A reinforced concrete railway trough bridge has been strengthened and loaded to failure. The aim was to test and calibrate methods developed in the European Research Project "Sustainable Bridges" regarding: (a) condition appraisal and inspection, (b) load carrying capacity analysis, (c) monitoring and (d) strengthening of existing bridges. The tested methods proved to be useful and to give accurate predictions. A failure in combined shear, bending and torsion was reached for an applied mid span load of 11,7 MN. This was well predicted by enhanced methods but 20 to 50 % higher than ultimate load evaluated according to predictions based on common codes and models.

This paper describes how the strengthening of concrete structures with fiber reinforced polymer materials has grown to be a widely used method over most parts of the world. As a way of higher utilization of the Fiber Reinforced Polymers (FRP) prestressing has proved to be beneficial. Most of the research done with prestressing Carbon Fiber Reinforced Polymers (CFRP) for strengthening has been done with surface bonded plates. However, in this paper a presentation is given where CFRP quadratic rods are bonded in the concrete cover in sawed grooves and then immediately prestressed. Testing has, proven this to be an advantageous way of bonding CFRP to the concrete. There is also a tendency that the shear forces between the CFRP and the concrete are transferred more efficiently compared to surface bonded plates and sheets. In the tests performed, no mechanical device has been used to keep the prestress during testing, which then means that the adhesive has to transfer all shear stresses to the concrete. These tests have then been compared with concrete beams strengthened with prestressed external steel and CFRP tendons and similar load carrying capacity has been obtained.

The report contains three parts:

- Strengthening of the Örnsköldsviks Bridge with Near Surface Mounted CFRP Rods

- Strengthening of the Sub-soil at Vitmossen, Avesta, Sweden

- Strengthening of the Frövi Bridge with CFRP Rods and Tubes

Retrofitting concrete structures with fiber reinforced polymer (FRP) has today grown to be a widely used method throughout most parts of the world. The main reason for this is that it is possible to obtain a good strengthening effect with a relatively small work effort. It is also possible to carry out strengthening work without changing the appearance or dimensions of the structure. Nevertheless, when strengthening a structure with external FRP, it is often not possible to make full use of the FRP. The reason for this depends mainly on the fact that a strain distribution exists over the section due to dead load or other loads that cannot be removed during strengthening. This implies that steel yielding in the reinforcement may already be occurring in the service limit state or that compressive failure in the concrete is occurring. By prestressing, a higher utilization of the FRP material is made possible. It is extremely important to ensure that, if external prestressing is used, the force is properly transferred to the structure. Most of the research conducted with prestressing carbon fiber reinforced polymer (CFRP) for strengthening has been on surface bonded laminates. However, this paper presents research on prestressed CFRP quadratic rods bonded in sawed grooves in the concrete cover. This method has proven to be an advantageous means of bonding CFRP to concrete, and in comparison to surface bonded laminates, the shear and normal stress between the CFRP and the concrete are more efficiently transferred to the structure. In the presented test, no mechanical device has been used to maintain the prestress during testing, which means that the adhesive must transfer all shear stresses to the concrete. Fifteen beams with a length of 4 m have been tested. The tests show that the prestressed beams exhibited a higher first-crack load as well as a higher steel-yielding load as compared to nonprestressed strengthened beams. The ultimate load at failure was also higher, as compared to nonprestressed beams, but in relation not as large as for the cracking and yielding. In addition, the beams strengthened with prestressed FRP had a smaller midpoint deflection. All strengthened beams failed due to fiber rupture of the FRP.

Rehabilitation and strengthening of existing concrete structures has become more and more in focus during the last decade. All over the world there are structures intended for living and transportation. The structures are of varying quality and function, but they are all ageing and deteriorating over time. Some of these structures will need to be replaced since they are in such a bad condition. However, it is not only the deterioration processes that make upgrading necessary, errors can have been made during the design or construction phase so that the structure needs to be strengthened before it can be used. The causes for repair and/or strengthening can be many, but normally deteriorated concrete, steel corrosion, change of use, increased demands on the structure, errors in the design or/and construction phase or accidents are governing factors. Many methods to repair or/and strengthen concrete structures exists such as concrete overlays, shotcrete, use of external prestressed tendons, just to mention a few. Prestressing is in particular interesting with several comparable advantages to other methods. In this paper the use of prestressing for repair and strengthening are briefly discussed and tests on concrete T-beams with external Prestressed tendons of steel or CFRP (Carbon Fibre Reinforced Polymer) are presented. The tests shown that prestressing is a very effective way to increase the existing load carrying capacity of existing concrete members. The presented project is a small part of a larger European funded project, the "Sustainable Bridges", where the aim is to evaluate the load carrying capacity and life of existing railway bridges with the purpose to increase existing load carrying capacity with 25% and the train speed to 350 km/h. The tests were carried out at Luleå University ofTechnology (LTU). The test specimens were concrete T-beams with a length of 6 meters, see figure 1. The beams were loaded under four-point bending, the load was applied with deformation control at 0.2 mm/s until failure or to a point where the beam no longer could carry any more load. A total of eight beams were tested during the series. The strengthening techniques used were externally prestressed steel rods, externally prestressed CFRP rods and Near Surface Mounted Reinforcement (NSMR) CFRP rods with and without prestress. For the steel tendons a traditional steel wedge anchor was used, but that was not possible for the CFRP tendons, as normal steel wedge anchors would crush the FRP tendons. For the tests an anchor was developed using a nylon wedge. To get better effect of the anchor, the tendons had quarts sand glued on them in the zone for anchoring. As those anchors would not be able to take as high forces as a steel anchor on a steel tendon six tendons were used instead of two. However, as the prestressing was applied it became clear that it would not be possible to achieve the same prestressing force as with the steel tendons. With the exception of the beam with external CFRP tendons all the tested beams behaved as expected. The strengthening effects for the prestressed beams were over 100% bom for concrete cracking and steel yielding. When looking at the post-steel yielding behaviour it is interesting to compare the beams strengthened with unbonded tendons and those with bonded tendons (Steel 3 and NSMR PS). The beams with bonded tendons and rods showed a better behaviour after steel yielding than those with (Graph Presented) unbonded tendons. The problems with the CFRP anchor during prestress continued during loading and the loads were much lower for that beam then predicted. In figure 2 the loads and displacements are shown. The tests show a large increase in crack and steel yielding loads. The increase in load for steel yielding can be very important for a constructions life, the fatigue behaviour will improve and as a consequence the crack widths will be smaller which can result in increased durability. Together with higher crack loads the cracks also go smaller, this should also indicate a more advantageous behaviour in the service limit state (SLS).

The advantages of Fibre Reinforced Polymer (FRP) strengthening have been shown time and again during the last decade. All over the world several thousand structures have been retrofitted using FRP. Buildings and civil structures usually have a very long life and it is not uncommon that the demands on the structure change with time. The structures may have to carry larger loads at a later date or fulfil new standards. In extreme cases, a structure may need repair due to an accident, or due to errors made during the design or construction phase. To guarantee the function of the strengthening properties, anchorage of the FRP is essential. Without sufficient anchorage lengths, full utilization of the strengthening material cannot be achieved, leading to possible premature failure. In this paper, experimental work and numerical analyses of three different Carbon Fibre Reinforced Polymer (CFRP) strengthening techniques have been carried out. The techniques are externally bonded plates, sheets and the use of Near Surface Mounted Reinforcement (NSMR). The aim is to find a critical anchorage length, where a longer anchorage length does not contribute to the load bearing capacity. Three different anchorage lengths have been investigated; 100, 200 and 500 mm. The finite element program ABAQUS has been used for the numerical study. The results show that a critical anchorage length exists for plates and sheets as well as for NSMR. However, the present study also shows that an exact critical anchorage length may be difficult to estimate, at least with the present test set-up. Further tests and investigations of the constitutive model for the concrete are needed.