Over the last decade, there has been a rapid and to some extent unpredicted growth in the volume of traffic using the worlds road networks. This, and the EC directive regarding the use of 40 tonne articulated lorries, has resulted in a number of bridges requiring upgrading. It is estimated that some 40,000 road bridges in the UK will need to be strengthened, replaced or subjected to restrictions by the year 1999.
Clearly the costs associated with replacing all of these bridges would be astronomical. Fortunately a technique has been developed which allows certain types of bridge to be upgraded, relatively simply. This technique, known as plate bonding or external reinforcement, makes use of steel plates bonded to the tension side of the existing reinforced concrete beams. The technique has been widely and successfully exploited in Europe, South Africa and Japan, since the late sixties. The first use in this country being on the M5 interchange at Quinton in 1975, and is still performing well, shown above. A recent project undertaken in the UK is the A23 Bolney flyover in East Sussex, where some 38 tonnes of steel was bonded to the underside of the bridge deck. More recently the Hythe Bridge in Oxford was repaired using similar bonding methods. The major advantages of this technique over other methods, is the fact that the reinforcement can be carried out whilst the structure is still in use. In addition the technique has relatively little effect on the headroom under the bridge.
A number of core samples were obtained from the bridges at Quinton to provide samples for this forensic investigation of the adhesive joint. The Quinton bridges appear to be performing satisfactorily, apart from areas subject to persistent leakage and consequent washing of water over plated areas. Clearly it is important to divert running water and to fill cracks in the deck, prior to plating, through which water might migrate to bondlines.
Concrete/ Steel
Filled cold curing epoxy
Grit blast
Room temperature cure
The bonding was carried out at the field location.
In simple terms the technique involves grit blasted steel plates, generally around 300 mm wide and 6 mm thick, bonded to the prepared surface of the concrete bridge, using a heavily filled cold curing epoxy adhesive. Typical bondline thicknesses are between 1 and 6 mm thick to allow for irregularities in the concrete surface.
Once hardened the dynamic loads from the bridge structure are transmitted into the steel plates through the adhesive. In order to reduce the amount of support required during the assembly of the joint the steel plates are generally bolted to the underlying concrete, whilst the adhesive cures. Not only does this provide an important second line of defence, it acts to inhibit peel stresses, and transmits any tensile loads which might otherwise develop in the adhesive where the structure goes into reverse bending, at supports for example.
The bulk mechanical properties of both Ciba and Shell adhesives were satisfactory, as judged by shear tests. There were clearly some mixing and application problems associated with both adhesives, but more particularly with the Shell adhesive as the latter contains a regular dispersion of air voids. Poor mixing of the Shell adhesive was inferred from changes in colour of the bondline from place to place.
Bondline thickness varied enormously. The steel/steel bondlines were found to be between 0 and 1.5mm thick; the steel/concrete bondlines were between 0.5 and 5.0 mm thick. Adhesive was absent in some areas, including an important zone close to the end of a plate. Such areas had not been detected in any of the tapping surveys. These involve monitoring the sound emitted by the plate when tapped, a change in tone indicating a change in bond quality.
The adhesion between the concrete and the adhesives was satisfactory; all failures at adhesive/concrete interfaces were in the concrete itself. Different steels were used over different parts of the structures. These probably confer different bond and adhesion characteristics, and different bond stabilities. The quality of gritblasting of the steel from place to place was also variable, from uniformly rough to quite smooth. Particles and/or dust were also present on the surface of the steel in places. The adhesion of the Shell adhesive to steel was better than the adhesion of the Ciba adhesive to steel. However, the only significant steel surface corrosion found, in places, was beneath the Shell adhesive. This suggests an adverse reaction between the constituents of the Shell adhesive and steel, perhaps where the adhesive had not been mixed adequately. Joint shear strength did not seem to be affected by the presence of localised steel surface corrosion.
The technique of plate bonding offers a cost effective method by which the load capacity of a bridge can be increased, whilst the structure is still in use, and with relatively little effect on the headroom under the bridge. Because of the nature of the problem, it is not possible to ensure perfect conditions for bonding. Nonetheless, the investigation showed that, over a twenty year service life, the adhesive performance remained satisfactory, with the concrete of the bridge representing the weakest link.
DTI MTS programme adhesives – Dissemination Guides BRIDGE.DOC
ESR Technology Limited/ NPL
Other relevant links:
