Partenaires et organisations internationales
(Anglais)
|
A, B, DK, FIN, F, D, GR, IRL, I, NL, N, P, RO, E, S, CH, GB
|
Résumé des résultats (Abstract)
(Anglais)
|
Calculation models suggest that the load bearing capacity of low grade timber beams could be greatly improved by strengthening them with prestressed artificial fibres of high strength. A study of the literature suggests however that in practice, only a small prestressing force can be applied because of the danger of delamination caused by the force transmission from the fibre at the corners of the beam into the main body of the beam. At the University of Vienna for example the doctorate thesis of W.F. Luggin (2000) was concerned with the application of prestressed carbon fibre laminates on glulam beams. Most of the work was devoted to the difficulties involved in attaching the prestressed fibres onto the wood and indeed most of the test specimens failed prematurely in delamination. The aim of the research was to help eliminate premature bonding failure and thus achieve the expected strengthening of the beam. In the research work, two main approaches to combat delamination were considered. The first approach involved a special machine developed by the project partner EMPA to prestress laminated carbon fibres. The prestressed laminate is bonded in stages with the beam, whereby the force anchored at the machine-head is decreased slightly at each stage. Thus the bonding stresses are distributed over the entire length of the beam. Although the machine had been successfully used to strengthen concrete beams with prestressed fibres, it had never been tested on timber. In particular, the 'hard' epoxide-based adhesive is seldom used in the timber industry. The second aproach to solving delamination was concerned with the modification of adhesives more commonly used in the timber industry. Theoretical work suggests that with a 'ductile' adhesive, which 'yields' when the local shear stresses attain high values, it might be possible to spread out the attachment force over a larger area at the ends of the beam and thus prevent delamination. Similar results were possible with a 'soft' adhesive which, although elastic, would deform much more than the classical 'stiff' adhesives. The aim of this part of the research work was therefore to find out if it might be feasible to develop such a 'soft or 'ductile' adhesive. The feasibility study on a suitable adhesive occupied most of the project time and resources. Acting on a tip by a French University Professor, two adhesive types on epoxide and polyurethane basis were selected. The two basic ingredients were mixed in different proportions and the mixes used to connect two small timber specimens together. Tensile shear tests showed that some adhesive mixes performed better than the others. The better performers were used to attach slack carbon fibres onto glulam beams. The laminate was placed under tension until the bonding failed. The force-deformation-curve was plotted: the adhesive mix 8, which comprised 70% epoxide und 70% polyurethane plus the respective hardeners, yielded the best results and was selected for the final test series. It was used to attach a prestressed carbon laminate onto glulam beams. A positive result of the research was the confirmation that the EMPA prestressing machine and the epoxide-based adhesive system could work quite well for timber beams as well. The carbon fibre-laminate was safely prestressed with a force of over 60kN, which corresponded to that used to strengthen concrete beams. The attempt to develop a 'soft' or 'ductile' adhesive yielded some promising results, particularly at the beginning. The final tests however showed clearly that the selected adhesive mix 8 exhibited too much creep: when the prestressing machine was turned off, the initial prestressing force of 30kN was slowly reduced to zero. Because of limited financial resources, no work could be done to try to improve the creep performance of the selected adhesive, which may have been caused by inadequate curing. It was possible to identify areas for further research to help improve upon the findings of the work.
|
