ML20083N546

Forwards Relief Requests 95-03 Re Leaks in Nuclear SWS Piping

ML20083N546
Person / Time
Site:	Catawba
Issue date:	05/11/1995
From:	Rehn D DUKE POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML20083N549	List: ML20083N546 ML20083N714 ML20083N732 ML20083N754
References
NUDOCS 9505230147
Download: ML20083N546 (66)

Text

'

Ib.. 7.r - pr.: - .,

,,3 y ,.

k. g.hy;vy. ' ,:n, .w

,W

,4 l~ .

' ~

\

f T,.  ;  ; ' : Duke PuuttCornpany L D. L Ran . >

.  : CatawbaNuclearGenerationDepannent'

, , VicePnesident MG ' ' 4800ConcordRoad - (803)U14205 Onice _ .i

,  ; , - . York, SC29745 ; . (803)DI3426 Fax'  ;

a<q _ q. v -

cuggpogVER m ,

i

~ '

LMay' 11,1995 i

' U.S. Nuclear Regulatory Commission '

1 Attention: . Document Control Desk g Washington, D.C. 20555 i,.

Subject:

Catawba Nuclear Station, Units 1 and 2  :

L Docket Nos. 50-413 and 50-414 - ,

Request for ReliefNumber 95-03 Leaks in Nuclear Service Water System Piping si

' Gentlemen:  ;  !

t

' The purpose of this letter is to request, in response to NRC Generic Letter 90-05, relief j from ASME Boiler and Pressure Vessel Code Section XI requirements pursuant to. l 10CFR50.55a(g)(6)(i). Accordingly, please find attached Request for ReliefNumber 95- .

03. This relief request is being submitted as a result of pinhole leaks that have.been l detected in the Heat Affected Zone (HAZ) of certain stainless steel piping in the nuclear -

. service water (RN) system. ' The leaks have been determined to have resulted from

[

Microbiological Influenced Corrosion (MIC).  !

4 Duke Power Company is requesting that th'e NRC review and approve this relief request [

at your earliest convenience. Ifyou have any questions pertaining to this reliefrequest,  ;

please call L.J. Rudy at (803) 831-3084. -

j Very truly yours, .

4

< D.L. Rehn y :i

. LJR/s. i I

Attachment - 950523o147 950511 .

P PDR .ADOCK 05000413 I.

PDR i

/ t n~nm-Jf pp I

e .-

[4,'

Document Control Desk Page 2 May 11,1995 xc (with attachment):

S.D. Ebneter, Regional Administrator Region II . '

R.L Freudenberger, Senior Resident Inspector R.E. Martin, Senior Project Manager ONRR 4

i 1

1 1

i I

a i DUKE POWER COMPANY

. RELIEF REQUEST FROM ASME' CODE SECTION XI REQUIREMENTS

' REQUEST FOR RELIEF, NUMBER 95-03

' UNIT: Catawba Nuclear Station Unit-l'& 2 DATE: 5/11/95 t

l '. 0 ORIGINATOR

1.1 DESCRIPTION

OF FLAW Pinhole leaks in stainless steel nuclear service water'(RN) system piping Heat Affected Zone (HAZ).

1.2 IMPRACTICALITY OF CODE REPAIB ,

Repair of the leaks will result in increased RN system unavailability. In addition, the $celding process itself may have resulted in the leaks. Delaying the repair will allow it to be scheduled with pre-planned accivities, increasing safety system availability and allowing for analysis and L

changes to the welding process to prevent possible j reoccurrence.

1.3 DESCRIPTION

OF PROPOSED TEMPORARY REPAIR No temporary repair is proposed. The pinhole leaks will be allowed to continue until a code repair is implemented during the Unit 2 end-of-cycle 7 refueling outage.

Temporary repairs are not necessary at this time.

1.4 SAFETY SIGNIFICANCE: SYSTEM INTERACTION EVALUATION Attachment 1 of calculations:

Flooding reviewed Jet Spray reviewed Loss of Flow: insignificant 1.5 ROOT CAUSE INVESTIGATION Root Cause

Description:

Based on Catawba's investigation and inspection to date, the pitting is limited to the HAZ. The pit initiation sites are interphase of interdendritic transition regions. The presence of delta-ferrite at the austenite phase boundary increases the critical pitting potential. The water quality plays a major role in the corrosion initiation in this area..

Catawba is located on Lake Wylie, which contains high suspended solids. Water bacteria and/or welding increases the possibility of random Microbiological Influenced Corrosion (MIC) attack. Polishing the pipe surface prior to-welding, using N: in the shield gas, and. controlling the ferrite content in the weld decreases the possibility / extent of MIC attack. These are corrective actions suggested in literature searches conducted to date. The MIC is attaching to the pipe in the HAZ at Cata'cba. Inspections of non-welded stainless steel pipe shcaed no attachment or growth of MIC tubercles. The adhered MIC starts to grow in the area beneath where the tubercle covers the pipe. The tubercle entraps water underneath it, changing the chemistry in this area. The presence of ferrite, altered chemistry, and the depletion of 0 under the tubercle does not allow the passive layer to reform. This passive layer provides A:\RR9503. DOC

W ^

DUKE. POWER COMPANY RELIEF REQUEST FROM ASME CODE SECTION XI[ REQUIREMENTS REQUEST FOR RELIEF NUMBER 95-03.

the key corrosion resistance. Therefore, continuous attack to the pipe wall occurs. The attack, and anodic reaction,

. forms a large cavernous subsurface pit in the pipe wall.and' propagates through it.

MIC attack on the HA2'of the piping hasLbeen determined to be the cause of,this pitting, rather than some other

,- mechanism acting on the piping. An exhaustive study,1 using an in-situ test stand, was conducted at Catawba in 1989/1990 to determine the best alternative material to use in lieu of the RN' system A-106 Grade B piping. Both piping and welded coupons were included for a variety of materials. Stainless steel suffered no degradation and no evidence of-p microbiological attack was evidenced in the study. This is an unanticipated attack mechanism on the material which was not' identified in the study. No change in water chendstry or biology has yet been identifed since the study. Thus, the study is believed to have not accurately simulated the conditions which have resulted in the'recent failures.

1.6 AUGMENTED INSPECTION Leaks were first identified as three pinhole-sized leaks in stainless steel piping associated with the RN lube injection simplex strainers. The leaks are located within the heat affected zone of the welds of this 4-inch piping.

Subsequent visual examination of other stainless steel piping and butt and socket-welded connections during a RN system walkdown has identified a total of 75 leak sites in both butt-welded and socket-welded connections of this-stainless steel piping. The piping involved is that associated with the RN lube injection simplex strainers, the l- containment spray heat exchanger cooling water side process radiation monitors (EMFs),.and two diesel generator starting air aftercoolers. The leak sites are very small and result in leak rates on the order of several drops per ndnute maximum. The leaks are primarily located on the bottom of the piping, but are not limited to horizontal piping. Some brown spots on the outside of the piping are evident which are leakage sites that have since plugged themselves with corrosion or silting deposits. Examination of the internal '

surface of the piping at the failure sites continues to.show very low silting or corrosion deposit layers in the general piping area. U e failure sites typically have a mound of d material on the ?.nside diameter of the piping over the failure site. This has been confirmed to be related to MIC, _

based upon Duke Metallurgical Laboratory analysis of samples cut from the system.

• A 100% walkdown of all stainless steel piping within the RN system has been completed. A total of 75 leak sites were-identified in 4-inch butt-welded and various socket-welded piping.

Characterization of samples from the identified leak sites has been completed by Duke's Metallurgical Laboratory.

• Structural assessment calcul&tions per Generic Letter 90-05 for all leak sites have been completed and it was determined that all acceptance criteria were met. j l

l A:\RR9503. DOC l

l

g~v g.

1 DUKE. POWER COMPANY RELIEF REQUEST FROM ASME CODE 'SECTION XI REQUIREMENTS '

REQUEST FOR RELIEF NUMBER 95-03

• Repair of socket-welded piping for 2 EMF 45's ' leak locations '-

, has been completed. Overall,' 39 of the 75 leak sites have i

been repaired"to..date.

• It was determined that ultrasonic exaraination of the leak sites is not a. viable method for characterizing the sites.

continued Metallurgical Laboratory work is being performed to support characterization of flaw sites,as they are ,.

removed from the system. This will provide additional assurance that the structural' evaluation'is current and acceptable.

2.0 STRESS ANALYSIS UNIT.

2.1 DESIGN DETAILS ASME CLASS 3 PIPING, Duke' Class C' CNC 1206.00-02-0007 CNC 1206.00-02-0008 2.2 FLAW CHARACTERIZATION >

Refer to Attachment 2A of calculations.

2.3 PRELIMINARY FLAW EVALUATION

SUMMARY

None: Initial flaws were isolated, removed, and~ flaw sized.

2.4 END OF CYCLE FLAW EVALUATION

SUMMARY

Flaw size is not expected to exceed flaw limit. size in '

greater than one year. Repair will be completed no later than by the end of the Unit 2 end-of-cycle 7 refueling- -e outage which~is scheduled to begin October 6, 1995.

2.5 FLAW MONITORING A walkdown of the RN system stainless steel piping will be conducted at least once every six months.

2.6 ADDITIONAL COMMENTS None. ,

L 2.7 EXCEPTIONS TO GL 90-05/ DRAFT ASME CODE CASE Calculations were performed in accordance with GL f 90-05, through-wall flaw approach. ,

2.8 REFERENCES

/ INPUTS Attachment 1: CNC 1206.00-02-0007 Attachment 2: CNC 1206.00-02-0008 Attachment 3: Metallurgical Laboratory Report 1812 PIP 0-C95-0527 i

A:\RR9503. DOC  !

1