Issue34

M. Ševčík et al, Frattura ed Integrità Strutturale, 34 (2015) 216-225; DOI: 10.3221/IGF-ESIS.34.23 221 F RACTURE CRITERION he fracture criterion for the crack initiation and crack propagation can be found in the literature [21]. The critical values of the strain energy release rate for the crack initiation and propagation in various crack propagation paths are shown in Tab. 3. More information about the experimental measurement of the fracture criterion can be found in [14] . specime n code lever dimensio n [mm] crack initiation crack propagation in Path I crack propagation in Path II crack propagation in Path III c c g G I G II G tot G I G II G tot G I G II G tot G I G II G tot MMB-01 227 54 205 85 290 1061 287 1348 102 4 144 1168 MMB-02 166 69 235 410 151 561 963 260 1223 MMB-03 133 55 188 837 226 1063 MMB-04 159 66 225 890 241 1131 MMB-05 163 67 230 497 183 680 924 250 1174 MMB-06 150 38 162 128 290 851 387 1238 383 89 472 MMB-07 210 165 375 550 374 924 1247 567 1814 MMB-08 182 143 325 462 314 776 945 430 1375 MMB-09 182 143 325 1190 541 1731 MMB-10 181 143 324 1012 460 1472 700 164 864 MMB-11 189 149 338 597 406 1003 1302 592 1894 Table 3 : Critical strain energy release rate at various paths under different mixed-mode loading in [J/m 2 ] from [14]. Single underlined numbers represent highest recorded values of the group, double underlined values represent lowest recorded values. R ESULTS he proposed analytical model of the asymmetric MMB test is analyzed in this section. Fracture behavior obtained using the proposed model is compared with experimental data recently obtained by authors and well documented in the literature [14,21]. Effect of fiber bridging length At first the sensitivity analysis of the fiber bridging length was performed. The total strain energy release rate for crack initiation and crack propagation was kept constant but the fiber bridging length L F was changed in order to study its effect on the load vs. crack length (P-a) diagram. The shape of the fracture criterion used for this analysis is shown in Fig. 6. Based on the analytical procedure described above it is possible to predict the load vs. crack length diagram assuming that the crack will propagate if the condition G tot (a) = G crit (a) , where G tot (a) is the calculated value of the total strain energy release rate and G crit (a) is the critical value of the total strain energy release rate that can be measured experimentally, Tab. 3. Example of such prediction is shown in Fig. 7 where experimentally measured and analytically predicted P-a diagrams are plotted when considering fracture in Path II. By inspecting the Fig. 7 it is possible to conclude that the analytical model is capable to predict the P-a diagram of the MMB test with sufficient accuracy. The fiber bridging length seems to have strong effect on the maximum loading force, however, it has to be mentioned that in reality the critical SERR for the crack propagation G prop depends on the fiber bridging length. Generally, the G prop = G ini when L F = 0 and G prop increases when L F increases. Therefore, if the values of G ini and G prop are known the fiber bridging length can be obtained indirectly using this analytical model in comparison with T T