PSI - Issue 12

78 A. Castriota et al. / Procedia Structural Integrity 12 (2018) 71–81 Castriota et al. / Structural Integrity Procedia 00 (2018) 000 – 000 loaded applying a load at the tip up to . Since no failure was observed, the component was unloaded. The normalized load-deformation curves both in loading and unloading phase are shown for the strain gauges placed in the compressed part of the spar (Fig. 8). 8

a

b

Fig. 8. Experimental results: Normalized load – strain curves of strain gauge on Upper Skin (a) and Lower Skin (b).

Analyzing the diagrams, it can be observed how the loading and unloading curves do not coincide perfectly and show a slight hysteresis, more or less marked according to the considered area, which involves a very small residual deformation at the end of the discharge. From what has been reported, it emerges that the variable amplitude fatigue cycles do not introduced significant consequences on the behaviour of the test article. It is also remarkable that at the end of the static test (Step 2) with load equal to , the test article showed no type of structural failure; this experimental observation derives both from the analysis of the deformations and from visual inspections on the spar during and after the test. A further result is the evaluation of the normalized experimental bending stiffness that has been compared to the numerical one (Fig. 9): from the numerical model, a normalized flexural stiffness of 0.9984 was obtained while experimentally there was a normalized stiffness of 1.0754, with a percentage gap of 7.71% compared to the numerical model.

Fig. 9. Comparison numerical/experimental stiffness

5.4. Static test until failure (Step 3)

A further static test (Step 3) was carried out until the breakdown was reached. From the observation of the normalized load – deformation curves of the strain gauges and rosettes near the hole (Fig. 11), it was possible to detect