PSI - Issue 25

Tiago Bento et al. / Procedia Structural Integrity 25 (2020) 234–245 Tiago Bento/ Structural Integrity Procedia 00 (2019) 000 – 000

9

242

0 1000 2000 3000 4000 5000 6000 7000 7 6 5 4 3 2 1

HybBragg450 - 2 Abaqus - HybBragg450 - 2 FEM model

Experimental specimen

Load (kN)

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

Displacement (mm)

(a)

(b) (c) Fig. 10. Lap joint: (a) Stress vs Strain graphic; (b) Experimental test; (c) Stress distribution.

The reasons behind the difference between the Abaqus and experimental curve are due to the fact that we are working with hybrid joints and a variation is introduced by the perfectly uniform adhesive layer in the numerical case, which was not verified in the macroscopic analysis of the real specimen. Also, the continuous flaws in the material and weld line cannot be easily modeled and are not taken into account. This smart joints’ model reinforced the idea that is possible to produce reliable simulations of FSW designs.

3.8. Bragg fibers measurements in T-joints

Once the manufacturing process was consolidated T-joints were produced for testing. Unlike the previous case, in this one, fiber’s location was well defined in order to model and compare the results of the fiber te sted and modulated. In Fig. 11 the evolution of the Load applied and Strain measured in the fiber are presented both in function of the superior actuator displacement in the bending test.

1000 1200 1400

0 0,5 1 1,5 2 2,5 3 3,5 4

0 200 400 600 800 Microstrain ( με)

Strain-Bragg Strain-FEM Load

Load [kN]

0

0,2

0,4

0,6

0,8

1

1,2

1,4

1,6

1,8

2

Displacement applied [mm]

Fig. 11. T-joint bending test Bragg fiber's reading.