PSI - Issue 24

Giuseppe Mirone et al. / Procedia Structural Integrity 24 (2019) 259–266 Mirone & Barbagallo / Structural Integrity Procedia 00 (2019) 000 – 000

262

4

̇ = ( )

(7) The evolving diameters have been obtained in static tests by means of standard video camera acquisitions while thanks to an ultra-high frame rate camera in SHTB tests.

Table 1. Summary of the A2-70 Experimental Campaign

Reference True Strain Rate [s -1 ]

Test Environment Temperature [°C]

Test Series

Test Name

S-T ROOM -01 0.003 S-T ROOM -02 0.003 S-T ROOM -03 0.003

19 22 22 80 80

Static T ROOM

S-T80-01 S-T80-02 S-T140-01 S-T140-02 S-T200-01 S-T200-02 S-T300-01 S-T300-02 S-T300-03

0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003

Static T80

140 140 200 200 300 300 300

Static T140

Static T200

Static T300

D-01 (15 kN) 700 D-02 (15 kN) 890 D-03 (26 kN) 1800 D-04 (26 kN) 1850

20 20 20 20

Dynamic T ROOM

3.2. Thermal softening and its influence on the necking inception Firstly, the static tests at different temperatures are analysed in order to obtain the reference static equivalent stress strain curve at room temperature (T ROOM about 20 °C) and to evaluate the differences between such curve and the others obtained at different temperatures regarding the necking phenomenon. The true stress-true strain data from static tests are shown in Fig. 1. In Table 2 are reported the mean yield stress and necking strain of each group of tests. From such data, it is already possible to note an important finding of this work, i.e. that the thermal softening has a great effect on the necking inception strain that changes from 0.44 at 20 °C to 0.18 at 300 °C. The static curves are then transformed into Mises curves by the MLR correction (Mirone (2004)), so that the ratio ( ) = ( )/ ( ) can be calculated for the tests at 80, 140, 200 and 300 °C at each strain level. The obtained data are presented in Fig. 2 in a 3D graph in which the three axis are the deformation, the temperature and the thermal softening. Such figure highlights a rather important finding not usually acknowledged in the literature: the thermal softening is clearly not strain-independent, as it is usually assumed. Therefore it is wrong to refer to a single value of ( ) while, on the contrary, a two variables function ( , ) should be defined and implemented in material models. Such finding also indicates that eq. (3) is not entirely correct, because such expression was obtained by considering the thermal softening and the strain uncoupled. In fact, if the softening really depended on the temperature alone, then ( ) under controlled temperature would be constant all over a test and the material curve at high temperature would be proportional to the reference static curve at room temperature, through the constant value ( ) . In such scenery, the necking onset pinpointed by the mathematical condition = / would not be affected at all by the temperature, which is against the experimental evidence shown in Table 2.