ML17262A899

Summary of Operating Reactors Events Meeting 92-08 on 920610 to Discuss Selected Events Which Occurred Since Last Briefing on 920520.Reactor Scram Statistics for Wks Ending 920524,31 & 0607 Encl

ML17262A899
Person / Time
Site:	Hatch, Mcguire, Point Beach, Grand Gulf, Braidwood, Summer, Brunswick, Ginna, Callaway, Farley, Big Rock Point, McGuire, Trojan  File:Consumers Energy icon.png
Issue date:	06/11/1992
From:	Chaffee A Office of Nuclear Reactor Regulation
To:	Rossi C Office of Nuclear Reactor Regulation
References
OREM-92-008, OREM-92-8, NUDOCS 9206180120
Download: ML17262A899 (20)

Text

June 11, 1992

~

P WgQ

/./~a/97 5TQ MEMORANDUM FOR:

FROM:

SUBJECT:

Charles E. Ross',

Director Division of Operational Events Assessment Alfred E. Chaffee, Chief Events Assessment Branch Division of Operational Events Assessment i

OPERATING REACTORS EVENTS MEETING JUNE 10, 1992 - MEETING 92-08 r~

, ii On June 10,

1992, we conducted an Operati'ng Reactors Events meeting (92-08) to inform senior managers from the Commission

Office, EDO,

AEOD, ACRS,

SECY, NRR, and regional offices of selected events that occurred since our last briefing on May 2

1992.

Enclosure 1 lists the attendees.

Enclosure 2 presents the significant elements of the discussed event.

Enclosure 3 contains reactor scram statistics for the weeks ending 05/24/92, 05/31'/92 and 06/07/92.

Two significant events were identified for input into the NRC performance indicator program (Enclosure 4)'.

(original signed by Robert L, Dennig for )

Alfred ED Chaffee, Chief Events Assessment Branch Division of Operational Events Assessment

Enclosures:

As stateQ cc v/enclosures:

See next page DISTRT U ION:

Central Files KBaumann PDR DGamberoni EAB R/F

LKilgore, SECY DFischer ORTRANS.RPT (NP/KAB)

E DOEA E

D K aumann:kab R

g 06/Il /92 06/Ir'/92 OFFICIAL RECORD COPY DFisch 06/(i /

2 AChaffe Oe/ ]t/9

""':gETllRtl TO RiGULATOP~Y CE lTRAL Fll.ES

~~C7b20-20b 1 i PDR OR G NRRB PDR

N

OPERATING REACTORS EVENTS BRIEFING 92-08 V

N LOCATION:

8 Bll, MHITE FLINT REDNESDAV JUNE, 10 1992, ll:00 A,N, G INNA PRE-SEPARATOR DRAIN TANK RUPTURE

~

~

h

> (k C'

92-08 G INNA PRE-SEPARATOR DRAIN TANK RUPTURE JUNE 9i l992 ZB ALFIE RUPTURE OF A PRE-SEPARATOR DtAIN (PSD)

TANK IN THE TURBINE

BUILDING, HULK TMO-PHASE FLOM INPINGENENT OH THE PSD TANK MALL IS EXPECTED TO BE THE CAUSE, DETAILED NETALLURGICAL ANALYSIS IS IN PROGRESS, H

HAZARD TO PERSONHEL AHD PLAIT EQUIPNENT.

PLANT AT 97X

POMER, DURING A ROUTINE PLANT TOUR, THE SHIFT SUPERYISOR (SS)

HO~EP HOT ""66t tl6 O'I f0'- "h" 050 PAND< Atlh MATg4 GQLLKQTING UNDERNEATH IT, THIS TANK FUNCTIONS AS A NOI STURE SEPARATOR TO NININIZE THE NOISTURE LOADING IH THE P4 FM HEATER, JUST NIHUTES AFTER THE SS LEFT THE AREA, AT 0449A A

RUPTURE OF THE "A" PSD TANK OCCURRED, OHE FIRE ALARN ACTIYATED IN THE TURBINE BUILDING MHEH LOM PRESSURE STEAN RAPIDLY FILLED THE BUILDING, CONTROL ROON OPERATORS HEARD THE LEAK AHD RESPONDED BY CONNENCING A

POMER REDUCTION AT 1X PER NIHUTE, AT 0530, THE LICENSEE;~NOTIFIED THE HRC RESlDENT INSPECTOR CONTACT:

REFERENCE:

D, GANBEROHI, NRR/DOEA NORHING REPORT DATED 06/09/92 AIT: ~N SIGEVENT: ~B

IP', I I

) ~

P Ih 4'

GINNA 92-08 AT 0535, THE RATE OF POWER REDUCTION WAS INCREASED TO 2X PER MINUTE DUE TO INTERMITTENT GROUNDS ON THE 480 VOLT NON-SAFETY BUSES.

AT 0612'HE MAIN GENERATOR OUTPUT BREAKERS WERE OPENED AND THE TURBINE WAS TRIPPED ONE MINUTE LATER, THIS ISOLATED THE STEAM LEAK, NO SIGNIFICANT EQUIPMENT DAMAGE (OTHER THAN THE "A" PSD TANK) OR PERSONNEL INJURIES WERE

REPORTED, AT 0928, THE REACTOR WAS SHUTDOWN,

~QWW LICENSEE INSPECTION REVEALED AN 8-INCH LINEAR RUPTURE IN THE WALL OF THE PSD TANK.

THE PSD TANK WAS DESIGNED BY BROWN-BOVERI. FABRICATED BY CONRING FABRICATORS, AND INSTALLED BY ROCHESTER GAS AND ELECTRIC (RGE)

IN 1983','N

1980, RGE ADDED AN INLET NOZZLE J-TUBE MODIFICATION WHICH REDIRECTED

FLOW, THIS MODIFICATION INADVERTENTLY ALLOWED INLET FLOW TO PARTIALLY MISS THE IMPINGEMENT PLATE, THIS APPEARS TO BE THE ROOT CAUSE OF THE EVENT, POINT BEACH, FARLEY, AND INDIAN POINT 2

MAY HAVE SIMILAR COMPONENTS AND MAY BE INSUSCEPTIBLE TO THIS PHENOMENON, NRR AND REGION I

ARE DRAFTING AN INFORMATION NOTICE, THE LICENSEE ESTIMATES THAT A MINIMUM OF 2

DAYS WILL BE REQUIRED FOR REPAIRS,

GINNA BRIEFING 92-08 INLETS OUTLET TO 84A FM HE I

j 1

t l

I I

RUPTURE 8 BAFFLE PLATES

GINNA BRIEFING 92-08

~A PRE-SEPARATOR DRAIN TANK 0

0 0

RUPTURE 0

0 k

O' 0

RUPTURE lipll INLET NOZZLE BAFFLE PLATE

bP

~

1t 1

I

6.0 ENG INEERING/TECIINICAI.SU I'POR'I'71707, 92701) 6.1 "A" I'resepa "ntor Drnirr Tank I(rrptrri'e 6.1.1 Everit Dcscriptiorr On the morning of June 9, 1992, a shij't supervisor (SS) noticed a sinall puddle ol'ater under thc "A" prescparator drain tank {PD'I') while conducting a routine plant tour.

'I'hrough comparison by touch, the SS found the "A" PDT I igging to be significantly hotter tlian tliat of thc "B" PDT.

He then returned directly to thc control rooni.

As he entered thr control room, hc heard a lo'id roaring noise in thc turbine building and saw steam filling the building.

Auxiliary operators werc dispatched to investigate.

I]pon reporting that steam was emanating from the area where thc PDTs and fccdwatcr heaters are located, control room operators commenced a 1%/minute load reductio>> at 4:49 A.M. As a result of tlic turbine building steam environment, an erroneous fire alarm on Z-32 (turbine building basement north) and intermittent 480 volt non-vital bus ground fault alarms werc rcccivcd (and rapidly cleared) in the control room.

In rcsponsc to tliesc iiidications of degrading conditions, the load reduction rate was increased to 2%/minute.

'I'he main generator output brcakcrs werc opened at 6:12 A.M. and thc turbine manually tripped one minute later.

Shutting thc turbine stop and intercept valves.~vlatcd the leak.

Subsequent inspection of thc "A" PD'I'cvcalcd a wall failure in the form of a fish mouth rupture on thc north side of thc PD'I'.

'I'hc fisli mouth was approximately 8 inches long with a maximuin opening width ol about I/2 inch.

Using the clock location systcin, thc failure area was located <<t thc 3 o'lock position wlien viewed from the east end of the "A" PDT. Thcrc werc no personnel injuries and no obvious damage to surrounding plant equipment.

Examination of'he ground fault indicating lights by thc auxiliary operators on all safety anc, non-safety related 480 volt buses indicated nU off-normal conditions.

6. 1.2 Equiprtrerrt 1)cscriptiorr The "A" PDT is of carbon steel (A-53-B) construction rneasurirrg 36 inches in diameter by approximately 11 fcct long.

'I'he nominal wall thickness of the shell iuatcrial is indicatu'o be 3/8 inch.

Thc tank serves as an in-linc inoisturc collection header for extraction stcam being routed to the 4A I'cedwater heater.

Stcaru cxhaustcd from thc eleventh stage of the high pressure turbine is extracted at two points and directed through scparatc 31-inch piping runs to the in-linc IB and 2B prcseparators prior to cntcring thc I B and 2B moisture scpara',or rchcaters (MSR), respectively.

Thc saturated stcam/water mixture captured in the prcseparators is channeled initially through two 14-inch lines, then through 16-iiich lines to the PDTs.

Steam/water enters the "A" PDT through two inlet nozzlcs at thc east end.

Elbows installed on the inlet nozzles inside the vessel direct the flow axially along thc length of the vessel.

Through impingement, flow direction changes, and voluinctric expansion, water is removed from the !low stream.

At full power, thc I'D'I s collect saturated water at

g P

S~

Ig P

E

~

lt p

~

150 psig and 350'F.

Stcam with improved quality is directed frown the vessels to the 4A and 48 feedwater heaters.

Water is dischh~r'gul to the heater drain tank'or, on high PDT level, dumped to the main condenser.

1~

The PDTs were originally installed in 1983 as part of the Enginccring Work Request (EWR) 3100 MSR Upgrade Project.

Thc vessel was designed by the Brown Bovcri Corporation and manufactured by Conring Fabricators.

'I'hc internal nozzlcs and baffle werc modified in 1984 to correct lcvcl control problems.

At that time, due to space limitations in the "A" PDT, u>>e of the internal elbows was orientated at an angle with rcspcct to the vessel axis, directing thc flow toward the north vcsscl wall.

The rupture occurred in this area of the "A" PDT vessel wall, which had bccn extensively fi thinned by an erosion/corrosion (E/C) mechanism.

6.1.3 Licensee Response Following the rupture at 4:49 A.M., operations personnel initiallyreduced electrical generation at 1%/minute.

Upon receiving intermittent alarms for ground faults on the plants 480 volt buses, indicating that the turbine building steam environt'nent could be potentially degrading balance-of-plant electrical supplies, the rate of power riduction was increased to 2%/minute.

Plant equipmcnt functioned as required.

'I'hc main generator output breakers werc opened at 6:12 AM and thc turbine was manually tripped one minute later.

Shutting the stop and intercept valves effectively isolated thc leaking "A" PD'I', permitting inspection of the wall failure and assessment of repair strategies.

The reactor was maintained at about 1% power until 9:28 A.M., at which time site management directed that it be placed in hot shutdown conditions pending completion of PDT repairs.

Fwrlier, site management convened a scrics of planning mcctings to identify specific tasks to be completed to affect the repairs, and additional tasks that could bc facilitated during the outage.

A team of corporate mechanical engineers was assigned to provide continuous on-site coverage to support repair activities.

Initially, the licensee's Materials Engineering and Inspection Services Group completed a

100% volumetric examination, using ultrasonic techniques, ot both PDTs, in order to determine ifadditional areas of these vessels were at or near thc minimum wall thickness.

'I'his inspection revealed that other areas were below thc minimum wall thickness requirements.

Accordingly, repairs were extended to include those areas.

The repair technique to be used was to provide reinforcing material to the thinned areas through either the use of 3/8-inch A-36 carbon steel plate overlay, or the use of weld buildup.

Thc actual technique used would depend on thc size/location of the area requiring repair.

Following these UT examinations, a section of the "A" PDT wall (approximately 3-inches wide by 8-inches long) containing thc rupture was cut out and taken to the RG&E Materials Engineering Laboratory for examination, verification of material spccilications, and determination of the failure cause.

4 I)

I Ag(e

7 Final repairs to thc "A" 1'DT consisted of welding three two-foot wide, 3/8-inch thick, rolled A-36 carbon steel plates to the outside of thc tank.

Full pcnctration wclds were provided between plates, These plates extend 180 degrees around the tank, encompassing the rupture as well as areas identified to bc less than minimum acccptablc thickness.

Additionally, two smaller 3/8-inch thick plate ovi:rlays were wcldcd to the tank exterior opposite the rupture to complete reinforcement of thinned portions of the tank.

(

~

Thinned wall areas in the "B" PDT werc rcinforccd through weld buildup in two internal locations <<nd by installing a 15-inch wide by 26-inch long, 3/8-inch thick, rolled A-36 carbon steel plate at onc external location.

Following weld inspections, integrity of the "A" PDT was tcstcd under operational conditions in accordance with Power Piping Code B-31.1.

Spccilically, with thc insulation off, aAcr the first delivery of steain, a visual leak check was pcrfornicdI for 10 minutes in thc vicinity of the rupture.

Additionally, witli the insulation on, a visualt1cak check was pcrfornicd over a 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> period after full power and maximum prcssure werc achieved.

No leakage was detected.

'1'hc justification that these repairs provided an adequate margin of safety w;is documented by RG&E corporate engineering in the following design analysis reports:

Prcscparator Tanks A 4 B, Evaluation of required minimum wall thickness, EWR 3100, DA-ME-92-001-01, June 10, 1992 Prcscparator Headers A & B Basis for Repair Methodology, EWR 3100, DA-ME 002-02, June 11, 1992"."

These design analyses give predicted values for minimum,'wall thickness, based on thc actual erosion rates, that can bc used in periodic inspections.

Thc PDTs arc scheduled for a complete reinspection against tliese predicted values during~ thc 1993 refueling outage.

Until then, the liccnscc plans to conduct routine UT examinatio(ts over limited areas of the PDTs.

As part of their rcvicw of the failure of thc "A" PDT, thc corporate engineering team identified several components in balance-of-plant systems that could be susceptible to similar E/C failures which were not included in thc E/C inspection program because they were not considered piping.

These components, which included thc heater drain tank, house heating steam condensate return tank, steam generator blowdown tank, and 2nd and 4th pass MSR drain tank, will bc examined for wall thinning during thc 1993 outage.

6,1.4 Inspector Findings The NRC resident and regional staff reviewed the liccnscc's engineering response in evaluating the failure mode of,',the PDT and its associated repair.

The inspectors agreed with the conclusion of thc licensee'k metallurgical rcport that cbvercd the analysis of the riipturcd

C p

I

• P I

~

I i'I 1

)

Vp

E I

8 tank wall. 'I'his analysis vcritied tint thc I'ailiirc occurred in;i>> are'i ot'he tiink tliat had been i'ignificantly thinned by erosion/corrosion as a result of water <<nd ste;in) inipinge>>icnt.

'I'he l.

I';iilure was of a ductile natiirc and origi>>atcd i>>;in area of tlie t'ink in which the wall thick>>ess was rcduccd from a iioininal 0.375 iiich to 0.014 incli,'I'he i>>aterial properties werc found to corrcspond to the chemi))ical and>>ietallurgical protx:rties ot'he specilied AS'I'M A53 carbon stccl.

No inetallurgic il anonialics were observed i)),the i>>aterial slxcilicatio>>s ot'hc failed tank.

13ascd o>> a rcvicw of the licensee's tcchnical justilication lor the repair strategy, tlie i>>spectors identified sevcriil limitatio>>s concenii>>g tlie long-term integrity ol'he repair.

'I'licsc limitations included:

The licensee maintained that,'ltlioiigh the MSRs were designed to ASMI: 13oiler;i>>il I'ress>>re Vessel Code Sectioii Vill rules, iio sucli design)tii))) w:i>> giv<<>> to tlie I'IXI's.

r't}

'I'hey were not considered to for>>)ally coine under the rules ot'Section Villor tlie I'ower Piping Code (1331.1).'Gb'cl stated that, I'or the original vessel built by I3row>>-Boveri, thc fabricatio'ri principles werc to Section VIII, but that no Code Stan)p was necessary.

'I'he inspects')'r was aware of no precede>>ce set in eitlicr Seclioi) Vill or 1331.1 to peri>>it liip weld,'joints acting as welds for Class A horizo>>tal or circumferential joints (Ret'crit)ce Section Vill,'I')ble UW12).

I'he abse>>ce of,i (:ode stamp on the, vessel (pipe) would indicate. thc Ix>ssibility'(hat the complete sct ol'ules for vessel or pipe design werc not followed.

ASME apIIroval ol'lie design would provide assurance that thc st'icccssful experience of Code'vessels would provide a measure of confidence in the design, materials, and fabrication used.

i

~

In reviewing the detailed calculations, thc inspector found that some ele>>ients ol'he analysis werc not clearly shown.

In particular, DA-ML)-92-001-01 showed a weld et'ficiency factor of 0.7, which, for a single lap welded joint, appears to bc high and provides non-conservative estiinates of erosion progress.

Conceptually, the analysis was not clear as to tlie participation of the reniai>>ing portion of thc burst plate with the patch plate in contai>>>>ient ot the pressure loatl.

'I'hc plates, when welded togcthcr, both act in pressure load restraiiit. II'uch is the

case, there is concern for long-term progression of thc burst r>>outh.

The licensee explained that,since thc patch plate neutral.axis is not coplaner witli that of thu shcii wall, a small hei%ding moment would re<nit i[ Discussion widt thc iicunsee indicated that a bending stress of approximately 5000 fix)t-pounds was present.

'I'lie inspector concluded that this>stress lcvcl should have bet.n given consideratioii in forniulating thc repair strategy.

'I'hc repair was necessitated because of flow iini)ingei>>ent upon a shell wall such tliat, over a period of time, the wall had enxled to I'ailiire thickness.

I'liis indicated;i design deficiency.

Good design practice would provide t'or i>>ateri;il;idded to tlie w;ill thickness or a sacrificial platy, i>>stalled to absorb the erosive et'I'ect ol'lie iinpingi>><s fluid.

j

consultation with Westinghouse, RG&E site engineering determined that by adjusting thc gain on the Advanced Digital Fccdwatcr Control System (ADFCS), the cycling of the main fecdwatcr control valves could bc brought into

phase, thereby lessening flow imbalaiiccs.

Following review and,'approval by thc Plant Operations Review Committee (PORC) of the station modification procedure, SM-4773.33, and supporting procedures to change the ADFCS sctpoints, the gain was adjusted on June 16, 1992.

By observing this evolution, the inspector dctcrmincd that thc activity was well controlled by plant operations management and the site engineering staff in accordance with liccnscc administrative procedure A-52.15, "Significant Infrequently Performed Evolutions."

Prior to making the setpoint change, control room operators werc briefed by thc Superintendent-Plant Operations and the cognizant engineer..

Briefings included a summary of individual responsibilities, confirmation of lines'of communications to various plant areas, expected ADFCS response, and anticipated ogrator actions should ADFCS response be off-normal.

The procedure controlling the station'odification, SM-4773.33, was found to bc sufficiently detailed, requiring verbatim compliance with individual step signoffs.

The changes were completed without challenging plant stability and effectively corrected the cycling phase for the main feedwatcr regulating valves.

The licensee is continuing in its root cause analysis of feedwater flow oscillations to identify measures to optimize secondary plant performance.

6.3 Licensee Action on Previous Inspection Findings 6.3.1 (Closed) Unresolved Item (50-244/90-10-02)

Licenscc to verify separation of AMSAC Cable Run in Cable Tray No. 372 from Cable Tray No. 23.

In RG&E correspondence to the NRC dated May 27, 1992, the liccnsce provided the actions taken to verify the separation of the ATWS Mitigating System Actuation Circuitry (AMSA(')

cable in cable tray No. 372, a non-safety related tray containing AMSAC cable, and cable tray No. 23, a safety related cable tray.

The ATWS (Anticipated Transient Without Scram)

Rule, 10 CFR 50.62, and RG&E Ah<SAC modification Design Criteria (EWR 4230) require