ML19322A563

Final Deficiency Rept Re Defect in Insulation of One of 12 Conductors in Number 14 Awg Class IE Cable.Defective Cable Removed & Returned to Manufacturer.Test Rept Encl

ML19322A563
Person / Time
Site:	North Anna
Issue date:	11/13/1979
From:	Brown S VIRGINIA POWER (VIRGINIA ELECTRIC & POWER CO.)
To:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
References
NUDOCS 7911150288
Download: ML19322A563 (6)

Text

r p_. . c 0C4 e

4

.Mr. Jame[ P. O'Reilly,' Director Serial No. 888A Office of Inspection & Enforcement PSE&C/JMD:adw:mc U. S. Nuclear Regulatory Commission

-Region II Docket No. 50-339 101 Marietta Street, Suite 3100 Atlanta, Georgia ^30303

Dear Mr. O'Reilly:

. Pursuant to the provisions of 10CFR50.55(e) and 10CFR21, you were notified on October 24, 1979, and by letter (Serial No. 888) dated October

~29, 1979, concerning discovery of a defect in the insulation of one conduc-

. tor of a twelve (12) conduc, tor el4 AWG Class IE cable.

North Anna Unit I was installing a twelve (12) conductor #14 AWG, Fire-wall III Class IE cable furnished by Cerro Wire and Cable Company, a type cable used in many Class IE areas of both North Anna Units 1 and 2, and dur-ing the termination process discovered a small area of bare conductor on one of the twelve conductors af ter stripping the jacket back.

The. defective section of cable was removed and returned to vendor for determination-of cause and probability of any further defects in procured and installed cable.

The defect was determined by the vendor to be the result of a under-cured insulation repair made during the manufacturing of single conductors.

The undercured repair was severely distorted during subsequent curing of the overall neoprene jacket covering the twelve conductors.

Vepco returne'd the remaining three (3) reels of 12 conductor #14 AWG Firewall III cable on site to the vendor for inspection and evaluation of any defects. The cable was disassembled, inspected and tested for any de- ,

fects. One ' repair was located and shown to be good. All testing proved no defects existed in the returned cable.

The vendor further sampled and tested insulation repairs in his exist-ing cable inventory to statiscally show further defects do not exist. The

-attached test report. from.the cable vendor details his inspections, tests and conclusions.

00 M Wjo0R Bo/'l

. ss lll 7911150 2 8f b

Mr. James P. O'Reilly We sgree that the conductor defect discovered at North Anna was a random and highly improbable occurrence and not indicative of other defects existing in the installed cables.

This is our final report on tnis item. Should you require further infortation please advise.

Very truly yours, 3ea C. 3rcwn, Jr. ' '

Sanior Vi:a President Pcwer Station Engineering and Construction Attacncen:s cc:- Mr. 'lic:ce 5tello, Director Office of *nspec:1cn a Enfor:arant

.tir. darold R. Centon, Director Office of nuclear Roactor ? gulatten t

9

').

e

• m ,

\ y, , , ;j n.

? ..

...... i 5a j "**'t.

, (p;g h *3'I I N 14( )CK1 il01-4*I'( >H C( ).N I I *,\ N Y ,;

i .g g i 3 .

Ma www.ce a.rcticutumus4 tritteer rentn.mo itm ris m.:to '

} '.

October 26, 1979

}

\1 V

}

Mr. Arthur J. Fit:: patrick Stone 6 Webster Engineering Corp. '

245 Su:::er Street #~

h; Boston, Mass. 02107 5~

2.

F

Dear Mr. Fitzpatrick.:

h

, , e

'. p l- In acicordance with the' test plan subraitted to you, d$ted October t 17, 1979 we have perfomed testing and analysis outlined in the ;n'

.. attached report. h.

, .. E4

' .-1 Picase contact Jo R. March or ne if you have further questiens. Si

i. 4 c;^

- Very truly yours, f-U;

. Tile ROCKBESTOS CCMPNi'( ~

.E; .

[h j g /j -

r G. C. Littlehales Ei)

'[ Manager, Quality Assurance ff

.z.

s.- .

k'0' ebn EI

. .; att. ~'

..' 5' 3=!

. . .Y1

~

.. -cc: E.' S. Reed b-l

y. F. RI Postea.  !

7 J. R. Marth -

(( .

1, ;( , A. H. Lybeck

• i

,g. E. J. D'Aquaano ,.

A. J. Rausch [.3 .

W. J. Patterson I. '

4

+6 A ulust e nt tut w.w> Gew 5

O

pt

.O

/ TEST PIPORT SUBJECT Defective insulation repair Stone 6 Webster F.O. NA 3187 ,

12/C 14 Firewall III

' Vendor Rcel 3187-153 Stone & Webster Reel 3-692 f

Obicetive: To determine the degree of probability that Rockbestos Firewall III cabics cay contain further defective insulation repairs, sicilar to the exampic reported.-

Analysis

1. Defective Repair A defective nolded insulation repair was taken fro: one conductor of reci ,

No. 3-692,12/C 14 AWG Firewall III cable, at the North Anca site. This ,

repair was severey.' distorted and contained openings exposing the copper COuductor. -

Upon exanination, the repair was found to be. extrencly undercured, causing ,

it to becene distorted during subsequent curing of the overall neoprene f-cable jacket.

The repair had received a minical cold cure causing the pre-cured taped  !

configuration to flow and become homogeneous, but not to becore cured. <

j The absence of a significant degree of curing van confirmed by a nolvent - '

extraction te s: per ASTM D2765. The uncured, or undercured repair showed

~

extraction of essentially 100%, whereas the adjacent cured insulation Il showed extraction of 19.4% which is nor=al for properly cured XL?I. .

i.

2. Heat Distortion Testing Thirteen repairs were taken at random fre= reels of Firewall III aingle conductor control cables presently in the New Haven plant. We dacntified -I 150 reels containing repairs out of an approximate total inventory of 600 reels, and used these reels for our sa:pling.

l This sacpling, in accordance with MIL-STD-105D, indicates that vich no [.

rejections, statistically the lot .contains fever than 1.0% defectives.  :

Heat distortion testing was perforned on the thirteen specimens per IPCEA S-66-524.-except that preheating was conducted at 100'C which is equiv-alent to the. cable jacket curing temperature of 210 # F., and is lO*C. In , t '

excess of the tc perature rcting of the cabic.

The following results were obtained: '

(see page 2)

e ,

• * . i ,

, ,, - Test Re g Specimen No.  % Distortion (requirement 30% Etz.)  ;

1 13.0 2 18.0 l, 3 24.0 1-4 20.0 i.

5 15.0 i 6 17.0 i 18.0 t 7

8 24.0 h 9 26.0 p 10 20.0  ;-

11 19.0 f 12 17.0 l 25.0 13 d

- All of the above specicens scre adequately cured and wuld sufficient 1, ,

resist significant defor=ation during jacket curing.

3. Curing Cycle Tolerance f'

. T:

. Repairs were produced in Firewall III single conductor control cables }.

using cure tires of 1 minute. 5 minutes, and 10 minutes at 150 FSI stea 4 pressure. A heat distortion test was then performed on each repair per p.

IPCEA S-66-524 at 100'C. F

!E The following results were obtained: 3 E

Curo Tine Specicen No.  % Distortion (requircsent 30% Max.) 0 ti 10 minutes 1 17.1 h-2 20.0 g '

'S ninutes 3 48.0 k 4 43.0  !

I minute 5 56.0 0 6 60.0 i F

A curve. derived from these data establishes eight ninutes cure time as the point at which the maxicum permissibic defor .ation of 30% occurs. Therefore,

[

r there is a 20% rargin of safety built into the standard ten ninute cure cycle. i U-

Cures of less than 5 minutes show visual evidence of insufficient cure as h the insulation tape overlaps are still visible. ,

}M 3I

4. Coeparison of Properly and Irproperly Cured Repairs I Yrom the data in item 3 above, it can be deronstrated that a properly cured j '1 repair '(specimens 1 and 2) vill not defona during exposure to the 210*F. i' hest of jacket curing. On the other hand, an ieproperly cured repair (speci- $

cens 5 and .6) will deform excessively at ticse te=peratures. h.,

n

. i l

1 1

, c e

• ' n st Report 3. L t ".

5. Exa _ination of 12/C 14 AVC Firewall III Cable fron Forth Anna Site The following three reels of 12/C 14 71reva]1 III cahic were returned to Ncv Itaven for evaluation on October 19, 1979.

Reel No. Length 3187-150 2550' 3187-151 2550' 3167-153 2562' (2030' on rect when received)

Overall jackets and underlying binder tapes vero conpletely rcreved from these Icosths and they were visually inspected on an inch by inch basis for presence of insulation repairs and general condition of singic conductors.

One repair was located in the orange / black conductor of reel 31?.7-153. This repair showed no distortion whatsoever, and was conpletely servicc961c. It was cut out and subjected to a dielectric breakdevn test, shoving no break-down up to 22 }NAC which is five times greater than the I?CLi requirecent, at which point arcing occurred. There was no distortion, sticking, or other abnottality in any of the cenductors in the three lengths exa=ined. s t:

CO?O!D.'TS: ,f l=j Each length of single conductor in P.O. NA 3187 cables successfully passed AC j:'

voltage testing afect it=crsica in water for six hours.  !

This testing provides a safeguard against further proecssing of single conductors -

which are electrically unsound at that stage of production.

f In addition, each length of cocpleted cable ecccessfully met or execeded II'CL\ L tequirenents for AC voltage testing, insulation resistance, and conductor resistance .

testing. n p

CONCLQION:O i c!

Undercuring of the repair is established as the cause of the defective condition. Ij Ihis Icd to distortion during curing of the overall jacket, and physical darene to the repair which ray have been caused durin;; stripping of the jachat. Ihi:.;

dar.sge vould not have occurred if the repair had bcon properly cured. 6j Ihe undercuring was caused either by nn extreme reduction of the colded repair  ;

curing cycle, or by a rmrentary loss of cteam prer. cure which caused the curing ,

cold to fail to heat properly. Neither of these possible causea is an event i 1 vith a significant possibility of recurrence. We have no history of internittent , .

loss of steam other than during a cajor steam line breakdown. Steam pressure is ;l pre-set, and is not controllabic by operntors. In addition, repair operators are 4 l' l trained in correct operating procedurcs, including the duration of the curing cycle.  ;

l l

Therefore, af ter careful review of all pcrtinent facts, it is our opinion that this .'

l defect was a highly unusual occurrence, and not indicative of a condition to be found elsewhere in Rockbestos Firewall III cables.

,l

\'

darE#r?A..l.w C. G. I,i t t.l e:b a l e rt i

The Rockbeston Co: pany ".anager, Wality Ar.nurunce 285 Nicoll St. '

Ncv Paven, Ct. 06511  !

October 29, 1979