Post-tensioning tendons in segmental bridge construction are often only anchored within the deviator and pier segments. The effectiveness of the post-tensioning (PT) system is therefore dependent on proper functioning of the anchorages. On August 28, 2000 a routine inspection of the Mid-Bay Bridge (Okaloosa County, Florida) revealed corrosion in numerous PT tendons. Moreover, one of the 19-strand tendons was completely slacked, with later inspection revealing a corrosion-induced failure at the pier anchor location. Anchorage failure caused all PT force to transfer to the steel duct located within the pier segment that in turn slipped and caused the tendon to go completely slack. After the application of PT force, the anchorage assembly and steel pipes that house the tendon are filled with grout. These short grouted regions could, in the event of anchorage failure, provide a secondary anchorage mechanism preventing the scenario mentioned above from occurring. This paper presents the results of a full-scale experimental investigation on anchorage tendon pull-out and a finite element model to support the experimental results and interpretation. The study focuses on the length required to develop the in-service PT force within the pier segment grouted steel tube assembly. Seven, twelve, and nineteen 0.6" diameter strand tendons with various development lengths were considered. Recommendations for pier section pipe detailing and design will be discussed.

Various anchorage systems including both cast-in-place and post-installed anchors have been developed for fastening both non-structural and structural components to concrete structures. The need for increased flexibility in the design of new structures and strengthening of existing concrete structures has led to increased use of various metallic anchors in practice. Although millions of fasteners are used each year in the construction industry around the world, knowledge of the fastening technology remains poor. In a sustainable society, buildings and structures must, from time to time, be adjusted to meet new demands. Loads on structures must, in general, be increased to comply with new demands, and the structural components and the structural connections must also be upgraded. From the structural connection point of view, the adequacy of the current fastenings for the intended increased load must be determined, and inadequate fastenings must either be replaced or upgraded. The current design models are generally believed to be conservative, although the extent of this behavior is not very clear. To address these issues, the current models must be refined to allow the design of new fastenings and also the assessment of current anchorage systems in practice. The research presented in this thesis consists of numerical and experimental studies of the load-carrying capacity of anchors in concrete structures. Two different types of anchors were studied: (I) cast-in-place headed anchors, and (II) post-installed adhesive anchors. This research focused particularly on the tensile load-carrying capacity of cast-in-place headed anchors and also on the sustained tension loading performance of post-installed adhesive anchors. The overall objective of this research was to provide knowledge for the development of improved methods of designing new fastening systems and assessing the current anchorage systems in practice. For the cast-in-place headed anchors (I), the influence of various parameters including the size of anchor head, thickness of concrete member, amount of orthogonal surface reinforcement, presence of concrete cracks, concrete compressive strength, and addition of steel fibers to concrete were studied. Among these parameters, the influence of the anchor head size, member thickness, surface reinforcement, and cracked concrete was initially evaluated via numerical analysis of headed anchors at various embedment depths. Although these parameters have considerable influence on the anchorage capacity and performance, this influence is not explicitly considered by the current design models. The numerical results showed that the tensile breakout capacity of headed anchors increases with increasing member thickness and/or increasing size of the anchor head or the use of orthogonal surface reinforcement. However, their capacity decreased considerably in cracked concrete. Based on the numerical results, the current theoretical model for the tensile breakout capacity of headed anchors was extended by incorporating several modification factors that take the influence of the investigated parameters into account. In addition, a supplementary experimental study was performed to verify the numerically obtained findings and the proposed refined model. The experimental results corresponded closely to the numerical results, both in terms of failure load and failure pattern, thereby confirming the validity of the proposed model. The validity of the model was further confirmed through experimental results reported in the literature. Additional experiments were performed to determine the influence of the concrete compressive strength and the addition of steel fiber to concrete on the anchorage capacity and performance. These experiments showed that the anchorage capacity and stiffness increase considerably with increasing concrete compressive strength, but the ductility of the anchor decreases. However, the anchorage capacity and ductility increased significantly with the addition of steel fibers to the concrete mixture. The test results also revealed that the tensile breakout capacity of headed anchors in steel fiber-reinforced concrete is significantly underestimated by the current design model. The long-term performance and creep behavior of the post-installed headed anchors (II) was evaluated from the results of long-time tests on adhesive anchors under sustained loads. In this experimental study, adhesive anchors of various sizes were subjected to various sustained load levels for up to 28 years. The anchors were also exposed to several in-service conditions including indoor temperature, variations in the outdoor temperature and humidity, wetness (i.e., water on the surface of concrete), and the presence of salt (setting accelerant) additives in the concrete. Among the tested in-service conditions, variations in the outdoor temperature and humidity had the most adverse effect on the long-term sustained loading performance of the anchors. Based on the test results, recommendations were proposed for maximum sustained load levels under various conditions. The anchors tested under indoor conditions could carry sustained loads of up to 47% of their mean ultimate short-term capacities. However, compared with these anchors, the anchors tested under outdoor conditions exhibited larger creep deformation and failure occurred at sustained loads higher than 23% of their mean ultimate short-term capacities. Salt additives in concrete and wet conditions had negligible influence on the long-term performance of the anchors, although the wet condition resulted in progressive corrosion of the steel. Based on the experimental results, the suitability of the current testing and approval provisions for qualifying adhesive anchors subjected to long-term sustained tensile loads was evaluated. The evaluations revealed that the current approval provisions are not necessarily reliable for qualifying adhesive anchors for long-term sustained loading applications. Recommendations were given for modifying the current provisions to ensure safe long-term performance of adhesive anchors under sustained loads.