PSI - Issue 12

6 7 8

30.10 28.91 28.88

- - -

- - -

mean (st.dev)

30.4 (1.7)

359 (12)

1.66 (0.09)

P.M. Giuliani et al. / Procedia Structural Integrity 12 (2018) 296–303 P.M. Giuliani, O. Giannini, R. Panciroli / Structural Integrity Procedia 00 (2018) 000–000

299

4

200

100 150 200 250 300 350 400

150

Stress [ MPa ]

Stress [ MPa ]

100

50

0 0 . 2 0 . 4 0 . 6 0 . 8 1 1 . 2 1 . 4 1 . 6 1 . 8 2 0 50

0

0

0 . 2

0 . 4

0 . 6

0 . 8

Strain [%]

Strain [%]

Figure 1. Left: Typical stress-strain curves for the flax/epoxy specimens. Right: Detail of the superposition of a monotonic (dashed) and a cyclic test (solid line). The plot also highlights the chord modulus (dotted line) and the chord reference points (round markers). ig. 1. Left: Typical stress-strain curves for the flax / epoxy specim ns. Right: Detail of the superposition of a monotonic (dashed) and a cyclic test (solid line). The plot also highlights the chord modulus (dotted line) and the chord reference points (round markers). Figure 1 shows some representative examples of the stress-strain curves obtained for the flax / epoxy specimens. These show the characteristic bi-linear be avior well known from the literatur , where the initial elastic modulus E 1 suddenly decre ses at about 0.2% strain. It later follows a second linear trend, with elastic modulus E 2 . We comment that the estimati n of these two characteristic elastic moduli can not follow the ASTM D3039, sinc the normative imposes t evaluate th chord modulus in the rang 1000 ÷ 3000 µε . Such range crosses the transition point and can not be utilized to estimate either E 1 or E 2 . We thus arbitrarily decided t utilize ε = 200 and ε = 1500 as chord reference points. The right plot in Figure 1 shows the superposition of the initial portion of the str ss-strain curv for a single test (dashed line), th chord modulus (dotted line), and the chord reference points (circular markers). Fig re 1 further show the initial portion of a cyclic test (solid line). The experimental r ults reveal that the upper bounds of the cyclic te t match with t e monotonic curve. The repeatability f the cyclic test is in line with the previous observations. It is also found that both ultimate stress and train match between m notonic and cyclic tests. M. Giuliani, O. Giannini, R. Panciroli / Structural Integrity Procedia 00 (2018) 000–000 5 superposition of the initial portion of the stress-strain curve for a single test (dashed line), the chord modulus (dotted line), and the chord reference points (circular markers). Figure 1 further shows the initial portion of a cyclic test (solid line). The experimental results reveal that the upper bounds of the cyclic test match with the monotonic curve. The repeatability of the cyclic test is in line with the previous observations. It is also found that both ultimate stress and strain match between monotonic and cyclic tests. Figure 2 shows the stress-strain curve for a full cyclic test. A single test is reported to maximize readability, but we note that tests are characterized by an extremely high repeatability. Cyclic tests were found to show inelastic strain accumulation at each cycle. To every unload corresponds an increment of the inelastic strain, as reported in the right chart in Figure 2, which shows the inelastic accumulated strain against the maximum strain attained at each cycle. Our results further supports other findings in the literature, which stated that the inelastic strain appears above 0.1% of applied strain. The inelastic strain is never recovered, and it is found that removing the inelastic strain at the beginning of each cycle, hence translating all the loading cycles to the origin, the initial portion of the stress-strain curve at the following loading cycle always presents an initial elastic modulus equal to E 1 . The initial modulus is preserved until the transition point, which appears to preserve its position, and it follows an increase of the second elastic modulus E 2 , until the previous peak stress is attained. It then follows a last portion of the curve tracing the monotonic trend, hence the original E 2 . Results thus show that at the overall modulus increases at each load cycle. Figure 3 shows an example of this behavior. The chart also reports a curve fitting the peak loads attained at each cycle, and the chord modulus E 1 (dash-dotted line). 0 . 3 0 . 4 0 . 5

100 150 200 250 300 350 400

0 . 1 Inelastic Strain [%] 0 . 2

Stress [ MPa ]

0 0 . 2 0 . 4 0 . 6 0 . 8 1 1 . 2 1 . 4 1 . 6 1 . 8 2 0 50

0 0 . 2 0 . 4 0 . 6 0 . 8 1 1 . 2 1 . 4 1 . 6 1 . 8 0

Strain [%]

Strain [%]

Figure 2. Left: Typical cyclic curve for the flax/epoxy specimens. Right: Accumulated inelastic strain versus the maximum strain attained at each load cycle. i . 2. Left: Typical cyclic curve for the flax / epoxy specimens. Right: Accumulated inelastic strain versus the maximum strain attained at each load cycle.

150 200 250 300 350 400

Figure 2 shows the stress-strain curve for a full cyclic test. A single test is reported to maximize readability, but we note that tests are characterized by an extremely high repeatability. Cyclic tests were found to show inelastic strain accumulation at each cycle. To every unload corresponds an increment of the inelastic strain, as reported in the right

ress [ MPa ]