PSI - Issue 33

M.F.M.O. Rosas et al. / Procedia Structural Integrity 33 (2021) 115–125 Rosas et al. / Structural Integrity Procedia 00 (2019) 000 – 000

122

8

Fig. 8 shows the P -  curves of the tubular joints of the adhesive Araldite ® 2015, studied by CZM, with the  modification at the ends of the tubular adhesive joint. From the P -  curves, it is visible that the aluminium tubes undergo the plastic loading region for all  , due to the good strength characteristics of this adhesive and the respective obtained bond. As a result, there are no changes in P m with the variation of   as it shows an insignificant increase of 0.02% for  =30º, over  =90º. On the other hand, t here is a change in the joints’ elastic stiffness, with smaller  giving lower stiffness, due to the change in the load patterns. 3.2.3. Adhesive fillet The adhesive fillet has as main objective to reduce the stresses of the tubular adhesive joint. This geometric change creates an increase in L O , such that the adhesive layer extends beyond x / L O =0 and 1. The values of adhesive fillet angle (  ), studied numerically, were 7.5  , 15  , 30  , 45  , 60  and 90  . Fig. 9 (a) shows the σ y / τ avg stress distribution curves as a function of  , for the Araldite ® 2015. The adhesive fillet leads to the existence of additional peak stresses, apart from the ones at x / L O =0 and x / L O =1, now taking place at the end of the different adhesive fillet angles, located at x / L O <0 and x / L O >1. However, there is a decrease in σ y / τ avg stresses outside the overlap region, because the stiffness of the adhesive is lower than that of the adherend, which results in lower stress. Overall, the σ y / τ avg peak stresses decrease for smaller  , with a maximum reduction of the peak value of 16.7% for  =7.5  , over  =90  . Fig. 9 (b) depicts the τ xy / τ avg stress distributions as a function of the  . The aforementioned phenomenon of additional peak stresses arising due to the adhesive fillets reflects also on τ xy stresses. Nonetheless, these are much lower in magnitude compared to those at the overlap ends ( x / L O =0 and x / L O =1). There is a stabilization of peak τ xy / τ avg for angles between  =7.5  and  =60  . Comparing all  gives the highest τ xy / τ avg for the joint without geometrical modifications (  =90  ), and a maximum reduction of 11.7% for  =7.5  , when compared to  =90  .

6

5

4

4

3

2

 xy /  avg

 y /  avg

2

0

1

-2

0

-0.5

0

0.5

1

1.5

-0.5

0

0.5

1

1.5

x / L O

x / L O

7.5º

15°

30°

45°

60°

90°

7.5º

15°

30°

45°

60°

90°

a)

b)

Fig. 9.  y (a) and  xy (b) stress distributions in the adhesive as a function of  .

50

40

30

P [kN]

20

10

0

0

0.1

0.2

0.3

0.4

0.5

0.6

 [mm]

7.5º

15°

30°

45°

60°

90°

Fig. 10. P -  curves as a function of  