Issue 48

V. M. G. Gomes et alii, Frattura ed Integrità Strutturale, 48 (2019) 304-317; DOI: 10.3221/IGF-ESIS.48.30 316 Specimen Series e 1/ D hole e 2 sup / D hole e 2 inf / D hole p 1 / D hole p 2 / D hole SSC2 1.94 2.22 - - - SSC3 1.94 2.22 - - - SSZ2 1.94 2.22 - - - SSZ3 1.92 2.23 - - - SSP2 1.94 2.23 - - - SSP3 1.91 2.22 - - - Average 1.93 2.22 - - - SMC2 1.95 2.24 2.24 2.79 2.80 SMC3 1.95 2.24 2.24 2.80 2.80 SMZ2 1.96 2.24 2.23 2.78 2.80 SMZ3 1.96 2.25 2.24 2.79 2.80 SMP2 1.96 2.25 2.24 2.78 2.79 SMP3 1.95 2.25 2.25 2.78 2.79 Average 1.96 2.24 2.24 2.79 2.80 Table 7 : Geometric ratios between hole distances and hole diameter for single and multiple bolted connections. C ONCLUSIONS he purpose of this research was to evaluate the bolted connections behaviour including ductility, strength and slip factors. Firstly, it was verified a high variation on slip factors (slip tests) for each test performed with a maximum coefficient of variation value of 21.65%. The slip tests revealed the lowest slip factor for the painted specimens, followed by the uncoated specimens, and the higher slip factor was verified for the zinc coated specimens. The comparison of the friction coefficients obtained with the slip and static tests showed significant deviations that may be partially justified by deviations in the coating processes including the coating thicknesses. The slip factors evaluated according to the EN 1090-2 standard should be preferable instead of using the static monotonic tests, since the first tests produced results that are more conservative. The static/monotonic tests showed that the increase of thickness increases significantly the failure load and that high preloads increases the sliding loads. However higher preloads have not much influence on failure modes. In addition, it was possible to verify that the 4+4 bolted joints are stiffer than 1+1 bolted joints but reversely the limit displacements of single bolted connections are greater than observed for multiple bolted connections. The experimental results showed a higher coefficient of variation on ultimate loads for specimens with preloaded bolts. Performed numerical simulations allowed understanding that for low preload levels, the friction coefficient value is not very significant, but it may have important influence for high preloads and low thicknesses. For these reasons, the friction behaviour law should be properly modelled, taken into account the objectives and accuracy desired. The numerical results were very satisfactory with a maximum error of 8.7% in the ultimate loads estimation. Finally, from numerical simulations was possible to predict the failure location and ultimate loads without the use of a damage model, but only from detection of the peak load which occurs due the necking. Single bolted connection specimens without coating (355MC steel grade) showed only shear fracture, however single bolted connection specimens with zinc coating or zinc plus paint coating (350GD steel grade) showed two possibilities, shear and/or splitting fractures. As regards to multiple bolts connection, it was only verified fracture through net cross-section. A CKNOWLEDGMENTS he authors express their gratitude to the FASTCOLD RFCS European Project (Grant Agreement No. 745982). Shelter S.A (Prokopis Tsintzos and Markos Mezari) is particularly acknowledged for the material supply. The SciTech (Science and Technology for Competitive and Sustainable Industries) R&D project NORTE-01-0145- T T