PSI - Issue 14

Vamsi Krishna Rentala et al. / Procedia Structural Integrity 14 (2019) 597–604 Vamsi Krishna Rentala et al. / Structural Integrity Procedia 00 (2018) 000 – 000

601

5

the constants, the remaining fatigue cycles of the component are calculated using Equation 6. In order to convert these remaining fatigue cycles in to engine operating hours, these fatigue cycles were divided by a factor of 3 or 5 by considering at least 3 to 5 throttle excursions during the flight per hour of a military aircraft engine (Koul et al. (1990)). Finally, these fatigue cycles were further divided by a safety factor of 2 for obtaining the safe inspection intervals (SII) to be adopted in the damage tolerance lifing methodology (Koul et al. (1990)). 3. Results and Discussion The HIT/MISS POD data obtained from the previous studies (Vamsi et al.(2017)) by the current authors were utilized in the current study. The obtained data was segregated according to the number of inspection sites, the number of flaws at a site and the corresponding flaw sizes along with FPI and ECI inspection data. For easy understanding, few of the segregated data was shown in Table 1. This data consists of 40 inspection sites and with a total of 69 fatigue cracks. The sizes of these fatigue cracks ranges from 0.1 mm to 2 mm. There are 17 out of 40 inspection sites containing multiple fatigue cracks. From Table 1, it can be observed that the maximum number of multiple fatigue cracks at a site is found to be 4 for site number 27. In addition, Table 1 also provides the FPI and ECI inspection data in the form of H and M corresponding to HIT and MISS, respectively. From the inspection data, it was observed that more number of sites were detected by the ECI technique in comparison with FPI technique. As discussed in the experimental procedure, the FPI and ECI HIT/MISS indications are segregated according to the three approaches and corresponding HIT and MISS indications were assigned for each individual flaw at a site.

Table 1. Number of sites, number of flaws at a site and the corresponding flaw sizes along with FPI and ECI indications. Site No. Number of flaws at a site Flaw Sizes, mm FPI ECI 1 3 0.09, 0.61, 0.61 H H 2 1 1.59 H H 3 1 0.68 M M 5 1 0.75 M H 7 2 0.65, 0.19 M M 16 3 0.43, 0.11, 0.09 M H 23 3 0.12, 0.39, 0.09 M M 27 4 0.22, 0.19, 0.22, 1.15 H H 30 2 0.47, 1.04 H H 35 2 0.38, 0.30 M H 40 1 1.78 H H

Figure 1 shows the POD curves plotted using Type 1, modified Type 1 and Type 2 approaches for FPI and Type 2 for ECI inspection data. From Figure 1, it can be observed that the a 90/95 value is the intersection point on the 95 % Lower Confidence Curve (LCL) curve at 90 % POD value. In addition, it can also be observed that the a 90/95 values for Type 1 and modified Type 1 approaches for FPI inspection data are 1.52 mm and 1.59 mm, respectively. These values are approximately matched and it can be understood that removing the miss indications in POD calculations does not affect the a 90/95 values. In other words, modified Type 1 POD procedure makes the FPI detectability slightly lower than Type 1. Figure 1(c) and 1(d) shows the POD curves plotted using Type 2 approach for both FPI and ECI inspection data. The a 90/95 values obtained using Type 2 approach for both FPI and ECI techniques are 2.38 mm and 1.12 mm, respectively. As the ECI technique’s sensitivity is higher than FPI, the a 90/95 value is better for ECI technique. In addition, it can also be observed that the a 90/95 value of the FPI technique for Type 2 is greater than the Type 1 and modified Type 1 approaches. All these a 90/95 values of both the NDT techniques obtained from three types of POD approaches are given in Table 2.