ML20058F519

Reconstitution of Design Bases & Design Documents, Presented at NRC Regulatory Conference on 890418-20 in Washington,Dc

ML20058F519
Person / Time
Site:	Millstone
Issue date:	04/18/1989
From:	Imbro E Office of Nuclear Reactor Regulation
To:
Shared Package
ML19311B205	List: B14692, Forwards Response to NRC Re Two Alleged Violations of Requirements Identified in Connection W/Insp Rept 50-245/90-01 & Investigation Rept 1-90-008.Proprietary Rept PED-12-0388 Withheld (Ref 10CFR2.790(a)(6)) ML20058F500 ML20058F504 ML20058F509 ML20058F519 ML20058F531 ML20058K108
References
NUDOCS 9312080160
Download: ML20058F519 (55)

Text

RECONSTITUi 010F DESIGN BASES AND DESIGN DOCUMENTS BY EUGENE V. IMBR0, CHIEF TEAM INSPECTIDM DEVELOPMENT SECTION A SPECIAL INSPECTION BRANCH OFFICE OF NUCLEAR REACTOR REGULATION U.S. NUCLEAR REGULATORY COMMISSION FOR PRESENTATION AT THE NRC REGULATORY INFORMATION CONFERENCE THE MAYFLOWER HOTEL WASHINGTON, D.C.

APRIL 18-20, 1989 i

9312080160 931130 PDR ADOCK 05000245 Q

PDR

1 RECONSTITUTION OF DESIGN BASES AND DESIGN DOCUMENTS INTRODUCTION The United States Nuclear Regulatory Commission (NRC) has been conducting Safety System Functional Inspections (SSFIs) and Safety Systen Outage Modification Inspections (S50 mis) since 1985. These inspecticns are somewhat different in approach. The SSFI is a system-oriented inspection while to the 550MI focuses attention on plant modifications. Both examine design in detail.

A common thread emerging from these inspections is that design basis informa-tion to support the as-configured system design and proposed plant modifica-tions is not available. As a result of this lack of design basis information, the NRC has found, in some instances, plant modifications being made that potentially jeopardize the ability of safety systems to perform their intended safety function due to the lack of full understanding of the original system design requirements.

These NRC inspection findings have prompted many nuclear utilities to evaluate the availability and control of design basis documents at their facilities as a part of an overall configuration management program.

In order to fill gaps discovered in the availability of design basis information, the reconstitution or regeneration of original design basis information has been necessary. How-ever, at this time, there is no regulatory guidance or clear industry consensus as to what constitutes an adequate set of design basis documents and many questions have been raised regarding the necessity, extent, and time table for recreating missing design basis documents.

DISCUSSION The term " design basis document" is commonly used but not well defined. Title 10 of the Code of Federal Reculations defines " design basis" in part as that information which identifies the specific functions to be perfonned by a struc-ture, system, or component of a facility, and the specific values or ranges of values chosen for controlling parameters as reference bounds for design.

Design basis documents are the documents which can be used to verify that the structures, systems, or components of a facility have been designed to perform their specific identified functions, and that the specific values or ranges of values for controlling parameters have been properly chosen and incorporated in the plant design. Although many consider design basis docume ts to be only design input documents as described below, for operating facilities we have defined the design basis documents as the set of documents that support the "as-built" plant configuration, forming the basis for future plant modifica-tions. Design basis documents include (1) design input documents such as those that specify the performance requirements of structures, systems, and compen-ents, licensee commitments to the NRC, industry standards, regulatory require-ments, and documented generally-accepted good engineering practices; (2) design analyses such as calculations or other engineering evaluations; and (3) design output documents such as facility drawings, lists of qualified equipment, equipment purchase specifications, and documents which contain interface infor-mation necessary to develop plant operating guidelines or provide information to other engineering organizations.

Design basis documents provide a necessary starting point for plant modifica-tions. For example, without design analyses which correctly reflect the

Y current plant configuration the impact of a plant modi!ication on the' existing design margins cannot be determined reliably. Design basis documents (in particular, cesign output documents such as facility drawings, piping and instrumentation diagrams, electrical single-line diagrams and schematics, and equipment purchase specifications) form a living record of the as-configured plant. These cesign output documents are necessary for plant modifications, as well as for plant operations and maintenance.

In addition, design input documents are also necessary. Plant modifications made without design input documents may unknowingly compromise initial design considerations, since the design input documents specify the functional requirements and design criteria for structures, systems, and components.

Many nuclear utilities are currently in the process of regenerating design bases. For many, this consists of oreparing upper-tier documents for each plant system or each plant-wide design consideration, such as environmental qualification, seismic design, pipe break outside containment, and soon. These system level and topical design documents vary widely in content from con-taining all pertinent design inputs, design analyses, and design output documents, to being a summary of pertinent design inputs with references to the design analyses and design output documents. There is currently no standard approach to the format or content of the system level or topical design documents; each utility has been satisfying its own needs. However, I

the development and use of these upper-tier documents should reduce the possibility for design errors during the modification process by making information accessible in a single location. These upper-tier documents should not be considered as a substitute for the specific plant design basis documents (i.e., design input documents, design analyses, and design output documents) as previously described. Therefore, for each planned modification to a structure, system, or component, sufficient design basis documentation should be available to support the final as-modified configuration such that a credible design verification as specified in ANSI N45.2.11-1975, " Quality Assurance Requirements for the Design of Nuclear Power Plants," can be performed.

Nuclear plants that have not appropriately controlled their design basis documentation or whose design basis documentation is unavailable, may find it necessary to recreate design analyses to support planned plant modifications.

The extent that other missing calculations and design basis documents must be regenerated is an area which the industry needs to address jointly.

Industry guidance which would promote the use of design basis documents in the modifica-tion process would also be of value.

Ont key (and probably the most elusive) attribute that is useful in the upper-tier system and topical design basis summary documents is the reasoning behind the design (i.e., why the design is the way it is).

In many cases, this will be very difficult to recapture, as corporate memories of why various decisions were made fade over the years. This information, however, if available, is extremely valuable to an engineer designing a plant modification.

Insights as to why certain components were selected or why components were installed in a particular location, for example, may not be immediately apparent. However, when the original intent is determined, systematic use of such information can prevent modifications plant that would inadvertently compromise a particular design consideration.

P CONCLUSION The cevelopment of system specific and topical design basis summary documents can markedly improve the modification process by providing a centralized access point for design basis information and a key to accessing the supporting design basis input, analyses, and output docurents. The existence of these upper tier design basis summary documents, however is not a substitute for the suppor ng design basis documents. Design basis documents that support the as-configured plant design must be controlled and readily available for use and reference by the engineering staff. A sufficient design basis must be available to support planned plant modifications such that a credible design verification can be perf ormed in accordance with the guidance in ANSI N45.2.11-1974,

" Quality Assurance Requirements for the Design of Nuclear Power Plants." If an adequate calculation basis does not exist to quantify the design margins remaining following a proposed modification, design analyses should be per-formed or recreated as necessary to support the final as-configured design of the system being modified.

i s

Qgcket No. 50-245~

B14692 r

f i

?

I i

i i

i L

- Exhibit 11

.i Millstone Nuclear Power Station, Unit No. I i

n t

l l

t i

i

.I December 1993 1

3.f.C.: ark

'"ffc.

Senin u,35

~ Ui 5. %

^?%jreab.s heam w1ai d'.11k n Fu) mm cane. values rnxW

~

NE8 i

JaeA Guin of uvl u inemy hos askd dg g.c

Pos thrn % h mir, <w<c FW va yes (1 Fw-tets Kctq c.), For

• f Pf1b4re,6L Colecck % I pbsthm is op. SW twr, his rivf

^

r bs k wcf pssem Scr nuct opirahk htlm a wcii. De w

.mt0rn5 Cof FWCToftrak dutinf a LOCH m oslfyUpug s} b ifR YM niin it ct(c. Vcd w ; y % k t

,$ lot.D b fiq, /fc,cler ve%<l w,II be ndoah pass'bly 6s gired 9,( Fwc5 m nhm u

p.) 5. FuJ pu)r s(or adusoL cp Mk bwhe pnys ]

4 i

1f skud he a&d H h. % sen w < F. 2

&ma1s re i, ans+u-S ea ruire umJim.1%.

mna

<drel sysk wHl trg to unaez Flo mychmmk & Te$r Wsv),

-fi., caus n

& Si S ca f,G{~ Iiv es<. k GL o>1 NsuArlr,

-.0seyJdgp-Ak cm,,u$htk dhan.H*M'.1<s$,af'9 PU SI HEPLY b

j 000391 acci u m sc, man o.ra

~.............

)

DIAGNOSTIC TEAM EVALUATIONS BY LEE SPESSARD, DIRECTOR DIVISION OF OPERATIONAL ASSESSMENT OFFICE OF NUCLEAR REACTOR REGULATION U.S. NUCLEAR REGULATORY COMMISSION FOR PRESENTATION AT THE NRC REGULATORY INFORMATION CONFERENCE THE MAYFLOWER HOTEL WASHINGTON, D.C.

APRIL 18-20, 1989

PRESENTATION OUTLINE DIAGNOSTIC EVALUATION PROGRAM DIAGt40STIC EVALUATION PROCESS DIAGNOSTIC EVALUATION RESULTS CONCLUSI0HS

. 7 e

w

. ~.

l DIAGNOSTIC EVALUATION A BROAD-BASED INDEPENDENT EVALUATION OF SAFETY f

PERFORMANCE:

I

• TECHNICAL PROGRAMMATIC MANAGEMENT ORGANIZATIONAL j

f I

i 9

-t P

h i

i I

i i

i 1

[

i f

9 i

P V

?

PROGRAM OBJECTIVES IMPROVED UNDERSTANDING OF LICENSEE SAFETY PERFORMANCE INDEPENDENT ASSESSMENT OF SELECTED PLANTS EVALUATE PROGRAMS AND PRACTICES DETERMINE ROOT CAUSES r

)

i l

SPECIAL FEATURES OF A DIAGNOSTIC EVALUATION EDO REQUESTS REVIEW EDO APPROVES TEAM AND PLAN SES TEAM MANAGER MEMBERS ARE INDEPENDENT MANAGEliENT CONSULTANTS USED COMPREHENSIVE PERFORMANCE EVALUATION PLANT AND CORPORATE MANAGEMENT AND ORGANIZATIONAL CULTURE ASSESSED INTERVIEWS USED EXTENSIVELY ROOT CAUSES EMPHASIZED NRC CONTRIBUTING CAUSES IDENTIFIED EDO TRANSMITS REPORT ED0 ASSIGNS FOLLOWUP ACTIONS

4 PLANT SELECTION DISCUSSION AT SENIOR MANAGERS MEETING PI, SALP, SIMS, AIT DATA MANAGERS' PERSPECTIVES SENIOR MANAGERS' EECOMMENDATIONS ED0 SELECTS PLANTS

e i

TEAM PLANNING AND PREPARATIONS COLLECT AND REVIE'A BACKGROUND INFORMATION REVIEW LICENSEE PERFORMANCE, :"PROVEMENT PROGRAMS AND NRC ACTIONS f

TAILOR TEAM COMPOSITION TO ASSESSMENT NEEDS CONDUCT TEAM BRIEFINGS SELECT A SYSTEM FOR EVALUATION PREPARE EVALUATION PLANS I

S t

J i

f 1

ONSITE EVALUATION SEOUENCE OBSERVE AND ASSESS SAFETY PERFORMAhCE i

ASSESS THE OUALITY, Il1PLEMENTATION OF PROGRAMS ASSESS MANAGEMENT AND ORGAtlIZATIONAL EFFECTIVENESS IDENTIFY CAUSES FOR PERFORMANCE PROBLEMS I

4 PLANT SELECTION 7

DISCUSSION AT SEh!OR fANAGERS MEETING i

MANAGERS' PERSPECTIVES PI, SALP, SIMS, AIT DATA SENIOR t1ANAGERS' EECOMMENDATIONS EDO SELECTS PLANTS t

't I

b t

i.

i a

O t

b 4

l l

b

)

EVALUATION AREAS PLANT OPERATIONS i

TRAINING AND QUALIFICATION MAI!4TENANCE SURVEILLANCE AND TESTIttG ENGINEERING SUPPORT QUALITY PROGRAMS AND SAFETY REVIEW GROUPS RADIOLOGICAL CONTROLS

• MANAGEMENT PRACTICES ORGANIZATIONAL CULTURE i

i i

f I

L

l x.

j j

COMMUNICATION OF RESULTS.

SENIOR MANAGEMENT BRIEFIt;GS l

LICENSEE EXIT MEETitiG REPORT TRAtiSMITTAL TO LICENSEE ASSIGNMENT OF f4RC FOLLOW UP ACTION t

i i

5 I

i

~

-. i l

i

-1 i

i 1

i l

t I

=-

^

i i

COMPLETED DIAGNOSTIC EVALUATIONS PLANT UTILITY REPORT DATE DRESDEN CECO NOV 1987 MCGUIRE DUKE MAR 1988 FERMI DECO MAY 1988 TURKEY POINT

• FP&L JUN 1988 PERRY CEI MAY 1989 BRUNSWICK CP&L JUN 1989**

l SPECIAL EVALUATION l

PROJECTED DATE r

e I

t e

o r

5 i

?

r h

i I

e '

ROOT CAUSES PLANT NEGLECTED In FAVOR OF OTHER PRIORITIES FOSSIL PLANT ATTITUDE LACK OF CLEAR PERFORMANCE GOALS INEFFECTIVE PLANNING FOR OPERATIONS LACK 0F OPERATING EXPERIENCE LACK OF ATTENTION TO HUMAN RELATIONS MATTERS CORPORATE MICROMANAGEMENT

.i I

PERFORMANCE STRENGTHS / IMPROVEMENTS CORPORATE LEADERSHIP, CVERSIGHT AND INVOLVE!1ENT INTEGRATED PERFORMANCE PLANS STAFF TECHNICAL CAPABILITIES i

POSITIVE fiANAGEMENT AND STAFF ATTITUDE TOWARD SAFETY MANAGERIAL AND ORGANIZATIONAL CHANGES j

PROGRAMS FOR IMPROVED ENGINEERING SUPPORT I

l r

b

't PERFORMANCE WEAKNESSES STRAINED RESOURCES ORGANIZATIONAL INSTABILITY COMMUNICATIONS PROBLEMS INEFFECTIVE ENGINEERING SUPPORT QA STAFF CAPABILITIES LIMITED UNREllABLE EQUIPMENT INADEQUATE CHECK VALVE TESTING INADEQUATE MOV MAINTENANCE OPERATOR OVERTIME SAFETY ISSUE IST PROGRAM DEFICIENCIES POOR EQUIPMENT FAILURE ANALYSIS

i l

EDO STAFF ACTIONS GENERIC ACTIONS

. APPROVE LICENSEE IST PROGRAMS L

EVALUATE INDUSTRY RESPONSES TO CHECK VALVE FAILURES EVALUATE FAILURES OF AUXILIARY FEEDWATER PUMPS EVALUATE MOV MAINTENANCE PROGRAMS PLANT-SPECIFIC ACTIONS j

HANDLED BY APPROPRIATE OFFICE / REGION i

)

V i

I i

t i

4 I

l

CCt:CLUS10:15 S;CCESS LEPEt:ES Ci: INTEt31'!E FPEPAFATICU, EXPEFIEtlCED TEAtt tiEMBERS At:D GOOD TEAM CCMMutlICATIONS I'At:AGEMErlT At:0 CULTURE EVALUATIONS E!;llANCE ROOT CAUSE 5

ASSESS! Et'T ROOT CAUSE DETERl!ItSTICil5 It'PPOVE UtiDERSTAt: DING OF PEEFORMANCE PROBLEMS, LIKELIHOOD FOR It1PROVEMEt!T AND THE NEED L

FOR ADDITI0ftAL fiRC ACTIONS DIAGNOSTIC EVALUAT10t!S GENERALLY HAVE CONFIRitED t1RC SENIOR ltANAGERS' PERFORMAt CE PICUTRE Tl!AT WAS BASED Oil EXISTit:C 1

SALP, PI EVALUAT10tlS HAVE BEEN WELL RECEIVED BY UTILITIES AND NRC

[

MANAGEt1EtlT i

. i 6

i i

i

^ !

f Docket No. 50-245 l

-B14692 i

t i

. - Exhibit 12 Millstone Nuclear Power Station, Unit No.1 e

h I

. i f

December 1993 I

s I

e y

---e

)

\\

8 v

s, CE Nucleat Eseryy Gennf fara o,.,

I?$ care Aame, san.Am CA 95f15 May 15, 1990 Mr. R. Matheny Northeast Utilities Service company P. O. Box 1928 Bartford, CT 06144-1928

Subject:

Millstone Unit 1 Teedvater Coolant Injection System (FWCI)

Flowrate Requirements Reference 1: Telecon, Niraal Jain and R. M. Matheny (NUSCo) with C. H. Stoll (GE), May 14, 1990.

Reference 2: NUSCO Purchase Order Number: 001256 Dear Mr Matheny a GE is pleased to provide to Northeast Utilities the attached letter documenting the results of an evaluation of the Millstone Unit 1 FWCI system flowrate requirements as authorized by the above referenced telecon.

The services provided were performed under the GE controlled quality assurance program as described in the current NRC ao-capted revision of Licensing Topical Report NEDO-11209.

~

CE appreciates this opportunity to be of service to Northeast Utilities. If we can be of any further assistance to you in the future, please do not hesitate to contact us.

e

Regards, R. W. Tobin Huolear Services Manager GE Nuclear Energy ec: M.P. Hills R.M.Mathony N.K.Jain C.L.Sorzi, GE C.H.Stoll, GE

9

, e.

'~

IVAWATION OF MILIA20NE UNIT 1 FNCI FmMEATE REQUIREMIDrrs Reference la NEDC-31740F,

.f Of-CoolantAccidentbrietaryInformation)111 stone (GE Fro

• Loss-al sis Report For M Unit 1 Nuclear Power 5te ion *,. September 1989.

j Reference 2: PED-12-0388, (GE Buclear Energy, "Feedwater Coolant Injection system Design) Basis Requirements",

September 1988.

-)

Review of MCA Licensina parety Analyses -

' i The Millstone Unit 1 licensing basis MCA analyses was recently updated and documented (Reference 1). In this analysis, no credit was taken for the feedwater coolant injection (FWCI) system in any limiting or near limiting LOCA scenario for Millstone Unit.1.

The low pressure-coolant injection (LPCI) system and the core spray systems in conjunction with the isolation condenser and the -

tion to meet all requirements of 10CFR50.46outomatic depressurizati 10CFR50.

and Appendix K of The isolation condenser is capable of removing decay heat from the reactor vessel, shortly atter scram,-without loss of reactor inventory, which Provides protection against potential cora uncovery during hig% pressure reactor vessel. isolation events. In particular the isolation condenser prevents core uncovery during a loss of feedwater abnormal operating occurrence.

u Review of FWCI Intended Functions The current intended ~ function of the FWCI system is to reduce the probability of oore unoovery during the lifetime of the plant.

The FWCI can perform this function by replenishing reactor inven-tory during-postulated high pressure events thereby avoiding unnecessary reactor blowdown and potentini partial core uncovery.

Such events include small line breaks (leaks and postulated line breaks outside the containannt tahtoh can resu)lt in some inventory removal from the vessel prior to the typical isolation of the vessel closure (of is,olation valves on the line brok(en). )

e.g.

main steam isolation valve MSIV closure ' and/or The intended function of FWCI stated above can' be sooonplished -

with a flow rate of a few thousand gym or less. This is substan-tiated by the fact that the capacity of the reactor oore.isolm-L tion cooling (RCIC)llar to Millstone Unit 1) system of EWRs witho (but otherwise sia is'400 to-600 gpa.

The RCIC system is. designed to provide adequate makeup for loss of feedwater' events. Therefore, a FWCI flowrate in excess of 600 is a more than adequate capacity to backup the isolation gpa 6

4

4ue,925 1490, Hay 16.90 16:1.6 P.04 en ws.

4e oondenser function of preventing core uncovery after scram at Millstone Unit 1. The magnitude of the flowrate in excess of 600 gpm provides additional capacity beyond decay heat " boil-off" requirements to mitigate significant inventory loss from Isaks or breaks.

Conclusion

- Based on the above facts and enginocring judgement, the most important function of the the Millstone Unit 1 FWCI system can be fulfilled with a system capacity in excess of 600 gpm. A FWCI capacity in arcess of 600 gpa provides some small break mitiga-tion capability that is not required in order to meet regulatory requirenents or to maintain safety limits, but will contribute to reducing the overall probability of core damage in the lifetime of the plant.

~

Prepared by:

C.H.Stoll Plant Performance Engineering l

GE Nuclear Energy Verified by:

h c.c. Allen Lund System Enginaar GE Nuclear Energy l

j l

l

• The high pressure coolant injection (HPCI) system provides a backup for the RCIC for the typical BWR without an isolation j

condenser.

. ~.

l Docket No. 50-215 B14692

.i I

i i

! - Exhibit 13 Millstone Nuclear Power Station, Unit No. I j

i

'I i

i i

Dect*ber 1993

e 1

January 27,1990 To:

Mike Bigiarelli Millstone Unit One Engineering From:

Willinm G. Noll Millstone Unit One Engineering Extension 4442

Subject:

RFP Minimum Flow Valve Accumulator Events To the best of my knowledge, I have outlined the sequence of events that resulted in the reportability evaluation and LER 89-022.

~'n

1. s***. W W Y t,/irj er -w s +r ass. ~ ~ " h seeb 1,

6/22/89: Jack Quinn submitted Izvel of Effort to John Ferguson to design accumulators for the RFP mimmum flow valves. Expected completion date was 9/1/89.

2.

7/06/89-7/28/89-I discussed the conceptual design with Andre Lassonde.

Original concept was to install air accumulators that would basically float with air system pressure.

3.

7/28/89 - Received memo from Andre Lassonde documenting the valve air requirements that were provided from the valve manufacturer. Since the valve air requirements could not be met using air accumulators, Andre recommended using compressed air bottles.

4 8/8/89 - Andre Lassonde supplied vendor information on bottle regulators and pneumatic switching devices.

00019e

6 5.

8/9/89 - received confirmation that LOE request to Berlin for system design was accepted (Reference LOE-1-89-JJ818) l 8/29/89 - Andre Lassonde provided final conceptual design information that would 6.

install high pressure bottles as a back-up air supply.

7.

9/1/89 - 10/01/89 - discussed conceptual design with Plant Operations and I&C i

personnel.

t 8.

10/01/89 - Project turned over to Elliot Abolana (Responsible Fngneer for Feed System) h 10/10/89 - NEO 2.25 REF 89-48(MP1) initiated to determine reportability of FWCI operability with RFP minimum flow valves failing open.

10.

10/20/89 - memo from Andre Lassonde to Elliot Abolfia regarding final conceptual design.

@ 11/17/89 - REF completed. Final determination stated that the design deficiency was reportable.

If you require any additional information or documentation regarding the sequence of events listed above, please feel free to contact me.

I f

I l

i

't Docket No. 50-245 B14692 l

I t

i 1

i - Exhibit 14 Millstone Nuclear Power Station, Unit No.1 i

+

1 0

?

December 1993 i

- i m

NORTHEAST UTILITIES r

33te:

/d:~

aco

=.

0 l

" =. ~.. r _ - -

i(

y.,'

~

E der /4 a REF Ongmator b" # / MO

- a:

REF Originator s Supervisor b #/ -N-O it/

-W4 6

From:

R.L McGuinness, Supervisor - Generanon Facilities Licensmg fpg 4 fru (Ext. 3297)

Subject:

Comrieted Reeortability Evaluation i

In accordance with NEO 2.25, attached is a copy of the completed Reportability Evaluanen REF W "T for vour information/ records. The supermtendent has determined that this item W

reportable under 10CFR50.72/73.

(Additional Distribution:)

___ Nuclear Records W.H. Becker N-020 R. Ferraro N-216 f.B. Regan N-020 i

f.A. Blaisdell W-141 C.J. G1 adding.

CY A.R. Roby N-020

_,_,_M.V. Bonaca W-141 M.P. Hills W-141

_R.C. Rogers W-122 W.J. Briggs W-053 M.S. Kai W-141 IJ. Roncaioli W-007

__._S. Chandra W-020 f.W. Klisiewicz NU-SO R.J. Schmidt RH-B4F3 X_C.E. Cornelius W-006 W.R. Koste W-020

_,_T.A. Shaffer N-030 R.A. Crandell W-122 M. Kupinski W-021 B.A. Tuthill N-017 l

._ TJ. Dente W-219 T.J. Mawson W-021 R.W. Wells NU-SO i.F. Ely W-021

.,&_R.E. McMullen W-030 R.J. Young N-020 T.M. Fackelman NU-SO

_ _,E.J. Molloy W-122 Y

P D. AN

• D

.,,X_J.H. Ferguson W-006 R.P. Necci W-020 Y

D 7-W n* er"*

CFL Dismbution:

X,_ R.M. Kacich All Units

_X. _P.A. Blasioil MP1/MP2

__C.P van Noordennen CY/MP3 i

X _REF File (Origmal TBS) loute To:

_E.P. Perkins CY R.C. Joshi MP3

__ P J. Miner MP1 i

__W.E. Hutchins MP2 Other

r:.n w

$ 891

FIGURE '.;

NORTHEAST U!LI!!ES REPORTABILITT E7ALUATICN FORM W -Me'-* /'/ )

'; nit:

vt,Ils f;n

' A ; T - G -4 REF Log n

Subject:

Pwc1 c as e. 4ll, n L., <.r

.A et le wen 1.0 ORIGINATOR 1.1 Describe potential significant implication or reportable item (attach additional documentation if required).

Swcr T -~ ;.,Kn

.la rin k n;i

- ~ ~

- - < r.

dve n /rn e4 l<

A

., a -

n

=-

~,.-

v /w ( t. c ~ - s * * /e h 3 A

CI smi 1.2 Describe potential safety implications.

t.r

. r a sv i a a t.

2 m.:, af c.a Loa. rt.

r-e,sLws 1L L.- s. sn s.

f,.. a t a4 u i-n2 (S< <- A r%44 l he.1 nns 1.3 Identify applicable sections of 10CFR50.72, 10CTR50.73.

or 10CFR50.9(b).

54.1A. S-k. 2 lrkTD s o.7 3 2.,db n 'td 0 1.4 Forvard REP to Immediate Supervisor.

(M (W to - to - H Originator Date

/0[/O N SUPERVISOR OF ORIGINATOR Date Received Suggested Maximum Processing Times 1 vorking day 1.5 Confirm that the Originator has satisfactorily completed Steps 1.1 through 1.3.

1 I

1.6 Forvard to Manager, CFL.

h

/Of/0 El l

Supervisor Date NEO 2.25 Rev. 1 Page 7.2-1 of 4 l

e-

'f i

i 2.3 "ANAGER. CFL Date Receiveo

/ v!/ 3 !77

.i 3uggestec Maximum Processing Ti e:

three working cays.

2.1 Contact Superintencent to determine cisposition of potential l

significant implication or reportable ites (select options).

NUSCO reportability evaluation requested.

5 JC0 evaluation requested.

i Potential significant implication or reportable ites will be dispositionec j

in accorcance with applicable station procedures.

/8!d 3 !f7

~

Person contacted:

8/// //a 6e [_

Date:

2.2 Designated Evaluator (s)

-I // F6VE 6vs w

~!

1 v ?u Elf,1/oe:ek i o ft 3 F 87

' Manager, GFL e

Date 3.0 TECHNICAL SUPERVISOR /

Date Received MANAGER Suggested Maximum Processing Times No more than 20 calendar days from Step l'.4, to complete all technical evaluations.

i MM@ m D.1.H.%um F4WW E

l f

Supervisor / Manager Date u-t,-es

-i Not Reportable JC0 Required

{

M rtable per

% oJrttt l

i 1

NEO 2.25 Rev. 1 Page 7.2-2 of a

f-1 j

=..

Oate j

Supervisor / Manager i

Not Reportacle JC0 Required Reportable per

+

i Date Supervisor / Manager Not Reportable JC0 Required Reportable per

//!/7!T9 4.0 MANAGER. CEN2 RATION Date Received PACILITIES LICENSING Suggested Maximus Processing Times three working cays.

4.1 Not Reportable M Reportable per f oc.R w 71[b)[rIGdD) 4.2 If yes, provide early notification to the Superintendent and the Supervisor, Nuclear Operations by telephone.

n/nkt

~ r4 as aa <.

M

((oldM' Manager, GFL Date

' i 5.0 SUPERINTDIDDrr Date Received

.. A t s le =

i 5.1 Ites determined to be reportable?

l No g Yes, Section(s)

, ca_ cn,, mt ;;; 3 i

I i

NEO 2.25 Rev. 1 Page 7.2-3 of 4 i

l i

i

c.

5.2 If item is determined to ce reportacle. follov applicaole station procedures for NRC reporting recuirements.

10CFR50.72 i'tems are vercally reportec within 1 hour:

10CFR50.73 items are suomitted under Station Superintendent signature within 30 days; and ICCFR30.9(b) items are ver-bally reported within 2 days then suositted under corporate of ficer signature within 2 veests.

Applicable PIPJLER Number. / - ?4_ g 7

%,,- L g i d n tp 7/

fff i, L 1e9 Unit Sapierintencent Date

//!19 ! 99 5.0 MANAGER. CFL Date Received 6.1 If JC0 is performed, send REF/JC0 to NRB/SNRB for review.

A/!8 Date Sent:

6.2 Ensure documentation completed, logged, filed, and distributed.

By:

M Date: /

2.

Originator Originator's Supervisor Nuclear Records NEO 2.25 Rev. 1 Page 7.2-4 of 4

1 2:

C.

-ct.e

.4

. t..t,, ;

3.4 i n. :

w ;.r e e t

.' ; s t ; r ; in ::,1. '..

C. i... J L..-

4-

.2

.uss ;- ' stre.ent sie accc ni...

t* ' 1 0 '... ".

~he C ur C CsE C*

this

'4TiO 1 "a Cr OV' de O aCl-QT tcuno

6, recuest 4 or ac c r t.a t t. :,.. av a s -astiun.

• here

.A e

1.1% J n r. I 1; aOr

..,ss

,1 S t I c.:n i e t <. ; terw.

a1r :ar 1,.

• o

.. s si.-. - -.; ;.

3 G we?r t '.* gyn t,

,4 4 v.c n

+r,

.c v ty ri ".. e r e ;; O ; *g r ' **

J '" c '.P

• • t r t n 1 t '*

a sm62. ; r ee,

.:ss cv :::...n t e.

c t, t %: cr. a e...

t.

5 u o ra Oui c cer ce**ct20.

1 icss ci instrument atr uson tne Avect,r The effects t.t e

Femowater Oumo iitntmum F i t.w Fec t r* cul ati cn Val ces 1-#W-14A.3 and C would CAuse all three velves

• 3 f&1i th the cOen 31 r ec t I cn.

• he ccnsecuenco ce such a fatIure coulo 1 ac.tc t the FWCI system av increastnc cemanc on the systers wovend the I

capsetl10/ ce i. h u system cumps.

The first i rr-c ac t wauld te that the selected feecwater pumo would be reaut ree to suopiv a flow bevonc the normal " runout flow control moce' uetuctnt by one minimum ficw cath.

The second impact would t,tr tr.at the Melected condensnte and condensata boostur cuma flow eevond the l

como s nat t er would be errout r ec to suop1v

'cunout +iow centret a.c c e by tne +cdstion of thr ee m a n t ir.urse 410w paths.

t i s n e t.

,newn wnether ;WCI woueJ e.ec e ~_ r tr r..

C a D a u l 'd C? tta emergenCV OOol a flu functicn thrOUGnout the i

anal v: L*d spectrum of Iatt of ecclant actsdent iIne ttea*.s.

I I

i l

l i

l i

1

~

1 s

i Reconability Evaluation so-'.R

Anaivsis Objecme The objecuve of de andvsis is to determine whether the Feedwater Coolant Injecuan System (FWG) will remam operable dunng a loss of coolant accident combined with a loss of normal power. The objeenve is limited to evaluation of dynamic response associateo with ie operanon of the feed pump mimmum flow valves 1-FW-14A.B and C.

2. Summnry of Results The loss of off-site power event causes a simultaneous loss of instrument air compressor.

The instrument air compressor which is fed from the gas turbine on-site supply will not automatically re-start upon the reload of the aunliary elecmc system when a LOCA signal

s present.

When the mmirmim flow valves reposition to the fail open position instead of positioning to control flow then FWG will be required to supply Dow beyond its design capability.

3. Method of almMon "Be system design basis is reviewed with respect to the expected flows for the system motor driven pumps. An esumate is made with respect to system response during a loss of coolant accident combined with a loss of off-site power. He failure of the air system is postulated to occur. At this time a calculation is performed to verify whether the FWG system is operating with flows beyond the runout mode. If FWG flows are within the bounds of system operation then the system should is czpected to be available.

l

4. Design Inputs / References

a. UFSAR Section 6.3

b. EBASCO Specification MPC-MI MS

c. EHASCO Specification MPC-MI-M7

d. EBASCO Speci5 cation MPC-MI-M4 1

e. General Electric Specification 22A1374

f. Copes-Vulcan drawing E 268831 (NUSCO 25202-293200) 6" Class 1500 Valve l

g. Nusco Drawing 25202-29006 sh.16: Condensate Booster Pump Curve l

r a

b. Susco Drawing 25202 29006 sn. O Concensate Pmnp Curve Test Procedure Tu-1-8 Feecwater Minim"m Rectreuiation Valves 1-FW.

MA.B,C Operactlity Test"

j. SP 83-1-10 "Feedwater Pump Runout Test"

k. MP 716.6 " Disassembly and Assemoly of RFP Mimmum Recirculanon Valves.

1. GEK 27655 ' Millstone Feedwater Cuerol System with FWCI"

5. Assumptions

a. FWCI is available to inject 8000 gpm within 90 seconds of a LOCA/I NP FWCI wiH provide flow over a range of 100 psig to 1125 psig.

b. Condensate and condensate booster pumps will operate to limits defined by their pump curves.

c. The reactor feed pumps will operate to 11,000 gpm capacity

d. Mimmum flow valves pass 1.1 million Ibm /hr at Feed pump head of 3600 ft-

6. Analysis 6.1 Current System Cbn5guration At entry to runout flow conditions feed-pump flow is 10,000 gpm as set by current I@C procedures. His corresponds to a setpoint of 120% of flow with turbine valves wide open.

He mtmmum flow valve setpoint corresponds to 1.1 x EE6 lbm/hr or appronmately 2291 gpn The total flow requirement corresponding to runout flow c;ntrol with the 4

mtmmitm flow valve failed open would appear to be 12.291 gpm. However referring to reference d, drawmg number 41587 an extrapolation at a flow of 11,000 gpm shows that the i

corresponding feed pump head is 2200 ft. This corresponds to 2200/3600 or.611. Usmg nominal flow relationships the mmtmum aow nive at this pressure drop is approrimmtely

.78 x 2291 = 1787 gpm. The difference in Dow results in a total Dow of 11,787 gpm. This j

Dow is not on the pump curve and it is therefore not expected that the feed-pump would contmue to operate.

A review of GEK 27655 indicates that the runout setpoint should be set at a value of 23.3 ma. This value is less than the value used to enter Oow control (29.2 ma.). If the lower setpoint is used. then the resulting feed Gow would be 9531 gpm (115% of flow with turbine

I valves wide opens. If the feeo pumn were operaung at the edge of its pu=p c::ve then 1757 gpm would be cotained for num um dow. Adding tuese values we oct:un 11.318 ep=

wnich is beyonc the etc of the pu=p c:rve.

Reference ig) provides pu=p curve data for tre cc::censate booster pump. At the Dow rate of 11.318 gpm wnich is reomred for steaoy state operation of the feed pumo dunng the i

an:dvzeo concition. the concensate oooster pu:no head is approximateiy 405 ft. Assummg a pressure drop of 200 ft. throuen the idle reactor feed pumps and associated piping, the mm m"m dow valves of the idle pumps are suoje.:ted to an inlet pressure oi 205 ft. Adding this Sow to the Gow supptied to the operanng Oed pump results in the following regtured Cow:

2 x.24 x 2291 = 1093 gpm 1093 + 11,318 = 12,411 gpm Following the analysis through the condensate system it appears that the condensate and booster pumps will be in a runout condition due to the sodition of Dows tnrough the other failed open minimum Sow valves. Even if the pumps are operstmg at tne coge of their pump curves at 11.500 gpm then it appears that the total suction head available to the feed pump will be 225,+ 392 = 617 ft. Herefore, the low suction pressure interlock for the feed pumps will not be satisEed.

7.0 Conclusion De FWCI system cannot be relied upon to operate during a simultaneous loss of offsite power coicident with a loss of coolant accident. Since the system is required to be operable by technical speci5 cations then this simation is reportable.

L Originator: M M "/" M Reviewed:

NTE.:. CF=CE MEMO

' CEPT.-LCCAT;cN R L '.icGuinness Generation Facilities Lcensina 3OA CEPT -LCCATiCN

D '.iason D.J. Yaoenanvn Balance of Plant Systerns. W-30 Berlin i_iLECT CATE Ceview ci '.iP '

WCl REF Evaluation REF 59-48 11-17-89

. Ess;GE We nave c0motetec a review of Eliot Abotafia's (MP-1 Engineering) evaluation of REF 89-48 ano at this time cannot refute his conclusion that the issue is recortable.

cc: J. H. Ferguson R. E. McMullen J. P. Stetz

/

Cecer

-l-9; gn

,/

1 SICMD EXTENTION 3623 REPLY

.EPT -LOCATCN SOED DATE

-}

i

.. --i o.

l Safety Evaluanon for Secooms Char.ge Reauest 189413 l

Re purpose of this semom change recuest is to decrease feed system Cow woen statulized in the Cow a:ntrol mode of eg.scs. Da reason is that analysis maicates that ccadcasate oooster and feed putap Cows may exceed the -@v of the

cennemar,

sta!!ad emnpment. The revised seapouns wars saleasd to allow the F#c system to r.ippsy a minhaam of 8000 gym over tbs rangs of rencer er=amana for wench FWG h now reqmrod. He seleczed setpourts have acopaned h instransat loop armr and instrmnant i

drift.

De onginal setooms for eury imo the Sow comed mods have not been changed. De wipoun rounda at 120%. ne reason for not moddying the eury seapoint is that the respoons to the Sow emural entry would be rapid enough to aDour tbs pumps to consmas OParadng un S sysema condhians stabilims at the meady stam Sour enamd conditions. The respones uma will be as the ortler of assural seemedL The Sow commt sospaints are based on a mas Sour rues eastespondag to main artsins valves wins open. For a magia injection smag read Sour correspanos m a Sour c( 3,976,788 i

Ibm./hr or about 8284 gun. Dis vales is commerveuve wink respost to rated Sow requand by F5AR and FWQ W-whhtg requks 8000 gym.

j De now enamn seapouns requemed by tiismes request 149 013 equass to 102.5% of ths artins valves wide open rused Sow russ. His anneados provides a margin af 6% (3JM

+ 23) to dessa basis required Sour ramas.

De h loop arram using sum of the squares methodoloy is L65%. A dieck wkh the Insernrasension and Casamis Departaset indkmes the a 1% insmunent band toleranos is acespable for calSession meshedL Doradora, a asspoint of 102.5% is "T*** whh he prendous wtSe providing asuranos that FWG denga basis Soir reus are not i

The astpost for two pump operanon is based soiety upon not eenseding limits for synsa pumps if two feed meings are in opennes at tbs thns of runos Sow emeral. As suik FWC2 avaSabGity is enhanomd sinos abs deed weser syusses is designed to operuns M during a LOCA wnhout sa Dr. bilukuma sous for FWC2 are not i

^*",*a =d sinos two panp operados esameds the FWC ndslama injeadon Sow rapar No other esnargsacy core comung or saisty syssna is a5ected by this setpoisa change.

g l.:. r 6 L & f ' n M h d

l

~

a

~~

,c[

.. e me :

t s c: asso. <

aaa"=os

$5=

t.=.,_ - -.

ww s,.

r 2.,

=.e a m.i., e.c.o..== a.

L*ENSEE E"v'ENT DEPCRT (LER) c

..e

.e u s w a w r

. n n.1_

oc Maas.i,. =

1 - _, y. e eu p. um. e satto.cies,

.mmer. a ma m.a o*t im.e.m CC Mb2:

YC, #v at iia

- 2* ~

• l00G84t.M'etad4e t s i es et ee r 1 Millstorie Nue: car Power 5:auen Urut !

J".n nx Feec.ater Cociant intecuen Declareo incoeracie fe d& 5'9-&K W ".$

~

t.,e :* v + e

..w...,

. :. ev,

c.z.n...ei. e s...om o...

• .cae v me s

_eman woc

• e.

tvt..e-t.ae gl Ig,.oc.e cav q viam q el s1 et et el 1 1 1l~

El0 8 l9 Ol2l*

0l 0 il 21l5 sl9 l ei st ei et of i i cremarse j '.s ae.oer is a:e4 sm v.su.ar te w as cumswecs o. $0 cea i scr

- = -. =

..i$,

1 M.#(M M wael 60.r34alG 84 1 73 71 ret WomTR M 484bf tlf51 60 M eettil X' 60.731al 25 tel 73.71 ect

a. *vs,. Il010 M.as.inisio go 33,,,,,

3, r,.,,, g3,,,,,,

ev. es c

. w

.=

' M.#dM811tiall 60 rSimmitt) l Se.73tMGit )De&&

M*"***

.~

==

M.38 5111ts.t Se.7%441Gif 88 94 7a(4)3)(eestial M aosessener 40 r3tmalf ano se 73raittoss UCE8.Stf Co.gvact som vs.g Lgm stri

~

ae na oe.u.een

~

.Ata CGO$

Eliot Abotaha. Seruar Engmeer (E.xt. 4429) 2;01 3ldi4g7l-q1;7;9Q cometrra o.a 6 4 oa..e co o.ewr eu.una oestaneo Tws armont <*si i

cause{rve

7f,48-

..,,; causa-evereu commonewr

• ^"ru*m'eA**M

.?;

co a.o a.n B l Bt J hPt ! g Gl0l8(0l

.-- - j

; gl o-mn.

)

i

!l l l l l l l O'~f

{

l l 1

{ { l l

I f

su u w.c = acae.t e m eres n..

uowm o*v iwaa "E vte en.

expecTeo smeene navei

""1 o

fa no 0g2 28' 9;0 1

asst =acTwa=i.co

,.1 2_

,ou On Novemoer 17,1989 at 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br /> vnh the plam at 100% power (530 degrees Fahrenhen and 1000 pug) an engmeerms anaivses mdtcated that the Feeduser ("antam Invecnoo Symem (FWQ) may be recutred to deliver Gow m excess of symem pump Dow capabdrues durms a loss of Mm acadern with a amunaneous loss of air. The scenu6ed pro 06em was the fadure to the open posman of the reaczar feed p.anp mirumum flow nives. Fadure to the open posmon wouid placa andmonal doir rew.s.s.s on the symem pumps.

Ahhoutfi the analyns us not fuDy ".' min'*d. plarn management toont the conservauve accan of declanng the FWG syssa moperable. The una was placed mto an admmr.nratrve tecnrucal spectficauoo hmams condnaan for opercoon (LCO) requirms entry mto a seven day ume hms for renorms FWG to operaethtv.

FWG c+.n.s remamed avadable durms the LCO penod. An trem.+.te nouhcanon was made to the Nucisar Reguianary 'hm m accordance wna manoo admamstranve proceeures. Operstmg procecures were reviewed and venhed to pve adequate guidance with respect to the idanched scenano. Wahm hve days a piarn mod:Scanon was completed wtuch ms:.alled a sessrmcally quahSed, safety related air supply to the reactor feed pump marumum Cow valves and FWG ns placed m an operable natus. In addason.

ptarn managemem ts pursumg further anahus in order to determine the consequences of th:s postulated tuusuon. No synams were required to fimcuen durtng stus event. No safety conseguences resuhed from this avent.

=:-~ ~

i

.3

.u,,u,., u om, e - - -- - ~-

am movt= en.=

o m o.c o

-mey m.

s.xasus eranan j

8 f P wi ehdP 8%

F 8 p.y 3 p.4.gp edWR 9.m.

I UCENSEE EVENT 83EPORT 4,,ER) gg,'_,","_,",*,*,*,,,8,,'",,,**'",",,*

TEXT CONTINUATION a

,m : _

a

.---.erw

.saosua

=

• r,cu.ns.- -n rien ocpaans v =

in w ei ca.

1_

e

. w nsa ocroam s *curv w m aoc eTeu.aban m

, to auwers...

..c.e o e vs aa l l"'""1 l'L"**i

%!!none Nuclear Power Stauon U ** I cl si of of 012 la i s

$ 19 012l:

010 0(

'CF O!3

~

E c --.

c 3.... m L

,.-.--...,., e..,

On November 17.1989 at 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br /> with the piant at 100% power (530 degrees Fahrennen anc 2000 pues an engmeerms analvus mercataa that the Feeowater Coolam intecuen Sntem tFWCI) i may be reoutree to cehver flow m excess of system pump flow capsedrues durmt a less of comant i

acctcent sita a umunaneous loss of att. The idenuhed prooiem was the fadure to the open posuon of the reactor fees pumo mirumum flow valves. Fadure to the open posazon wouad place accruonal flow requsrements on the system pumps.

i Ahhougn the analysts was not fully completed. piam managemem took the conservauve scoon of J

declarms the FWG synem moperaele. The unn was placed mto an adamanrauve techrucal i

spenhcanon hmmng conoroon for operanon (LCol.requirms crurv mte a seven dar ume hma for renorms FWG to opersodar. FWG componerus remameo anuable durmt tne LCO pened. An unmensate nouhcanon was made to the Nuclear Regulatorv '"~===on m accordance wah nauan acmmastranve proceoures. Operaung procecures were renewea ano venhed to give adequats guacance wna respect to tne idenched scenano. Withm hve cays a plam modthcanon was completad wtuch menti d a setsrnacally quahbed. safety related aar suppty to the reactor feed pump

~ ~.. ~ Gow valves and FWG was placed m an operable satus. In addioon. plant managemeen u pursums further analpas an order to determme the consecuences of stus posudated srtumnne No systems were required to funcuan curmt this evern. No safety consequences resuhed frcen stus evem.

II.

NueiFwm Venous plarn revwws resulang from Genene Lener 88-14 and a self-trunated Safety Synem Funcuanal 1rupecuan (SSFI) focused aneruson on svsem mueracuans between the FWC and msn'umern air sy=amt As a readt of these renews a became re--a>.M that the effects of a lots of artarumern azr upon the reactor feed pump........ flow recircutauan valves 1-FW-14A.B and C would cause all three vahes to fail in the open direcuan. The consequence of such a fadure could hupact the FWG syneen bv meresses cemand on the rysem beyond the capacity of the synem pumtw. On October 10, 1989 e formal reportatninv evaluauan was inraated m accornance 51th npany proceaurus (Nuclear Enoneerms and Operanons Prococure 2.25).

Recogmung that an emensrve eynem moochng effen would be required to address the operabshty concern. a prehmm.ry engmeerms mvesogauon began to assess the potenual for FWC to be rencered moperable due to a loss of atr. he resun of this prehmmary review mdscated that FWCl pump curves may not envelope the anucapated range of flows and pressures that woutd be encournered.

Stauen managemera took the conservauve acuan of dectann FWC inoperable ahhough n was apparent that addruonal ume was neeoed to assess the data and/or obtam aed:uonal data.

111.

enhw nf Evem The determmauon of FWO system mopersednv recutred entry into plant Techmcal SpectScauen 3.5.C.3. Ahhougn the analyns of the postulated event is not get completec, nauon management is taking the conservauve scuan of reporung this condruon as havmg the potenual of rencerms a safety rynam moperable. Therefore the evem is reportable uncer 10CFR50.73(a)(2)(v).

The rentits of LOCA analyns provided in the UFSAR remamed nhd even tf FWCI is not erecited in the analyns. The basas for this concluuon is prooded below:

Laree Wenk I N"aa Because of the miecuen locanon (downcemers all FWC1 injecuan 3111 bypass the core and sedl om throuen the break, in a taree ereak. FWC1 does not centndtite to core cochne. Adeouate core

• G'""wso a w.

rs es.ws e w.co.tyswun m w e 4t., pamps.. -si s a w's. *W e t

• mke pwa.ansac.W.MN 'OCA anaivus provided m the UFS AR remains utad even if FWC1 is not creened.

cream L

I

..,i..-- e -.e,.

- c

.e v.o

. - + + -

txposa eaome j

t e cw== m-a======== ====us.

==

1 WCENSEE EVENT CEPOR (LET))

gy yagyng m m'*===gyggigegg'g' yg m

i

- = = = = =

• o * '

• • =

J

' EXT

,NTINUAT1CN

,, - _ __ _ _ e,,,,,,o eans. w s n=ar

+

c. a==== a.= mi maa.es.n.a.ns. we..

=====m oc a

=

o==

1_

a cs.i oc rancs I

8.a v aae su

aocar aumen m re an.nere.e.

• *cd L '

taa i N:

t=,

%1! stone. Nuclear Power Si.auon nn 1

~

O f s ! 310I 012 l a i s 8 l 9 01:10 010 013 CF 0:3 t r-

.,...anc me= an.... m c e c...k im4 The small bream LOCA anarvus provicec m the UFSAR assumes that the gas turonne generator fans f

to operate tassumen worn smpe fauures. Tharetore, the anasvus ammas no flow from me FwC1.

wruen is powered by the ps nartune generator. Even af the FWCI were mopernole. the wara j

smpe fadute wowd have remamec fadure of the gas turome generator. This is because the gas c.treme generator surt ume 98 secs) is sigmiscanuy less than the ADS delev ame (120 secsl. In a small breas LOCA miecuan flow trom the Low Pressure Coolant intecnon and Core Spray pumps 1

becomes avatiable long before the tsactor pressure vessel ts depresmartzed by the Automauc Depressurunntwt Symem to allow low pressare m)ecuan. This ts the case whether the pumps are powered by the diesel generator or the gas turtune. Therefore, whosher FWC1 is credaad or not the small **

. LOCA anairns provideo m the UFSAR would remam vahd.

Addmona. 41vs s of the evem and the fmal deterrrunauon cf the smzanon will be provided after i

furtner anasysis is compresea.

IV.

(.-e--.ve 4,.m Operaung p.h n were revizwed and venhed to pve adeouate ruxtance wnh respect to the tdenuhed scenano. A piarn desgn change nas mrusted to prmide a back-up att supply to the affected valves. The back-up aar supply uss dessned to satsnuc. safety class one cruena. De 1

g supply constas of two high pressare air boezies mith necessary pressure reducers. tubmg and valva to provtce separauon and redundaner from tne non-seismic piam instrumern aar sistem. The back-up a2r supply was made GP.uchal and FWC1 sss declarea operatne before the teP+'ement specz5 canon LCO expired.

The course of acuan taken by managemem recogruzad that further anah's:s was warramed m order to fub cetermme whether the idenafied set of concruens watld acusally rencer FWCI inope:sble.

Therefore. further analyns u be.ng pursued to resolve the usue ahhougn the unoctf31ng conewon has been corrected.

v.

> $8..~n s 1-1-nem A @-5, CM repWs will De g!ven upon FM"'i8 Don of further anaIvsts.

l i

i j

)

1

.anc aes i

i l

l

i i

l

_ Docket No. 50-245 B14692 l

1 i

l l

t a

k l

l

! - Exhibit 15 Millstone Nuclear Power Station, Unit No.1 1

4 i

t December 1993 1

I l

I

,a.

+ + -

w.+.,

1 4

i g * * ** ** ? 'S f * ? f 8 $ *'g'.[a'.g'{* g'.g'.g t*

E OPlER COVER LETTER 9

8

". -"*.e

.-m~-

l.

-(

. -.. ~

J u -.u

.c... v.m.

l"2 TOTAL NUMBER CF FAGES BEING SENT.

q^

i

'I~

?'-

7rj' h (inclacantt Inns Cactl g

[> T A ArasMITTTNG To i

j

,7, y 3 c y

~ l

E 8 A a T w E N f,Cm $$og a T T i% T ION PMONTNuwSER

[

M..

e, A

(

)

i q

f ESECOPrER PaONE NuuSER C

ARE A COOE (

)

f(( @

(( '

i

> T A ANSMITTING FROM t

,, outuvuagm

/

[

[

(

) $4-2 4 O

& f L T[A OEPAR1utNT/OswissON V>t P -/

1. r %

TELECCPiER PwoNE NuuSER AaE A CCCE (

)

• t! YOUDID NOT RECEtYt ALL Of THE 9A GES. 9 LEASE CALL TNts NUNSUL

,> C O M M E NTS/DIR E CTIO NS WEd 22r Ev~ f D&f i

r

-s

.... -r 4

-r e

% 1 :-

6

1 i

l Reoortability Evaluation So-18

1. Analysis Objective The objective of the analysis is to determine whether the Feedwater Coolant Injection System (FWCI) will remain operable during a loss of coolant accident combined with a loss of normal power. The objective is limited to evaluation of dynamic response associated i

with the operation of the feed pump mimmum flow valves 1-FW-14A.B and C.

2. Summary of Results The loss of off-site power event causes a simultaneous loss of instrument air compressor.

The instrument air compressor which is fed from the gas turbine on-site supply wtil not automatically re-start upon the reload of the auxiliary electric system when a LOCA signal is present.

When the mmimum Cow valves reposition to the fail open position instead of positioning to control flow then FWCI will be required to supply flow beyond its design capability.

3. Method of Calculation I

The system design basis is reviewed with respect to the expected Dows for the system motor driven pumps. An estimate is made with respect to system response during a loss of coolant accident combined with a loss of off-site power. The failure of the air system is postulatea to occur. At this time a calculation is performed to verify whether the FWCI system is operating with flows beyond the runout mode. If FWCI flows are within the bounds of system operation then the system should is expected to be available.

4. Design Inputs / References

a. UFSAR Section 6.3

)

b. EBASCO Speci5 cation MPC-MI h6

c. EBASCO Spect5 cation MPC-h0 M7

d. EBASCO Speci5 cation MPC-MI M4

e. General Electric Speci5 cation 22A1374

f. Copes-Vulcan drawing E-268831 (NUSCO 25202 293200) 6" Class 1500 Valve

g. Nusco Drawig 25202 29006 sh.16: Condensate Booster Pump Curve

-.;...j - -

. n c>

~~

!. -, ?

i

r

h. Nusco Drawing 25202-29006 sh. 22: Condensate Pump Curve

i. Test Procedure TS4-18 Teedwater Minimum Recirculation Valves 1.FW-14A.B.C Operability Test *

j. SP 83-1-10 *Feedwater Pump Runout Test *

k. MP 716.6

• Disassembly and Assembly of RFP Mimmum Recirculation Valves.*

1. GEK 27655

• Millstone Feedwater Control System with FWCl*

t

5. Assumptions

a. FWCI is available to inject 8000 gpm within 90 seconds of a LOCA/LNP FWCI will provide flow over a range of 100 psig to 1125 psig.

b. Condensate and condensate booster pumps will operate to limits defined by their pump curves.

c. The reactor feed pumps will operate to 11,000 gpm capacity

d. Minimum Dow valves pass 1.1 million Ibm /hr at Feed pump head of 3600 ft.

6. Analyus 6.1 Current System Con 5guration n CW l

At entry to runout flow conditions feed-pump flow is JD;006 gpm as set by current I@C procedures. His corresponds to a setpoint of 120% of flow with turbine valves wide open.

e rnmimum Dow valve setpoint corresponds to 1.1 x EE6 Ibm /hr or approximately pt h Mgpm. The total flow requirement corresponding to rtgopt flow control with the nummum flow valve failed open would appear to be 32;29fgpm. However referring to reference d, drawmg number 41587 an extrapolation at a flow of 11,000 gwp,jhoQat the corresponding feed pump head is 2200 ft. This corresponds to 2200/360(For.611. Using

=""!)ow tionships the minimum Dow valve at this pressure drop ,y roximately

.78 x DH' =

m. The difference in flow results in a total flow of gpm. This flow is not on the pump curve and it is therefore not expected that the feed pump would contmue to operate.

A review of GEK 27655 indicates that the runout serpoint should be set at a value of 28.3 ma. This value is less than the value used to enter flowgntrol (29.2 ma.). If the lower

]

serpomt is used, then the resulting feed flow would be.&S3jTm (115% of Dow with turbine

valves wide open). If the feed pump were operating at the edge of its pump curve then pP87 gpm would be obtained for Imnimum Dow. Adding these values we obtain ll.2dff gpm which is b+y#Nend the end of the pump curve.

vo*

Refep;fs.(g) provides pump curve data for the condensate boosten pump. At the Dow rate of 146% gpm which is required for steady state operation of the feedgJ/Jft.

ump during the analyzed condition, the condensate booster pump head is approximately-a pressure drop of 200 ft. through the idle reactor feed pumps and associated piping, the nunimum Cow valves of the idle pumps are subjected to an inlet pressure ofdidTft. Adding this Cow to the Dow supplied to the operating feed pump results in the following required dow:

9' 95 i+3 7 2 x,Nx 229r = JO97gpm

'a,-se om a o.w it.se-o 1993 + 11:3T8 = 12,411 gpm e um m ~ - " 9

• r)

Following the analysis through the condensate system it appears that the condensate and booster pumps will be in a runout condition due to the addition of Dows through the other failed open mimmum Dow valves. Even if the pumps are operating at the edge of their pump curves at 11,500 gpm then it appears that the total suction head available to the feed Pump will be 225 + 392 = 617 ft. Therefore, the low suction pressure interlock for the feed pumps will not be satisfied.

7.0 Conclusion The FWCI system cannot be relied upon to operate during a simultaneous loss of offsite Power coicident with a loss of coolant accident. Since the system is required to be operable by technical specifications then this situation is reportable.

Originator:

Reviewed:

g n

es e

s o

EN'

it 29189 3

,jf

~

FwCI I

p' ' ** ya d'

B C

i O

iE F

G H

i J

f K

A i

I COfD BOOSTER tttu RX I

2 ?

t 3 i HEAD 2P(Feet HEAD 2P(Feet)

HEAD

}

l 2MINL 1 MIN j

l J i i

['

l 5 1 6 :GFM Al f

rese):

+S 8 _ r.J t

7 1 12 0 0 0 MJ.l.5 0' 26.11 5001 84.82'5! N 21001 9600: '20% Checx ij'

-(1005.3) 1553.1!

121581r, 8 IFLOW 12000:

9 linlet in 0s:

23.2331 i %,< 202.191 1107.4:

o,i24 i

469.081 10t in Ft M*l53.91 1

22001 i

96004 120%; Check l 11! 118001 110! 25.237 5301 82.0211 12320 t

i 1117.91 i 1601.81 12 IFLCW 118001 i

' 1177.9!

13 linlet in oss i 36.5361 i

i 229.631 14I in Ft i 84.763!

l

! 532.741 I

t l

15I i

i i

I f

i i

16I i

i i

i e

i 17I i

i t

i i

I i

i 18l t

1 i

I I

i l

i 19iDATA !

I i

l i

i i

2 O ISHUTOFF HEAD DATA IN FEET I

I I

i i

i i

l i

I 21 l CDO iS/OFF l 480lNOTESOFF FOR FLOW OF 2200GPM I l

I i

2 2 BMERSOFF t 9751NOTESOFF FOR FLOW OF 2200GPM i 2 3 l FEED IS/OFF i 3700lNOTESOFF FOR FLOW OF 2200GPM i l

i I

1 ^"d'E CIl[f 4 TOTAL IS/OFF I 5155iNOTESOFF FOR FLOW OF 2200GPM i o # :S 25 Man Flow set at i 2200lopm i

l i

i l

I MEb,j l

l l

l i

I 26 l

l l

i 27 aP DATA (PSI) 1100%FLQW l

i l

l I

I 28I COND h itR i (2pt) 1 I

I I

I i

i 29 BOOSTERFEED I 6 51 )

i I

I I

I I

I 301 I

i i j i

i l

i l

i 1

70 t*

9 ser"X Q /G1./1~ m g n.c'l 7 2M)Y C 19 r / # #'

9 7d'A t S f1*5 3Iff te J

^ (:[

~

~

m;.

iNb-f$

4 AS EddE. " $1'! E.* *=me I:

• k1

'l 6...

. - s. _ i s

i

.1 i :as 4

2 iAI 4.,g i,

=

.-1 l

l I

. l l

j.!! !

I l

l

,Et i

I i

c......v

. l l : p, y t l i.I. FTD il l l l.. I.,-t l!

si "ijN[N; Vi

)

4l I'

j!

5 I

i

}

lj l

E i

l il 5

ej Miti

-' i t E' 1i

m. s, n,

-2 t

x.-

Ii 2 same s

i a

m 1

I 3

E I

E I e

,I I

=

[

?.

a 3

i O

s a

i

~*-

g 1

E l

s

=

m N

E r

D E

I lb i

0 F

f.

I A

. k.

3 b

A 4 e g

I i

1 l

s Docket No. 50-245 i

B14692 1

i i

s i

i T - Exhibit 16-j Millstone Nuclear Power Station, Unit No. 1 l

l t

i t

P b

t t

i 1

December ?t3 I

h f

i i

F i

i 4

h I

i

Tc.cr.-oc MEMO CEPT -LCCATCN Generation. acristies L.:ensino L. '.icGuinness CEPT..LCCATiCN

., D. Mascn / D.J. YaoenanvK Balance of P8 ant Systems. W-30 Eertin iLELECT

-eview ct '.iP.s :WCl REF Evaluanon REF 5948 11 17 89 vm We have cometetec a review of Eliot Abotatia's (MP-1 Engineenng) evatuation et REF 59-48 ano at this time cannot refute his conclusion that the issue is reportacle.

cc: J. H. Ferguson R. E. McMutten J. P. Stetz j

R. W. Vogel IA C

/

. B. Siko

/

/

EXTENTION 3C3

=,. SOED pgggy g

EFTAOCATON g

1

k Docket No. 50-245 B14692 T

)

i 1

j a - Exhibit 17 i

i Millstone Nuclear Power Station, Unit No. 1 s

i f

.l 1

i I

i I

1 i

e l

I December 1993-l 1

i I

t s

-l

kI Er f)/:2-

?,/ [ NM

// - 2 0-o3

~ ? F P iM i,~

= G h~ UfGus U

/ J4h /c G /f0u y Of 6.IR Peance c w o,,

pqy 1c 7,9 cr a

s a tr-yg ",

J. 6 C t CHOCK \\

4gg 33Z c.

5.A. HAe T G l413 5736

& 'V ' E ""'"" "

Gg-& 241 ego (o /

05. E\\wg As 24c yn-p L 3 \\ a d id \\ l' L tN "'

36IT yi, s Ki Tiss,e~tAcQ, Mi yxf.bal~

$$ MU>

3 A.m./4 M7 Rf E J3 3 3 L 4 AQs,i 3097 N w. h

/

f.7. Mh1" L i m,,y n9c

/ 14 3 b l G I A r-G L u p i suc, a gy cczff Jsx7sd Gmf - FA=a.me_c s iir-E4 n Eh rn A/,

cry; g pg /,ftr

[860 W b1bD hfI" 7

000609

Y a-,

,,e ir-pgg,C iv l.

CvBn/

u g p ec

/w-p ag 4. asso a

},g, fj.gg'//AUSEA A&E-Sa'r'E3

5. A. 4kr SM D" C"

c;. n. n e n.c e T 3 E nc ao/:

J, 6. C i c 4 c e K \\

4MD ff/4on-U n 3. y g f doNW' LWin)

Nas ze r

,, s a.

m

, % k= lhpre //i

[hi 8y, nr? 3 000810

i Docket No. 50-245 B14692 i

I i

f t

i i

I - Exhibit 18 t

Millstone Nuclear Power Station, Unit No. 1 6

l i

l

.j i

~

i December 1993 i

i 1

1

.i i

4

.e.AC.AAm as,.,g,t y,c:3aaggan,, e

.,s.unovtp cmas.ao 3 5.ogi.e.

ms acom t,.

m, vi.

33.o.ev.

we w.e.t

_. _- sw gnuy

. i== a U:':ENSEE EVENT REPORT (LER) a m.1_

_.er.=a..-mass a s r

. u.

oc an.na asso.eios o.

- -. _ _ _ sys 3.ammt.

• 4peur. oC 7:Bol

..ca ~v p,ocas:* - a a.

Millstone Nuclear Power Stacon Unn 1 I e t s i of of 012 i s i si deel 013 av ~ a..

Feecwater Coolant intecuen Declareo inoceraele

1. df d& F9-W i - e.c :. v...

..w...,

,e., e,

-.oz.e...e tne s

.et vue...

wo c* cavi-taaevtami FQM lg,uc>eig oav i vtan,

g

~

• *wv mans l of sI el of of I f 1l1 1l*

$l0 8 l9 0l2l2 0l 0 il 2115 elg of si et of el i 1 l N[ '.es sco, se osno s,.n nyveo n.new.u, to me seoumendanas op to era i

ecp

==.raim.

w.-

_.ui onematwo macce foi

o n esassi ao.ae m to.ntaiens.re m rites aowem l 11010

"'"' to auss.icism se.asmos y

sa.rataieneet ta rim a vu. ' i ro eas.aisinus

~

' 'o

,,,,3,,,,,,,,,

' 20 ambaanstHan 30 ragalmm AS.f5tasaltwe.BsAs tema. W per.=a.aa, s*

==

x 70 acEsanatneer 04.ratasel1st to.ratalSitwo. FBI to.cse. net.e,

torat ient.:

se fatalaisst ucchsett condael 60d ?*.s Lee 81F' wa.

• A6aCQot Ehot Abotaba. Seruar Engmeer (Ext. 4a29) 2l0q3l414t 7I-t 1I 78 9 I!

ccma.trvu one two som sac ecam.c

.c eika cescmeen w v=

.co v osi caussl:

co w

'T #

ca w eversw comanownwr

• • "rt""eAC- %

w.,

m B l Bt J lPt l }

GjoI8JOI Ow l

l l l l t { {

-6 'N

=- 6 li i i i iii i

l

+- -

1,

i i i i i i

[

~

" '#'4 su.=t.rsar.cai me.ao re crave *ro es. __

eme To 2il9o m

e - c,e -. _

ve, Ho

• N 0 isi i

i

. 4 ;

aa m.cr - -

On Novemoer 17,1989 at 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br /> wth the plam at 100% power ($30 degrees Fahrenhen and 1000 puu an engmeerms analvns meicated that the Feeesiter Coolarn inwrcuan System (FWC1 may be recuzree to cettver Gow m eacesa of eynem pump Gow capabdines curms a k>ss of coolant acc2cem wTth a nuntaaneous loss of asr. The noenufted prootem was the fadure to the open pospon of the reactor feed 1

pumo n.mtmum flow naves. Fatlure to tra open penuan wowd p!. ace accmonal Dow reamremerns on tne j

system p'u=ps.

Ahhough the analyns mis not fuDy comoleted. plars management took the conservauve accon of declarmg i

the FWG sysem moperable. The unn was placed into an adnurustrsuve techracal speczfacauen Lmnmg condruon for opersoon (LCO) regurrms emrv mio a seven day ume hmn for renorms Fwa to opermothtv.

i FWG c=*e remamed avadable cunnt the LCO penod. An immediate nouficauon was mace to the i

Nuclear Retuintory Comrrussion m accorcance witn ::auon acmmistranve procedures. Operaung proeecures were revtewee and venhed to grve acequate gsudance with respect to the idenufied scenano. Wrthm five days a plant moetficanon was completed stuch mstalled a seistrucally quah5ed, safety related air suppty to the reactor feed pumo marumum Dow valves and FWC s2s placed m an operable status. In addnion.

plant managemern u pursuang further anahsu m order to determme the consequences of this postulated snusuon..No synema were requzred to funcuen dunnt thts evern. No safety consequences resuned frem thxs evern.

i

.a.me

. /

-. i i

N

1 l

~

y.- ma.

.,s -...= w on,e ----

" ug gno-o*

t.,- -

M*n.==.au.",.',,*.,.","e,,r".',,**,*,.'",,,,*.*.*".,

C UOENSEE EVENT REPORT fLER)

TEXT CONTINUATION c

. a eans u s a m.,

are a w.s= e oc naes i.

m

-W e P.W. d4.hS.EE.R N. 4at&O Otthat C

1_

oc nean j

1

..c m

,oc e.u

. ui p.u.e.

..c a l

%11none. Nuclear Power Stauca

' ^" I c;sto!etot2laIs 819 0l:{2 010 01 :

CF 0:3 t r m --.

nc

m.... o i L

a.e. ~.. - 4 cy,y On Novemoer 17.1989 at 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br /> with the p: ant at 100% power (530 degrees Fahrennen anc 1000 put) an engmeennt analysts mezcates that the Feecwater Coolam intecuan Synem (FWCll rnar be recuarea to cemver Dow m excess of svmem pump Dow cassenhues cunnt a loss of coc4 ant acocent m.ita a simunaneous loss of ast. De idenuhed problem was the fadure to the open pomuon of the reactor feec pump -...

Dow vatves. Fadure to the open posuon watud place addmonal flow requzremens on the symem pumps.

Akhough the analysis was not fully completed, plam managemem took the cornervauve acuan of declarms the FWG svmem snopersons. The unn was placed into an admmastratrve t ean3est spenhcanon hmaxng conmuon for operanon (LCol.requinng ernrr mto a seven car ume hms for remannt FWG to opersonsv. FWG components remameo annatne dunng tne LCO pened. An tmmechats nouhcanon ras made to the Nuclear Requiatory Commasson m accorcance vnh sauan acannistranve procsourts. Operstmg procecures were renewee anc venhee to give aceounts guicance with respect to the idenufsed scenano. Wnhm hve days a plant mochcanon sus completed stuch mstalled a seismically quauhed. safety related azr suppty to the reactor feed pump

- Oor valves and FWO was placeo m an opersone manas. In addnaon. plant managemem a pursuing further anarists m order to determme the consecuences of this postulated stunuan. No svaams were required to hmcuan curmg trus evern. No safety consequences resumed from this evera.

II.

O mte M Fvem Vanous plant renews rendcmg from Genene Lener 88-14 and a self-mnsated Safety Svmem Funcuanal Inspecoon (SSF1) focused anencon on svmem mterscoons between the FWG and msmunem azr symams. As a remak of these reviews'n became recogrused that the effects of a loss of msmunern air upon the rsactor feed pump...... Gow recirculanon valves 1-FW-14A.B and C would cause all three valves to fail in the open carscuan. ne coruequence of such a fadure could impact the FWG symem by mereasmg demand on the system beyone the capacnv of the svmem puanos. On October 10. 1989 a formal reportandry evaluauan was iruuatec m accorcance

• '2 """"ty procacures (Nuclear Engmeenns and Operacons Proceours 2.25).

Recogntems that an emansrve symmm modetms effon would be requsred to address the operaesty t

concern. a prehmanary engmeennt mvesogauan began to assess the potennal for FWC to be rencered moperable cue to a loss of air. De resuit of thu prehmmary review mdicated that FWCI pump curves may not envelope the anue: pated range of flows and pressures that would be enecr.unered.

Stauen management took the conservauve acuen of dec!annt FWCI inoperable ahhougn rt was apparem that acdsuonal ume was neeced to assess the cata anc/or octam sodiuonal cata.

III.

Aanhw M Fve~

ne deternunauon of FWC1 synem moperabday reoutred entry mte plant Technical Speci6cauen 3.5.C.3. Ahhougn the analysis of the postulated event is not yet completed. stauen management :s takang the conservauve acuen of reporung thts cendruon as havmg the potenual of rencerms a safety symam mopersole. Therefore, the event is reporuble unoer 10CFR50.73(a)(:l(v).

He resuns of LOCA analysts prended m the UFSAR remamed vahd even tf FwCI a not creenec m me analysu. He bass for thn, conclusaon is pronced below:

L2rce n enk t ev a_.

Because of the intecuon locanon (downcomert all FWCI miecuan usil bvpass the core anc spiil cut throurn the tresk. In a taree eresa. FWG does not centrietne to core cootme. Adeounte cere ace m estisN a etw'~*e 4 w.

rie nwa w,w..c.. a.fc a.wn r.e w.,a sp.i P mpo...s a w' * * * *

=reas LOCA anaivsts omween m the L'FSAR remams vaud even if FWCI is not :reeneo.

e s** *

.6.. < s.

• mg La.47osv cchamasecy.

• • • 'apweD M $e*

e c

- m,== m,.ame

= a-.-

===

UOENSEE EVENT REPORT fLER)

,. amme TEXT CONTINUAT10N m

__ _ eram seeent u a ar

, -1.

.,es.i.oie.s.c asses m..

wet o oiesi ca==.

f

__ _e

- oc acnos i

e. a v

cesc **.amea m

.r. amer i

• .as o.

saai N l'""*

%I1none Nucisar Power Stauon

+

' n" I

~

~

ci si of 0; ett la Is 819 0l 21:

0{0 01 3 OF 0 3

• ac< --.

ce--ansa.s.iri C=sl D*Psk

! N" 4.

• o The smati break LOCA ananvas provided m the UFSAR asannes that the gas eurome generator fads l

to operate satsumee wars unpe fadures. There ore the anaivsa======= no Gor trcan the FWC1.

wtuch a powerec by the gas naronne generator. Even af the FWC1 were moperstne. the warm smpe fadure wouse have retnames fadure of the gas turtune gernerator. This a because the gas turome generator mart tune (48 secs) is agrubcantiy less than the ADS deley ame (120 secs). In a small bream LOCA intecuan Gow tross the Low Pressure Coolans lapacman and Core Spray puesps becomes avadalde long before the reacier pressure vennel a depressertsad by the Atmosasoc Depressurtsanon Symem to allor tour pressure mpecuan. Tha as the case wheder tne puseps are powered by the shesel generator or the gas untune. Therefore, whosher FWCI is credmed or not the srnati break LOCA annives primded m the UFSAA wou6d renam vahd.

Addenonal analysis of the everst and the tmal determmanon of the serusuon wdl be primded after hartner analyms is er=na=tes.

n..

~.. -. _. _

Operaung procedures were reviewed and vertised to grve adestuste guadance wnh respect to the idenufied scenano. A piara denen change mis truusted to primde a back-up air supph to the affected valves. The back-up aar mapper sis desgned to na====e safety class one cruana. The g

suppt) consens of two high pressure aar bonies mish necessary pressure reducers, tubung and nives to prtmos separanon and rocksnataner troen the non-sensense plans irrsmannent aar s51:em. The back-up air supph was made opersoonal and FWCI mas declared opernoie before the 'a'*==e=1 specthcanon LCO expired.

t The course of acnon taken br managemem recogmsed that further analysis was warramed m order j

to fully detennme whether the idenu6ed set of conesaons would actually rencer FWC1 inoperates.

Therefore, further Snalysts is bemt pursued to resobe the assue akhough the undertymg consboon has been corrected.

V, A etc me= n t 1 in-e, ew.

e i

A stwementaa report wel be pven upon ccmp:euon of f.m.ner analvsn.

I i

?

y t

i i

M Mg O

i

[

t l

r Docket No. 50-245 i

B14692

{

t 1

I I

i 4

l P

! - Exhibit 19 l

Millstone Nuclear Power Station, Unit No. 1 i

l 4

i

.l

-l December 1993

e s

EEM 3emre omose semen erset. m cemenme l

" O L Y e ~=

1

• c. sex m L

M*UL.~. ~~ca noisonc : Nuncw.;* auu.cm

, *

• N 5-acsioe w ooc i

.%Isr':ary 28, 1990 MP490-219 i

Re: 10CFR50.73(a)(2)(v) i U.S. Nuclear Regulatory Comm:ssion Document Contre Dess Wash:ngton. D.C. 20555

Reference:

Facilitv Operat:ng Ucense No. DPR-21 Docket No. 50-245 T9 Licensee Event RepertM-022-01 Gentlemen:

This letter ferwards update U:ensee Evem Report 89-022-01 required to be rubmined pursuant to the requirements of 10CFR50.73(a)(2)(v).

Very truly yours.

NORTHEAST NUCI.ZAR ENERGY COMPAST FOR: Stephen E. Scace Director. Millstone Skation BY:

Jotm P. Stetz Millstons Unit 1 Director y, fad i,,,,.,.,.,.,.

. 3..cith NOP.R-t0 SES/EA:mo-jf/f/ff Attacnment: LER 89-0:2-01 LW W. T. Russell. Region I Administrator W. J. Raymond. Senior Resident inspector. Millstone Unit Net.1, 2 and 3 M. Boyle. NRC Project Manager, Millstone Unit No.1 qf$St ShbYS

~'

9 e

• c **= Joe

, s wc.go of 3,1,4cey ::see640, vencWD one oc 3 ec.<ies 6-ess twqsse aseme esmesse swam

=m

=

.wamme======

ase UCENSEE EVENT llEPORT l'JR)

Esume="_"'""er*.*""" " " *!""",

w eumwee. a.nm.

ai..wan visies.eie.e,an a.

a

.--wca

==.

m. oc m

=

o..

emum =

_ oc men

..c o,., u p~ -, amen.a.

=> >

Mistene %:aar Pewer Sucen Un:t !

I e t s i eI eI ei: I e i a i i K:mi at !

-.s...

Feeowster Cociant In:ecuen Dee! arse !ncoarsele e w :. e s.

. r..o,

. em,,,

m..c,,, s =vese e,

ao <.4 c. i -taa : -t.a F fumaniiL=m.g j q uo.g.,

3,,, y, u i

..c.;r,

• 4e g

I el si el el el i i 1l1 1l? ll9 e l9 3l:{:

ol 1 ol2 2} 8 9l 0 el si el el el i I

~

~

carma.wo l 1

• -e me.o=* s etwa e.airve sume e ve *.e neoumsi.evre ce,e era e er t

INI ' a mi mei m.mmes

n. rim ao.eeionim se sere m y

so.nruem

n. rum

.ioi0 n

..n n...

.e..,,e

_ o,

,s e

N assas jp as ensim o...a so.nemm es.nimeinnai l-S ao.asto o so.nrumim es.remai.ensi 6 N'gY n as....n,.i

+

se rareerun so.tsenmim xemese co.nact.o= me un nn a.naas vammense an ne.n ama cooi Ehot Adelana. Somer Engmeer (F.n. 4429) 2q0;3 4q 4g 7l-l 1l 9l 91 1 ecum,r 3 e we som e.e eensessee eaume eseeseso a me me ioer ns, causel eve % couscurin MS W

cause em a-m s % '8*

3 l BlJ lPI l l G101810:

L

[l Jtll l l l

i L

!! ll l l l1 1 I

]

l t ;l l 1 i eussuwevs asson, twesens swi usame sa' i *eaa

,s..

_ e ee m a _.a.e, mo i i i i,

i-

.e e tmact wam i.ao -- -.... _ _ _ __. m u..

coi On Novemoer 17.1989 at 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br /> wnh the plam at 100er, power (530 degrees Fahrenhen and 1000 pno an engmestmg analysa edacates that tas Feedwater Coolam injecuan System (FWCI) c:ay De teamrea to esirver Ocw in excess of system pump flow esaS*'3***s durms a loss of coolant accsdam wnh a umunansous less of a:r. De idenuhed proelem was the latlure to the open posinon of the reacter feed pu p mm:=um flow valves. Failure to the open posanon would place addrdonal flow requiremenu on the sys:em ;n:mps.

Although the analys:s was not fu".y completed. p!am rnanagemem took the conserveuve accen of decianng the FWC1 system moperable, ne umt was ptacea irno as sommistranve techrecat specmcauen umazng cenemen for opermana (LCO) requirmg entry into a seven day ume lumt for resorms FWCI to opersettnv FWCI c:mponstus sessamed avslaabte curms the LCO permd. An immeaste nouncauon was made to sne r

Nuclear Regulatory f*a'=====aa m accordance wah stanon admmisuratrve aures. Operaung procecures were reviewes and eartaed to grve adequate gedance wnh respect to the d scenano. Withm five cays a ;1arn amaaan=t was completed which matalled a seismically quaufted safety related att m.typty to ins reseter feed 7.uny Y r. mum flow valves and FWCI was placed m an opersale status. No systems were reqmtes to funcuan cung:g t!us event. No safety consequences resuand trian tsaa event.

As a resuh cf the soove event ;! ant management decided to pursue further analyra m order to determme tne ceraequences of ine pcznnaten utcauen. n;a s.rpplemental repen ec:sams the results of the analves.

1

I l

~

l 4

i l

]y m**

enegas.

..s c ame mesa.,6tomv ccan es o.

se L;CINSit IVENT REPCRT ILER)

.*""* a"",",,,, g',,,, 4a*

)

• 1XT, NTINWATlCN se assens _ _ _ e,se essi.wa.eaan i

D. =anameEe===eMoM j

issuspeo =emen se mass

• • as 4 1

• • a v w m oocaresemeen :

~D WI!nene Nucient Power Stanon U^^ l

~

~

et si sj cl alt la ls

$19 Ol 28 2 011 0;; 0F:Oi' 1

v m--.

ca,-.aan...oti z.

~.n--....u..,

On Novemoer 1? 19f 9 at 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br /> wnh the plars at 100% power ($30 degrees Fahrennen ans 1000 ests) an ensmeenne analysts metated that the Foesesser Coelant truecnon Sysem (FWCII 1

r.sv et recuaroc to oeuver flow in sucess of system 60w capabstmee ourms a loss of cootant scescent with a simunaneous loss of air. The estaan prooies was the failure to the open

cuuon of the reactor feed pump mmarnum flow vnhos. FaGure to the open posson wound place acdntonal flow requirements on tne sysem pumps.

I Although tne analysts was not fully completed. plant manarmaar took the conse.rvsovo acnon of declarms tne FWCI synom moperaele. The uns was placed tone sa adsumarauwe tocassoal l

specificauen lumtmg coneuen for coernoon (LCO) requeng omry;sso a seven day tune ilma for remorms FWCI to operseday. FWCI campaaa=== remassed availette durag the LCC pened. An immemate nonficanon was made to the Nustaar Anguissesy Comesseman m accordance wak asoon ademmrauve procedures. Operaung procedures were revsewsui and venSed to pve adequata gualance with respect to the noennSed seenano. Wldda Sve days a pleas moencanon was comp!sted wtuch insalled a sensaucany quah8ed asfety reissed air supply to the reactor feed pump r.trumum flew vams and FWCI was placed in an operabis naam. 'No symams were requesd to funcuen cunns in:s event. No safety eaaaayaaaaa reeuand from this evens.

I As a result of the above evem plant manapmase corsamed the Geheral Electnc Company, the onpr.a1 nucasar steam tupply symem vender. m order to perform harther analyms to estersune the conseeuences of the postutetes sinaeuca.

l l

This supplememal Licensee Evem Report eammma the reauk of ths;saalyes and ts provided as a resun of the co - - - cornamea m LER 59-011-00 to provide a ayya===' report.

t%

< hag Vanous plata revmws resukms from Genene Leser 88-14 and a ad-mansed Safety Symen Funcuanal Inspecuan ($$FI) focused asenoon on synes insersonas essween the PWC1 and marumem air rymeras. As a resuk of these reviews a became roeigmaad that the effects of a loss j

of mstrument tar upon the reactor feed pump muumpun flow rearcutanon vslves 1-FW-laA.3 and i

C wousa cause all three valves to fad a the open direseen. The conseguance of suon a fadure could impact tne FWCI synem by meressas demand on the synem toyond the capocay of the symem pumps. On Omotor 10.19I9 a fonaal repenatesy evehanden was vuestais a socorcance i

wna conquany proceeares (Nuenaar Enaneenne and Operanens Presseurs 1.13).

I Recogrumme ther an maansa synom medaitng effort would to regeered to aderem the opera 6 day concern, a Meenmary engmeenng mvesnessen began to assess the posennal for FWC1 to to i

rendered asportese due to a loss of air. The resus of this pressumary revmv mecated that FWCI pums curves may not enveiope the sancipated range of Sows and 1pressures that would be encounteroa.

Stauen management took the conservauve acnon of decianns FWCI inoperseis shhoush a was apparent snat acamonal ume was needed to assess the esta and/or osnasa asemanal asia.

I The root cause of stus event has been deterramed to be madequese destga review dunns msinal plant cesign. These volves nave two domen funcnons, one for punap prosecoon and one for N g

--...~v

t..

.a w.. e,d nae sedress noen seemes.

'I

t O-'*-"*S*

.a %.a.as co. c:mmeeman

= m om e.. e.e,e, useos.sseer JCENSEE EVINT REPO8t? f!.ER) asumense a6 rum, sur ammuuano n. so.us,== wie I

• * * ' ' * * * " ' " ' * * = ' * * * * *m==

• EXT CONTINUAT1CN 7eumen.'"""l*""eE[asE*"""

-= M.ammsi-E_ _nd"m"'e.C w. e.ammi.

oc seems

..c.,. v %,.

poc.c.ueen ;as

>=

in

• .en +

v1he b 4 - -=}

Mdr.one Nuclear Power Stauen Umt I

~

of si el of oss 14 Is 819 0121 011 of3.0F 0t!

s n. - = =...

e m., m,

11]

.1....

.,e....

The ong:nst ceter nmauen of FWCI synem mopermothty required emry into plant Tecnrucai Specshcauen 31.C.3. Further analysa of it.e possulated event has been completed. It u tr.e Opmten et stauen management tnat the pesculated conazion couad have me potencal of rencenng a tafety nstem moperseas. Theretors, me event u reportante unoer 20CFR*0.73(a)(2)(v).

I The resu2u of LOCA analysts pronced m me UFSAR remamed vald even af FWCIis not crocaea a the anatwss. The basu for tras conclusaca is prowood below:

Lme beak LN A.

Because of the miscuen locanon (downeamer) all FWC! miecuan will bypass the core and spd! out througn the brsax. In a large erosir FWCI does not counbute to core cochng. Acequate core cochng is provices by Low Pressure Coolarn Injeccon and Core Spity pumps. Therefore, large breas LOCA analysts prowded m the CFIAR renams vahd enn t' FWCI a not creened.

0.a s

.w tm o The small break LOCA ar.alysts prowded in the UFIAR assumes that me gas curtans gesarator fans to operate tassumso worn smane fanures. Therefore. the analyms assumes no Aow from me FWC1.

wruch is powerec ev sne gas tureme generator. Even tf the FWC1 :were moperable. the waru smgle fatture wou)d haw remamed falksre of the gas martsne generator. His la because the gas i

n:rnme generator start cme (48 secs a s=ade==w less than the ADS delav tune (120 secsi. In a i

sf.all cream LCCA.rnacucn new rrem tr.aI Lov Pressure Cac;arn trueccon anc Core sprav pumps recomes svauspie song Defore me reactor preesare vesset a espresserued ey the Automacc j

Deoressuruauon System to anoe now pressure tryottaca. This is the case wnether the pumps are powerea ey the caesel generator er the gas rJrtnne. Derefore, whether FWCI u creened er not 2e srnail brent LOCA analysis prowded m the UT5AR wound renam vahd.

r. m.- e -am. A.i Addtnenal analvs s to determme system response was begun m conhmccon with the submmal cf

}

LER 89-012 00. Generna Electne Company was grven ad necessary and available data relaur.g to i

cent.raracon and tasang of the FWCI rysem.

i The analysa posadates that an evern req.unns FWCI is trunased. The everu resuha m a low resster water tovat causes the feeerster regulaung valves to go Mly open dunng the intual transient.

Ftrare 1 is gripneed for backgrm.na. An assumpoon a maae such that me miramurn flow vatves of all three ressem'leed pumps fad open. Under these condsmone theisysem runaut Aow cernrol serpomt of 130% is never reached. The anatyas evaluates two aJpecu of synom operstnhty:

1) The abery of the synes to operate contmuously wnh mmunues $ow valves failed open, and l

21 The staltry of the rysem to miset grsater than a.000 pm. The analvsu exammes the eperetthrv corscerne uomt a FWCI symem mency-nase moest dovsnopeo spec 26cally to amarnme trus taras anc ey roweeing FWCI symem performance tasang.

The analysts was developed usmg a model with certata hastanons as fonows-For waam flew rates grsater tnan 11.000 pm the mocal exzrapotstes the pump hand and now curves The feed re r'.uaur: varve throttang charactensuc (CV) used is the... - - vatue.at wruen runout wi!! occur The mocet simuates nnngs of condensate. consonaate Doomer anc feeoweser pumps The mocal

.a...a-...

j i

a -.

. e _....e s connee...

amamme m, -.

i.JCENSgg T_ VENT atPORT ttIR)

= =

• • =*

-man

'u==='

"=

same

• 1XT ~NTINUATICN amas

.e ie.i easi. u a a ---

esem -- _ _ _r

.ma oc seems c_ _

m iaise.4sesi o=

___ _ we name==eweisi oc seem

• ~e m oocar 'u esa m

.sp u mme s.

• • ma.a.

Was (3 ~ =-[f3 -

Wi!nene.%ctaar Pcotr Stauen I

' I# I

~

ci si el el att la is 819 01212 0l1 01 4 OF Oi!

• awe smaan * =en===e

.moeeneena. eaca,. 3esae,ein e -e-e..---._, a - > Nm (Cenunuec i r

ne mecol was usec to ceterrn:ne FWC1 performance asaarrang the trummum Cow vajves fatl open. i Se mecel calcu stes tnat a flow of 8.540 spm ts cettvered to the vessel whale cris enure rugn pressure sumg cenunues to operate. T'herefore. based upon the fo$owing t

tne FWC1 system remamse avatiatie cunng prewous tesung at ranout conratacra. and 1

tne synem remains capacia of $sliverms greater than rated fices to the veuel assumes the mammum Sow vaive failes open.

l ine failure of the feedwatsr marumum flow valves does not agruScam!y impem the rapaM*y of the FWCI rystem to perform ns intanced fanc=an, tvem r--<

~,eme..

Even tacugn tne stauen analyma atecussed above primdes some can5dence that the FWC! ryisem would remam operable under tna pommiated circumstances (if the W Flew Valves faded opent. NNECc coraervauvely cannot ta!!y credt tirs anstras smce dynamne effects could not be

=ocaled, anc tne smcy was necesurJy constramed a scope and 4W.

I

!: was inerefers censervatrvely assumed that the faihre of the Mantmum Flow Eves could have

susec tne FWCI system to be incapable of moeung as denen funcdona. De ufery consequences 1 of this potennal less of hath pressure (tugh Bow) inpocoon would be a decrease m the marge to

{

m:ugate tne potennas effects of a small brsant loss of coonsat acadera.

[

l v

wre m l

Operaung procacures were renewed and venfled to sve adequate cadance w:tn respect to the isennhed scenano. A plarn desgn change was trutaaned to prtmda a Dack-up aar supply to the affectec v Jves. The back-up att supp6y was deessmed to sansme, safety class cas cruana. The supply consists of two high pressure at Docues with noosemary preemare reaucers, tutung and valves to previce separanon and rW>ey frtxn the non-seierme plars tastrumeru tar rymam ne back-up air supply was made operaconal and FWCI was declared operante before the techmcal specWeauen LCO expered, t

i

_A d 88ar**al t-_deruptanren I

/

1 1

A report v2 not be g:ven.

j

. m mo= e aim

,si

. a e as m.a.4rcar,c - - --

osmosi.

i AEEEE" U2NSEE EVENT mgPCRT (LER)

• • ==mius

%"c -'"~4.""2. oc%ElI-

• EXT CONTINUA

• ION

~

~~.

tammessaseesmer = seems suun l

ssen es,

~ ~ *"'* **

Docan m as i'

WhE ne Nucw Power St.auon 01 si el si el:Ia ls 1l 9, 01212,~ 011,015 OP. Oi!

• • M8M

• a8 A W WF%As W C SWu.* W 33a.73 I

i l

1

)

i I

a i

16=u==uum W E M 1)$

setsamt 4 A

lous gg U

N M_

e., w i

4 1 I i

O i

J b WW **lME W$

$N o,

m It.. ~

==

q,

MN d

r

. u...

i

-tMl>

W"" 'T

'"U" %

(#

?

8P. 9.

M.

s

h

• y -[m, r,

O

, 1 e mamma w,,

-Q_,.A,,.

una.

h 8

h a k a N

y 8f "Y% r

.i an= =,s.

{

TB (3)

&E tabem l

h I

t FEEDWATER CCCLANT LNIECTION SYSTEM FLOW DIAGRAM

f EAS-05-0190 DRF A00-03470-27 January 1990 DBAFT Killstone 1 NCI Analysis' Assuming Feedwatar Pump Minimum Flow Valves Fail Open 9

I.D. Poppel i

D.M. Ks11y t

5 i

l Reviewed by C.H. Stoll, Principal Engineer f

Plant Performance Engineering i

Approved by G.L. Sozzi. Manager Plant Periousance Engineering.

.(

t i

l I

t

=

PuRAFT 7

p IMPORTANT NOTICE REGARDING CONTENTS OF THIS PIPORT I

Please Raad Carefully The only undere= kings of the General Electric (CE) Company respectir4 information in this document are contained in the purchase ordar between

^

Northeast Utilities Services Company and the Canural Electric Company, and nothing courninad in _this document shall be construed as changing ths purchase order. The use of this information by anyone other than Northeast i

Utilities Services company, or for any purpose other than that for which it is intended under such purchase ordar is not authorized; and with respect to any unauthorized use, the Canaral Electric Company makas no representation or varranty, and asstanas no liability as to the completeness, accuracy, or usefulness of the information cone =f nad in this document.

i The information contained in this juse4N=eion.is.balieved by Canaral t

Electric Company to be an accurate and true representation of the facts kncvn, obtained or provided to Caneral Electric at the time this information was preparad.

i J

i i

f

4 DRAFT TABLF,OF Cv M S Section Title 2.111 1.0 Introduction 1

2.0 Impact on the Capability of WCI to Perform Its 1

Intended Function 2.1 Intended Function of the WC1 Systan 1

2.2 Min 4=m Flow Valve Failura Concerns -

2 2.3 Evaluation of WCI System Parformance 3

3.cr Conclusion 5

1 DRAFT 1.IST OF TASIIS I!Jda 2311 Table 1 - Input Used to Develop WCI Model 6

Table 2 - Modal Limitations 7

i Table 3 - WCI Performance - Min 4== now Valves Closed 8

l Table 4 - WCI Parformance - Mini== new Valves Open 9

I e aneeune,** mum-e-

e ese I

-111-

i 1

4

'!ST OF FITJFys Title ZAgg Figure 1 - Feedvater Coolant Injection System Flow Diagram 10 I

i Figure 2 - Model Output - Min Flow Valves Closed "WCI 11 i

Availability Test" Figure 3 - Model output - Min Flow Valves Open - Runout 12 Condition

.-e l

i l

i l

l l

I.AS.05-0190 DRA T

1.0 INTRODUCTION

A study of the f2edvater coolant injection system (WCI) indicates that prior to recent modifications by the Northeast Utilities Services company, the minimum flow valves of the feedwater system could fail open on a loss of plant air. Therefore, for a postulated accident condition, where loss of offsite power results in a loss of control air, the feedwater a

mini:n:a valves are assumed to be open during WCI operation.

This report examines WCI system performance with the minimum flow valves failed open and its i= pact on the capability of the W CI system to perform its intended function.

s 2.0 IMPACT ON TEr. CAFABILITY OF WCI TO PERFORM ITS INTENDED FUNCTION 2.1 Intended Function of the WCI System The purpose of the WCI system -(see Figure d) is to provida adequata core cooling durings small break loss of coolant accidants for reactor coolant temperatures greater than 330*F.

In accordance with the IDCA Analysis Report for the Millstone Unit 1 Nuclear Power Station, the WCI system is assumed to be capable of delivering 8000 gpm to the vassal during such events.

For a IDCA event coincident with a loss of offsite power, power is lost to the feedwater system causing the running feedwater, condansato and condensa'te booster pumps to stop.

The subsequent drop in the feedwater pump discharge pressure causes the automatic closure of the Feedwater Regulatin5 Valves (RVs).

The gas turbine generator is started and the selected WCI loads are sequenced on.

The nonessential feedwater flow paths are automatically secured and each of the selected condanzate, condensate booster and feedwater pumps are started in sequence given that the proper permissives are met (e.g.,

each suction pressure permiss!ve is satisfied). Once sufficient feedwater pump discharge pressure is developed the FRVs automatically open to permit the makeup of reactor vessel i

-05 0190 inventory.

~he feedwater pu..p mini =m flow valves are assumed to operate automatically to provide adequate pump protection (opening below pu=p 6

suction flow rates of 2.5*10 lb/hr).

3e FRVs operate in the level control mode until the feedwater flow rate has increased above 120% of the rated flow (i.e.,

9600 gym).

At this point the control system switches into a flow control modo demanding a flow rate of 115% of rated (i.e., 9200 gpa). Ones the reactor high level reset point is reached (+37 inches) the FEVs automatically return to the level control mode.

2.2 Minimum riow Valve railure Concerns If the minimza flow valves fail open, the WCI system availability and the effective capacity of the system (i.e.,

flow to the vessel) may be affected.

To better understand the resultant impact, the following feedvatar control system response is postulated following WCI system initiation:

a)

The reduction in watar level. due to tha -postulated IDCA avant causes the feedwater control system to de=ami the FRVs to be full open.

b)

The FRVs respond to a full open position once the feedvater pump r

discharge pressure exceeds the FEV valve open permissive, c)

Due to the flow split between the minimme flow linas (of both the operating and non-operating feedwater pumps) and the feedwater

\\ inj ection path, sufficient flow is not atesinad throu6h the feedwater flow elements to activate the flow control mode of the WCI system at 120% flow.

I d)

The TRVs therefore remain full open until level is sufficiently recovered such that the difference between the level serpoint and the actual reactor level A *= mds the FRVs to a throttled position.

i 0190 I

For the conditions above, the postulated opening of the minimu:n flow valves increases the total system flow (i.e.,

the flow through the condensate portion of the system) which reduces the suction pressure available to the pu=ps within the system.

~herefore, this condition could potentially result in the trip of the selected feedwater pump on low suction pressure-and thereby impact the availability of the system.

Additionally, the condition splits the total system flow between the minimum flow lines and the feedwater injection path to the vessel. This could potentially result in flow rates to the vessel that are less than d===a4d, thereby having an impact on the effective capacity of the WCI system (i.e.,

the flow delivered to the vessel).

Canaral Electric has ev==4nad these concerns using a WCI system steady _stata modal and by reviewin6 WCI system performance testing. The WCI system staady stata model was developed using the system performance information supplied by Northanat Utilities, shown in Table 1,

and is subject to the limitations dascribed in Table 2.

2.3-Evaluatiotr of-WCI Systam Performance The WCI system performance test conrheted. 6/84, T84-1-12 "WCI Availability Iast",

demonstrated that the WCI system is capable of injecting into a reactor vessel at atmospheric pressure without tripping the feedwater pump on low succion pressure (at 80 psig). htrther review of the test data shows that a paah flow of approdantaly 12,400 gpa was attained with a single TRV full open.

This testad condition certainly represents a minimum resistance to flow condition and indicates that a runout how condition most likely occurs at approximately 12,400 gpa.

A review of each of the WCI system pump performant e curves further supports this deduction. Of the three pump curves, the pump performance curve for the condensate pump (when extrapolated) exhibits practically no increase in pump flow at flow rates near 12,400 gpa for a relatively large decrease in pump developed head.

Ihis indicates that the condensata pump is the portion of the WCI system in runout for those events that result in the reactor and/or the FEVs offering an insignificant amount of hydraulic,

T.AS-05 0190 resistance.

~c.e WCI steady. state model was used to simulata such a condition.

(i.e.,

conditions similar to the 'WCI Availability Test").

Figure 2 is the output of the WCI system model issuming one FRV is full open and the reactor vessel is at atmospheric conditions.

The results show that the total system flow is near tha 12.400 gym demonstrated during the WCI availability test.

The model was then used to determine the system flows, shown in Tabis 3, at various reactor pressures assuming the minimum flow valvss are closed and both one and two mVs are full open. If runout flow is defined as 12,400 gym, it is clear that two full open EVs have the capacity to put one condensate, condensate booster and feedwater pump into runout at reactor.pressuras below 600 psi 5 and for one full open EV, runout occurs at pressures below 20,0 psig.

The model also demonstratas that condansats boostar pump and feedwater pump suction pressures (61 and 265 psia, respectively) remain above trip setpoints (20 psig and 80 psig, respectively) at the assumed runout flow. Nevertheless, the NCI control system is designed to prevent a runout.candition as.dascribed in Section 2.1.

Givert the assumptions er==ined above, the system flow to the vessel exceeds 1204 (9600 gpm) of the rated flow. At that system flow, the WCI cEtrol systam will "thansfer from a level control mode (which may initiall'~

~

~

y d===nd the FRVs to be full open) to a flow control mode that throttles the FRVs to a total systan flow of 115% of rated (as datermined in Section 2.1).

The model was than used to determine the systam performance assuming the aid== flow valves have failed open.

Figure 2 is the modal output assuming a runout condition occurs at 12,400 gpa, the minimum flow valves have failed open and the reactor is at atmospheric pressure. Under thans assumptions the modal demonstrates that greater than rated flows are delivered to the vassal (i.e., 8,400 gpm).

Also note that the condensats booster pump and feedwater pump suction pressures are the same as those reported above (i.e., the case where the mini nne flow valves are closed and a single EV is full open). The results of this case demonstrates that the WCI system, in a assumed runout condition and the minimum flow valves 4

DRAFEDWm dEAS-05-O.?O

~

b failed open, is capuole of delivering greater than rated flow ro the vessel with out tripping the feedwater pumps on low suction pressure.

Table 4

)

reports the WCI system perfernance at variote reactor pressures for the same assu=ed concitions.

3.0 CONCIISION A review of the WCI availability test results, system ptzp performance curves and the results of steady-state system modeling indicates that a system runout condition occurs at approximately 12,400 gpm.

Additionally, the performance test demonstrated that the WCI selected feedwater pump did not trip on low suction pressure at this runout condition.

The staady-state model demonstrates that flows greater than rated are delivered to the vessel assumitig the

4ni=

flow valves have

~

failed open.

Consequantly, more than ample capacity is available to the vessel (i.e. + Ereater than 8000 gpa) assn =i ng the failure of the mini==

flow valves for the postulated -aidae.conditiona.. Tharafore, based on the following;

~

that the WCI syscam~ils Tesined ~available during previous

'~

tasting at runout conditions, and that the system rammina capable of delivering greater than rated flows to the vessel assuming the minimum flow valves have failed open.

N the failure of the feedwater minimum flow valves does not significantly impact the capability of the WCI system to perform its intended function.

t i

l

?).5 - 0 5 -01 ? O 4

DRAFT TABLE 1 j

INFUT USED TO DEVElDP F7CI EODEL a)

I1LV C versus position and total C 560 per valve, y

y b)

Calibration temperature for reactor feedvater flow elements.

c)

Feedwater, Condensate Booster and Condensate pump flow curves, d)

Feedwater pump minimum flow line orificing characteristics, 1

e)

System line losses based on computer edits showing the feedwater pump, condensate booster pump and condensate pump discharge

.. praasuras..ac..ratad.rasctor. -d ' ** ans.

f)

Feedwater piping and instrument diagrams, and g)

Feedwater system isometries.

l l

l

\\

1

\\

6-j

I.AS-05-0190 DRAFT MODEL I.DCITATIONS t

a)

For system flow rates Ereater than approximately 11,000 gpm the model extrapolates the pump head and flow curves.

b)

The model determines the FRV Ce required for a given system flow.

In this

==nnar.

the model matches system conditions on the condensate /feedvater side of the HIVs to the conditions on the reactor side of the FEVs for a given system flow.

[ Note: For a runout condition the C shown in the model outpurs represents the y

mi nimm FRV position at which runout will occur. ]

c)

The model similates strings of cond/cond booster /feedvater pumps (i.e., it is unable for.axample. to simulate 2 condensate pumps P

operating with..ons. condensats boostar pump and one feedwater pump).

d)

Because sufficient data was not available, the flow resistance of the non-operating (vindmilling) feedwater pumps are conservatively neglected by the model, e)

This model ignoreo dynsmic effects such as control system dynamic response sad the time responses of pressure switches, all of which would be highly dependant upon the actual control system tuning.

1 I l

TA.S.05-0190 DRAFT TASLE 3 WCI Performance - Mini =u:n Flow Valves Closed Condensate System Flev Remeter Pressure 1 FRV Oeen 2 W.V Oren (PSICS (KGPM)

(KGPM) 1200 9.9 10.8 1100 10.3 11.2 1000 10.6 11.6 900 10.9 11.9 800 11.2 12.1 600 11.8 12.4 400 12.2 12.6 200-12.4 12.7 100 12.5 12.7 0 --

12,6 "12^.8~~ - ~~

(

l 1

8-j l

ts.05-0190 4

DRAFT WCI Perforr.ance - Minism Flow Valves Open 1 IZZ 2 ZE2 PJACTOR COND SYSTDi HDU TO RFV COND SYSTDi FIDW TO PEI PRESSURE FIFJ AND CCRRESPONDING FIFJ AND CCRRESPONDING (PSIG)

(KCFM)

C, (KGPM)

C 1

y 1200 11.95 7.70/MAI 12.4 8.34/1056 1100 12.24 8.11/mut 12.4 8.34/658 1000 12.24 8.34/518 12.4_

8.34/518 800 12.4 8.34/391 12.4 8.34/391 600 12.4 8.34/326 12.4 8.34/326 400 12.4 8.34/286 12.4 8.34/286 200 12.4 8.34/258 12.4 8.34/258 100 12.4

~

8.34/2I6" 12.4.

8.34/246 I

i i

'l

.g.

EAS 03 0190 k 5g dL h 1 g

3

$s m

3-M km$l L

J t

z

/////// '////////W///

Z 1

d E

g 3

@M @M i

as g_

s V

u 2nb xp

,i-e "8

X E

i i

in<

Eia E

(

5a k

T z

E 5

8b

~

n

.E 5

5 l

~~

52 5

5 V

g u

u 9

5

-8 m-N k

A I

o a

i K

=

e si a

en-es- !.

l 4

4 U

b sw 6m

_gng se sm-lin y y

+

+

h'O E$<

t$a zE 1

an

!*Eh i

5 E

=O j$

Eg mb8 ha a

g

-Oc-em

(

sg L

J X

Figure 1 - Feedwater Coolant Injection System Flow Diagram i

i l

i Docket No. 50-245 i

B14692 1

-l i

i i - Exhibit 20 i

Millstone Nuclear Power Station, Unit No.1 l

l l

-u 4

i i

December 1993 I

1 l

l j