The goal of this research is to evaluate the immediate deflections of concrete beams reinforced by carbon FRP grids with various fiber placement designs. Manufacturing and testing various FRP grid designs are the first steps toward the long-term goal of developing FRP reinforcement with optimized strength and serviceability performance. Four grid designs with stiffnesses that ranged from 47.1 kN/mm to 33.6 kN/mm, as determined from stand-alone tensile tests, were used as reinforcement in concrete beams and tested in flexure via three-point loading. The flexural results are in good agreement with the deflections as determined from a modified version of the ACI 318 flexure equations.

One of the major disadvantages of the plate bonding technology is the premature and brittle debonding failure of the bonded plate. It has been assumed, quite logically, that the stress concentration at the plate end is the primary cause of such premature plate debonding failure. However, there is no direct evidence of the validity of these stresses as to whether the predicted stresses agree with the experimental data or not. Also there is concern if they can form the basis and criteria for the design and prevention of debonding failures. This paper presents a critical analysis of the calculated peak shear and normal stress values at the plate end using Roberts’ approximate model, and derived from a wide range of published data involving steel, glass fibre reinforced polymer (GFRP) and carbon fibre reinforced polymer (CFRP) plates. It is shown that these calculated stresses are far too high, and cover unacceptably wide range of values, without any consistent pattern of variation with the plate stiffness. It is clear that the peak stresses are influenced by other parameters which are not taken into account in the approximate model used in the calculations. The wide range of the peak stresses obtained from a large number of tests seems to indicate that these stresses cannot form a reliable basis to explain or design the prevention of premature plate debonding failures.

Large-scale (80%) tests were conducted on one "as-built" and four composite jacketed rectangular flexural bridge spandrel columns to assess the effectiveness of different retrofit schemes using fiber reinforced polymer composite jackets. Retrofit challenges were in (1) the unknown response of the inclined interface between spandrel column and the arch rib and (2) the behavior of the column reinforcement lap splice located at the top of the spandrel column pedestal. Three of the four FRP retrofit systems only addressed the lap splice region, where as the fourth system connected the column jacket to the arch rib to improve the column/arch rib interface response. Final damage patterns and failure modes showed that only the latter scheme improved the seismic response whereas the other systems resulted in a sliding failure mode without improving the displacement capacity which for the prototype bridge response is less desirable than the original “as-built” lap splice debonding failure. All retrofit schemes successfully clamped the column reinforcement lap splice above the column pedestal construction joint. The tests showed that fiber reinforced polymer composite jacketing systems clearly can be installed without affecting the overall geometry or appearance of the structure, and emphasizes the importance of designing retrofit strategies to control the mode of failure. Retrofitting of one weakness without considering the next mode of failure can lead to ineffective and poor designs.

The objective of this paper is to develop the effective retrofitting method applicable to reinforced concrete columns connected with monolithically cast wing walls by CFRP (Carbon Fiber Reinforced Plastics) sheet jacketing. To clarify the improving mechanism of deformation capacity under earthquake loading by applying CFRP sheet jacketing, eight column specimens with or without wing walls were tested under reversals of horizontal load mainly in the large deformation zone. Test results show that CFRP sheet jacketing methods proposed in this paper are useful for retrofitting works of existing apartment buildings designed in accordance with the old building code. Based on the test evidence, a proposal for estimating the ultimate strength and the deformation capacity of columns with wing walls strengthened by CFRP sheet jacketing are presented.

Investigation of the long-term characteristics of the Highly Expansive Material (HEM) anchorage for CFRP strands is very important. In the post-tensioning type of prestressed concrete structures, considerations should be made for the loss of prestressing force due to the pull-out displacement which is caused by the creep of the HEM. The long-term characteristics of the HEM anchorage were investigated by creep test on five specimens. From the creep test, some important characteristics of HEM anchorage were observed, for example, time-dependent change of pull-out displacement at the loaded end, unit shear “q” distributions and the tensile force distributions “Tp”. An analytical relationship on how the long-term behavior of prestressing force can be predicted by using the measured values for the time-dependent change of pull-out displacement at the loaded end is presented. Also from the simulated results of this relationship, it was found that the loss of prestressing force is negligible in practice when the CFRP strand is 10 meters long. Normally the expansive pressure of HEM at prestressing is 50MPa. However, when the expansive pressure is 100MPa, the pull-out displacement at the loaded end and the loss of prestressing force can be reduced to more than the half of one with 50MPa.