ML20199K483

Submits Response to Request for Addl Info Re Power Uprate Facility Operating Licenses & Tech Specs Change Request

ML20199K483
Person / Time
Site:	Farley
Issue date:	01/23/1998
From:	Dennis Morey SOUTHERN NUCLEAR OPERATING CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NUDOCS 9802060194
Download: ML20199K483 (20)

Text

.

~

U , Dave Mor:y Southern Nucleat Vo bestrit Operating Company Fadey Propet PO B a 1295 Bmineam.Natiama 35201 tel 205 932 5131 SOUTHERN January 23, 1998 ggg Energy to Serve l'our %rld" Docket Nos. 50 348 10 CFR 50.90 50 364 U. S. Nuc! car Regulatory Commission A'ITN.: Document Control Desk Washington, DC 20555 Joseph M. Farley Nuclear Plant Response to Request for Additional Information Related to Power Uprate Encility Operating Licenses and Technical Specifications Change Reauest Ladies and Gentlemen:

Dy letter dated February 14,1997, Southem Nuclear Operating Company (SNC) proposed to amend the Facility Operating Licenses and Technical Specifications for Joseph M. Farley Nuclear Plant (FNP) Unit I and Unit 2 to allow operation at an increased reactor core power level of 2775 meg .,vatts thermal (Mwt). NRC letters dated July 1, lo9?; August 21,1997; and October 14,1997 requested SNC provide additional information. SNC responded by letters dated August 5,1997; September 22, 1997; and November 19,1997 respectively. SNC letters dated December 17 and 31,1997 responded to NRC questions resulting from conference calls on December 9,10,11 and 15,1997. By telephone conference calls on December 19 and 22,1997 and Janusry 7 and 13,199f, SNC respondod to addit ional NRC StalTquestions. Attachment I provides the SNC responses to these questions.

Attachment 11 includes corrections to the power uprate BOP Liwnsing Report (page 25) and Attachment I to SNC letter dated August 5,1997 (page 31 If you have any questions, please advise.

Respectfully submitted,

[f)77ttW Dave Morey Su orn to and subscribed befbre me thi.cf[by o

.,1998

\

YLNotary Pub'llcV Q-ie $ C2 r '

/

My Commission Expires: kO & r l;c200/

MGFJmaf:pwrup28. doc f[Dl Attachments y gyQ cc: Mr. L. A. Reyes, Region 11 Administrator Mr. J.1. Zimmerman. NRR Project Manager h Mr. T. M. Ross, Plant Sr. Resident inspector hjk , gg{.}

• 9002060194 900123 PDR ADOCK 05000340 P PDR ,

J

ATTACitMENT SNC Response to NRC Request For AdditionalInformation Related To Power Uprate Submittal . Joseph M. Farley Nuclear Plant. Units I and 2 SNC RESPONSES TO NRC QUESTIONS RESULTING FROM NRC/SNC CONFERENCE CALLS ON DECEMBER 19 AND 22,1997 AND JANUARY 7 AND 13,1998 l

i i

M, -

f- -

SNC Response to NRC Request For Additloaalinformation Related To Power Uprate Submittal . Joseph M. Farley L : clear Plant, Units I and 2 NRC_Qugstion No.1 (ReferenecEssember 19.1997 & January 13.1998 NRC/SNC Conference Calli)

The SNC response to Question No. 5 in SNC letter dated December 17,1997, indicated that a plastic analysis had been performed to demonstrate the acceptability of the steam generator disider plate and that this analysis was documented in Westinghouse "$1 Series Steam Generator Stress Report," Section 3.3, dated April 1972. Describe the analysis methodology and, if applicable, state what computer code was used. Also, provide more detail about the disider plate fatigue calculations for the Farley uprate.

SNCESJP903e No. l The computer program used for the Farley ste m generator divider plate is described in Westinghouse Research Report 67 ID7 GENCO R3,"A Finite Element Computer Program for Elastic-Plastic Plane Solutions," S. E. Gabrielse and C. Visser. Ec program uses the fmite element displacement method and triangular elements. The assumed displacement functions result in coastant strains and stresses on the elements. First the clastic solution is obtained, and the equivalent stress in each element is computed The element is considered in a plastic state if the exiuivalent strain is larger than the equivalent strain at yield point under a uniaxial state of stress.

The plastic strains are then obtained and an improved solution is computed by an iterative procedure. The program is designed as a general purpose plane clastic-plastic program and can determine the displacements and stresses in arbitrary planc shapes by replacing the actual shape by an assemblage of triangular elements. The program is limited to a maximum of 700 element and 500 nadal points. He program will accept a variety of boundary conditions and loadings, such as prescribed displacements, boundary tinctions, body forces and temperaturc variations. Several plotting routines have been incorporated into the program to facilitate the interpretation of the results.

In the onginal stress report [ Reference it the stress intensities for three of the load conditions (Primary llydro, Secondary 1lydro, and Lo s of Load at 15 sec) were calculated using plastic analysis. He equivalent plastic strains were multiplied by a stress concentration factor of 1.5 and the clastic modulus to obtain the stress intensities used in the fatigus evaluation. The stress intensities from Reference I for each of the load conditions contributing to the divider plate fatigue usage are listed in the following table.

Page1

Number Load Condition Number Stress intensity Elastic / Plastic of Cycles (Sn - ksi) 1 Prima.y Side !!ydrotest 5 362.678 Plastic 2 Secondary Side 11ydrotest 5 165.791 Plastic 3 Loss ofload at 15 sec 600 270.691 Plastic 4 less oflead at 100 sec 4810 70.237 Elastic 5 Plant Loading 18300 128.199 Elastic 6 Full Load 18300 128,980 Elastic 7 lleatup (0% Power) 18300 126.578 Elastic 8 Ileatup (Cold) 200 0.000 Elastic Ac resulting fatigue usage was then calculated according to the following.

Load Alternating Number Allowable Usage Condition Stress Intensity of Cycles (n) Cycles (n/N)

Combination (ksi) (N)

I-2 264.235 5 77 0.0649 38 135.346 200 490 0.4082 3-4 100.227 400 1300 0.3077 4-6 29.372 4410 460000 0.0096 6-7 1.201 13890 Infinite 0.0000 57 0.811 4410 Infinite 0.0000 Total 0.7904 The alternating stress intensities in the above table were obtained by taking one-half the difference of the stras intensities for the approprirte load conditions given in the first table. Note that combination 3 4 does combine stresses obtained from a plastic analysis with stresses obtained from an clastic analysis. This is a valid operation since the stresses obtained by multiplying the equivalent plastic strains by the clastic modulus are consistent with the method used to obtain trw ASME Code fatigue curves [ Reference 2]. (Note that since case 8 has zero stress intensity, the combination 3 - 8 uses the plastic stress intensity value only )

For the uprate evaluation, the scale factors due ta the uprate were applied to the appropriate stresses and the fatigue evalt.ation was performed. The revised fatigue usage is less than 1.0, satisfying the ASME Code requirements.

Etsoonse No. I Refereness

1. "$1 Series Steam Geaerator Stress Report, Section 3.3, Divider Plate Analysis," Westinghouse Tampa Division, Tampa, FL, April 1972.

2. ASME Boiler and Pressure Vessel Code Section ill," Nuclear Power Plant Components," 1971 Edition.

W/ alt . 01/14/98 Page?

NRC_Qugjlion No. 2 (ReferengfErstmber 19.1997 NRC/SNC Confsrence Call)

Please estimate margins associated with Steam Generator LOCA hydraulic forcing functions, loop forcing fune. ions, and reactor vessel forcing functions for the Farley power uprate analysis, include a discussion, where appropriate, of specific Unit I and " nit 2 differences, including a discussion of computer codes used.

SNC Regense No. 2 Stear.: Generator LOCA Ilydraulic Forcing Functions Prior to uprating, model $ 1 steam generator LOCA hydraulic forcing functions for Farley Units I and 2 were generated using MULTIFLEX l.0 for branch line break sizes using conservative branch line modeling. Model $1 steam generator specific sensitivities to RCS operating temperatures r.ud more accurate branch line modeling were developed for the power uprate analysis. These sensitivities looked at peak LOCA pressure differentials across the ll-bend, across the tubes at the tubesheet, and across the vertical disider plate. The conservatively calculated LOCA forces used in the current steam generator structural analysis are compared with those calculated for the uprated power with more accurate branch line modeli~,in the following table. i Location Original Power Peak AP ' Uprated Power Peak Net Margin Remainin?, j (psi) AP (psi)

U Bend 59 45 24Y.

Tube Sheet 350 294 16 %

Divider Plate 225 229 0%*

i

' Difference of 4 pi is judged to be negligible since small changes (i.e., < 2%) in LOCA forces have been shown to have an insignificant effect on the structural analyses.

As a result, the original power steam generator forcing functions are judged to be applicable to the uprated conditions, it should be noted that the uprated power forcing functions are computed at ,

the minimum allowable terr.peratures and maximum allowable operating pressures including uncertainties. The net margin remaining in the above table only refers to margin in the forcing functions.

Loop Forcing Functions The loop forcing functions for Farley Units I and 2 consist of double-ended guillotine (DEG) and limited displacement (LD) breaks of the main loop piping calculated with the BLOWDOWN 2 and TilRUST codes. The most limiting break is the LD break of the reactor vessel inlet nozzle (RVIN). The most limiting cold leg brak area that must be addressed under leak-before-break (LBB)is the accumulator branch line. The reduction la teraperr.ture associated with the uprated power (including uncertainties) is calculated to result in an increase of 13.2% in loop forces. The reduction in break area associated with LUB is calculated to result in a decrease of approximately 30% in ieop forces. No specific benefit is expected from use of MULTIFLEX 1.0 over DLOWDOWN2 because no flexible wall modeling is credited in the loop forces calculation (a conservative assumption).

Page 3

As a result, the main loop piping breaks are judged to remain bounding and are estimated to remain conservative by more than 16% due to break area margin alone.

Vessel Forcing Functions

{

%c vessel forcing functions for Unit I were presiously generated using MULTIFLEX l.0 for the l

upflow conversion program. De vessel forcing functions for Unit 2 were generated using BLOWDOWN 2. During the upflow conwrsion program, the upflow configuration was identified to be more limiting than the downflow configuration with respect to the peak honzontal forces on the core carrel. %c upflow conversion program modeled a lirr.ited displacement break of the reactor vessel inlet and cutlet nozzles as the limiting breaks.

l For the uprating program, leak before break (LBB) was credited in excluding the main loop piping from consideration for LOCA forces. He largest branch line on the cold leg modelod in the uprating snalysis was the accumulator line.

De increase in forces estimated to result from the uprating RCS condition changes was less than 14%. He decrease in forces estimated to result from the use of the branch line breaks was approximately 30%. He decrease in forces resulting from use of MULTIFLEX 1.0 instead of  !

BLOWDOWN 2 was estimated to be 30%.

For Unit I, the uprated forces are estimated to be 16% (+14%, -30%) lower than the forces previously analyzed. For Unit 2. the uprated forces are estimated to be 46% (+14%, -30%,

-30%) lower than the force nresiously analyzed.

W/knk A res 1/lW8 NRC Ouestion No. 3 (Reference December 19.1997 NRC/SNC Conference Call)

With the exception of valves in the Main Steam, Main Feedwater and Auxiliary Feedwater Systems, confirm that the safety related valve design basis parameters (pressure, temperature, differential pressure (AP), and flow rate) are unchanged by the Farley power uprate.

SjiC Response No. 3 De FNP NSSS and BOP safety-related valves design parameters were demonstrated D91 to be impacted by the power uprate system process parameter changes. Bis fact was confirmed by verifying that any change (e.g , system operating temperature and/or pressure) was bounded by the requirements of the associated equipment specifications. He Westinghouse NSSS fluid systems (RCS, CVCS, Si and RHR) and associated components e,aluation summaries are presented in Sections 4.1 (pages 4 1 - 4 12) and $.9 (pages $ 18 - $ 19) of the NSSS Licensing Report. De results of addnional HHSI & LHSI system reviews are provided in Section 2.11 of the BOP Licensing Report (page 27). He component cooling water and service water systems evaluation sumnaries for power uprate process parameter changes are provided in Sections 2.8 and 2.9 (pages 20 - 23). Because the Farley MOV Program resulted in changes to the design basis operating requirements for certain MOVs, additional reviews confirmed that MOV motor-operator Page 4

j l

a capability and switch settings would not be adversely affected by any system process parameter changes. Section 2.19 (page 72) of the BOP Licensing Report conclud : that the uprate has no impact on the FNP MOV Progiant (Additional information pertaining to the MOV reviews is presented in our responses w NRC Additional Question No.12 in SNC Leiter dated August $,1997 and NRC Question No. 6 in SNC Letter dated December 17,1997.) Rese evaluations, as supported by the uprate control system and safety analyses, demonstrate that the safety related valves in the RCS, CVCS, Si and RilR Systems will perform their design basis functions during normal uprated system operating conditions and during postulated events.

SCS/gid.1/8/98 A SNC/mge.1/15/98 NRChuerttion Nc. 4 @cfstentsfecember 19.1997 NRC/SNC Confetmec Call) 4 Provide a detailed description of the snalyses/ evaluations performed for the safety related valves in

, the Main Steam, Main Feedwater and Auxiliary Feedwater Systems.

SNC Response No. 4

%c Main Steam and Main Feedwater Systems were evaluated for power uprate impacts associated with increased flow and reduced temperature and pressure. %c evaluations explicitly addressed the MSSVs, SO ARVs, MSIVs, FRVs, and FIVs in Section 4.2 (pages 4 13 - 4 17) of the NSSS Licensing Report and in Sections 2.2,2.7 and 2.19 (pages 5 - 8,18 - 19 & 72) of the BOP Licensing Report. He Auxiliary Feedwater System was evaluated for power uprate impacts associated with reduced temperature and pressure. De AFW system evaluations are presented in Section 4.2.4 (pages 4 18 - 4 19) of the NSSS Licensing Report and in Sections 2.12 and 2.19 (pages 28 - 31 & 72) of the BOP Licensing Report. %cse evaluations, as supported by the upate control system and safety analyses, demonstrate that the safety-related valves in the MS, MFW and AFW Systems will perfonn their design basis functions during normal uprated system operating conditions and during postulated events.

SCSigid. l/8/98 A SNC/mge l/l$/98 NRC Ouestion No. 5 (Reference December 19.1997 NRC/SNC Conference Call)

C *m that the change in Main Steamline Isolation Valve closure stroke time from 5 to 7 seconds doa, not adversely affect compartment pressurization loads, LOCA loads, and equipment inside

, and outside of containment.

SRC Response No. 5 As discussed in NSSS Licensing Report Section 4.3.1.3, the change in Main Steamline Isolation Valve (MSIV) closure stroke time was assessed in applicable power uprate safety analyses and evaluations, including mass and energy (M&E) release analyses. He analyses and evaluations demonstrated that the MSIV closure time increase is acceptable, r

Page5

~_ _ _. . . _ , _ _ _ _ ___ _,

Subcompartment pressure and ternperature responses tc, pipe breaks were assessed for power uprate. The mass and energy releases were determined for LOCA and Steam Line Breaks (SLB) as described in Sections 6.4 and 6.5 of the NSSS Licersing Report. While the chance in MSIV closure time does not affect LOCA forces, the LOCA blowdown rnass and energy analyses are discussed on pages 6 209 - 6 220 and 6 250 - 6 251. The SLB blowdown mass and energy analyses for inside and outside containment are discussed on pages 6-253 268. Note t' at the SLB M&E analyses conservatively assumed an MSlY stroke time of 10 seconds; see page 6 257.

The impact of changes in blowdown mass and energ" was evaluated as described in Sections 2.13 and 2.15 of the UDP Licensing Report. The assessrient of M&E on the containment structure and subcompartments is presented on pages 32 - 37. Tae asses. ment of safety related electrical equipment qualification for inside and outside containment is presentcd on pages 52 - $3.

SCS/ jaw 12/19/97 A SNC/mge 1/7/98 MRC_ Question No. 6 (RefsRace December 22.1997 NRC/SNC Conference Call & NRC Eactimils) in the response, datM August 5,1997, to the staffs Request for Additional information (RAI) regarding spent fuel pool (SFP) cooling for Farley plant operations at the proposed uprated power level, Southern Nuclear Operating Company (SNC) stated that full core omond is a general practice during all routine / normal (planned) refueling outages. Ilowever, SFP temperatures were not calculated for the routine /nonnal refueling outage with full core omoad. SNC further stated l that the heat load (30.3 x 10' Btu /hr) resulting from routine' normal refueling outage is bounded by the heat load (37.0 x 10'Bru/hr) resul ting from emergency full core omoad at the beginning of-cycle (DOC).

Please proside the following infonnation:

(a) The SFP temperature as a function of time during a routine / normal (planned) refueling outage with full core omoad assuming a failure of one SFP cooling train. Information should clearly indicate the duration of time that the temperature exceeds the SFP design temperature of 150'F.

(b) Provide / compare (in detail) the input parameters / assumptions (i.e. time after reactor shutdown prior to core omoaded, fuel assemblics discharged rates, etc.) used to calculate the above cited (routine and emergency) heat loads.

SNC Respsnse No. 6 (a) Farley operates the SFP cooling system with one SFP cooling pump and heat exchanger insenice and the remaining train on standby; simultaneous operation of both trains is not a none' operating practice. As such, the limiting analysis results are based on single train operations.

SNC does not perform time / history transient SFP temperature calculations. Instead, conservative assumptions are used in determining a worst case pool temperature at 150 hours0.00174 days <br />0.0417 hours <br />2.480159e-4 weeks <br />5.7075e-5 months <br /> follow.ng reactor shutdown for beginning of cycle (BOC) and end of cycle (EOC) refucting cases. The EOC full core omoad with one SFP cooling train in senice is considered as the Farley design basis case for nonnal refueling. As such, the maximum Page 6 jj .

calculated SFP temperature calculated for the EOC full core omoad for uprate is 174'F at the SFP heat exchanger inlet. (See pages 30 31 of Attachment I to the August 5,1997 SNC response to NRC Question No. 2, " Questions Regarding A'tachment 6, Section 2 -

Balance of Plant Program Description.")

The BOP Licensing Report (Section 2.10) and the SNC August 5,1997 RAI response (Attachment 1) present engineering analysis results for operation with one and two trains of SFP cooling pumps and heat exchangers. Because two train operation is not a normal plant practice ud parallel SFP pump operation is not advisable with the present Farley design configuration, these two documents must be revised. Specifically, page 25 of the BOP Liecnsing Report and page 31 of Attachment I to the SNC August 5,1997 submittal are bemg resisod to remove references and information pertaining to two train operation.

The corrected pages are included in Attachment 11 to this letter.

(b) The principle consenatisms in the Farley SFP refueling calculations are summarized by the following.

• It is conservatively assumed that the decay heat load and SFP cooling system have instantly reached steady-state equilibrium. No effort was made to take credit for the omoad rate. Instead, the analysis models an instantaneous full core omoad at l the minimum allowable procedural time to commence fuel movement (i. c.,150 l hours after shutdown). '

l

• Cooling water temperature is assumed to be equal to the CCW normal operating design temperature of 105'F, which is based on the peak senice water temperatures that could occur in the summer. Normally, refueling is perfonW in spring and fall when lower senice wat:r temperatures exist.

l l

• Decay heat is maximized with respect to as:embly burnup and the number of assemblies discharged per fuel cycle. A conservative assembly average bumup of 60 GWD/MTU is assumed. The decay heat uncertainty defined in NRC BTP ASB 9 2 is applied for the EOC and BOC full core omoad refueling cases.

No credit is taken for evaporative or convective heat loss from the pool. All spent fuel pool decay heat is assumed to be removed by the SFP heat exchangers.

• Reponed pool temperatures are taken at the SFP heat exchanger inlet. The average pool temperatures are lower than this ieported value.

• For the refueling cases, the spent fuel discharge schedules are continued until the assemblics discharged for the last refueling exceed the pool capacity.

SFP temperatures were not calculated for the "best estimate" full core omoad case (30.3 MBlu/Ilt) since this cas: does not apply the decay heat uncertainty defined in NRC BTP ASB 9-2.

SCS/jvi & mam- t/R/98 A SNC/mge 1/23/98 Page 7

M'_ Question No. 7 (Reference Decernkgr 22.1997 NRC/SNC Conference Call & NRC Eacsimile) I The staff, in its Safety Evaluation dated June 23,1982 and April 15,1983 for Farley SFP re rack application, approved operation of the SFP up tc a maximum SFP temperature of 150'F under routine / normal refueling outages. It is expected that with the assumption of a failure of one SFP cooling train, the maximum SFP temperature resulting from plant operations at the proposed uprated level will exceed the SFP design temperature of 150'F. Provide detailed justification to demonstrate why it is acceptable to perform routine / normal refueling with the potential for the SFP temperature to exceed the design temperature of 150*F. l ENC Response No. 7 l Farley refueling practices and SFP cooling operation are consistent with (and/or more restrictive than) the Farley Technical Specifications and FSAR requirements. Farley does not perform fuel l handling evolutions unless SFP temperatures are below the 150'F limit. Nevertheless, given worst case conditions and the very conservative arsumptions outlined in the response to Question Number 6 above, SFP temperature rould increase to as much as 174'F, tallowing a full core off-load with one train of cooling in senice. Engineering evaluations demonstrate the acceptability of SFP cooling operation at temperatures below 180*F. Additional evaluations and operating procedures address a potential loss of cooling. *Ihe Farley Technical Specifications limiting conditions for SFP operation proside assurance that spent fuel cladding integrity is maintained; there are no Technical Specification Limiting Condition for Operation stipulations for either SFP temperature or thr. cooln.g system.

Plant procedures ensure that the SFP temperature does not exceed the l$0'F assumed in the fuel handling accident (Fila) analysis during fuel handling evolutions. Specifically, the Fila analysis has addressed all required loads on the racks up to 150*F. Since the plant procedures require all fuel handling operations to be secured at the SFP high temperature alann of 130'F, the 150'F temperature ascumption for the Fila is not exceeded. Also, it shoula be noted that the SFP demineralizer is isolated by the plant operators before reaching 140'F to prevent resin degradation.

Per cooling system operation above 150*F, with the demineralizer isolated and fuel handling operations prohibited, engineering evaluations have demonstrated acceptability of the SFP system components, including the pump, pool racks, liner and concrete structure, provided 'emperatu4es remain at or below 180*F. It is therefore concluded that 180'F is an acceptable temperature limit for the SFP cooling system operation for full core ofiload with one cooling train in senice.

For a complete loss of SFP cooling event, it has also been determined that adequate design margin exists in the concrete stmeture, racks, and liner for temperatures corresponding to SFP boiling. As discussed in the response to Question Number 8 below, plant procedures provide guidance to mitigate such an occurrence and ensure that SFP minimum water levels are maintained.

With regard to the FNP Technical Specifications and SFP design, the spent fuel pool (free surface) and pool storage racks (natural circulation) ensure that fuel cladding integrity is protected without forced coolingm in the event ofloss ofinventoru due to evaporation, makeup water sources are available to compensate for pool inventory losses. Makeup water can be provided by the demineralized water system or by the seismic category 1 portion of the reactor makeup water system. In addition, borated water can be supplied from the RWST or from the reactor makeup &

Page 8

boric acid system blender. Evaporative losses are normally made up from the demineralized water system. Since SFP coolant inventory is assured by these multiple sources and plant operating procedum, safe storage of the irradiated fuel is riot directly affected by the maximum SFP temperatures which can occur with the coolinsystem in operation. As discussed above, the limiting concerns for SFP cooling system operating temperatures are maintaining structural integrity and remaining within equipment limits, it is concluded that it is neceptable to perform routine / normal refueling evolutions prosided CFP temperaturer continue to satisfy the FHA analysis assumptions. It is also acceptable to operate the SFP cooling system up to 180'F. Plant design and procedures provide assurance that minimum SFP water level is maintained.

SCS/jvi & mam !/8/98 & SNC/mge,jmg & rwa 1/22/98 NRC Ouestion No. 8 (Refqtrage December 22.1997 NRC/SNC Conference Call & NRC Facsirails)

In the above cited response, SNC also stated that plant procedures require a minimum of one operable cooling tinin during routine / normal refueling outages. Smcc full core ofiload is normal practice d ning all routine / normal (planned) refueling outages, this requirement is incorristent with the guidance described in SRP Section 9.1.3 which recor mends that a single failure proof SFP I l cooling system be available prior to routine /nomul outages. Providejustification as to how this l operational practice meets single failure guidance.

SNC Response No. 8 The Parley SFP Cooling System is designed to meet the General Design Criteria (GDC) in 10 CFR 50, Appendix A Criterion 61," Fuel storage and handling and radioactivity control." This criterian requires that systems which may contain radioactivity shall be designed to assure  !

adequate safety under normal and postulated accident conditions. The FNP PSAR stated that the spent fuel pool cooling system is designed to remove the heat generated by stored spent fuel clements from the spent fuel pool. NRC Question 9.11, during the PSAR stage, noted:

[The design basis given in the] PSAR for the Spent Fuel Pool Cooling System state [s] that an alternate cooling capability will be provided for anticipated malfunctions of the system, which contain only one heat exchanger, one pump, and a single pipe. Describe the procedures that will be used for connecting other equipment into the system in order to obtain the alternate cooling capacity.

In response to this question, FNP stated that the "rufundant heat exchangers and pumps each wil 100 percent capacity" are part of the system design.

The NRC stated in Farley SER dated June 1975 that they had:

evaluated the spent fuel pool cooling and cleanup system design and determined that the cooling subsystem meets the single active failure criterion' including redundancy of emergency buses. We also have determined that the system conforms to the n girements of GDC 61. . (and) have concluded that the system design is acceptable.

Page 9

4 . ,

De FNP FSAR (refer to Sections 9.1.3.2 and 9.1.3.3) states that:

The spent fuel pool cooling and cleanup system has na emergency function during an accida This manually controlled system may be shut down for limited periods of time ior maintenance or replacement of malfunctioning co.nponents. In the event of a failure of a spent fuel [ pool cooling) pump or loss of cooling to a spent fuel heat exchanger, the second cooling traiu provides 100 percent backup capabi'ity . .

. . In the unlikely event of the failure of both trains of the spent fuel pool cooling system and the demineralized water system, reactor makeup water can be added to the spent fuel pool by the use of a temporary hose connection 1

ne Farley operating and surveillance procedures assure compliance with Technical Specifications l SFP minimum level requirements. As long as SFP inventory is maintained (i.e. m:nimum level),

l fuel integrity is assured.

i Plant policy and philosophy regarding availability of safety related equipment during outages is detailed in FNP 0 AP 94," Outage Nuclear Safety," an administrative procedure, and in FNP M-47, the " Outage Planning Manual." The outage nuclear Safety philosophy has been developed in accordance with guidance provided by NUMARC (now NEI). He outage schedule is developed to provide " defense in depth" capability to ensure backups are availabic for key safety functions whenever possible, and the schedule is resiewed prior to the outage by an Independent Safety i Assurance Committee. SFP cooling is specifically identified in the above documents as a key safety function. The outage schedule is developed using ' system windows" to ensure .inimum downtime for safety systems and to ensure the level of system availability for systems supporting key safety functions is maintained above Technical Specifications minimum requirements as much as possible. The condition of each key safety function is assigned a green, yellow, orange, or red classification Green signifies the highest level of avai: ability. Yellow denotes a reduced but acceptab!c level, while orange indicates a severely reduced level, and red is an unacceptable level.

A risk assessment and contingency plans are required for any situation resulting from an intentional entry into a yellow condition, or for situations where an orange or a red condition could be entered unintentionally. For loss of spent fuel pool cooling, abnormal operating procedure FNP- j 1/2 AOP-36.0 provides specific instructions to restore cooling and/or maintain inventory. The procedure first attempts to start the standby train ifit is available. If this is unsuccessful, any fuel handling in progress is suspended, and the SFP is isolated from the refueling cavity by closing the transfer tube gate valve. An attemate cooling scheme is then attempted whereby hot SFP water is pumped to the RWST and replaced with cooler RWST water, making sure that the SFP level is maintair ed within specification during the transfer evolution. if the SFP temperature continues to rise to the point that irwentory is being lost due to evaporation, then either the demineralized water or the reactor makeup water systems will be used to replace the loss and maintain SFP level.

SNC/wTm t/R/98 & mge A jmp 1/14/98 NRC Ouestien No. 9 (Refettnce January 7.19981997 NRC/SNC Conference Call & NRC Eassimile)

Regarding the interruption of ECCS flow in the swapover process, your response to question 18 in

~

your Nov-mber 19, M97 submittal indicates that the low head safety injection pumps are tripped in the recirculation process. Your response goes on to state that the initial version of the FSAR included this sequence. The staff safety evaluation (NUREG Oll7 Supplement 2) discusses the Page 10

swapover process, however, it does not discuss manual tripping of the LPI pumps. It does state "that automatic tripping of the pumps was unacceptable" because it did not assure a continuous supply of water to the core.

Please provide references where the manual tripping of the LPI pumps v,as proposed and subsequently approved by the staff. Additionally, please verify that the LPI pump vendor recommendations indicate that re-starting the LPI pumps less than four minutes after being tripped is acceptable.

Your response to question 18 also states that "To quantify the potential impact of the tempo'ary low head ECCS flow interruption would require a new analysis." long term cooling must be shown to be acceptable. As a result, if the swapover process causes the core to uncover or cause a s cond heat up, the impact must be evaluated Please provide an evaluation that evaluates the impact of the recirculation process or show that there is n- nd heat up.

SNC Respsnse No. 9 Table 6 3 3," Sequence of Changeover Operation from injection to Recirculation," which specifically ioentificd that the RilR pumps (i e , LilSi pumps) are temporarily stopped during the initial steps of the switchover process, was included in the original Farley FSAR. 'lhe NRC issued the initial SER in 1975 which included the following:

We have reviewed the applicant's sequence of change over operation from injection to recirculation (FSAR Table 6.3-3) and have concluded that there must be an 3 automatic bacbp to the manual switchover from injection to recirculation cooling , ,

We require this design change to be implemented prior to issuing an operating license for the Farley plant.

A!though the NRC requested that Farley consider a design change and simplified procedure for starting the recirculation phase of ECCS after the injection phase is completed, the only outstanding issue was the automatic backup for switchover. The Farley design was modified to ensure that this automatic requirement was implemented and, as a result, Supplement 2 to the Farley SER, issued on October 15,1976, addressed the resolution of this issue. This SER states that:

Upon annunciation of a low level alarm, the operator will perform a manual switchover in accordance with the emergency operating procedures established for the original design We conclude that the design bases for an automatic backup to the manual switchover of the emergency core cooling system from the injection mode of operation to the recirculation mode of operation are acceptable for the Farley plant.

The Farley emergency procedures are consistent with the temporary ECCS flow redation resulting from the specified sequence (i.e.,1111S1 flow is continuously maintained during the short interval LilSI flow is secured), Farley system operating procedures are also in agreement with the manufacturer';; motor start limitations such as two successive starts from ambient and fifteen minutes run time for subsequent starts. In that the switchover to ECCS recirculation occurs

- greater than fifleen minutes following initiation of ECCS injection, restarting the RiiR pumps is acceptable.

Pagei1 a ~

o As noted in the response to NRC Question No.18 of the November 19,1997 SNC submittal, a new analysis would be required to quantify the potential impact of the temporary low head ECCS Cow interruption. Be:ause the NRC approved Farley licensing basir included the temporary low head ECCS flow interruption and because the duration of the temporary interruption did not change, SNC does not plan to perform such a detailed analysis.

W/rcs 1/14/98 A SNC/rnre,un &png 1/14/98 NRC Outs 11on No.10RsfrigorgAriuary 1.1998 NJ1C/SNC Conferengs_QillA NRC Facsimile)

With regard to the loss of offsite power (LOOP) assumptions in the accident analysis, your response to question 20 in your November 19,1997 submittal describes the mechanistic LOOP due to grid instability following the r: actor / turbine trip and an associated two second delay. He chapter 15 accident analysis assumptions are generally intended to bound all the possible scenarios.

The LOOP is generally assumed to occur coincident with the event or a more limiting time for the LOOP is chosen to bound all possible scenarios. For example, SRP section 15.1.5 " Steam System l Piping Failures inside and Outside of Containment" indicates that " Assumptions as to the loss of offsite power and the time ofloss should be made to study their effects on the consequences of the I accident." Although different bounding or mechanistic usumptions have been approved, these are l

done for specific applications (i.e., for the locked rotor event a mechanistic delay for the LOOP has been approved). For each of the transients please state the LOOP assumption, provide a justi6 cation and indicate ifit has received NRC approval for that application at Faricy.

SNC Response No.10 With respect to loss of offsite power, no non-LOCA event safety analyses assumption was : hanged for the Farley Power Uprate Program. For any event modeled in the non LOCA safety analyses that relies on ESF actuation for protection (with the exception of the locked rotor event), it is necessprv to demonstrate that the acceptance criteria are met with and without a loss of offsite power. For Farley this includes the main steamline break, feedline break, and loss of normal feedwater events. For these events, the loss of offsite power is assumed to occur as a consequence of the reactor trip and is modeled to occur two seconds following the turbine trip. (Sec response to NRC Question No. 20 in SNC letter dated November 19,1997.) This analysis assumption is consistent with the original NRC approved Farley licensing basis, subsequent NRC-approved Farley licensing submittals, such as upgrade to VANTAGE 5 fuel and steam generator level tap relocation, and other licensed Westinghouse PWR's with non LOCA analyses performed by Westinghouse.

f As required to support application of the NRC approved Westinghouse Best Estimate LOCA

(

methodology for large breaks, the response to NRC Question No.12 in SNC letter dated November 19,1997 indicates that a LOOP study confirmed that ofTsite power available is more limiting than LOOP.

For SBLOCA, the response to NRC Question No.14 in SNC letter dated November 19,1097 and Section 6.1.2 of WCAP-14723 (page 6-9) provide the basis of the LOOP assumption. This SBLOCA analysis as>umption is cons 5 tent with the post-TMI NRC-approved Farley licensing basis and subsequent NRC approved Farley licensing submittals, such as upgrade to VANTAGE 5  !

fuel.

1 Page 12 l

For SOTR, Section 6.3 of WCAP 14723 (page 6 202) defines the LOOP assumptions, which are consistent with the NRC approved methodology for Farley.

For LOCA/MSLB ht&E releases, Sections 6.4 and 6.5 of WCAP 14723 (pages 6 214 - 6 215 &

6 255) delineate the LOOP assumptions consistent with the standard methods previously used for Farley, While new containment, subcompartment, main steam valve room, and rudiological analyses and/or evaluations were performed for the Farley Power Uprate Project, the LOOP assumptions were not changed.

W/rjm + 1/1$/98 A SNC/mge t/1$/98 & AS/ jaw 1/15/98 NRC Ouestion No. I1 (RticiturcAnuary 7.1998 NRC/SNC Conference Call & NRC Facsimile)

Because the pressurizer PORV is credited to open automatically after the associated block valve is opened in the FS AR Chapter 15 transient analysis, please verify that the automatic opening function of the PORV is performed with only safcty related components and that the function can be accomplished assuming the most limiting single failure.

SNC b2gnse No. ll NUREG 1316 and GL 90-06 do not require upgrade of the PORVs and block valves (and associated control circuits) to full safety grade qual, ication status. Farley has implemented the backfit delineated in Generic Letter %-06 for improving the reliability and availability of the PORVs. The NRC-approved Technical Specifications changes included: enhanced PORV and block valve surveillance testing and expanded LCO requirements; and maintaining electrical power to closed block valve (s). These valves are included in the plant QA, IST and Phi programs, and certain replacement parts are procured in compliance with 10 CFR 50 Appendix B. The block valves are also included in the plant ht0V program. Periodic testing and calibration of the valve control circuits, including the autonutic pressurizer pressure control circuits, are accomplished by plant maintenance and/or surveillance testing procedures. In addition, as part of the .csponse to Thil action plan, PORY improvements were implemented at Farley, such as electrical power supplies from Class IE buses .md qualified valve position indication.

NUREG 1316 also concluded that the PORVs and block valves can be used to perfomi certain safety functions or "any other safety-related function that may be identified in the future." The significant hazards evaluation (10 CFR 50.92) included with the SNC October 13,1992 Technical Specifications change submittal, which was approved by the NRC, stated that manual PORV operation as required by the emergency operating procedures is a safety function performed by the PORVs. For an inadvertent SI, Farley FSAR section 15.2.14.1D states,"[mlanual operation of the PORVs and block valves is assumed." This analysis assumption is in concert with the analysis modeling and the Farley emergency response procedures, which require manual operator actions, if necessary, to open at least one block valve and the associated PORV. hianusi actuation of the PORVs is performed by use ofindependent trains of safety related equipment. Based on the above, it is acceptable to credit operation of the PORVs; furthermore, it can be concluded that automatic PORV actuation by the pressurizer pressure control system is n01 required to mitigate the ccnsequences of an inadvertent Si at power.

Page 13

.o it is expected that the Farley PORVs will automatically open if required. %c normsl system operating alignment is with both block valves open and both PORVs closed in the automatic cont of mode. Automatic actuation ci,cuits for the PORVs, although coraidered nonsafety-related, are routed separately. Two separate pressure transmitters are utilized, and the cables are routal in accordance with requirements described in FSAR section 8.3.1.4. The cable and transmitters are the same as used in Class IE applications. All control system components are supplied from Class IE power supplies. De signal processing components (part of Westinghouse 7K0 Process Control System) have been procured as safety related. De actuation circuit uses Class IE relays ,

for isolation and non Class IE designated relays for control, which are reocured as safety-related. j nerefore, the automatic actuation circuits provide a highly reliable back-up actuation path for ac  ;

safety related manual actuation circuits associated with the Farley PORVs.

IWwp A jet .1/16/98 & SNC/mge 1/15/98 NRC Ouestion No.12 (ReferenceJanuary 7.1998 NRC/SNC Conference Call & NRC FacsimilA With regard to the analysis of the main steam line break, RAI dated October 14,1997, question 29 asked if the plant cooldown caused by a main steam line break (MSLB) could cause the plant to cool down to the point where brittle or ductile failures of the reactor vessel could occur. Your response indicated that the main steam line break does not provide the basis for the evaluation of brittle and ductile failures. Section 15.1.5 of the SRP states that " Pressure in the reactor coolant and main steam systems should be maintained below acceptable design limits, considering potential brittle and well as ductile failure." Please verify that the MSLB will not cause potential brittle or ductile failures or provide a technical argument indicating that this is not necessary.

Your response to question 29 indicates that the flow area assumed in the MSLB analysis is the l cold, nominal flow area. Your response indicated that although this assumption is not conservative, other factors (assuming Fl/d=0 and no moisture carryover) make the overall analysis I conservative. Please verify that using a flow area associated with cold, nominal dimensions of the integral nozzle is consistent with your licensing basis. Ifit is not, to s.,pport this conclusion please evalusta how much the flow area is under predicted if the area were calculated using the hot (normal operating temperature), upper tolerance dimensions. Den using more realistic v0ues for the head loss and moisture carryover, evaluate how much the flow is being [overpredicted] and compare the results. Additionally, please verify that the conservatisms in the head loss and moisture carryover assumptions are not being credited elsewhere in the analysis to iustify other non-conservative assumptions.

SNC Response No.12 The steamline break analysis performed for Chapter 15 is performed to evaluate core response and not to explicitly evaluate brittle or ductile failure of the RVs. liowever, the Main Steam Line Break (MSLB) does contribute to the Pressurized Hennal Shock (PTS) risk associated with brittle failure of the Reactor Vessel which was evalunted for Uprate conditions.

A Pressurized nermal Shock (PTS) Screening Criteria was established by the NRC in the early 1980s (10 CFR 50.61). Generic analyses were performed by the Westinghouse Owners Group and the NRC in the early 1980s to establish a risk basis fo. the PTS Screening Criteria. Those analyses included Probabilistic Risk Assessment (PRA) and nermal and liydraube Transient System Evaluations and Structural Reliabi ty and Risk Analysis (SRRA) of the Reactor Vessel (RV). He generic analyses identified the events that contributed to RV failure risk, e g., MSLB, Page 14

0

• , e and then assessed the cumulative cfTect of all contributors to RV (1ilure risk. The NRC identified an acceptable RV failure risk as part of the basis for the PTS Screening Criteria. The NRC then identified a RV Irradiation embrittlement level (RTpts) that would assure that the RV failure risk would not be a significant contributor to core melt. This " acceptable" RV irradiation embrittlement level is the PTS Screening Criteria dermed in 10 CFR 50.61.

%e intent of the PTS Screening Criteria is to enable a cumulative assessment of the efket of all system transient events on the integrity of the RV. The PTS Rule (10 CFR 50.61) is a regulatory requirement that assures that the cumulative contribution of all system events to reactor vessel integnty will be acceptable as long as the RTpts (irradiation damage level) of the RV is maintained below the PTS Screcuing Criteria. As part of the upmting evaluations, the RTpt., values in support of Fower Uprate for the Joseph hl. Farley plant have been evaluated and am substantially below the PTS Screening Cnteria defined in the PTS Rule.

Since the Joseph hl. Farley PTS evaluations have shown that the plant specific RTpts values in support of Dower Uprate are below the PTS Screening Criteria, tM RV is being maintained bel?w acceptable "desir,n limits,"i.e., the PTS Rule requirements. Since the RTpt: values are

! acceptable, i.e., below the PTS Screening Criteria, no other RV brittle fracture evaluations are required to demonstrate acceptable RV integrity per the definition of the PTS Rule.

The Large Steam Line Break is classified as a Faulted (Level D) Condition in the design transient specifications. As such only one occurrence of the transient must be considered in the rea: tor vessel stress report, and only the primary stress (pressure) effects must bejustified with the Faulted (Level D) Condition limits. ash 1E Section 11! does not require evaluation of the seconda.y stress (thermal) effects since there cannot be " cycling" to failure by definition of Faulted (Level D) conditions. Even though the thermal stress can be very high, only gross distortion will result in the ductile mode as the thennal stress is relieved. Faulted condition criteria permits such gross distortion of a component.

The Small Steam Line Break is classified as an Emergency (Level C) Condition. AShtE Section til limits the total number of occurrences of Emergency Condition transientr to 25. As with I Faulted Conditions, the secondary thermal stress efTects are not evaluated for Emergency Condition transients due to the low number of cycles for the infrequent transients. Only the primary stress effects must be evaluated, and the small number of cycles are considered " free" and need not be considered m the fatigue analysis.

The manmum break area used in the h1SLB uprate analysis is consistent with the Farley licensing basis and was not changed for the uprate analysis. Based on our conversation on January 7,1998 this satisfies the concem regarding the break size assumed in the h1SLB analysis.

J W/knk & gis & tcs .1/13/98 Page 15

o .

)

ATTACitMENT 11 SNC Response to NRC Request For AdditionalInformation Related To Power Uprate Submittal . Joseph M. Farley Nuclear Plant, Units 1 & 2 CORRECTED PAGE NO. 25 "FARLEY NUCLEAR PLANT UNITS 1 AND 2 POWER UPRATE PROJECT BOP LICENSING REPORT" (ATTACllMENT 6 TO SNC SUBMITTAL DATED FEBRUARY 14,1997) l

[ CORRECTED PAGE NO. 31 1

"FARLEY NUCLEAR PLANT RESPONSE TO REQUEST l FOR ADDITIONAL INi'ORMATION RELATED TO POWER UPRATE FACILITY OPERATING LICENSE AND TECliNICAL SPECIFICATIONS CilANGE REQUEST" (ATTACllMENT I TO SNC SUBMITTAL DATED AUGUST 5,1997) i 2

__ = == -

o ,

to the dxty heat results. Case 5 estimates the maximum spent fuel pool heat load after an outage for use in evaluating the component cooling water system and ultimate heat sink.

Acceptability of Spent Fuel Pool Bulk Water Temocratures For the partia'-core ofiload case, the pool bulk water temperature is maintained sl50'F. For the full-core offload cases, pool bulk water temperature remains s180'F. These analyses results are based on conservative input assumptions with one SFP cooling pump and heat exchangcr :n operation. For the "best estimate" fAl-t. ore omoad case with no uncertainty factors applied, the pool bulk water temperatures are bounded by the BOC and EOC full core omor.d cases due to the lower total heat load on the spent fuel pool.

The results for the partial core omond case (pool water temperature sl50'F ) are within the l maximum normal operrting temperature limit for the pool, rack, and cooling equipment, but above the demineralizer resins maximum operating temperature limit of 140'F. A temperature switch, which activates a high temperature alarm whenever the pool temperature reaches 130'F, can be used to alert the operators to manually initiate action to maintain the integrity of the cleanup '

system. Plaat procedures require demineralizer isolation before the 140'F temperaturt limit is cr.ceeded.

Engineering evaluations have demonstrated that the SFP and associated cooling equipment arc acceptable for use up to 180 *F. De behavior of the SFP structure under the appropriate combined dead, live, scismic, and a creased thermal badings was evaluated, and sufficient margin is maintained to ensure structural integrity. For the spent fuel racks, the revised themial loading was considered, and it has been judged that the resised thermal condition will not adversely affect the qualification of the spent fuel racks. Thus no changes to the SFP cooling or cleanup systenu are required to support uprate, SFP Makeuofrime Available Before Bulk Pool Bailing on Loss of SFP Cooling '

Although a complete loss of cooling is a beyond design basis-event, the time available from the l high temperature alcrm before bulk pool boiling occurs exceeds 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />. For the maximum decay heat load case (BOCfull core ollload), a maximum makeup rate of 76.4 gal / min was calculated to be required to maintain pool level if boiling occurs. Makeup can be provided directly from the RWST by the refueling water purification pump, by the demineralized water system, cr by the reacter makeup water hose station located on the operating area adjacent to the pool, which  ;

requires only a minimal hookup time and can be supplied by either one, or both reactor makeup water pumps.

Fuel Assembly Clad Temperatures With Loss of SFP Cooline For assembly exposure of 60,000 MWD /MTU with a peaking factor of 1.7, the maximum fuel assembly clad temperature at 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> after shutdown was calculated to be 257.I'F. This ten.perature is well below a temperature at which damage to the fuel cladding occurs. He maximum heat flux in the fuel assembly was determined to be approximately 1600 BTU /hr ff, which is considerably less than the critical heat flux (onset of film boiling) for these conditions.

BOP UPRATE LICENSING REPORT 25 FNP - UNITS I AND 2 lay 98 l

b. Is full core omoad a general practice for routine refueling? Ifit is, how many trains of the SFP cooling system will be available/ operable prior to refueling operation?

SNC Response No. 2

a. He following table provides the requested information. Note that temperatures were on!y calculated for cases 13 to demonstrate compliance with the Standard Resiew Plan.

Temperatures were not calculated for the Best Estimate Full Core Omoad or post-l refueling cases. He total heat load for the Best Estimate Full Core Omond case is l bounded by the DOC and EOC Full Core Omoad cases. He post refueling cases were i analyzed only for heat loads used in evaluating component cooling water system performance and the ultimate heat sink maximum post accident temperature.

Case # 1 2 3 4 5A Sil SC Partial Core full Core Full core Full Core Post Post Post Case Omond Omood Omond Omoad Refueling Refueling Refueling Desenption (IKX') (EOC) (llest Estimate) (25 days) (40 days) (65 days) lleat toad 22.1 37 0 36.5 30.3 l$.4 13.6 12.0 (MifftJ/hr) l Max. SFP 147 175 174 - - - -

Terrperature l

(

• F)

b. Full core omoad is a general plant practice. Plant procedures require a minimum of one operable cooling train. Plant procedures further require that fuel handling operations be suspended and actions taken to restore cooling upon receipt of the high temperature alarm, j SCS/pi . 7/24/97 SCS/dwm 1/23/98 l

(

1/23/98 Page 31 pwrupt 8. doc l