PSI - Issue 22

Maksym Zarazovskii et al. / Procedia Structural Integrity 22 (2019) 305–312 Maksym Zarazovskii and Yaroslav Dubyk / Structural Integrity Procedia 00 (2019) 000 – 000

309

5

In all countries with developed nuclear industry, the function (3) for deterministic integrity assessment of RPV is defined as the lower envelope of a large number of experimental data. Therefore, deterministic evaluation of the RPV brittle strength by (3) will always contain certain conservatism. For the manufacture of WWER-1000 RPVs, 15Kh2NMFA (15Kh2NMFA-A – for radiated parts) steel is used, which provides strength characteristics at the level 400 MPa in thicknesses up to 650 mm at 350°C. Fracture toughness for 15Kh2NMFA steel is specified by Strength Calculation Norms for Nuclear Power Plant Equipment and Piping – PNAE G-7-002-86 (1989). Fracture toughness curves of the WWER-1000 RPV steel and its welds are specified by the four different documents, that used in Ukraine to deterministic RPV brittle strength assessment namely: PNAE G-7-002-86 (1989), MT-D.0.03.391-09 (2009), VERLIFE (2008) and MRKR-SKhR-2004 (2004). Fig. 2 shows that FT curves for welds of WWER-1000 RPV (MT-D.0.03.391-09 (2009) FT curve is as much as MRKR-SKhR- 2004’s (2004) multiplied on 1.1). You can see, that for the same material we have three different curves. Therefore, using the different codes we can obtain different safety margin in terms of time of safety operation by brittle fracture criteria.

Fig. 2. FT curves for welds of WWER-1000 RPV according to different normative codes

The certification tests results (Timofeev et al., 1994) and our own experiments (Orynyak et al., 2013) shows, that the PNAE G-7-002-86 (1989) fracture toughness curves for WWER-1000 RPVs metal was built as a lower bound of all data. So, it is not understandable why the conservative PNAE G-7-002-86 (1989) FT curves needs to be replaced by the more conservative ones? Besides, according to the PNAE G-7-002-86 (1989), FT toughness in the initial state of RPV metal is determined separately for the base metal (BM) and the welds (for this purpose the different institutions of the USSR carried out large-scale fracture toughness experiments, including large - size specimens for elevated and high temperatures). Moreover, all modern codes (MT-D.0.03.391-09, 2009; VERLIFE, 2008; MRKR-SKhR-2004, 2004) specify the so called “ upper shelf ” of FT curves at the level of 200 MPa √ m. The strange situation we can obtain in practice: if we postulate a big enough crack we obtain stress intensity factor (SIF) more than value of its upper shelf, and consequently – we get formal unfulfillment the brittle fracture criterion. From the other hand, it is well known that FT within LEFM do not inherent such restriction as upper shelf postulation not saying about experiments for WWER RPV (see Brumovsky ’s (1993) experiments for WWER-1000 RPV base metal and Nosov and Chirkin (1988) – for the WWER 1000 RPV welds). So, we can conclude, that this artificial upper shelf of FT curve must be excluded at all, as it was in the origin at WWER-440 and WWER-1000 RPV design calculations. And finally, issue related to the WWER RPV FT – a few words about Master-Curve (MC) need to be noticed. On Fig. 3 the comparison of classic FT curves indexed CTBs and MC indexed T 0 for BM and weld of RPV of Unit #3 of Zaporizhzhya RPV are depicted. It is good bright example of practical inapplicability of MC, otherwise all knowledge about WWER RPV FT curves that was before MC invention must be considered as mistaken!