PSI - Issue 2_A

A. Sancho et al. / Procedia Structural Integrity 2 (2016) 966–973

971

6

A. Sancho et al. / Structural Integrity Procedia 00 (2016) 0 0–000

a)

b)

180

0.25

Sample #4 Sample #6 Sample #9 Sample #11

170

0.20

160

0.15

150

0.10

140

Ductile damage [-]

Bonora model Sample #4 Sample #6 Sample #9 Sample #11

Elastic modulus [GPa]

0.05

130

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Equivalent plastic strain [-] 120

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Equivalent plastic strain [-] 0.00

Fig. 3. (a) Elastic modulus vs. equivalent plastic strain and; (b) ductile damage vs. equivalent plastic strain experimental data of four hourglass shaped round bar specimens under interrupted tensile tests and preliminary Bonora (1997) model calibration.

a)

b)

180

0.25

Sample #19 Sample #20 Sample #13 (DIC)

170

0.20

160

0.15

150

0.10

140

Ductile damage [-]

Elastic modulus [GPa]

0.05

130

Sample #19 Sample #20 Sample #13 (DIC)

120

0.00

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

Equivalent plastic strain [-]

Equivalent plastic strain [-]

Fig. 4. (a) Elastic modulus vs. equivalent plastic strain and; (b) ductile damage vs. equivalent plastic strain experimental data of four hourglass shaped flat specimens under interrupted tensile tests.

( p eq > th ). As the strain is further increased, new voids may appear at a lower almost constant rate, and at the same time, the existing voids grow in size also at a rate fairly proportional to the applied strain. This process continues up to necking and failure when the equivalent plastic strain reaches its critical value, cr ( th < p eq < cr ). In Fig.3(b) a preliminary calibration of Bonora (1997) model is also plotted. The flat samples (Fig. 4) also present a significant initial increase of damage (decrease of sti ff ness) that saturates afterwards, although these results are considerably noisier than those of Fig. 3, making the interpretation more di ffi cult. It is interesting though to compare these results with those obtained using DIC technique. It can be observed that although the results are also very noisy, they have the same order of magnitude and follow a similar trend. Regarding the void formation phenomenon, a similar behaviour to that exposed in the case of round samples can be observed here, being the di ff erences attributable to the di ff erent material production method.