Evaluation of the Influence of Fatigue Crack Closure on Eddy Current Testing Signals
公開日:
カテゴリ: 第13回
1. Introduction
In nuclear and thermal power plants, it is significant to manage fatigue damage. Fatigue crack can be detected and sized by eddy current testing (ECT). However, various influential factors affect ECT signals of fatigue crack, and evaluation of the ECT signals including error may lead to underestimation of fatigue crack. In the previous study [1], it has been found that amplitudes of ECT signals of fatigue cracks decrease after heat treatment in the both cases of type 316 and 316L austenite stainless steels, and it is independent of the type of austenite stainless steels. However, the mechanisms of the ECT signal change is still under discussion. In this study, dependences of loading conditions of fatigue test, such as stress ratio R and stress intensity factor K on ECT signal change after heat treatment are investigated.
2. Method
In order to introduce fatigue cracks, specimens with a V-notch as shown in Fig. 1 were designed and used in this study. Fatigue test was done by the four-point bending method, and the support span and loading nodes are shown in Fig. 2. Materials used in this study are type 316 austenite stainless steels. 40kN and 49kN bending loads were set in different specimens and the frequency is 10Hz. In order to figure out the dependence of stress ratio R on ECT signal change, R was set to 0.1 and 0.5. After fatigue test, the upper V-notch of each specimen was removed by electric discharge machining (EDM). According to the previous study [3], there is no effect to the ECT signal of fatigue crack by EDM. After ECT, destructive testing was carried out to measure crack depth. Then the specimens were put into an electric furnace, and specimens were heated to 600°C for 24h. This heat treatment condition can induce the reverse martensitic Fig. 1 V-notched specimen for fatigue test Fig. 2 Configuration of fatigue test by four-point bending method transition effectively. For purpose of opening the fatigue crack mouth, additional bending moment cycles were applied. ECT was conducted using the pancake type probe which is shown in Fig. 3 with an EC instrument (ASSORT PCII, Aswan Electronics Co., Ltd.). 3. Results and Discussion Typical three specimens are picked up among 12 specimens prepared in this study, and specifications are shown in Table 1. The results of ECT signals of the three specimens A-1, A-2 and A-3 are shown in Fig. 4. The amplitude changes of ECT signals are different depending on different loading conditions. Amplitude of ECT signal of A-1 decreased the most. Amplitude of ECT signal of A-2 just decreased a little comparing to the original signals. Amplitude of ECT signal of A-3 decreased roughly by half comparing to the original signals. Even though the stress ratio changed, decrease Fig. 3 Overview of an EC probe Corresponding author: Tetsuya UCHIMOTO E-mail address: uchimoto@ifs.tohoku.ac.jp 10mm 20mm 120mm Length:148mm Length:148mm 40mm - 137 - Table 1. Specification of fatigue crack specimens Specimen Crack depth ID [mm] B9 40 0.1 1.1 B13 49 0.1 2.1 B15 49 0.5 4.3 Cycle number: 93,500 (a) ECT signal of A-1 Cycle number: 59,000 (b) ECT signal of A-2 Cycle number: 275,500 (c) ECT signal of A-3 Fig. 4 ECT signal of type 316 austenite stainless steel of amplitude of ECT signal still exist. In order to discuss the dependence of loading - 138 - Load [kN] Stress ratio conditions on ECT signal change, crack tip stress intensity factors were calculated by using a semi-elliptical surface crack under bending model [2]. Relation between amplitude change of ECT signals and stress intensity factors are shown in Fig. 5. Fig. 5 Relation between amplitude change of ECT signals and stress intensity factor range The change of ECT signal after heat treatment is defined by (A0-A1)/A0, where A0 is the amplitude of the peak of ECT signal before heating and A1 is the amplitude of the peak after heating. Majority of the specimens locate in the region where theΔK is from 20 to 35. After adjusting the stress ratio R, the point still locates in the region mentioned above. All specimens are categorized into two groups; one group is marked as solid triangles that the depth is greater than 3.5mm, another group is marked as solid circle that is smaller than 3.5mm. According to the graph, fatigue crack depth is independent from ECT signal change. 4. Summary In this study, the dependences of stress ratio R and stress intensity factors K on ECT signal change were investigated. Amplitude change does not depend on the value of stress ratio R or crack depth w. In order to discuss the relation between crack closure and stress intensity factor K, conductivity is preferred. The results will be will be presented on the conference. Acknowledgements This work was partly supported by the JSPS Core-to-Core Program, Advanced Research Networks, “International research core on smart layered materials and structures for energy saving” and by the FY2014-2015 and FY2016-2017 Research Exchange Program between the Japan Society for Promotion of Science and Hungarian Academy of Sciences. References [1] X.Y. Wu, T. Uchimoto, T. Takagi, Effect of Thermal Aging of Fatigue Cracks in 316 Austenite Stainless Steels and 316L Austenite steels on Eddy Current Signals, China-Japan Joint Workshop on Flow Dynamics and Transport Phenomena, Tsinghua University,2015. [2] J.C.Newman, Jr. and I.S.Raju, Engng. Frac. Mech., Vol.15, No.1-2 (1981), pp.185-192. [3] H.Feng, Influential Factor Evaluation for Sizing Fatigue Crack of Austenite Stainless Steels in Terms of Eddy Current Signals, Master thesis,2014“ “Evaluation of the Influence of Fatigue Crack Closure on Eddy Current Testing Signals “ “,XiaoYang WU,,Tetsuya UCHIMOTO,,Toshiyuki TAKAGI,,Ryoichi URAYAMA
In nuclear and thermal power plants, it is significant to manage fatigue damage. Fatigue crack can be detected and sized by eddy current testing (ECT). However, various influential factors affect ECT signals of fatigue crack, and evaluation of the ECT signals including error may lead to underestimation of fatigue crack. In the previous study [1], it has been found that amplitudes of ECT signals of fatigue cracks decrease after heat treatment in the both cases of type 316 and 316L austenite stainless steels, and it is independent of the type of austenite stainless steels. However, the mechanisms of the ECT signal change is still under discussion. In this study, dependences of loading conditions of fatigue test, such as stress ratio R and stress intensity factor K on ECT signal change after heat treatment are investigated.
2. Method
In order to introduce fatigue cracks, specimens with a V-notch as shown in Fig. 1 were designed and used in this study. Fatigue test was done by the four-point bending method, and the support span and loading nodes are shown in Fig. 2. Materials used in this study are type 316 austenite stainless steels. 40kN and 49kN bending loads were set in different specimens and the frequency is 10Hz. In order to figure out the dependence of stress ratio R on ECT signal change, R was set to 0.1 and 0.5. After fatigue test, the upper V-notch of each specimen was removed by electric discharge machining (EDM). According to the previous study [3], there is no effect to the ECT signal of fatigue crack by EDM. After ECT, destructive testing was carried out to measure crack depth. Then the specimens were put into an electric furnace, and specimens were heated to 600°C for 24h. This heat treatment condition can induce the reverse martensitic Fig. 1 V-notched specimen for fatigue test Fig. 2 Configuration of fatigue test by four-point bending method transition effectively. For purpose of opening the fatigue crack mouth, additional bending moment cycles were applied. ECT was conducted using the pancake type probe which is shown in Fig. 3 with an EC instrument (ASSORT PCII, Aswan Electronics Co., Ltd.). 3. Results and Discussion Typical three specimens are picked up among 12 specimens prepared in this study, and specifications are shown in Table 1. The results of ECT signals of the three specimens A-1, A-2 and A-3 are shown in Fig. 4. The amplitude changes of ECT signals are different depending on different loading conditions. Amplitude of ECT signal of A-1 decreased the most. Amplitude of ECT signal of A-2 just decreased a little comparing to the original signals. Amplitude of ECT signal of A-3 decreased roughly by half comparing to the original signals. Even though the stress ratio changed, decrease Fig. 3 Overview of an EC probe Corresponding author: Tetsuya UCHIMOTO E-mail address: uchimoto@ifs.tohoku.ac.jp 10mm 20mm 120mm Length:148mm Length:148mm 40mm - 137 - Table 1. Specification of fatigue crack specimens Specimen Crack depth ID [mm] B9 40 0.1 1.1 B13 49 0.1 2.1 B15 49 0.5 4.3 Cycle number: 93,500 (a) ECT signal of A-1 Cycle number: 59,000 (b) ECT signal of A-2 Cycle number: 275,500 (c) ECT signal of A-3 Fig. 4 ECT signal of type 316 austenite stainless steel of amplitude of ECT signal still exist. In order to discuss the dependence of loading - 138 - Load [kN] Stress ratio conditions on ECT signal change, crack tip stress intensity factors were calculated by using a semi-elliptical surface crack under bending model [2]. Relation between amplitude change of ECT signals and stress intensity factors are shown in Fig. 5. Fig. 5 Relation between amplitude change of ECT signals and stress intensity factor range The change of ECT signal after heat treatment is defined by (A0-A1)/A0, where A0 is the amplitude of the peak of ECT signal before heating and A1 is the amplitude of the peak after heating. Majority of the specimens locate in the region where theΔK is from 20 to 35. After adjusting the stress ratio R, the point still locates in the region mentioned above. All specimens are categorized into two groups; one group is marked as solid triangles that the depth is greater than 3.5mm, another group is marked as solid circle that is smaller than 3.5mm. According to the graph, fatigue crack depth is independent from ECT signal change. 4. Summary In this study, the dependences of stress ratio R and stress intensity factors K on ECT signal change were investigated. Amplitude change does not depend on the value of stress ratio R or crack depth w. In order to discuss the relation between crack closure and stress intensity factor K, conductivity is preferred. The results will be will be presented on the conference. Acknowledgements This work was partly supported by the JSPS Core-to-Core Program, Advanced Research Networks, “International research core on smart layered materials and structures for energy saving” and by the FY2014-2015 and FY2016-2017 Research Exchange Program between the Japan Society for Promotion of Science and Hungarian Academy of Sciences. References [1] X.Y. Wu, T. Uchimoto, T. Takagi, Effect of Thermal Aging of Fatigue Cracks in 316 Austenite Stainless Steels and 316L Austenite steels on Eddy Current Signals, China-Japan Joint Workshop on Flow Dynamics and Transport Phenomena, Tsinghua University,2015. [2] J.C.Newman, Jr. and I.S.Raju, Engng. Frac. Mech., Vol.15, No.1-2 (1981), pp.185-192. [3] H.Feng, Influential Factor Evaluation for Sizing Fatigue Crack of Austenite Stainless Steels in Terms of Eddy Current Signals, Master thesis,2014“ “Evaluation of the Influence of Fatigue Crack Closure on Eddy Current Testing Signals “ “,XiaoYang WU,,Tetsuya UCHIMOTO,,Toshiyuki TAKAGI,,Ryoichi URAYAMA