ML17290A731

LER 93-029-00:on 930805,discovered Primary Coolant Steam Leak.Caused by Weld Defect.Weld Crack Repaired on 930806. Weld Record for Applicable Weld Also Reviewed & No Discrepancies Identified

ML17290A731
Person / Time
Site:	Columbia
Issue date:	11/05/1993
From:	Fies C WASHINGTON PUBLIC POWER SUPPLY SYSTEM
To:
Shared Package
ML17290A730	List: ML17290A729 ML17290A731
References
LER-93-029, LER-93-29, NUDOCS 9311100114
Download: ML17290A731 (5)

Text

LICENSEE EVEOREPORT (LER)

ACILITY NAHE (I) DCKET HUHB R ( ) PAGE (3)

Washin ton Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 I DF 4 ITLE (4)

STEAM LINE FLO% ELEMENT SENSING LINE PINHOLE LEAK EVENT DATE (5) LER NUNBER 6 REPORT DATE 7 OTHER FACILITIES INVOLVED 8 NONTH DAY YEAR SEQUENTIAL EV I 5 ION HDNTN DAY YEAR FACILITY NANES CKET NUMB R (S)

HUHBER UNBER 05 0 0 0 0 8 0 5 9 3 9 3 0 2 9 0 0 0 5 9 3 05 000 P ERAT ING HIS REPORT IS SUBMITTED PURSUANT TO THE REQUIREHEHTS OF 10 CFR E: (Check one or more of the following) (11)

DDE (9) 3 OWER LEVEL 0.402(b) 0.405(C) 0.73(a)(2)(iv) 77.71(b)

(10) 0.405(a)(1)(i) 0.36(c)(1) 0.73 a)(2)(v) 73.73(c) 0.405(a)(1)(ii) 0.36(c) 2) 0.73 a)(2)(vii) THER (Specify in Abstract 20.405(a)(1)(iii) X 50.73(a) 2)(i) 0.73 a)(2)(viii)(A) elow and in Text. NRC 20.405(a)(l)(iv) 5D.73(a) 2)(ii) 50.73 a)(2)(viii)(B) Form 366A) 0.405(a)(1)(v) 50.73(a)(2)( iii) 50.73(a)(2)(x)

LICENSEE CONTACT FOR THIS LER 12 AHE TELEPHONE HUHBER C. L. Fies, Licensing Engineer REA CODE 0 9 7 7 - 4 1 4 7 COHPLETE OHE LINE FOR EACH COHPOHEHT FAILURE DESCRIBED IH THIS REPORT (13)

CAUSE SYSTEH COHPOHEHT HANUFACTURER EPORTABLE CAUSE SYSTEH CONPOHEHT NAHUFACTURER REPORTABLE 0 HPRDS TO HPRDS SUPPLEHEHTAL REPORT EXPECTED (14) EXPECTED SUBHISSIOH HOHTH DAY YEAR ATE (15)

YES (If yes, coapiete EXPECTED SUBHISSIOH DATE) X HO TRACT eel On August 5, 1993, with the plant in Mode 3 (Hot Shutdown) a system engineer discovered a small steam leak of reactor coolant located upstream of the "A" Inboard Main Steam Isolation Valve inside the Primary Containment. The steam flow was from an unisolatable pinhole leak at a flow element sensing line weld.

Control Room personnel immediately initiated a plant cooldown from Mode 3 to Mode 4 (Cold Shutdown) to allow repair of the leak.

The root cause of the steam leak was a weld defect. A defect introduced into the root of the weld during installation served as the initiation point with subsequent crack propagation due to fatigue.

The weld crack was repaired on August 6, 1993.

This event posed no threat to the safety of the public or plant personnel.

9311100114 931105 PDR ADOCK 05000397 S PDR

LICENSEE EVENT REPORT 'R)

TEXT CONTINUATION ACIL1TY NANF (1) DOCKET NUMBER (2) LER NURSER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 3 029 i 00 2 OF '4 ITLE (4)

STEAMLINE FLOW ELEMENT SENSING LINE PINHOLE LEAK Pl t ni'n Power Level - 0%

Plant Mode - 3 (Hot Shutdown)

Even D ri tion On August 5, 1993, with the plant in Mode 3 (Hot Shutdown), a system engineer discovered a small unisolatable steam leak of primary coolant. The leak was discovered during ongoing work associated with recovery from a reactor scram (see LER 93-027). The leak was located in the Containment Drywell upstream of the "A" Main Steam Isolation Valve (MSIV), MS-V-22A. The steam flow was from an unisolatable pinhole leak emanating from the "A" Main Steam Line Flow Element, MS-FE-SA, sensing line weld.

Imrr,edi rr iv A in On August 5, 1993, at 0846 hours0.00979 days <br />0.235 hours <br />0.0014 weeks <br />3.21903e-4 months <br />, Control Room personnel initiated a plant cooldown from Mode 3 to Mode 4 (Cold Shutdown) to maintain compliance with Technical Specifications associated with PRESSURE BOUNDARY LEAKAGE in Modes 1, 2, or 3.

Further Ev 1 ti n R e nd rr iv A in Further Ev lu ti n On August 5, 1993, at approximately 0838 hours0.0097 days <br />0.233 hours <br />0.00139 weeks <br />3.18859e-4 months <br />, this event was reported to the NRC by telephone in accordance with 10CFR50.72(b)(2)(i). This event is also reportable under 10CFR50.73(a)(2)(i)(A), "The completion of any nuclear plant shutdown required by the plant's Technical Specifications." The WNP-2 Technical Specifications do not permit any reactor coolant pressure boundary leakage.

2. An Engineering review of the instrument line calculation was completed. Stresses were calculated to be well below the ASME Code allowables for all deadweight, thermal, and dynamic loading conditions.

3. The weld record for this weld was reviewed and no discrepancies were identified.

L LICENSEE EVENT REPORTR)

TEKl CONTINUATION ACILITY NAME (1) DOCKET NUMBER (2) LER NUMBER (8) AGE (3)

Washington Nuclear Plant - Unit 2 ear umbel'v. No.

0 5 0 0 0 3 9 7 3 029 00 3 OF 4 iTLE (4)

STEANLINE FLOW ELEMENT SENSING LINE PINHOLE LEAK

4. Materials and Welding personnel performed a failure analysis on the weld and determined that the cracking had initiated at an undetectable construction defect at the root of the weld. The propagation of the crack from the root was attributed to fatigue. WNP-2 has had fatigue failures of socket welds in the past. The stress concentrations in a socket weld are at the root of the weld and the toe of the weld. Ifan anomaly exists at the root of the weld; the cyclic loading, if high enough, will tend to propagate the defect. If no anomalies exist at the root of the weld, the cyclic loading, if high enough, will initiate cracking at the toe of the weld. In this case, the root defect, which was not detectable by the required surface examinations, had propagated by fatigue to the weld surface.

5. No intergranular stress corrosion cracking was identified at this weld joint.

~Retype The root cause of the steam leak was a weld defect. A defect'introduced into the root of the weld during installation acted as an initiation point for the fatigue failure.

F her rrective Ac ion The weld crack was repaired in accordance with Maintenance Work Request AP4900 and ASME Section XI Plan 2-0975 on August 6, 1993.

2. Engineering has an on going program for-identifying candidates for fatigue cracking on the small break LOCA boundaries, with the main emphasis on the primary coolant/containment pressure boundary. The program, however, focuses on high probability failure locations. Socket welded process piping similar to this failure have not historically been a problem area. Cantilevered socket welded vent, drain, and test connections continue to be replaced on a priority basis during annual outages.

f i nifi n The steam leak was very small and it was concluded the weld defect did not challenge plant safety in that it represented a leakage well within the ability to provide makeup of primary coolant inventory. In addition, the steam plume did not challenge safety-related equipment. Plant records documenting drywell floor drain leakage from August 2, 1993, to August 6, 1993, report zero leakage confirming the character of the leak.

Leak before break was demonstrated and if the crack had opened up during further plant operation the unidentifiable leak rate would have eventually increased identifying a problem within the containment.

LICENSEE EVENT REPORT QR)

TEXT CONTINUATlON ACILITY NANE (I) DOCKET NUMBER (2) LER NUNBER (8) AGE (3) eer umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 3 029 OO 4 OF 4

.ITLE .(4)

STEAMLINE FLOW ELEtIlENT SENSING LINE PINHOLE LEAK imilar event The Supply System has had other small bore fatigue failures associated with socket welded vent, drain and test connections which are a cantilever beam type design as reported in LERs90-028 and 91-030.. These, as mentioned above, are being addressed under an ongoing engineering program. There have been only two other instrumentation line failures inside containment, one failure mechanism was indeterminate and the other was due to intergranular stress corrosion. These two failures were not reportable as LERs because they were found during plant outages.

EII Informa i n

• f

/@~tern ~monent Main Steam Isolation Valve SB V Primary Containment BT Steam Line Flow Element, MS-FE-5A SB FE