PSI - Issue 12

A. Castriota et al. / Procedia Structural Integrity 12 (2018) 71–81

76

Castriota et al. / Structural Integrity Procedia 00 (2018) 000 – 000

6

to the strain gauges positioned on the spar cap in a symmetrical position with respect to the symmetry plane. The absolute difference and percentage is contained, considering also the presence of the repair. It can be said that the load is well centered with respect to the symmetry plane and any warping of the spar during the subsequent tests is avoided. This makes it possible to state how the numerical model is very close to reality and how the repair does not affect the overall behavior of the spar for loads far from the expected breaking load.

a

b

Fig. 4. (a) Experimental vs. numerical comparison of the values ( ε x ) of strain gauges; (b) longitudinal position of the strain gauges

a

b

Fig. 5. (a) Comparison of normalized s train ε x side A vs. side B; (b) longitudinal position of the strain gauges

5.2. Variable fatigue test (Step 1)

At the end of Step 0, the variable amplitude loading history has been applied. Load history is obtained combining in a random way blocks of flights. Each block is constituted by a repetition of a flight. Six different flights have been considered, which are representative of the real load condition of the spar. Each flight that constitutes the load history is codified as following: Flight – X – Y – n/n tot where: X (from 1 to 6) indicates the flight typology; Y (from 1 to 490) is a progressive number which identifies the block of flights in the load history; n identifies the effective repetition of the flight among the planned n tot of the block. As an example, the reading versus time of the E41A strain gauge, which is close to the repair area of the hole, is showed for Flight 1 and Flight 4 (Fig. 6). In particular for Flight 1 (Fig. 6a), which is the most simple and widespread of the load history, the strain gauge readings are compared for Flight-1-1-1/225 and for Flight-1-489-1125/1125,