PSI - Issue 2_A

Nedeljko Vukojević et al. / Procedia Structural Integrity 2 (2016) 2982 – 2988 Author name / Structural Integrity Procedia 00 (2016) 000–000

2987

6

Statistical analysis shows the reliability of the obtained results of fracture toughness parameters. The higher value of the Weibull's coefficient m means higher reliability and less wastage of value, while the value K o represents mean value of the distribution. Reliability curve obtained by Weibull's distribution shows the interval 99% probability of the measured fracture toughness appear in the distribution interval. Thus, from the curve presented in Fig.2, Fig. 3 and Fig. 4, its obvious the widest interval for the measured values of toughness have specimens taken out from top vessel head "T" (K Ic = 20  180 MPa√m) in Fig. 4. Reliability interval is narrowest for the tested specimens taken out from the bottom of the vessel head "B" (K Ic = 120  155MPa√m), Figure 3. For the middle part of the vessel, specimens marked with "M", the reliability interval of K Ic is within 90  200 MPa√m, as shown in Figure 2. Based on statistical analysis could be considered to be the most critical part is top vessel head ("T" specimen). 4. Metallography of tested specimens Metallographic tests were performed on three samples to determine the microstructure in the plane of the fracture. Charpy specimens taken from three characteristic parts of the vessel were used. The tests were performed according to standard, indicate reference ASTM E407(2007) on the optical microscope OLYMPUS type PMG3 with increment x100. Fig. 5 shows the microstructure of the analyzed samples.

c

b

a

Fig. 5. (a) specimens from vessel middle "M"; (b) specimens from bottom vessel head "B"; (c) specimens from top vessel head "T".

On all three analyzed samples was observed ferrite-pearlite microstructure. A sample from plate were located on the top vessel head "T" is visible presence of Widmansttaten ferrite, and individual presence of something larger ferrite grains compared to samples from the middle of the vessel "M" and samples taken out from the bottom vessel head "B. The presence of coarse grain structure and the structure of sharp edges, typical Widmansttaten's structure, are caused lower values of fracture toughness samples of the "T" in comparison with the samples from the bottom vessel head "B" and samples of "M" located in the middle of the cylindrical part of the vessel. Part of the plate taken from top vessel head (samples "T") was, in the production process of hot plastic deformation, exposed the least deformation, which was caused coarse structure. Other parts were exposed longitudinal and radial deformation and multiple plastic transformation created a fine grained ferrite-pearlite structure and longitudinally orientated grain. This fine grained structure is reflected in the higher values of fracture toughness as shown results of tested samples taken from middle of the vessel and from bottom vessel head. 5. Conclusion The analysis of the results according to the location of specimens sampling, it is clear that the results of calculated critical cracks lengths are worst for specimens taken out form top vessel head (plate of the top vessel head marked with "T"). This part of the structure was only possible to take the test samples, without destruction of structure. As shown in Table 1, row 3 critical crack length is 21,2 mm (min. value) it's significantly lower than the results of the minimal critical crack length obtained from samples taken from the