Issue 48

V. M. G. Gomes et alii, Frattura ed Integrità Strutturale, 48 (2019) 304-317; DOI: 10.3221/IGF-ESIS.48.30 313 Figure 6 : A) Experimental vs. numerical load-displacement curves; B) Failure loads obtained in the monotonic static tests for two preload levels (25%  70% Fu and 70% Fu , or 140 and 560 MPa, respectively). A set of experimental and numerical load-displacement curves were chosen to explain in detail the friction and mechanical behaviour of bolted joints. Fig. 7 illustrates the multiple bolted joints behaviours made of S350GD steel with zinc coating plus painting until the moment that plate starts to yielding. These curves were chosen because they are representative of the behaviour of all studied specimens (single and multiple bolts for all preload ranges). Also there was a good agreement between the numerical and experimental curves. Initially, specimens were assembled in such a way to allow the maximum possible sliding (1). From (1), the load increases and the static friction coefficient prevails until (a). After (a), the friction coefficient decreases to the dynamic friction coefficient and consequently, the recorded load decreases, becoming almost constant until the contact between the middle plate and the bolt shank is established (2). Then the load increases again due to slippage resistance of fastening set until the applied load exceed the total friction load (b), inducing the movement of middle plate and fastening bolt until disappearing all gaps (3) and the load increases due to deformation of middle plate (4) until its failure. It is worth noting that the two load plateaus visible in Fig. 7 are similar to the ones observed by Rodrigues et al. [1]. The only difference is that in the current research two friction coefficients ( stat µ and dyn µ ) are used, which induces extra oscillations in the reaction loads [1]. Comparing the experimental and numerical curves, it is noted that there are differences between experimental and numerical friction behaviours. Firstly, there is a difference between instant the gaps disappeared. Theoretically, the gap is equal to 4 mm for each side of specimen, resulting on 8 mm, but only after about 6 mm of experimental displacement there is again an increasing of load. This fact may be explained by the contact S355MC (Not coated) 0 50 100 150 200 0 10 20 30 40 Load (kN) Displacement(mm) 0 50 100 150 200 0 10 20 30 40 50 Load (kN) Displacement(mm) 0 50 100 150 200 0 10 20 30 40 Load (kN) Displacement(mm) 0 50 100 150 200 0 10 20 30 40 Load (kN) Displacement(mm) SSC2 SSC3 SMC2 SMC3 SSC2_FEM SSC3_FEM SMC2_FEM SMC3_FEM 0 50 100 150 200 0 10 20 30 40 Load (kN) Displacement(mm) SSZ2 SSZ3 SMZ2 SMZ3 SSZ2_FEM SSZ3_FEM SMZ2_FEM SMZ3_FEM 0 50 100 150 200 0 10 20 30 40 Load (kN) Displacement(mm) SSP2 SSP3 SMP2 SMP3 SSP2_FEM SSP3_FEM SMP2_FEM SMP3_FEM Preload = 140 MPa Preload = 560 MPa S350GD (Zinc coating) S350GD (Zinc plus paint coating) A B 36.69 59.18 100.57 154.02 39.03 63.05 105.10 161.66 0 40 80 120 160 200 SSC2 SSC3 SMC2 SMC3 Load (kN) P140 (SnugTight Bolts ) P560 (Preloaded Bolts ) 39.81 55.16 101.18 142.84 42.84 56.28 110.37 141.71 0 40 80 120 160 200 SSZ2 SSZ3 SMZ2 SMZ3 Load (kN) P140 (SnugTight Bolts ) P560 (PreloadedBolts ) 37.96 52.09 100.18 139.17 39.85 55.45 99.91 140.35 0 40 80 120 160 200 SSP2 SSP3 SMP2 SMP3 Load (kN) P140 (SnugTightBolts ) P560 (PreloadedBolts )