented which dramatically improve the efficiency of the DFT procedure, and the relative merits of the two techniques are examined. The concepts involved are developed by reference to single‐degree‐of‐freedom systems, and they are illustrated with the aid of judiciously selected numerical solutions.

Several design criteria for isolated, simply supported welded rectangular box‐column subjected to compression and combined with biaxial bending are examined. The load carrying capacity of the columns is presented in terms of nonlinear interaction equations. In the present development, four major subjects are considered: (1) Uniaxial bending of columns; (2) biaxial strength of short columns; (3) biaxial strength of long columns; and (4) the role of residual stress.

A general analysis of cross sections subjected to axial force and biaxial bending is presented. The method is based on Green's theorem and allows for exact determination of stress resultants and tangent stiffnesses for a given set of deformations when the section boundary is rectilinear. The method may be applied to cross sections of any material provided that the stress‐strain relationships are integrable. The method is extremely rapid because stress integrals need only be evaluated at a small number of points on the section boundary. Unlike existing tangent stiffness methods, discontinuous stress‐strain relationships may be used.

Nonuniform normal longitudinal stress distribution in a trapezoidal box beam with lateral cantilever is discussed. The problem is solved by the variational method assuming as unknown the displacement of the beam axis and three functions which describe the warping of the horizontal flanges; the unknowns are reduced in case of vertical walls. The state of stress of stress in each structural element is analyzed and an example is given to illustrate the results.

Nonlinear finite element analysis (NLFEA) systems are rarely used in practice for the analysis of concrete structures even though several systems have been developed to date. The main reason appears to be a lack of agreement concerning the numerical description of material behavior characteristics. A successful NLFEA system is developed by incorporating realistic descriptions of material characteristics into a basically standard nonlinear finite element package. The system applies to the analysis of concrete structures under shortterm plane stress and axisymmetric loading conditions. Simple, yet realistic, numerical models are presented to describe important features of material behavior such as failure criteria and deformational properties of concrete and steel, fracture processes of concrete and steel‐concrete interaction. Procedures to represent crack propagation in concrete appear to influence significantly the quality of predictions. Three structural configurations of known experimental performance are finally analyzed. The predictions show good agreement with the experiment which is typical of several analyses covering a wide range of concrete structural forms.

A simplified analysis procedure is developed to consider the beneficial effects of foundation‐mat uplift in computing the earthquake response of multistory structures. This analysis procedure is presented for structures attached to a rigid foundation mat which is supported on flexible foundation soil modeled as two spring‐damper elements, Winkler foundation with distributed spring‐damper elements, or a viscoelastic half space. In this analysis procedure, the maximum, earthquake induced forces and deformations for an uplifting structure are computed from the earthquake response spectrum without the need for nonlinear response history analysis. It is demonstrated that the maxr imum response is estimated by the simplified analysis procedure to a useful degree of accuracy for practical structural design.

The effect of frequency content of the ground excitation on the optimum design of a base isolation system and a first‐story damping system is investigated. The ground excitation is modeled as a filtered white noise with Kanai‐Tajimi power spectral density. The stationary response, including peak response, is derived in closed form. Small “ground frequencies” result in considerable differences of the optimum design as compared with the one based on a white noise model of the excitation. Examples show the manner of applying the results to an ordinary two‐story structure. Time‐history analyses of the optimized structures confirm the results of the stochastic analysis.

This paper illustrates the design of a ten‐story, single bay, earthquake‐resistant friction‐braced steel frame using a computer‐aided design system called DELIGHT.STRUCT. Linear and nonlinear time history analyses are built into the design procedure itself rather than serving as a check at the end of the design process. The frame's performance is assessed on the basis of its response to gravity loads alone, gravity loads plus a moderate earthquake, and finally gravity loads combined with a rare severe earthquake ground motion. A preliminary analysis is conducted first to find the most active frame performance constraints. The method of feasible directions is employed to solve the constrained optimization problem. Objective functions include minimum volume, minimum dissipated energy and minimum sum of squared story drifts. A sensitivity analysis of frame response for perturbed ground motion and modeling parameters is included.

A distortional mode of buckling is described for cold‐formed lipped channel columns that have additional flanges attached to the flange stiffening lips. The purpose of the additional flanges, called “rear flanges,” is to permit bolting of braces to the channel section so as to form upright frames of steel storage racks. Theoretical and experimental evidence of this mode of buckling is provided for perforated and unperforated columns. A set of design charts is included to permit calculation of the distortional budding stresses of a range of sizes of practical channel sections with rear flanges.

Functional relationships between cross-sectional properties of standard sections are useful for preliminary design, and for automated optimization. Expressions which relate the area, moment of inertia, and the torsional constant for steel sections are used for saving efforts associated witn numerous searches for section properties in a table or database. They also provide means for deriving recurrence relations for optimizations. This paper presents a modified set of relations for commonly used steel sections. A distinction is made between beam and column sections, resulting in less scatter of discrete points around the correlation functions, as compared with other expressions published so far.

The usual criteria to fix minimum reinforcement in concrete structures require first, that the cross section of reinforcement be dimensioned so that the tensile stresses acting on the ``embedment section'' of concrete, when concentrated on the steel after cracking, do not exceed its elastic limit, and second, the crack width be kept within established limits. But these rules neglect some important aspects as soon as cracking provokes significant redistribution of stresses. The incidence of redundancy is decisive in the presence of imposed deformations. An interesting example is the case of a well in which thermohygrometric effects generates a diffused state of tension. From a finite elements analysis it appears that, in the mid-vertical section, a limited opening of the crack enduces a marked relaxation, the sum of the tensile stresses along the section being much smaller than before crack formation. This result suggests the idea to replace evenly distributed bars with series of well-reinforced chords (or prestressed prismatic elements) spaced so that they may keep within acceptable limits the opening of cracks in the intermediate zones and allowing to exploit the release of stresses to cut down load effects in restraining elements.