ML20236P882
| ML20236P882 | |
| Person / Time | |
|---|---|
| Issue date: | 06/16/1998 |
| From: | Collins T NRC (Affiliation Not Assigned) |
| To: | Maloney S AFFILIATION NOT ASSIGNED |
| References | |
| NUDOCS 9807170184 | |
| Download: ML20236P882 (6) | |
Text
- _ _ _ _ _ _ _ _ _ _ - _ _ - - - - - _ _ _ _ _ _ _ - - _ _ _ - - - - _ - - - _ - - _.
lQ g.,o 4 p 9 , ,l e- _4 UNlTEn @TATyp, lo
- j NUCLEAR REGULATORY CEMMISSION WASHINGTON, D.C. 20555 4001 3
( ....+ / JM315N13)B a D 51 FUBUC DOCUMEIU i10CM Mr. Stephen Maloney Devonrue LTD 38 North Street Suite 200 Hingham, Massachusetts 02043
Dear Mr. Maloney:
Thank you for your letters of May 14,1998 and June 1,1998, concerning a number of issues l related to the large-break loss-of-coolant accident (LBLOCA) analysis methodology developed by Siemens Power Corporation - Nuclear Division. The purpose of this letter is to address the technical issues discussed in your letter, and to apprise you of the actions that have been taken by the NRC to ensure continued protraction of public health and safety. The responses below are keyed to the issues discussed in y our May 14 letter; the additional issue regarding NRC's treatment of proprietary information is specifically addressed, as well.
Your May 14 letter cites those sections in Title 10 of the Code of Federal Regulations, Parts 50 and 21 (10 CFR 50 and 10 CFR 21) that detail the responsibilities of boti licensees (10 CFR 50.46), such as Northeast Utilities, and vendors (10 CFR 21), such as Siemens, in reporting errors in LOCA analysis codes and licensing-basis analyses. The NRC takes the enforcement of these regulations vary seriously, and has not hesitated in the past to impose restrictions on licensees to ensure their compliance with 10 CFR 50.46 limits on peak clad temperature (PCT) and other parameters covered by that rule. In the case of analyses performed by Siemens for licensees or by licensees using Siemens' methodology, the NRC l staff has been actively involved in assessments of the analysis codes and their results, as l detailed below. I believe that you will see that actions you attribute in your letter to Siemens !
were taken, in fact, as a result of the NRC staff assessment.
The NRC first became aware several years ago of possible errors in part of Siemens' LBLOCA analysis methodology for pressurized water reactors (PWRs). At that time, the NRC staff requested that Siemens prc ide detailed information for the staff's evaluation, demonstrating that the methodology complied with the requirements of 10 CFR 50.46 and Appendix K of 10 CFR Part 50. Following an in-depth assessment of the part of the methodology in question, the staff informed Siemens in September 1996 that the m-thodology did not meet NRC requirements, and required that Siemens implement changes to comply with those requirements. As a result of these changes, severallicensees using Siemens' PWR LELOCA ;
methodology immedia'.ely reduced power or the power distribution (" peaking factors") in their reactors, prior to a detailed evaluation of the effects of the changes, to ensure that their calculated PCTs were below the 2200 'F regulatory limit. Further evaluation by the affected licensees allowed some of them to remove the restrictions and remain within regulatory limits.
These issues were documented in letters sent by the NRC to all affected licensees, copies of /
l which were placed in the NRC's Pub lic Document Room. Sarnples of these letters are enclosed. .
.L
.:8 o
.9907170184 900616 PDR TOPRP EMVEXXN C PDR t- _ _ _ _ _ _ _ - .- - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - . _ _ -_ _ _ _ _ _ _ _ _ _ _ _ _ __ __
d 2
During its evaluation of Siemens' LBLOCA methodology, the staff became aware of other potential problems with Siemens' analysis codes, including the " variability issues" referred to in your letter. As a result of these concerns, the staff conducted a major inspection at Siemens,
' beginning in February 1997. Over a period of two months, four week-long inspections were performed at Siemens* facilities in Richland, Washington; a total of 23 inspectors took part in these activities. The inspections covered quality assurance (QA) and technical issues related j to Siemens' PWR LOCA methodology, boiling water reactor (BWR) LOCA methodology, design j methodologies for BWR and PWR fuels, and fuel fabrication activities. The staff's inspection report, issued in October 1997, identified 5 non-conformances in Siemens OA activities and 6 unresolved technical issues related to PWR LOCA analyses and PWR and BWR fuel designs.
A copy of the inspection report is enclosed for your information. The "Part 21" notifications to which your May 14 letter refers were issued by Siemens as a result of issues identified by the NRC staff in these inspections and technical evaluations. They were not tied in any way to the Millstone Unit 2 Independent Corrective Action Verification Program (ICAVP).
I The staff has also provided copies of the inspection report to all licensees using Siemens* fuel l and/or analysis methodologies. These letters and the inspection report itself are publicly available. 1 As a part of its inspection raport to Siemens, the NRC staff issued a Demand for Information (DFI) to Siemens to demonstrate its capability to provide fuel and analysis services to NRC licensees, at a level of quality in which the NRC could have confidence that regulatory requirements would be met. Siemens has provided its response to the NRC, covering the DFI, non-conformances, and unresolved technical ise'es. Numerous corrective actions have been taken or are proposed, addressing the issues rused by the staff. The staff is currently reviewing Siemens' response, and wiH be following up to ensure that appropriate corrective I actions are taken to resolve all outstanding issues. The staff anticipates that follow-up activities willinclude a re-inspection of Siemens at an appropriate time.
Siemens has taken actions to correct the errors in its LOCA methodologies, and the staff has reviewed those changes. Each affected licensee has also provided an assessment of the impact of correcting the errors, as required by 10 CFR 50.46, and the staff has reviewed those assessmen'.s. By means of these reviews, the stati has determined that the LOCA analysis coh meet regulatory requirements, and that licensees using Siemens' methodologies are currently operating in compliance with 10 CFR 50.46 requirements. We have received a notification, i.e., the Part 21 notification referred to in footnotes 3 and 5 on p.1 of your letter, that correction of the error in the PWR LBLOCA methodology that caused the variability problem, discussed above, resulted in PCT changes for several plants. While the changes did, as you indicate, apply to all PWRs using Siemens LBLOCA methodology, Millstone Unit 2 was the only licensee for which those changes caused the calculated PCT to exceed 2200 'F at projetd full-power operating conditions. Since Millstone Unit 2 is not operating, no immediate action was necessary as a result of Siemens' findings. However, prior to startup, Millstone Unit 2 witi be required to demonstrate compliance with 10 CFR 50.46 PCT limits. Note that Siemens uses a different analysis methodology for BWR LOCA calculations, so these issues do not apply to 6"/Rs whose licensing-basis analyses use Siemens' codes.
1 Y_-- . _ _ _ _ . _ - _ _ . _ _ _ . . .
i 3
Siemens has also committed to proics a ccmplete update of its PWR LOCA analysis 1 methodology to implement improved models in a number of areas. The staff expects this submittalin August 1998. These models must be approved by the staff prior to use in licensing '
applications.
I would like to note that a number of the issues that you discuss in your letter do not involve LOCA analysis methodologies, and are thus not covered by requirements of 10 CFR 50.46. l These include: '
- Change to Siemens methodology, footnote 2, p.1. This is related to analysis of a steam )
line break, i.e., a break in the secondary side of the steam generator, not a pipe carrying j reactor coolant. The code used for these types of analyses is different from the {
LBLOCA methodology. j e Part 21 report and event notification 32379, footnote 7, p. 2. This refers to BWR fuel design methodology.
- Event notification 34119, footnote 8, p. 2. This is also related to BWR fuel design methodology, i
The repods and notifications indicated in the second and third bullets are related to the )
technicalissues identified by the staff concerning BWR fuel designs. The staff has required that Siemens modify its methodology for calculating critical heat flux / critical power for new 9 x 9 and 10 x 10 fuel designs, including performance of additional tests to improve the databases for these correlations. Only one licensee actually operated a plant with one of the new designs in the core prior to the staff's findings, and enforcement action is being taken against that licensee ,
for violation of technical specification requirements. Other licensees who were planning to use 4 these fuel designs were not allowed to load the fuel into their reactors until Siemens and the licensees provided assurance that the plants would operate within appropriate regulatory and technical specification limitations. With regard to the first bullet, any change in Siemens' methodology for steam line break calculations will be reviewed, when provided to the staff, to ensure that Millstone 2 (and any other affected licensee) complies with appropriate regulatory i 1
requirements.
The information provided above addresses many of the issues and questions that you raise in your May 14 letter, including your questions regarding the discovery of errors in Siemens LOCA methodology (item (4), p. 3); NRC actions to prevent recurrence of these problems (item (6),
- p. 4]. and staff actions taken pursuant to the requirements of 10 CFR 50.46 (item (7), p. 4).
Conceming justification for continued operation for licensees using Siemens' LOCA methodology (item (3), p. 3), licensee operation when parts of the emergency core cooling system (ECCS) are inoperable is govemed by each plant's technical specifications. This refers to hardware (pumps, valves, power supply, instrumentation, support systems) only. Technical specifications detail how long systems may be out of service, and what rctions must be taken (e.g., plant shutdown) if the system cannot be returred to service within the allowed time. j Errors in LOCA analyses, or PCTs greater than 2200 'F, are covered oy 10 CFR 50.46. In that
! regard, there is no justification for a plant to operate under conditions for which a licensing- )
l basis LOCA analysis gives a PCT in excess of the regulatory limit. ;mmediate action must be l I l l
l i
\
-_-___________________a
4 I taken to bring the plant into compliance. As stated above, the staff has determined that all l plants currently operating using Siemens methodology comply with 10 CFR 50.46; Millstone Unit 2, which was shut down for reasons unconnected to its LOCA analyses, must likewise demonstrate compliance with 10 CFR 50.46 prior to its return to power operation.
With regard to your request for an assessment of the likelihood of " comparable errors" in other ,
safety analyses (item (5), p. 3), the staff is required by 10 CFR 50.46 to review and approve ell codes used for LOCA analyses. Those reviews are technically complex, and the staff makes every effort to ensure that there are no known errors in approved LOCA methodologies. It is not unusual for the staff to impose conditions or limitations on the use of certain LOCA models to ensure that analyses are conservative. Experience has shown, however, that errom are occasionally detected, as analytical models are applied to new problems or are used on different computer platforms from those on which they were originally developed and run, or if changes are made consistent with the regulations. The NRC's regulations acknowledge that this may, in fact, occur, and as you have noted in your letter, requirements are included for reporting changes and errors, and their estimated effects, by licensees, on schedules commensurate with their significance, and with immediate action required for cases in which compliance with the regulations cannot be demonstrated. The staff believes that there is a low likelihood that significant errors exist in other methodologies; nevertheless, should such errors be identified, the staff would take immediate action as appropriate to ensure that licensees comply with applicable regulations and that public health and safety are protected.
i With regard to public access to information concerning errors in Siemens' LOCA methodology and NRC staff interactions with both Siemens and affected licensees, these issues were discussed in the NRC's Information Notice 97-15, issued in April 1997, entitled, " Reporting of Errors and Changes in Large-Break Loss-of-Coolant Accident Evaluation Models of Fuel Vendors and Compliance With 10 CFR 50.46(a)(3)," which is available on the NRC's website.
Further, communications and minutes of meetings involving the NRC, Siemens, and affected licensees, which address these issues in great detail, are available for review in the NRC's Public Document Room, and most meetings with Siemens and licensees have been open to the public. The NRC is committed to providing public access to information regarding its activities.
Information is withheld from the public only under certain conditions as required by regulations, such as the protection of proprietary information provice,d by a licensee or venour (in whic' case a redacted version of the document with proprietary information removed must be made available), protection of the identity of an individual who has made an allegation of wrongdoing against a licensee or vendor, or when release of information would imperil an ongoing agency investigation. In inis regard, I would like to note, in response to the issue of proprietary treatment of information related to Siemens LOCA analysis codes in your June 2 letter, that the )
information withheld by Siemens is typical of the kind of information accepted by the staff as ]
proprietary in similar applications. l
. Attached to this letter is additional information in response to your requests for a listing of l licer sees using Siemens* analytical methodologies, and copies of Part 21 reports and other l reports and correspondence relevant to the issues discussed in your letters (items (1) and (2), i respectively, on p. 3 of your May 14 letter). You will see from the list in Attachment 1 that j l
Siemens' clients comprise approximately 17% of the U.S. operating reactor fleet. With regard t l to item (2), I have included notifications provided to the NRC pursuant to the requirements of r l l
U
-e ,
5-
- 10 CFR Part 21, and a representative sample of communications between the NRC and Siemens, and between the NRC and licensees using Siemens' LOCA methodologies, which address the issues that you discuss in your letter. Note that letters such as the one dated October 11,1996 signed by Brian W. Sheron, concerning 10 CFR 50.46 large break LOCA analyses, and those cent in June and July 1997 signed by the Project Manager for the
, addressee and concerning the results of the NRC's inspection at Siemens Power Corporation, were sent to all licensees affected by the issues discussed in the letters. All such letters are' available in the Public Document Room.
Finally, your letters 'suggest that the NRC permitted widespread violations of 10 CFR 50.46
. without prompt action to ensure compliance with regulatory requirements. Your June 2 letter also suggests that the NRC dealt inappropriately with information provided by Siemens as
" proprietary." As a result, your letters have been referred to the NRC's Office of the Inspector General (OlG) for action as deemed appropriate.
Sincerely yours,
/ original signed by/ )
Timothy E. Collins, Chief Reactor Systems Branch Division of Systems Safety and Analysis Office of Nuclear Reactor Regulation Enclosure.
. As stated-l 1
DISTRIBUTION File Center OGC- '
SRXB R/F IG NRR Mail Room ' GHolahan BShero.i TCollins !
AThadani RCaruso I HThompson Alevin l PNorry Elmbro l
. JBlaha !
SRXB:D SR A SRXB:DSSS ALjV RpA USO TECOLLINS E i
. Su /98 - 6/////98 6/4/98 l 1 6f//9 DOCUMENT NAME: G:\MALONEY.LTR
- i. 1 1
4
- i
'Z____.____._________________._______ .
____._..U
e Attachment 1 List of Licensees Using Siemens Fuel and/or LOCA Analysis Services Since 1993 l
EWBS Texas Utilities: Comanche Peak Unit 1 l Comanche Peak Unit 2 l
Carolina Power and Light: H. B. Robinson Unit 2 Shearon Harris Unit 1 Northeast Utilities: Millstone Unit 2 Consumers Power: Palisades Florida Power and Light: St. Lucie Unit 1 Wisconsin Public Service: Kewaunee Note: These 8 operating units represent approximately 11% of operating U.S. PWRs.
BWRs Commonwealth Edison: La Salle Unit 1 La Salle Unit 2 Quad Cities Unit 1 Quad Cities Unit 2 Dresden Unit 2 Dresden Unit 3 Washington Public Power Supply System: WNP-2 Pennsylvania Power and Light: Susquehanna Unit 1 Susquehanna Unit 2 Entergy Operations: Grand Gulf Consumers Power: Big Rock Point (note: plant has been permanently shut down)
Note: These 10 operating units represent approximately 28% of operating U.S. BWRs.
l l
_ _ _ _ _ _ _ _ _ _ _ . _ . _ _ _ _ _ . _ _ _ _ .