ML20081J046

From kanterella
Jump to navigation Jump to search
Rev 1 to Problem Investigation Rept 2-C90-0294:on 900914, Vendor Personnel Drilled Through Modified Blind Hub Into RCS Pressure Boundary,Resulting in Cooldown to Cold Shutdown. On-line Leak Repair Procedure Revised
ML20081J046
Person / Time
Site: Catawba Duke Energy icon.png
Issue date: 06/12/1991
From: Hampton J
DUKE POWER CO.
To:
NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
2-C90-0294, 2-C90-294, NUDOCS 9106210028
Download: ML20081J046 (8)


Text

_.

',o h

l Duke hunt Coinpany -

- v:

Catau'ba NuclearStation

'~

PO Box 256 L M

. Chur, SC29710 ;

DUKEPOWER a;y

.x -

June 12,;1991 c

Document Control Desk U.J S. Nuclear
Regulatory Commission Washington, D. C.

20555-

Subject:

Catawba-Nuclear Station Docket-50-414 Special Report, Revision 1 IIR C90-067-2; PIR 2-C90-0294 Gentlemen:=

Attached is a revision to the report submitted October 15,

-1990 concerning-UNIT COOLDOWN TO COLD SHUTDOWN DUE TO r

= REACTOR-COOLANT SYSTEM LEAKAGE.

This report is being

' submitted as~a."Special Report" to ensure industry awareness of.this-event =-

The health'and-safety.of.the public.were not affected by this-incident.

Nary trull' yours,

. W., Har pto Station Manager.

ken: REPORT.SP xc Mr.'S. D. Ebneter M & M Nuclear Consultants Regional Administrator, Region II 1221 Avenues of the Americas

' U. S. _ Nuclear Regulatory Commission

_New York, NY 10020

.101 Marietta Street, NW, Suite 2900 Atlanta, GA 30323 R.:E.. Martin INPO Records Center U. S. Nuclear Regulatory Commission Suite 1500

' Office of Nuclear Reactor Regulation 1100 Circle 75 Parkway

. Washington, D c.

20555 Atlanta, GA 30339 f4 OA Q.

' ~ ' ' EOU #

.Mr. W.-T. Orders NRC Resident' Inspector Catawba Nuclear. Station-9106210028'9106f.2 --

[]Uhi /

PDR. ADOCK 05000414 I.

S-PDR s~-

s

+-

n

DlWF p0WEP, COMPANY E

CATAWBA NUCIEAR STATION g-PROBLEM INVESTICATION REPORT NO, 2-C90-0294, Revision 1 I

UNIT COOLDOWN TO COID SHitfDOWN DUP 10 RF AC10R Cn0LATI SYSTEM l EAKAGF ABSTRACT On September. 14,-1990 at approximately 0200 hours0.00231 days <br />0.0556 hours <br />3.306878e-4 weeks <br />7.61e-5 months <br />, with Unit ? in Mode 3, llo t j

Stanlby, vendor personnel were in tie process of performing a leak repair on a Cote Exit.Thermncouple Nozzle Assembly (CETNA) on the Peactor Vessel head.

in order to perform the leak repair, it was necessary t o drill t hrough a modi fied blind hub (part of the Grayloc finnge at the CETNA) into the kenetor Coolant

.(NC) System pressure boundary.

A sealant injection valve was installed duiing

-this process to provide isolstion at-the pressure boundary.

Af ter the final drilling was completed, the scalant injection valve would not close A second valve was t hr eaded into the back of - the first valve, but before it could be closed the entire assembly ejected f rom the hole.

A conservative decision was made to consider the leakage as NC system pressure boundary leakage.

Unit 2 commenced cooldown-to Mode 5 Cold Shutdown, and an Unusua l Event was declared.

I An analysis was performed to determine the cause(s) of this failure The original CETNA leaks are att ributed to a Mannfneturing Deficiency in that the modified blind hubs did not meet dimensional requit ement s during fabricat inn.

' The unacceptable hubs we re identifled and replaced.

This report -is being submitted as a Special Report, t

4 W

4

-u.

a.__.____..__._.._.__.-_._,...._-.________._.._....

1

..1 l

DUKE POWCR COMPANY / CATAWBA NtiCI.r5R STATION PIR-2-C90-0294/Special Report, Revision 1 l

4

page 2 I

.l BACEGROUND

. l The Reactor Coolant [ETIS: AB} (NC) % stem is designed to transport heat f1om the

- J Reactor t o t he Steam Generators [ Ells:llX] (S/Gs), wher e heat is t ransf erred to the Feedwat er [EIIS:SJ] (CF) Systrm and ' fain Steam [ Ells:SB} (SM) Syst em of t he secondary side.

The NC System consists of four identical heat t ransf er loops connected in parallel to the Reactor Vessel [E11S:VS1}.

The Incore Instrumentation [r11S:1G] (ENA) system provides information on the neutron fTux dist ribut ion and fuel assembly outlet temperatures at selected cote locations.

Chromel-alumel thermocouples are threaded into guide tubes that penetrate the Reactor Vessel head thsough seal assemblies, and t erminat a at the exit flow end of the fuel assemblies.

Catawba Unit 2 ut ilizes a Core Exit Thermocouple Nozzle Assembly (CETNA) t o provide the NC System pressure bound 9ry seal where t he t hermocouple leads penetrate the Reactor Vessel head, Thete are five CETp' on Unit 2 ident ified as connections 074 through 78 The CETNA, supplied by < mbustion ingtneering, Inc. (CE), consists of a nonle assembly which is thread-l and seal welded to the Reactor Vessel head nozzle, a Grayloc clamp set, a modi fied blind hub t o form the second half of t he flange and house Graf oil packing rings, di ive sleese and nut, thrust-bearing and washers, and a hinged split collar to ret ain t he assembly.

Gray Tool Company manufetured t he original Grayloc clamp set s,

-including the modified blind huhs.

Gray Tool lat er gave up thei r N-st amp and turned.over t he manuf ac turing' responsibil it y' for t hese part s to CE.

During the Catawba Unit 2 End-of-Cycle 2 (200C2) refueling outage, discoloration was found on three of the original modified blind hubs (manuf act ured by Gray Tool),

As a conservative measure, these hubs were replaced with spare hubs manuf actured by CE.

'eakapa inspections were conducted during start up, ami no leakage was noted.

Tnhhical Specification (T/S) 3.4.6.2 requires NC system leakage to be limited to no. pressure boundary leakage during Mode 1, Power Operation, Mode 2, Startup, Mode 3, Ilot Standby, and Mode 4, Hot Shutdown, Pressure boundary leakage is defined as-leakage (other than S/G tube leakage) t hrough a non-isolable fault in a NC system component body, pipe [ETIS: PSP] wall, or vessel wall With'any

. pressure -houndary leakage present, the required act ion is t o be in at least Hot Standby wit hin 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and in Mode 5, Cold Shutdown, wi t hin t he following 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />, The T/S Bases states that pressure boundary leakage of any magnitude is unacceptable since it may be indicative of an impending gross failure of the pressure boundary.

' EVENT DESCRIPTION-On; June 10, 1990, Uni t 2 was shutdown f or t he End-of-Cycle 3 (2EOC3) refueling

. outage.

Following couldown, it was noted that Core Ex i t-Thermocouple Nozzle Assemblies (CETNAs) had been leaking at; the Grayloc flaner connection.

w

q:

h :

DUKE p0WER COMPANY / CATAWBA NUCI EAR STATION Fili 2-C90-0244 /Special Rel ort kevision 1 l

i page 3 Maint.enance Engineering Services (MES) contacted Design Engineering and the

.CETNA suppliet, Combust ion Engineeritte (CE) to evaluat e the tenson for the leaks so that corrective actinn could be t aken prior to restart.

During disassenhly of t he - CETNAs, a breakaway t orque check W5ts pet formed oij the Gruylor clump nuts.

The breakaway torque r esult s indicated that t im bol t ing mat er ial had relaxed during plant operation, therefore, CE recommended a new torque procedure for the Gr tyloc f1 mge connect ion, which Design apptnted fot use at Cataktm.

I On August 19, 1990 the CETNAs were reassembled using t he new torque procedure per Work Request.(W/R) 5312 SWE On Sept ember 4, 1990 with Un;t 2 in Mode 4, Maintenance inspect ed t he React or Vessel head and CETNAs for leakage CETNA #74 had baron residne present, indicating leakage had occurred. but did not appear to be leaking at this time CE was contacted and recommended that

-t h e CETNAs he rechecked in Mode 3 due t o -

a design feat ure t hat provides better seating at higher ptewures.

On Sept ember 5 at. 2155 hours0.0249 days <br />0.599 hours <br />0.00356 weeks <br />8.199775e-4 months <br />, Unit 2 ent ered Mode 1.

r on September 6, the CETNAs were reinspected and no leakage was ident11ied.

On September 8, anot.her inspection was performed and boron residue was found on CETNAs #74 and #76 in the area of the lower Grayloc clamp seal.

On September 9,

[

Maintenance cleaned the boron residue f rom bot h leaking CETNAs and observed a l

small st eam lenh at a76 (d74 did not appea r t o be leating at this time).

CE was-contact ed and recommended that a torque check pass he performed on the Gtayloc l

flanges.

Maintenance performed the torque check'on #76 and did not note any nut movetnen t, indica t ing the prop (,r t orque existed, On September 10,-

the CETNAs were inspected several times, and both #74 and #76 were leaking a small amount.

of steam.

At the time, leak repair by injection of a sealant at t he Gray loc L

-connection was considered, This proross is regularly used t hroughout. the-industry for repair of leaks under pressure.

MES, Design, and CE conducted extensive evaluations t o establish a sealant injection leak repair procedure for L

the CETNAs.

During this evaluation process on September 12 at 0500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br />, Unit 2 entered Mode 2 f or complet ion of 7ero. Power Physic s Test.ing (7 ppt),

l On September 13, Catawba Nuclear St at ion Modi f.icat inn (NSM) CN-20626 und the l

required 10CFR50.59 evaluation were approved t o allow injection of a sealant into the CETNA seal ring areas on 474 and C76.

The process consisted of

-pat tially drilling into the modified blind hub at the Grayloc flange, tnpping t he hole,- inserting a threaded sealant injection valve [FTTS:V}, and drilling through the remaining hub thickness ta the NC system pressure boundary.

The sealant. injection valve would then be closed to isolate the pressure boundary, and reopened when the sea 1ing process was to begin.

Mocirup training sessions were conducted to.fami1iarize workers with the repu11 procedure and the configuration ot"the involved components.

W/E 3109 MES was originated to implement this mod firation.

a -.

... w

... _ _.. _..., _ _.. -. _ -,., --_, _,., ~, -,,

1 6

e DUlT POWER COMPANY /CA1AWi;A NFCI 1.AR ET A'l ION Pilt 2-C90-0294/Special Repont, Revisien 1 l

rare 4 i

Oil h8'p t t'ilikit' r } 3 :it 1200 litairs, Uli i ! 2 e-tit e! eul Mculet t ii ; t l liiw Cl1NA t <>[i:1 i t k t il I to begin.

At algit ox imu t ely 2110 houts, a ju e job mer t itm w m held Imtwern Radiation P t o t ec t. i nn (1:P). MES, Maint-n nwe, a:nl Et ilit ies Suppmt <pec t il t s t.

Inc (USs1), who had been cont iact ed by Inike Powet ta pri f a t in the ieptir wo k.

u (Mt the RP re41ue s t ed t ha t the area he ' loaned pitot to uoth beginning init ial pat t of the j ol could be performed w i t hout t espira t or = and addi t ional pr ot ect is e weat The cleaning wa' comple t ed at a ppi m. i u t e l y 2100 hom s Also du t itte Pt e-j ob pt epai n t ions. emch sealant i nj ei-t t ish Valte w1 W c)* led wve ra l times to ensore ptopei opetation.

On Septen. bet 14 at approximately on2O htmrs, U%) 3"twinnel entoted the aten to Imgin wot k on CI TN A 976 ihe initial diillieg w e performed, the hole wn tapped, atul the sealant injection valve was inserted.

At app r ox inn t e l t 0200 houts the final diilling inta !he NC %) stem ptessute bounlary was t 'omp l e t ed.

As the drill hit was ret rac t ed f rom the hole, C%1 per sonne l at t empt ed t o eler the They then setewed a sealant inject ion valvo, bllt the valve ktsuld nat i

.w second sealant injection salve into the back of the first talte while holding the first valve in place When the first t ais e was telee>cd in oider to close the seconi valve, the entire as wmbly ejected itom the hole Wtthin minutes, a centei punch was tapped inta the hole to at i empt to plug the opening 1.c ala ge was rethtced but did not stop completely.

0%I personne1 le1t the aten at tbls time to loent e anot het size cent er punch t o better plug t he hole At app!oximately 0305 houts, USS1 a t. t cap t e i t<>

insert a centet punch with a i87 inch diomelet, but ww unsuccessful.

The at iginal cent et punch um :"i nse rt ed,

again reducing the leak but not complet e l y stoppiny it Saatt af t er t he pl oblem occu t i ed, the 5h i f. t Maniger, Shiff Supervisot, M1 ;, and Maintenance met to evaluate the NC system leak %e A ionsetvattse decision w %

reached to sonsider the leakage as NC sy st em ptwsttre houndary leakane The T/S 3,4,b. 2 act ion s t at ement was entered and preputations began foi cooldowit to % ute 5.

On September 14 at 0710 hunts, Operations commenced coaldown to Mo I" 5

.Also nt this t itne, an Unusual Event was declared and proper notifications were made per datawha Fmergency Response Procedure-E.ven tbough leakag+ h'id been reduced by ins t allat ion of t he cent er punch, et t at I s continned ta minimtre the led age to prevent contaminnt. ion of ot he r component s in t he ar ea.

On Sept mhe t 16 at 142R hours. Unit 2 entered 44dr 4.

On September 14 at a pp rox i mi.t e l y 1530 honi', 0%1 p"rsonnel inst all< d a mulitied C-c i uap in place of the center punch at CPTNA v76 ir, older to minimive l e aka r.e This reduced leukat, even futthet, but not complete 13 On Septemhor 14 at 2338 hoors, Unit enieie1 Made ; and the Unusual Frenf um t+ rm i n a t ed,

~ ~... - -. _. -..

~. -

' DUKE p0WER. COMPANY /CATAWIR NUCl. EAR S'I ATION -

PIR.2-C90-0294/ Spec 1al Report,.

l Kevision 1

-page.5 Following Unit 2 couldown to Mode 5, CE and Gray Tool representetives were onsite t o assist in the final CETNA repair voik.

While inspecting the spare modified blind hubs manufactured by CF, it was determined that the hubs did not meet.the dimensional tolerances required f or these pat t s.

Af t e r evaluat ion, i t was also determ:ned that the utmensional error would have ptevented proper seal seat ing cont act when inst alled.

The original hubs manufactured by Gray Tool were checked and no dimensional problems were found, All five CETNAs were subsequently checked to verify which hub (CE or Gray Tool) was currently installed.

CETNAs #74, 76, and 78 had hubs manufactured by CE, and #75 and 77 had t he original hubs manufact ured by Gray Tool.

The original Gray Tool hubs for connectors n74, 76, and 7R were located, cleaned, and Inspected, No dimensional problems were found, and the hubs were approved for reuse on September 21, the original modi f ied hiind hubs manuf act ured by Gray Tool wer e installed at-CFTNAs #74, 76, and 78.

After this replacement was compleIc, al1 five Unit. 2 CETNAs had dimensionally acceptnble hubs in place On September 25 at 2318 hours0.0268 days <br />0.644 hours <br />0.00383 weeks <br />8.81999e-4 months <br />. Unit 2 entered Mode 4.

On Sept ember 26 at approximatel) 1530 hours0.0177 days <br />0.425 hours <br />0.00253 weeks <br />5.82165e-4 months <br />. Maintenance inspected all five CETNAs for leakage.

No leakage was noted.

On September 27 at 0449 houts, Unit 2 entered Mode 3.

At approximately 2110 hours0.0244 days <br />0.586 hours <br />0.00349 weeks <br />8.02855e-4 months <br /> Maintenance again inspected all five CETNAs.

No leakage was noted.

. CONC 1.USION The original CI:TNA leaks at e at t r ibut ed to a Manuf acturing Deficiency in t hat the modified blind hubs manufactured by CE did not meet requ.ed dimensioual l.

tolerances. -The original hubs manufactured by Gray Tool were checked and met the dimensional requirement s.

Inspection of all five CETNAs revealed that #74, 76, and 78 had hubs supplied by CE.

These CETNAs were ident i fied as the ones that had experienced leakage.

The original hubs manufactured by Gray Tool were jc

-locatedi-cleaned and inspected.

No dimensional problems were not ed and t hey i

l

'were installed on Unit 2.

All five Unit 2 CETNAs now have dimensionally I

acceptable modtfied blind hubs inst alled.

During subsequent ' st art up preparat. ions. Maintenance inspected all five CETNAs in Mode 4, and again in Mode l1 3,- for leakage.

No leakage.was noted.

Atter investigations concerning the Jmpro'perly manufactured hubs, CE feels that thelimIroper hubs-are isolated to Cat.awba-2, Millstone-3, and CE St ock All i

improper hubs at Catawba have been located.

In addit inn, CF contact ed Mi:11 stone-3 to advise them of tbe buproper hubs The Duke Power Qua1i t y Assurance-Vendor Division has been contacted and advised of this problem.

l L

Whi]e at t.empting to repair t he leaking CETNAs, the sealant injention valve Installed at the NC syst em pressure boundary would not close A second valve was threaded into the back of the nriginal valve, but before the second valve could be closed the ent' ire assembly ejected from the h, il e.

MES completed a root l

_. _ - ~,. - _... _. -. _ _ _. _ _. _ _ _, _..

DUKE POWER COMPANY / CATAWBA NUCilAR STATinN l

plR 2-C90-0294/Special Report, Revision 1 f

Page 6

)

L

-cause urutlysis ain! conc luded t he pr oblem was t he inabil'ity to achieve proper thrend engagement,.This was due to the angle of t he hole not allowing full use of t he available t hreads on the inject ion valve and t hread lengt h on t he valve not being cottsidered in t he original review i

The decision to coimider t he leakage us React m Coolant. System pressute botmdary leakage, and the resul t ant. Unit cooldown atui declat at ion of an Unusual I vent,

was a conservative action not st rict ly i equi rel by Catawha Technica l Specifleations.

Pressure. boundary 1ankage 1imit s are est ahlished to preclude growth of defects to the point where coolant leak _tates pose a threat to nuclear safety.

Unit shutdown is t equit ed be f ore crack pr opagat ion rectil t s in potentially limiting leak tates.

The drilling of a 3/16 im b diamet er hole is not considered to have pon_d a demonstiahle porential ior propagatlon.

I.enkage t hrough t he drilled hole was well wit hin the capability of the normal chaining system and posed no danger t o unit safety, This event was conservat ively report ed to the NRC under the provisions of 10CIRSO 72.

This r e po r t is

- submit ted as a special Report and is not requi r ed un:le r 10Cf7t30,73 This event was consl.dered for report abili t y under t he requirement s of 10CFR part 21 with the conclusion that existence of a subst ant ial sat et.y hazard was not demonstrated. Ieakage through the CETNA was of a magnitude that would have l

remained well within the capabilit y of t he normal. charging syst em.

The potential for a cata.; trophic failute of the CCTNA. leading to o loss of coolant accident and safety system challenge, was not et eat ed as a result of the i

out-of-toletanee fittings.

Further, the vendor concluded t hat only t wo plant e,

Catawba Unit 2 and Millstone Unit 3..could pot ent ially have the out -of-t olerance l

parts. Thus, it was concluded that a Part 21 Yepott was not tequityd.

ihls report is submitted as a Special Report to ensure industry awareness of fhis event.

l I

-A review of the OEp database for the past 24 motiths revealed one event in which an operating Unit was shutdown due to a Manufacturing Deficiency (IIR 413/90-24).

This incident involved a Nuclear Service Watet [ElTS:Bl] (HN)

System pump [ETTS: Pl motor {EITS:M0} for which a stator had been manufactured

.appt oximat ely 0,1 inches shorter than required.

Two other events in this time l

period were att ributed to %nufacturing Deficiencies.

IIR 414/89-01 invoked a Reactor Trip' due to a fuse that failed on a feedwater cont rol valve.

LER

.l 413/89-26 documented an unexpected Hydrogen Skimmer ran (EITS:BLn] breaker

'l

{EITS:BRKl trip due to a defective Westinghouse ht aker.

'lhese two incidents I

are different-from-the incident in this re po r t - in t hat they were not attributable to dimensional requirement inaccuracies.

per Nucleat Safety Assurance guidelines, this is not a recortlug problem CORRECTIVE ACTION I

sttBSEQUENT l

1) After the sealant injection valve assembl3 ejected from the NC system pressure boundary, Unit 2 come nced couldown to Mode 5 and au l'nusual Event ww declared.

l L

i

\\

-e

' V

, DUKE POWCR COMPANY / CATAWBA NUCLEAR STA110N P!N 2rC90-0294/Special Report,' Revision 1 l

-Page 7 l

l

2) During subsequent itwest igat ions, the spare modi f ied Itlitul hubs manuf actured by CE were inspect ed and it was det eunined that -t hey did not meet the dimensional requirements for these part s.
3) The original hubs manufactured by Gray Tool wer e inspected.

No dimensional

. probleins were noted.

4) All five Unit 2. CE'INAs were inspected to determine whether CF or Gray Tool.

~

hubs were installed.

CETNAs #74, 76, and 78 were identified as having CE hubs installed.

5) The original Giny Tool hubs for CETNAs #74, 76, and 78 were located,-

cleaned, and inspected.

All were accept able f or inst allat ion on tinit 2, and were installed per W/R 5312 SWR.

4

6) Maintenance inspected all five CPTNAs for leakage during Mode 4, and again in Mode 3.

No leakage was notrd.

7) The Duke Power Quality Assurance Vendor Division was contacted and advised of the problems concerning the hubs manufact ured by CE.

'8) Revise on-line leak repair proceshu e t o inclinie caut ion against using an angled hole for any leak t epair operat lon.

The procedute revision has been approved to ada a sign-aff for verification of act.ual thread engagement.

-9) An analysis was performed to determine the cause(s) of the failures associated with the sealant injection valve.

SAFETY ANA1.YSIS 4

Following-this incident, Performance conducted a review of Operator Aid Computer (OAC) data.to establish the leakage ret e at the repnir locat ion.

An initial

-review indicated that NC system leakage at the COTNA was 6 to 9 gpm.

Atter n

-detailed. review of plant evolut ions in progress during t he incident, the actual leakage rate was-determined to be less -t han 5 epm.

A leak of t his magnit ude is well within normal charging pump capability (approximately 150 gpm)-an<l falls well below the limits of a small break Loss of Coolant Accident (1.0CA ).

The in h s act:ual hole diameter (11875 inches) was less than the hole sir.e of '.375 c e considered in Sect ion 15.6.5 of the-Catawba Final-Saf ety Analysis Report L(FSAR) for'which NC system inventory can be. niaintained by one charging pump.

The henIth and safety of the public were not affected by t hi s.. incident.

1

'*+4*'t-

-4w fr rip

-'===++r+F%*r-

  1. ++N'tet r my y-w e.. wee.yrMew 1--

-ip-N3$

9***

uh6B-*W"W*1t9'-4-dWWW

-'tF"'EN"-*

-tTg-*r-W 1*"WTr

% WTr r -wa+->vg-----mmye yg13 tgy-w

+8'P

q fM