PSI - Issue 28

Thomas Bergmayr et al. / Procedia Structural Integrity 28 (2020) 1473–1480

1475 3

C.V. Thomas Bergmayr et al. / Structural Integrity Procedia 00 (2020) 000–000

- ε nm

2 β 1

2 α 2

ε xx

2 α 1

ε 1

ε 2

ε nn

ε yy

2 β 2

ε xy

Fig. 1. Mohr’s circle with the zero-strain directions β 1 and β 2 and the principal strain directions α 1 and α 2 for a given plane strain state ε xx , ε yy and ε xy .

face layer debonding

spoiler

inboard edge tting

center hinge tting

outboard edge tting

center hinge tting

edge ttings

Fig. 2. The figure shows the original A340 aircraft spoiler, the location of the spoiler on an aircraft as well as the idealized aircraft spoiler model, which was build for our tests.

measures no signal if the structure is undamaged, therefore no data storage is necessary. The second advantage is the high sensitivity, if the structure gets damaged, the strain along the trajectory deviates from a zero baseline to a finite value. In addition, the strain state near the damage changes to a larger extent. With increasing distance between the sensor and the damage the strain signal converges to zero. Therefore, it is possible not only to detect damages but also to monitor their changes and to localize them (Riedl (2018), Kralovec (2020)).

3. Experimental set-up

The tested composite structure should imitate an aircraft spoiler of an A340 aircraft on the scale of 2:1. The idealized spoiler model with simplified shapes and the dimension of dimensions 1000 × 380 × 16 mm 3 depicted, in Fig. 2 is composed of glass fiber reinforced plastic face layers and a Nomex honeycomb core. The GFRP face layer laminate is built up of four fabric plies [0 , 45 , − 45 , 0], with a total thickness of 0.5 mm. The material sti ff ness properties of the ply, the aluminum fittings and the Nomex honeycomb core are listed in Tab. 1. In this investigation, a debonding of the top face layer at the trailing edge is considered. The face layer debonding is semicircular shaped. Di ff erent debonding states and its corresponding sizes, as listed in Table 2 are considered. The spoiler model was attached to a test rig via the center hinge fitting (fixed) and the edge fittings (with rotational degree of freedom around x -axis). The load was applied with a whi ffl etree such that the spoiler model deforms like the real A-340 spoiler when used as air brake during aircraft landing. The ideal force introduction points of the whi ffl etree are found by means of a linear FE shell model and parametric optimization. The objective was to minimize