IR 05000423/2008010

IR 05000423-08-010; on 12/15/2008 - 02/06/2009; Dominion Nuclear Connecticut, Inc. (Dominion); Millstone Power Station, Unit 3 (MP3); Special Inspection Team Report

ML090820433
Person / Time
Site:	Millstone
Issue date:	03/23/2009
From:	Doerflein L Engineering Region 1 Branch 2
To:	Christian D Dominion Resources
References
FOIA/PA-2011-0115 IR-08-010
Download: ML090820433 (27)

Text

rch 23, 2009

SUBJECT:

MILLSTONE POWER STATION - UNIT 3 - NRC SPECIAL INSPECTION TEAM REPORT 05000423/2008010

Dear Mr. Christian:

On February 6, 2009, the U.S. Nuclear Regulatory Commission (NRC) completed a special inspection at the Millstone Power Station, Unit 3. The enclosed inspection report documents the inspection results, which were discussed on February 6, 2009, with Mr. A. J. Jordan, Site Vice President, and other members of your staff.

The special inspection was conducted in response to the October 20, 2008, discovery of an air void in the 24-inch diameter pipe connecting the refueling water storage tank to the suction of the emergency core cooling system (ECCS) pumps. The NRC's initial evaluation of this condition satisfied the criteria in NRC Inspection Manual Chapter 0309, Reactive Inspection Decision Basis for Reactors, for conducting a special inspection. The basis for initiating this special inspection team is further discussed in the teams charter that is included as Attachment B to the enclosed report. The inspection examined activities conducted under your license as they relate to safety and compliance with the Commission's rules and regulations and with the conditions of your license. The team reviewed selected procedures and records, technical evaluations, calculations, and construction documentation, and interviewed site personnel.

This report documents one self-revealing finding of very low safety significance (Green), which was determined to involve a violation of NRC requirements. However, because of the very low safety significance of the violation and because it was entered into your correction action program, the NRC is treating it as a non-cited violation (NCV) consistent with Section VI.A.1 of the NRC Enforcement Policy. If you contest the NCV documented in the enclosed report, you should provide a response within 30 days of the date of the inspection report, with the basis for your denial, to the Nuclear Regulatory Commission, ATTN: Document Control Desk, Washington, DC 20555-0001; with copies to the Regional Administrator Region I; the Director, Office of Enforcement, United States Nuclear Regulatory Commission, Washington, DC 20555-0001; and the NRC Resident Inspectors at Millstone Power Station. In accordance with 10 CFR 2.390 of the NRC's "Rules of Practice," a copy of this letter, and its enclosures, and your response (if any) will be available electronically for public inspection in the NRC Public Document Room or from the Publicly Available Records (PARS) component of NRC=s document system (ADAMS). ADAMS is accessible from the NRC Website at http://www.nrc.gov/reading-rm/adams.html (the Public Electronic Reading Room).

Sincerely,

/RA/

Lawrence T. Doerflein, Chief Engineering Branch 2 Division of Reactor Safety Docket No: 50-423 License No: NPF-49

Enclosures:

Inspection Report 05000423/2008010 w/Attachment A: Supplemental Information w/Attachment B: Special Inspection Charter In a

REGION I==

Docket No. 50-423 License No. NPF-49 Report No. 05000423/2008010 Licensee: Dominion Nuclear Connecticut, Inc.

Facility: Millstone Power Station, Unit 3 Location: P. O. Box 128 Waterford, CT 06385 Dates: December 15, 2008 through February 6, 2009 Inspectors: W. Schmidt, Senior Reactor Analyst, Division of Reactor Safety (DRS),

Team Leader K. Mangan, Senior Reactor Inspector, DRS M. Davis, Resident Inspector, Division of Reactor Projects (DRP)

A. Rao, Project Engineer, DRP (in training)

W. Lyon, Senior Reactor Systems Engineer, Office of Nuclear Reactor Regulations (NRR)

M. Holmberg, Senior Reactor Inspector, DRS, NRC Region III State Observer: D. Galloway, Connecticut, Department of Environmental Protection Approved by: Lawrence T. Doerflein, Chief Engineering Branch 2 Division of Reactor Safety Enclosure

SUMMARY OF FINDINGS

IR 05000423/2008010; 12/15/2008 - 02/06/2009; Dominion Nuclear Connecticut, Inc.

(Dominion); Millstone Power Station, Unit 3 (MP3); Special Inspection Team Report.

The report covered three on-site inspection visits by a special inspection team consisting of a Senior Reactor Analyst, Senior Reactor Engineer, a Project Engineer, and a Resident Inspector, with support from a Region III Senior Reactor Inspector and staff members of the Office of Nuclear Reactor Regulation. One finding of very low safety significance (Green) was identified.

The significance of most findings is indicated by their color (Green, White, Yellow, or Red) using Inspection Manual Chapter (IMC) 0609, Significance Determination Process (SDP). Findings for which the SDP does not apply may be Green or be assigned a severity level after NRC management review. The NRCs program for overseeing the safe operation of commercial nuclear power reactors is described in NUREG-1649, "Reactor Oversight Process," Revision 4, dated December 2006.

Cornerstone: Mitigating Systems

Green.

The team identified a noncited violation (NCV) of Technical Specification (TS)3.5.2.d which requires an operable residual heat removal (RHR) pump for each train of the emergency core cooling system (ECCS). The team found that Dominion did not maintain the 24-inch outside diameter piping connecting the refueling water storage tank (RWST) to the suction of the ECCS pumps sufficiently full of water to ensure operability of the RHR pumps following a large break loss-of-coolant accident (LLOCA).

Additionally, the team determined that TS Surveillance 4.5.2.b requires that every 31 days Dominion verify the ECCS piping full of water but this section of piping was not checked. While performing actions to address NRC Generic Letter 2008-001, Dominion identified the air void and determined the piping did not have sufficient slope to allow venting back to the RWST. The team concluded the air void had the potential to air bind and make the RHR pumps inoperable during a LLOCA event. Following identification of the air void during the 2008 refueling outage, Dominion isolated and drained the piping, installed a vent valve, refilled the piping, and confirmed that the piping was full using an ultrasonic testing (UT) measurement.

The performance deficiency was a failure to maintain the common ECCS suction piping sufficiently full of water, as required by TS surveillance 4.5.2.b, to ensure RHR pump operability in the event of a LLOCA, as required by TS 3.5.2.d. The finding is more than minor because it is associated with the design control attribute of the Mitigating Cornerstone and affected the cornerstone objective of ensuring the availability, reliability, and capability of systems that respond to initiating events to prevent undesirable consequences. In accordance with NRC IMC 0609, Attachment 4,

"Phase 1 - Initial Screening and Characterization of Findings," the Phase 1 screening identified that this issue was a design/qualification deficiency which resulted in the loss of the RHR system low pressure injection (LPI) safety function and required a Phase 2 evaluation.

ii

In accordance with IMC 0609, Appendix A, Determining the Significance of Reactor Inspection Findings for At-Power Situations, a Region I senior reactor analyst determined that the finding was of very low safety significance (Green) using a modified Phase 2 analysis and the MP3 plant-specific Phase 2 Notebook worksheet for a LLOCA.

This assessment resulted in an increase in the core damage frequency on the order of low E-8 per year, which was dominated by the LLOCA frequency of E-5 per year and the probability of high pressure injection (HPI) failure, due to some other unrelated cause.

The safety injection, charging and recirculation spray systems were still available to prevent core damage following a LLOCA initiating event, by performing the HPI and high pressure recirculation safety functions.

The finding did not have a crosscutting aspect.

Licensee-Identified Violations

None.

iii

REPORT DETAILS

1. INTRODUCTION

1.1 Background On October 20, 2008, with Millstone Power Station Unit 3 (MP3) in a refueling outage (Operating Mode 5), Dominion Nuclear Connecticut, Inc. (Dominion or the licensee)detected a gas void in the 24-inch outside diameter pipe that connects the refueling water storage tank (RWST) to the suction of emergency core cooling system (ECCS)pumps (residual heat removal (RHR), safety injection (SI) and charging (CHS)1). The ultrasonic testing (UT) measurements that identified the gas were being performed to address potential voiding concerns as outlined in NRC Generic Letter (GL) 2008-01, Managing Gas Accumulation in Emergency Core Cooling, Decay Heat Removal, and Containment Spray Systems, dated January 11, 2008. In Mode 5 the Technical Specifications (TS) did not require the ECCS suction from the RWST to be operable.

The 24-inch ECCS supply pipe exits the side of the RWST and proceeds underground where it then enters through the wall of the engineered safety feature (ESF) building. In the ESF building the pipe splits to supply the ECCS pumps. The SI system supply is taken off the bottom of the 24-inch pipe, and the 24-inch pipe then proceeds to a 24 to 16-inch reducer. The RHR suction is taken off the 16-inch pipe at a 45 degree angle while the 16-inch pipe continues to a 16 to 8-inch reducer, and the CHS suction is directly off the 8 inch pipe. The gas void was quantified as 15% of the internal diameter of the 24-inch diameter pipe (approximately 3.5-inch at the top of the pipe), or a 9%

static void fraction ()2.

Dominions examination of this location in the ESF building for the GL 2008-01 piping review was based on their evaluation that there was a potential for a small amount of air at the top of the pipe after reviewing as-welded piping configurations. The measured void depth was substantially larger than the expected depth. Once identified, Dominion prepared a modification to install a vent valve in the 24-inch piping in the ESF building.

Dominion drained the line, installed the vent valve, and refilled the line prior to the RWST needing to be operable to support the outage activities. Dominion also confirmed that the piping was full with a UT measurement.

The observed void existed in the 10-foot section of 24-inch pipe within the ESF building (from the wall to the 24 to16-inch reducer). The actual total gas void volume could not be accurately quantified because of the inaccessibility of the underground header between the ESF building and RWST. Construction (as-built) isometric drawings showed that an 85-foot section of the buried 24-inch header was at approximately the same elevation as the exposed piping in the ESF building. For the remainder of the 24-inch header the as-At MP3, the residual heat removal (RHR) system functions in a low pressure safety injection (LPI) mode, and the safety injection (SI) and charging pumps are considered the high pressure safety injection (HPI)system. The RHR system is not used for containment sump recirculation. Instead, the recirculation spray system (RSS) provides low pressure recirculation (LPR) from the sump and supplies the SI and charging pumps for high pressure recirculation (HPR).

- Void fraction is expressed in percent (%) and calculated as the ratio of the internal cross sectional area of a pipe that is voided to the total internal cross sectional area of the pipe.

built drawings showed the pipe to be sloped upwards towards the RWST or vertical, with the exception of the short horizontal run exiting the RWST. Dominion conservatively assumed that the air void extended at the same elevation, as in the 10-foot section within the ESF building, through the 85 foot of horizontal underground piping section. With the RWST at a lower than normal operating level during the outage, the as-found static of 9% in the 95 feet of 24-inch pipe resulted in a calculated volume of approximately 63 standard cubic feet (scf) of air. This equated to an approximate static of 8% if the RWST was at its normal TS required operating level (higher pressure).

Dominions preliminary operability review concluded that, assuming a consistent distribution to all ECCS pumps, the pumps would experience a worst case entrained dynamic suction of 7%, given the high flowrate in the 24-inch piping following a large loss-of-coolant accident (LLOCA). They further assumed that this dynamic of 7%

would cause air binding and damage the RHR, SI and CHS pumps if a LLOCA occurred.

This dynamic of 7% assumed some attenuation of the original static of 8%. For the other assumed small and medium loss-of-coolant accidents (SLOCA, and MLOCA) the licensee determined that the flowrates in the RWST supply piping would be low enough so that a lower entrainment rate of the air would occur. This would result in a low void fraction at the pump suctions and pump operability would not be impacted.

1.2 Preliminary Conditional Risk Assessment Using NRC Inspection Manual Chapter (IMC) 0309, Reactive Inspection Decision Basis for Reactors, the NRC staff recommended and NRC management approved the conduct of this special inspection team (SIT) on November 20, 2008. The identified suction gas void was determined to be a significant unplanned degraded condition, due to the past potential common cause operability impact on the RHR, SI and CHS pumps following a LLOCA. Several deterministic criteria were met and the risk assessment, conducted by a Region I Senior Reactor Analyst (SRA), estimated the incremental conditional core damage probability (ICCDP) to be in the low E-6 per year range, which was in the overlap region between normal and a special inspection review of this condition. This risk assessment, which used the MP3 Standardized Plant Analysis Risk (SPAR) model, in a bounding case assumed that, given the frequency of a LLOCA over a years time, that the gas void would become entrained and cause damage to all the RHR, SI, and CHS pumps. This assumption of RHR, SI, and CHS pump failure would result in core damage in the event of a LLOCA. The SRA discussed this result with the Dominion risk assessment staff who had calculated a similar result with similar assumptions, using their plant specific risk model.

1.3 Special Inspection Scope As outlined in the SIT charter, provided as Attachment B, the team assessed the impact of the as-found conditions on the accident mitigation functions of the ECCS and reviewed the timeliness and effectiveness of the licensees corrective actions for this and any prior similar events related to voiding in safety related piping systems.

The SIT used NRC Inspection Procedure 93812, Special Inspection, as a guide to complete their review which included: procedures, corrective action documents, modifications, work requests, engineering calculations and analyses, initial plant construction records, the apparent cause evaluation prepared by Dominion and scale model hydraulic testing results. The team also interviewed key plant personnel regarding the discovery and resolution of the condition. A list of site personnel interviewed and documents reviewed are provided in Attachment A to this report.

2. SPECIAL INSPECTION AREAS

2.1 Sequence of Events

a. Inspection Scope

The team developed a complete sequence of events related to Dominions discovery of the voided piping and their follow-up actions to address the condition. The team also included events that occured prior to the identification of the void, which put the issue in context, relative to the total potential length of time which the condition existed and the potential points at which Dominion could have identified the condition.

b. Condition Identification and Resolution Chronology 1997 Refueling Outage ECCS supply line from RWST isolated, drained and subsequently refilled due to outage work.

April - June 2004 Air identified in the A RHR piping system, due to outage activities and inadequate fill and venting.

April 2005 Dominion issues Technical Evaluation M3-EV-05-0008; Determination of Allowable ECCS Gas Accumulations in Support of Surveillance 4.5.2.b.1.

January 11, 2008 GL 2008-01, Managing Gas Accumulation in Emergency Core Cooling, Decay Heat Removal, and Containment Spray Systems, issued.

January - October 2008 Millstone Engineering staff conducted an intensive review of ECCS piping configurations in response to GL 2008-01, documented in M3-EV-08-00261. Based on the as-built drawing it was recommended that the ECCS supply line from the RWST be UT examined, because of a noted potential small difference in elevation in the piping at the ESF building.

October 12, 2008 MP3 shutdown for Refueling Outage 3R12.

October 20, 2008 Gas void identified in the 24-inch piping, in the ESF building, connecting the RWST to the suction of the ECCS pumps. Condition Report (CR) 115088 initiated.

October 27, 2008 New vent valve installed. Work Order 53102195865 completed.

October 28, 2008 24-inch piping connecting the RWST to the suction of the ECCS refilled and verified by UT examination to be full.

November 13, 2008 Technical issues were discussed in a conference call with Dominion, the NRC Office of Nuclear Reactor Regulation (NRR) staff, and Region I Staff. The consensus view of the NRC staff, based on the licensees information, was that there was doubt as to past ECCS pump operability during potential LLOCA scenarios.

November 13 - 20, 2008 IMC 0309 review completed and SIT charter approved by NRC management.

November 24, 2008 MP3 restarted following 3R12.

December 8, 2008 Dominion completed Technical Evaluation M3-EV-08-0035 Evaluation of Gas Void Discovered in the 24-inch RWST ECCS Supply Line (CR115088).

December 15 - 19, 2008 SIT onsite, debriefed with site management on Friday.

December 19, 2008 Licensee Event Report (LER) 2008-004, Gas Void in ECCS Suction Line, submitted. This LER reported a condition prohibited by Technical Specification (TS) 3.5.2, in that, given the identified gas void, the RHR pumps were not operable to respond to a LLOCA.

January 27 - 29, 2009 SIT onsite, debriefed with site management on Thursday.

February 3 - 6, 2009 SIT onsite, exited with site management on Friday.

Outstanding issues included: 1) SIT review of the results from Dominions planned 1/4 scale system flow and testing and 2) SIT review of potential stresses in the 24-inch piping between the ESF building and the RWST, due to the potential that the piping had settled from its as-welded configuration.

February 10 - Dominion conducted and reviewed 1/4 scale testing to March 11, 2009 validate the use of RELAP 5 Mod 3.3 to evaluate the transport of the air void.

NRC Region III Senior Reactor Inspector observed portions of the testing, on February 11.

Dominion provided and the team reviewed information concerning the potential deflection and stress in the 24-inch piping between the ESF building and the RWST. This resolved the outstanding issue from the February 6, 2009 exit meeting concerning the potential stresses in the 24-inch piping between the ESF building and the RWST.

March 11, 2009 The results of the 1/4 scale testing and RELAP analysis were discussed in a conference call with Dominion, members of the SIT, and Region I and NRR technical staffs. The consensus view of the NRC staff, based on the scale testing, was that Dominion adequately predicted the distribution of the air void to the ECCS pump suctions.

This resolved the outstanding issue from the February 6, 2009 exit meeting concerning the 1/4 scale testing.

2.2 Review of Operating Experience

a. Inspection Scope

The team reviewed operating experience concerning air voids in the suction of ECCS pumps. This included MP3 specific issues from the past. The team examined the specific issues associated with Dominions recent void discoveries to assess any new generic issues of industry interest for prompt communication and dissemination.

b. Findings and Observations

No findings of significance were identified.

In 2005, following other issues with air in ECCS piping, Dominion completed an evaluation to ensure compliance with Technical Specification (TS) Surveillance 4.5.2.b.1, and to clarify what full of water means in the TS Basis. This change was implemented to assess the potential areas where gas could accumulate and to allow the use of UTs to determine that piping was full enough to support ECCS pump operability.

As part of this assessment the 24-inch piping was not evaluated and was assumed to be full.

The team did not identify any new generic issues.

2.3 Review of Root Cause and Extent-of-Condition

a. Inspection Scope

The team examined the licensees apparent cause evaluation and engineering technical evaluation for the ECCS suction pipe air void and assessed the condition for evidence of inadequate design and/or system operations. The team independently evaluated plant isometric and piping drawings, plant records including as-welded piping configuration, historical photographs from construction, ECCS pre-operational and outage surveillance testing, and engineering technical evaluations. Additionally the team conducted a detailed field inspection of the ECCS piping in the ESF building. The team also reviewed the work order used to install a new vent valve in the 24-inch pipe in the ESF building.

b. Findings and Observations

No findings of significance were identified.

Dominion characterized the root causes as latent design errors because there was no vent valve installed in the 24-inch piping and the 24-inch piping did not have sufficient slope to allow venting back to the RWST. These design errors allowed gas to remain in the piping following system restoration after draining and refilling of the 24-inch piping for maintenance activities. Dominion ruled out settling of either the ESF building or the RWST as a cause. They also stated that settling of the 24-inch pipe between the ESF building and the RWST was unlikely because this area was refilled with QA Cat1 backfill.

The team agreed with Dominions assessment of the latent design errors. However, based on review of construction photographs and welding information, the team believed that the nearly 3.5-inch deep void indicated that a portion of the 24-inch piping between the ESF building and the RWST was lower than indicated by the as-welded isometric drawing. This piping was installed when the area between the ESF building and the RSWT foundation was excavated, the pipe welds were completed and as-built drawings recorded the centerline elevation of sections of the pipe run to be 15-feet 5 1/8-inches above sea level, for the 85-foot horizontal portion of the pipe. The team noted that the piping up to the RWST was installed and backfilled prior to construction of the RWST.

There are no permanent pipe supports installed or documentation of temporary supports for the pipe during backfill. Additionally, pipe elevation was not verified during backfill of the excavated area.

At the teams request, Dominion provided information on calculated pipe deflection and stress due to the weight of the piping assuming the 24-inch pipe was only supported at the ESF building wall and at the RWST foundation. This calculation indicated that a maximum deflection of approximately 3.5-inches would occur at the upstream end of the 85-foot section of the originally assumed horizontal piping.

The extent-of-condition for this issue was enveloped by the GL 2008-01 review conducted for other similar configurations of suctions from the RWST though buried pipe. This included the quench spray system, where the licensee determined, and the inspectors verified, that given the flowrates involved in the normal monthly test and the size of the piping any air in the suction piping would be be removed during the test.

The team found that the modification to install the vent valve in the ECCS suction piping was conducted properly, after the piping was isolated and drained. The use of the vent valve during refilling and subsequent UT demonstrated that this corrective action had been successful. The team also found this corrective action acceptable even if a portion of the 24-inch piping between the ESF building and the RWST had sagged and was lower than as indicated by the as-welded isometric drawing.

2.4 Independent Review of Engineering Calculations and Operability Determination

a. Inspection Scope

The team reviewed the detailed engineering evaluations and calculations used to assess the transport of the void to the suction of the ECCS pumps, completed subsequent to the initial operability determination, discussed in section 1.0. The detailed calculations were completed using RELAP 5 Mod 3.3. Additionally, the team reviewed Dominions scale model testing of a piping configuration similar to that of the plant, to validate the results of the RELAP calculations. The team conducted independent calculations to validate the flowrates used during the 1/4 scale testing and a Region III senior inspector observed portions of the testing.

The team also reviewed the information provided by Dominion concerning the possible deflection and associated stresses on the 24-inch piping between the ESF building and the RWST.

b. Findings and Observations

No findings of significance were identified.

Dominion conducted a detailed technical evaluation to determine the dynamic at the suction of the ECCS pumps for different sizes of LOCAs using the RELAP 5, Mod 3.3 thermo-hydraulic code to model the piping and the void transport to the suction of the ECCS pumps for large, medium and small LOCA (LLOCA, MLOCA, and SLOCA). To validate the RELAP results Dominion contracted the conduct of 1/4 scale model fluid flow testing. The team reviewed the Dominion analysis and scale model testing, determining the following:

• Use of the 63 scf void was conservative. The team noted that assuming that the entire 24-inch pipe that was designed to be horizontal had the static of 8%

was conservative, because of the teams assessment that the piping was not horizontal and likely at some point lower by at least 3.5-inch. This would tend to make the size of the air void in the pipe approximately half what was assumed in the analysis. This would result in less potential for the RHR pump to be damage following a LLOCA. However, without any specific information concerning the actual configuration of the buried portion of piping this assumption was not included in the teams analysis.

• The RELAP input data adequately modeled the in-plant piping configurations, which included the assumption of the initial size of the air void and an appropriate pumps start transient.

• Based on the relatively low flowrates in the 24-inch piping, during normal RHR, SI, and CHS system in-service testing and outage integrated testing, the team validated that the air void would not have passed to the suction of the pumps during normal operations and maintenance testing.

• The team agreed that the air void was likely in the piping since the 1997 RFO, when the line was isolated, drained, and refilled.

• The estimated dynamic near the suctions for both RHR pumps following system actuation in response to a LLOCA exceeded the 5% limit provided by the pump manufacturer and was greater than the current NRC proposed acceptance criteria for centrifugal pump of less than or equal to 2% for steady state operation (greater than 20 seconds) and less than 10% for 5 seconds. For example the A RHR pump was estimated to receive a distributed dynamic suction above 5%

for 10 seconds with an average of 9% and a maximum of 13%. This included a 13 second period where the dynamic was above 2%, with an average of 8%

and above 10% for 5 seconds with an average of 11%.

• The SI pumps would experience essential no void transfer to the suction of the pumps for any LOCA situation, due to the location of the suction piping tie-in point at the bottom of the 24-inch pipe.

• The CHS pumps would receive suction of less than 2% steady state for SLOCAs and MLOCAs, and for LLOCAs where the RCS depressurizes quickly.

• Dominion also analyzed several, specific LLOCA situations where the RHR pumps would be injecting with the reactor coolant system pressure just below their shutoff head of approximately 200 psig, along with SI and charging pumps, to determine if there were any situation where more than just the RHR pumps could have been affected. In these cases, Dominion found, and the team agreed, that the suction void fraction to the RHR pumps would be less than a steady state 2% and that the void fraction to the CHS pumps could increase, but would not likely damage these multi-stage pumps. Additionally, the team concluded that if the CHS pumps were damaged due to air voiding the RHR pumps would not be damaged, so there could not be a potential impact on both high head and low head injection sources.

• The scaled testing setup, including the established flowrates, represented the Froude number calculated for the in-plant piping and system flow conditions.

• The RELAP calculations and the 1/4 scale model testing validated Dominions position that only the RHR pumps operating in response to a LLOCA would be impacted by the air void.

• The technical evaluation also covered and properly analyzed that there would have been no impact on HPR, if the A RHR pump became damaged and the air volume was distributed in the common section of recirculation spray system (RSS) and RHR pumping used for both LPI and HPR. The B RHR pump was not of concern because the common section of piping would not have trapped the air void.

With respect to the potential that the buried section of 24-inch piping had deflected (see Section 2.3 above) following welding but before backfill of the excavated area the team found:

• The lack of observed settling in the area above the piping indicated that the piping has not continued to deflect significantly since the area was backfilled during construction.

• Based on the additional information provided by the licensee, including seismic loads, the two connection points (at the ESF wall and the RWST) would not be stressed above ASME allowable values and the original piping seismic analysis would remain valid given the assumed deflection.

2.5 Final Risk Assessment of the Condition The team assessed the final risk of the air void in the ECCS supply piping from the RWST finding that the only likely impact on the ECCS pumps would be on RHR pump operations following a LLOCA. The final ICCDP for this condition over a year of exposure was on the order of low E-8 based on the Phase 2 SDP assessment documented below in Section 2.6. This was a significant reduction, in the range of two orders of magnitude, from the initially assumed risk of the condition. This was because the air void was determined not to have impacted the SI and CHS pumps, as was originally assumed. Functionality of the SI and CHS pumps allowed crediting the HPI safety function and, in conjunction with RSS, the HPR safety function to prevent core damage in the event of a LLOCA.

2.6 Review of Technical Specification Compliance - Gas Void Discovered in the ECCS Suction Line - (Closed) LER 05000423/2008004-00

a. Inspection Scope

The team reviewed Dominions compliance with applicable TS and the licensee event report submitted subsequent to the discovery of the air void in the ECCS suction piping, which reported a condition prohibited by TS, in that, given the identified gas void, the RHR pumps were not operable to respond to a LLOCA.

b. Findings and Observations

Introduction:

The team identified a Green, self-revealing, non-cited violation (NCV) of TS 3.5.2.d which requires an operable (RHR) pump for each train of the ECCS. The team found that Dominion did not maintain the 24-inch piping connecting the RWST to the suction of the ECCS pumps sufficiently full of water to ensure operability of the RHR pumps following a LLOCA.

Discussion: The team reviewed Dominions assessment of the operability of ECCS suction piping and associated ECCS pumps following the identification of the 3.5-inch gas void in the 24-inch piping that connects the RWST to the suction headers of the ECCS pumps. The team observed that Dominion assumed a conservatively high 63 scf void for their analysis and used the RELAP 5, Mod 3.3 thermo-hydraulic code to model the piping and the void transport to the suction of the ECCS pumps for LLOCA, MLOCA, and SLOCAs. The resulting estimated dynamic near the suctions for both RHR pumps following system actuation in response to a LLOCA exceeded the 5% limit provided by the pump manufacturer and the NRC staff criteria (ML083250536) of 10%

for 5 seconds limit. Additionally, Dominion evaluated that the SI and CHS pumps would not have been impacted for any LOCA scenarios nor would the RHR pumps be impacted for SLOCAs or MLOCAs. The team reviewed Dominions analysis and agreed with the assessment. Given the presence of the air void, absent any specific pump testing to prove operability, the team determined that the RHR pumps were inoperable for response to a LLOCA, with respect to the pump manufacturer and NRC suction limits.

The team also reviewed procedures SP 3610A(B)-3, RHR System Venting and Valve Lineup - Train A(B) performed to meet the intent of TS surveillance 4.5.2.b. This TS surveillance requires that Dominion verify every 31 days that the ECCS piping is verified full of water. The team concluded that this procedure was inadequate because Dominion had assumed that the as-built drawings were correct and, therefore, did not verify portions of piping that they believed self-vented back to the RWST were full of water. Because of this assumption the portion of piping where the air was found had not been checked by the procedure.

Finally, the team found the corrective actions to install and use a vent valve in the 24-inch piping located in the ESF building and to conduct confirmatory UT verification that the piping is full, were acceptable to prevent recurrence.

Analysis:

The performance deficiency was a failure to maintain the common ECCS suction piping sufficiently full of water, as required by TS surveillance 4.5.2.b, to ensure RHR pump operability in the event of a LLOCA, as required by TS 3.5.2.d. The finding was more than minor because it was similar to NRC IMC 0612, Appendix E, Examples of Minor Issues, Example 2e. Specifically, vendor limits for allowable RHR pump suction flow air void fraction of 5% would have been exceeded, in the event of a LLOCA accident, which would have adversely impact the pumps safety function. The finding is associated with the design control attribute of the Mitigating Cornerstone and affected the cornerstone objective of ensuring the availability, reliability, and capability of systems that respond to initiating events to prevent undesirable consequences. Traditional enforcement does not apply because the issue did not have any actual safety consequences or potential for impacting the NRCs regulatory function, and was not the result of any willful violation of NRC requirements. In accordance with NRC IMC 0609, 4, "Phase 1 - Initial Screening and Characterization of Findings," the Phase 1 screening identified that this issue was a design/qualification deficiency which resulted in the loss of the RHR system LPI safety function and required a Phase 2 evaluation.

In accordance with IMC 0609, Appendix A, Determining the Significance of Reactor Inspection Findings for At-Power Situations, a Region I SRA determined that the finding was of very low safety significance (Green) using a modified Phase 2 analysis and the MP3 plant-specific Phase 2 Notebook worksheet for a LLOCA. The modified Phase 2 assessment was needed to account for the assumed complete loss of the low pressure injection (LPI) safety function of RHR. The following assumptions were made during the evaluation: 1) the LPI safety function of RHR would fail due to air ingestion following a LLOCA; 2) the exposure period was one year (>30 days). This assessment resulted in an increase in the core damage frequency on the order of low E-8 per year, which was dominated by the LLOCA frequency of E-5 per year and the probability of HPI failure, due to some other unrelated cause. The SI, CHS, and RSS systems were still available to prevent core damage following a LLOCA initiating event, by performing the HPI and HPR safety functions.

The finding did not have a crosscutting aspect, because it represented a latent design issue and there was not a reasonable opportunity to identify the issue within the recent performance guideline period.

Enforcement:

TS 3.5.2.d requires one operable RHR pump for each train of ECCS.

Contrary to the above, on October 20, 2008, Dominion identified that an air void existed in the suction piping of the RHR pump which caused both trains of RHR pumps to be inoperable. Dominion concluded, and the team agreed, that the air was in the pipe for several years. Dominion corrected the latent design error by installing a vent valve and refilling the piping on October 28, 2008. Because this violation is of very low safety significance and has been entered into the licensees corrective action program (CR 115088), this violation is being treated as a non-cited violation, consistent with Section VI.A.1 of the NRC Enforcement Policy: (NCV 05000423/2008010-01, RHR pumps inoperable in the event of a LLOCA, due to a suction air void)

4OA2 Identification and Resolution of Problems (IP 71152)

a. Inspection Scope

The team reviewed problems that Dominion had identified and entered into their corrective action program associated with this issue. The team reviewed these issues to verify an appropriate threshold for identifying issues and to evaluate the effectiveness of corrective actions. In addition, CRs written on issues identified during the inspection were reviewed to verify adequate problem identification and incorporation of the problem into the corrective action system. The specific corrective action documents that were sampled and reviewed by the team are listed in the attachment.

b. Findings

No findings of significance were identified.

4OA3 Event Followup

(Closed) LER 05000423/2008-004-00, Gas void in ECCS Suction Piping. The team reviewed the LER associated with this event submitted on December 19, 2008 and did not identify any additional issues other than the NCV discussed in section 2.6 of this report. This LER is closed.

4OA6 Meetings, Including Exit

On December 19, 2008, and January 29, 2009, the team conducted a debrief meeting with Mr. A. J. Jordan and other members of his staff to discuss the status of the teams inspection activities, to date. On February 6, 2009, the team presented the inspection results to Mr. A. J. Jordan and other members of his staff. Proprietary information that was reviewed during the inspection, as noted in Attachment A to this report, was returned to Dominion or destroyed.

ATTACHMENT A

SUPPLEMENTAL INFORMATION

KEY POINTS OF CONTACT

Licensee Personnel

J. Jordan Site Vice President

G. Closius Licensing Engineer

A. Ghanakhanian Safety Analyst

M. Van Haltern Mechanical Engineering

R. DeConto Mechanical Engineering

LIST OF ITEMS OPENED, CLOSED, AND DISCUSSED

Opened/Closed

05000423/2008010-01 NCV RHR pumps inoperable in the event of a LLOCA, due to a suction air void.

Closed

05000423/2008-004-00 LER Gas void in ECCS suction piping