ML090370666
| ML090370666 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 02/06/2009 |
| From: | NRC/RES/DE/CIB |
| To: | |
| References | |
| Download: ML090370666 (5) | |
Text
ENCLOSURE 2 Detailed Results Received from ORNL Problem Statement In a recent application to NRR, the Calvert Cliffs (CC) Nuclear Power Plant (Unit 2) requested an extension of its in-service-inspection interval from 10 to 20 years. As part of this request, NRR has required that CC assess if flaws found in the RPV beltline are bounded by the flaw tables in 10CFR50.61a. The documents provided by CC demonstrate that one flaw is not bounded by the 10CFR50.61a flaw tables.
The goal of this analysis is to use FAVOR to determine if this single flaw found in CC, which is not bounded by the 10CFR50.61a flaw tables, would significantly contribute to the risk assessment results of Calvert Cliffs.
FAVOR (07.1) Analysis Details The 07.1 version of FAVOR was applied to perform a probabilistic fracture mechanics (PFM) analysis for the specified flaw geometry, residing in a plate region with the specified embrittlement-related properties, subjected to the 30 transients for Palisades that were applied in the PTS re-evaluation program.
The 07.1 version of FAVPFM was modified such that each RPV would contain exactly one flaw, in every case the specified flaw geometry: an embedded flaw with its inner crack tip located 0.40 inch from the clad-base interface, a through-wall thickness of 0.60 inches, and a length of 0.95 inch (aspect ratio (l / a) of 3.17). Also, the 07.1 version of FAVPFM was modified such that all of the flaws would be axially oriented (normally half of the embedded flaws in plate material are assumed to be axially oriented and half are assumed to be circumferentially oriented).
The PFM analysis was performed for 50,000 RPVs; therefore, there were a total of 50,000 flaws in the analysis. The flaw geometry was constant for all RPVs; however, RTNDT was calculated for each RPV using appropriately sampled values of chemistry and neutron fluence. Figure 1 illustrates the probability distribution of RTNDT at the inner crack tip of the specified embedded flaw geometry determined by sampling from the specified embrittlement-related properties.
FAVOR (07.1) Analysis Results The conditional probability of crack initiation (CPI) was zero for all transients except transient
- 48. The CPImean for transient 48 was 5.52e-11. The CPFmean was the identical value, i.e., all initiated flaws propagated thru the wall to failure. The integrated results were that the mean value of the frequency of crack initiation and thru-wall crack frequency were identical and were equal to 3.57e-17. The 95th percentile of both the CPI and the CPF distributions were zero.
As a check for consistency of the full FAVOR analysis discussed in the preceding paragraph, the deterministic minimization option of FAVOR was used for each of the 30 Palisades Enclosure 2
transients to determine how high the simulated value of RTNDT needs to be such that the conditional probability of crack initiation becomes greater than zero of the flaw/transient combination analyzed. This value is called the Minimum RTNDT for Crack Initiation, or RTNDT(min-init), a name intended to say that all higher simulated values of RTNDT will produce some finite probability of crack initiation and, depending on the transient, vessel failure. Figure 2 illustrates, for Transient 48, that at RTNDT(min-init) the applied KI time history becomes tangent to (or just touches) and time history of the minimum value of fracture initiation toughness (KIc).
Transient 48 is a late re-pressurization transient, as illustrated in Figure 3.
Table 1 contains the incipient value of RTNDT(min-init) for each of the 30 Palisades transients.
Transient 48 has the lowest value of RTNDT(min-init). Of the 50,000 flaws simulated in the FAVOR analysis, there were two flaws predicted to have a conditional probability of crack initiation (CPI) greater than zero, i.e., there were only two simulated values of inner crack tip RTNDT that exceeded the values of RTNDT(min-init) in Table 1. These two simulated inner crack tip RTNDT values were 199.22 °F and 208.84 °. Each of these exceeds the RTNDT(min-init) for transient 48, but they are both below the RTNDT(min-init) values for all other transients. Consequently, only transient 48 contributes to the crack initiation and vessel failure probabilities in this analysis.
4 sample size = 50,000 lowest simulated RTNDT = -135.53 F RTNDT mean = 108.97 F highest simulated RTNDT = 208.84 F neutron fluence = 7.22 x 1019 Copper = 0.11 wt %
3 Nickel = 0.56 wt %
Phosphorus = 0.007 wt %
Manganese = 1.45 wt %
RTNDT(u) = 10 F \
denstiy (%)
2 Eason 2006 RTNDT correlation applied 1
0
-150 -100 -50 0 50 100 150 200 250 RTNDT (F)
Figure 1 - Probability distribution of RTNDT at the inner crack tip of the specified embedded geometry determined by sampling from the specified embrittlement-related properties Table 1 - Values of incipient crack tip RTNDT for 30 Palisades transients (warm pre-stress in effect) for axially oriented flaw Transient ID Minimum RTNDT for number Crack Initiation RTNDT(min-init) (°F) 2 673.51 16 423.11 18 477.30 19 342.91 22 407.48 Enclosure 2
Transient ID Minimum RTNDT for number Crack Initiation RTNDT(min-init) (°F) 24 376.34 26 376.34 27 322.13 29 378.21 31 288.73 32 348.02 34 339.33 40 230.52 42 673.51 48 198.99 49 378.21 50 312.60 51 281.32 52 339.09 53 354.38 54 271.38 55 332.15 58 235.87 59 302.21 60 294.94 61 356.29 62 237.38 63 224.68 64 246.80 65 224.84 Enclosure 2
Illustration of incipient value of inner crack tip RTNDT when specified flaw geometry is subjected to Palisades transient 48 70 minimum KIc represented 60 by Weibull "a" parameter for inner crack tip = 199 F 50 KI and KIc (ksi in 1/2) 40 30 20 10 applied KI at inner crack tip 0
0 50 100 150 200 250 transient time (minutes)
Figure 2 - Illustration of the incipient value of RTNDT for Palisades transient 48. The incipient value of RTNDT creates a point of tangency between the applied KI time history and the minimum value of fracture initiation toughness KIc time history as represented by the Weibull a parameter.
Enclosure 2
Palisades transient 48 600 3.0 500 2.5 P(t) coolant temperature (F) 400 2.0 pressure (ksi) 300 1.5 T(t) 200 1.0 100 0.5 0 0.0 0 50 100 150 200 250 300 transient time (min)
Figure 3 - Palisades transient 48 is a late re-pressurization transient Enclosure 2