Journal of Bridge Engineering

For integral abutment bridge (IAB) design, there is no consensus on the preferable orientation of the piles supporting the abutment, although variations can affect overall bridge behavior and forces in the piles. This paper details a parametric study of steel girder IABs that used finite-element models to compare the effects of pile orientation on bridges of various lengths and skews. For this study, the bridge lengths investigated were 15.2 m (50 ft), 30.5 m (100 ft), and 45.7 m (150 ft) with skew angles of 0, 15, 30, and 45°. H-pile orientation of webs parallel and perpendicular to the abutment centerline were investigated and compared. The results apply to the thermal response of the models and report abutment and pile displacements, pile weak-axis and strong-axis bending moments, and ratios of maximum bending moments to yield moments. Results show that there is not one optimal pile orientation; rather, optimal orientation depends on other factors. Beyond the effects of length and skew angle, choice of orientation should consider expected temperature range, construction temperature, and backfill conditions.

As important load bearing members of cable-stayed bridges, stay cables may experience corrosion, fatigue, and accidental or intentional actions that may lead to possible breakage failure. The current bridge design guidelines require that a cable-stayed bridge be designed against single-cable breakage. A holistic reliability assessment framework is developed for a long-span cable-stayed bridge and traffic system subjected to breakage of stay cables considering service load conditions from both traffic and wind. In addition to the bridge structural ultimate limit state, a new bridge serviceability limit state is also introduced by focusing on the overall traffic safety performance of all vehicles of the traffic flow on the bridge following cable breakage incidents. Random variables are defined by considering uncertainties related to structural material properties, sectional properties, traffic, wind condition, and cable breakage parameters. The Latin hypercube sampling technique is adopted to sample the random variables and establish the simulation models by considering various uncertainties of parameters. Nonlinear dynamic analysis is conducted in each experiment to simulate the breakage of stay cables, during which various sources of nonlinearities and dynamic coupling effects from traffic and wind are incorporated. Fragility analyses of the bridge subjected to cable breakage events in terms of the ultimate limit state and the serviceability limit state are finally conducted.

This paper presents a practical method for automatically generating optimal strut-and-tie models (STMs) in anchorage zones. In the proposed approach, a new microtruss element pattern is created to build ground structures. In certain conditions, the approach could also be improved by combining the microtruss grids with solid elements. Furthermore, the formula for the cross-sectional areas of bar elements in the microtruss grid is deduced for the purpose of replacing the continuum structures built by solid elements with discrete structures built by truss elements. Then, three examples are illustrated to demonstrate the capability of the proposed optimization procedure to find the optimal STMs of anchorage zones. The existing solutions and numerical experiments indicate that the proposed approach accurately reflects the actual load-transfer mechanism in anchorage zones. The method can be used in practice, especially in the design of some disturbed regions (D-regions) in which optimal STMs cannot be obtained by traditional optimization procedures. It should be noted that at present, the application of the proposed model in this paper is limited to homogeneous material with linear constitutive laws.

Perforated shear connectors (PSCs) in steel–concrete composite joints are usually deeply embedded in concrete, and the perforated plate thickness varies significantly, which is different from those in a steel–concrete composite beam. In this study, two types of push-out tests (24 specimens in total) were conducted to compare the mechanical behavior of PSCs used in steel–concrete composite joints with that in composite beams. The testing variables include the push-out test arrangement, plate thickness, and concrete compressive strength. The experimental results are presented and discussed, focusing on the shear resistance, load-slip behavior, and failure modes. Finally, the experimental results were compared with existing shear resistance equations to evaluate their applicability for PSCs in composite joints.

The numbers and sizes of permitted vehicles that consist of at least four wheel lines have been increasingly crossing on highways and bridges. For safety purposes, it is necessary to understand how those truck loads are distributed to the primary bridge elements. It is widely accepted that the lateral live-load distribution factor (LDF) is affected by the spacing of adjacent wheel lines of truck loads. However, current specifications for the LDF are applicable only for vehicles that have standard axle configurations, whereas oversized vehicles usually have nonstandard wheel-line spacing. Furthermore, Iowa’s 5-ft (1.5-m) spacing requirement, which specifies that dual-lane trucks with interior-pair wheel-line spacing ≥5 ft (1.5 m) are allowed to have an axle weight up to 20 kips/lane (90 KN/lane), should also be evaluated. The objective of this paper was to investigate the impact of the wheel-line spacing of four-wheel dual-lane loads on lateral live-load distribution on slab-on-girder bridges. Twenty prestressed-concrete and 20 steel-girder bridges from the Iowa bridge inventory were sampled randomly. Two-dimensional linear elastic finite-element (FE) models of the selected bridges were established, and 22 types of single-axle four-wheel-line loads with variable adjacent wheel-line spacing were applied. The LDFs for the investigated bridges were determined on the basis of the extracted internal shears and moments of girders from the FE models and were compared with those estimated using current LRFD and load factor design (LFD) equations. For practical needs, axle weight limits for those dual-lane loads were determined on the basis of the calculated LDFs. It was found that both moment and shear LDFs decrease with increases in the outside and inner wheel-line spacing of the dual-lane loads. The current Iowa DOT practice with Iowa’s 5-ft (1.5-m) spacing requirement appears reasonable and adequate. The current LRFD equations significantly underestimated (or overestimated) moment LDFs for the interior (or exterior) girders of the four-girder bridges. Both the LRFD and LFD equations mostly underestimated (or overestimated) shear LDFs for interior (or exterior) girders. For safety concerns under dual-lane loading scenarios, moment LDFs for the interior girders of four-girder bridges and the exterior girders of more-than-four-girder bridges should be determined by averaging the estimations according to the lever rule and LRFD equations, and shear LDFs for the interior girders of bridges should be determined on the basis of the lever rule.

Welding defects and fatigue cracks in full-penetration welded beam–column connections, resulting in partial-penetration welds, are found in existing steel moment-resisting frame bridge piers. The welding defects may have an unfavorable effect on the seismic performance of a large number of existing piers, most of which have been in operation for over 50 years. This study aims to clarify the effects of the weld profile, e.g., weld leg length, penetration depth, and incomplete penetration depth, on seismic performance of the beam–column connections in existing piers with welded box sections. The effect of gusset stiffeners (termed fillets in this paper) at the beam-web-to-column-web joint on the improvement of seismic performance is also studied. In this study, experiments on 10 specimens with different incomplete penetration depths and fillet radii are performed under quasi-static cyclic large displacement loading. Based on the test results, cracks may initiate either at the beam end or at the weld of the beam-flange-to-column-flange joint, and the sum of the external and internal total weld leg lengths is a dominant parameter that affects crack propagation at the joint. In addition, the fillet at the beam-web-to-column-web joint can delay the decrease in the load-carrying capacity.

Steel–concrete composite frame bridges have larger structural stiffness, lower cost, and faster construction speed than traditional bridges because a combination of steel–concrete composite girders, concrete piers, and steel–concrete composite connections is used. In this type of composite bridge, the most important part is the connection joint between the steel–concrete composite girder and the concrete pier, which is formed by a steel plate, core concrete, shear connectors, and rebars. Taking a five-opening (110 m × 5) steel–concrete composite frame bridge with five 40-m-high piers as a prototype structure, three specimens, including two steel jacket–concrete composite connections with different construction details and one traditional rebar connection, were designed to study the load transfer mechanism in different types of composite connections. Results of the reversed cyclic load tests on the three specimens indicated that the steel jacket composite connections have better structural performance than the rebar connection, and bending moments in steel–concrete composite girders can be directly transferred to concrete piers via steel jackets. Furthermore, the interrelationship between the anchorage length of longitudinal rebars in the core concrete and the length of the steel jacket is revealed by strains measured by strain gauges, from which the design method of the steel jacket–concrete composite connections is recommended.

This paper studies distortion-induced fatigue cracks in a 1960s-era welded plate girder bridge that developed cracks in the web-gap region shortly after completion of a comprehensive seismic retrofit. Finite-element analysis (FEA) and field testing were conducted to investigate the cause of cracking and then to recommend appropriate repair measures for the bridge. The field test results validated the high-stress concentrations in the web-gap region of the bridge that led to fatigue cracking after only a limited number of cycles. FEA results indicated that the high-stress concentrations were principally a result of replacing K-type diaphragms with stiffer cross diaphragms as part of the seismic retrofit. Two conventional repair measures in addition to two innovative repair measures were evaluated. The proposed innovative repair measures focused on ease of installation compared with more conventional options and were found to reduce the distortion-induced stresses well below the constant amplitude fatigue threshold (CAFT).

The highly predictable and unusually heavy loading on bridges carrying mine haul trucks should be accompanied by advancement in the current bridge design code provisions. Mine haul trucks are optimally designed to match the capacity of large loading shovels, which provide a set number of consistent volumes of ore at efficient loading frequencies. In this research, reliability-based design procedures were implemented to calibrate live-load factors for bridge systems conveying extremely heavy mining trucks (gross operating weight of approximately 700 t). Field measurements of the gross vehicle weight of trucks were used in the calibration process. Analysis results suggest a live-load factor of 1.33 for trucks moving on short-span composite steel girders or NU precast girder bridge systems. Analysis results also recommend the adjustment of the load combination in the design code to treat the barrier dead load as a separate random variable with a load factor of 1.36. Simplified live-load and barrier-load distribution factors for interior and exterior girders in bridges transporting heavy trucks were developed based on a regression analysis. The developed distribution factors are recommended to be used in alignment with the calibrated live-load and barrier-load factors. Modifications to the live-load provisions in the current design code are recommended based on the conclusions drawn from this study.

One of the major drawbacks with most bridge weigh-in-motion (BWIM) methods that adopt the Moses algorithm is the need for acquisition of vehicle speed and axle spacing by using additional devices, such as free-of-axle detectors. This study presents a novel virtual simply supported beam (VSSB) method, which uses weighing sensors to directly identify the speed and the axle spacing of passing vehicles on bridges, making it very desirable for commercial BWIM systems. Numerical simulations and model tests are carried out to study the effectiveness and accuracy of the proposed method. The results show that the proposed method can successfully identify the vehicle speed and axle spacing with good accuracy. The proposed method also proves reliable under noisy conditions.

This study focuses on modifications in bridge aerodynamic and aeroelastic behaviors due to stationary (nonmoving) vehicles for various arrangements of vehicles on a bridge deck. Dimensionless aerodynamic force and moment coefficients, as well as flutter derivatives, are analyzed experimentally in a wind tunnel in the model of a bridge with a semibluff cross section. A comparison is provided for stationary vehicles placed in leeward, windward, and all traffic lanes, as well as for an empty bridge-deck section. Drag, lift static forces, and moment coefficients are determined for various wind incidence angles. Flutter derivatives are obtained using the free-vibration technique for both one- and two-degrees-of-freedom systems. Stationary vehicles on the bridge deck are generally observed to change aerodynamic coefficients of the bridge. This is particularly exhibited for the pitch moment. Galloping sensitivity of the bridge does not change considerably in the presence of vehicles, as all studied configurations remain stable. When considering one-degree-of-freedom motion of the bridge-deck section, an empty bridge without the vehicles proved to be more prone to torsional flutter than the bridge with various arrangements of the vehicles. For the two-degrees-of-freedom system, nonmoving vehicles have a rather beneficial effect on the bridge stability. This is particularly exhibited in the experiment with the vehicles placed in the windward traffic lanes, as well as in all traffic lanes. Although these findings cannot be completely generalized because every bridge has its unique geometry, some indications are provided with respect to important design aspects.

Vibration testing is an effective approach to obtaining the dynamic parameters of a bridge and evaluating its state of health. It is carried out mainly on the basis of wired or wireless sensor networks, which consist of expensive sensors and data-acquisition and -transmission systems, and professional installation is required. Integrated with networks, sensors, and storage capability, smartphones can be convenient to use for vibration testing, and the dependence on professional installation can be reduced. Toward this end, an innovative smartphone-based mobile testing method is proposed and demonstrated in this paper. First, software with two acquisition modes for the iPhone 4s was developed. Second, smartphone-based cable-force tests were performed in the laboratory to validate the feasibility of the proposed technique. Last, the technique was applied to the Hualu and Sifangtai bridges to validate the feasibility of using this technique on real bridges. The results confirm the convenience and efficiency of the proposed technique for quick cable-force measurement.

Failures of strand–concrete bonds have been widely observed in load tests of precast-pretensioned concrete I girders. Through a review of 22 test programs, 15 terminologies were identified to describe failures associated with bond loss. In many cases, previous researchers used different terms to describe the same failure behavior. In response to the wide-ranging and sometimes inconsistent terminologies used in the literature, this technical note makes two contributions. First, the different types of failure that involve strand–concrete bond loss are characterized, and a consistent labeling scheme is proposed. The 15 labels given in the referenced test programs are condensed into four primary behaviors. Second, a flowchart is presented for assisting future researchers in characterizing and labeling bond-loss failures. Decision points in the flowchart are based on a synthesis of the reviewed test programs, which included 120 unique load tests having some type of bond-loss failure.