This thesis describes the experimental appraisal of a series of 10 'non-sway' steel column subassemblages, each comprising a 6m long column with up to three 1.5m long beams, together with two full-scale 3 storey, 2 bay, single span, non-sway steel frames (typical overall dimensions 9m x 10m x 3.5m). The subassemblages tests were conducted in the Department of Civil and Structural Engineering at the University of Sheffield whilst the much larger frame tests were carried out at the Building Research Establishment. In all cases, the beam and column elements were connected using 'simple' bolted steelwork connections. The aim was to investigate the effect of the inherent rotational stiffness (semi-rigid characteristics) of such connections on the behaviour of steel frames in which the columns were loaded biaxially and were not restricted to in-plane deformation. The appraisal of the results from these experiments clearly shows that the stiffness of even the most modest connection can have a significant influence on the distribution of bending moments, the ultimate column capacity and deflection of frame members. The experimental data were subsequently used to validate the predictions of a sophisticated finite-element computer program which was developed specifically to analyse 3-dimensional column subassemblages employing semi-rigid connections. This thesis documents this validation and reports the findings of an extensive parametric study which was then conducted to investigate the influence of semi-rigid connection behaviour on a wide range of subassemblage configurations. Comparisons with the experimentally observed and analytically predicted ultimate capacities of the subassemblage and frame tests showed that 'commonly used' methods of frame design are unduly conservative. The author has therefore proposed a number of design approaches for both ultimate and serviceability limit state loading conditions which take into account the inherent benefits of semi-rigid joint action.

Steel bridges and structures often need strengthening due to increased live loads, or repair due to corrosion or fatigue cracking. This thesis explores the use of adhesively bonded prestressed carbon fibre reinforced polymers (CFRP) strips in retrofitting intact steel girders, through experimental and analytical investigations. The first part of the research program investigates the behaviour of CFRP-strengthened steel beams comprised of W Structural Sections (W ) with cover plates welded to the tension flange. Six beams, 2000 mm long, were tested under cyclic loads to examine the effects of CFRP strip strengthening on the fatigue life. The CFRP strip prestressing process, type of CFRP strip, level of prestressing, and the location of the CFRP strips were the main parameters examined in this study. Debonding at the end of strip was a significant problem that can be controlled by applying a proper end clamp. The maximum increase in fatigue life observed in the experiments was 125 percent, for a specimen strengthened using high modulus CFRP strips bonded onto the cover plates with the highest level of prestressing. An analytical model and a finite element model were developed for analyzing the strengthened beams. A fracture mechanic analysis was performed to investigate the effects of prestressing on the crack growth rates at the critical weld toe. The models were verified using experimental results, and then used to perform parametric studies. It is shown that the effectiveness of reinforcement is greatest for beams with strips on the cover plate, higher CFRP elastic modulus, and higher prestressing level. In general, this study demonstrates that steel beams can indeed be successfully strengthened or repaired using prestressed CFRP materials.

Steel bridges and structures often need strengthening due to increased live loads, or repair due to corrosion or fatigue cracking. This thesis explores the use of adhesively bonded prestressed carbon fibre reinforced polymers (CFRP) strips in retrofitting intact steel girders, through experimental and analytical investigations. The first part of the research program investigates the behaviour of CFRP-strengthened steel beams comprised of W Structural Sections (W ) with cover plates welded to the tension flange. Six beams, 2000 mm long, were tested under cyclic loads to examine the effects of CFRP strip strengthening on the fatigue life. The CFRP strip prestressing process, type of CFRP strip, level of prestressing, and the location of the CFRP strips were the main parameters examined in this study. Debonding at the end of strip was a significant problem that can be controlled by applying a proper end clamp. The maximum increase in fatigue life observed in the experiments was 125 percent, for a specimen strengthened using high modulus CFRP strips bonded onto the cover plates with the highest level of prestressing. An analytical model and a finite element model were developed for analyzing the strengthened beams. A fracture mechanic analysis was performed to investigate the effects of prestressing on the crack growth rates at the critical weld toe. The models were verified using experimental results, and then used to perform parametric studies. It is shown that the effectiveness of reinforcement is greatest for beams with strips on the cover plate, higher CFRP elastic modulus, and higher prestressing level. In general, this study demonstrates that steel beams can indeed be successfully strengthened or repaired using prestressed CFRP materials.

Current specifications of the American Association of State Highway and Transportation Officials (AASHTO) include restrictions on the live load deflections of highway bridge girders. Conventional practice, which utilizes hand calculations to estimate girder deflections, assumes that all girders of a highway bridge deflect to the same degree. In addition, the conventional equations do not account for AASHTO specifications requiring the evaluation of extreme force effects. As such, the accuracy of the conventional approach for calculating girder deflections is under question. The purpose of this study is, therefore, to check the accuracy of the conventional approach by testing the two aforementioned assumptions made by the equations. A composite steel girder bridge example has been selected from Design of Highway Bridges: An LRFD Approach, Third Edition by Richard M. Barker and Jay A. Puckett. The design example specifies the dimensions for all structural elements, as well as the girder type and spacing. The design example does not include specifications for the bridge bearings, and so bearing pads are designed according to the Illinois Department of Transportation (IDOT) Bridge Manual (2012). This study consists of two steps. First, a hand-calculated live load deflection for the bridge example is derived from the conventional approach (assuming all girders deflect to the same degree and without consideration for extreme force effects). Next, the finite element analysis software, NISA/Display IV, is utilized to model and analyze the real-world deflections of the bridge model. Three live loading conditions are applied to the finite element model, in accordance with AASHTO specifications. For first live load condition, the live loads are positioned at the center of each traffic lane. The second and third conditions apply extreme force effects to an interior girder and exterior girder, respectively. The results for each finite element analysis are then compared with the conventional, hand-calculated deflection. The results of this study contradict the two aforementioned assumptions made by the conventional equations for calculating girder deflections. Firstly, this study demonstrates that interior girders experience a significantly greater live load deflection than interior girders. More importantly, the results indicate that the conventional equations underestimate the live load deflection of an interior girder subjected to extreme force effects. None of the results, however, suggest that the bridge example is at risk of excessive deformation, and so the extent to which these drawbacks present a concern can be left to the discretion of the engineer.

The Utah Transportation Center, in conjunction with the Mountain Plains Consortium, sponsored a study that investigated the distribution factors and load ratings of a continuous, steel I-girder bridge. The SH-52 Bridge over the Snake River is located on the Idaho-Oregon border near Payette, Idaho. The bridge was built in the 1950’s and presently supports two lanes of traffic. A finite-element model of the bridge was calibrated with the results from a liveload test. For the live-load test, the bridge was instrumented at nine longitudinal cross section locations with 62 strain gauges attached on the girders, stringers, and intermediate diaphragms. The live-load was applied with two heavy trucks that were driven along three predetermined load paths. The calibrated finite-element model was used to quantify moment distribution factors and load ratings for the bridge. The finite-element distribution factors were compared to those calculated according to the AASHTO Standard and AASHTO LRFD Specifications. The distribution factors from both AASHTO codes were found to be unconservative for the girders and overly conservative for the stringers. The model was also used to quantify the effect of the transverse diaphragm members on the live-load distribution. Distribution factors were calculated with and without the diaphragm members. The diaphragms were found to increase the distribution of moments by over 20% for both positive and negative moments.

Economical constraints on the design of bridges usually necessitate the use of as few girders as possible across the bridge width. The girders are typically uniformly spaced transversely with the deck extending past the fascia girders, thereby resulting in an overhang. While designers commonly employ rules of thumb with regard to the geometry of the overhang, these rules of thumb generally lack research justification and the actual girder behavior is not well understood. Overhang construction often produces torsinally unbalanced loading on the girder system, which can lead to problems in steel and concrete girder bridges during construction. The main issue with concrete girder bridges is excessive lateral rotation in the fascia girder, which can cause potential problems of construction safety and maintenance. Field problems on concrete bridges have been reported in the state of Texas where the fascia girders experienced excessive rotation during construction. For steel girder bridges, the unbalanced overhang loading can lead to both local and global instability. Locally, the overhang brackets often exert a large force on the web plate that can distort the web and increase the magnitude of the plate imperfection. Global stability problems have also occurred primarily on bridge widening projects when a few girders are added to an existing bridge system. The girders in the widening are usually isolated from the existing bridge and the unbalanced load from the overhang can cause excessive twist that intensifies the global stability of the girder system. The objective of this study was to improve the understanding of the bridge behavior due to the unbalanced loading from the overhangs and to identify critical factors affecting the girder behavior. The study was also aimed at developing simple design methodologies and design recommendations for overhang construction. The research included field monitoring, laboratory tests, and parametric finite element analyses. The data from the field monitoring and laboratory tests were used to validate finite element models for both concrete and steel girder bridges. Based on the validated models, detailed parametric studies were conducted to investigate the effects of the unbalanced loading. Results from the parametric studies were used to identify the geometries of girder systems that are prone to problems with the overhangs as well as to provide design suggestions. In addition, a closed-form solution for lateral rotation in the fascia girder in a concrete girder bridge was derived using a rigid-body model, and was used to develop design methodology and design recommendations for overhang construction. / text

Lateral bracing systems consisting of cross frames and their connections play a significant role in the elastic buckling strength of steel girder bridges. By providing lateral and torsional stability, they prevent lateral torsional buckling of the girder during bridge construction prior to the concrete bridge deck curing. To perform this function, the bracing system must possess adequate strength and stiffness. And since each component of the bracing system acts in series, the overall stiffness of the system is less than the least stiff component. In skewed bridges, cross frames at the ends of the girders are installed parallel to the bridge skew angle, and their connection to the girder requires that the cross frames be at an angle that prohibits welding a stiffener from the cross frame directly to the girder web. To make this connection, many states use a bent plate to span the angle between the web stiffener and cross frame. While this bent plate connection is now being widely used, it has never been rationally designed to account for its strength or stiffness in the bracing system. Results from field studies show that the bent plate connection may be limiting the cross frame stiffness thereby hampering its ability to provide stability to the girder during construction. The result is significant girder end rotations. The purpose of this research is to classify the impact of the bent plate connection on the end cross frame stiffness in skewed straight steel girder bridges and propose methods to improve the end cross frame’s structural efficiency. This research uses laboratory testing, finite element modeling, and parametric studies to recommend design guidance and construction practices related to the end cross frames of skewed steel girder bridges. In addition to recommending methods to stiffen the existing bent plate connection, an alternative pipe stiffener connection is evaluated. The pipe stiffener not only offers the possibility of a stiffer connection, but can also provide warping restraint to the end of the girder which may significantly increase the girder elastic buckling capacity. / text

The scheduled replacement of the 8th North Bridge, in Salt Lake City, UT, presented a unique opportunity to test a pre-cast concrete deck, steel girder bridge. A live-load test was performed under the directions of Bridge Diagnostic Inc (BDI) and Utah State University. Six different load paths were chosen to be tested. The recorded data was used to calibrate a finite-element model of this superstructure, which was created using solid, shell, and frame elements. A comparison between the measured and finite-element response was performed and it was determined that the finite-element model replicated the measured results within 3.5% of the actual values. This model was later used to obtain theoretical live-load distribution factors, which were compared with the AASHTO LRFD Specifications estimations. The analysis was performed for the actual condition of the bridge and the original case of the bridge, which included sidewalks on both sides. The comparison showed that the code over predicted the behavior of the actual structure by 10%. For the original case, the code's estimation differed by as much as 45% of the theoretical values. Another opportunity was presented to test the behavior of a cast-in-place concrete box girder bridge in Joaquin County, CA. The Walnut Grove Bridge was tested by BDI at the request of Utah State University. The test was performed with six different load paths and the recorded data was used to calibrate a finite-element model of the structure. The bridge was modeled using shell elements and the supports were modeled using solid elements. The model was shown to replicate the actual behavior of the bridge to within 3% of the measured values. The calibrated model was then used to calculate the theoretical live-load distribution factors, which allowed a comparison of the results with the AASHTOO LRFD Specifications equations. This analysis was performed for the real conditions of the bridge and a second case where intermediate diaphragms were not included. It was determined that the code's equations estimated the behavior of the interior girder more accurately for the second model (within 10%) than the real model of the bridge (within 20%).

The majority of the bridges in the United States are already reaching the years that the design process took into account when determining the time the structure would be functional. This means that many of the bridges in the nation are in need of increasing maintenance, and in some cases, major retrofitting. Researchers at Utah State University in conjunction with the Long-Term Bridge Performance (LTBP) Program, under the direction of the Federal Highway Administration’s (FHWA’s) Office of Infrastructure Research and Development, directed dynamic testing on the New Jersey Pilot Bridge, structure number 1618-150. The purpose of the LTBP Program is to monitor the nation’s highway bridges for a 20-year period to analyze and understand the behavior over time of the selected bridges and then promote the safety, mobility, longevity, and reliability on those bridges. In order to perform the monitoring of the bridge, ambient vibration analysis was selected for this structure, which was instrumented with an array of velocity transducers to record the response coming from the excitation. A finite element model was also created to compare the results from the ambient vibration testing. The results of this testing will be used with the LTBP Program to improve the knowledge of the bridge performance and foster the next generation of bridges and bridge management in the nation.

Cross frames are critical components in steel bridge systems. Cross frames brace girders against lateral torsional buckling and assist in distributing live loads to girders during the service life of the bridge. In curved bridges, cross frames also serve as primary structural members in resisting torsion generated by the traffic loads. The conventional cross frames are often constructed in X- or K- type shapes with steel angle sections. However, the actual stiffness of these cross frames are not well understood or quantified, leading to potentially inaccurate prediction of bridge behavior and safety during construction and in service. Previous studies have shown the possibility of employing new sections, such as tubular members and double angles, in cross frame designs. In addition, a type-Z cross frame, or single diagonal cross frame was also found to be a potential use to simplify the design. However, the effectiveness of these innovative cross frame types has not been completely examined. And these new cross frames have yet compared with the conventional ones in terms of their stiffness and strength capacity. This dissertation documents the results of a study on the stiffness of various types of cross frame systems. Full size cross frames were tested to establish actual stiffness of the cross frames specimens. The tests results revealed a significant discrepancy between the actual measured stiffness and the stiffness calculated using methods commonly employed by bridge designers. The research showed that the major source of this discrepancy was eccentricity in the connection. The stiffness reduction was quantified by employing analytical derivation and finite element modeling. As a result, methods were developed to account for the stiffness reduction. / text