IR 05000382/2006009

IR 05000382-06-009 on 03/06/2006-08/09/2006 for Waterford Steam Electric Station, Unit 3; Discrepant or Unreported Performance Indicator Data and Notice of Violation

ML062420513
Person / Time
Site:	Waterford
Issue date:	08/30/2006
From:	Howell A T NRC/RGN-IV/DRP
To:	Venable J E Entergy Operations
References
EA-06-136 IR-06-009
Preceding documents:	ML062930302 (18 October 2006)
Download: ML062420513 (33)

Text

August 30, 2006

EA-06-136 Joseph E. VenableVice President Operations Waterford 3 Entergy Operations, Inc.

17265 River Road Killona, LA 70066-0751

SUBJECT: WATERFORD STEAM ELECTRIC STATION, UNIT 3 - NRC INSPECTIONREPORT 05000382/2006009 AND NOTICE OF VIOLATION

Dear Mr. Venable:

On August 9, 2006, the NRC completed an inspection at your Waterford Steam Electric Station,Unit 3. The enclosed report documents the inspection findings which were preliminarily discussed on May 11, 2006, with Mr. Kevin Walsh and other members of your staff. A followup call concerning the disposition of the related enforcement activities was conducted with you and other members of your staff on June 23, 2006, and a final telephonic exit meeting was conducted on August 9, 2006.This inspection examined activities conducted under your license as they relate to safety andcompliance with the Commission's rules and regulations and with the conditions of your license.

Within these areas, the inspection consisted of selected examination of procedures and representative records, observations of activities, and interviews with personnel.The report documents one violation of 10 CFR 50.9, with two examples, for reporting inaccurateinformation to the NRC associated with the NRC's Safety System Unavailability (High Pressure Injection and Residual Heat Removal) Performance Indicators. The performance indicator information was inaccurate because your staff improperly concluded that the Train B high pressure safety injection and Train B containment spray systems were still available for the fullrange of potential accidents during an extended period when the containment safety injectionsump suction valve was partially open during the period of November 11, 2003 to September 9, 2004. If this data had been accurately reported, then the High Pressure Injection Performance Indicator should have been Red and the Residual Heat Removal Performance Indicator should have been Yellow up until the point that the safety system unavailability performance indicatorswere replaced by the mitigating systems performance index (MSPI) on April 1, 2006.The Enforcement Policy, Supplement VII, specifies that issues such as reporting inaccurateperformance indicator data that would have caused a performance indicator to change color from Green to Yellow or Red would be a Severity Level III violation. However, the NRC hasdetermined that a Severity Level IV violation was more appropriate in this particular circumstance. This determination was made primarily on the basis that the risk significance Entergy Operations, Inc.-2-associated with the valve being mis-positioned was determined to be very low (Green) asdocumented in NRC Inspection Report 05000382/2004005. Additionally, under the NRC'srecently implemented MSPI program, which replaced the safety system unavailabilityperformance indicator program on April 1, 2006, a similar set of circumstances would have resulted in a Green outcome because of the differences in the way fault exposure is treated under the former safety system unavailability performance indicator program and the currentMSPI program. While the overall risk significance of the underlying performance deficiency was low, the failure to provide accurate information calls into question the ability to effectivelyand consistently implement the performance indicator program, particularly for circumstances in which the performance indicator outcome may result in a degradation of the cornerstone safety objective.The Severity Level IV violation is being cited because not all the criteria specified in SectionVI.A.1 of the NRC Enforcement Policy for a noncited violation were satisfied. Specifically,Entergy failed to restore compliance and report corrected or accurate performance indicator information, or otherwise notify NRC that the data reported in the past was not accurate, withina reasonable time after the potential violation was initially identified. For the purposes of thiscriterion, you were notified of the violation during the May 11, 2006, exit debrief meeting. Eventhough the safety system unavailability performance indicators were superceded by MSPIbefore the date of the exit meeting, the failure to accurately report the data was still materialbecause the NRC used the information to determine its level of oversight of Waterford 3.

Additionally, as stated in the Enforcement Policy, 10 CFR 50.9 violations are not evaluatedunder the Significance Determination Process because violations of this nature impact the ability of the NRC to properly regulate its licensees. You are required to respond to this letterand should follow the instructions specified in the enclosed Notice when preparing your response. The NRC will review your response, in part, to determine whether furtherenforcement action is necessary to ensure compliance with regulatory requirements.The NRC Action Matrix is described in NRC Inspection Manual Chapter 0305, "OperatingReactor Assessment Program." In accordance with the Action Matrix, one Red and one Yellow performance indicator input would have placed your facility in the "Multiple/Repetitive DegradedCornerstone," column. This would have required the implementation of Inspection Procedure 95003, "Supplemental Inspection for Repetitive Degraded Cornerstone, Multiple Degraded Cornerstones, Multiple Yellow inputs, or One Red Input," at your facility, as well as a number ofadditional regulatory actions. For the same reasons as those specified for categorizing the violation at Severity Level IV, the NRC also determined that a deviation from the NRC ActionMatrix was warranted. In lieu of these actions, we will implement actions as specified in the Licensee Response column because, in this circumstance, this will provide the appropriate levelof regulatory response. This deviation from the NRC Action Matrix is documented in amemorandum dated July 25, 2006.

Entergy Operations, Inc.-3-In accordance with 10 CFR 2.390 of the NRC's "Rules of Practice," a copy of this letter, itsenclosures, the July 25, 2006, Action Matrix Deviation Memorandum, and your response will bemade available electronically for public inspection in the NRC Public Document Room or fromthe NRC's document system (ADAMS), accessible from the NRC Web site at http://www.nrc.gov/reading-rm/adams.html. Should you have any questions concerning thisinspection, we will be pleased to discuss them with you.

Sincerely,/RA/Authur T. Howell III, DirectorDivision of Reactor ProjectsDocket: 50-382License: NPF-38

Enclosures:

Notice of Violation NRC Inspection Report 050000382/2006009

w/Attachment:

Supplemental Informationcc w/

Enclosure:

Senior Vice President and Chief Operating Officer Entergy Operations, Inc.

P.O. Box 31995 Jackson, MS 39286-1995Vice President, Operations SupportEntergy Operations, Inc.

P.O. Box 31995 Jackson, MS 39286-1995Wise, Carter, Child & Caraway P.O. Box 651 Jackson, MS 39205General Manager, Plant OperationsWaterford 3 SES Entergy Operations, Inc.

17265 River Road Killona, LA 70066-0751 Entergy Operations, Inc.-4-Manager - Licensing ManagerWaterford 3 SES Entergy Operations, Inc.

17265 River Road Killona, LA 70066-0751ChairmanLouisiana Public Service Commission P.O. Box 91154 Baton Rouge, LA 70821-9154Director, Nuclear Safety & Regulatory Affairs Waterford 3 SES Entergy Operations, Inc.

17265 River Road Killona, LA 70066-0751Richard Penrod, Senior Environmental Scientist Office of Environmental Services Northwestern State University Russsell Hall, Room 201 Natchitoches, LA 71497Parish President St. Charles Parish

P.O. Box 302 Hahnville, LA 70057Winston & Strawn LLP1700 K Street, N.W.

Washington, DC 20006-3817 Entergy Operations, Inc.-5-Electronic distribution by RIV:Regional Administrator (BSM1)DRP Director (ATH)DRS Director (DDC)DRS Deputy Director (RJC1)Senior Resident Inspector (MCH)Branch Chief, DRP/E (DNG)Senior Project Engineer, DRP/E (VGG)Team Leader, DRP/TSS (RLN1)RITS Coordinator (KEG)OEMAILD. Starkey, OE D. Solorio, OE J. Luehman, OE M. Vasquez M. Haire Only inspection reports to the following:DRS STA (DAP)J. Lamb, OEDO RIV Coordinator (SCO)ROPreports WAT Site Secretary (AHY)SUNSI Review Completed: __DNG__ADAMS: Yes G No Initials:_DNG____ Publicly Available G Non-Publicly Available G Sensitive Non-SensitiveR:\REACTORS\WAT\2006\WT2006-09RP-GDR.wpdRIV:SRI:DRS/EB1NRRSPE:DRP/DACESC:DRP/ED:DRPGDReplogleMAJungeMCHayGMVasquezDNGravesATHowell/RA/E-GDR/RA//RA//RA//RA/6/16/06 & 6/23/066/12/066/20/068/24/068/24/068/30/06OFFICIAL RECORD COPY T=Telephone E=E-mail F=Fax Enclosure 1NOTICE OF VIOLATIONEntergy Operations, Inc.Docket No. 50-382Waterford Steam Electric StationLicense No. NPF-38EA-06-136During an NRC inspection conducted on March 6 through May 11, 2006, a violation of NRCrequirements was identified. In accordance with the NRC Enforcement Policy, the violation islisted below: 10 CFR 50.9 requires, in part; "Information provided to the Commission by a licenseeshall be complete and accurate in all material respects." Contrary to the above, from approximately November 1, 2004, (when the licenseeinitially submitted the subject performance indicator information) to May 11, 2006, information provided to the Commission in the form of system unavailability statistics forthe high pressure safety injection (Train B) and containment spray (Train B) systemswas not complete and accurate in all material respects. The licensee significantly

under-reported the unavailability hours for each train. This was material because theNRC used the information to determine its response (e.g., inspection) to the data.This is a Severity Level IV violation (Supplement VII).

Pursuant to the provisions of 10 CFR 2.201, Entergy Operations is hereby required to submit awritten statement or explanation to the U.S. Nuclear Regulatory Commission, ATTN: DocumentControl Desk, Washington, DC 20555 with a copy to the Regional Administrator, Region IV, and a copy to the NRC Resident Inspector at the facility that is the subject of this Notice, within 30days of the date of the letter transmitting this Notice of Violation (Notice). This reply should be clearly marked as a "Reply to a Notice of Violation; EA-06-136" and should include: (1) the reason for the violation, or, if contested, the basis for disputing the violation or severity level,(2) the corrective steps that have been taken and the results achieved, (3) the corrective stepsthat will be taken to avoid further violations, and (4) the date when full compliance will beachieved. Your response may reference or include previous docketed correspondence, if the correspondence adequately addresses the required response. If an adequate reply is not received within the time specified in this Notice, an order or a Demand for Information may be issued as to why the license should not be modified, suspended, or revoked, or why such other action as may be proper should not be taken. Where good cause is shown, consideration willbe given to extending the response time. If you contest this enforcement action, you should also provide a copy of your response, withthe basis for your denial, to the Director, Office of Enforcement, United States Nuclear Regulatory Commission, Washington, DC 20555-0001. Because your response will be made available electronically for public inspection in the NRCPublic Document Room or from the NRC's document system (ADAMS), accessible from the NRC Web site at http://www.nrc.gov/reading-rm/adams.html, to the extent possible, it shouldnot include any personal privacy, proprietary, or safeguards information so that it can be made available to the public without redaction. If personal privacy or proprietary information is necessary to provide an acceptable response, then please provide a bracketed copy of your

-2-Enclosure 1response that identifies the information that should be protected and a redacted copy of yourresponse that deletes such information. If you request withholding of such material, you mustspecifically identify the portions of your response that you seek to have withheld and provide in detail the bases for your claim of withholding (e.g., explain why the disclosure of information willcreate an unwarranted invasion of personal privacy or provide the information required by 10 CFR 2.390(b) to support a request for withholding confidential commercial or financial information). If safeguards information is necessary to provide an acceptable response, please provide the level of protection described in 10 CFR 73.21. Dated this 30th day of August, 2006 Enclosure 2U.S. NUCLEAR REGULATORY COMMISSION REGION IV Docket No.:50-382 License No.:NPF-38 Report No.:05000382/2006009 Licensee:Entergy Operations, Inc.

Facility:Waterford Steam Electric Station, Unit 3 Location:Hwy. 18 Killona, Louisiana Dates:March 6 through August 9, 2006 Inspectors:G. D. Replogle, Senior Reactor Inspector, Division of Reactor SafetyM. A. Junge, Mechanical Engineer, Office of Nuclear Reactor RegulationApproved By:Arthur T. Howell III, Director, Division of Reactor ProjectsATTACHMENTS:Supplemental Information Enclosure 2-2-

SUMMARY OF FINDINGS

IR05000382/2006-009; 03/06/2006-08/09/2006; Waterford Steam Electric Station, Unit 3;Discrepant or Unreported Performance Indicator DataThe report covered a 5 month period of inspection by a Region IV senior reactor inspector anda mechanical engineer from the NRC's Office of Nuclear Reactor Regulation. The inspectoridentified one Severity Level IV violation, which is not subject to the Significance Determination

Process. Findings for which the Significance Determination Process does not apply may be Green or be assigned a severity level after NRC management review. The NRC's program foroverseeing the safe operation of commercial nuclear power reactors is described in NUREG-1649, "Reactor Oversight Process," Revision 3, dated July, 2000. A.

Cornerstone: Mitigating Systems

Severity Level IV. The inspector identified a violation of 10 CFR 50.9, with twoexamples, for the failure to provide accurate information to the NRC associated with the Safety System Unavailably High Pressure Injection and Residual Heat Removal Performance Indicators. The performance indicator information was inaccurate because the licensee improperly concluded that the Train B high pressure safety injection and Train B containment spray systems were still available during an extended period whenthe containment safety injection sump suction valve was partially open. The inspector found that the licensee had underestimated the size of valve (SI 602B) opening when assessing system availability and failed to address inconsistencies between their fielddata, diagnostic test data and their own informal calculations. Further, a second analysis performed by a contractor (to determine the as-found valve position) was inadequate, as it contained several errors and inappropriate assumptions. The licensee also provided inadequate contractor oversight with respect to this effort. The erroneous valve position determination resulted in the licensee reporti ng system availabilityinformation that caused the performance indicators to be Green when the High Pressure Safety Injection System Unavailability Performance Indicator should have been Red andthe Residual Heat Removal System Unavailability Performance Indicator should havebeen

Yellow.

The failure to provide accurate information to the NRC in accordance with 10 CFR 50.9requirements was a performance deficiency. The issue had more than minor significance in that, had the information been accurate, two performance indicators would have changed color. Per the NRC Enforcement Policy,Section IV.A.3, theseissues are not subject to the Significance Determination Process. The Enforcement Policy, Supplement VII, specifies that a Severity Level III violation would be appropriatefor these issues. However, considering: 1) the NRC's recently implemented Mitigating Systems Performance Index program, which would have resulted in the subject performance indicators returning to the Green threshold; and 2) the risk associated with 2-3-the underlying valve performance issue was of very low safety significance (Green), theNRC determined that a Severity Level IV violation was more appropriate. This findinghad problem identification and resolution crosscutting aspects, in that theimplementation of the licensee's Corrective Action Program did not result in a thorough evaluation of the identified condition such that information reported to the NRC wasverified to be complete and accurate (Section 4OA1).

B.Licensee-Identified Violations

None.

2-4-

REPORT DETAILS

4OA1Discrepant or Unreported Performance Indicator Data (71150)

a. Inspection Scope

The purpose of the inspection was to evaluate the accuracy of performance indicatorinformation submitted by the licensee for the High Pressure Injection and Residual Heat Removal Performance Indicators. The accuracy of the performance indicator information was questioned after an operator found containment safety injection sump suction Valve SI-602B partially open on September 9, 2004. This condition could have affected the availability of the Train B high pressure safety injection and Train Bcontainment spray systems. The licensee determined that the valve had been partiallyopen since November 11, 2003, when valve technicians incorrectly set one of the Valve SI-602B limit switches. The partially open valve posed potential challenges to the availability of the noted trains in response to medium break and large break loss ofcoolant accidents. The availability of the high pressure safety injection system affectedthe Safety System Unavailability High Pressure Injection Performance Indicator while availability of the containment spray system affected the Safety System Unavailability Residual Heat Removal Performance Indicator. The inspector reviewed the licensee's evaluation of the as-found valve position, the rootcause evaluation report and two contractor-furnished analyses. One contractor had performed an evaluation of system impacts based on different valve positions. The second contractor performed an analysis to determine the valve's as-found position. In addition, the inspector interviewed plant personnel who were involved with issue identification and problem evaluation. Finally, the inspector obtained motor-operatedvalve drawings and valve test data and performed independent calculations to determine the valve's as-found position.The underlying valve performance issue was documented as a licensee-identifiedviolation of very low risk significance (Green) in NRC InspectionReport 05000382/2004005, Section 4OA7. Therefore, the focus of this inspection dealt solely with the determination of the accuracy of the data submitted to the NRC regardingthe two subject performance indicators.

b. Findings

Introduction.

The inspector identified a violation of 10 CFR 50.9, with two examples, forthe failure to provide accurate information to the NRC associated with the High Pressure Safety Injection System Unavailability and Residual Heat Removal System UnavailabilityPerformance Indicators. The performance indicator information was inaccurate because the licensee improperly concluded that the Train B high pressure safety injection and Train B containment spray systems were still available during an extended period whenthe containment safety injection sump suction Valve SI-602B, was partially open. The inspector found that the licensee had underestimated the size of the valve opening whenassessing system availability and failed to address inconsistencies between their field 2-5-data, diagnostic test data and their own informal calculations. Furthermore, a secondanalysis performed by a contractor to determine the valve's position was inadequate, as it contained several errors and unsupported assumptions. The licensee provided inadequate contractor oversight with respect to this effort. The erroneous valve position determination resulted in the licensee reporting system availability information thatcaused the two performance indicators to indicate Green when the High Pressure Injection System Unavailability Performance Indicator should have been Red and theResidual Heat Removal System Unavailability Performance Indicator should have beenYellow.Background: On September 9, 2004, during a system piping integrity test, the testoperator found the containment safety injection sump suction Valve SI-602B partiallyopen. The valve is a 24 inch Fisher 9200 series butterfly valve driven by an SMB-00 electric actuator through a quarter turn H2BC gear operator. The licensee determined that the condition had existed since November 11, 2003, when one of the valve limitswitches was set incorrectly and prevented the valve disc from reaching the fully closed and seated position. In response to medium and large break loss of coolant accidents that pressurizecontainment, the partially open valve posed two potential challenges to systemavailability. First, the pressurized containment would force air, steam and water throughthe valve and into the system flow streams. Air entrainment can cause excessive pumpvibration and pump air binding which can render the system unavailable. Second, thesystem suction lines could be pressurized sufficiently as to force and hold closed therefueling water storage pool discharge check Valve SI-107B. This would starve the system pumps of a suction source of water, rendering them unavailable. Depending onthe position of SI-602B, this condition had the potential of rendering the high pressure safety injection system Train B, low pressure safety injection system Train B, andcontainment spray system Train B unavailable. However, in this case, the unavailabilityof the Train B low pressure safety injection system would not affect a performance indicator value. The licensee performed a detailed analysis to determine the equipment's sensitivity toair entrainment and suction header pressurization. The licensee determined that the total valve opening could not exceed 0.41 square inches. In addition, the licensee attempted to replicate the valve position and then took physical gap measurements todetermine the affected valve disc area. The licensee determined that the open areawas 0.25 square inches, which was within the 0.41 square inch acceptance limit. The licensee then took additional valve measurements and determined that the valve's as-found position was approximately 3.2 open. In response to the NRC questions andconcerns regarding the valve's actual as-found position, the licensee contracted with an industry engineering firm to again determine the as-found valve position.Summary: The following bullets summarize the inspector observations and findings. Additional details concerning each identified problem are provided in subsequentsections of this report.

2-6-*The licensee's evaluation of system impacts, based on different possible valvepositions, was conservative and effective. The evaluation concl uded that systemavailability could be maintained with a maximum valve opening of 0.41 squareinches. The inspector had no findings of significance related to this evaluation.*The licensee's re-creation of the as-found position of Valve SI-602B wasinadequate because it failed to account for direction-dependent differences in valve performance. Specifically, valve leakage started and stopped at different handwheel positions depending on whether the valve was being opened or closed. The licensee's re-creation concluded that the valve was 3.2 open.*A subsequent engineering evaluation performed by a contractor was inadequatebecause:The contractor's evaluation of the licensee's re-creation of the valve's as-foundposition also failed to account for direction-dependent differences in valve performance. The licensee had not provided the contractor with the pertinent test data. This affected one of five contractor valve position assessments.The contractor inappropriately assumed that the valve was over-traveling past the 0 position. The contractor provided no objective evidence to support thisassumption and the inspector identified evidence to the contrary. The two calculations that used this assumption concluded that the valve was 3.4 and 3.5 from the 0 position. However, when the inspector applied correctionfactors to these calculations to account for the erroneous assumption, the calculated valve positions were 5.4 and 5.6, respectively.The contractor used an inappropriate value for the "Modulus of Rigidity" (anengineering term related to the stiffness of a round bar when torque is applied) in one calculation. This made two calculations that were based on different values of handwheel turns (0.55 and 1.0) come to the same final result. This inappropriate factor was used in one calculation but not the other.The contractor's estimation of valve position based on HBC operator stop nutmeasurements was not credible. The data, which was provided by the licensee,was obtained under informal and uncontrolled conditions, was not specific as to the conditions under which it was obtained, and was inconsistent with other information. This affected one contractor calculation.*The licensee reviewed the contractor's analysis prior to providing it to the NRCbut they failed to identify the noted problems. *The inspector performed two additional independent calculations and determinedthat the valve's as-found position was slightly greater than 5 from the zeroposition. The licensee had performed similar calculations and had obtained similar results, but had not formalized these efforts. In addition, the licensee had 2-7-not attempted to explain the inconsistencies between these calculations andother results that they utilized. The licensee's evaluation of these issues lacked sufficient engineering rigor.A summary of the calculated valve positions, including inspector corrected values andresults from independent NRC calculations, is provided below:Contractor Calculations and NRC CorrectionsCalculation Method/ Referenced PositionChange in DiscPosition from Referenced PositionInspectorCorrected ValuesHBC Stop Nut2.915Not Credible,Inconsistent DataHandwheel Closure (one fullhandwheel turn)2.8703.9Least AccurateStroke Time Comparison3/10/1999 Trace (diagnostic system traces)3.3635.4Reasonable ApproximationStroke Time Comparison11/6/2003 Trace (diagnostic system traces)3.5495.6Reasonable ApproximationDisc Measurement3.186Inadequate Re-creationAverage 3.177Independent NRC CalculationsMethodResultsRelativeAccuracyInservice Stroke Time Tests 5.9ReasonableApproximationDiagnostic System Test Data5.5Most AccurateThe detailed supporting sections of this report are organized into three main categories:*Assessment of Licensee's Initial Evaluation*Assessment of Licensee's Contractor Evaluation

• Independent NRC Calculations 2-8-Assessment of Licensee's Initial EvaluationLicensee's First Field Test: The licensee performed field testing to estimate thevalve's as-found position.The first field test was a local leak rate test. The test involved slowly opening the valveusing the manual handwheel in 1/8 turn increments to the 2-1/4 turns open position. Then an operator closed the valve in 1/8 turn increments until it was fully closed. The valve operator stated that he did not start counting the handwheel turns until heobtained resistance. That means that the operator did not count the first part of thehandwheel operation before the handwheel dogs engaged into the drive sleeve slots.

Until the handwheel dogs are engaged, the handwheel would spin freely. The inspector considered this approach appropriate for counting handwheel turns.The inspector noted that the apparent positions of the valve when leakage started(going open) and stopped (going closed) were sensitive to the direction of valve movement. For example, in the open direction gross leakage initiated between 1-3/8 and 1-1/2 handwheel turns from hard-seat contact. In the closing direction, gross leakage stopped between 7/8 and 3/4 handwheel turn from hard-seat contact. In response to inspector's questions, the licensee stated that they had also noticed this discrepancy but had not attempted to evaluate it.The inspector reviewed the valve's diagnostic testing system torque traces anddetermined that the significant difference between the points where leakage started andstopped was due to gear play. The traces revealed that, when the valve changed direction, the motor and gears operate for 1.34 seconds before the valve stem started tomove. In the more pertinent part of the trace (from "hammer-blow" to the initiation of valve movement) the time delay was 0.8 seconds. This is the more pertinent trace section because it mimics the valve/actuator response when it is operated with the handwheel. The delay of 0.8 seconds equates to about 0.6 handwheel turn worth of motion, which turns out to be the approximate difference in the handwheel positions noted above. In reality, for both the opening and closing directions the valve disc was in about thesame position when gross flow initiated and stopped. The indicated position in the open direction was misleading because the licensee did not account for the gear play contribution. Accounting for this type of gear play is only necessary when changing the valve's direction and will usually only affect the estimated position (using handwheelturns) in the open direction.The table, below, illustrates the affect of gear play on the valve position. In thisexample, the valve in question is repositioned from full closed to 1.8 handwheel turns open. Then the valve is repositioned back to the closed position. For this example, one full handwheel turn is equal to 5.14 of valve movement.

2-9-OpeningHandwheelturns from fully closed, going openActual valveposition in equivalent handwheel turns from fully closedValveposition in degrees open0000.2000.400 0.6000.80.21.01.00.42.01.20.63.01.40.84.11.61.05.11.81.26.2 ClosingHandwheelturns from fully closed, going closedActual valveposition in equivalent handwheel turns from fully closedValveposition in degrees open1.81.26.21.61.26.21.41.26.21.21.26.21.01.05.10.80.84.10.60.63.00.40.42.00.20.21.0000As demonstrated above, when the valve is being repositioned in the open direction,handwheel position is not an accurate indicator of valve position unless the amount of gear play is known and accounted for. Conversely, moving in the closed direction, once the valve moves through the gear-play region, the handwheel position is a relatively accurate indicator of valve position. This fact is important and has direct ramifications on the licensee's evaluation as discussed below.

Licensee's Second Field Test: The second field test involved positioning the valve tothe one handwheel turn open position and taking internal gap measurements. The gapmeasurements were then used to estimate the approximate valve opening. For clarity, each major step is discussed separately:1.The licensee opened the valve one full handwheel turn, against resistance, andmatch-marked the valve's position indicator. The indicator is directly aligned with the valve stem. It is connected to the stem through a splined shaft, which has a small amount of gear play. The inspector noted that because of the position error in the opening direction(due to gear play), the valve was not the equivalent of one full handwheel turn open (5.12) but was about 0.4 handwheel turn open (2). 2.An operator opened the valve to the 40 percent open position (about 34).

2-10-3.The operator then closed the valve to align the match-markings that were madein the first step. The inspector noted that the licensee had not counted thenumber of turns back to this position to verify that the valve was actually one handwheel turn from the seat. Instead, the licensee aligned the valve indicator to the match marks that corresponded to the 0.4 handwheel

(2) open position. If there had been no gear play between the indicator and the valve stem, then the valve would have been only 2 open. This gear play between the indicatorand the valve stem caused the indicator to lead the actual valve position very slightly. When reversing the valve's direction (going in the closed direction) the gear play would cause the valve to be slightly more open at the match-marked position than when the match-marks were made. Since this gear play constituted only a portion of the total gear play in the valv e/actuator system, itwas not sufficient to cause the valve to be the equivalent of one full handwheel turn open (5.12). 4.The licensee took gap measurements between the valve disc and body.

5.The licensee took additional internal measurements and established that thevalve disc was approximately 3.2 open. 6.The licensee repeated the procedure and calculated the total valve open area at0.25 square inches (using the larger of the two measurements). The two measurements were approximately 50% different in magnitude. The licensee identified small gaps between the Tee-Ring and the seat in one lower quadrant and in a small section of one upper quadrant, where the Tee-Ring had torn.The inspector noted that, between the two tests, the licensee had to reposition the valveto the full closed position in order to start the procedure over for the second set of measurements. The licensee could have easily verified that the valve was actually one full handwheel turn from hard-seat contact at this point in their procedure. The licensee did not perform this verification. If they had done so, then they would have found that the valve was not one full handwheel turn away from the fully closed position.Assessment of Licensee's Contractor EvaluationAs discussed later in this report, the inspector had performed independent calculationsand determined that the valve was slightly more than 5 open. In response to theinspector concerns, the licensee hired a contract engineering firm to evaluate the valve's as-found position. The licensee provided the inspector's with the contractor's report, "Independent Assessment of Disc Position of SI-602B Between November 3, 2003 and September 11, 2004," dated May 8, 2006. The engineering firm determined that the valve was likely at the 3.2 open position,consistent with the licensee's claims. The engineering firm calculated the valve's position using four different methods, five estimations in all. However, the inspector identified several inappropriate assumptions and errors in the analysis. The following table summarizes the vendor's results as well as the inspector corrected values:

2-11-Calculation Method/ Referenced PositionChange in DiscPosition from Referenced PositionInspectorCorrected ValuesHBC Stop Nut2.915Not Credible,Inconsistent DataHandwheel Closure (onefull handwheel turn)2.8703.9Stroke Time Comparison3/10/1999 Trace (diagnosti c systemtraces)3.3635.4Stroke Time Comparison11/6/2003 Trace (diagnosti c systemtraces)3.5495.6Disc Measurement3.186InadequateRe-creationAverage3.177The inspector findings are discussed in detail below:HBC Stop Nut Method: This method is based on gap measurements between one ofthe HBC operator stop nuts and the adjacent enclosure wall. The stop nuts are used inthe HBC-2 operator to limit valve travel. While the licensee did not utilize the stop nutsfor this purpose, the gap measurements provided some information with respect to the as-found valve position. Most importantly, the stop nut is on the same shaft as the handwheel. Stop nut movement correlates directly with handwheel movement (when the handwheel is engaged). Since the stop nuts had seven threads per inch, one handwheel turn equates to 143 mils in stop nut travel.The licensee had stated that when the valve was checked, after the operator had closedit, the as-found stop nut gap measurement was 70 mils. After the valve was opened and closed electrically (with the inappropriate limit switch setting) the gap measurement was 150 mils. The licensee then adjusted the valve limit switch until the stop nut measurement was 71 mils. The licensee provided this data to the engineering contractor.The contractor calculated that the operator had turned the valve handwheel 0.55 turnswhen he found it. Considering stem deflection (based on the torque at the final valve position being 965 foot-lbs) the contractor's calculated as-found valve position was 2-12-2.915 open. The contractor's report indicated that the 0.55 handwheel turns appearedinconsistent with the operator's statements (that he had closed the valve approximatelyone full handwheel turn).The inspector noted the following problems and inconsistencies with the data and thecontractor's assessment.

• As noted by the contractor, gap measurements did not correlate with operatorstatements. The HBC stop nut movement should correlate exactly with handwheel movement. The components are on the same shaft. But, the gap measurements indicated that the operator had only turned the valve 0.55handwheel turn while the operator had indicated that he had repositioned the valve approximately 1 full handwheel turn.*The HBC gap measurement taken after the operator closed the valve was70 mils. The gap measurement taken after the valve limit switch was adjusted so that the stop nut gap approximated the same position was 71 mils. These two gap measurements are almost exactly the same. Accordingly, the torque on the valve stem that coincided with each measurement should have been about the same. They were not. The torque value after the operator closed the valve was 965 ft-lbs while the torque value measured after the limit switch was adjusted was 454 ft-lbs.*The gap measurements were not performed in accordance with procedures andthe existing documentation was unclear with respect to what was measured and when it was measured. The licensee had stated that they took gap measurements after each of 15 limit switch adjustments (when they were attempting to properly adjust the value's limit switch on September 11, 2004).

However, no data was apparently recorded for 14 of the gap measurements.

When asked on more that one occasion for "any and all" documents associated with these gap measurements, the licensee provided only two documents that contained gap values. One document, Work Order 51581 stated, in part:After adjustment 145 mils Desired HW set 70 mils using ETT as ASswitched to sts. gap = 71The work order contained no steps to take these measurements and themeaning of some of the terminology was not clear. For example, 145 mils was obtained "after adjustment." The document does not state what adjustment was made. Additionally, there was no formally controlled record of the 150 mil value which the contractor used in the calculations (the 145 mil value was not used in the calculations). The 150 mil value was recorded in the computer system butnot in the hard copy of the work order.The second document, a print of a computer screen (print date May 5, 2006)stated, in part:WO# 51581 150 mils Desired HQ set at 70 mils.

2-13-While this document provided some of the same measurement values utilized bythe contractor in their calculations, the 150 mil value was inconsistent with thevalue actually documented in the work order (145 mils after adjustment). In addition, this value was contained in an uncontrolled computer file and was not subject to any quality controls such that it could have been changed in an uncontrolled manner.*The results from the stop nut measurements were inconsistent with results thatthe inspector obtained from other calculations, such as those based on diagnostic system torque traces and Inservice Testing stroke times. The timesignature data in the diagnostic system torque traces was of a high quality andwas very accurate with respect to the identification of artifacts and the time they occurred (provided later in this report).*A representative from Limitorque informed the inspector that the HBC stop nutthreads were 1-1/4 X 7 UNC - coarse. This means, in part, that the stop nut has7 threads per inch and they are coarse, versus fine or medium grade. The inspector also noted, while inspecting the threads on a much smaller HBC unit, that the stop nuts could move slightly from side to side (without any stop nut rotation). Limitorque had not designed the stop nuts to be used for the purpose of taking precise measurements for the purpose of estimating valve position.

They were only provided as a means to stop valve movement. Making precise measurements, to within a few mils, would be very difficult.Based on the above, the inspector had a low level of confidence in the accuracy of theHBC stop nut gap measurements and it was not clear that calculations based on thesevalues provided meaningful results. Compared to the quality of the other available data, such as the valve torque traces, the HBC stop nut measurements were not credible.Handwheel Turns: The contractor calculated the as-found valve position based on theoperator's statement that he had repositioned the valve approximately one fullhandwheel turn. The contractor based this calculation on 1.0 handwheel turn (exactly)and then deducted the contribution from stem deflection. Stem deflection is the amountof stem rotation observed without corresponding valve movement (the twisting of the stem due to torque). This would occur when the operator continued to turn the valve's handwheel after the valve disc had contacted the hard-seat. The inspector identified that the contractor's calculation for stem deflection utilized aninappropriate factor ("torsional stiffness") instead of the standard classical equation that is used for these purposes. This error affected the calculation sufficiently as to make the final result (2.9 from the backseat) appear the same as the calculation based onHBC stop nuts (the contractor had utilized the classical engineering equation in lieu ofthe torsional stiffness factor in the HBC stop nut calculation). Getting almost exactly the same value in the two calculations was unexpected because the stop nut calculationwas based on 0.55 handwheel turn worth of movement while the handwheel turn calculation was based on 1.0 handwheel turn worth of movement. Since the other factors should have been the same, the calculations should have produced different results.

2-14-The contractor derived the "torsional stiffness" factor from a portion of the diagnosticsystem torque trace between hard seat contact and motor-cuttoff. The contractor hadperformed no validation testing of the torsional stiffness factor to verify that it properly predicted system deflection. The stem deflection equati on used by the contractor was:Stem Deflection = Torque/Torsional StiffnessThe inspector noted that the "torsional stiffness" factor included deflection from all of theactuator/HBC gear trains (including a sliding worm gear in the Limitorque actuator) when only a much smaller portion of the gear train is in service when operating the valve with the handwheel. The inspector observed that the use of the torsional stiffness factor, in lieu of the classical equation, resulted in a higher calculated stem deflection value and a smaller valve opening for the handwheel calculation.Calculated as-found position = calculated position - stem deflectionFurthermore, the contractor acknowledged that the torsional stiffness factor was notlinear, as it appeared to change based on the amount of torque applied to the system. However, the contractor calculated and utilized the value as if it were linear. Thecontractor used the one calculated value for all circumstances.The inspector also observed that the contractor had calculated the torsional stiffness forthree different closing strokes and had obtained three different results, as provided below:431 ft-lbs/degree310 ft-lbs/degree358 ft-lbs/degreeThe relatively large degree of variability in this factor called into question its usefulnessin any of the stem deflection calculations. Any factor used to calculate stem deflection should demonstrate good repeatability. The contractor chose to use the310 ft-lbs/degree value, which also provided the most optimistic (lesser degree of valve opening) results for the "handwheel turn" calculation. The contractor calculated the stem deflection to be:

2.273The contractor then calculated the amount of valve opening based on this method as:

5.12 - 2.273 = 2.9 NOTE: The contractor also used the torsional stiffness term in their "stroke time"calculations. The effect of using this term in these calculations makes the calculated valve position more open, which is conservative. The magnitude of the adjustment, however, was much smaller, as a much smaller torque difference was applied. As noted later in this report, the contractor had used a different inappropriate assumption in these calculations which made the calculated valve position appear more closed.

2-15-Various text and reference books provide a classical equation for calculating the angleof deflection of a round bar with a given torque (for example, "Mechanical Engineering Design," Fourth Edition, Shigley and Mitchell). Since the stem torque was known (from a direct measurement on the stem) and the diameter, material and length of the stemwas also known, the contractor should have utilized the standard equation for thispurpose. Considering the configuration of the other components in the affected geartrain, deflection of the other components should be negligible. The equation, along with applicable values, is provided below: = Twist Angle in radians = (T*l)/(G*J)*T = Torque (use the torque difference in this case) = 705 ft-lbs*l = Length of Stem = 28 inches

• G = Modulus of Rigidity = 11,500,000 psi

• J = Polar Moment of Inertia = *d 4/32 = 0.92*d = Diameter of Stem = 1.75 inchesThe stem deflection based on the above is: 1.27, compared to the 2.273 used by thecontractor. The difference is:

2.27 - 1.27 = 1.0Adjusting the contractors calculation to account for this difference, the valve positionshould be:

2.9 + 1.0 = 3.9 position.In addition to the above, converting this data to the same units as used in thecontractor's "torsional stiffness" factor, the apparent torsional stiffness using the widely accepted engineering equations would be:555 ft-lbs/degree, which is substantially greater than the 310 ft-lbs/degree valueused by the contractor. While the inspector agreed that this calculation provided the valve's position based onthe operator's recollection, the operator had only estimated the amount of handwheelmovement at "approximately," one handwheel turn. The actual amount of travel could have been a little more or a little less. The inspector considered other calculations, based on diagnostic system torque traces, for example, to provide a much higher levelof accuracy and confidence. Stroke Time Measurements: The contractor performed two calculations based on thetiming of artifacts in the diagnostic system torque traces. The contractor referred tothese as stroke time measurements. These were similar to a calculation performed by the inspector in a later part of this report (Diagnostic System Test Data Method). They should not be confused, however, with another calculation performed by the inspector that was based on Inservice Testing stroke time measurements.

2-16-The theory behind the position estimation calculation is that by comparing the as-found(partially open valve) torque trace with a torque trace of the same valve where it did fully close (to the 0 position), the angle of the valve opening with respect to the 0 degreeposition can be calculated. The inspector agreed that this method of analysis should provide an accurate as-found valve position. The inspector found that instead of comparing the as-found torque trace to traces wherethe valve was at the fully closed position, indicated by the presence of hard seat contacton the trace, the contractor had compared the as-found torque trace to other valve traces where the valve did not reach the 0 position. This resulted in significantly under-estimating the valve's position. The contractor used the closing stroke dated March 10, 1999, and the closing stroke dated November 6, 2003, as reference traces.For the 1999 torque trace, the valve stopped significantly before the 0 position. Theinspector calculated the distance that the 1999 valve stroke was from the hard seat byutilizing a trace that was known to reach the 0 position, as evidence by a characteristicshape when the valve reaches the hard seat. Based on this approach, the inspector determined that the valve in the 1999 trace had stopped at the 1.9 position. The valvewas leaktight but was still not fully seated. The valve vendor's drawing indicated that thevalve could be leak tight out to the 3.2 position. To find the actual as-found valveangle, the inspector added this additional amount of travel to the contractors results.

3.36 + 1.9 = 5.4For the November 6, 2003, reference trace, the valve had traveled closer to the valveseat than in the 1999 reference trace. However, the contractor chose a point substantially before the valve had stopped moving as the reference point and treatedthis point as if it were the 0 position. This reference point was 0.57 seconds short ofhard seat contact. That equates to 2.15 degrees of additional travel. Adding thisamount to the contractors calculation resulted in:

3.549 + 2.15 = 5.6NOTE: For the two calculations above, the inspector neglected any additionaladjustment for stem deflection, as the torque in these traces was substantially smaller than in the handwheel case and any additional adjustment would be very small. Additionally, the stem deflection would cause the valve to be slightly more open (making the problem worse, not better).The inspector determined that the contractor had utilized an inappropriate assumptionwhen performing these calculations. The contractor had assumed that the valve stopwas not in the correct position, or was absent altogether, and that the valve was over-traveling past the 0 position. This would make all of the calculated valve openingangles smaller by the amount of over-travel. The contractor's report indicated that the valve was at the 0 position when it was at the first large torque peak and had furtherstated that this information came from the valve vendor. Some torque traces had two torque peaks (one at the 3.6 position and one at the 2.2 position). The inspectornoted that the contractor had no objective evidence to support this claim. In addition, the inspector disagreed with this contention for the following reasons:

2-17-*The inspector called the valve vendor and the vendor did not agree that the 0position was at the first large torque peak. In addition, the vendor indicated that for over-travel to occur, the valve stop set-screw, which is attached to a larger hard stop, would have to be misadjusted or missing. The vendor stated that the set-screw was either staked or welded into position at the factory. *The licensee had also informed the inspector that they had measured the gapbetween the set-screw stop and the larger hard stop during the past outage and had verified with the valve vendor that it was the proper size (0.25 inches). This indicated that the valve stop set-screw was in the correct position.*The inspector reviewed a similar valve trace for the identical valve in Train A,Valve SI-602A. The inspector noted that the valve displayed the same two hump-like features as were evident in the Valve SI-602B traces. If Valve SI-602B was over-traveling past the 0 position, as described by thecontractor, both valves were over-traveling. This was not likely, given the vendor information regarding set-screw installation.*As a bounding exercise, if the set-screw was totally missing, the valve could onlytravel an additional 0.25 inches at the outer edge. That equates to only anadditional 1.2 degrees of valve travel. The inspector calculated the angular distance between the large hump and the hardseat (without correction for stem deflection at 3.6). In other words, the hump was too far from the hard stop forthis theory to be credible.Independent NRC CalculationsThe inspector used the licensee's test data and two calculation methods to determinethe as-found position of Safety Injection Valve SI-602B. Based on the calculations, the inspector determined that the valve was open slightly more than 5, which wassignificantly more than the 3.2 assumed in the licensee's analysis. The following tableprovides a summary of the results:MethodResultsRelative AccuracyInservice Stroke Time Tests 5.9Reasonably AccurateDiagnostic System Test Data 5.5Most AccurateThe calculations are provided below:*NRC Calculation 1: Based on Inservice Stroke Time Tests: Since theinappropriate limit switch setting also affected the valve's stroke time, theinspector calculated the valve position based on the difference in the recorded stroke times. The inspector compared the average of the stroke times when the limit switch was set incorrectly to the average of the stroke times when the limit switch was set properly.

2-18-Stroke Times (Close Direction), Limit Switch set Incorrectly:11/7/200322.2 Seconds1/6/200422.5 3/30/200422.5 6/25/200422.5Average = 22.4 Seconds Stroke Times (Close Direction), Limit Switch set Properly9/11/200423.912/7/200423.7 3/1/200524 5/25/200524 6/16/200523.9 8/16/200523.9 8/17/200523.9 11/8/200523.8 1/31/200624.2Average = 23.9 Seconds Calculate Difference in Averages= 23.9 - 22.4 = 1.5 seconds Calculate Valve Speed in Degrees per Second (based on 1800 RPM motorspeed)*Given Information:*Limitorque actuator overall ratio (OAR) = 41.0*Limitorque HBC operator ratio = 70

• Motor speed = 1800 rpm (based on motor speed curve)*Calculate the time for the valve to rotate 90:((41*70) motor rev. per valve rev.)(60 seconds per minute)(90) (1800 motor revolutions per min)(360)= 23.9 seconds*Determine the valve speed in degrees per second:= (90)/(23.9 seconds) = 3.77/second 2-19-Calculate the Difference in Valve Travel (based on valve speed and difference instroke times)= (1.5 seconds)

• 3.77/second = 5.66 openCalculate Stem Deflection Adjustment (from stem torque traces)*T = Stem Torque Difference, Between As-Found Torque when PartiallyOpen (285 foot-lbs) and the Torque when Valve Disc Starts to Move (150 foot-lbs)= 265 - 150 = 115 foot lbs*l, d, G and J are the same as used in an earlier calculation.

Based on the above, = (115 foot-lbs

• 28 inches

• 12 inches/foot)/(11,500,000 psi*.92) = 0.00365 radians * (180/ radians) = 0.21Calculate Valve Opening Based on the Difference in Inservice Stroke TimesSince stem deflection causes the valve disc to lag slightly, stem deflection affectcauses the valve to be slightly more open. = 5.66 + 0.21 = 5.87 5.9The accuracy of this calculation is affected by the method used for measuringthe individual stroke times. When testing the valve, operators used a stop watch. There is some variability in operator response times when stopping thewatch. This uncertainty is limited to some degree because the calculation uses the difference in the average stroke times, versus comparing any two points directly. Considering that the data scatter was very limited, the response times for the opening and closing tests should approximately cancel each other in the calculation. Therefore, this calculation should provide a reasonable estimation of valve position.*Calculation 2: Based on VOTES Torque Traces: This method ofapproximation utilizes data taken directly from diagnostic system torque traces.Diagnostic system torque traces are very accurate when identifying the locationof artifacts in the traces with respect to time. Artifacts such as the start of valvemovement and the point where the valve disc hits the hard seat are easilyidentified by commonly known characteristic shapes on the traces. Accordingly, the timing between artifacts, which is performed in this calculation, is easily determined and the results are very accurate. When combined with relatively minor adjustments to account for stem deflection and valve coastdown (the amount of valve travel after the motor is de-energized), the final calculated valve position is very accurate when compared to other estimation methods.

2-20-The inspector considered this method the most accurate of all of theapproximation methods used in this report.Given Information:*Limitorque actuator overall ratio (OAR) = 41.0

• Limitorque HBC operator ratio = 70

• Motor speed = 1800 rpm

• Valve Speed = 3.77 degrees/second (calculated earlier)

NOTES:

1.The calculation assumes that the 1800 rpm motor experiences noslippage. This is generally a safe assumption considering the applicable part of the motor curve (torque = less than 1 foot pound). As part of a sensitivity analysis, however, the inspector also performed a calculation assuming worst case slippage (the motor rotates at 1775 rpm). This is considered more than bounding for this problem.2.The inspector used data from two motor-operated valve torque traces. Inone trace (2005), the closing limit switch was set properly while in the other (2003) the limit switch was set short of the target (and is the genesis of the discrepant PI inspection).Part A: Determine the size of the valve opening (in degrees) in 2003 when thelimit switch was set incorrectly (based on 1800 revolutions per minute motorspeed).1.Determine the valve travel time when the limit switch was set correctly(information from 2005 MOV torque trace):Time valve starts moving (from MOV trace) = 1.36 secondsTime close limit switch opens = 23.55 seconds Travel time = 23.55 - 1.36 = 22.19 seconds2.Calculate the distance traveled (in degrees) when limit switch was setcorrectly:(22.19 sec)(3.77/second) = 83.7Therefore, when the valve was full open, it was 83.7 away from hardseat contact.

2-21-3.Determine the valve travel time when the limit switches were setincorrectly (information from 2003 MOV torque trace):Time valve starts moving (from MOV trace) = 1.34 secondsTime close limit switch opens = 22.02 seconds Travel time = 22.02 - 1.34 = 20.7 seconds4.Calculate the distance traveled (in degrees) when the limit switch was setincorrectly:(20.7 sec)(3.77/second) = 78.05.Calculate the valve's open angle (in degrees) from the seat when theclosing limit switch opened in 2003:= total distance traveled in 2005 (from full open to seat) - total distancetraveled in 2003= 83.7 - 78.0 = 5.76.Determine the time that the valve continued to move after the closing limitswitch contact opened (2003 trace):Limit switch opened at 22.02 secondsValve stopped moving at 22.22 seconds Time of additional travel = 22.22-22.02 = 0.2 secondsThe most that the valve could have continued to travel was:(0.2 seconds)(3.77) =

.75 Since the valve was not traveling at full speed for the entire time, but was,instead, slowing to a stop, use 1/2 of the above.1/2 * .75

=

.387 .Determine the 2003 distance from the seat considering additional valvecoasting after the limit switch contact opened:

= 5.7 -

.38

= 5.38.Adjust for Stem Deflection (use same value as in NRC Calculation 1)5.3 +

.21

= 5.5 2-22-Part B (sensitivity case): Determine the size of the valve opening in 2003 whenthe valve closing limit switch was set incorrectly (assuming 1775 revolutions perminute motor speed, maximum slippage).Part B was completed in exactly the same manner as Part A above, with theexception that the motor speed was changed from 1800 revolutions per minute to 1775 revolutions per minute. The resultant valve angle was:

5.3The inspector asked the plant's engineers if they had performed other calculationssimilar to those performed in this report (by the inspector) to determine valve position.

The engineers stated that they had performed similar calculations and that they hadobtained similar results (the valve was approximately 5 open). However, the engineersdid not formalize these calculations or formally address the inconsistencies betweentheir field data, the diagnostic system data and the calculations.Performance Indicator ConclusionsDuring the first 2 years of Reactor Oversight Process implementation, NRC staff andindustry identified problems with the safety system unavailability performance indicatorsand worked to implement incremental changes to the indicators. Over the next 4 years, a working group developed a performance indicator that addressed the identified problems. This indicator, the Mitigating System Performance Index, was implemented April 1, 2006, and is addressed in NRC Regulatory Issue Summary 2006-07, "Changesto the Safety System Unavailability Performance Indicators."Based on a valve opening of approximately 5, which is a close approxiation of thedegree of valve opening as determined by the NRC, the Train B high pressure safetyinjection and both Train B containment spray pumps were unavailable between November 11, 2003, and September 9, 2004, for a range of medium and large break loss of coolant accidents. The licensee calculated the maximum tolerable valve opening area at 0.41 square inches. According to the valve vendor, a valve opening of 5corresponded to an opening size of 3.16 square inches, over 7 times the acceptable opening.The inspector determined that the High Pressure Injection and Residual Heat Removal Performance Indicator information provided to the NRC was not accurate. Based onrevised data, the High Pressure Injection Performance Indicator should have been Red while the Residual Heat Removal Performance Indicator should have been Yellow up until the point that the safety system unavailability performance indicators were replacedby MSPI on April 1, 2006. Specifically, revising the reported performance indicator data to reflect the appropriate system unavailability would have caused the High PressureInjection Unavailability Performance Indicator to turn White in the fourth quarter of 2003with a value of 2.5%, Yellow in the first quarter of 2004 with a value of 6.9%, and Red in the second quarter of 2004 with a value of 11.2%. The indicator would have remained Red until the changeover to MSPI on April 1, 2006. The Residual Heat Removal System Unavailability Performance Indicator, which includes Containment Spray, would 2-23-have turned White in the first quarter of 2004 with a value of 3.6%, and Yellow in thesecond quarter of 2004 with a value of 5.8%. The Indicator would have remained Yellow until the changeover to MSPI on April 1, 2006.Analysis. The failure to provide accurate information to the NRC in accordance with10 CFR 50.9 requirements was a performance deficiency. The issue had more than minor significance because, had the information been accurate, two performance indicators would have changed color. Per the NRC Enforcement Policy,Section IV.A.3,these issues are not subject to the Significance Determination Process. In determining the violation's severity level, the NRC considered a recent revision to the Mitigating Systems Performance Indicators. Specifically, under the revised process fault exposure hours would not be counted and the affected performance indicators would return to the Green performance level. Additionally, the initial performance deficiency, Valve SI-602B being left slightly open, had been evaluated under the Significance Determination Process and found to be of very low safety significance (Green). Following identification of the mis-positioned valve, the licensee promptly placed the valve in the correct positionwhich reestablished the functionality of the affected systems. The lic ensee conduc ted aroot cause evaluation of the underlying performance deficiency, and took actions to ensure that the condition was not present on similarly operated valves. Field verifications were conducted to ensure that similar valves were in fact, fully closed.

Accordingly, the NRC concluded that the concerns are more consistent with a SeverityLevel IV violation in lieu of the Severity Level III recommendation made in theEnforcement Policy (Supplement VII). This finding had problem identification and resolution crosscutting aspects in that the licensee's Corrective Action Program did not thoroughly evaluate the identified condition such that information reported to NRC wasverified to be complete and accurate.Enforcement. 10 CFR 50.9 requires, in part, that "Information provided to theCommission... by a licensee... shall be complete and accurate in all material respects."

Contrary to the above, from approximately November 1, 2004 (when the licensee initially submitted the information) to May 11, 2006, information provided to the Commission in the form of unavailability statistics, for the Train B high pressure safety injection andTrain B containment spray systems, was not complete and accurate in all materialrespects. The licensee significantly under-reported the unavailability hours for eachtrain. This violation did not qualify for consideration as a noncited violation because the violation did not meet the Enforcement Policy,Section VI.A.1.a criterion in that thenoncompliance was not corrected within a reasonable time following the identification of the violation. For the purposes of this criterion, the violation was identified on theinspection exit debrief date of May 11, 2006. As of the date of this inspection report issuance, the licensee had not corrected the inaccurate information (NOV 05000382/2006009-01).

2-24-4OA6Meetings, Including Exit

Exit Meeting Summary

On May 11, 2006, the senior reactor inspector presented the inspection results toMr. Kevin Walsh, General Manager, Plant Operations and other members of the licensee's staff, who acknowledged the findings. The inspector updated the licensee by telephone on June 23, 2006, with respect to proposed enforcement conclusions. A final telephonic exit meeting was conducted on August 9, 2006, with respect to a subsequent review of additional information provided by the licensee. The conclusions remained the same. Some proprietary information was reviewed during the inspection but none of the information was documented in this report.

ATTACHMENT:

SUPPLEMENTAL INFORMATION

KEY POINTS OF CONTACT

Licensee Personnel

J. Venable, Vice President, Operations

T. Gaudet, Manager, Quality Assurance

A. Harris, Acting Director, Nuclear Safety Assurance

J. Holman, Manager, Nuclear Engineering

B. Lanka, Supervisor, Engineering Design

T. Mitchell, General Manager Plant Operations, Arkansas Nuclear One

R. Murillo, Manager, Licensing

R. Osborn, Manager, Engineering Programs and Components

K. Peter, Manager, Design Engineering

P. Stanton, Supervisor, Engineering Programs and Components

T. Tankersley, Director, Safety Assurance

K. Walsh, General Manager, Plant Operations

B. Williams, Director, Engineering

ITEMS OPENED, CLOSED, AND DISCUSSED

Opened

05000382/2006009-01NOVInaccurate Performance Indicator Information

LIST OF DOCUMENTS REVIEWED

Condition ReportsCR-WF3-2000-00048CR-WF3-2004-02847CR-WF3-2006-01567Drawings52B8124, "24 body,

SMB-00-HBC-1 Actuator Limitorque - 9200 Electric Actuated ControlValve," Revision BVarious piping and instrumentation diagrams for the safety injection and containment spraysystemsK11131, "Cross-Section Adjustable Elastomer Tee Ring in Disc Edge," dated February, 1972

"Butterfly Valve Information Sheet," Sheet B-3, dated October 10, 1995

Vendor supplied drawing of valve configuration, no date, title or revision number AttachmentA-2Work Orders51581556216654150686649Miscellaneous*MPR draft letter report titled "Waterford Unit 3 Safety Injection and Containment SpraySystem Air Entrainment Reanalysis for Large and Medium Break LOCA," dated February 21, 2006.*Licensee document No. 2493C, Rev. 1, Appendix D: Response to NRC Comments ofMay 11, 2006*MPR-2758, "Waterford 3 Safety Injection and Containment Spray Systems PastAvailability Evaluation," Revision 0*Calculation Number ECM05-007, "Waterford 3 Safety Injection and Containment SpraySystems Past Availability Evaluation for Medium Break LOCA," Revision 0*Calculation Number ECM05-006, "Waterford 3 Safety Injection and Containment SpraySystems Past Availability Evaluation," Revision 0*Root Cause Analysis Report, "SI-602 B Not Fully Seated,"

CR-WF3-2004-2847, DatedSeptember 9, 2004, Report Date:

November 10, 2004.*MPR-2469, Appendix J, "Verification of Hydraulic Analysis Program," Revision 0

• Various safety injection Valve

SI-602A and

SI-602B motor-operated valve VOTEStraces, dated November 6, 2003; September 11, 2004; May 5, 2005; May 7, 2005; andMay 12, 2005*White Paper, "Standard of Proof for Assessing Past Operability," no date*Limitorque AC Motor Curve, 10 Ft-Pound motor

STI-W3-2005-0004-00. "Special Test Instructions (STI):

Leak Rate Test of

SI-602B,"performed on May 25, 2005*Informal calculation to determine as-found backpressure, assuming 0.25 square inchopening, no date*Table of trended Inservice Test data for Valve

SI-602B, no revision or date

• Estimated valve opening areas for 4 and 5 open positions, dated April 4, 2006*"Independent Assessment of Disc Position of

SI 602B between November 3, 2003, andSeptember 11, 2004," dated May 8, 2006, Kalsi Engineering