ML20095E481
| ML20095E481 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 04/21/1992 |
| From: | Devine J GENERAL PUBLIC UTILITIES CORP. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 5000-92-3020, C321-92-2080, NUDOCS 9204270455 | |
| Download: ML20095E481 (6) | |
Text
- - _ _ _ _ _ _ _ _ _ _ -
l I
NJ Nuclear GPU Nuclear Corpc. ration Parsopany, New Jerwy 07054 201 316-700')
TELEX 136-482 Writer e Direct otal Numter April 21, 1992 C321-92-2080 5000-92-3020 U. S. Nuclear Regulatory Commission Att:
Document Control Desk Washington, DC 20555 Gentlemen:
Subject:
Oyster Creek Nuclear Generating Station Facility License No. OPR-16 bocket No. 50-210 Deferral of Isolation Condenser Piping Stress improvement In order to resolve the remaining open issue under Systematic Evaluation Program Topic 111-5.B, " Pipe Break Outside Containment," GPU Nuclear submitted an evaluation of the newly-replaced Isolation Condenser (IC) System piping outside containment by letter (C321-91-2260) dated October 28, 1991.
The evaluation was based on the guidance in draft Standard Review Plan (SRP)
Section 3.6.3 to demonstrate that the probability of rupture of this piping is extremely low.
The draft SRP Section 3.6.3 quidelines required an assessment of the new piping': resistance to intergranular stress corrosion cracking (IGSCC). Consisteno with the staff's guidance, we planned to implement residual stress improvement treatment to applicable weldments within two years of service. To meet this schedule, the stress improvement work would have to be completed by the end of the next (Cycle 14R) refueling outage.
In light of the very large number of candidate work tasks being considered for the 14R outage, it is clear that some adjustments to the planned workscope must be made in order to stay within established resource constraints.
For that reason we recently reassessed the schedular importance of IC piping stress improvement. Our conclusion is that the combination of material composition of the new piping, welding processes which minimize sensitization and our good reactor water chemistry would ensure that the possibility of IGSCC initiation during an extended service period without stre;s improvement is remote. These IGSCC mitigating factors, when viewed in relation to current industry experience with resistant material and projected service periods without cracking, provide at least a 12 year period during which crack initiation would not be expected.
On that basis, we propose to defer stress improvement of the new IC piping to the Cycle 15R outage.
Any further schedular change will be based on our continuing review of industry studies and experience as well as our spec 1 Tic experience at Oyster Creek.
'3,@V r 'l 9204270455 920421 A 'D l
'I PDR ADOCK 05000219
\\
j P
PDR i
- wam, GPU Nuclear Corpora 0cn is a subsid.a'y of General Puchc Un%es Corporabon
AUACUMUf1 J 50LM10!LC0WUMULUME DM[.5 f 0R SlRLSS IMPRQ1[MtH1 DUlfLAL 1.0 DACKGROUND 1.1 Intergranular Stress Corrosion Cracking (IGSCC)
The original isolation condenser (10) piping material was Type 316 austenitic stainless steel with a carbon content of 0.05 to 0.07 percent.
Welding procedures used during the construction phase were not controlled to minimize heat input.
The absence of these controls sensitized the weid heat affected zone and introduced high residual stresses.
- Thus, the piping was susceptible to IGSCC.
Oyster Creek began commercial operation in 1969.
IGSCC was first observed in 1982 in the isolation condenser condensate return lines.
It was detected via thru-wall leakage and ultrasonic examination, from 1982 until the 1991 Cycle 13R refueling outage 1G500 repairs to 10 piping were made using pipe spool replacement or weld overlays.
1./ Piping Replacement During the 13R outage, the Isolation Condenser System steam supply and condensate return piping outside containment was replaced with piping fabricated from 316 Nuclear Grade stainless steel.
This was done to avoid IGSCC so that the requirements of General Design Criterion 4 could be met as they pertain to thi application of a leak before break analysis to eliminate from the design basis the dynamic effects of pipe rupture.
1.3 Piping Haterial The alloy composition and fabrication processing requirements governing the piping used in tDe replacement c' fort conform with the EPkl guidelines established to ensure resist.',n e to IGSCC.
Carbon content was limited to A maximum oi 0.02 percent which minimizes the possibility of 105C0.
The formation of a sensitized structure as a result of the welding process is signific,9tly reduced.
1.4 Piping Installation Welding heat input was controlled in accordance with the EPRI guidelines.
Specifically, tgese controls recommend an energy density maximum of 2.0 MJ/in,
field installation records show that the actual heat input levels were lower.
These practices are a significant improvement over the original piping installation where similar controls were not utilized, men s
i ATTACHMEUJ_{CMill{ED) 110LA.110RIDMDESIR PIPitlG EASES FOR STRESS IMPR01[ MENT DEf GRAL Reference 3 shows the calculated probabilities of leaks in 304SS and.,16HG piping material as a function of time.
These results were generated using the PRAISE computer code which is
)
based on a probabilistic fracture mechanics model using Monte Carlo simulation techniques.
In tN se calculations worst u se stress conditions were assumed and no credit for any mitigating mea,ures was taken.
Therefore, these results are considered conservative.
The results of calculations are l
summarized below:
1)
The time required *.o reach a given leak probability is about six times as long in 316HG as it is in 304.
2)
Where failure in 304 piping is always dominated by initiated cracks (i.e., resulting from stress corrosion),
in 316NG the initiated cracks dominate the probability of leak only after about 12 years.
i 3)
The probability of a leak in the 304SS weldment exceeds IE 4 after approximately three years, while this value is reached only after 15 years with 316NG.
Therefore, it is concluded that 316HG material owes its l
resist:ms to the fact that fewer cracks initiate than in 304S5 and when they do initiate, it is later in plant life.
These results, even with the cited conservatism, support deferring stress improvement for an additional fuel cycle and does not subject the isolation condenser piping to the risk of crack initiation, in addition, of three other nuclear generating stations with replacement nuclear grade material in service without stress improvement, one has a service period of approximately 7 years with no evidence of IGSCC.
The field evidence alone supports the performance projection.
3.0 CONCLUSION
GPU Nuclear believes it is very unlikely corrosion damage would occur to the extent of Gaw initiation within a 12 year service period.
From this assessment it is concluded that stress improvement can be deferred at least to the Cycle 15R refueling outage as the current piping without stress improvement already provides substantial resistance to crack initiation resulting from stress corrosion.
I Miu3nIbr
_.... ~. _,....., _.... _. _ -.. _.. _,. _ _ ~,..
._..._,,__._,._._.,,,.__-,_,,___.,_,,,,_....m.._,__,.,
,__m.,,_-
l a
ATTACHMJfLLICQRU!!QIDJ 1101 A110N CONDUf1LILPlf111G
+
BASES FOR STRESS IMPROVEMENT DEFERRAL
4.0 REFERENCES
1)
EPRI Report NP-2671-LD, October 1982, " Alternative Alloys for BWR Pipe Applications" 2)
GPU Nuclear Topical Report No. 050, Revision 3. Section 3.2 3)
NUREG/CR-5486, " Application of Reliability Techniques to Prioritize BWR Recirculation Loop Welds for in-Service inspection" l
l b
a l
l' i
3219Emb ht
AllACHMUi1R!{ll!MD1 IMLM10!LCD!MtiMR PIElffG fBELf 0R STRtSS IMPl10MLNT DEfflifLAL l
With the impicmentation of heat input controls, it is unlikely that residual tensile welding stresses or sensitization would occur to the extent where they would adversely effect the material's inherent resistance to IGSCC.
These practices are consistent with those implemented at other utilities for pipe replacement programs.
l i
2.0 DISCUSSION 2.1 Hitigative Hoasures As discussed on the previous page, material composition and welding controls significantly reduce the possibility of IGSCC.
The carbon content limit of the material provides about a four-fold improvement in resistance to IGSCC (Reference 1).
Since 1985 reactor water chemistry controls have been tightened allowing additional margin against the occurrence of IGSCC.
Currently, water conductivity is controlled to less than 0.2 S/cm which yields a factor of 4 improvement in crack growth (Reference 2),
2.2 Expected Functional Performance During installation of the original isolation condenser piping there was an absence of IGSCC mitigation measures.
In fact, factors promoting IGSCC were presut, i.e., high residual strasses, a sensitized material and-a more aggressive environment. Under this combination of factors, it took approximately 13 years for cracks to initiate and propagate to the extent they were readily detected.
Since crack growth is significantly faster than initiation, it is estimatec' that initiation occurred after 8 to 10 years of service.
The crack growth rate of greater than 90 mils /yr discussed in our previously submitted Topical Report 050 supports the premise that cracks can extend from initiation to leakage within 3-4 years in the absence of preventative measures. Tha new piping, installed in the 13R outage along with improved water chemistry is expected to have_a much longer time frame, compared with that of the old piping,- for-IGSCC to initiate.
T 321974 hr
.-. -...-- --..-m.
__.___.....-.._........_.._.,.,_~.m,_...._,-.,__--
Pege 2 The attachment to this letter is a more detailed summary of the bases of this proposal.
We request the NRC staff's concurrence that our leak-before break evaluation of the IC piping outside containment is not impacted since IGSCC mitigating factors are already in place and stress improvement will be implemented at an appropriate time.
Very'tryibo s,
_ {f i
J. C. DeVine, Jr.
Vice Pre::ident and Director, Technical functions
' JCD/PFC/amk Attachment cc: Administrator, Region 1 NRC Resident inspector Oyster Creek NRC Project Manager I
l e
[
o l
- 3289Km0.tr
,...-,--.-..-.n.
..,.,..-~-.-,.-.-,,,,.,-....,.,...,n.,
..,,,..,..n,,,..
.-,,,-,;,.,,..,,,,,,n,-_,.
.