PSI - Issue 82

Chiara Bedon et al. / Procedia Structural Integrity 82 (2026) 65–71 Chiara Bedon et al. / Structural Integrity Procedia 00 (2026) 000–000

68 4

3. Experimental results and discussion 3.1. Experimental observations

The analysis of experimental results was maximized to characterize both the mechanical properties of the interlayer in use, and the bending response of partially fractured 2-ply samples. To this aim, Digital Image Correlation techniques were also used to integrate the traditional instrumental acquisitions (Fig. 3).

(a)

(b)

Fig. 3. Digital Image Correlation analysis: examples of (a) tensile tests on interlayers (strain) and (c) 3-point-bending test on partially fractured 2 ply AN laminated glass samples (deflection).

3.2. Equivalent modulus of elasticity for cracked glass The parametric experimental results were specifically elaborated to derive – fitting the average experimental stiffness EI – the equivalent modulus of elasticity for the cracked glass layer ( E g,cr ) for the examined 2-ply LG sections (Table 1). In particular, the E g,cr values were compared towards the nominal modulus of elasticity of intact glass ( E g = 70 GPa (EN 572-8:2004)). A similar approach was used for example in (Bedon et al., 2025), to quantify the residual capacity of LG in Stage 2, where the analysis of 2-ply (8+8 mm) partially fractured AN glass specimens bonded by EVA film (500 mm × 1000 mm their size) resulted in E g,cr = 69.9 GPa ( D E = -0.14%) in SII and E g,cr = 37.9 MPa ( D E = -45.8%) in SIII. For small-scale LG samples, where the local effect of cracks can be further magnified, literature experiments are reported in (Bedon & Fasan; Bedon & Santos) for AN plates (40 mm × 100 mm). The average E g,cr was calculated in 32.1 GPa ( D E = -54.1%), which is rather in close agreement with SIII results recalled from (Bedon et al., 2025). Most importantly, a progressive decrease in E g,cr was observed with increasing the number of imposed impacts (from an initial value of 53.4 GPa ( D E = -23.7%), and down to 10.8 GPa ( D E = -84.7%) after 25 imposed impacts). In the present experimental analysis, E g,cr was derived from the average cyclic test results, for each series of specimens. The Wölfel-Bennison analytical model (Wölfel, 1987; CNR-DT 201/2013; ASTM E1300-09a) was specifically adapted to a 2-ply LG beam with a broken layer (Fig. 4). The original method, as known, quantifies the equivalent flexural stiffness EI eq of a given 2-ply LG beam-like member, and implicitly accounts for the coupled response of the bonded components. This means that the corresponding moment of inertia can be expressed as: = + + ∗ 2 (1) with: