PSI - Issue 54

I.R.S. Araújo et al. / Procedia Structural Integrity 54 (2024) 406–413 Araújo et al. / Structural Integrity Procedia 00 (2019) 000 – 000

412

7

However, all other P m predictions by the XFEM are close to the experimental results. For the 2015 and 7752, in general the XFEM predictions closely reproduced the experimental P m values. Table 3 shows the P m values obtained experimentally and numerically by the XFEM as a function of  for the three adhesives. Data analysis shows that the predictions made by XFEM produced values close to the experimental ones. In the scarf joint bonded with the AV138 and with  =3.43º there is a notable difference between the numerical and experimental P m values. The relative difference is 20.25%, with the numerical value being lower than the experimental P m . However, it should be emphasized that, despite the discrepancy between values, this joint configuration is the one with the highest P m . The relative differences for the other  values are quite acceptable, with the largest relative difference (6.56%) being observed for  =20º and the smallest relative difference (0.18%) for  =30º. For the 2015 adhesive, only in the joint configuration with  =30º is the numerical P m lower than the experimental one. In the other configurations, the numerical values are always higher than the experimental values. Since this is a small difference, the results are considered adequate. The largest relative difference is 9.59% for  =20º and the smallest is 2.22% for  =10º. Regarding the 7752, the XFEM was able to approximately predict the P m values for this adhesive, although the numerical P m are always smaller than the experimental results. The biggest relative difference is 7.41% for  =15º, but this disparity is considered acceptable. Overall, the XFEM was able to predict strength to an acceptable degree. The AV138 is the adhesive with the highest strength for this type of joint, followed by the 2015 and, lastly, the 7752.

40

25

20

30

15

20

P m [kN]

P m [kN]

10

10

5

0

0

0

10

20

30

40

50

0

10

20

30

40

50

α [º]

α [º]

a)

b)

Experimental

Numerical

Experimental

Numerical

20

16

12

P m [kN]

8

4

0

0

10

20

30

40

50

α [º] Experimental

c)

Numerical

Fig. 6. P m evolution as a function o  for the scarf joints bonded with the adhesives AV138 (a), 2015 (b) and 7752 (c).

Table 3. P m evolution as a function of  with the different adhesives.

 [ o ]

3

10

15

20

30

45

AV138

Experimental

29700 23685 -20.25 21903 22823 13500 12868 -4.68 4.20

10648 10467 -1.70 7509 7676 2.22 4677 4396 -6.01

7926 7701 -2.84 4858 5144 5.89 3132 2900 -7.41

5715 6090 6.56 3567 3909 9.59 2378 2240 -5.80

4480 4488 0.18 2832 2718 -4.03 1543 1526 -1.10

3325 3487 4.87 1868 1997 6.91 1142 1111 -2.71

XFEM

Relative deviation [%]

2015

Experimental

XFEM

Relative deviation [%]

7752

Experimental

XFEM

Relative deviation [%]