ML17309A909

LER 97-011-00:on 971102,non-conservative RAS Set Point Resulted in Operation Prohibited by Ts.Caused by Inadequate Set Point & Instrument Loop Scaling Process.Revised ESFAS Functional Tp W/Proper Set point.W/971202 Ltr

ML17309A909
Person / Time
Site:	Saint Lucie
Issue date:	12/02/1997
From:	Stall J, Weinkam E FLORIDA POWER & LIGHT CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
L-97-300, LER-97-011, LER-97-11, NUDOCS 9712090200
Download: ML17309A909 (15)

Text

CATEGORY REGULA ZORY INFORMATION DISTRIBUT N SYSTEM (RIDS)

ACCESSION NBR:9712090200 DOC.DATE: 97/12/02 FACIL:50-335 St. Lucie Plant, Unit 1, Florida Power & Light Co.

NOTARIZED: NO DOCKET ¹ 05000335 AUTHN~ AUTHOR AFFILIATION WEINKAM,E.J. Florida Power & Light Co.

STALL,J.A. Florida Power & Light Co.

RECIP.NAME RECIPIENT AFFILIATION

SUBJECT:

LER 97-011-00:on 971102,non-conservative RAS set point resulted in operation prohibited by TS.Caused by inadequate set point &, instruement loop scaling process. Revised ESFAS fuctional TP w/proper set point.W/971202 ltr.

DISTRIBUTION CODE: IE22T COPIES RECEIVED:LTR TITLE: 50.73/50.9 Licensee Event Report (LER), Incident Rpt, etc.

1 ENCL I SIZE: I 0 NOTES:

RECIPIENT COPIES RECIPIENT ID CODE/NAME LTTR ENCL ID CODE/NAME LTTR ENCL PD2-3 PD 1 1 WIENS,L. 'OPIES 1 1 INTERNAL: ACRS 1 1 AE RAB 2 2 AEOD/SPD/RRAB 1 1 LE CE 1 1 NRR/DE/ECGB 1 1 EELB 1 1 NRR/DE/EMEB 1 1 NRR/DRCH/HHFB 1 1 NRR/DRCH/HICB 1 1' NRR/DRCH/HOLB 1 1

'RR/DRCH/HQMB 1 NRR/DRPM/PECB 1 1 NRR/DSSA/SPLB 1 1 NRR/DSSA/SRXB 1 1 RES/DET/EIB 1 1 RGN2 FILE 01 1 1 D EXTERNAL: L ST LOBBY WARD 1 1 LITCO BRYCE,J H 1 1 NOAC POORE,W. 1 1 NOAC QUEENER,DS 1 1 0

NRC PDR 1 1 NUDOCS FULL TXT 1 1 U'

Ei N

NOTE TO ALL "RIDS" RECIPIENTS:

PLEASE HELP US TO REDUCE WASTE. TO HAVE YOUR NAME OR ORGANIZATION REMOVED FROM DISTRIBUTION LISTS OR REDUCE THE NUMBER OF COPIES RECEIVED BY YOU OR YOUR ORGANIZATION, CONTACT THE DOCUMENT CONTROL DESK (DCD) ON EXTENSION 415-2083 FULL TEXT 'CONVERSION REQUIRED TOTAL NUMBER OF COPIES REQUIRED: LTTR 25 ENCL 25

Florida Power & Light Company. 6351 S. Ocean Drive, Jensen Beach, FL 34957 December 2, 1997 L-97-300 10 CFR 50.73 U. S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, D. C. 20555 Re: St. Lucie Unit 1 Docket No. 50-335 Reportable Event: '97-011 Date of Event: November 2, 1997 Non-Conservative Recirculation Actuation Signal Set Point Resulted in Operation The attached Licensee Event Report is being submitted pursuant to the requirements of 10 CFR 50.73 to provide notification of the subject event.

Very t ly yours, J. A. Stall Vice President St. Lucie Plant JAS/MSD Attachment cc: Regional Administrator, USNRC Region II Senior Resident Inspector, USNRC, St. Lucie Plant 97i2090200 97i202 PDR ADOCK 05000335 S PDR lllllllillllllllllllllllllNlflllllll an FPL Group companY

NRC FORM 366 U.S. NUCLEAR REGULATORY COMMISSION APPROVED SY OMS NO. 3160<104 ExreRES 04lsolss (4-95)

ESTIMATED BURDEN PER RESPONSE TO COMPLY WITH THIS MANDATO INFORMATION COLLECTION REQUEST: 60.0 HRS. REPORTED LESSON LEARNED ARE INCORPORATED INTO THE UCENQNQ PROCESS AND F BACK TO SIDVSTRY. FORWARD COMMENTS REGARDINQ BURDEN ESTIMAT LICENSEE EVENT REPORT (LER) To THE INFORMATION AND RECORDS MANAGEMENTBRANCH IT% F33)

VB. NUCLEAR REGIAATORY COM MlSQON. WASHINGTON, DC 20666~1 AND TO THE PAPERWORK REDUCTION PROJECT 1316OCII041, OFRCE 0 (See reverse for required number of MANAGEMENTAND BUDGET, WASHINGTON, DC 20603.

digits/characters for each block)

FACIUTYNAME III DOC KEl'ISASEII 12) PAGE isl ST LUCIE UNIT 1 05000335 1 OF 12 TITLE fei Non-Conservative Recirculation Actuation Signal Set Point Resulted in Operation Prohibited by the Technical Specifications k FACIUTY NAME DOCKETNVMBER MONTH DAY YEAR YEAR SEOUENTIAL REVISION Nu M8ER NUMBER DAY YEAR N/A 05000 FACIUTY NAME DOCKEFNVMBER 11 02 97 97 011 0 12 02 97 05000 OPERATING MODE (9) 20.2201(b) 20.2203(a) (2) (v) 50.73(a)(2)(i) 50.73(a)(2)(viii)

POWER LEVEL (10) 0 20.2203(a) (2) (i) 20.2203 (a) (3) (ii) 50.73(a)(2)(iii) 73.71 OTHER 20.2203 (a) (2) (iii) 50.36(c) (1) 50.73(a)(2)(v) Specify In Abetrect below or in NRC Form 3BBA 20.2203(a)(2)(iv) 50.36(c)(2) 50.73(a) (2)(vii)

TELEPHONE NUMBER Bndude Aree Codei E. J. Weinkam, Licensing Manager (561) 467-7162 CAUSE REPORTABLE REPORTABLE SYSTEM MANUFACTURER TO NPRDS CAUSE SYSTEM COMPONENI'ANUFACTURER TO NPRDS N/A MONTH DAY YEAR EXPECTED YES SUBMISSION (If yes, complete EXPECTED SUBMISSION DATE). X NO DATE (15)

ABSTRACT (Umit to 1400 spaces, i.e., approximately 15 single-spaced typewritten lines) (16)

On October 27, 1997, Unit 1 was defueled in support of the steam generator replacement refueling outage. During the outage, obsolete Engineered Safety Feature Actuation System (ESFAS) bistables were being replaced to improve system reliability. During this effort on November 2, 1997, non-licensed utility personnel determined that the ESFAS Recirculation Actuation Signal (RAS) bistable set point for refueling water tank level had been set less conservative than the Technical Specification set point.

The apparent cause for the non-conservative RAS set point was an inadequate set point and instrument loop scaling process. The process associated with the implementation of the refueling water tank instrument set point calculation resulted in not all required related plant procedures being revised.

Corrective actions include: revising the ESFAS functional test procedure with the proper set point; implementing and calibrating to the proper set point; reviewing other set points on Unit 1 and 2; revising and training on a new Engineering set point calculation process; and issuing of a technical alert to include lessons learned.

NRC FORM 398 (4-95)

NRC FORM 386A U.S. NUCLEAR REGULATORY COMMISSI

<. <4-96)

LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION YEAR SEQUENTIAL REVISION St. Lucie Unit 1 05000335 2 OF 12 97 011 0 TEXT (Ifmore spaceis required, use additional copies of fVRC Farm 366Ai I17)

On October 27, 1997, St. Lucie Unit 1 was defueled in support of the steam generator replacement refueling outage. As part of the outage, obsolete Engineered Safety Features System (ESFAS) bistables [EIISIJE:EA] were being replaced to improve system 'ctuation reliability and calibration methods. The equipment replacement included all four channels of the Refueling Water Tank (RWT) [EIIS:BQ:TK] low level bistables. A low RWT level on two of the four (2 out of 4 logic) channels will initiate the Recirculation Actuation SignaI (RAS).

This signal changes the operating mode of the safety injection system [EIIS:BQ] from injection, with suction from the refueling water tank, to recirculation, with suction from the containment sump [EIIS:BQ], in the event of a Loss of Coolant Accident (LOCA).

As a result of a RWT level change on the ESFAS cabinet meter [EIIS:BQ:MTR] during the replacement of the RWT level bistable, the System Engineer performed additional verification to ensure that the RWT level set point agreed with the instrument loop scaling requirement.

The review showed that the St. Lucie Unit 1 Technical Specification Table 3.3-4, "Engineered Safety Feature Actuation System Instrumentation Trip Values," set point of "48 inches from the tank bottom" correlated to a bistable trip set point of 5.28mA, while the functional test procedure 1-1400052, "Engineered Safeguards Actuation System - Channel Functional Test," required an assigned set point of 4.96mA. Based on the instrument loop scaling, 4.96mA corresponded to an RWT level of 36 inches above the tank bottom.

Following the engineer's checks, along with an independent check by Instrument 5 Control personnel, the System Engineer concluded on November 2, 1997, that the bistable setting of 4.96mA corresponded to a less conservative RAS set point of 36 inches above the tank bottom, and that the less conservative set point had been applied to all four channels of the RWT level instrument loop bistables. It was determined that the set points were in place prior to the current plant refueling shutdown. The System Engineer generated Condition Report 97-2092 to have this as-found condition investigated for cause and corrective actions.

NAC FOAM 366A <4-95I

NRC FORM 366A U.S. NUCLEAR REGULATORY COMMISSIO I4-95I LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION YEAR SEQUENTIAL REVISION St. Lucie Unit 1 05000335 3 OF 12 97 011 0 TEXT ilfmore spece is required, use edditionel copies of NRC arm 366Ai I17)

The apparent cause for the non-conservative RAS set point was an inadequate set point and scaling change process, in that, the process associated with implementation of the RWT level instrument set point calculation resulted in not all required related plant procedures being revised. No formal process existed. for the identification and tracking of action items associated with the issuance and revision of set point calculations.

As part of enhancing the St. Lucie Set Point Program in 1992/1993, Engineering issued calculation PSL-1FJI-92-011, Revision 0, dated January 22, 1993, changing the span of the RWT level measurement loop, as well as identifying a new set point. The effect of this change was to correct the RWT measurement and indication loop to indicate zero feet tank level being on the bottom of the tank with the instruments located at 1 foot above the bottom of the tank. Prior to the loop rescaling, the indicated level of, 0 to 50 feet (4 to 20mA level transmitter output) corresponded to 1 to 51 feet of actual tank level. For this previous scaling condition, the corresponding set point contained in functional test procedure 1-1400052 was listed as 4.96mA. The 4.96mA corresponded to a tank level 36 inches above the instrument tap which is 1 foot from the tank bottom. Therefore, 4.96mA corresponded to 48 inches actual level. This set point value accurately corresponded to the Technical Specification required set point of 48 inches above the tank bottom.

On February 25, 1993, Refueling Water Storage Tank Level Calibration procedure 1-1400153H was revised to implement the span conditions provided in calculation PSL-1FJI-92-01.1, Revision 0. The level span change was implemented on Unit 1 during the 1993 Spring refueling outage on April 19, 1993. However, set point procedure 1-1400052 was not revised by the Instrument 5 Control personnel to reflect the new scaling conditions.

This caused the set point to be incorrect since the 4.96mA set point corresponded to 36 inches actual level, based on the new scaling condition of zero foot tank level being at the bottom of the tank. The RAS set point in procedure 1-1400052 should have been changed to 5.28mA in order to be consistent with the new scaling calculation. This procedure was not changed and thus the existing set point (4.96mA) corresponded to RAS actuating at 36 inches from the tank bottom instead of the Technical Specification required 48 inches.

NAG FOAM 366A I4 95)

NRC FORM 366A U.S. NUCLEAR REGULATORY COMMISSIO I4.95)

LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION YEAR SEQUENTIAL REVISION St. Lucie Unit 1 05000335 4 OF 'I2 97 011 0 TEXT fifmore spece is required, use edditione/ copies oi iVRC Form 3M4/ (17)

The RWT level calculation was prepared as part of the 1992/1993 Set Point Enhancement Program and issued separately from the normal plant change and safety evaluation process.

Currently, most of the set point and calibration changes are associated with plant modifications or safety evaluations. They are processed through the plant configuration control group and are sent to the affected site departments for review, and action items are assigned and tracked to completion. The plant change, including procedures, are reviewed by the site's Facility Review Group (FRG) and approved by the Plant General Manager. This ensures that all design, maintenance, and operational requirements are properly addressed and implemented. Determination of the affected procedures is accomplished through electronic searches of documents and databases, PASSPORT (Total Equipment Database),

and personnel experience and knowledge of the affected system.

Calculations may also be performed as a result of emergent conditions. These calculations would normally be associated with Condition Reports (CR) or Plant Management Action Items (PMAI). The CR and PMAI processes include a formal tracking mechanism for follow-on activities.

Contributing issues associated with the non-conservative RAS set point included:

1) Though the engineering calculation PSL-1FJI-92-011, Revision 0 was issued with the necessary information to revise both the loop calibration and the ESFAS set point, specific implementation requirements were not issued to provide the Instrument &

Control maintenance personnel clear instruction. An informal Engineering punch list only specified making the scaling change and not the set point change.

2) Cross-functional communication between Engineering personnel and Instrument &

Control personnel did not occur. Communications existed between the Engineer performing the calculation and the RWT loop Instrument & Control Supervisor, but there was no evidence of communication between the calculation preparer and the ESFAS system supervisor. Engineering did not perform any follow-up regarding the new calculation and Instrument & Control maintenance was not required to provide any feedback to the calculation preparer regarding implementation.

NRC FORM 366A I4-95I

~ '

I ~ 4~ ic: 4~

Hl

~

I II ~

EEEI

~ ' ~ . ~ ~ I ~ 8 ~: 4 ~ ~ ~

~ ~ ~ ~ ~ ~ ~

' ~ ~ ~

~ I ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~

~ ~ ~ ~ ~ ~

~

~

~

~,

~

~

~ - ~

~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~

~ ~

~

~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ' ~ ~

~ ~ ~

' ~

~ ~ ~ i'

~ ~ ~ ~ I ~ ~ ~ ~ I

~ ~ ~ ~ ~ ~

~ ~ I ~ ~ ~ ~

~ ~ ~ ~

~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~,

~ ~

~ ~ ~ ~ ~ ~

I ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ I ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~

~

' ~ ~ ~ ~ ~

~ . ~ ~ ~ ~ ~ ~ ~ ~

~ I' ~

~ ~

~ ~

~ ~ ~ ~

~ ~ ~ ~ ~

NRC FORM 388A U.S. NUCLEAR RMULATORYCOMMISSI I4-95)

LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION YEAR SEQUENTIAL REVISION St. Lucie Unit 1 05000335 6 OF 12 97 011 0 TEXT /ifmore speceis required, use edditionel copies of ilRC Form 3MAJ I17I A review was performed of the remaining set points for the RPS, Auxiliary Feedwater Actuation Systems (AFW) [EIIS:BAJ, and the ESFAS for both St. Lucie Unit 1 and Unit 2.

A comparison of the Technical Specification set point values, the Engineering set point calculations, the individual process measurement calibration procedures, and the bistable surveillance procedures was performed. It was concluded from these reviews that the remaining ESFAS, RPS, and AFAS set points and loop calibrations meet the Technical Specification requirements.

FPL concluded that this RWT set point error was an isolated incident where design information was not correctly transferred into maintenance procedures and ultimately the plant instrumentation. The processes that were in place to initiate and control set point changes were not formal and relied heavily on human performance for success. Where calculations are prepared in support of plant change modifications, safety evaluations, or Condition Reports, the current processes include formalized action item tracking systems.

This event is reportable in accordance with 10 CFR 50.73 (a)(2)(i)(B) for any operation or condition prohibited by the plant's Technical Specifications. Additionally, this event met the conditions of 10 CFR 50.73 (a)(2)(ii)(B) as a condition that was outside the design basis of the plant, and 10 CFR 50.73 (a)(2)(v) as any event or condition that alone could have prevented the fulfillment of the safety function of structures or systems that are needed to:

... (B) Remove residual heat....or (D) Mitigate the consequences of an accident.

St Lucie Unit 1, utilizes an automatic signal to accomplish the transfer of the emergency core cooling system (ECCS) suction at the end of the injection phase from the refueling water tank (RWT) to the containment sump to begin the recirculation phase. On reaching the recirculation actuation signaI, the low pressure safety injection pumps (LPSI) are secured, the containment sump isolation valve is opened in approximately 30 seconds, and the RWT ECCS suction line isolation valve is closed in approximately 90 seconds. Effectively, the suction supply source is transferred to the sump within a few seconds of RAS, when the sump isolation butterfly valve starts to open. This automatic function is backed up by manual action specified in the plant emergency operating procedures. At its minimum level allowed by plant Technical Specifications, there is sufficient water in the RWT to support-injection at full ECCS flow (i.e., two trains of containment spray, two trains of High Pressure Safety Injection (HPSI), and two trains of LPSI for at least 20 minutes into the LOCA scenario. If less than full ECCS flow is available or not required or if run out conditions do not exist (as in the case of small break LOCAs) substantially longer times (up to several hours) would pass prior to RAS initiation.

NRC FORM 368A I4-95)

NRC FORM 366A U.S. NUCLEAR REGULATORY COMMISSIO I4-95I LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION YEAR SEQUENTIAL REVISION St. Lucie Unit 1 05000335 7 OF 12 97 011 0 TEXT /ifmore spece is required, use edditional copies of NRC Form 388A/ I17)

The nominal set point for RAS is 48 inches from the bottom of the RWT. The ECCS suction pipe outer diameter is approximately 24 inches with the top of the pipe inner diameter at 42.25 inches and the pipe center line at 30.75 inches from the tank bottom. The set point error would have resulted in a nominal set point of 36 inches or approximately 6 inches below the top of the ECCS suction line.

The RAS set point is not reached until after the Injection phase is complete (an absolute minimum of 20 minutes into the LOCA assuming the minimum RWT volume and the highest possible full ECCS flow). At this point in the post LOCA analysis, the containment peak pressure and temperature have already been mitigated as well as the fuel peak clad temperature (PCT). At the start of the recirculation phase, the core has been ref looded and the ECCS function is decay heat removal and makeup for boil off.

The potential for vortexing at levels slightly above the ECCS suction has been previously evaluated and it was determined that operation of the ECCS pumps would not be adversely impacted. However, during the Unit 1 Cycle 14 refueling outage, a vortex suppression device (PC/M 96085) was added to the suction line to minimize any vortexing. There is approximately 25 feet of elevation difference between the RWT and the ECCS pump suctions. This serves to compress any air entrained and to provide NPSH to pumps for RWT levels above the top of the ECCS suction line. Therefore, air entrainment due to vortexing at RWT levels above the top of the ECCS suction line is not of concern.

Since the set point error would have resulted in a RAS actuation point at a RWT level below the top of the ECCS suction line, open channel flow would have occurred. Based on calculations performed by both FPL and ABB-CE, open channel flow would not be capable of supporting full ECCS flow (approximately 13,000 gpm). The mismatch between the available open channel flow from the tank and the ECCS full flow would result in a drain down of water level in the suction piping. This drain down would result in air ingestion and reduction in the available NPSH at which point the ECCS flow would be reduced and the piping might momentarily refill. If operation was to continue under these conditions, this could result in a chugging flow during which significant air ingestion could have occurred. This flow condition could have resulted in the ECCS becoming air bound which could have prevented further drain down of the RWT to the lower automatic RAS actuation point.

Although high ECCS flows could have resulted in the chugging flow described previously, lower ECCS flows of approximately 7000 gpm initially and 1000 gpm at the RAS set point could have been adequately supported by the open channel conditions without adversely impacting pump performance. For these lower ECCS flows, the automatic RAS function would have been successful at a RAS set point of 36 inches.

NRC FORM 366A I4.95I

NRC FORM 366A U.S. NUCLEAR REGULATORY COMMNSIO I4-95)

LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION YEAR SEQUENTIAL REVISION St. Lucie Unit 1 05000335 8 OF 12 97 011 0 TEXT (If more speceis required, use edditionel copies of NRC Form 368Ai I17)

Although RAS actuation is designed as an automatic function, manual actuation from the control room in the event that automatic RAS fails is clearly directed in the plant emergency operating procedures (1-EOP-03, "Loss of Coolant Accident" ). Initial and requalification operator training stresses manual actuation of RAS.

Four channels of RWT level indication are available to the control room operators in the form of Versatile indicators. Each of the four Versatile indica<ors has an analog display consisting of lighted bars in one half foot increments and a digital readout in tenths of a foot. These indicators were correctly spanned and were not affected by the set point error. RAS may be manually initiated by simultaneous operation of a switch and push-button located on the same vertical panel as the RWT level indicators.

At a 'RWT level of 6 to 8 feet, 1-EOP-03 (step 41) directs the operator to perform specific actions to ensure power is available to the minimum flow isolation valves. The applicable switches are located on the horizontal panel below the RWT level indicators. Therefore, by procedure, the operator is in the required vicinity to initiate RAS manually and is monitoring RWT level minutes before reaching the RAS set point. The next step in the procedure (step

42) directs the operator to manually initiate RAS if it does not occur automatically at 4 feet.

The RWT level is calculated to be falling at a maximum rate of about 10.7 inches per minute conservatively assuming all six ECCS pumps operating at run out flow. Under full ECCS flow, the operator would have a minimum of 90 seconds to initiate RAS before HPSI pump degradation would occur and 240 seconds before LPSI and containment spray pump degradation occurred. These estimates are based on the following: (1) a minimum of 30 seconds until the level reaches the top of the ECCS suction line and open channel flow begins, assuming full run out ECCS flow; (2) 30 seconds minimum at full ECCS flow and open channel conditions until the ECCS pumps start drawing air resulting in chugging flow; and (3) an additional 30 seconds of operation during unstable flow conditions for the HPSI pumps and 180 seconds for the LPSI and containment spray pumps, before significant pump degradation occurs.

If the operators failed to manually initiate RAS at 4 feet, as directed in the emergency operating procedure, several indications of degrading pump performance would be available including motor amps, discharge pressure and loop flow. Based on vendor discussions, while this would most likely be too late to prevent damage to the HPSI pumps, one to two additional minutes of operation would remain before damage was done to the LPSI or containment spray pumps.

NRC FORM 366A I4.95)

NRC FORM 366A U.S. NUCLEAR REGULATORY COMMISSI I4-96I LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION YEAR SEQUENTIAL REVISION St. Lucie Unit 1 05000335 9 OF 12 97 011 0 TEXT (Ifmore speceis required, use edditionel copies of NRC Form 86EA/ I17I A test was conducted on the training simulator to confirm that the operators would manually initiate RAS within 90 seconds, as expected, if RAS did not automatically actuate at the nominal set point. Without prior knowledge or briefing of the set point issue, a large break LOCA scenario, with the automatic RAS function defeated, was run on an operating crew.

The operators maintained positive control over RWT level and manually initiated RAS as directed in the emergency operating procedures approximately 40 seconds after the nominal RAS set point was reached.

The peak containment pressure and temperature are reached at approximately 10 seconds (St. Lucie Unit 1 UFSAR section 6.2) following the postulated large break LOCA. The earliest time for RAS actuation is 1200 seconds (20 minutes) into the accident, long after the peak temperature/pressure has been mitigated. The containment fan coolers are not impacted by the RAS issue and therefore would be capable of removing heat at this point in the accident without containment spray. Therefore the containment would not have been challenged during a large break LOCA as a result of the RWT set point error.

For small break LOCA in which LPSI flow is not required, and for which the operators are expected to secure one or more containment spray pumps, a significant amount of time is available before the RWT RAS set point is reached. The open channel flow that would have existed prior to RAS would have been capable of supporting the ECCS flow under these conditions and automatic RAS would have been successful.

The worst case for air entrainment resulting from the lower RAS set point would have been with all trains of ECCS operating (i.e., no single failure of an ECCS). However, full availability of the ECCS trains improves overall margins for the accident scenario by reducing containment pressure and temperature and rapidly ref looding the core. Each train of ECCS is 100% capacity.

Emergency operating procedure 1-EOP-03 directs the operators to secure spray when containment pressure falls to 5 psig and to reduce HPSI flow and secure LPSI under specific conditions. Additionally, actual pump run out conditions would not have existed for most LOCA scenarios. Therefore, it is likely that for most postulated LOCA scenarios the ECCS flow at RAS would have been on the order of 6000 gpm, As stated earlier, the open channel flow conditions would have been capable of supporting a 7000 gpm ECCS flow initially and therefore, additional time (several minutes) would have been available to the operators to manual initiate RAS.

NRC FoAM 366A I4-95)

~

~

I~ R~e!gQ m

~ ~ ~ I~ ~o

~ III IREI I ~

I ~ ~ 4 ~ .'

~ ~ ~ ~ ~ ~

I ~

~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ t ~ ~

~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~, ~ ~ ~ ~ ~ ~

~ ~ ~

' ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ I ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~

' ~ ~, ~ ~ ~ ~

~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~

~ ~ ~ ~ a ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~

' ~ ~ ~ ~ ~ ~

~ ~ e ~

~ ~ ~ ~ ~ ~

' ~ ~ ~ ~ ~ ~ ~ ~ ~ ' ~ ~

~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~

- . s

~ ~ ~ ~ ' ~ ~ ~ ~ ~ ~ ~

' ~ ~ ~

' ~ ~

~ ~ ~ ~ ~ ~

' ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~

~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~

KRC FORM SSSA U.S. NUCLEAR REGULATORY COMMISSI I4-9S)

LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION YEAR SEQUENTIAL REVISION St. Lucie Unit 1 05000335 11 OF 12 97 011 0 TEXT iifmore spaceis required, use edditionel copies of NRC Form 866tAJ I17I The RAS set point error has existed since 1993. During this period, the automatic function of RAS would have been successful for small break LOCA scenarios. In the unlikely event that a large break LOCA had occurred during the period that the set point error existed, automatic RAS may not have been successful. This would have been dependent on the actual ECCS flow rates prior to recirculation. In the worst case scenario, manual action would have been required to initiate RAS. Because manual initiation of RAS is clearly proceduralized and easily accomplished from the control room there is a high degree of confidence that manual actuation of RAS would have been successful. In the unlikely event that the control room operators failed to manually initiate RAS, time would still have been available prior to core darriage (M one hour) for recovery actions to restore flow.

As demonstrated by the PSA estimates, the corresponding increase in core damage frequency is 3.0E-5/yr. The PSA estimate reflects confidence in operator action for the large break scenario and the fact that the probability of small break LOCAs is much greater than for large break LOCAs.

1. Procedures 1-1400052, "Engineered Safeguards Actuation System - Channel Functional Test," and 1-1400166, "Engineered Safeguards Actuation System - ATI Alignment Check," have been revised to include the correct RAS set point. Implementation and calibration of the correct set point is required prior to Mode 3.

2. Remaining set points for the Unit 1 ESFAS and all other set points for the RPS and AFAS were reviewed. It was verified that set points and loop calibration data meet the Technical Specification requiremerits.

3. Set points for the Unit 2 ESFAS, RPS, and AFAS were also reviewed. It was verified that set points and loop calibration data meet the Technical Specification requirements.

4. The Engineering procedures governing the preparation of set point calculations are being revised to include formal instructions regarding the identification and transmittal of plant action items as the result of calculational changes.

5. Procedures 1-1400052, "Engineered Safeguards Actuation System - Channel Functional Test," and 1-1400153H, "Refueling Water Storage Tank Level Calibration," have been revised to include cross-references between the two procedures. In addition, the remaining process instrument loops with input to the ESFAS, RPS, and AFAS monthly functional procedures were also revised to include cross-references. This will also be completed on Unit 2.

NRC FORM 366A (4.96)

NRC FORM 366A U.S. NUCLEAR REGULATORY COMMSSIO I4-851 LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION YEAR SEQUENTIAL REVISION St. Lucis Unit 1 05000335 12 OF 12 97 011 0 TEXT ilfmore speceis required, use edditionel copies of NRC Form 366Ai I17I

6. Training will be provided to Engineering personnel regarding the revised procedures resulting from Corrective Action 4.

7. An Engineering Technical Alert was issued to include lessons learned from this event.

8. A Maintenance Training Bulletin was issued to describe this event and to emphasize the importance of the procedural requirements for the design change review process.

9. An Engineering Technical Alert has been issued to define the responsibilities of Engineering and Instrument L Control Maintenance for the development of safety-related instrumentation calibration sheet information.

None Recent LER related to non-conservative set point issues:

LER 335/97-002, "Operation in Excess of Maximum Rated Thermal Power (RTP) Due to Digital Data Processor Calorimetric Error" (Voluntary Report).

LER 335/90-001, "Low Steam Generator Pressure Trip to Main Steam Isolation Signal Set Below Technical Specification Allowed Minimum Value Due to Procedural Error."

NRC FORM 366A (4-95)