ML20134H052

Submits Extent of Region Understanding of Problem Involving Deterioration of Insulation on Containment Penetration Electrical Circuits in Unit 1

ML20134H052
Person / Time
Site:	San Onofre
Issue date:	06/23/1987
From:	Kirsch D NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
To:	Rossi C NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
Shared Package
ML20127A865	List: ML20127A863 ML20134H015 ML20134H052 ML20134H101 ML20134H141
References
FOIA-96-354 NUDOCS 9611130508
Download: ML20134H052 (2)

Text

'} .

1. jtMac UNITED STATES

%g '

NUCLEAR REGULATORY COMMISSION 0 p L'9

[ E REGION V o 5

D 1450 MARIA LANE.SulTE 210 WALNUT CRE E K, CM,lFORNIA 94596 '

%*****,o June 23, 1987 MEMORANDUM FOR: C. E. Rossi, Director Division of Operational Events Assessment FROM: Dennis F. Kirsch, Director Division of Reactor Safety and Projects

SUBJECT:

POTENTIALLY GENERIC ISSUE RV-87-04, DETERIORATION OF INSULATION ON CONAX CONTAINMENT PENETRATIONS AT SAN ON0FRE UNIT 1 On June 15, 1987, Southern California Edison (SCE) advised Region V of a

, potentially significant problem involving the deterioration of insulation on containment penetration electrical circuits in Unit 1. The licensee is still attempting to determine the magnitude of this problem but the Region has been informed of the current developments and the following is the extent of our understanding of this issue.

During the current outage, SCE performed megger testing of the control rod drive mechanism (CRDM) coil stacks to determine if any damage resulted from

, previous cooling fm malfunctions. During this testing, although the CRDM coils tested satisfactorily, unaccentably low resistance readings were found on 11 CRDM cables. Troubleshooting determined that the low megger readings

were the result of insulation damage on " pigtails" associated with new containment electrical oenetrations that were recently installed during the last refuelino outaae (about I vear of operation). These new penetrations were installed as part of committed EQ upgrading. The new penetrations are manufactured to include pigtails on either end, which are field spliced onto existing plant cabling. The pigtails are multi-conductor cables that are factory spliced tn the single conductor of the penetration. The pigtiils are insulated by a very thin dielectric called DuPont "Kapton" (about 1 mil thick) with an additional thin teflon coating (about 0.5 mil thick). The insulation is installed as a double wrapping of Kapton tape, which is installed by the vendor, CONAX. The entire insulation is about 8 mils thick. Since the insulation is so thin, the pigtails are also protected by a polyolefin sheathing.

To date, SCE has visually inspected about 800 of approximately 7000 cables and discovered about 30 defects. It is our understanding that these inspections have been somewhat limited in their degree of detail. Many of the involved cable trays contain energized cables and inspections only covered those portions of cables that were immediately visible without any repositioning.

Other than the initial CRDM cable meggering, all inspection has been visual.

All of the damaged cables appear to involve cables for which the polyolefin sheathing was cut back to facilitate field splicing at the cable trays (e.g.

all of cne damage has been observed ac the field splice end, not the factory spliced end). The cables with the most severe insulation damage (e.g. the CRDM cables) are located outside of containment. SCE has performed laboratory evaluation of portions of these cables which appears to confirm salt water attack of the Kapton. The mechanism appears to involve salt water infusion 9611130508 961028 PDR FOIA WHITE 96-354 PDR

through what appear to be knife cuts in the Kapton/ Teflon coating. Once the salt water gets to the copper conductor, copper chloride and sodium hydroxide are produced. The sodium hydroxide has been shown to easily dissolve the Kapton tape. This mechanism seems to be confimed by the fact that none of the cut Kapton obse,ryed inside of_containmertt_appe.ars ts_i_nlolve Kapton ~

deterioratiog. j Although SCE has yet to confim their evaluatinn of the cause and correction i of this problem, it is our understanding that the leading theory involves accidental cutting of the Kapton during removal of the polyolefin sheathing.

SCE has agreed to perform a thorough visual inspection of every bare Kapton wire associated with all* the containment penetrations and provide the results to Region V.

This potential generic issue has been discussed with Messrs. Lanning and Merschoff. Region V is not aware of other' potential uses or users of the ,

subject coating material.and recommends that NRR evaluate whether other l nuclear applications of the coating material may be subject to similar  !

degradation or environmental conditions inducing degradation.

{

l YennisKirsch, F. M'h Director Division of Reactor Safety and Projects cc: E. L. Jordan, AE00 C. H. Berlinger, NRR G. W. Knighton, NRR R. F. Dudley, NRR D. Crutchfield, NRR W. Lanning, NRR ,3)p E. Merschoff, NRR g P. H. Johnson, RV P. Qualls, RV F. R. Huey, RV l

T

- l

{

t s I

• l

{

l

l j

i July 10, 1987 l ,

4 i

' I l

U. $. Nuclear Regulatory Comission I Document Control Desk l l Washington, D.C. 20555 - I 1

Subjects cceket No. 50-206 Im ormation LER t

j Licensee Event Report No.87-008 -

q San Onofre Nuclear Senerating tation Unit 1 1 l l

! The attached Licensee Event Report (LER) provides information regarding damage to i the insulatior, of conductors which connect field cabling to containment 1

' electrical per.etrations. Neither the health and safety of plant personnel nor the health anc safety of the public was affected by this event. ,

If you require any additional information, please so advise.

I l

j Sincerely, ,

3 i

i I

. M H. E. MORGAN y

STATION MANAGER RCDouglas j

Enclosure:

LtR No.87-008 cc F. R. Huey (USNRC Senior Resident Inspector Units 1, t and 3)

I J. 8. Martin (Regional Administrator, USNRC Region V) I Institute of Nuclear Power Operations (1NPO) j g

1 l

ncr n. us, numana soeum , o wanei.

6 Wai, -

maeven sus ,a ..

misussi as m LICENSEE EVENT REPORT (LER)

. ....... m .........i.i .......

SAN ON0FRE NUCLEAR _ GENERATING STATION, UNIT 1 . 0 15101010: 21016 I lofl 013

.. i.

KAPTON INSULATION DAMAGE ON CONTAINNENT PENETRATION CABLIS L. a m u .e n _rei ...on,oa,aen ovnem racius... savo6vse m

_r_vswt o.sva is l y ... ... i.u. .... g .J::.. 1,,::;.. ..... ... .... .u.. ..... .~ - - - l e I

o15l0lo10l tl 016 0 12 81 817 01018 010 0J7 110 871 _ 0 15101010i l l

. . . . . , . . ma as,oc a even rrrio sunsume tw meouino ms or se en ii cen.o . . -... ., ea. wi... .: im i

• • • • • • 5 so aoncei j .0..asi,3 _

... ,3,.ic a n, , r,,ico, w 20 4C5 aH381s) 50.36tc)(a l $0.n(entu.) FJ P4ts)

., , . o . . 010 f 0 n miimi, 20 *esmmimi -

a.umm so.ntenem wa.un n. i> { oP.p' .'laay g.g g .c

'$ w.nmmt.innai m

4  ?@yig' 4 dp, h-'

~ ~ '

se ent.ntm.) .o.nmismen u.n m mt.iein.i ,

SS! .b'l so 40Sf6.in.) 90. 73ta h 2)(lee) S0.Hte)( Hal INFORMAT10N 4.C3 pgg COuT Att PO. .Me. LS A (IS)

~

me. ... uL. .

. . . . . .y H. E. NORGAN, STATION MANAGER 711]4 Si 6181 -16l 2! 4 11 ca. e o.er una non amew so m An.une oseca.eso a me aspon ini u . .. ...... .........

]Q

. m... '..... ........, .::::..- ,,.:::::. gg~'

w, w ,.. .e 1 [ ] l lJl M5h l l l l l l l  % II t

I JlI 1_ j__J

$ MU$l.

% nre; L Il l fl 1 llY?f h ',. . .e

. u . u . . , . . . . . . . .. c . . . n . , -- ... l .. !

7 u r,g/g g,q n g '*c = Tho .'v".'J.*.Im?!

=^"

i i i-i 4 e... .n cu i. i . oo .= c.. . . . Tn,. . -.i... mi . ..ni. . . .. .... .a . .> a e s on June 2,1987 with Unit 1 in Mode 5, moqge-ing of Control Rod Drive Mechanism (CRDN) '

coil circuits revealec low insulation resustance in 22 circuits. The source of the low resistance was later found in Conax containment electrical penetration " pigtails" which c nnect the field cables to the penetration conductors, i

Laboratory analysis and subsequent evaluation of the damaged CRDN pigtails revealed that tho Kapton insulation appeared to have been cut or punctured through to the conductor.

The analysis further determined the insulation damage did not resu' t from electrical l causes. Subsequent visual inspection of those portions of approximately 8,000 p:netration pigtails which are not protected with a polyolefin ,jecitet, revealed that less thar. 2 per cent were damaged (including the CRDN cables). Osmaged pigtails in circuits which are safety related or which are important to safety have been repaired.

Investigation of the cause of the insulation damage indicates that most of the damage occurred during initial installation of the Coner penetrations in 1988/1386. Corrective action will be developed and implemented to prevent recurrence of insulation damage.

Th) Units 2 and 3 electrical penetrations are of a different design and thus are not sia11erly affected.

I y,c,, ... u. I

i s

. LICENSEE EVENT REPORT (LER) TEXT CONTivalATl0N  !'

i 5AM DNOFRE NUCLEAR GENERATIUR 5TATION potKET NUMBER LER NUMER PAGE i UNIT 1 05000206 87-008 00-- 2 0F 3 i I

i i On June 2,1987 with Unit 1 in Mode 5, meggering of Control Rod Drive Mechanism (CRDM) i roll circuits revealed low insulation resistance b 22 circuits. The source of the low ,

! resistance was later found in Conax containment electrical penetration ' pigtails",

i ranging in length from four to twenty feet, which connect field cables to the solid l copper penetration feed throuqhs. These pigtails ar. insulated with a double wrap of .

Kapton insulation approximate'y 9 mils thick. A leflon-based adhesive is used to bind I l the Kapton wrap and to provide a moisture barrier. As provided by the manufacturer, the i pigtails originating from each feed through are grouped together and jacketed with a i j loose fitting polyolefin s,eeye which protects the insulation from mechanical damage.

Visual examination of the CADN pigtails outside containment, which are in an outside l j marine air environment., identified points at which the insulation was damaged and the

! copper conductor displayed signs of corrosion. Laboratory analysis and subsequent i evaluation of two samples of damaged cable revealed that the insulation appeared to have i j been cut or punctured through to the conductor. j 1

! Subsequent visual insroction of those portions of approximately 5,000 penetration l j pigtails which are not protected with the polyolefin sleeve, revealed that less than 2 per cent were damaged (incluoing the CRON cables).

! The inspection further revealed a i

! pattern of physical danage (i.e., cuts, punctures, or abrasion) to the exposed portion I of the pigtail insulattor, shere the pigtails are connected to the field cables. None of i j the damaged pigtails 'iocated inside containment revealed evidence of conductor '

3 corrosion. I i

j .ne laboratory analysis, further determined that the failures did not result from i electrical causes. The postulated corrosion mechanism was intrusion of moisture and l Nacl from the marine air anvironment outside of containment, through a cut in the i

insulation. Reaction with the conductor formed CuC1 2 and Na0H. The NaOH, in a

! saturated water soluta.on, degraded the Kapton insulation around the cut. I 1

investigatton of the cause of the insulation damage indicates that most of the damage h occurred during initial installation of the Conax penetrations in IM5/1986. The balance of the damage appears to be from other causes such as dropping or placing gl; objects on the cables, and stepping on the cables during subsequent construction, O aaintenance and inspection activities.

j A sa ple of exposed polyelefin jacketing was also visually surveyed. The survey revealed three instances of jacketing damege extending through to the conductor insulation. The damage to the jat.keting appears to be random and unrelated to the Kapton insulation damage described above.

04: aged pigtails in circuits which are safety telated or which are important to safety I were repaired by cutting out the damaged section and reconnecting the field cable to the !l re2aining portion of the pigtail, or sy covering 'he damaged section with an approved heat shrink insulating tube. The remaining damaged pigtalls are part of non essential lighting and communication circuits, which will be repaired during the next refueling l1!!

i>

outage.

i l

f

• l J

! LICENSEE EVENT REPORT (LER) TEXT CONTINUATION i M NERATION STATION DOCKET NUMBER LER NLmer.R PAGE I

UNIT 1 -

05000206 87-008-00 3 0F 3 l l i

i C rrective action to prevent recurrence will include providing cable tray covers where

. appropriate, and administrative controls on activities which could potentially damage i l j conductor insulation. Additionally, appropriate procedures will be revised to address I protection and handling of Kapton insulated cabla. Implementation of corrective action i uill be completed by the end of the next refueling outage.

i These instances of installation damage to Kapton insulation are similar to the i installation damage.r/  ;

i i Unit I (Docket 50 450)inported January, ,by1985.

Gulf 5tates Utilities company for the River Bend Station ,

i I l The Units 2 and 3 electrical penetrations are of a different design and thus are not  ;

similarly affected.  !

.i

', l 1

ii i

l l

i a

I 1

I