PSI - Issue 18

V. Dattoma et al. / Procedia Structural Integrity 18 (2019) 719–730 Author name / Structural Integrity Procedia 00 (2019) 000–000

726

8

The value of ultimate bearing strength F bu , offset bearing strength F by and bearing chord stiffness E b were calculated in the direction specified by the subscript. From a qualitative point of view, the presence of data variations may be determined by dimension variability of specimen hole diameter and of laminate composite thickness in correspondence of bolted zone (Fig. 8b). 4.2. Model A - effect of the friction and preload In order to examine the effect of the friction, different friction coefficients μ have been considered (Chakherlou et al., 2013). In Fig. 9a it can observe as the slope of the Bearing Stress/ Bearing Strain curve does not change so significantly. The stiffness for the coefficient of friction values between 0.15 and 0.45 assumes values from 12.250 GPa to 14.034 GPa; consequently, it can be deduced that increasing the friction coefficient between the parts, the stiffness increases but not enough to so much as to be the determining factor for the resistance of the junction. In Fig. 9b the curve Bearing Stress/Bearing Strain curve for different values of preload is shown. For values between 2000 N and 6000 N it can be observed that the stiffness does not change significantly and keeps around 12.250 GPa values.

(a) (b) Fig. 9. (a) Comparison of stiffness for different values of friction coefficient; (b) Comparison of stiffness for different values of preload.

Figure 10a and 10b illustrate the distribution of the stresses σ X on the entire model. The higher values are reached near the holes, exactly in the area where the shank rivet reaches the wall of the hole. In these zone the compressive stress reaches values around 4390 MPa. Figure 10c shows the distribution of Shear Stress (XY plane). The highest values are obtained near the hole where the contact pressure between the rivet and laminate reached maximum values. In this area, the shear stress reach values around 2903,5 MPa. 4.3. Model B – Analysis of the damage In this model the 3-D Hashin-type failure criteria were used to predict the damage of the matrix and the fibers. Once the stress reaches the damage limit, the material stiffness is immediately reduced to a user-specified value. Different Axial Force / Displacement curves of the bolted joint obtained by varying the amount of material degradation are reported in Fig. 11. In particular, only matrix degradation factor was varied because the matrix remains the main cause of the failure of the joint. The results are compared with the experimental curve. In this work, a macroscopic damage was assumed beginning where the curve slope changes abruptly and in some cases up to negative slope. For a degradation factor equal to 0,75 the damage occurs around an Axial Force equal to 8726 N and an axial displacement equal to 1,15 mm. Decreasing the value to 0,63, the damage begins around to 10605 N and the axial displacement equal to 1,68 mm, which are values higher than experimental data where the damage begins around to 1,25 mm and 11060 N. Figure 11b is represented the curve Axial Force/Displacement varying the compressive stress limits of the lamina in y-z direction. It can be observed as decreasing the strength of the lamina of the 30 % the damage begins