ML20085D235
| ML20085D235 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 10/10/1991 |
| From: | Feigenbaum T PUBLIC SERVICE CO. OF NEW HAMPSHIRE |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| REF-GTECI-023, REF-GTECI-NI, TASK-023, TASK-23, TASK-OR NYN-91164, NUDOCS 9110150361 | |
| Download: ML20085D235 (4) | |
Text
New Hampshire Yankee =:e Chief Executive Officer NYN-91164 October 10, 1991 United States Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Document Control Desk
Reference:
Facility Operating License No. NIT 86, Doch! No. 50-443
Subject:
Response to " Solicitation of Public Comments on Generic issue 23, Reactor Coolant Pump Seal Failure" Gentlemen: New Hampshire Yankee (NHY) is pleased to have the oppertunity to comm :nt on the proposcd resolution of Generic Issue 23, " Reactor Coolant Pump Seal Failure
- While we agree that the RCP seals are important components of the primary loop and that seal failures may be important contributors to severe accident risk, we offer the following insights for consideration:
RCP seal rnaintenance and operation has significantly improved since the early 1980s as a result of licensees' own efforts to improve safety and reduce downtime due to seal problems. Although Seabrook has only operated for one yeer, we have instituted significant enhancements (de:cribed below) as a result at our review of industry experience and vendor recommendations. the utility. generated improvements in seat maintenance and operation are the most cost effective in assuring high reliability. Additional NRC requirements would provide only marginal improvements but would impose significant paperwork burdens in contro: and reporting. plant speciGc design differences make reasonable generic solutions difficult. For exam},le, the Scabrook seal cooling system includes a separate closed cooling loop that cools cach RCP thermal barrier and that is, in turn, cooled by both trains of PCCW. This added system improves the reliability of the seal cooling for normal operation. the severe accident risk is subject to significant uncertainty in the areas of seal i performance (leak size, time to initiation, effect of depressurization) and time to recover failed equipment (electric power and component cooling water j pumps). In both of these areas, the uncertainty is due to the lack of data seal performance with hot fluid and long-term (greater than 4 hour) on recoverability. While the focus of this issue is on the seal, assumptions about recoverability can significantly affect risk and change the relative importance of fixes to seal leakage. 9110150361 911010 k O PDR ADOCK 05000443 Nh) b3 p PDR New Hampshire Yankee Divisican of Public Service Company of New Hampshire 3.) t " P.O. Box 300
- Seabrook, NH 03874
- Telephone (603) 474 9521
,m. - United States Nuclear Regulatory Commission October 10,.1991-Attention: Document Control Desk - Page two while it may be possible to show that a seal LOCA fix is cost beneficial,--it does not follow that it is the most cost-effective risk reduction measure for each individual plant. This demonstrates the difficulty of trying to address severe accident issues one at a time. Generic Letter 88 20, Individual Plant Examination, was set up by the Commission to systematically examine plants for severe accident vulnerabilities and identify cost effective changes to design and operation to ' reduce the vulnerabilities. The following comments ' are in response to specific areas requested in : the notice and 'uumbered accordingly: 1. Seal Experieuce Because of' our limited operating experience, Seabrook has not-had problems that - forced changes triscal maintenance / operation. Instead, as a result of industry experience and vendor recommendations.and prior to operation, NIIY instituted an extensive RCP Seal Reliability program, that includes the following: .An interdisciplinary task force was set up, consisting of members from maintenance, operations, technical support, engineering, health physics,.and training. This group meets to review seal operation, trends, and opticns to 3: deal with any degradation. ' A 'RCP_ Seal Reliability Analysis" study was completed in January 1989. This study' evei"sted industry experience with seal failures in order to identify the -potential railure modes and what could be done to prevent or detect them. This identified such areas as the need for clean room conditions when handling b the scal - in storage, rebuild, or replacement; the isnportance of seal injection filte:.eplacement to assure that the seals are not contaminated during the switchover; and a trending program to allow for early identification of potential _ seal degradation; - A monitoring program is conducted by the Technical Support Group which L periodically issues a trend report to the task force members. As a result of this monitoring, degraded seal flow was detected in mid 1990 on the "C" RCP
- 1 seal leakoff flow trending low.. Data was r.vai!able over a sufficient length g
L - of time to allow the identification of this slow trend. Several recovery options were tried as directed by the pump vendor, including small. changes in VCT - temperature and seal injection flow. None of these options was successful and lz .in late 1990, when the plant was shutdown, the. pump seal was inspected. This . revealed wear on the # 1 insert and a small amount of crud. Also, the #1 insert was made of stainless steel. The replacement insert is Chrome Carbide coated to give longer service life, as recommended by the vendor, Subsequently, the decision was made to inspect the other three pumps at the l next refueling. In the future, the plan is to inspe-two pump seals per refueling outage, including changing out the cartridge seal package, i m e ,, ~,. r,
'4 United States Nuclear Regulatory Commission October 10, 1991 - Attention: - Document Control Desk. Page three - A full? scale _ mockup' is used in. training for maintenance, health physics ' technicians, and' support. engineers.. This hands-on training -was also used to. optimize the procedures used to remove, rebuild, and replace a seal cartridge. The training also included classroom training for operations, maintenance, and Technical Support personnel on the theory of seal operation and the ways the ~ ~ seal' can possibly fail. NHY believes the above initiatives will ensure reliable seal pe formance and early identification of potential seal degradation.- '2. Experience in Degraded Cooling None at Seabrook. 3. Procedures and Instrumentation 3.1 The Abnormal and Emergency Operating Procedures at Seabrook Station are consistent with vendor guidance which includes responding to a-scal leak, i.e. shutting down - the RCP on abnormal sea! ' indications and closing the seal leakoff line-on high flow, j
- 3.2-The RCP seal instrumentation is fully in compliance with the vendor
~ recommendationsi 3.3 - Vendor information is used as the basis for procedures and maintenance practices. Configuration control assures that vendor updates are reviewed by - the appropriate groups, 3.4 ' IPer vendor ' recommendations,. the abnormal procedure (OS1201.01 RCP Malfunction") calls for RCP trip and plant shutdown for any of the following:
- 1 seal leakoff flow abnormal + high or low i-
- high RCP vibration high RCP motor temperature - loss of thermal barrier cooling and seal injection In the case of. a degradation-of the first stage, such as the low seal leakap ~ experienced earlier at Seabrook, or loss of the second stage flow, the task force and pump p vendor are. consulted for recommendations. 4. ' PRA Seal Models.. -NHY' has no f additional information regarding seal leakage models beyond what is i ' available in the literature. From a PRA standpoint, the NUREG 1150 model seems to bc the best.in that it attempts to account for the range of expert opinion. l 6 M Y'*(Y P" N**# 1"t, ? 9 g +.n,,. p-e a e er-i D-a.-e' *
- e
- -tu
J i Un!ted States Nuclear Regulatory Commission October tu,1991 Attention: Document Control Dtsk Page four The NUREG 1150 scal LOCA probability model has been used in the curren: Seabrook PR A analyp:6. Sensitivity studies have been done to investigate the value in adding j the new high temperr,ture seal O. rings. The new 0. rings resulted i:, decrease in the best estiwate seal leakage rate from about ;$0 gpm per pump to about 25 gpm per pump. flowever, this order.of. magnitude change in leak rate did not significantly affect the core damage frequency for azal LOCA due to two f actors: (1) the small credit given to recovery t of electric power in the long term (greater than 4 hours) and (2) no credit ghen for covery of PCCW. Thus, core damage sequence durat ons are longer but are not i gnificantly changed in frequency. In this sase, the long term electric power recovery and he potential for PCCW recovery need to be investigated to account for conservatism Inherent in PRAs. 5. Testing It is not clear that enough is known about the seal performance to set up a test or series of tests that would resolve this issue. Tests would prebably be inconclusive until the parameters that affect seal performance are better-understood. 6. NRC Action The information presented above demonstrates the exlent of the care that is taken, with no regulatory action, to maintain the seuls. The additional requircraents covered by proposed Regulatory Guide DO>1008, 'Reattor Coolant Pump Seals", would not substantially irnprove seal performance, and would likely only result in resources spent to document and justify activities presently occurring. New llampshire Yankee believes that the assurance that NRC is looking for could be erhieved without the burden to the t. ilities of additional requirements and that generic i rulemaking is not necessary. If you have any questions on this, pir use contact Mr Larry Rau, Reliability and Safety Enginecting Mgnager, at (603) 474 95/1, extension 4305. Very 'ruly yours, 1 / / til llb /m( T.d d). Peigenbaum TCF:Jilli/ss cc: Mr. Thomas T. Martir, Mr. Noel Dudley [ Regional Administtator NRC Senitu Resident inspector U.S Nuclear Regulatory Co;nmission P.O. Box *.149 i Region 1 Seabrook, Nil 03874 475 Allendale Road King of Prussia, PA 19406 Mr. Gordon E; Edison, Sr. Project Manapr Project Directorate 13 -t Division of Reactor Projects l U.S. Nuclear Regulatory Commission ~ Washington, DC 20555 1 -'wtev e-t-me se v -sw v var-ev - ry w-*vy g v + 9 9 ym y wav' ew. w we e e-sev" - m eev v e w gr v - wr eerw e er -v'=e-
- rv e e er w se v-t'w we* w - T e
---s-f+c+eew+wo-wow ~'-*'wv**rv-+**e-+awm**m C}}