Journal of Structural Engineering

Buildings that are located near transportation corridors often experience floor vibrations induced by passing trains or traffic, which causes building owners some concern. In this paper, a mathematical, impedance-based (wave propagation) model is presented for predicting train-induced floor vibrations in buildings. The model analytically predicts velocities, velocity ratios, and impedances. The analytical predictions of the model were compared and validated with the measured floor vibrations in a 4-story scale model building constructed by the writers. These predictions closely mimicked the measured responses. Using the results from the method presented indicate that the vibrations on the upper floors can be mitigated by increasing the thickness of a floor at a lower level in the building. This lower-level floor with the increased thickness is called a blocking floor. The scale model building was tested with and without a blocking floor. The predicted and measured responses of the scale model building using floor slabs with various thicknesses on the first floor are compared. It is concluded that the use of a blocking floor can mitigate the transmission of structure-borne vibration to the upper floors.

Vibration serviceability criteria are governing the design and determining the cost of modern, slender footbridges. Efficient and reliable evaluation of dynamic performance of these structures usually requires a detailed insight into the structural behavior under human-induced dynamic loading. Design procedures are becoming ever more sophisticated and versatile, and for their successful use, a thorough verification on a range of structures is required. The verification is currently hampered by a lack of experimental data that are presented in the form directly usable in the verification process. This study presents a comprehensive experimental data set acquired on a box-girder footbridge that is lively in the vertical direction. The data are acquired under normal operating conditions and are presented using a range of descriptors suitable for easy extraction of desired information. This will allow researchers and designers to use this bridge as a benchmark structure for vibration serviceability checks under the vertical component of the pedestrian loading. In addition, capabilities of a sophisticated force model (developed for walking over rigid surfaces) to predict vibrations on this lively bridge are investigated. It was found that there are discrepancies between computed and measured responses. These differences most likely are a consequence of the pedestrian-structure interaction on this lively bridge. The interaction was then quantified in the form of pedestrian contribution to the overall damping of the human-structure system.

Fatigue performance of stiffened pole-to-base plate socket connection in high-mast lighting structures was evaluated by experimental and analytical studies. In absence of sufficient experimental data, constant amplitude fatigue limit (CAFL) of longitudinal stiffener terminations on pole walls is classified as Category E′ in the existing American Association of State Highway and Transpiration Officials (AASHTO) specification. Fatigue performance of this connection in the thin-walled tubular structures is controlled by localized secondary bending stress, the magnitude of which depends on the relative stiffness of the connecting elements. Ten full-size high-mast specimens having stiffened socket connection were tested under constant amplitude reversal loading to determine the fatigue resistance of this connection in finite and infinite life regimes. The specimen design was optimized by parametric finite-element analysis (FEA) and local stress approaches. FEA results were verified by conducting static tests on an elaborately instrumented specimen. The test results demonstrated that an optimized stiffened connection can provide a CAFL of AASHTO fatigue Category D in commonly used high-mast structures.

The Vierendeel mechanism is always critical in perforated steel beams with single large web openings, where global shear forces and Vierendeel moments coexist. Thus far, the main parameters that affect the structural behavior of such beams are the depth of the web opening, the critical opening length of the top T-section, and the web opening area. A comprehensive finite-element (FE) study of four sizes of perforated steel sections with three different sizes of 11 standard and novel nonstandard web opening shapes was undertaken, and their primary structural characteristics are presented in detail to provide a simple design method for general practice. The different geometric parameters were isolated and studied to understand the significance of their effects and in turn advance the knowledge on the performance of perforated steel beams. An elaborate FE model was established, with both material and geometrical nonlinearity, allowing load redistribution across the web openings and formation of the Vierendeel mechanism. The reduction of the global shear capacities, because of incorporation of the local Vierendeel moments acting on the top and bottom T-sections, was obtained directly from the FE analysis. Following that, a comparison of the global shear-moment ($(V/M)$) interaction curves of the steel sections with various web opening shapes and sizes was established, and empirical generalized $V/M$ interaction curves were developed. Moreover, the accurate position of the plastic hinges was determined together with the critical opening length and the Vierendeel parameter. This work has shown that the shape of the web opening can also significantly affect the structural behavior of perforated beams, as opposed to the equivalent rectangular shape predominately used so far. In addition, the effect of the position of the web opening along the length of the perforated beam was revealed. The importance of the parameters that affects the structural performance of such beams is illustrated. The thorough examination of the computational results has led to useful conclusions and an elliptical form of a web opening is proposed for further study. The outcomes are considered relevant for practical applications.

This work presents an experimental and numerical analysis of the lateral forces exerted by the inner core of a buckling-restrained brace (BRB) when, after buckling in compression, it arrives into contact with the external restraining case, from which it is separated by a small empty gap. Experiments are carried out on reduced-scale specimens to measure the thrust acting on the containment struts. Refined, nonlinear two-dimensional (2D) FEM models are adopted to obtain numerical estimates of the thrust. The results show (1) the capability of 2D, plane stress FEM models to correctly reproduce the cyclic behavior of the studied BRBs, and (2) the difficulty in accurately measuring/computing the thrust associated to a given axial load/displacement.

Part I of this work was concerned with the experimental and numerical analysis of the lateral force, or thrust, exerted by the inner core of a buckling-restrained brace (BRB) when, after buckling in compression, it arrives into contact with the external retaining case. As Part II of that study, this paper considers the case of a steel-bolted BRB, with an empty gap between core and containment struts, to attempt the derivation of engineering analytical expressions for such a thrust. It is found that the relationship between applied axial force and buckled wavelength does not obey Euler’s formula.

Equations for the design of a posttensioned seismic retrofit of unreinforced clay brick masonry walls are discussed, and results from an associated experimental program are reported. A total of eight full-scale multiwythe vintage solid clay brick masonry walls were subjected to uniformly distributed one-directional and reverse cyclic out-of-plane loading, of which two walls were tested as built and six walls were seismically retrofitted using unbonded posttensioning. Test wall configurations and constituent masonry materials were selected to replicate typical characteristics of historic clay brick masonry walls. The test walls were seismically retrofitted by applying varying magnitudes of posttensioning using a single tendon inserted into a cavity located at the center of each test wall. Several aspects pertaining to the seismic behavior of posttensioned masonry walls were investigated, including damage patterns, force-displacement behavior, tendon stress variation, wall secant stiffness, hysteretic energy dissipation, toughness modulus, and damping ratio. Finally, measured performance parameters of the test walls were compared with the corresponding values predicted using the proposed design equations.

A shear failure model for nonductile RC columns that sustain flexural yielding prior to shear failure is proposed. The model, which relates shear failure to column end rotation, was derived by performing a forward stepwise linear regression on a database compiled from column tests. Critical parameters that affect rotation capacity at shear failure were found to be axial load, spacing (rather than amount) of transverse reinforcements, and nominal shear stress. The model is suitable for use in nonlinear frame analyses using either a lumped-plasticity or fiber-section column idealization. The model also can be used to determine rotation limits for performance-based assessment of existing buildings.

Shear lag effect in thin-walled box girders has been studied over several decades. However, the methods adopted in many papers have some deficiencies. In the present work, an improved displacement function for shear lag warping in a box girder with cantilever slabs is established. Based on the concept of generalized force corresponding to the generalized displacement for shear lag and the relevant geometrical properties, an improved finite-segment method is proposed to simplify the shear lag analysis of complex box girders. The homogeneous solution of the governing differential equation for shear lag is adopted as the element displacement function. The formulas of the element stiffness matrix and the equivalent nodal force vector are derived. A general formula expressed in terms of the generalized moment is presented to calculate the stress. A finite-element computer program is developed by using FORTRAN language and is used to analyze a cantilever box girder model and a continuous prestressed concrete box girder. The theoretical results are in good agreement with the test results, validating the proposed method and formulas. This paper also reveals for the first time the characteristics of the generalized moment for shear lag.

This study relates to the topic of anchorage of shear walls. At times, eccentric forces between the sheathing and the anchoring devices may be introduced in the sill plate. In severe cases, such forces may cause the sill plate to split and to fail in a brittle manner. In this study, fracture mechanics are applied to develop a simple closed-form hand-calculation expression for estimation of the ultimate load capacity of the sill plate. Finite-element analyses using both linear elastic fracture mechanics (LEFM) theory and a nonlinear fictitious crack model are also used to predict the ultimate load-bearing capacity of the sill plate. The hand-calculation model is compared with the finite-element models, and good agreement is obtained. The results obtained with the various fracture mechanics models are compared with results available from previously performed experimental tests, and again good agreement is obtained. A general conclusion is that the LEFM theory is an adequate approach for the case studied and that the hand-calculation expression developed could be useful for structural design.