James A. Swanson

Advanced finite element modeling of bolted T-stub connection components

Abstract The results of a finite element investigation of the behavior of T-stub flanges are presented and compared to experimental data. Two types of models were employed; a three dimensional T-stub model consisting of brick and wedge elements and several two dimensional T-stub flange models consisting of rectangular and triangular elements. All models incorporated non-linear material characteristics, non-linear geometric behavior, and several contact interactions. The two-dimensional models were used to conduct a parametric study that augmented the experimental program and led to simpler behavior models.

Cable-Stayed Bridges: Case Study for Ambient Vibration-Based Cable Tension Estimation

A cable-stayed bridge recently constructed by the Ohio Department of Transportation incorporates measures put forth by a Federal Highway Administration study to mitigate stay motion. In following recent trends, the stays at this bridge are built without the use of grout for the purposes of inspection and, if necessary, replacement. Several experiments were performed to determine the viability of using traditional vibration techniques, which assume an integral sheath, to estimate cable tension with this new configuration.

Prying Models for Strength in Thick-Flange Built-Up T-Stubs with Complete Joint Penetration and Fillet Welds

AbstractThe results of a series of finite-element (FE) simulations and experimental studies are used to develop two prying models that predict the failure strength of thick-flange built-up T-stub connections with complete joint penetration (CJP) and fillet welds. A parametric study based on identifying the major geometric and force-related parameters that vary within current acceptable steel fabrication and design practices is used to select the cases for analysis. The strength prying models predict the capacity of thick-flange built-up T-stub connections with CJP and fillet welds for the failure limit state of the partial yielding tension flange followed by bolt fracture. The accuracy of the developed strength models are verified with FE results of built-up T-stubs designed for full strength compared with existing models reported in the literature. Based on the results of this study, a refined design procedure that takes into account the partial yielding in thick-flange T-stubs is proposed.

PREDICTING DEFLECTIONS IN CONCRETE-FILLED GRID DECK PANELS

Laboratory test results from six different concrete-filled steel-grid bridge deck panels of one and two spans with approximate midspan loadings were compared with finite difference solutions of a governing differential orthotropic plate equation. The orthotropic plate equation was based on assumptions related to fully composite behavior. The effects of changes in properties due to increasing loads were examined. A means of predicting deflections in other laboratory panels was developed, and the results were extended to permit the estimating of deflections in actual bridge decks.

A Finite Element Study of Non-Orthogonal Bolted Flange Plate Connections for Seismic Applications

The current design approach for moment connections in many seismic specifications worldwide focuses on providing energy dissipation to achieve a collapse prevention objective. Most specifications only provide guidance for the design of beam-column connections in orthogonal configurations. However, it is not uncommon that for architectural reasons beams be required to frame into columns at an angle, either with respect to the horizontal (sloped connections) or with respect to a vertical plane (skewed connections), for the design of which little guidance is available. This paper presents a numerical study focusing on Bolted Flange Plate connections in non-orthogonal configurations. The response of these connections was compared to extant experimental results, using orthogonal configurations as a baseline. Investigating the connections response in terms of moment-rotation capacity, stress distributions, plastic strain demands, and plastic hinge locations, some recommendations for the design of non-orthogonal Bolted Flange Plate connections are presented. Secondarily, a study of the influence of specific details on the response of the connections was performed, resulting in practical detailing recommendations.

FINITE ELEMENT ANALYSES ON SEISMIC RESPONSE OF PARTIAL STRENGTH EXTENDED STIFFENED JOINTS

Extended stiffened end-plate bolted joints are widely used in seismic resistant steel frames. In the United States (US) this type of joint is seismically pre-qualified according to AISC 358. In Europe within the framework of the ongoing EQUALJOINT research project, prequalification criteria for different types of bolted joints are under development. Differently from the US approach, in the EQUALJOINT procedure both full and partial strength joints are seismically qualified. The experimental tests carried out within the EQUALJOINT project confirmed the effectiveness of these intermediate strength levels. Therefore, the aim of this work is investigate the possibility to extend this design approach to US joints. The results of a comprehensive parametric finite element investigation are described and discussed, showing the effectiveness of the proposed design performance criteria. 4952 Available online at www.eccomasproceedia.org Eccomas Proceedia COMPDYN (2017) 4952-4964

High Resolution Modeling and Modeling of Connections in Pony-Truss Bridges

Pony-truss bridges are usually modeled using simplifying assumptions, such as pinned connections and analyzing the trusses separately from the rest of the structure. Pony-truss bridges are also generally designated as possessing many Fracture Critical Members (FCM), a characteristic that requires more rigorous inspections. Eliminating or reducing the number of FCM is desirable as it reduces inspection costs. A high-resolution model of a pony-truss bridge was assembled in ABAQUS CAE that explicitly models the behavior of the gusset plates in the connections and the angles in the framed stringer connections. This model indicated alternative load paths that are potential sources of redundancy useful for retrofit or reclassification of bridges with FCM designation. Higher-resolution models of framed stringer connections were also made in ABAQUS CAE using solid elements explicitly modeling the bolted connections. The connections were loaded in tension; what the stringers must hold if they supply an alternative load path for the truss in the event of fracture in the bottom chord. The primary objective of the work described is to provide high-resolution finite element analysis results that can be used to create general behavior for lower-resolution finite element models used in fracture critical analyses to reclassify FCM as non-fracture critical. The results of high-resolution modeling reveal the behavior that is necessary to include in these lower-resolution models, as well as potential alternative load paths. Lower-resolution models are assembled in SAP2000 and matched to the higher-resolution models.

Influence of Composite Slab on the Nonlinear Response of Extended End-Plate Beam-to-Column Joints

Extended stiffened end-plate connections are widely used in seismic area due to their good performance in terms of both resistance and ductility. The most of existing studies focused on the all-steel behavior of these joints, disregarding the composite action of the concrete slab that is generally disconnected. However, the presence of the concrete slab can have beneficial effects on the structural stiffness for both gravity and lateral loads. Hence, most of the building frames are usually designed considering steel-concrete composite solution. However, the slab can strongly influence the hierarchy between beam and column and the ductility of the joint. In this paper the influence of composite deck on the response of extended stiffened end-plate joins has been investigated by means of finite element analyses (FEAs). In particular, the following details have been investigated: (i) all steel joints without slab; (ii) steel joint with disconnected slab; (iii) composite joint.

Fatigue Behavior of Prestressed Tubular Bridge Deck of Fiber-Reinforced Polymer

An experimental study was undertaken to assess the flexural fatigue performance of a new deck system of fiber-reinforced polymer (FRP) that has been installed on a bridge in Delaware County, Ohio. The deck consists of a series of 4-in. (102-mm) square pultruded FRP tubes with a thickness of 1/4 in. (6.4 mm), laid side by side on existing stringers, perpendicular to the direction of traffic. The tubes are posttensioned up to 20 kips (90 kN) with 0.6-in. (15-mm) diameter seven-wire strands at midpoints between the stringers in the direction of traffic. Four double-span FRP deck specimens with two different tube sizes and three different span lengths were tested in flexural fatigue under AASHTO-prescribed footprint of wheel loads for an HS20 truck. Panel action in the deck system was inadequate for the most part, as only the tubes that were directly under the load carried the majority of the load. Subsequent slippage between these tubes and their adjacent tubes can cause cracking in the asphalt overlay. However, no such crack has been observed on the actual bridge after 1 year in service. Panel action is generally improved at higher prestress levels. Prestressing also offers additional redundancy and reserve strength for the system. While longer span decks fail in bending, shorter span decks generally suffer from local shear failure as a result of stress concentrations at the corner of the tubes under the applied load or at the support. The fatigue problem is less critical for longer span decks and smaller tube sizes.