Segmental post-tensioned (PT) bridges are major structures that carry significant traffic. Recent investigations of these bridges have identified voids in their ducts. and some of these exposed strands at these void locations are undergoing corrosion. The corrosion of strands may lead to the failure of tendons. As such, an effective inspection process for identifying these voids is needed. From a literature review, several non-destructive testing (NDT) methods are compared for applicability in inspecting voids in external tendons. The impact echo (IE), ultrasonic pulse velocity (UPV), and sounding inspection methods were selected and assessed for identifying voids in preliminary test setup. The sounding inspection method is further assessed for its effectiveness in identifying voids in a full-scale, external tendon system. The results indicate that the sounding inspection slightly underestimates the size of the voids. However, the inspected size and locations of voids have a close correlation with actual voids in ducts. Thus, the sounding inspection can be an effective tool for identifying voids because of its easy application in the field. Recently, the investigated failures of segmental post-tensioned (PT) bridges called attention to the rehabilitation and mitigation methods of voided ducts in PT structures. Although controversy exists on how to best protect PT tendons from corrosion, filling these voids with grout may be one option. An optimized grouting procedure for repairing these voids is needed how best to protect the strands from corrosive environments. This research investigates three grouting methods for efficiently repairing the voids in PT duct systems. These methods are (1) vacuum grouting (VG), (2) pressure grouting (PG), and (3) pressure-vacuum grouting (PVG). Each method is being evaluated for filling capability, filling performance, and economic feasibility. Also, three different pre-packaged grouts for repair are assessed in this research to propose the most suitable material for repairing voided PT ducts. The results indicate that the PG and PVG methods are more constructible and likely more economical than the VG method. However, the PVG and VG methods seem to be more effective than the PG method in filling the voids. As a result of these tests, the PVG method is recommended for filling voids in tendons. The results also show that C-1 and C-2 grouts have better filling capability than C-3 grout. Although experimental tests using prototype specimens of external PT tendons are performed to propose an effective repair grouting method and material, the experimental conditions cannot cover all voids types, duct types, and other effects of repair grouting methods in the field. Thus, the grout flow in voided ducts is predicted using a commercial Computational Fluid Dynamics (CFD) program. The simulation of the flow is challenging due to the complicated geometry of voided ducts, but the simplified model in this research shows close correlations with experimental results. Thus, various parameters of repair methods and materials are assessed in this research, and the PVG method with grouts having low viscosity exhibited the best performance. If it is determined that filling voids with grout is appropriate and prevents future corrosion, it is recommended that voids in the field be filled using the PVG method with grouts exhibiting low viscosities.

[pt] Este trabalho apresenta um estudo experimental sobre a redistribuição de momentos em vigas contínuas de concreto construídas em aduelas pré-moldadas, com juntas secas, e protendidas com cabos sintéticos externos. Para tanto foram ensaiadas quatro vigas contínuas, com dois vãos iguais, sendo uma monolítica e as demais em aduelas pré-moldadas. Todas as vigas têm a mesma seção transversal tipo I, com dois vãos, altura da seção de 30 cm, largura do flange e espessura da alma igual a 30 cm e 10 cm, respectivamente. A viga monolítica tem relação 1/dp (vão/altura) igual a 18,75 e nas vigas em aduelas os valores desta relação foram entre 12,5, 18,75 e 25. As vigas são protendidas com dois cabos sintéticos externos com traçado poligonal. Os cabos utilizados são feitos com fibras sintéticas de alta resistências (2700 MPa) e alto modulo de elasticidade (126000 MPa) conhecidas comercialmente como Kevlar. O objetivo deste estudo é examinar a redistribuição de momentos em vigas contínuas com sistemas construtivos diferentes e com diferentes relações 1/dp frente à utilização de cabos sintéticos. Os resultados mostram que a redistribuição de momentos em vigas em aduelas pode ser obtida a partir de uma viga monolítica semelhante e que o comportamento da redistribuição de momentos nas vigas em aduelas é pouco influenciada pela relação l/dp. / [en] An experimental investigation on the behavior of continuous segmental concrete beams, prestressed with external aramid tendons, was carried out. Four beams were tested; one was monolithic and the others were constructed in precast dry- jointed segments. All the beams had the same overall dimensions, with two spans and an I section 30 cm high, flange width of 30 cm and web thickness of 10 cm. The 1/dp span/section height) ratio was 18,75 for the monolithic beam and 12.5, 18.75 and 25 for the segmental beams. The beams were post-tensioned with two external aramid tendons, which are made of high strength (2700 MPa) high modulus (126000 MPa)Kevlar 49 yarns. The main objective was to study the moment redistribution in the post-cracking stage. Test results have shown that the moment redistribution in segmental beams can be obtained from a similar monolithic beam and that the moment redistribution is only slightly affected by the 1/dp.