Issue 35

O. Demir et alii, Frattura ed Integrità Strutturale, 35 (2016) 340-349; DOI: 10.3221/IGF-ESIS.35.39 344 M ODE - I / II FRACTURE CRITERION DEVELOPMENT n this section, development of a mixed mode-I/II fracture criterion by making use of SIFs from finite element models and experimental results and comparisons with other existing criteria are presented. In the first sub-section, details of mode-I/II fracture criterion development is presented. In the second sub-section, comparisons of developed and some of the existing mode-I/II fracture criteria in terms of fracture loads are presented. This is followed by comparisons of the developed criterion with other criteria in the literature in terms of crack deflection angles. Procedure for Mode-I/II Fracture Criterion Development In an effort to develop an improved empirical mixed mode-I/II fracture criterion, mixed mode SIFs obtained from detailed finite element models of CTS specimen and fracture results from the tests for different loading angles are used. These data are used in a regression analysis using Datafit TM [13], from which the empirical mixed mode-I/II fracture criterion formula is determined by performing regression analysis (curve fitting). Comparisons of Mode-I/II Fracture Criteria In Terms of Fracture Loads In this subsection, the predicted fracture load values from the developed criterion are compared with other existing criteria in the literature. Applying the method mentioned above, equivalent stress intensity factor ( Keq ) in the developed criterion is defined by: 2 3 eq I II II II K a (b K ) (c / K ) (d / K ) (e / K )       (12) This equation is developed by using CTS specimen data obtained for 15°, 30°, 45°, 60° and 75° loading cases. 0° loading case data is excluded to develop the improved in-plane mixed mode-I/II equivalent SIF equation. Coefficients in Eq. (12) are given in Tab. 3. a b c d e 3.0731 1.0311 -7.3269 5.7100 -1.4163 Table 3 : Coefficients of developed equivalent SIF equation (CTS specimen data used). Experimental loads and predicted fracture loads obtained from the developed criterion for the CTS specimen along with other existing criteria in the literature for different loading cases are given in Tab. 4. The results are also plotted and shown in Fig. 1. It can be seen from the Fig. 1 that for Erdogan and Sih, Richard and Pook criteria, divergence occurs after 45° loading case and increases with increasing loading angle. Tanaka criterion is good agreement with the experimental results up to 60° loading case but the criterion does not produce accurate result for 75° loading case. Critical load values obtained from the developed criterion and experimental results are almost identical with an average error rate of 2.8% for all loading angles performed. Experimental and predicted T-specimen fracture load values obtained from the developed criterion along with other existing criteria in the literature for different loading cases are given in Tab. 5. Critical load values of T-specimen according to developed criterion are acquired by substituting mixed mode SIFs obtained from detailed finite element analyses of T-specimen for different loading angles in the equivalent SIF equation developed by using aonly CTS specimen data. Comparisons of experimental and predicted critical fracture load values of T-specimen are presented in Fig. 2. As can be seen in the figure, all criteria are in good agreement with the experimental results up to 60° loading case. Divergence occurs after 60° loading case for existing criteria. Critical load values obtained from the developed criterion and experimental results are almost identical with an average error rate of 5.5% for all loading angles performed. Thus, criterion is validated with T-specimen fracture results. Furthermore, T-specimen is also proposed as a new specimen type, since its experimental results are validated with the developed and existing criteria. I