ML20236K645
| ML20236K645 | |
| Person / Time | |
|---|---|
| Site: | Nine Mile Point  |
| Issue date: | 06/24/1998 |
| From: | Lochbaum D UNION OF CONCERNED SCIENTISTS |
| To: | Collins S NRC (Affiliation Not Assigned) |
| Shared Package | |
| ML20236K643 | List: |
| References | |
| NUDOCS 9807100022 | |
| Download: ML20236K645 (6) | |
Text
. - _ - - ~. _.
'g 4
UNION OF Y /
CONCERNED SCIENTISTS June 24,1998 Mr. Samuel J. Collins, Director Office of Nuclear Reactor Regulation United States Nuclear Reguhtory Commission Washington, DC 20555-0001
SUBJECT:
COMMENTS ON REQUEST BY NIAGARA MOIIAWK POWER CORPORATION TO EXTEND CORE SIIROUD INSPECTION INTERVAL, NINE MILE POINT UNIT 1 - DOCKET NO. 50-220 I
Dear Mr. Collins:
l l
By letter dated February 27,1998, the Niagara Mohawk Power Corporation requested relief from its commitment to inspect the vertical welds on the Nine Mile Point Unit I core shroud within 10,600 t
operating hours after restarting from its 1997 refueling outage. UCS reviewed that request, along with recent reports that the horizontal welds on the Unit 2 core shroud are cracked, and submits the enclosed comments for consideration by the Office of Nuclear Reac:nr Regulation in its evaluation of Niagara Mohawk's request.
Several questions were raised during our review that we feel need to be formally answered before the inspection interval can be extended. We respectfully ask that the NRC provide UCS with its answers, or forward UCS the licensee's answers, to these questions before the inspection interval request is granted.
Additionally, since the Unit I core shroud issue generated considerable interest in the local community, including one large public meeting, we request that the NRC consider holding another public meeting on core shroud issues at Nine Mile Point before granting this extension request.
I UCS recommends that the NRC ng approve Niagara Mohawk's request to extend the inspection interval l
for the Nine Mile Point Unit I core shroud vertical welds until a series of questions, included in the attached document, have been satisfactorily answered.
i i
Sincerely, b
I h&h.
David A. Loch um Nuclear Safety Engineer i
9807100022 980630 PDR ADOCK 05000220 P
PDR i
i Washington Office: 1616 P Street NW Suite 310. Washington DC 20036-1495 202 332-0900. FAX: 202 332-0905 Cambridge Headquarters: Two Brattle Square. Cambridge MA 022384105 617-547-5552. FAX: 617-864-9405 California Office: 2397 Shattuck Avenue Suite 203. Berkeley CA 94704-1567 510-843-1872. FAX: 510-843-3785
l.
j June 24,1998 Page 2 of 6 Background - Unit 1 By letter dated February 27,1998, the Niagara Mohawk Powe: Company, the owner and operator of the Nine Mile Point Unit 1 (NMPI) nuclear plant, requested an exten.on from the Nuclear Regulatory Commission in the inspection interval for the NMP1 core shroud's vertical welds. During the most recent refueling outage at NMP1 (spring 1997), cracking had been identified in the vertical welds holding the stainless steel core shroud pieces together. Cracking of the NMP1 core shroud's horizontal welds had previously been identified. Niagara Mohawk had installed tie-rods designed to hold the core shroud segments in place irregardless of the integrity of the horizontal welds.
Through correspondence in late March, April, and early May 1997, Niagara Mohawk sought approval from the Nuclear Regulatory Commission to restart NMP1 with the core shroud's vertical welds cracked.
That request was based on an evaluation which concluded that the identified cracks would not propagate to the point where the shroud's functional integrity would be impaired. That evaluation was contingent on several factors - most notably, that the plant would be operated with excellent reactor water chemistry and that the plant be operated for no longer than 10,600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> before the next inspection of the vertical weld cracks. The NRC approved NMPl's restart on May 7,1997 based on those conditions.
Niagara Mohawk has now requested that the NMP1 core shroud vertical weld inspection interval be extended from 10,600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> to 14,500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br />. This extension, if approved, would allow NMP1 to operate until its next scheduled refueling outage and perform the inspection during that shut down. That requert was based on a more recent evaluation which also concluded that the identified cracks would not propagate to the point of causing the shroud to fail. The key difference between the 1997 analysis and the 1998 evaluations was the assumed crack growth rate. The 1997 evaluation assumed that the existing 4
l cracks would grow (deepen) at a rate of 5x10 inches per hour. The 1998 evaluation reduced the 4
assumed growth rate to 2.2x 10 inches per hour.
The crack growth rate in the 1997 evaluation came from an NRC report (NUREG-0313," Technical Report on Material Selection and Processing Guidelines for BWR Coolant Pressure Boundary Piping,"
Rev. 2, January 1988). While the core shroud is neither part of the reactor coolant pressure boundary nor piping, the NRC's data were conservatively valid when applied to the NMP1 core shroud.
The 1998 evaluation pointed out that the NRC's crack growth rate data were obtained from laboratory tests. These tests were conducted in water having much higher conduc tivity than routinely experienced at NMP1 (0.3 to 0.7 S/cm in the tests vs. approximately 0.09 S/cm at NMPI). Crack propagation rates increase as conductivity increases when all other factors are constant.
The 1998 evaluation justified the lower crack growth rate based on analytical results of samples taken from the cracked vertical welds during the 1997 refueling outage and a CE amputer model for core shroud crack growth rates. The lower crack growth rate assumed that the reactor water conductivity at NMP1 remained below 0.19 S/cm.
y+E'.
l4.;
e Ij M
June 24,1998 Page 3 of 6 Background'- Unit 2 In its Preliminary Notification of Event or Unusual Occurrence PN I-98-018 dated June 2,1998, the L
NRC reported that Niagara Mohawk found 0.25 to 0.65 inch deep cracks in the 2.0-inch thick core shroud on Nine Mile Point Unit 2.
. By memorandum' dated June 10,1998, the NRC relayed Niagara Mohawk's data'on cracks discovered in the horizontal welds on Unit 2's core shroud. Weld H4 had flaws of up to 0.65 inches deep indicated over 55.1 percent ofits length. Weld H5 had flaws of up to 0.65 inches deep indicated over 42.1 percent -
. ofits length.
By letter dated August 23,1994, Niagara Mohawk reported the results of visual inspections it had conducted of the Unit 2 core shroud during October 1993. According to the licensee,"No recordable indications were identified." This letter also reported reactor water conductivity results ofless than 0.20 S/cm for Unit 2 during its first three operating cycles.
I UCS Evaluation - Unit 1 UCS reviewed Niagara Mohawk's letter dated February 27,1998, along with its non-proprietary
' attachments (the letter also had attachments which were claimed by General Electric to contain proprietary information and were thus withheld from public review). We had the following comments -
listed in order of decreasing significance:
Md Attachment I to Niagara Mohawk's letter is Technical Report No. 97181 TR-03, "EPR Testing of L
_ Boat Samples from Core Shroud Vertical Welds V-98 and V-10 at NMP-1," issued by General l-Electric in February 1998. Attachment 6 is Technical Report No. 97181-TR-02, " Interpretive l.
Metallurgical Report on Core Shroud Vertical Weld Boat Samples from Nine Mile Point Unit 1,"
L also issued by General Electric in February 1998. Both of these GE reports were originated by R. E.
Smith and approved by R. E. Smith and D. S. Williams. Neither report had a design verification performed. The stated justification was that it was "Not Required" because the reports were "Non QA." Since the core shroud performs a function that is important to safety, if not safety-related, it l'
would seem that design verifications should have been performed for these GE reports (and any proprietary GE reports). These GE reports comprise the primary basis for Niagara Mohawk's extension request. Appendix B to 10 CFR Part 50 would seem to require that a design verification for these reports be performedprior to their results being used tojustify any such extension request.
1-
.. Attachment 5 to Niagara Mohawk's letter is the non-proprietary version of Report No. GE-NE 523-B13-01869-113NP," Assessment ofCrack Growth Rates Applicable to Nine Mile Point-1 Vertical Weld Indications," issued by General Electric in February 1998. Page 12 of this report stated that t'...the initial crack growth rates were higher due to poor water chemistry...." Page v reported that l(
the cracked vertical welds on the Nine Mile Point Unit I core shroud would grow at an acceptably
'J' low rate as long as its reactor water conductivity was maintained less than or equal to 0.190 S/cm.
. Table I showed that Niagara Mohawk has operated Nine Mile Point Unit I with its reactor water conductivity less than 0.190 pS/cm since 1982. Prior to 1983, the Unit I reactor water conductivity ranged up to 0.656 S/cm.
NRC Daily Event Report (DER) No. 34323 dated June 1,1998, reported that Niagara Mohawk had just identified cracking in two horizontal welds on Nine Mile Point Unit 2. The NRC issued Niagara
- Mohawk an operating license for Nine Mile Point Unit 2 on July 2,1987, Although no reactor water t
t June 24,1998 i
Page 4 of 6 chemistry data for Unit 2 has been reviewed, it is assumed that Unit 2 benefited from the Unit I reactor water chemistry learning curve and was thus able to also keep its conductivity less than 0.190 S/cm.
The proposed conclusion that the Unit I core shroud cracking was caused by poor water chemistry during its early years of operation seems seriously undermined by the discovery ofcore shroud cracking on Uni' 2. Until the root cause of the cracking on Unit 2 is definitively determined, its seems possible that both units may be experiencing cracking to some heretofore unidentified mechanism. to Niagara Mohawk's letter is Repoit No. RDD:98:55863-004-000:01, " Niagara Mohawk's Nine Mile Point Unit 1 Boat Sample Analyses - Part III: Tension Tests," 6 sued by McDermott Technology Inc. in February 1998. Footnote 3 on page 2 of this report revealed that Niagara Mohawk did not have core shroud archive materials for use as control specimens in the tension testing. Consequently, control specimens were made frc~ a Type 304 stainless steel plate.
The report was not specific on how these control specimens were later used. Depending on how these control specimens were used, the substitution of contemporary material for archive material may produce questionable results.
Attachment I to Niagara Mohawk's letter is Technical Repon No. 97181-TR-03,"EPR Testing of Boat Samples from Core Shread Vertical Welds V-98 and V-10 at NMP-1," issued by General Electric in February 1998. Page 6 of this GE report describes that several test results for the NMP1 vertical weld samples obtained during the 1997 refueling outage were discounted because they were
" anomalous." AJ of the " anomalous" results were 'high.' All of the ' low' results werejudged not to be " anomalous." The results of this report would be significantly different if the ' low' results were "anomslous" or if al_l of the results were used.
In the middle paragraph on page 4 of Niagara Mohawk's transmittal letter, the company states that
"...the preliminary plants for the core shroud inspection during the RFO-15 are to complete a 100%
reinspection of all the accessible vertical w elds and selected accessible horizontal welds." Since reinspection implies that welds were accessible before, it is assumed that all of the previously I
inspected vertical welds will be cxamined during the next refueling outage.
) to Niagara Mohawk's letter is the non-proprietary version of Report No. GE-NE-523-l e
B13-01869-113NP," Assessment of Crack Growth Rates Applicable to Nine Mil ePoint-1 Vertical Weld Indications," issued by General Electric in February 1998. Page 12 of this report stated that Figure 3 "... clearly displays that the initial crack growth rates were higher due to poor water l
chemistry and the higher residual stresses on the outside diameter of the shroud." Since Figure 3 was not included, this conclusion could not be verified.
The first sentence on page 3 of Niagt.ra Mohawk's transmittal letter states that the testing of samples taken during the 1997 refusling outage "... demonstrate that the NMP1 core shroud material retains excellent ductility." Considering that NMP1 core shroud's vertical and horizontal welds are cracked so seserely that tie-rods were installed to hold the horizontal shroud cylinders in place, the NMPI core shroud material could have, at best, adequate or acceptable ductility. Ifit truly had excellent i
ductility, it would not be cracked.
l
)
1
\\
i I
June 24,1998 Page 5 of 6 The second naragraph on page 2 of Niagara Mohawk's transmittal letter states that NMP1 is currently operating in fuel cyc:e 13 and its next refueling outage will be refuel outage 15. The numbering scheme appears disjointed. When NMP1 initially started up, it was in fuel cycle 1. After that cycle of operation, it shut down for refuel outage 1. It restarted in fuel cycle 2 and ran until it shut down for refuel outage 2. It's not apparent how NMP1 experienced two more refueling outages than fuel c;cles.
The second paragraph on page 2 of Niagara Mohawk's transmittal letter sta'tes: "The depth of evaluation of the material condition of the NMP1 core shroud at the vertical welds as documented in the attached reports in unprecedented in the industry." It is also true, although not mentioned by Niagara Mohawk, that the depth of cracking of the NMPI core shroud's vertical welds is unprecedented in the industry.
UCS Evaluation - Unit 2 UCS reviewed NRC's notice I-98-018 dated June 2,1998, NRC letter of June 10,1998, and Niagara Mohawk's letter of August 23,1994. Our evaluation of core shroud cracking on Unit 2 is preliminary because a formal root cause has not been made publicly available. An unofficial root cause has been proposed by a Niagara Mohawk spokesperson in a local media report dated June 20,1998. The spokesperson attributed the Unit 2 cracks to " stress put on the metal while the steel pieces were being lifted, formed into circular shapes and welded together." We had the following faitial conanents:
If the purported root cause (i.e., stress occurring early in life of the core shroud) is valid, then it appears that the visual inspections conducted during October 1993 should have identified flaw indications. Yet no such indications were reported.
The purported root cause for cracking on Unit 1 (poor reactor water conductivity) could not have caused the cracking on Unit 2, according to the reactor water conductivity data reported by Niagara Mohawk.
Recommendations UCS recommends that the NRC not approve Niagara Mohawk's request to extend the inspection interval for the Nine Mile Point Unit 1 core shroud vertical welds until the following questions have been satisfactorily answered:
- 1. Why wasn't the design verification performed by General Electric for its technical reports 97181-TR-02 and 97181-TR-03?
2.
Does the recent discovery of core shroud cracking on Nine Mile Point Unit 2 challenge the conclusion that the core shroud cracking on Nine Mile Point Unit I was caused by poor reactor chemistry during the plant's early years of operation?
- 3. How were control specimens used in the boat sample tension tests?
4.
Was it appropriate to rely on control specimens made from "off-the-shelf" Type 304 stainless steel just because Niagara Mohawk did not have core shroud archive meterial samples?
- 5. Did General Electric properly discount the high results during its EPR testing of the boat samples?
.?-
l 4 June 24,1998 l
Page 6 of 6
- 6. Can Figure 3 from General Electric report GE-NE-523 B13-01869-113NP be made publicly available?
- 7. Why is Nine Mile Point Unit 1 in operating cycle 13 with its next refueling outage being Refuel 15?
- 8. What cracked the horizontal welds on the Unit 2 core shroud?
- 9. If the Unit 2 cracking was caused by stress from " lifting, forming, and lifting," why were the flaws not detected until the 1998 outage?
9