Journal of Structural Engineering

The problem of the wind buffeting of bluff bodies, in this instance bridge decks, is reexamined in the present paper from the standpoint of linearized theory. The links of flutter derivatives to aerodynamic indicial functions are first recalled and extended. The possibility that these functions offer of simplifying investigative aerodynamic tests of bridge models is discussed. Finally, the gust response problem is set up for a two‐dimensional model in the time domain, employing aerodynamic indicial functions. The equations are then Fourier transformed to the frequency domain, preserving in the process the correct relations between the frequency‐dependent flutter coefficients and the aerodynamic buffeting force spectra. Methods for estimating maximal excursions of the model are then recalled. The paper offers a theoretical framework for future wind tunnel model investigations under turbulent flow.

The working stress, load factor and autostress methods are applied to the rating of composite and noncomposite steel beam and girder bridge structures. These three methods are developed and used to rate a series of 16 representative bridges with a wide range of span lengths. The rating factors resulting from these analyses are compared to each other and correlated with the span length by means of a linear regression. The results of the regression allow the engineer to predict the increase in the rating factor obtained by utilizing load factor and autostress methods over that given by the universally recognized working stress method.

Methods are examined for the estimation of extreme wind speeds from short records. They involve monthly winds (proposed method) and yearly winds (conventional method) and assume that these wind series are stationary, independent, and extreme type I distributed. The analysis of numerous simulated and actual wind data shows that the proposed method is superior when the available wind record is limited to about 3 yr of observations. The method generally provides reliable estimates for 50‐yr wind speeds. Developments in the paper also include confidence limits on extreme wind speeds. The limits are determined at over forty wind stations in the U.S.

Mean crossings rates are determined for translation series, which can be obtained from correlated Gaussian series by nonlinear transformations. Translation series have a specified marginal distribution and correlation structure. It is shown that mean crossing rates depend weakly on correlation, particularly for relatively large thresholds. This weak dependence shows that approximations can be developed for mean crossing rates if correlation is disregarded. Theoretical developments are applied to determine the reliability of chain systems with links having correlated, non‐Gaussian resistances and to estimate extreme wind speeds.

Reliability analysis of chain‐like series structural systems was performed for the special cases of lognormal stress and both Weibull and lognormal component strength. The goal was to relate component reliability to system reliability. This problem has application to establishment of design requirements, e.g. for fatigue avoidance in tethering components for marine structures or disks for turbine engines. When the events of failure of each member in a series system are assumed to be mutually independent, a simple expression for risk results. But component failure events are not independent, because of the presence of the same stress term in each event. Correction factors to be applied to the simple expression are derived for the stress and strength distributions considered. It is demonstrated, that the conservative assumption of independent component failure events can produce weight penalties of 10% or more. An example is provided which illustrates application of the results to establish a design requirement for a tendon of a tension leg platform.

The response of slab structures after initial failure is investigated in order to determine a means of preventing progressive collapse. Analytical models for predicting the post‐failure response of slabs are presented and the predictions are compared with experimental results. These analytical models along with an experimental investigation enabled the development of simple design and detailing guidelines for bottom slab reinforcement which is capable of hanging the slab from the columns after initial failures due to punching shear and flexure.

In current practice, design response spectra are normalized by the three peak ground motions. Alternative scaling factors are evaluated to reduce the dispersion encountered in normalized spectral ordinates. The scaling factors comprise two major groups, one based on ground motion data, and the other on response quantities. Within the group based on ground motion values are the integrals of the squared acceleration, velocity, and displacement, and the associated root‐square, mean‐square, and root‐mean‐square motions. Included within the group based on response quantities are the spectrum intensity and the mean Fourier amplitude. The scaling parameters are evaluated statistically using response spectra for elastic, elastoplastic, and bilinear hysteretic systems, computed from a set of twelve representative earthquake recordings. The results show that a three parameter system of spectrum intensities, computed within low, medium, and high frequency regions, may provide a better means of scaling earthquake response spectra. Significant reductions in dispersion may be realized if elastic spectra are normalized by the spectrum intensities rather than the peak ground motions. The spectrum intensities also reduce the scatter for normalized inelastic spectra, for low to moderate displacement ductilities.

An approximate method for estimating the drift of multi‐bent structures is presented. Structures that are singly or doubly symmetrical in plan and comprising any combination of shear walls, coupled walls, rigid frames and braced frames, can be considered. Results for structures that are uniform with height compare closely with results from stiffness matrix computer analyses. The method is developed from coupled‐wall deflection theory which is expressed in terms of non‐dimensional structural parameters. The parameters involve three structural properties: the individual bending stiffness of the walls, the overall bending stiffness related to axial deformations of the walls and racking stiffness caused by reverse bending of the beams. Similar properties are calculated for rigid frames, braced frames and shear walls and then combined to determine values of the two parameters for the total structure. These values are then substituted into a generalized equation to obtain the deflection profile. This method accounts for axial deflection of the vertical components and is, therefore, more accurate for very tall structures. The method provides a rapid estimate of the drift in a high‐rise structure as well as allowing an easy means of comparing the suitability of different structural solutions for a tall building. The method also illustrates the fundamental dependence of the behavior of continuous type cantilevers on two characteristic parameters.

Deterioration of bridges in Western Pennsylvania accelerated after the inception of the use of deicing agents two decades ago. These bridges were constructed primarily of reinforced concrete and steel with no provisions in their design for protection from deicing agents. These agents chemically attack unprotected steel and concrete and ultimately cause weakening of the bridge structure. The benefit of the resistance of timber to the effects of deicing agents led to the reconstruction of three highway bridges using modern timber design. Details of two cases of highway bridge reconstruction are presented: (1) A deck replacement over steel stringers; and (2) a beam and deck replacement using existing abutments. The case of replacement of a railroad overpass is also presented. A primary consideration in this project was the minimization of railroad service disruption time. This was accomplished by fabricating timber members on site to accommodate uncertain field conditions. In this case, the use of steel or concrete would have resulted in an unacceptably lengthy service disruption.

A computer approach to the optimal design of structural concrete beams is presented. The approach allows the optimization for a variety of possible merit functions, considers all relevant limit state constraints by any standard code (as well as other desirable engineering and construction requirements) and is valid for reinforced, prestressed and partially prestressed concrete members. The problem formulation and a nonlinear programming technique for its solution are briefly described. Four numerical examples illustrate some of the capabilities of the optimization program: (1) Minimizing the amount of reinforcement for a given concrete section; (2) assessing the effect of steel cost ratios on optimal solutions; (3) optimizing the section shape of partially prestressed concrete beams; and (4) analysis of given designs with regard to all relevant criteria.

There are many bridges in North America that are inadequate for modern traffic loads because of deterioration or insufficient original design strength, or both. Although many of these bridges do not meet the present standards for roadway width and other geometrical considerations, they are serving well, probably because of low traffic volumes on these bridges. Thus, rehabilitation and reinforcement, rather than replacement of these bridges, can be very cost‐effective. This paper describes a reinforcement system suitable for many through truss bridges, particularly those built around the turn of the century. The reinforcement system investigated consists of superimposed arches with additional hangers and additional floor beams. This reinforcement was implemented on a one‐seventh (1:7) scale model of a steel truss bridge. A series of static load tests was conducted on the reinforced model to determine the effectiveness of the reinforcement system. The test results indicate that the reinforcement system is very effective to the extent that the model was still stable even after severing the bottom chords in one panel of the bridge model.

This paper documents the structural design and analysis as it evolved in the planning of the new Four Allen Center Tower in Houston, Texas. The 1,440,000 sq ft $\left(133\text{,}920m^2\right)$ tower consists of 50 levels above grade and two levels below grade. Aeroelastic wind tunnel models predicted that extreme dynamic oscillation of the tower may occur due to the configuration of existing tall buildings surrounding the site. Dynamic wind force resistance for the silver reflective glass structure was, provided by a four‐celled “bundled framed tube” system. The bundled tube frames are comprised of two‐story high “tree column” modules located at 15 ft (4.58 m) on centers around the building perimeter. The interior tubular cell partitions are framed by three “tree beam” diagonal truss frames quadsecting the tower's curvirectangular plan shape. The superstructure gravity and lateral loadings were transferred to the soil substratum through the medium of an 8 ft‐6 in. thick concrete foundation mat.

The problem of evaluating the probability that a structure becomes unsafe under a combination of loads over a given time period is addressed. The loads and load effects are modeled as either pulse (static problem) or intermittent continuous (dynamic problem) processes. The load coincidence method is extended to problems with both nonlinear limit states and dynamic responses. Results are compared with other methods, namely, methods based on upcrossing rate, simpler combination rules and Monte Carlo simulation. It is found that the load coincidence method is a versatile method for load combination of many types. It generally gives conservative results with good accuracy.

Numerical analysis and experimental investigation of the ultimate strength of tapered beam‐columns are performed to obtain information concerning practical design formula. In order to develop the factor of equivalent length, the ultimate strength of tapered columns and beams are analyzed separately, using the transfer matrix method, and they are derived by statistical technique. Using these factors, the familiar empirical interaction formula for prismatic beam‐columns can be extended to tapered beam‐columns, and this formula is compared with the test results of the writers and others. From this, it is found that the recommended formula can be used to estimate the lower bound of test values.

The historical background and the design and construction of the Northampton Street Bridge are documented. This structure is a well preserved example of an unusual steel cantilever bridge of the 1890s. It spans the Delaware River at Easton, Pennsylvania, and was built to replace Timothy Palmer's wooden covered bridge of 1806. Its designer, James Madison Porter 3rd, was a prominent citizen of Easton and a professor of civil engineering at Lafayette College. Keenly aware of the townspeople's desire for a handsome structure, Porter selected a bridge configuration and proportions well suited for the surrounding area. The substructure and superstructure and the construction sequence are described, on the basis of information given in the contract documents and in the design and detail drawings, and in construction photographs. Traffic and maintenance over the years, and damage caused by Hurricane Diane in 1955 and subsequent repairs, are reviewed. The bridge remains in service and is readily accessible for informal inspection. The most important documentation records are available in the nearby Lafayette College library and in the Easton Public Library.

Reinforced concrete columns with annular section, called R.C. pipe columns, have frequently been used in building construction. Because of the centrifugal process of making concrete pipes and the difference in arranging the reinforcements along the circular cross section, the behavior of a pipe column may be different from that of conventional R.C. rectangular columns. The aim of this paper is therefore: (1) To develop a computer model for the analysis of R.C. pipe columns using the Chebyshev numerical integration procedure; and (2) to determine a proper stress‐strain relationship for centrifugal concrete. Comparisons of the present computer solutions using the proposed stress‐strain relationship of centrifugal concrete with test results on 39 full scale pipe columns are also presented.

In crane design an accidental action that should be considered is the failure of one of the hanging mechanisms during transport operations of large masses. In this paper, the case of a crane transporting a large mass suspended by two sets of cables is considered. The dynamic behavior and the forces induced in the crane, due to the failure of one of the hanging mechanisms, are studied using two 3 degrees of freedom mathematical models. The highly nonlinear differential equations of motion are derived and integrated by Runge‐Kutta algorithm. The influence of the stiffness of the cables, the initial configuration of the system, the value of the hanging force corresponding to the cable release, and the stiffness of the braking system of the trolley are analyzed.

A graphical approach is presented for the design of timber columns subjected to combined bending and tension. The proposed set of charts are used to determine the allowable bending stress and stresses in combined bending and axial tension. The method and the equations are based on the National Design Specification for Wood Construction, and Timber Construction Manual. An illustrative example is presented to show the step-by-step design approach.

A graphical approach is presented for the design of timber columns (struts) subjected to combined bending and axial compression. A step-by-step procedure is given for the construction and use of design charts. An illustrative example is provided.