ML18096A786

LER 91-009-02:on 901220,discovered Unsealed Portion of Heba Barrier Between Inboard Mechanical Penetration Area & Chiller Room.On 901101 Concern Raised Re Adequacy of Evaluations.Missing Portion installed.W/920611 Ltr

ML18096A786
Person / Time
Site:	Salem
Issue date:	06/11/1992
From:	Pollack M, Vondra C Public Service Enterprise Group
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-91-009, LER-91-9, NUDOCS 9206240010
Download: ML18096A786 (8)

Text

• -* **** *----* .:_ ***

__ _..,,i...

____ J,;.L.* ** o..._ __ _::,'7; *.*

..

.,.._._,,_,..,_.._

-,_...._*c

__ ...._._ ** ._,.,._, ___ , _ _._, __ , ___ **---

I.. e .e *.Public Service Electric and Gas Company P.O. Box 236 Hancocks Bridge, New Jersey 08038 Generating Station {\, '.** u. s. Nuclear Regulatory Commission Document Control Desk Washington, tic-*_ .20555

Dear Sir:

SALEM GENERATING STATION LICENSE DPR-70 DOCKET NO. 50-272 UNIT .NO. 1

• June 111 1992 . SUPPLEMENTAL LICENSEE EVENT REPORT 91-009-02 This Supplemental Licensee Event Report is being submitted pursuant to 10CFR50.73.

This supplement clarifies the assessment of the event significance based upon completed engineering investigations and corrects editorial mistakes contained in the prior supplement.

It is also being submitted as per NRC Inspection Report 272/92-03.

MJP:pc Distribution.

T !'1P r=n:.:>rov Penr1ic 920624001'0 9206!'1'*'

PDR ADOCK 05000272 S PDR ':'"'. Sincere y yours, . A. Vondra Gener.al Manager -Salem Operations 95-2189 (10M) 12-89 NRC FORM366 (6-89) U.S. NUCLEAR REGULATORY COMMISSION APPROVED OMB NO. 3150-0104 LICENSEE EVENT REPORT (LER) FACILITY NAME (1) Salem Generating Station -Unit 1 EXPIRES: 4/30/92 ESTIMATED BURDEN PER RESPONSE TO COMPLY WTH THIS INFORMATION COLLECTION REQUEST: 50.0 HAS. FORWARD COMMENTS REGARDING BURDEN ESTIMATE TO THE RECORDS. AND REPORTS MANAGEMENT BRANCH (P-530). U.S. NUCLEAR REGULATORY COMMISSION.

WASHINGTON.

DC 20555. AND TO THE PAPERWORK REDUCTION PROJECT (3150-0104).

OFFICE OF MANAGEMENT AND BUDGET. WASHINGTON.

DC 20503. 'DOCKET NUMBER (2) I PAGE 131 0 I 5 I 0*1 0 I 0 I 21 7 12 1 I OF 0 I 7 TITLE (4) 1High*Ene£gy Line Break Concern Between Mechanical Pene. Area & Chiller Room EVENT DATE (5) LEA NUMBER (61 REPORT DATE (71 OTHER FACILITIES INVQLVE_I;>

(8) MONTH DAY YEAR YEAR rm

{\

MONTH DAY YEAR FACILITY*NAMES DOCKET NUMBER(Sl Salem Unit i OPERATING THIS REPORT IS SUBMITTED PURSUANT TO THE Rl:.QUIREMENTS OF 10 CFR §: (Cht1Ck on* or moro of th* following) 111) MODE 'Ill : J. _ :20.402lb)

.. **20.4051ci 50.73(i)f2lfiv)

--POWER *1 . . 20.40li(o)f1)(il 110 I 0 ,..._ 20.4051o)f1)(1il

• -150.3111clf1 I --50.3111clf2)

--1--x ......_ ......_ &0.73l*H211¥i 50.73l*ll2)fvii) 73.71lbl 73.71lcl OTHER (Specify in Abstr*ct below end in Tttxr, NRC Form 366AI llillllllll=

20.40lllo)f1

)(Ill). 20.405(0)(1

)(lvl 20.4051*Jf1lM 1---......_ 60.7310)(2)(1) 50,731*1f2)fiil 60.73l*lf21 fiiil &0.73loll2HviillfAI

......_ 50.73(ol(2)(viiilfBI 1--60.73lolf2ll*I LICENSEE CONTACT FOR THIS LER 1121 NAME TELEPHONE NUMBER AREA CODE * ** M.. *J. *Pollack LER Coordinator 6 D i 9 . 31 3 19. I 2 IO *r 21 2: COMPLETE ONE LINE FOR EACH COMPONENT FAILUR 0 E DESCRIBED IN THIS *REPORT 1131 CAUSE SYSTEM COMPONENT I I I I I I I I TURER I I I I I I SUPPLEMENTAL REPORT EXPECTED 114) n YES (If Y*S. compl*t* EXPECTED SUBMISSION DATE/ -Fxl NO ABSTRACT (Limit to 1400 spaces, i.e., approximately fifttHJn single-space typewritten lintJs) (16) .. I I I l I .MANUFAC*

TUR ER I I I. I I I EXPECTED SUBMISSION DATE (151 MONTH DAY YEAR l I I On 2/15/91, a Probabilistic Risk Assessment of a portion of the Unit*1 seismic gap seal, between the Inboard Mechanical Penetration Area and the Chiller. Room, showed a calculated change in core damage frequency for Salem Unit 1 from 5.BE-5/Yr to 8.5E-5/Yr (assuming a worst case Main Steamline break in the Inboard* Mechanical Penetration Area). Assessment shows that the health and safety of the public was not affected due to available mitigating capabilities.

A similar Unit 2 seismic gap section had been found unsealed prior to discovery of the Unit 1 concern. The root cause of both Units' seismic gap seal(s) not being fully installed is "Design, Manufacturing, Construction/Installation" error (per NUREG 1022). The subject Unit 2 seismic gap section was sealed on 2/22/91. The missing portion of the Unit 1 seismic gap seal was installed on 4/12/91. The remaining seismic gap areas (both Units) have been .. _in.spected and. those seals requiring

.. repaiZ'. have bee11 repaired.

controls for.Appendix R penetrati9ns haye been e}!':tended to* include. HEBA barrier impairments.

NRC Form 366 .(6-891 LICENSEE EVENT REPORT {LER) TEXT CONTINUATION Salem Generating station Unit 1 DOCKET NUMBER 5000272 PLANT AND SYSTEM IDENTIFICATION:

Westinghouse

-. Pressurized*water Reaq1:9r LER NUMBER 91-009-02 PAGE 2 of 7 : __ ,c-.: Energy .Industry.

• .are .. -identified.;.in the text as {xx} IDENTIFICATION OF OCCURRENCE:

-Concern for effects of-a High---Energy Line Break *with a missing barrier between the mechanical and Chiller Room Event Date: 12/20/90 Discovery Date: 2/15/91* Report Date: 6/11/92 Thi_s .report was .initiated by Incident Report No.91-101. CONDITIONS PRIOR TO OCCURRENCE:

December 20, 1990: Mode 1 -Reactor Power 100% February a, 1991: Mode 1 -Unit shutdown in progress in support of ninth refueling outage preparation

• DESCRIPTION OF OCCURRENCE:

on December 20, 1990, during normal power operation of Unit 1, a Penetration Seal Review Group {PSRG) walkdown of Appendix R fire barriers identified an unsealed portion of a HEBA barrier between the *Inboard Mechanical Penetration Area and the Chiller Room, _at e'ievation-120'. The 6" by 5.5' opening was located where the*concrete slab above the Chiller Room approaches the outside wall of the Containment.

The slab -is a nominal distance of 6" from the Containment wall providing a seismic gap. The Chiller Room Handling Area) is atmospherically open to the Electrical Penetration Area {el .. 100') . The seismic gap seal at this location would prevent the steam environment from a postulated Main Steamline Break {MSLB) in the Inboard Mechanical Penetration Area from entering the mild environment of .the Chiller Room and Electrical Penetration Area. The PSRG initiated a work order to seal the opening. _ I A similar condition was identified previously on Unit No. 2 in May,_ 1990. Deficiency Report (DR) identified 104 penetrations . witb i_mpa.j.red seals in .f fre

• __ barriers that are also HEBA This number included one missing portion of the Unit 2 seismic-gap . ..seal-similar-to the Unit -1 opening described-above.

These impairments were not adequately evaluated against HEBA considerations.

On November 1, 1990, during the.SORC review of Design Change Package (DCP) No. 1SC-2183, which included two of the penetration seals identified in DR NFP-90-002, a concern was raised as to the adequacy of evaluations for openings in HEBA barriers.

Investigation resulted I.

\ . ' --......

.......

--.;;-*'

"' r' **.;:-; .* ;;,----*-*-.....

LICENSEE EVENT REPORT (LER) TEXT CONTINUATION Salem Generating Station Unit 1 ... DESCRIPTION OF OCCURRENCE:

DOCKET NUMBER 5000272 Ccont'dl LER NUMBER .91-009-02

• PAGE 3 of 7 ---* ._. I in the issuance of six ( 6) _ Discrepancy For.ms (.DEFs) ., one of * -.. * -. --which addressed:

the .*unit.. 2***seislliic openinCJ .:This DEF.,,. ** --DES--9o--;0:1573, :erroneously.

concluded .that no equipment -was aj: *risk whose failure would result in increased C-ore damage_f.requency

• . This occurred because it was not.recognized by the engineer making the I -assessment .that. the "Air Handling Area" contained vital motor control ** centers (MCCs). ' Following the Unit 1 finding in December 1990, DEF No. DES 91-00066 was initiated (on February 8, 1991} requesting.

an .evaluation of the safety significance of the finding and a Probabilistic Risk Assessment (PRA) analysis.

on February 15, 1991, the PRA assessment was completed.

This assessment sh.owed a calculated change in core damage frequency for Salem Unit 1 of 2.67E-5/Yr.

This raises the core damage fr.equency from s.:sE--5/Yr to Due to the potential for the identified condition to challenge the .... operability

.... of. .related .. equipment, the Nuclear. Regulatory Commission (NRC) was notified on February 15, 1991, at 1743 hours0.0202 days <br />0.484 hours <br />0.00288 weeks <br />6.632115e-4 months <br /> as required by Code of Federal Regulations lOCFR 50.72{b) (2) (iii). APPARENT CAUSE OF OCCURRENCE:

The root cause of the seismic gap seal(s) (both Units) not being installed is "Design, Manufacturing, Construction/Installation" error (as per NUREG 1022). The required seal(s) were not installed during plant construction (per design* prints). ANALYSIS OF.OCCURRENCE:

The leak-tightness design basis for the seismic gap seal between the Chiller Room and the Inboard Mechanical Penetration Area is to ensure that the Chiller Room (and Electrical Penetration Area) temperature remains < 120°F. Without the seal, a postulated 1 ft 2 MSLB in the Inboard Mechanical Penetration Area (allowing steam to enter through the deficient seal) could cause (i.e., worst case sce*nario) the three * (3) vital motor control centers (MCCs) in the Chiller Room to become inoperable.

Engineering Evaluation S-C-ZZ-MEE-0622 analyzed the effects of a MSLB* on the MCCs in the Chiller Room. The evaluation required extensive

.. thermal calcul_ationf:f for _modelipg the MSLB in the Inboard_ Mechanical-

• *-Penetration Area (seismic gap). Assumptions for the seismic gap . analysis include: 1) A _pressurized Inboard Mechanical

penetration Area --with steam temperature reaching 375°F when blowdown ends at 10 minutes; and 2) Inlet ventilation flow to the Chiller Room as permitted by the pressure therein. The Salem Unit 2 seismic gap was reinspected on February 19, 1991. A portion of the seal was found to be *improperly installe.d; however, I I

. *e.,.., ,. LICENSEE EVENT REPORT (LER) TEXT CONTINUATION Salem Generating Station Unit 1 DOCKET NUMBER 5000272 LER NUMBER 91-009-02

• PAGE 4 of 7 ANALYSIS OF OCCURRENCE: (cont'd} . anchored fiashing *was. in_P,iabe

• .

(3). cases.of the -.seismic gap .. .. ....... --*.--..:.,_

c::a57 No.: i *cunit.: __ *2) no,credit*

.. for*th7 .. , .. , .. f.lashµig

(.1.e.,

ar.ea -of 680 in )

• Case .. No. 2 (Unit 2) models partial blockage of the steam flow by the flashing-(280 in 2). Case (Unit 1) *models the December.

20, 1990, as-found condition (400 in 2). .Case No. 1 yielded an internal MCC temperature of 230°F--(worst .. case).:':

'for 96. hours.at * ** . 131°F and 90% relative humidity.

The evaluation concluded that operability of the targeted safety related lE electrical equipment could riot be demonstrated due to a lack of supporting test data. The MCCs control various ventilation and cooling equipment including: . . . 1) . 2) 3) Room Cc;>olers for: a) .Nos *. 11, 12 & 13 Auxiliary.

Feedw'ater Pump Rooms;' b) Nos. 11 & 12 Safety.Injection Pump Rooms; .c) Nos. 11, 12 &.13 Component Cooling Pump Rooms; d) Nos. 11 & .. _12 .. Residual .Heat .RemovaLP.ump

.. e) * .Nos *. 11 ,.& .12 Containment Spray Pump Room; and, f) Nos. 11, 12 & 13 Charging Pump Rooms Fans for the: a) Nos. 11, 12 & 13-Switchgear Room Exhaust, el. 64'; b) Nos. 14, 15 & 16 SWGR Room Exhaust, el. 84'; and c) Nos. 11 and 12 Electrical Penetration Exhaust No. 13 Chiller Condenser Water Recirculation Pump *An engineering review of the "loss" of the.above equipment shows the most critical equipment to be the room coolers for the: Charging Pumps -Nos. 11 and 12 charging pumps are required for high head safety injection in the mitigation of a design base accident.

They are 100%*redundant of each other. Safety Injection Pumps -both pumps (100% redundant to each other) would be affected; they would be used to mitigate the consequences of a design.base accident by providing intermediate head safety injection.* .RHR Pumps -there are 2 pumps (100% redundant to each other) which are used :to remove residual core heat during shutdown conditions .and to* mitigate the consequences of a .

base *accident-by providing low head safety injection.

The room coolers, .in conjunction with the once ventilation system, are designed to limit ambient temperature at vital pumping equipment.

This *helps assure* long-term reliable operation of the vital equipment.

Based upon PSE&G Engineering Evaluation S-C-ABV-NEE-0504E, "Engineering Evaluation on the Effect An Inoperable*

., '.'

LICENSEE EVENT .REPORT (LER) TEXT CONTINUATION.

Salem Generating Station Unit*l DOCKET NUMBER 5000272 .ANALYSIS OF OCCURRENCE: (cont'd) LER NUMBER 91-009-02.

-. PAGE 5 of 7 . Room Cooler.Has on the Operability'.of.

• vital';'"Pumps' 1:, .. operabi1it,y of .the .;.'. vita:l .pymps. is .affected cby-the* availcibilitj of**:associated*room

"**-*,:,coolers.

with :an inoperable .room -cooler, and without an* -analysis to show otherwise, the vital pumps ... in that room are considered inoperable

• ....

.. above **Engineering . Evaluation,.

among _.other conservatisms, .:assumed* -a. heat load generated by the protracted operation of the ECCS*pumps and other electrical equipment in the Auxiliary Building.

In the scenario for the HEBA in the Inboard Mechanical Penetration Area that impacts the Chiller Room through the missing seismic gap seal, the Steam Generator

.(S/G) blowdown through* the break is completed and auxiliary feedwater flow to the affected S/G isolated, by Control Room operator action, in approximately ten (.10) minutes (reference UFSAR Section ... 15. 4. a .2. 2 and '!'ables 15. 4-29_ and 15. 4-30) *. The Control_ Room .* . operators, using Emergency Operating.Procedures (EOPs),would shortly thereafter reset the safety injection signal, terminate operation of. the -ECCS.pumps

--and stabi-li-ze

• • Reactor Coolant. System -operating parameters.

Operator action times have been adequately demonstrated.*

on.the Sa:lem simulator during training and EOP validation.

Should the above scenario include a loss of offsite power, the ECCS pumps, by design,. would run for the first twenty (20) minutes without room coolers. The room coolers are locked out (on a loss of power event) by the Safeguards Equipment Control Cabinets.

Termination of injection provides the ability to only operate pumps necessary for normal* plant cooldown/shutdown and maintenance

• thereof.

-These actions would significantly lower the heat load on .the* Auxiliary Building Ventilation Syst.em (ABVS) *

• The ABVS operates during all postulated ac6ident scenarios loss of offsite

• power). Conservative calculations, which assume no room cooler operable, no condensation losses, no structural heat. sink losses, ABVS availability and all ECCS pumps running 6 show that affected pump room temperatures stabilize at less than 180 F. Based on NUMARC 87-00 projections, a pump can be reasonably expected to .operate for at least four (4) hours under ambient temperature conditions of < 180°F. Within thirty (30) minutes of the postulated steam line break, the plant would be stable and actions could be initiated to bring the p],ant to Modes (Cold Shutdown) in a controlled fashion. As *identified above,.*the available*

time for continued ECCS. pump availability is .at least four .(4) hours everi .with the pump rooms* heat rise calculation.

Within the first thirty-(30} minutes,.

Operations would be aware of Room Cooler failures because of Control Room indications.

Operator action can be taken to mitigate a loss of the Room Coolers prior to four (4)-hours having elapsed. Therefore, had this event occurred, the health and safety of the public would not have been affected.

However, this event is reportable per Code of Federal Regulations lOCFR 50.73(a) (2) (v).

• * .9 LICENSEE EVENT.REPORT (LER) TEXT CONTINUATION Salem Generating Station Unit 1 DOCKET NUMBER 5000272 LER NUMBER 91-009-02

.. ANALYSTS OF SUBSEQUENT HEBA BARRIER PENETRATION INSPECTIONS PAGE 6 of 7 --seals

• common to bo;th Appendix R and High_. Energy-Line. Break Accident -*.::: _ ---,.,-< (HEBA) barriers .(734:*:sea.1S).

were inspected as .part_ of.:the -Program ---.Ana-lys-is Group (PAG) program. *Approximately 10% (74) -of the inspected

• _:__-*---oseals were discovered to have openings_

requiring repair .* -Of these 74 seals, it was determined that 24 have no safety significance and an additional 14 have negligible .affect on-the environmental parameters

--,, **0-0f-' the *-targeted -areas. -Erigirieer.ing EValuation s-c-zz-MEE-0622

-assessed the environmental for some 221 penetrations including the remaining 36 impaired seals. As committed to in the original LER (272/91-009-00), PAG was tasked to* inspect HEBA penetration barriers (which are not Appendix R penetration barriers) to ensure adequacy.

of Results show that 113 Unit 1 penetration seals and 54 Unit 2 penetration seals required repair *. The nature of the impairments ranged from minor seal .damage to missing seals *. A DEF, DES-91-0086-,-was issued and the resulting PRA analysis showed no significant increase in core damage frequency.

Therefore, the ___ health,.and-safety .of .-the .. public was _not affected as a result of these impaired penetration seals. These penetration seals have been included in Engineering_

Evaluation S-C-ZZ-NEE-0622.

PRIOR EVENT ASSESSMENT:

j ' l I The Updated Final Safety Evaluation Report (UFSAR), Section 3.6.5.10,, j _ discusses leak tight areas of the contiguous zone; however, it does not detail specific requirements for the Inboard Mechanical Penetration Area *. Engineering Field Directive No. S-C-VAR-MFD-0508'-1 (issued in 1988) addressed this area and other areas of similar concern. The Field Directive was originally*

issued.in

• response.to LER. 272/87-017-02 which identified a concern pertaining to leakage paths in the steam Driven Auxiliary Feedwater Pump encl9sures (for Units t & 2). The design basis break.in the Mechanical Penetration Areas, at 78'.and 100 1 elevations and* the Pipe Alley, is in a 6 11 main steam line to the Auxiliary Feedwater Turbine Driven Pump. Another corrective action inthe above LER was walkdown of accessible penetrations in other pipe rupture enclosures.

Although this . inspection was performed, the seismic gap penetration areas were not specifically identified in the scope of that inspection.

The visual walkdown found no Unit 1 unsealed penetrations.

Two (2) unsealed areas in the Unit 2 Letdown Heat Exchanger Room were identified.

The .design. break is iil a 2 11--CVCS line and the break locatib!lS -are .. sleeved and restrained.

_ .Also, Section 3. 6. 5. 10 of the UFSAR

• indicates that all steam generated by this_-assumed break .can -be. carried :away by ' the norinal ventilation exhaust. Based ori the above.discussion, it is judged that a postulated HELB will have no significant effect on equipment

?perability in the areas.

• . ... e

• LICENSEE EVENT REPORT (LER) TEXT CONTINUATION Salem Generating Station Unit 1 CORRECTIVE ACTION: DOCKET NUMBER .. 5000272 LER NUMBER 91-009-02 PAGE 7 of 7 The unsealed portion of the vertical of the Unit 2

_gap .* * *1

• was **sealed on* November*

7, 1990. .Following the* February, 1991 ** . ** re'.""'insPection, the remainder .o'f .the Unit 2 seismic qap was* sealed on . February 22, 19.91. ** * -.,-* .. The missing portion of the Unit 1. se,ismic gap seal was. installeci.

on *April* 12'

• 1991.* .. . ' . Engineering Evaluation S-C-ZZ-MEE-0622 has been completed which assessed.

the operability of the Electrical Penetration Area MCCs with the*seismic gap not sealed. This evaluation concluded that operability of thermally sensitive electrical component designs such as relays, switchgear, motor control centers and control instrumentation could not be assured. However, as discussed in*the l . Analysis of occurrence Section, .this would not . have arf ected the * .: health or safety of the public. All Unit 1 and Unit .2 HEBA penetration .seals. have,been visually inspected and deficient seals have been repaired.

The .inspection and analysis of HEBA barrier penetrations not associated with Appendix R has been assigned to the Program Analysis Group as part of the Penetration Seal Program. This event has been reviewed by Engineering management

.. The PRA assessment engineer was counseled.

The need to maintain attention.

to detail and to challenge assumptions used in performing calculations was stressed.

• A 10% random sampling of DEFs (61 from a population of 605) was* reviewed to ensure that the PRA analyses were correctly performed.

Added to this sample were an additional six (6) DEFs associated with the HEBA barrier concerns first.identified in May.1990.

Results of the random sampling assessment indicate a confidence level in excess of 95%. MJP:pc SORC Mtg ..92-065 General Manager -Salem Operations have