ML17334B558
| ML17334B558 | |
| Person / Time | |
|---|---|
| Site: | Cook  |
| Issue date: | 10/20/1995 |
| From: | Rosenthal J NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD) |
| To: | Chaffee A, Strosnider J NRC (Affiliation Not Assigned) |
| References | |
| NUDOCS 9511200217 | |
| Download: ML17334B558 (10) | |
Text
~y,0 REGS 0
Cy OO Op qO
++*++
UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20555-0001 October 20, f995 5o -3(4 HEHORANOUH TO:
Jack R. Strosnider, Jr., Chief Haterials and Chemical Engineering Branch Division of Engineering Office of Nuclear Reactor Regulation FROH:
SUBJECT:
Alfred E. Chaffee, Chief Events Assessment and Generic Communications Branch Division of Reactor Program Hanagement Office of Nuclear Reactor Regulation Jack E. Rosenthal, Chief Reactor Analysis Branch Safety Programs Division Office for Analysis and Evaluation of Operational Data ASSESSMENT OF INSPECTION RESULTS FOR D.C.
COOK UNIT 2 REACTOR VESSEL UPPER HEAD PENETRATION
Reference:
Structural Integrity Evaluation of Reactor Vessel Upper Head Penetrations to Support Continued Operation:
D.C.
Cook Unit 2, WCAP-14118, Revision 1,
- 1994, Westinghouse Electric Corporation.
(Westinghouse Proprietary Document)
AEOD staff and contractors have evaluated the reference document submitted by the licensee to justify the continued operation of D.C.
Cook Unit 2 and believe there is an error in calculations that results in the underestimation of crack growth rate.
The inspection of the D.C.
Cook Unit 2 vessel head penetrations revealed three axial flaws in a peripheral penetration.
The flaws were on the inside surface of the penetration between the 140 and 160 degree locations.
The longest flaw was 45 mm in length and 6.8 mm in depth (about 42.5 percent through-wall),
with its upper end near the attachment weld elevation.
This flaw bracketed the other two flaws, which were below the attachment weld elevation.
The Westinghouse (W) document concluded that the growth of the 45 mm long flaw will not exceed the 75 percent through-wall limit during the next fuel cycle of 18 months.
Therefore, continued plant operation for the next cycle is acceptable.
Repair or replacement of a penetration is required when the maximum flaw depth exceeds the 75 percent through-wall limit.
CONTACT:
Chuck Hsu, AEOD/SPD/RAB (301) 415-6356 c-6 G g'('j 95112002'17 951020 PDR ADQCK 05000316 P
'f
J. Strosnider Our evaluation questions this conclusion.
It appears that there is an error in the use of the crack growth. model for the flaw growth prediction.
The model was developed by P.M. Scott of Framatome for crack growth evaluation at the PWR coolant temperature of 310 'C.
However, the W document uses this model as if it were for crack growth evaluation at a temperature of 330 'C instead of 310 'C and, therefore, provides non-conservative results for the flaw growth at the D.C.
Cook Unit 2 operating temperature of 318 'C.
This use of the Scott model requires further explanation.
We believe the proper use of the Scott model, which is presented in the Attachment (prepared by V.N. Shah, INEL), gives a crack growth rate 2.54 times higher than is estimated in the W document; as a result, the 42.5 percent through-wall crack could exceed the acceptance limit before the end of the next fuel cycle.
We continue to believe that it will take a long time to reach a critical crack
- size, and that detectable leakage will occur before that point.
- Hence, rupture of the nozzle is extremely unlikely.
Attachment:
As stated Distribution:
Public
'.C.entral Fi 1e~
RAB R/F EJordan DRoss CRossi FCongel DHickman PBaranowsky KRaglin MMayfield, RES
- FCoffman, RES
- REmrit, RES
- RSavio, ACRS
- LSoffer, EDO DOCUMENT NAME:
C:WP51tiWPDOCSilASMCOOK.CH To receive a copy of thIs document, Indicate In the box:
C I Copy without attachment/enciosure E" ~ Copy with attachment/enciosure "N ~ No copy OFFICE RAB RAB C
C:RAB D:SPD NAME DATE CHsu:mmk~-
10/>>
/95 GLanik 10/
95 JRosen al 10/Zo/95 CRos 95
Attachment Flaw Growth Evaluation for D.C. Cook Unit 2, Penetration 75 We have discussed here the application of the flaw growth model, which is developed by P.
M. Scott of Framatome (Scott 1991), because it could have been incorrectly used in the D. C.
Cook 2 evaluation.
The main question is whether the model is for the PWR coolant temperature of 310'C or 330'C; this temperature consideration has a significant impact on the estimated fiaw growth.
The Scott model is based on PWSCC growth rate data obtained by Smialowska et al. of Ohio State University. The data were developed at 330'C and include the effects ofseveral different water chemistries.
Only those data associated with standard primary water chemistry of 2 ppm Li, 1200 ppm B, and pH = 7.3 were considered in developing the model.
The equation fitted to these data is da/dt = 2.8 x 10" (K-9)'"
m/sec (K in MPa.m").
The model presents the PWSCC crack growth rate at 330'C as a function of the applied crack tip stress intensity factor K. The equation implies a threshold value ofK,<<c = 9 MPa.m"; no crack growth takes place when the applied crack tip intensity factor is less than K,<<c.
This value ofK<<appears to be reasonable because some other test results also indicate that Ki<<c for Alloy 600 in primary water would be in the range of 5 to 10 MPa.m 'Rebak et al. 1992).
The specimens used by Smialowska et al. for crack growth tests were machined from flattened halves of a short length of steam generator tubing.
These specimens are likely to have a
significantly higher degree of cold work than that found in steam generator tube roll transition regions (maximum of 2%).
Some stress corrosion crack growth rate tests for Alloy 600 performed in 400'C hydrogenated steam environments and in 360'C primary water environments have shown that 5% prior cold work leads to growth rates between 5 to 10 times faster than those observed in materials without cold work (Cassagne and Gelpi 1992). Another factor affecting the crack growth rate is test temperature; crack growth rate is higher for a higher test temperature.
Scott made corrections to the above crack growth equation
[(Equation (1)] to take into account the absence of cold work (or a presence of a small amount of cold work) and the difference between the hot leg operating temperature and the test temperature.
Generally, the rates given by Equation (1) are divided by a factor of 5 for a primary water temperature of 320'C and by 10 for 310'C.
So, the crack growth equation for primary water temperature of 310'C is da/dt = 2.8 x 10'K-9)"
m/sec.
(2)
This equation may be used for predicting crack growth in the CRDM nozzle material, because in the CRDM nozzles cold work is present only in a thin layer of material on the inside surface
of the nozzle, whereas the remaining subsurface material has little cold work.
However, an appropriate temperature correction needs to be made.
Based on the laboratory test results and field data for steam generator tube materials, the D. C. Cook 2 nozzle analysis uses an activation energy of 33 Kcal/mole for the temperature correction.
We have used the same value of the activation energy for estimating the temperature correction for 318'C, as follows:
damage rate cc e+
where (3)
Q = 33 kcal/mole R = 1.986 x 10'cal/mole 'K T = Temperature in Kelvin ('C + 273) damage rate)>>,.c = (damage rate(>>~c) e~
= 1.4708 (damage rate]3ipoc).
So, the temperature correction factor is 1.4708, and the crack growth rate at 318'C is, da/dt = 1.4708 x 2.8 x 10" (K-9)ii'4)
= 4 1182 x 10" (K-9)'"
(5)
Hunt (1994) has used the Scott model to derive the following crack growth model for a primary water temperature of 316'C.
da/dt = 3.67 x 10" (K-9)'" m/sec.
(6)
Equation (6) can be derived from Equation (2) in the same manner as Equation (5) is derived.
Thus, Equation (6) confirms our use of the Scott model [Equation (2)] that gives the PWSCC crack growth rate at 310'C.
There appears to be an error regarding the crack growth model used in the D.C. Cook 2 CRDM nozzle analyses as presented in a document by Bamford and Prager (1994).
The document refers to the Scott model [(Equation (2)] for PWSCC growth rates in Alloy 600, but it considers that the model represents the PWSCC.growth rate at 330'C instead of at 310'C.
Other flaw growth analyses for Alloy 600 components also make a similar reference to Equation (2) (Killian 1995, Briceno and Lapena 1994).
However, this appears to be an error
because Equation (2) already includes a correction for the primary water temperature of 310'C instead of330'C.
So, for the PWSCC growth rate at 318'C, the Westinghouse document gives a correction factor of 0.579, and the resulting crack growth rate equation is da/dt = 1.62 x 10" (K-9)'"
m/sec.
(7)
The crack growth rate given by Equation (7) is about 2.54 times slower than that given by Equation (5).
Figure A-1 shows the crack growth rates for Alloy 600 material as given by Equation (2) representing the Scott model for 310'C, Equation (5) representing the crack growth model for 318'C as derived in this attachment, and Equation (7) representing the crack growth model used in the analysis of D. C. Cook 2 Penetration 75.
It appears that Equation (5) should have been used, instead of Equation (7), for the analysis of crack growth in penetration 75. According to Equation (7), the crack in penetration 75 will grow to 75% through wall in 1.6 years, as shown in Figure A-2 (Bamford et al.
1995).
However, according to Equation (5), the crack will grow at a rate faster than the one used in the report and will reach 75% through-wall limit in about 0.63 years, several months before the end of the next fuel cycle.
Thus, the use of Equation(7) is nonconservative.
References Bamford, W. H. and D. E. Prager 1994.
Structural Integrity Evaluation ofReactor Vessel Upper Head Penetrations to Support Continued Operation:
D.C. Cook Unit 2, WCAP-14118, Revision 1, Westinghouse Electric Corporation.
(Westinghouse Proprietary Document)
Bamford, W. H., et al.
1995.
"Inspection and Evaluation of the Reactor Vessel Head Penetrations at D. C. Cook Unit 2," Service Experience, Structural Integrity, Severe Accidents, and Erosion in Nuclear and Fossil Plants, PVP-Vol. 303, pp. 75-85.
Briceno,D. G. and J. Lapena 1994. Crack Growth Rates in Vessel Head Penetration Materials, presented at the PWSCC Workshop:
Primary Water Stress Corrosion Cracking of Alloy 600 in PWRs, November 15-17, 1994.
- Cassagne, T. and A. Gelpi 1992.
"Crack Growth Rate Measurements on Alloy 600 Steam Generator Tubes in Steam and Primary Water,"
Proceedings of the 5th International Symposium on Environmental Degradation ofMaterials in Nuclear Power Systems - Water
- Reactors, August 25-29, 1991, Monterey, California, pp. 518-532.
Hunt, E. S. 1994. "CRDM Nozzle Strategic Planning," presented at the PWSCC Workshop:
Primary Water Stress Corrosion Cracking of Alloy 600 in PWRs, November 15-17, 1994.
Killian, D. E. 1995.
Fracture Mechanics Assessment of Palisades Alloy 600 Components, BWNT Calc No. 32-1235177-00, prepared for Consumers Power Company, Palisades
- Plant, Docket 50-255, p. 22.
Rebak, R. B., et al. 1992.
"Effects of pH and Stress Intensity on Crack Growth Rate in Alloy 600 in Lithiated + Borated Water at High Temperatures,"
Proceedings ofthe 5th International Symposium on Environmental Degradation ofMaterials in Nuclear Power Systems - 5'ater
- Reactors, August 25-29, 1991, Monterey, California, pp. 511-517.
Scott, P. M. 1991.
"An Analysis of Primary Water Stress Corrosion Cracking in PWR Steam Generators,"
Proceedings of the Specialists Meeting on Operating Experience with Steam Generators,
- Brussels, Belgium, September, 1991, Paper 5.6.
1.00e-09 1.00e-10 E
C 0
Ul O
C5 O
1.00e-11 rr
////////
r II
/
II
/
II
/
I
~
I I
I I
I I
I I
II, I
Equation 2
Equation 5
- - - - Equation 7 1.00e-12 20 40 60 K (MPa m")
80 C129-WHT.10S541 Figure A-l.
Models for stress corrosion crack growth rates in Alloy 600 in primary water environment:
Equation 2 - Scott's model for 310'C; Equation 5, model for 318'C, based on Equation 2; Equation 7, model used in D.
C. Cook 2, Penetration 75 analysis.
0.8 75% through-wall (flaw acceptance limit) 0.6 0.4 td lL 42% through-wall 0.2 0 0 Time (y) 8 Service life 1.6 y 10 c 1~wHT.s 09542 Figure A-2.
PWSCC flaw evaluation chart for D. C. Cook 2, Penetration 75 (Bamford et al.
1995).