Journal of Bridge Engineering

Although long-span bridges are usually designed with sufficient structural redundancy, particular concerns arise about the abrupt breakage of bridge cables, which may cause progressive failure such as zipper-like collapse. The time-progressive nonlinear dynamic analysis approach is proposed to investigate the abrupt cable-breakage event of a cable-stayed bridge. Compared with existing studies, the proposed methodology focuses on the simulation of cable loss scenarios in a more realistic manner through incorporating stochastic moving traffic loads, dynamic bridge–vehicle interactions, and associated dynamic initial states of the abrupt cable-breakage event. Several important issues associated with the proposed simulation methodology, such as the finite-element modeling option of cable breakage, different initial states of cable breakage, nonlinearity, and traffic loads, are investigated through a prototype bridge example. Finally, the response envelopes in terms of moments and stresses along the whole bridge are obtained and compared by means of the proposed nonlinear dynamic simulation approach, the static approach, and the pseudodynamic approach with a dynamic amplification factor of 2.0 as recommended by the Post-Tensioning Institute.

The multiscale modeling method is newly introduced to the research on cable anchorage systems for self-anchored suspension bridges with steel box girders to improve the reliability of the conventional modeling methods. Two kinds of boundary conditions (BCs), the hard BCs and flexible BCs, are defined and illustrated. The strategy of the modeling method of the cable anchorage system for multiscale analysis is then introduced. Based on the proposed strategy, a multiscale model of the cable anchorage system is developed for the Taohuayu Bridge, which is the largest self-anchored suspension bridge in the world. For comparison, a scale model test is carried out, and two elaborate finite-element (FE) models, including a traditional scale model and a full-scale model, are established according to the traditional scale modeling method and full-scale modeling method. For further validation, a full-scale whole-bridge elaborate FE model is established, determined as the full-scale G model. In the test, a complex stress distribution of the anchor structure is newly observed. Intensive comparison study of various models demonstrates that the fuzzy region exists in the test model, scale model, and full-scale model, caused by the use of the assumption of hard BCs and the Saint-Venant principle. The fuzzy region cannot be quantitatively determined, leading to unreliable analysis results and a large amount of trial-and-error work. In the multiscale model, the complex BCs of the cable anchorage system are accurately simulated by the flexible BCs, and the influence of the fuzzy region is eliminated. The results of the multiscale model matches well with the full-scale G model. It is concluded that the multiscale modeling method is the most suitable modeling method to simulate the mechanical behaviors of the cable anchorage system of self-anchored bridges with steel box girders, reliably and efficiently.

Five one-third scale segmental bridge columns with plastic hinges incorporating different advanced materials were designed and tested on one of the shake tables at the University of Nevada, Reno. The columns were subjected to the Sylmar Earthquake record with increasing amplitudes until failure. All the models were cantilever with longitudinal steel dowels connecting the base segment to the footing. Unbonded posttensioning was used to connect the segments and to minimize the residual displacements. Energy dissipation took place mostly through the yielding of the longitudinal bars in the base segment. Conventional RC was used in the plastic hinge of a reference column. In one of the models, a built in elastomeric pad integrated with the footing and a concrete segment constituted the plastic hinge. The other two columns incorporated engineered cementitious composite (ECC) and unidirectional carbon fiber reinforced polymer (CFRP) fabrics at the lower two segments. The effectiveness of repair with CFRP wraps was also studied by repairing and retesting the reference column. The test results showed that the proposed models with advanced materials are suitable for accelerated bridge construction in high seismic zones because of their fast construction, high energy dissipation, minimal damage in the plastic hinge zone, and minimal residual displacement.

Collisions between overheight vehicles and bridges are becoming more commonplace. Collision damage, however, is generally far from catastrophic, although sound repair techniques are critical if additional damage (typically related to corrosion) is to be mitigated. Steel posttensioning, fiber-reinforced polymers (FRPs), strand splicing, and other repair techniques have been used throughout the United States to repair damaged prestressed girders. Whereas these techniques individually are quite effective, the combination of repair techniques has far greater impact. Physical and mechanical limitations associated with individual repair techniques clearly led to the adoption of hybrid or mixed repair approaches in which techniques are used in combination, leveraging the strengths of each repair technique and maximizing the degree of damage that may be repaired. This paper addresses the practical case of augmenting an internal strand-splice repair with externally bonded carbon fiber–reinforced polymer (CFRP) materials, thereby overcoming the limitations of each method. This paper describes design considerations for strand-splice and externally bonded CFRP repair techniques both individually and in combination. A prototype example is presented. The hybrid repair approach is shown to maximize the degree of damage that may be repaired.

A new, double-tensioned prestressing steel strands anchorage system (DTPSSAS), characterized by low prestress loss caused by prestressing tendon retraction, is described. When it is used for short prestressing tendons in concrete box-girder webs, this anchorage system significantly reduces prestress loss. To investigate its efficiency in reducing prestress loss, a rectangular thin plate using the DTPSSAS was devised and tested; the results demonstrated that the average rate of instantaneous prestress loss decreased from 32.35% after the first tension to 2.68% after the second tension. Further, long-term observations of prestress loss with the DTPSSAS were conducted in the laboratory and in a field test. Based on the field test of a continuous concrete box-girder bridge, the average loss rate of instantaneous prestress loss to the control tensile stress of two strands in the web is 7.61%, and the average rate of prestress loss at 150 days is 10.53%. Concurrently, the time-dependent prestress loss was calculated with Models GL2000 and CEB-FIP90 for different concrete shrinkage and creep prediction by the step-by-step calculation method. In comparison with the laboratory test results, the calculated results with the two models were greater than the test values. In addition, the effect of stress redistribution caused by the constraint of reinforcing bars to concrete shrinkage and creep was evaluated. The results revealed that the stress redistribution effect increased the vertical compressive prestress loss of concrete, indicating that the time-dependent compressive prestress loss rate of concrete was greater than the tensile prestress loss rate of prestressed tendons. In this regard, bridge engineers may need to pay attention to the stress redistribution phenomenon in vertical prestressing design.

Steel-fiber-reinforced polymer (FRP) composite bars (SFCBs) have been proposed as a new form of reinforcement for concrete structural elements, such as bridge columns. SFCBs have high initial elastic stiffness provided by the inner steel bars before yielding, positive postyield stiffness owing to the outer FRP after the inner steel bars yield, and superior anticorrosion performance. Furthermore, the postyield stiffness of SFCBs can be fully tailored by changing the steel-to-FRP ratio. Consequently, concrete columns reinforced by SFCBs have exhibited good initial stiffness and stable postyield stiffness experimentally. One potential benefit of the stable and designable postyield stiffness exhibited by SFCB-reinforced columns is to reduce the residual displacement, which is a vital index for evaluating the postearthquake recoverability of bridges. In this paper, the mechanical properties of SFCBs and pushover behavior of concrete columns reinforced by SFCBs are first simulated numerically and validated with experimental results. The influence of FRP types is further evaluated in terms of column deformation capacity. Concrete columns reinforced by steel-basalt FRP composite bars (SBFCBs) demonstrate a better performance-to-cost ratio than that of steel-carbon FRP composite bars (SCFCBs). Nonlinear dynamic analyses of SFCB columns are subsequently conducted under a suite of near-fault ground motions with noticeable acceleration and velocity pulses. The numerical results show that the residual displacement is closely correlated with the peak ground velocity (PGV) and that it decreases with an increase in the postyield stiffness ratio $r_{\mathit{sf}}$, whereas the peak drift of columns stays almost the same. Finally, a design equation for residual displacement is updated with a drift-dependent displacement coefficient.

Modular bridge expansion joints (MBEJ) are large-capacity systems placed between two superstructure segments designed to provide safe joint crossing based on anticipated bridge movements. Locations of discontinuity in bridges are often recognized as weak links and therefore characterizing the behavior of expansion systems installed at deck joints under various excitations is critical to support the forecasting of bridge functionality. This paper presents the development of an analytical model representative of a common modular bridge expansion joint including its critical components, such as friction elements, equidistant devices, support bars, and center beams. The model is then validated through full-scale experimental testing of the joint. The results of this study offer a predictive model for the longitudinal motion of bridge joints excited through anticipated service or extreme events, which can be used to help determine local and global failure within the joint and make inferences as to how a bridge system could be affected. Such models provide a key step toward aiding design efforts, enabling more accurate specification of MBEJs, and supporting functionality-based risk assessment for bridges.

AASHTO defines fracture-critical members (FCMs) as components in tension whose failure is expected to lead to bridge collapse. Accordingly, the bottom flanges of a twin steel box-girder bridge are considered to be fracture-critical members in the positive-bending-moment region. In the event of a fracture propagating through the entire depth of a box girder, the shear studs connecting the fractured girder to the bridge deck play a crucial role in the performance of the bridge. To characterize the response of these connections, a series of laboratory tests were performed to determine the capacity and behavior associated with different stud layouts. Based on the test results, modifications to the current American Concrete Institute (ACI) equations to predict the tensile strength of shear stud connections are proposed.

A strain gauge–based force transducer methodology has been developed to identify magnitudes and locations of loads on noncontinuous, nonhomogeneous, slender beams with welded and bolted joints. Slopes of bending moment curves on two sides of a load are calculated from measured strains on a beam. Four uniaxial strain gauges are mounted to the bottom surface of the beam, with two strain gauges on each side of the load, to form a force transducer. A previously developed calibration method can be used to account for discrepancies between theoretical and actual scaling factors. Four or more force transducers are needed for calibration in this work. The force transducer methodology is experimentally validated on a continuously tapered aluminum beam with a series of welded joints, a half-aluminum and half-steel beam with two different cross sections and a bolted joint, and a full-scale portable Army bridge at the U.S. Army Aberdeen Test Center.

Assessment of the flexural and shear capacity of two V-shaped RC bridge bents has been validated against experimental results from tests on 1:3-scale specimens. A numerical procedure based on iterative analyses taking into account the variation of vertical loads in the bents’ columns provides a satisfactory prediction of the bents’ overall performance. Based on its general agreement with the experimental data, the numerical model was also used to investigate the uneven distribution of axial load and shear force between the two columns of each bent. Three models were compared for the assessment of the shear capacity of the specimens. All the models identified the compression column as the most critical in terms of shear capacity, as confirmed by the experimental results. The University of California at San Diego (UCSD) model, which introduces an additional contribution due to the compression strut along the column length, appears to be less sensitive to axial load variations and to provide a reliable assessment of the shear capacity of the bents.

The lever rule method is often used to determine live load distribution factors (LDFs) in two-girder bridge systems because of the ranges of applicability implicit in the simplified equations of the AASHTO LRFD bridge design specifications. This method is typically more conservative than the simplified equations, which take into account key parameters such as beam spacing, span length, longitudinal beam stiffness, and slab thickness. The lever rule method for yielding LDFs does not take into account these key parameters and only considers simple span distribution; therefore, a certain degree of conservatism is implied. This study demonstrates that the lever rule method in determining LDFs in two-girder bridge systems using precast trapezoidal U-girders is not overly conservative, as comparisons between LDFs from finite-element analysis (FEA) and the lever rule method of the AASHTO LRFD show that the lever rule method produces values for LDFs that are closely reflective of the actual response for shear and to a lesser degree for flexure. The effects of live load on parameters that include flexural response, slenderness, and span length are also considered.

Large-diameter drilled shafts embedded in stiff materials (e.g., clay and weak rock) experience larger lateral resistance compared to the resistance values obtained from available soil-pile models, which usually produce a conservative design. The axial skin friction (i.e., vertical side shear) developed on the side of the large-diameter drilled shafts in stiff clay and weak rock (i.e., cohesive intermediate geomaterials) improves shaft resistance to the lateral loads (i.e., the shaft-head lateral stiffness). This paper presents a model that calculates the vertical skin friction induced by the vertical-displacement component of the shaft deflection and section rotation through the use of the $t\text{-}z$ relationship. The model determines the shaft-deflection resisting moment caused by the axial skin friction on the passive side of the drilled shaft. Up to a 40% increase in the shaft-head lateral stiffness ($K_d$) (i.e., stiffer foundations) could develop as a result of the consideration of the vertical side shear resistance. In addition, the lateral response of the superstructure would be influenced by the variation of $K_d$. This study also shows the degrading effect of vertical side shear on the shaft-head lateral stiffness with the increase in shaft deflection. A number of full-scale drilled-shaft load tests have been used in this comparison, and the results obtained from the presented model highlight the contribution of vertical skin friction caused by the shaft deflection/rotation to the lateral resistance of large-diameter shafts.