PSI - Issue 3

D. Triantis et al. / Procedia Structural Integrity 3 (2017) 346–353 6 D. Triantis, E.D. Pasiou, I. Stavrakas, I. Dakanali and S.K. Kourkoulis / Structural Integrity Procedia 00 (2017) 000–000

351

to random generation of micro-cracks of low acoustic activity, especially when low loading rates are applied (28 N/s and 55 N/s in Fig.5), leading to increased Ib-values during the first loading stages. Afterwards, as the load increases further, Ib-value decreases systematically since more micro-cracks are formed and the existing ones grow. Then Ib value attains an almost constant value close to 1, a few seconds before the catastrophic macroscopic fracture of the specimens. The as above described behaviour of Ib-value is strongly related to the formation of macro-cracks which lead the specimen to failure. It is interesting that approximately at the same time PSC reaches its maximum value since the conductive path between the electrodes is interrupted due to the presence of macro-cracks. 3.4. PSC and AE energies and their correlation Following an alternative approach, the amount of absolute energy (measured in aJ) emitted and recorded by the acoustic sensors was taken into account. Absolute energy is a quantity characterizing each acoustic hit and it is calculated by the AEwin software as the integral of the signal voltage at a power of two over the reference resistance (10 kΩ). In the present study, the rate at which AE energy ( AE E  ) is released was considered since it is related to the size distribution of micro-cracks in all such materials (Rao et al. 2011). The specific quantity was calculated as:  t =1 s, t i = 0, 1, …, t f -1 (t f is the test’s duration) and E i the absolute energy of the hits recorded within every second. As a second step, the energy released as it was indirectly measured by the PSC technique was also calculated as:   i i t t 2 PSC t E PSC t dt     (4) where the symbols are the same as in Eq.3. The units of PSC E  are (pA) 2 ∙s since PSC is measured in pA. In Fig.6 the two energy rates are plotted against each other in logarithmic scale for typical specimens of each group. It is clear that a very good correlation exists between AE E  and PSC E  . More specific, all plots of Fig.6 obey a power law, i.e. m (5) The exponent m in Eq.(5) varies between relatively narrow limits in the 0.86 and 1.08 range with a correlation coef ficient R 2 equal to 0.99 for all experiments, without any exception, indicating an excellent correlation between the two energy quantities. The m-value for each group of specimens is presented in the last column of Table 1. i i t t AE i t E E     (3) AE PSC E E         

(a)

(b)

Fig. 6. The correlation of the released energy determined according to the data of the AE and PSC techniques.