PSI - Issue 33

A.F.M.V. Silva et al. / Procedia Structural Integrity 33 (2021) 138–148 Silva et al. / Structural Integrity Procedia 00 (2019) 000–000

146

9

increase of P m with L O , with bigger P m increments for lower L O . As it would be expected, the increase of L O translates into an increase of the shear-resistant area of the adhesive, resulting in higher P m . However, with the increase of L O , τ xy and  y stress concentrations also increase in magnitude at the overlap edges, whose justification was presented in the previous section. Usually, depending on the chosen adhesive type, it may be possible that the peak stresses become plasticized regions by the adhesive entering the plastic loading regime. In these cases, major P m improvements are typically observed with L O . However, the specific adhesive used in this work is highly brittle and fails with undergoing plasticity. Thus, the increasing peak stresses highly limit the joints performance averaged to the bonded area. Compared to the joints with L O =10 mm, relative P m improvements of 33.3% and 43.7% were obtained for L O =15 and 20 mm. Ideally, L O =20 mm should render a 100% P m increase, but the real behavior was far from that mark. Conversely to L O , t SE showed only a minor effect on P m (Fig. 8 b). Actually, taking as basis the previous stress analysis, the variations of  xy /  avg and  y /  avg with t SE were actually small and, in terms of P m , the variations were negligible. Compared to the base geometry with t SE =2 mm, reducing this value to 1 mm led to a 2.0% P m improvement, which is not relevant, while the increase of P m resulted in a P m reduction of 6.0% for t SE =3 mm and improvement of 1.9% for t SE =4 mm. The lack of a clear and univocal tendency through the range of tested t SE , given that the failure mode was identical between all four joint configurations, can be attributed to the dynamic loading and corresponding higher result variation, in comparison to static loading analyses. Thus, for the range of tested t SE , it can be concluded that this geometrical parameter has little effect on the joints’ strength, unlike L O .

40

40

30

30

20

20

P m [kN]

P m [kN]

10

10

0

0

0

1

2

3

4

5

0

5

10

15

20

25

L O [mm]

t SE [mm]

a)

b)

Fig. 8. P m vs. L O and t SE for the tubular joints.

3.3. Dissipated energy In addition to P m , U was used to evaluate the performance of the tubular joints. This is a highly relevant parameter, which indicates the joint’s ability to absorb energy on impact and gains special relevancy in crash absorption structures, for example. U was calculated by the areas under the respective P -  curves, using trapezoidal integration in the RStudio ® software. The obtained data is presented in Fig. 9 for the parametric studies on L O (a) and t SE (b).

10

10

8

8

6

6

U [J]

U [J]

4

4

2

2

0

0

0

5

10

15

20

25

0

1

2

3

4

5

L O [mm]

t SE [mm]

a)

b)

Fig. 9. U vs. L O and t SE for the tubular joints.