ML18152B452

Forwards Comments on Review of Preliminary Accident Sequence Precursor Analysis of Operational Event That Occurred at Plant,On 980508,as Reported in LER 98-009

ML18152B452
Person / Time
Site:	Surry
Issue date:	09/14/1999
From:	Christian D VIRGINIA POWER (VIRGINIA ELECTRIC & POWER CO.)
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
99-424, NUDOCS 9909170224
Download: ML18152B452 (6)

Text

i ..

e e VIRGINIA ELECTRIC AND POWER COMPANY RJCHMOND, VIRGINIA 23261 September 14, 1999 United States Nuclear Regulatory Commission Serial No.99-424 Attention: Document Control Desk NL&OS/GDM R1 Washington, D. C. 20555-0001 Docket No. 50-280 License No. DPR-32 Gentlemen:

VIRGINIA ELECTRIC AND POWER COMPANY SURRY POWER STATION UNIT 1 REVIEW OF PRELIMINARY ACCIDENT SEQUENCE PRECURSOR ANALYSIS OF OPERATIONAL EVENT - SEAL INJECTION LINE LEAK By letter dated August 11, 1999, the NRC provided Virginia Electric and Power Company (Virginia Power) a copy of the preliminary Accident Sequence Precursor (ASP) analysis associated with the "C" Reactor Coolant Pump seal injection line leak that occurred at Surry Unit 1 on May 9, 1998. The leak was reported in Licensee Event Report No. 280/98-009.

We have reviewed the ASP analysis and provide the following attached comments that'we believe directly affect the ASP analysis results. The comments specifically address the initiating event frequency and also include considerations regarding the Surry Individual Plant Examination (IPE) recovery models and success criteria.

Please contact us if you have any questions or require additional information.

Very truly yours, David A. Christian Vice President - Nuclear Operations Attachment Commitments made in this letter: None.

9909170224 990914 PDR ADOCK 05000280 S PDR

e cc: US Nuclear Regulatory Commission Region II Atlanta Federal Center 61 Forsyth Street,-S.W., Suite 23T85 Atlanta, Georgia 30303 Mr. R. A. Musser NRC Senior Resident Inspector Surry Power Station

Attachment Virginia Power Comments for

. "Review-of Preliminary.AccidentSequence Precursor Analysis of-Operational -

Event at Surry Power Station, Unit 1" Virginia Power has reviewed the subject NRG document and provides the following comments on the Accident Sequence Precursor (ASP) analysis of the Surry Power Station Unit 1 event on May 9, 1998, as reported in Licensee Event Report (LER) No.

50-280/98-009. These comments are arranged into the following three categories:

• More realistic initiating event frequency

• Surry IPE recovery models not utilized, and

• Surry IPE success criteria not utilized More Realistic Initiating Event Frequency The failure probability estimate used in the ASP LER No. 50-280/98-009 analysis made use of historical information developed for the Swedish Nuclear Power Inspectorate (SKI). The estimate used known leak information (world data) and Bayesian techniques to estimate a pipe rupture probability. The probability of occurrence dete'rmined was

• 2.4E-2. It is noted on page 3 that the estimate is uncertain, since no actual ruptures have occurred. Page 4 indicates that the uncertainty in the pipe failure probability can be up to three orders of magnitude using probabilistic fracture mechanics and varying parameters in the piping such as materials and crack location.

Assuming there are 50 Westinghouse nuclear plants in the world with an average of three Teador coolant pumps, the probability of occurrence used above, if realistic, would imply that on average one or more ruptures per year should occur at Westinghouse plants at the proposed location. The historical database used by the analysis does not support this conclusion. The discrepancy is probably caused by the large uncertainty acknowledged.

Probabilistic fracture mechanics codes have been used at Surry to determine piping failure probabilities. They were used in the risk informed inservice inspection (RI-ISi) pilot program developed and implemented for Surry Unit 1. WGAP-14572, Revision 1-NP-A,-and Supplement 1 (References 1 and 2) describe the process and software code used in the pilot program. These documents received a satisfactory safety evaluation from the NRG in December 1998, for selecting and categorizing piping components into high and low safety-significant groups, pertaining to risk-informed inservice inspection. The information used in calculating the estimated failure probabilities reflect actual plant conditions specific to the location in question. As such, a significant amount of the uncertainty found in the world data is removed.

At' the lo*cations in question, the pilot project estimated small leak (through-wall crack) and large leak (100 gpm) failure probabilities at 40 years. These probabilities are as follows:

Small leak = 6.63E-03 Large leak= 5.61 E-03 Converting the probabilities to a per year basis follows:

Small leak = 1.66 E-04 Large Leak = 1.41 E-04 These best estimates are based upon the median value calculated and assumed a vibratory fatigue component. The estimate takes no credit for inservice inspection or leak detection (if these were credited it would result in approximately a factor of 2 reduction in the probability). Supplement 1 of WCAP-14572, Revision 1-NP-A, provides on page 54, Figure 4-5, uncertainty work using a similar fracture mechanics code (PC-Praise). The figure, though based upon the fatigue failure of a 6" stainless steel pipe, indicates an upper bound uncertainty of about one order of magnitude.

Therefore, it is recommended that the large leak probability of 1.41 E-04 be used in lieu of the value currently used for failure probability in the preliminary ASP analysis.

Surry IPE Recovery Models Not Utilized The Surry Power Station IPE submittal (Reference 3) provides several system recovery models not addressed in the ASP Program Surry model. System recovery models pertinent to LER No. 50-280/98-009 include:

1. High Head Safety Injection (HHSI) recovery through cross-tie from opposite unit, using equipment from opposite unit (see IPE Sections 3.2.12 and A.7, see especially Table A.7-2, item 4). This was modeled with IPE basic events for cross-tie valves, human recovery (HEP-1 FRC:1-2) and Unit 2 HHSI pump equipment. Subsequent Virginia Power PRA models have expanded the Unit 2 HHSI pump equipment models.

2.

• Refueling-*water Storage Tarik {RWST) recovery through cross-tie from opposite unit, using equipment from opposite unit (see IPE Section A.7, Table A.7-2, item 3, and Table 3.3.3-3). This was modeled with IPE single recovery basic events combining human recovery, cross-tie valves and Unit 2 RWST (REC-XTIE-RWST). Subsequent Virginia Power PRA

• models have incorporated the recovery directly into the applicable fault trees.

l e

These IPE recovery models were not observed in the ASP LER No. 50-280/98-009 analysis 'dominant sequence, SLOCA Sequence 13, and would decrease the conditional core damage probability of cut sets 1, 3, 4 and 6 in Table 5 of that analysis.

Surry IPE Success Criteria Not Utilized The Surry Power Station IPE submittal (Reference 3) assumes a small LOCA success criteria that does not require the Residual Heat Removal (RHR) pumps. Surry IPE Table 3.1.1-16 presents the following core heat removal small LOCA success criteria:

Small LOCA Success Criteria Core Heat Removal Early Late 1 of 3 Charging Pumps (Ref. 1 of 3 Charging Pumps, and

4) 1 of 2 Low Head SI (LHSI)

Pumps in Recirculation Mode 1 of 3 Charging Pumps, and 1 of 3 Charging Pumps, and 2 PORVs (Ref. 5) 1 of 2 Low Head SI Pumps in Recirculation Mode 3 of 3 Accumulators, and 1 of 2 Low Head SI Pumps in 1 of 2 Low Head SI Pumps Recirculation Mode (Ref. 6)

References:

4. WCAP-9601

5. WCAP-9744

6. WCAP-9754 and Surry Specific Analysis Similar small LOCA success criteria without requiring RHR pumps were generated in an earlier Surry PRA as part of the NUREG-1150 program. It is noteworthy that the PRAs of this program received extensive peer review throughout the industry. The specific NUREG-1150 Surry small LOCA modeling is addressed in Section 4.4.9, Table 4.4-11, and Figure 4.4-10 of NUREG/CR-4450, Vol. 3, Rev.1, Part 1 (Reference 7).

It should be noted that unlike many other Westinghouse PWRs, Surry has separate RHR and LHSI pumps. The Surry IPE and current at-power PRA use the LHSI pumps for core heat removal in all transients and LOCAs except the Steam Generator Tube Rupture event (SGTR).

These Surry IPE success criteria were not observed in the ASP LER No. 50-280/98-009 analysis dominant sequence, SLOCA Sequence 04, and would decrease the conditional core damage probability of all of the cut sets of that analysis. In particular, SLOCA Sequence 4 focuses upon the electrical support system single point failures in the RHR system, which do not exist for the Surry LHSI system.

e e Finally, ASP SLOCA Sequence 04 and Sequence 07 assume that both of the RWST isolation' MOVs, 1-CH-MOV-11158 and D, must close for HHSI recirculation success.

However, the Surry IPE determined that these MOVs are not required to close to isolate the RWST during piggyback recirculation (LHSI -takes suction from containment sump and supplies the HHSI pumps). There is sufficient flow from one LHSI pump to reseat either check valve in the RWST suction line (see IPE Table A.7-2, item 21).

References:

1. Westinghouse Energy Systems Topical Report WCAP-14572, Rev. 1-NP-A, "Westinghouse Owners Group Application of Risk-Informed Methods to Piping lnservice Inspection," February 1999.

2. Westinghouse Energy Systems Topical Report WCAP-14572, Rev. 1-NP-A, Supplement 1, "Westinghouse Structural Reliability and Risk Assessment (SRRA)

Model for Piping Risk-Informed lnservice Inspection," February 1999. *

3. W. L. Stewart to USNRC Document Control Desk, "Virginia Electric and Power Company, Surry Power Station Units 1 and 2, Response to Generic Letter 88-20 and Supplement 1, lndivjdual Plant Examination (IPE) for Severe Accident Vulnerabilities," Serial No. 91-134A, August 30, 1991.

4. Westinghouse Electric Corporation Topical Report WCAP-9601, "Report on Small Break Accident for W-NSSS System," Volumes 1-3, June 1979.

5. Westinghouse Electric Corporation Topical Report WCAP-9744, "Loss of Feedwater Induced Loss of Coolant Accident Analysis Report," May 1980.

6. Westinghouse Electric Corporation Topical Report WCAP-9754, "Inadequate Core Cooling Studies of Scenarios with Feedwater Available Using NOTRUMP Computer Code," June 1980.

7. U.S. Nuclear Regulatory Commission, NUREG/CR-4550, Vol. 3, Rev. 1, Part 1, "Analysis of Core Damage Frequency: Surry, Unit 1 Internal Events," April 1990.