PSI - Issue 17

3

R. Branco et al. / Procedia Structural Integrity 17 (2019) 177–182 Author name / Structural Integrity Procedia 00 (2019) 000–000

179

Table 1. Monotonic mechanical properties of the 7050-T6 aluminum alloy.

E (GPa)

σ YS (MPa)

σ UTS (MPa)

∆ (%)

ν

546

621

14

71.7

0.33

Pre-strain (PS) was applied by monotonic tensile loading assuming three different levels, namely: 0% (i.e. no pre strain), 4% and 8%. For each case, strain amplitudes ( ∆ε /2) varied in the range 0.6% to 1.5%. Stress-strain data were recorded using an electrical extensometer with a 12.5mm-long gauge clamped to the specimen via two separated knife edges. After fatigue testing, fracture surfaces were examined by scanning electron microscopy. 3. Results and discussion The typical cyclic stress-strain responses recorded in the strain-controlled low-cycle fatigue tests for the different monotonic tensile pre-strain histories are presented in Figure 2. This figure shows three tests performed at the same strain amplitude ( ∆ε /2 = ±1.5%) with no pre-strain (PS=0%), 4% pre-strain (PS=4%), and 8% pre-strain (PS=8%). Pre-strain was applied by monotonic tensile loading until the prescribed pre-strain was reached. After that, loading was reduced to zero. Effective pre-strain values induced in the specimens were, respectively, 3.1% and 7.0% for the cases of PS=4% and PS=8% (see Figure 2). In the second stage of the tests, i.e. strain-controlled tests, fully-reversed conditions were assumed (R ε = -1). The specimen diameter considered in the calculations corresponded to the values measured at the beginning of the second stage (i.e. after pre-straining). The differences in fatigue behaviour are assumed to be due to the pre-strain stage. As can be seen in Figure 2, irrespective of the pre-strain history, it can be clearly observed a cyclic strain-softening behaviour. Without pre-strain (PS=0%), tensile and compressive stresses vary in a soft manner during the test. On the contrary, as the pre-strain increases, the variations of both the tensile and the compressive stresses tend to be more relevant in each cycle and, therefore, along the test. This is evident by comparing the positions of the mid-life cycles of the three tests. The more intense changes at higher pre-strain values can be explained by the means stress relaxation behaviour. At lower pre-strain levels, relaxation phenomenon is more limited which results in higher mean stress levels (and, therefore, lower changes in peak stresses) during the entire

800

σ (MPa)

600

400

200

ε (%)

0

-2

-1

0

1

2

3

4

5

6

7

8

9

-200

-400

-600

PS=0% PS=0%

PS=4% PS=4%

PS=8% PS=8%

Full test: Mid-life:

-800

Fig. 2. Cyclic stress-strain response for various pre-strain levels.