ML20153A981

Forwards Response to NRC Bulletin 88-004, Potential Safety- Related Pump Loss. Core Spray Sys Designed to Provide Min Flow of 320 Gpm Per Pump

ML20153A981
Person / Time
Site:	Peach Bottom, Limerick, 05000000
Issue date:	06/30/1988
From:	Kowalski S PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC
To:	Rossi C NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM), Office of Nuclear Reactor Regulation
References
IEB-88-004, IEB-88-4, NUDOCS 8807120673
Download: ML20153A981 (16)

Text

e: -

-r . .

^ PHILADELPHIA ELECTRIC COMPANY  :

2301 MARKET STREET P.O. BOX 8699 I PHILADELPHI A. PA.19101 g215) 8414502

"?$s7, ,$8$ June 30, 1988

.m..............

Docket Nos. 50-277 50-278 50-352 50-353 Mr. Charles E. Rossi, Director Division of Operational Events Assessment Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555

Reference:

NRC Bulletin No. 88-04 Potential Safety-Related Pump Loss

Dear Mr. Rossi:

.\

NRC Bulletin (NRCB) 88-04, "Potential Safety-Related Pump Loss", issued on May 5, 1988, requested licensees to investigate two potential. design concerns involving safety-related centrifugal pumps.

The concerns include the potential for a pump to dead-head when it is operating in the minimum flow recirculation mode in parallel with another pump, and also includes concerns over the design adequacy of pump minimum flow capacities.

The NRC requested that within 60 days of receipt of NRCB 88-04, Philadelphia Electric Company provide a written response:

a) to describe the problems and the systems affected, bi to identify any necessary short-term or long-term modifications to plant equipment or to plant operating procedures which are being implemented to ensure safe plant operations, c) to provide a schedule for long-term resolution of significant problems identified as a result cf thia bulletin, and bR OC U -

[f fi  ;

~--

~*-

• Mr . Cf. E. Rossi June 30, 1984 Page 2 d) to provide justification.for continued operation of the plants with particular attention being placed on the General -

Design Criterion 35 of Appendix A to Title 10 of the Code of Federal Regulations (10CFR50), Emergency Core Cooling" and 10CFR50.46, "Acceptance Criteria for Emergency Core Cooling System for Light Water Nuclear Power Reactors."

Attachment "A" to this letter provides our response and affidavit to the NRC for both Limerick Generating Station, Units 1 and 2, and Peach Bottom Atomic Power Station, Units 2 and 3.

Should you have any questions or require additional information, please do not hesitate to contact us.

Very-truly yours, Attachments cc: William T. Russell, Administrator, Region I, T. P. Johnson, USNRC Senior Resident Inspector T. J. Kenny, USNRC Senior Resident Inspector l

t

PHILADELPHIA ELECTRIC COMPANY DOCKET NOS. 50-277 ATTACFf4ENT "A" 50-278 50-352 50-353 60-DAY RESPONSE TO NRC BULLETIN 88-04 "POTENTI AL S/FETY-RELATED PUMP LOSS" NRC Bulletin 88-04, "Potential Safety-Related Pump Loss",

requested Ilcensees to investigate two potential design concerns involving safety-related centrifugal pumps. The two broad NRC concerns discussed in the bulletin include:

1) the potential for a pum to dead-head when it is operating in the minimtm flow recirculation mode in parallel with another ptmp (pum-to-ptmp interact lon), and

2) the adequacy of the minimtm flow capacity.

Philadelphia Electric Ca pany's understanding of these two concerns is suTmarized here in greater detail.

1) Punp-to-Ptmp Interaction When the minimtm flow Ilnes from two or more ptmps join at some point to form a calmon line, there is a potent tal for interaction between the pums. If the design of the piping configurat Ion has not considered the pum unique performance characteristics, the pump with the higher discharge pressure (stronger pum) could reduce the flow through the ptmp with the lesser discharge pressure (weaker ptmp) to the po!nt where it is incdequate for 1cng-term Integrity.

If the pums' minimtm #10w lines are orificed (back-loaded) in the Individual pum discharge lines upstream of the Junction and if the camion line has a large enough flow area such that its resistance is a relatively small part of the overall hydraulic resistance, there should be little adverse pum-to-pum interact lon. The parallel pums can be expected to operate Individually or in unison with adequate minimtm

-flow.

Haveve r, if the m!nimtm f discharge lines are not Individually orificed, but the cmmon line is orificed or contains no orifice, interaction between the two pums may occur. The severity of the degradation of mintmtm flav through a pum depends on the shape of the pum characteristic head-flow curves and the mismatch between the pums. If the characteristic curve is such that a stall change in flow results in a relatively large change in developed head, it is probable that little operational difficulty would result frcm an undesirable piping configuration. However, if a relatively large change in flav resulted in only a small change in developed head, some problems could be expected in satisfying the minimtm flow requirements.

c. : ,. - ,

PHILADELPHIA' ELECTRIC CCNPANY DOCKET NOS. 50-277 ATTACWENT "A" 50-278

. 50-352 50-353

2. Adequacy of Minimun Rt. circulation Ficw The original design basis fcr sizing the minimun ficw recirculation lines for the safety-related ptmps at Limerick Generating Station (LGS) and Peach Bottom Atomic Power Station (PBAPS) was to provide sufficient flow to avold.

overheating the puros due to low flow. However, . nore recently, purp vendors' guidelines for minimum flow are based on avoiding hydraulic Instability in addition to avoiding pum overheating, Iceding to higher reccanended L minimun flow values than those used in original system  !

( design. Hydraulic Instabilities can occur at low flow rates due to flow separatton across the Irme11er vane, which can l

(  !

lead to asynmetrical shaf t and bearing loads in addition to purp and piping vibration. Since the purp vendor guidelines are only applicable for "centinuous" or "Intermittent"'

operation, there are no new guidelines which specifically address low flow 11mits for Infrequent operation, as is the case for BWR Emergency Core Cooling . System Pums.

Philadelphia Electric Ccmpany, the holder of an operating Ilcense for LGS Unit 1 and PBAPS Units 2 S 3, and holder of a

' Construction Permit for LGS Unit 2, has cortpleted a preliminary investigation into the appilcability of the two NRC concerns to PBAPS and LGS. From thIs investIgatlon, we have ident!fIed the safety-related systems potentially affected, and have developed both short and long tonn resolutions for the concerns. This infonnation, including Justification for j continued operation is included in this response.

l I. Affected Systems l

A. Limerick Generating Station I

The LGS safety-related systems which contain centrifugal I pumps, and are therefore potentially affected are the Residual Heat Removal (RHP.) System, the Core Spray System, the High Pressure Coolant Injection (HPCI) System, the Reactor Core Isolation Cooling (RCIC) System, the Safeguard Piping Fill System, the Diesel Generator Fuel Oil Transfer System, the Emergency Service Water (ESW) System, the Residual Heat Removal Service Water (RHRSW) System, the Reactor Recirculation System, and the Control Structure Chilled Water l- System.

l l The ESW and RHPSW purps are vertical, wet pit turbine

) pums . No mininun flow recirculation lines are required for these puros because the systems are always aligned to provide an open flowpath; i .e. the purps are not operated dead-headed.

l

phi ADELPHIA ELECTRIC COMPANY DOCKET NOS. 50-277 ATTACm ENT "A" 50-278

'50-352 50-353 The Diesel Generator Fuel Oil Transfer punps do not require minimtm flow bypass .l ines. No m!nimtm flow recirculation IInes are required for these ptmps because the systems are always aligned.to provide an opM flowpath; i .e., the ptmps are not operated dead-headed.

/ The Control Structure. Chilled Water pums are vertical, centrifugal, in-l i ne punps. No minimtm flow recirculation j IIne is required for these pums because the system is ,l designed for enntinuous flow. Most of the cooling coils which 1 l.

are "loads" on the system are provided with bypass lines, so j that, if cooling is not needed, enough ficw is bypassed around I the colls to provide adequate minimtm ficw.

The Reactor Recirculation pums are variable speed pums I controlled by a trotor-generator (M-G) set. The punps are

! prevented frcm running below approximately 28% rateri speed by the master speed 1imiter. Since there are no parallel

< pump paths, there is no potential for pum-to-ptmp Interaction.

The Limerick Control Rod Drive (CRD) hydraulic system punps are not safety-related. The minimum flow bypass lines for

!= the "A" and "B" CRD pums join together into a ccmnon line.

However, each Individual minIntm flow line contains an orifice. Therefore, as described in the stmnary, there should be ' little adverse pum-to-pum interaction. In addition, normally only one CRD pu m is in service.

The renalning safety-related LGS systems which may be affected are discussed here.

Core Spray System

1. Punn-to-Pum Interaction The minimum flow bypass lines for the "A" and "C", and for the "B" and "D" Core Spray pums are joined into ccrrbined minimtm ficw lines with a single orifice in the combined line. Therefore, there is a concern for possibly dead-heading the "weaker" of the two parallel pums due to pum-to-punp Interactlon. An asseLsment of the design adequacy of the mir,lmum ficw bypass lines for both Unit 1 and Unit 2 Core Spray syst in 1 cops has been ccanpl eted. The results of this assessirent have shcwn the system design to be adequate. Additionally, minimtm flow testing will be perfonned to assure required I minimtm ficw values are obtained on all Core Spray f loops. The Unit 1 "A" Core Spray Loop was tested on l

June 23-, 1988. The results of this test are presently being reviewed.

PHILADELPHIA ELECTRIC COMPANY DOCKET:NOS. 50-277

. ATTACFNENT "A" '50-278 P 50-352 50-353 The design calculations for the Unit 2 Core Spray punps l demonstrated that, with both puTps in a Core Spray loop

! operating through the minimtm flow bypass IIne, the flowrates are expected to be approximately 358 and 340 i grm per puro. (The flows in the other loop are ccrnpa rable. ) This corresponds roughly to a 6% margin for the "weaker" punp ccynpared with the mininun specified fIcw of 320 gyn.

2. Adequacy of Mininun Recirculation Flow Tha Core Spray system has been designed to provide the specified minimun flow of 320 gpm per puro, or 10% of rated pump capacity of 3175 gpm. The flow restricting orifices in the ccrnbined line were originally sized for 400 gpm per puro, or 800 ggn through the orifice with both purps In a loop running to provide a margin.

Residual Hoat Removal System

1. Purn-to-Puno Interaction Each of the four RHR purps for each unit at Linerick is-provided with an Individual mininunsflow bypass line.

Therefore, there Is no concern for purp-to-purp Interactlon. The mininun ficw line;s for the "C" and "D" pums are routed to the suppression pool directly.

The mininun flow Ilnes for the "A" and "B" puTps are routed to the suppression pool via the full flow test return lines.

2. . Adequacy of Mininun Recirculation Flow The RHR system has been designed to provide the specified mininun ficw of 1000 ggn per pum, or 10% of rated flow. The ficw restricting orifice in each RHR purp l

minirrt.m flow line was sized for 1100 ggn flow. Thus, there is approxinetely a 10% tmrgin above the specified mininun flow.

l Reactor Core Isolation Cooling System I

1. Purp-to-Puro Interaction The mininun flow line for the RCIC purp is not interconnected with any other Ilne; therefore, there Is no possibility of puro-to-purp Interaction.

PHILADELPHIA ELECTRIC COMPANY . DOCKET NOS. 50-277

' ATTACFfENT "A" 50-278 50-352 50-353

2. Adequacy of Minimtm Recirculation Flow.

The minimtm flow line from the RCIC ptmp discharge to the suppression pool is provided with two orifices and a normally-closed valve which opens after the pum

~

starts and develops discharge cressure If flow in the

. min-RCIC flowpath Is Icw and automatically recloses once adequate flow is estabilshed. The system has.been designed to provide the specified minimtm flow-of 60-120 gpm, or 9%-19% of rated punp capacity of1610 gpm.

High Presst're Coolant injection Systerr

1. Pum-to-Ptmp Interaction The minirntm flow. line for the HPCI punp is not interconnected with any other line; therefore, there is 'I no possibility of ptmp-to-pum Interaction. {

1

2. Adequacy of Minimtm Recirculation Flow The minimtm ficw line from the HPCI pum discharge to L. the suppression pool is provided with two-orifices and a normally cic3ed valve which opens after the pum starts and develops discharge pressure if flow In the main HPCI flowpath is low and automatically recloses once adequate flow is established. The system has been

! designed to provide the specified mlnlmtm flow of 300-600 gun, or 5%-10% of rated pum capacity of SE00 gpm.

Safeguard Piping Fl11 System

.1. Pum-to-Pum Interaction Each of 'the Safeguard Piping Fill pums at Llrnerick is provided with an Individual, orificed mintmtm flow bypass line. There is no possfb!11ty of punp-to-punp Interaction.

2. Adequacy of Minintm Recirculation Flow l

The system has been dos!gned to provide the specified minirntm flow of 20 gun per pum, or 20% of maximtm punp capacity of 100 gpn.

l I

I I

i

PHILADELPHIA ELECTRIC COMPANY ~ -DOCKET HOS. 50-277

. ATTACifENT "A" 50-278 50-352 50-353 B. Peach Bottom Atcmic Power Statto,1 The P8APS safety-related systenu which are' potent.lally .

affected are the Residual Heat Removal (RHR) System, the Core Spray System, the High Pressure Coolant Injection (HPCI) Systan, the Reactor Core Isolation Coollng (RCIC)

Systnn, Diesel Generator Fuel 011 Transfer System, the Reactor Recirculation Systan, the Emergency Service Water System, the High Pressure Service Water System, and the Emergenc'f Cooling Water System.

Of the safety related systems mentioned above, only RPR, CS, HPCI, and RCIC have the potent!al of being affected by one o- both of the two concerns raised in this bulletin. The other safety related systems mentioned do not require or enrloy mininun flow lines on their punos for the same reasons described in the LGS discussion.

The four safety-related PBAPS systuns which may be offected are discussed here.

Core Spray System

1. Ptrrp-to-Ptmp Interaction The minimtm flow bypass lines for the "A" and "C", and for the "B" and "D" core spray ptmps join into caTbined mintrn_rn flow lines. However, each Individual pum mininun flow line contains a restricting orifice. With this type of configuration, as discussed previously, l ittle adverse ptmp-to-pttro Interaction is expected.

2. Aceouactcf Minimtm Rect rculatIon Flow The Core Spray System has been designed to provide the speciflW mlnliram flow of 312.5 gpm per pum, or 10% of rated ptrio capacity of 3125 gun.

Rosto j Heat Removal System

1. Pum-to-Ptrnp Interaction The minimtm flow b'/ pass I!nes for the "A" and "C" and for the "B" and "D" RHR pums join f rito et. - ined mintmtm flow lines. Eacn Individual pum minimum f sw Ilne contains a restricting orifice. Little adverse pum-to-pump Interact lon is expected. This will be confirmed through testing as stated in the short term resolution sectlon.

L e % e

1 t

W . .

PHILADELPHIA ELECTRIC COMP #1Y. 00CVET NOS.-50-277

. ATTACtNENT "A" 50-278 50-352 50-353

2. Adequacy of Minimun Recirculation Flow The RI-R system has been designed to provide the specitind O mininun flow of 500 gpm per pum, or 5% of rated pump capacity of 10,000 gun.

Reactor Core Isolation Cooling System

1. Purp-to-Purp Interaction The mininun flow line for the RCIC purp is not Interconnected with any other lina; therefore, there is no possibility of pum-to-pum Interaction.

2. Adequacy of Mininun Recirculatica Flow The RCIC System has been designed to provide the specified mininun flow of 60-120 gpn, or 10%-20% of rated pum capacity of. 600 gun.

High Pressure Coolant.Injectton System

1. Pum-to-Purip Interaction The mininun flow IIne for the HPCI purps is not Intercennected with any other line; there. ore, there is no possibility of pum-to-oump Interaction.

2. Adequacy of Mininun Recirculation Flow The HPCI System has'been designed to provide.the_specified mininun flow of 600-1200 gun, or 12%-24% of total purp capacity of 5000 gun.

II. Short Term Resolutions A. Limerick Generating Station No significant degradation of the Limerick Unit 1 Core Spray, RhR, RCIC or HPCI pums is expected to have occurred. These pums undergo routino In-service Inspections as required per ACHE Boller and Pressure Vessel Code Section XI, in addition to the Techrical Specification surveillance testing requirements. Furthernure, all the affected pums are provided with continuous, on-1Ine vibration nonitoring which processes the data obtained for alarming and trending. Thus, no design changes are necessary in the short-term. The following short-term actions are Intended mainly to reinforce this position.

4

'^

PHILADELPHIA ELECTRIC CmPANY  : DOCKET NOSo 50-277 ATTACFf1ENT "A" 50-278L '

50-352 50-353 Purp-to-Punp Interaction (Core Spray System)

As noted above, the Core Spray system punps are the only safety-related punps which present a concern for potential pum-to-purp Interactlon. Although the syatem design.has been demonstrated analytically to be adeonte frcm this standpoint, the following short-term actions will be taken to provide further assurance.

1. Monitor the actual flowrate for each punp during mlnlmun flow recirculation with both'punps in a Core Spray loop running:

Unit 1 "A" Core Spray loop was tested on 6/23/88.

"B" Core Spray loop will be tested in July, 1988.

Unit 2 - Will be performed during pre-operational testing.

Adequacy of Minlmt.m Recirculation Flow The following short-term actions apply to the Core Spray, RHR, RCIC ar.d HPCI putps.

1. The rou;Ine In-service Inspections required per ASME Doller and Pressure Vessel Code Section XI, along with the Technical Specification surveillance testing requirements will identify any punp degradation.

2. Contact putt vendors to determine whether their recormlendations for minimum recirculation flow have been revised upward since the original design and under what operating repimes such restrictions apply.

B. Peach Bottom Atanic Power Station Pum-to-Puro Interact.lon Actual ficwrates for each punp during minimun flow recirculation with both pumps in the loop operating in

' parallel will be obtained through testing.

Adequacy of Minimun Recirculation Ficw Current minimun flow parameters for each punp will be investigated. Mininun flow and deration Information from purp vendor:. will be ccrmared with previously acquired data.

"Reccmmnded" and "required" minimun ficws will be requested fran vendors. T'e duration of operation at the "reccmnended" and "required" minimun flows will also be required.

3:

? 3,

._. j PHILADELPHIA ELECTRIC COMPANY- ' DOCKET NOS. 50-277

• ATTACMENT "A" 50-278 50-352 50-353 o

III. Long Term Resolution A. Limerick Generating Station Pum-to-Pum Interaction (Core Spray System)

Because the margin available on the "weaker" Core Spray pum in a' loop betvmen.the calculated minimtm flow and the-manufecturer's recmmendation is sriell,' the follcwing design modifications are und7r consideration:

m' 1. Providing an orifice in each pum's minimtm flow. path upstream of the conmon mintmtm flow bypass line.

2. In addition, modify the existing orifice in the cmmon line to minimize the associated pressure drop, thus maximizing the effectiveness of restricting ortflces 'n the Individual minimtm flow lines. This will ensure that the effect of any difference in the pums' g,arformance with regards to minimtm flow is minimized.

The final design is contingent upon the results of actual minlmtm flow measurement. The following schedule applies:

1. Declslon o implement one of the above alternatives will .bo cmpleted by Septenter 30, 1988,

2. Physical Implementa?.lon of reconmended in difIcat lon:

Unit 1 - Prior to restart from the third refueling outage.

Unit 2 - Design will be ccmpinte by September 1,1988, at which time a camletion date for the physical impleraentation of the modification will be determined.

Mequa:y of M!nintm Recirculation Flow

1. In order to reduce pump duty and minIntm rmintenance re.aut renents, the time that the pums are running in the minimtm flow mde should be minimized. Because all the affected L!merick systems are provided with valves in the mintmtm fles s bypass lines which close autcmatically once flaw in the main pum discharge path is established, no ch*nges in procedures are required for system testing or operat'on of the systems in non-accident rnodes (e.g.,

RHR 2hutdown cooling). Mcwever, the low pressure ECCS systems (i .e., RHR-LPCI ar'd Core Spray), as a result of an ECC3 initiation signal while the reactor is at h;gh pressure, nay operate in the millntm flow node. Therefore, ,

plant 6mrgency operating procedures should provide guidance to the opecators to place the systems in the full flow test mode, if autcmatically Init'ated on a LOCA signal and sypotoms positively confirm delayed injection, until necessary to inject.  ;

_g_

y ,-

9  % ,

7 .;

M'(; PHILADELPHIA ELECTRIC COMPANY . DOCKET NOS. 50-277-ATTACl+EhT "A"' 278 50-352 50-353 s -q

2. Depending on the results of-the short-term actions to

. evaluate the adequacy of the specified mininun flow,- the sizes of the orifices in the minimun flow lines' could be increased on a case-by-case basis.

B. Peach Bottom Atomic Power Station Punp-to-Puno Interaction -

Should the actual flowrates obtained from the tehting of the Core- Spray and RHR punps prove to be unsatisfactory, modifications to the system will be performed.

Testing of both units and a decision to laplarent any rrodifications will be ccmpleted in the fall of 1988, based on the current PBAPS Restart Schedule.

Physical Irtpla. men'.ation of any system rrodifications (if required) will be completed prior to return to service frcm each ur its' next refuel outage.

Adequacy of rdinimum Recirculation Flow Should the pum vendors determine that minimun ficw rates are not adequate, the recirculation lines wl11 be analyzed to verify the lines are capable of handling the higher flow. If approoriate, rrodifications to the recirculatf or lines will be cor.sidered.

A decision to waluate any nodifications will be trade by October 30, 1988.

Should mod fications be required, physical limlementation will be conpleted prior to return to service from each units' next refuel outages.

IV. Justification for continued Operation The concerns stated in NRC Bulletin 88-04 are surmarized as:

1. With two purps operating in parallel in the minimum flow rrode, one of the pums may be deadhnded resulting in punp damage or failure.

2. Installed mininun pum ficws may not be adequate to preclude puno damage or failure.

, These concerns are addressed by the responses below whir.h provide the basis for concluding that continued operation of both PBAPS and LGS is justified.

s,4--- ,--e r,e- .- , h--,. ,, , , , ~ , - . , , - - , ,w, ,p-,

s; 3 .-

-- c , ;,

' PHILADELPHIA ELECTRIC COMPAt4Y DOCKET NOS 50-277-ATTACFfENT "A" 50-278 50-352 353-Pum-to-Ptrnp Interactions i

The potential for excessive pum wear attributable to inininun l flow' operation and/or deadheading at either LGS or PBAPS is negligible. An assessment of Limerick Units 1 and 2 concluded that pum-to-pum Interaction of the Core Spray pums in a loop does not result in Individual pum flowentes celow those specified by the manufacturers. To date, testing of the LGS Unit 1 "A" Core  !

Spray loop has been ccrroleted. Actual minimtm flow rates of each i

• tum ere obtained through the use of ultrasonic floemeters. The re M ts of this test are being evaluated to verify tLat pum-to-pum lateraction does not exist. Continuous tronitoring of  !

Core Spray pum vibration and routine maintenance end Inspections will provide early indication of pum degradation. The Core Spray and RHR systems at Peach Bottom Units 2 & 3 contain a restricting orifice In each pums Individual minimtm flow bypass line to minimize pum Interaction. Additionally, as was trentioned in Section II, actual flowrates for each pum during minimtm flow recirculation with both pums operating will be confirmed through testing.

Adequacy of Min 1mtm Recirculation Flow BWR operating experience does not inoicate any excesolve wear to pums when operating under the currently specified minimtm flow conditions. That is, no such reported wear has resulted in Indicated degi adation in pum performance. Any wear observed (per normal pum Inspection requirements) has taken place over relatively lor,g periods 'of time (on the order of ten years), and has been I talted to gradual detectable changes in pum performance, rather than a sudden significant degradation in performance.

The following discussion is based on work done by Ceneral Electric Co. for the BWR Ovners G.oup ar.d is appilcable to both Limerick Generating Staticr1 and Peach Bottcm Atomic Pcwer Station:

l Recent inspection of scme BWR RHR pums have Indicated no pum I

impeller damage (due to minimtm flow or any other adverse conditions) that could potentially de: grade pum performance over the Inspection period. It is estimated that the pums had been operated in the minIntm flow node for up to 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br /> during this period. This further substantiates that short-term operation in

> the m* nsn flow node has little or no irrpact on punp Ilfe.

l There have been occasions when pums have operated deadheaded inadvertently Cl.c., deadhaading was not caused by minimun flow operation but, for Instance, by incorrectly closing a valve).

These purps have continueo to function normally.

x- ..- - - _ - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

3. ,'

. , PHILADELPHIA ELECTRIC COMPANY DOCKET NOS. 50-277 ATTACHENT "A" - 50-278 ,

50-352 50-353 System operatico in the mII,Inun flow mode is Ilmited to ptsnp start transients during monthly survelliance testing and during a postulated system start from a LOCA signal.

The total expected time in the minimtsn flow mode over the plant lito is approximately one percent of the 30,000 hour0 days <br />0 hours <br />0 weeks <br />0 months <br /> maximtsn given by the purp vendors-for Intermittent operation. The maxinun expected continuous duration in the mininun flow mode is 30 minutes for postulated small break LOCAs. Therefore, the potential for excessive wear attributable to minimum flow operation is negligible.

All Class 1, 2, and 3 centrifugal and displacement-type pums Installed In BWR's and required to. perform a specific function in thutting dcwn the reactor or in mitigating the consequences of an accident, and provided with an emergency power source, must undergo routine In-service inspection (tests) per ASME Boller and Pressure Vessel Code Sectlon XI, Article IWP-1000. These quarterly tests are in addition to the Technical Specification ,

surveillance requirements Intended to dennnstrate cam 11ance with the plant safety analysis.

The Section XI tests are Intended to detect changes in ptinp performance; Article IWP-1500 ("Detection of Change") states:

"The hydraulic and mechanical condition of a punp, relative to a previous condition, can be determined by attenuting to dupilcate, by test, a set _ of basic reference narameters.

Deviations detected are synptcrns of changes and, depending upon the degree of deviation, Indiccte need for further tests or corrective action."

In general, the in-service tests measure speed, Inlet pressure, differential pressure, flow rate and vibration amplitude. Alert ranges and required action ranges are strictly defined, and require either increased frequency of testing or declai ig the punp as inoperative, respectively. Performance cucside of the required action range would place the affected system in a Limitleig Condition for operation.

Although these tests themselves would not detect punp dead-heading or inadequate mininun flow, any deleterious effects of operating with inadequate flow would be detected in advance of significant puno performance degrac'ation. Therefore, any changes in punp performance would be detected and corrected por routine pum testing in advance of purc degradation due to cunulative low ficw effects from ptinp surveillance testing and nornal system starts.

f

y v -

, '%^ .g -

,, 4

' PHILADELPHIA ELECTRIC COMPhlY DOCKET NOS.-50-277 V ATTACI-fENT "#' ~50-278 50-352 50-353 Limerick Unit 1, which was IIcensed in. October 1984, has been operating for more than 3 yeers; and, based on this operating experience, no appreciable pep degradation is expected to have occurred. (Limerick Unit 2 is not IIcensed.) This conclusion is further supported by the fact that no vibration

-problems attributable to inadequate mininm flow have been detected. The Limerick Vibration Monitoring System continuously acquires data on speed, vibration and shaft axial position for all the safety-related pmps of concern. A "smart monitor" provides alert and danger alanns. at- pre-selected vibration levels. The system offers historical trending of the data and can generate displays, logs and plots.

Pmp wear attributable to mlninm flow operation is not a significant contributor to total system unavailability ecmpared to other contributors (such as loss of emergency power, loss of cooling, etc.). This Is based on BWR operatlng history, which Indicates no occurrences of system unavailability upon demand due to punp wear incurred in minium flow operation, s

For the RHR and core spray ptmps, the only design basis events that would lead to ptmps running in the minimun flow nude and/or deadheading are events that result in an ECCS Initiation signal while the reactor is at high pressure (above the punp shutoff head). These events are normally snail break LOCAs and loss of drywell cooling isolativn events. Of there, only certain small break LOCAs actually r(quire ECCS Injection from LPCI or core spray. Once initiated, the maximtm duration that a LPCI or core spray punp may operate in the minfrun flow mode for the spectrun of hypothetical LOCAs is less than 30 minutes. This is derived frem postulated small break LOCAs, whernin a reactor depressurization to below the shut-off head of these purps is delayed. For large break LOCAs, the reactor inherently depressurizes more rapidly through the break. The present mininm flow bypass line is expected to provide adequate protection for these puTps for the short durations postulated during both the small and large break LOCAs.

For other scenarlos, there is adequate time to secure the RHR and core spray ptmps, and restart them as necessary, precluding extended operation in the mininun ficw node.

As discussed above, only certain small break LOCAs actually require ECCS InJoction for LPCI or core spray wherr, the puros may be operated in the mininm flow node. Mcwever, because of the excess ECCS capacity that is available, Ilmiting LOCA scenarios do not depend on both purps of a pair of parallel punps to operate in order to satisfy 10CFR50.% reaut rerrents and General Design Criteria 35 of 10CFR50 Appendix A.

4

(

_^

' 7-r

. PHILADELPHIA' ELECTRIC COMPANY DOCKET HOS. 50-277-ATTACFNENT "A" 50-278-50-352 50-353 b The smaller break sizes result In calculated peak clad terrperatures (PCTs) well below the Ilmiting calculated PCT.

Since the Ilmiting failures for these break sizes result in one or more LPCI systems available, the effect of a core spray flow reduction due to punp unavailability is reduced; therefore,- it is concluded that based on the large PCT margins for the small break.

sizes and the' availability of LPCI for the limiting single failures the current calculated maximtm PCT would be unaffected.

For the next most Ilmiting single failure in the small break range, only two core spray systems are available for low pressure ECCS Injection. The maximtm duration in mininun flow for this scenario is found to be less than four minutes for break sizes down to 0.3 sq. ft. It is expected that the core spray puros would demonstrate no performance degradation for operation at Icw flow for this period. Hcwever, a ficav reduction due to the loss of one ptrro in each core spray loop would not increase the calculated PCT above the limiting PCT, based on the large PCT margin availab)e Cup to 900 deg F) and the spray cooling effect of the available core spray system capacity.

For BWR plants such as Limerick 1 & 2 which are limited by the single failure of an emergency diesel generator there are at least three LPCI pumps available for ECCS injection for all breaks and single failures; therefore the Impact of reduced core spray flow is'much less severe than for plants limited by the LPCI Injection valve failure. Since the Ilmiting large recirculation line breaks - ,

result in a maximtrn duration in mininun flow of approximately one minute (In fact, the maxinun line break would shov essentia11v no -

time in minimtm flow mode), it is included that there is no Impact on the Ilmiting IIcensing LOCA analysis.

MGK/pdO6228805

- 1!+ -

. _ - . . - _