ML20235A733

Safety Evaluation Re Installation of Alternate Rod Injection (ARI) Sys & Adequacy of Plant Reactor Coolant Recirculating Pump Trip (RPT) Sys,In Compliance W/Atws Rule 10CFR50.62. ARI & RPT Acceptable

ML20235A733
Person / Time
Site:	Brunswick, 05000000
Issue date:	09/18/1987
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20235A701	List: ML20235A699 ML20235A733 ML20235B312 ML20236P640
References
NUDOCS 8709230401
Download: ML20235A733 (6)

Text

. _ _ _ . . . . . .

ENCLOSURE 1 y

h 7, UNITED STATES NUCLEAR REGULATORY COMMISSION y ;j WASHING TON, D. C. 20555

\*...+/

SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION l l

RELATED TO COMPLIANCE WITH ATWS RULE 10 CFR 50.62 CAROLINA POWER & LIGHT COMPANY BRUNSWICK STEAM ELECTRIC PLANT, UNITS 1 AND 2 DOCKET NOS. 50-325 AND 50-324

1.0 INTRODUCTION

On July 26, 1984, the Code of Federal Regulations (CFR) was amended to include Section 10 CFR 50.62, " Requirements for Reduction of Risk from Anticipated Transients Without Scram (ATWS) Ei.ents for Light-Water-Cooled Nuclear Power Plants," (known as the "ATW5 Rule"). An ATHS is an expected operational transient (such as loss of feedwater, loss of condenser vacuum, or loss of offsite pcwer) which is accompanied by a failure of the reactor trip system (RTS) to shutdown the reactor. The ATWS Rule requires specific improvements in the design and operation of commercial nuclear power facilities to reduce the likelihood of failure to shutdown the reactor following anticipated transients, and to mitigate the consequences of an ATWS event.

For each boiling water reactor, three systems are required to mitigate the consequences of an ATWS event.

1. It must have an alternate rod injection (ARI) system that is diverse (from the reactor trip system) from sensor output to the final actuation device. The ARI system must have redundant scram air header exhaust valves. The ARI system must be designed to perform its function in a reliable manner and be independent (from the existing reactor trip system) from sensor output to the final actuation device.

2. It must hcve a standby liquid control system (SLCS) with a minimum flow capac'ty and boron content equivalent in control capacity to 86 gallons pet minute of 13 weight percent sodium pentaborate solution.

The SCLS and its injection location must be designed to perform its function in a reliable manner.

3. It must have equipment to trip the reactor coolant recirculating pumps automatically under conditions indicative of an ATWS. This equipment must be designed to perform its function in a reliable manner.

87092304o1 07091s PDR ADOCK 05000324 P PDR

By letters dated October 10, 1985 and April 14, June 18, July 22, July 24, and August 24, 1987, Carolina Power and Light Company (CP&L or the Licensee) provided information to comply with the ATWS Rule. This safety evaluation report addresses the ARI system (Item 1) and the ATWS/RPT system (Item 3). The SLCS (Item 2) will be addressed in a separate document.

2.0 REVIEW CRITERIA The systems and equipment required by 10 CFR 50.62 do not have to meet all of the stringent requirements normally applied to safety-related equipment. However, this equipment is part of the broader class of structures, systems, and components important to safety defined in the introduction to 10 CFR 50, Appendix A, General Design Criteria (GDC).

GDC-1 requires that " structures, systems, and components important to safety shall be designed, fabricated, erected, and tested to quality standards commensurate with the importance of the safety functions to be performed." Generic Letter 85-06, " Quality Assurance Guidance for ATWS Equipment that is not Safety Related," details the quality assurance that must be applied to this equipment.

In general, the equipment to be installed in accordance with the ATWS Rule is required to be diverse from the existing RTS, and must be testable at power. This equipment is intended to provide needed diversity (where only minimal diversity currently exists in the RTS) to reduce the potential for comon mode failures that could result in an ATWS leading to unacceptable plant conditions. The criteria used in evaluating the licensee's submittal include 10 CFR 50.62, " Rule Considera-tions Regarding Systems and Equipment Criteria," published in Federal Register Volume 49, No. 124, dated June 26, 1984, and Generic Letter 85-06, " Quality Assurance Guidance for ATWS Equipment that is not Safety Related."

3.0 EVALUATION OF ARI SYSTEM CP&L is participating in the BWR Owners' Group ATWS implementation alter-natives program. The BWR Owners Group submitted a licensing topical report NEDE-31096-P, " Anticipated Transients Without Scram, Response to NRC ATWS Rule 10 CFR 50.62," (Reference 1) for staff review. The staff accepted the licensing topical report, NEDE-31096-P, in a letter to the Chairman of the BWR.0wners Group dated October 21, 1986 (Reference 2).

In the staff' safety evaluation which approved this topical report, the staff developed an Appendix A for the purpose of itemizing those ARI features approved by the staff in order to facilitate prompt review of plant specific ARI' designs. The safety evaluation stated: "The licensees or applicants who comit to fully implement or have implemented an ARI design incorporating these features covered in Appendix A will be consid-ered to be in conformance with the ATWS Rule 10 CFR 50.62 paragraph (c)(3) on ARI requirements." In letters dated April la and August 24, 1987, the licensee summarized,their method of compliance with the ARI portion of the ATWS Rule.

t 4 As stated in the staff SER on BWROG Topical Report NEDE-31096-P, the staff does not intend to repeat its review of the design information described in the GE Topical Report and found acceptable when the report appears as a reference in a specific license application. Based on our review of the licensee's submittal, the staff finds that one aspect of the Brunswick ARI design, i.e., testability at power, is not in conformance with the BWROG Topical Report NEDE-31096-P and the staff's guidance in their ATWS SER on this topical report.

The ATWS Rule guidance states that the ARI system should be testable at power. The BWROG licensing topical report NEDE-31096-P specifies the method of compliance as follows:

The ARI system is designed such that a periodic surveillance test can be performed during normal plant operation.... This testing includes the relay logic to initiate ARI valves actuation....

Testing of final actuation devices (ARI valves) while the reactor is at power is not required since this could affect plant avail-ability. Surveillance testing should not prevent the ARI system from responding to an automatic ARI initiation signal.

The staff has reviewed the licensee's ARI system circuitry. The staff finds that the Brunswick design has two test switches (one for each trip system) which will block the automatic ARI initiation signals during surveillance testing. This is not in conformance with the design description present in the BWR0G's topical report, the review guidelines of the ATWS SER or the requirements of the ATWS Rule guidance.

3.1 CONCLUSION

ON ARI SYSTEM The staff concludes that the testability of the ARI system while at power is not acceptable. The licensee should modify the testability portion of the ARI design so that surveillance testing while at power does not prevent the ARI system from responding to an automatic or manual ARI initiation signal.

3.2 ARI TECHNICAL SPECIFICATIONS The equipment required by the ATWS Rule to reduce the risk associated with an ATWS event must be designed to perform its function in a reliable manner. A method acceptable to the staff for demonstrating that the equipment satisfies the reliability requirements of the ATWS Rule is to provide ARI Technical Specifications which include operability and surveillance requirements. The staff will provide guidance on a generic basis regarding Technical Specification requirements for ARI at is later date.

g r-

g. ? . (. y > -. s y

, m y/ ,.  ;

-9 v -y( ,

g ,

by ,

r l n_ r , ,

,  !? , ,

I t G'

y ,

, -4 t ,

, 4.C -EVlltJATION OF RPi- !;YSTEM E - During the courre of rev4w ci the ATWS long-term program, General Electric duermined; and the staff has concurred,'tnat a recirculating

, pump trip can sigMfiat,nt\y limit the consequences of an ATWS event.

^

3.

Tripping'betW recircGating pumprJ. in the event of high reactor vessel prersure or low vessel water lavek will cause an 16 crease in the moderstor voids in the reacts core which will significantly mitigate the ceasequences of an ATES event.

3 . - .

Tho 4taff'1 SER on BWROG Topical ReperC NEDE-31006-P endorsed the Matice1# desic6 and the modified Hatch desiga. The staff required th se M 111 ties with other types cf.RPT cesigns to submit their schedule fir upgrad'ns to either of the tbove two appro(cd designs or to 6emonstrate that their presert cesign can perform its function in a i 3'. reluble'mmier equivalent to the two approved designs. The Brunswick j Mrv.t has, da triginal BWR/4 RP1 system which uses one-out-of-two trip 1 iogic.totripeachindividual' motor-ger.erator(MG)setdrivemotorfeeder breder. The breaker usfd for tripping the pump only has one trip coil.

By letter dated April 10. 1907. CPAL counit ted to modify its original SVR/4 RPT design so that it Mll utilize redundant tr!p logic. The t orfpint.1 pressure Nitch m.scrs will oc, replaced by transmitters and {

anaTog trip units, such ti.at eJch indiviaual le ml or pressure instrument  !

4 chame? can be tested ducir.g pinot operation. Each logic train will trip l

bot.h retircult. ting pumps. CP&L's propcsed cesion to upgrade the RPT  !

sens6r hardware aild to trip both puurs try toch logic train is acceptable.

Howeser, r.P&L believes that the original RPT design, which iii their case utilizes a A W (4160 Y ectual) breakir, car be as reliable as the Monticello RTT design. The Monticello design. utilizes two trip coils in each recirculation syster !% set fiel1 breaker, GE Model Wo. AKF-2-25, 480 V. CP&C pcrtorme6 a nronahili! tic risk assesscent er,alysis (PRA) to  !

demonstrate that the Brnswick RP1 design wn1d tt as reliable as the  !

Monticelb WPT doign. CP&L's bas.ic crgumer.t was that the breaker failure i

' is the de,silant factor and that the contributlen of the coil failure to  !

'the breaker faiTurt rate-is insignificant. A detailed m iew of the PRA J wbmittais wa pw formed.cy a consul at to the staff, EG&G Idaho, Inc. l

- +

The mvled exanined the accuracy arN completeness of the logic models i devehped by Cf'&L and the failurf. data utilized in the quantification of l the logic models to assure that the availability estimates a re valid. The  !

" 'ronsulta't 'cencluded in Referer'ce 2, ( Attachment I to this Safety Evaluation), j t

a' Sth qualitatively and ydtitatively, that the Brunswick design was not

]- ;quivalent to the Monticello /.nico. '

v'+

, i b

r s

..tW AhCm___ _ _ _ _ _ . &____

A meeting was held at NRC's Bethesda office on July 1G, 1987 between the staff and the licensee to discuss the consultant's conclusions. CP&L committed to provide additional information regarding the PRA performed to support the Brunswick RPT design. By letter dated July 22, 1987, CP&L provided data for the new fault tree quantifications and the justification for data selection. CP&L c theMonticelloRPTdesigntobe8.2x10~giculatedtheunavailabilityfor

, based upon experience data from the field breakers at the Brunswick and Pilgrim Plants, and calgulated the unavailability for the Brunswick RPT design to be 2.2 x 10~ , based upon 4 KV data from draft NUREG/CR 4126, January 1985. EG&G, in Reference 4 (Attachment 2 to this Safety Evaluation), reviewed the additional information and, as a result, two problem areas were' identified with CP&L's revised analysis:

(1) The limited failure data available for estimating the failure probability of the field circuit breaker model AKF-2-25.

(2) The majority of AKF-2-25 breaker failures can be attributed to improper lubrication and/or misadjustment of the breaker internal parts.

The staff recognizes that there have been problems with RPT field breakers (GE AKF 2 - 25 model) in the past. However, the reliability has been improved when the utilities follow GE Service Information Letter SIL - 448.

EG&G, in Reference 4, concluded that the inherent reliability of the Monticello and the Brunswick RPT designs were identical except for the breaker and trip coil arrangement. Given equal mean time to failure for the field breakers and the 4 KV breakers, the Monticello design was considered inherently more <

reliable due to the redundant trip coi'is. The CP&L study does not demonstrate a convincing argument that the present Brunswick RPT design can perform 1ts functions in a reliable manner equivalent to the two approved designs.

The Monticello (or modified Hatch) design is the minimum acceptable design. The staff recognizes that many later plants use a design which has two breakers in series and which is more reliable than the Monticello or the modified Hatch design. This design has been accepted by the staff.

Consequently, the staff encourages utilities to install two breakers in series to improve the reliability of the RPT system.

4.1 CONCLUSION

ON RPT SYSTEM Based on the discussion above, the staff finds that CP&L's proposed design to upgrade the RPT sensor hardware is acceptable. However, the original RPT design employing a single trip coil for each breaker is not i acceptable. The licensee should upgrade the RPT b  ;

trip coils (previously approved Monticello design)yorproviding providingredundant two )

breakers with single trip coils to trip each recirculating pump. j I

l 1

L_----__--- .- )

l I

I 4.2 RPT-TECHNICAL SPECIFICATIONS 1 l

The equipment required by the ATWS Rule to reduce the risk associated l with an.ATWS event must be designed to perform its function in a reliable manner. A method acceptable to the staff for demonstrating that the equipment satisfies the reliability requirements of the ATWS Rule is to 3 provide RPT Technical Specifications which include operability and 1 surveillance requirements. The staff will provide guidance on a generic I basis regarding Technical Spe.ification requirements for RPT at a later date.

5.0 REFERENCES

i I

(1) BWROG Topical Report NEDE-31096-P, " Anticipated Transients I '

Without Scram; Response to NRC ATWS Rule 10CFR50.62,"

dated December, 1985.

(2) Staff SER on BWROG Topical Report NEDE-31096-P. Letter from Gus'Lainas (HRC) to Terry A. Pickens (BWR Owners' Group Chairman), dated October 21, 1986.

(3) EG&G Information Report EGG-REG-7766, dated July 1987 (4) EG&G Interoffice Correspondence from John Poloski to Art Nolan, dated August 17, 1987

!