Experimental research was conducted to investigate the influence of axial load and prestress on the shear strength of web-shear critical reinforced concrete elements. The ability of two design codes, the ACI code and the CSA code, to accurately predict the shear strength of web-shear critical reinforced concrete elements was investigated through two sets of experiments performed for this thesis, the panel tests and the beam tests. The experimental results indicated that the CSA code provided better predictions for the shear strength of web-shear critical reinforced concrete members subjected to combined axial force and shear force than the ACI code. A total of six panels, reinforced almost identically, were tested under different combinations of uni-axial stress and shear stress. In addition to the panel tests, a total of eleven I-shaped beams, with the same web thickness, were tested under different combinations of axial force and shear force. The parameters for these beams were the amount of longitudinal reinforcement, the amount of transverse reinforcement, and the thickness of the flanges. The beams were simply supported, but the loading geometry was specially designed to simulate the loading conditions in continuous beams near points of inflection. The experimental results from the panel tests and the beam tests followed a similar trend of variations. Both the inclined cracking strength and the ultimate shear strength were increased by compression and were reduced by tension. The specimens subjected to very high compression failed explosively without developing many cracks. The inclined cracking strength could be predicted with good accuracy if the influence of the co-existing compression on the cracking strength of the concrete and the non-uniform distribution of the stresses over the depth of the cross-section were considered. The strength predictions using the ACI code for these tests were neither accurate nor consistent. The ACI code was unconservative for members subjected to compression and was excessively conservative for members subjected to tension. In contrast, the strength predictions using the CSA code for these tests were generally conservative and consistent. The CSA code accurately predicted the response of specimens subjected to compression and was somewhat conservative in predicting the shear strength of specimens subjected to tension.

Doctor of Philosophy / Department of Civil Engineering / Robert J. Peterman / This dissertation describes a testing program to evaluate the bond and splitting propensity characteristics of 5.32-mm-diameter prestressing wires. Prestressing wire reinforcement is used primarily in the production of prestressed concrete railroad ties. Twelve different 5.32-mm-diameter wires were tested in this study in order to measure bonding characteristics of the reinforcement. Establishment of the bond-slip characteristics of these reinforcement at both transfer of prestress (transfer bond) and under flexural loading (flexural bond) is necessary to enable the accurate modeling of these ties using finite elements. Transfer bond and flexure bond of various indent patterns were tested using tensioned pullouts. Specimens of various sizes with single or multiple wires were tested to determine the effects of cover and wire number on bond. Indents were machined on smooth prestressing wires to accurately compare indent geometries. Lateral expansion was tested to determine which wires have higher propensity to cause cracking or splitting. Crossties were instrumented to compare resulting lateral expansion with results found in the laboratory. The results from the testing program showed that the tensioned pullout test was able to be used to predict the transfer length of prisms made with the same reinforcement. The results also showed that the indent geometries were able to be used to predict the splitting of specimens based on the amount of slip the wire had experienced. The testing also showed the importance of concrete cover with the relation to splitting potential.

Master of Science / Department of Civil Engineering / Robert J. Peterman / With aging and deterioration of bridges, evaluation of existing conditions of their structural elements becomes vital to engineers and public officials when deciding how to repair or replace the structures. The ability to obtain necessary information on these conditions is often expensive and time consuming, especially for concrete bridges where the reinforcement is not available for inspection. Employing the surface-strain relief method could allow for accurate evaluation of aged or damaged prestressed members. The surface-strain relief method was developed to measure initial or pre-existing strains in a concrete member. It involves relieving the strain in the member and measuring the change in strain. Two methods were tested—one used a linear electrical-resistance strain gage and a three-inch-diameter diamond concrete core bit to cut around the gage, and the second method used a laser-speckle imaging device and a diamond cutting wheel to create notches perpendicular to the axis of maximum strain. Both methods measured the change in strain and related it to within 10 percent of the actual fse. The method of cutting notches and the laser-speckle imaging device provided a simpler method to be implemented in the field, while the coring method achieved a higher level of accuracy and precision.

During the mid 70's and early 80's, several precast deck panel bridges were constructed in Florida. These utilize prestressed precast panels as stay-in-place forms and are designed to act compositely with a cast-in-place deck which is poured subsequently. Such bridges offer advantages of quicker construction and lower costs. However, several such bridges built in Florida developed extensive cracking and spalling. Following localized failures, the Florida Department of Transportation have decided to replace all 127 precast panel deck bridges in Districts 1 and 7. Since deck replacement is contingent on funding, it is necessary to develop a rational procedure to decide the order in which they are replaced. This work describes the calibration of a software program developed to assist in the replacement prioritization of panel bridges. Prioritization is based on information available with FDOT in the form of biannual inspection reports over the past 20+ years. A new computer language called BRAILE ( BRidge Annual Inspection LanguagE) was developed to input the data from the inspection reports. A compute program PANEL was used to process the BRAILE data to identify trends that could be used to assign weights to observed deficiencies. The prioritization developed based on the method presented here was found to correctly identify high risk bridges and was reliable than one based solely on National Bridge Inventory (NBI) condition rating.

The prestress loss estimation provision in the AASHTO LRFD Bridge Design Specifications was recalibrated in 2005 to be more accurate for "high-strength [conventional] concrete." Greater accuracy may imply less conservatism, the result of which may be flexural cracking of beams under service loads. Concern with a potential lack of conservatism and the degree of complexity of these recalibrated prestress loss estimation provisions prompted the investigation to be discussed in this dissertation. The primary objectives of this investigation were: (1) to assess the conservatism and accuracy of the current prestress loss provisions, (2) to identify the benefits and weaknesses of using the AASHTO LRFD 2004 and 2005 prestress loss provisions, and (3) to make recommendations to simplify the current provisions. These objectives were accomplished through (1) the fabrication, conditioning, and testing of 30 field-representative girders, (2) the assembly and analysis of a prestress loss database unmatched in size and diversity when compared with previously assembled databases, and (3) a parametric study investigating the design implications and sensitivity of the current loss provisions. Based on the database evaluation coupled with the experimental results, it was revealed that the use of the AASHTO LRFD 2005 prestress loss provisions resulted in underestimation of the prestress loss in nearly half of all cases. A loss estimation procedure was developed based on the AASHTO LRFD 2005 provisions to greatly simplify the procedure and provide a reasonable level of conservatism. / text

Prestressed concrete construction relies on the application of compressive stresses to concrete elements. The prestressing force is typically applied through the tensioning of strands that react against the concrete and induce compression in the concrete. Loss of prestress is the decrease of this pre-applied stress. The conservative estimation of the prestress losses is imperative to prevent undesired cracking of the prestressed element under service loads. A large fraction of the prestress losses is a consequence of concrete deformations. This fraction of the losses can be identified as strain-related losses, and these occur due to instantaneous elastic shortening, and time-dependent creep and shrinkage. Creep and shrinkage of concrete depend on many factors that are extremely variable within concrete structures. The time-dependent behavior of concrete is not well-understood, but recent findings in the topics of concrete creep and shrinkage provide a better understanding of the underlying mechanisms affecting the nature of these two phenomena. However, current design practices and prestress loss estimation methods do not reflect the state-of-the-art knowledge regarding creep and shrinkage. The main objective of this dissertation was the study and estimation of strain-related prestress losses in simply supported pretensioned bridge girders. Simply supported pretensioned girders are widely designed, produced and frequently used in bridge construction. Due to this common use, pretensioned concrete bridge girders has become fairly standardized elements, which results in a reduced variability in the behavior of pretensioned bridge girders, as compare to that of less standardized concrete structures. Hence, a simplified method was calibrated to estimate prestress losses within pretensioned girders to an adequate level of accuracy. To achieve an acceptable accuracy experimental data from the monitoring of pretensioned simply supported girders was used for the calibration of the method. The accuracy of this simplified method is comparable to that achievable using more elaborate methods developed for generic concrete structures. / text

Corroded steel in concrete is a structural issue that plaques concrete structures in coastal regions. Traditionally corroded steel strength is calculated from a distributed area loss due to corrosion over the entire surface of the steel and reducing the capacity accordingly. In reality, corrosion attacks localized regions creating pits and reducing the cross section in a small region which amplifies the effects of corrosion. Stress concentrations at the corrosion pitting damage may further reduce the tensile capacity of the steel. A study of corrosion damage and strength associated with pitting damage can assist in understanding the ultimate tensile capacity of corroded steel strands, better correlations are needed to estimate actual strength of damaged steel. The focus of this thesis is on seven-wire prestress steel strands with various stages of induced corrosion. Each strand has been documented, profiled, and measured in order to correlate physical damage with ultimate capacity.

Aramid fiber reinforced polymer (AFRP) composite materials show promise for prestressed concrete bridge applications. However, there are still some knowledge gaps due to lack of sufficient data to assess the long-term performance and therefore sustainability of beams prestressed with AFRP composite materials. The objective of this research is to effectively characterize the material properties based on the short-term and long-term characteristics of AFRP bars. Tensile, creep-rupture, and relaxation tests are experimentally conducted using AFRP bars to validate testing procedures and expand an existing limited database. Previous results from tensile tests show that the stress-strain behavior of Arapree® AFRP bars is linear until failure with tensile strength of approximately 210 ksi (1448 MPa) and strain of 2.1%. For the creep-rupture tests, three specimens are tested and monitored at four different load levels (50, 60, 75 and 85% of maximum tensile strength) throughout a period of 14 days (short-term evaluation) and 42 days (long-term evaluation). From these tests, it is expected that for a 100-year life span, 55% of the ultimate load, Fu, must be applied as an initial stress to obtain a long-term residual strength of 0.80 Fu. For the relaxation tests, six specimens at four different strain levels (50, 60, 75 and 85% of maximum tensile strain) are tested and monitored throughout a period of 14 days and 42 days. Relaxation loss profiles of the AFRP bars are developed based on the experimental data collected from prestressed AFRP bars, which have been less well understood given lack of sufficient experimental data. Overall, the results of this study provide more insight as to the reliability and potential long-term performance of AFRP bars embedded within prestressed bridge structures.

Experimental research was conducted to investigate the influence of axial load and prestress on the shear strength of web-shear critical reinforced concrete elements. The ability of two design codes, the ACI code and the CSA code, to accurately predict the shear strength of web-shear critical reinforced concrete elements was investigated through two sets of experiments performed for this thesis, the panel tests and the beam tests. The experimental results indicated that the CSA code provided better predictions for the shear strength of web-shear critical reinforced concrete members subjected to combined axial force and shear force than the ACI code. A total of six panels, reinforced almost identically, were tested under different combinations of uni-axial stress and shear stress. In addition to the panel tests, a total of eleven I-shaped beams, with the same web thickness, were tested under different combinations of axial force and shear force. The parameters for these beams were the amount of longitudinal reinforcement, the amount of transverse reinforcement, and the thickness of the flanges. The beams were simply supported, but the loading geometry was specially designed to simulate the loading conditions in continuous beams near points of inflection. The experimental results from the panel tests and the beam tests followed a similar trend of variations. Both the inclined cracking strength and the ultimate shear strength were increased by compression and were reduced by tension. The specimens subjected to very high compression failed explosively without developing many cracks. The inclined cracking strength could be predicted with good accuracy if the influence of the co-existing compression on the cracking strength of the concrete and the non-uniform distribution of the stresses over the depth of the cross-section were considered. The strength predictions using the ACI code for these tests were neither accurate nor consistent. The ACI code was unconservative for members subjected to compression and was excessively conservative for members subjected to tension. In contrast, the strength predictions using the CSA code for these tests were generally conservative and consistent. The CSA code accurately predicted the response of specimens subjected to compression and was somewhat conservative in predicting the shear strength of specimens subjected to tension.

The purpose of this study is to investigate how accurately the distribution factor method estimates the live load deflections under the principles of the 2012 AASHTO LRFD Bridge Design Specifications (AASHTO LRFD specifications) compared to the results of the NISA finite element analysis software. The simple span bridge model analyzed is developed very similarly to the design example of the PCI Bridge Design Manual. The main difference is a shorter span length and smaller AASHTO-PCI bulb tee sections. Three main finite element models are created to estimate the live load deflections under the recommended live load conditions as per AASHTO LRFD specifications. The first model is simulated with simple support conditions. The purpose of this model is two-fold: compare the deflections to the distribution factor method and to the deflections of the second model that is simulated with elastomeric steel reinforced bearing pads. Thus, the stiffnesses of the elastomeric bearing pads of the second model are varied within the AASHTO LRFD specifications acceptable limits and under low temperature conditions the stiffness is increased accordingly for two cases. The purpose is to investigate if the stiffness have any significant affect on the deflections of the girders. Then a third model is created to investigate if the removal of the intermediate diaphragms have any affect on the deflections. The results of the first and second models, including the models with the allowed varied stiffnesses of the bearing pads, found only the interior girders deflecting up to 4% more and the exterior girders were deflecting up to 5.55% less than the estimates of the distribution factor method. In the case when the diaphragms are removed, the deflections of the inner most interior girders are deflecting up to 10.85% more compared to the same girders of the model which includes the intermediate diaphragms and the bearing pads. In the unique case of the second model where the bearing pads may stiffen significantly under low temperatures, the girders are deflecting up to 23% less than when at room temperature conditions. All these findings and other summarized results are discussed in greater detail in this study.