ML17187A322

SER Accepting Flaws in Head Flange Weld & Recirculation Line Welds W/O Requiring Repair,Iaw ASME Code,Section XI

ML17187A322
Person / Time
Site:	Dresden
Issue date:	02/15/1996
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML17187A321	List: ML17187A320 ML17187A322
References
NUDOCS 9602220108
Download: ML17187A322 (6)

Text

UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20f566..4001 SAffJY'~EVALUAIION BY THE OFFICE OF NUCLEAR REACTOR REGULATIQN*

RELATED TO FLAW EVALUATIONS COMMQNWEALTH EPISON COMPANY PRESDEN NUCLEAR PQWER STATION. UNIT 2 POCKET NO. 50-237 1.0 INTRQQUCTIQN On October 16, 1995, CoA1110nwea 1th* Ed is on Company ( ComEd, the 11 ten see)*

submitted two reports (Reference 1) for NBC approval.

The reports contained' the licensee's evaluation of flaw indications in the head flange weld. and\\

recirculation line welds from ultrasonic (UT) examinations conducted dur*tng:;-.

the current D2R14 refueling outage at Dresden, Uni_t 2.

The examinations; went*

performed in accordance with the requirements of American Society of *._,*.. *

~;- *.

Mechanical Engineer$ (ASME) Boiler and Pressure Vessel Code (Code), Seetfon* !i-.. :

XI, 1989 Edition. The first report indicates* that one UT indication fr~- t~:::,*. ',.

head flange weld exceeds the allowable flaw size specified in IWB-3500* of.. the-

• ASME Code,Section XI, and requires flaw evaluat.ton using IWB-~610 and the ASHE Code, Appendix A.

The second report ind;cates that one UT indication frOll the elbow-to-pipe weld on the Loop A recirculation pump discharge and two UT indications frOll the valve-to-tee weld on the Loop B recirculation pump suction exceed the allowable flaw sizes specified in IWB-3500 and require flaw evaluation using* IWB-3640, the ASME Code, Appendix C, and NUREG-0313 (Reference 2)".

The following evaluates each of the reports.

2.0 8ACKGRQUNQ CHEAQ: FLANGE WELD>

In-accordance witb-tbe requirements of the ASME Code,Section XI, 1989 Edition, the licensee has provided the results of the current D2R14 refueling outage UT ex*inattens performed on the reactor pressure vessel (RPV) head-to-fl ange. we1~~~,\\!fl~~':;~~1!ttl1. Also provided there are the evaluation and dhposttt. *.*

ti~t.JMl**** flaw that exceeds the ASME acceptance criteria~

This. flt.W(

.*'. *:';Q\\er:ii~ as a subsurface flaw with length of 3.4 inches and

. depttt:_ of.' :

~:-~:.:._:f': Inell/ts 1 ocated O. 53 inch fr~ the outer surf ace of the flange.~)...., __.

natt* subJl.tt ta 1 is based on the UT examinations on 50 percent of tlw tae*?;fl~M9ttW&ld~:.; The remaining 50 percent of the weld has been ex111ined* and: no flltlf* indications outside the ASME Code,Section XI, acceptance cri tert a were detected-.

2.1 EVALUATION AND INDEPENDENT VERIFICATION Methods and acceptance criteria that are acceptable to the staff for evaluating flaw indications exceeding the allowable flaw size in IWB~3500 are described in IWB-3610 and Appendix A of Section XI of the ASME Code for this ENCLOSURE.

(

'9602220108 960215 *"

. j

. ~DR A~O~K,.O~OO~ij~7

• r- \\?- ** 1 -v -

• 1 l *

. l j

l

• l j

. ";.if;~~'..~*;~~. >

2 subsuriac.~: ~.;.-~)f,ln:*thtfrelited fatigue analysis, the crack growth* rate curves: usecf wai!e:: fl"Gll'**alr".environment curves in Figure A430~1 of Appendix A.*

Using the bounding hydrostatic pressure test load (Reference 3)f. the licensee arr heel at a tot a 1 stress intensity factor, Kl' of 35. 7 ks i (in) 12 for th 1s subsurface flaw.

The stresses considered are those due to: (1) pressure stress - 18.4 ksi, (2) bolt preload ~ 18.5 ksi, and (3) weld residual stress -

8 ksi.

In calculating the bending stress Qf 18.5 ksi, an attenuation factor of 0.73.had been used.

The overall procedure and the calculated values in this submittal are consistent with.those in General ElectriC Report, SASR 9fJ-77 (Reference 4), and are considered adequate by the staff. However, the effect due to the clad stress, which was conside'red in SASR 90-77, was not 11entioned in this submittal.

In low upper-shelf energy vessel material

.*l

'~j

.:j

....... f

• • • ~

evaluations conducted recently by the NRC as part of the effort of Genert~..

Letter (Gt) 92-01 (Reference 5) reviews, the staff always considered t...

~n-_.. ('~~~:;,_:~:;,;_>:~{

cl adding effect. Further, since the cladding effect 1s not negligibllt..

_.. Jt:;~F_:.*' ':*:.. T::~::-:* : ;

when CQllP&rld to= that due to the weld residual stress, the staff bel iel-., 'tj)ltt*:c?;.-: -'"?*-:*

. ~

this effect sheuld be included.

Io the review, the staff perf~raed

~;~.i~~;~-*-:'.i:*~\\

-*'.,:~; ** J Independent calculations and obtained a value of 32.83 ksl(ln) n for ~i;.>c;. :;*:;,)'* '

applied K1 using the same stre.ss values empl. oyed by the licensee.

The-c.j:~?*.. \\**~.,_~i :,_:-.: __ ~:~,,~*.:,~"\\'::*.* ;

discrepancy uy be frOll the *inor difference in defining the* flange gePll&~f; ~~ff-*.'.*:*;~:~~-

The submittal lacks that infonution and the staff used generic valuesfio#:;'a _:f!t.. ':*'f.~~\\.' *. "

• typical BWR vessel. Since the discrepancy 1s rather small, the staff did!not request a calculation frO. the licensee.

To mke the evaluation complete, the staff also estimated the applied ic1 due*

to c 1 adding effect.

Al though the applied K1 formula for the* subsurface crack 1s not available in the literature, its bounding *value can be estimated by using the corresponding formula for surface cracks.

Employing Taia's formula-

. (Reference 6), the staff found. the* applied_ K1 to be 5.66 ksi (in) 11 for a surface crack deptb o.f 0.52* inch.

For the* current case of a subsurface.crack, the applied K value..y. hA around one half of 5.66 ksi(in).112*

Considering the.additiona\\ 3. ksi (in) 17f' due to the cladding, the allowable flaw depth would beca. 0.68 hac:h, and the corresponding values for one and 10 fuel.

. cycles would be 0.675 and' 0.643.

2. z -- maum*

The ii~~~

.2~ review of the licensee's report entitled,

• Struct

~.

.Il*~: -.... J~ of Head Flange Weld at Dresden Nuc 1 ear Power Stati*~-~ :*:*-~ '"'. :f~._,..{djt.ei'llined that the evaluation methodology in the report is appropri:ltir-iiiid'.{*tfi&;:tt~iteria used are in accordance with the ASM£ Code.

The current detected flaw depth of 0.52 inch for the RPV head flange is*

smaller than the revised allowable flaw depths for one and 10 fuel _cycles by

.adequate margins.

Hence, the flaw 1s acceptable without repair. However, the area containing this flaw shall be reexamined during the next three inspection

. periods as. required by IWB-2420 of the ASME Code.

3.0 8ACKGRQUNQ CR£ACTOR RECIRCULATION PIPING WELPSl During the current refueling outage (02Rl4/1995), the licensee performed intergranular stress corrosion cracking (IGSCC) inspection in accordance with Gl 88-01.

Raytheon Inc. performed ultrasonic examination.s for the 1 icensee, with EPRl qyiJ:,tf.:1edCexaminers, using both manual and automated techniques.

Circumferenttak cr:"ack'".. ind1cat ions were found in the heat affected zones {HAZ) of two recirculatfon system piping welds {PDlA-014 and PS2-TEE/202-4B). These crack indications exhibited the characteristics of branching, which is typical

• of IGSCC.

The crack indications exceeded the acceptance criteria in IWB-3514 of ASME Code,Section XI.

The licensee performed flaw evaluation to support continued plant operation without repairing the subject flawed welds.* The results of the flaw evaluations were submitted to NRC for review and approval on October 16, 1995.

The licensee also stated in its submittal that the subject flawed.welds will be re-inspected in accordance with :the Technical Specification {TS) during the next refueling outage.

3.1 EVALUATION During the current refueling* outage D2Rl4, the licensee performed ultraso~.fc;;*.

examinations on a total of 148 IGSCC susceptible welds.

The inspected. wel:ds*-. *..

consisted of 8 Category A welds, 35 Category C welds, all 67 Category.Q~'~)~elds;{..... : )'..

all 37 Category E welds and 1 Category F weld.

The staff has determined~-tlfati.."-.... e the scope of the IGSCC examination performed during this outage met thtf:!:~;[:~~~~~j~ :. *.,

guidelines of GL 88-01.

',_.;.:;'f.:<; __ i,.~:-*

Weld PDlA-014 is an e.lbow to pipe weld on a 28-inch recirculation pum~~*~;{;~;~-~~:~~V discharge line {Loop A). This weld is* an IGSCC Category F *weld* which is required to be inspected every refueling outage.* A circumferential crack, located *on the elbow _side of this weld, *was originally discovered during* 1986 inspection.

The wall thickness of the elbow was about 1.35 inches. The licensee reported that the depth of the crack had increased from 0.15 inch to.

0.25 inch since the last inspection performed in *1993 {D2Rl3).

However, there was no change in crack length (one inch long).

• Weld PS2-TEE/202-4B* is a valve to tee weld on a 28-inch recirculation pump*

suction line (Loop B).

Two circumferential cracks about 28 inches apart were found on the tee side of the weld during this outage. The wall thickness of the tee was about 1. 2 inches.

One of the cracks was reported. to have a depth of 0.35 inch.(29 percent through-wall) and a length of 2.85 inches (about 3;3 percent of the:.~Jrc~fer:~nce), and for the other* crack, its depth was 0.33 inch; (2li~\\~J:'Pj~ti~~~l\\.~*ia9brwall) with a length of 1. 75 inches (about 2 percent of the;~ct~- *. i{'!~~~tf$~J;;.f!:is weld was an IGSCC Category D weld which was requtre~r,:t*

-~!t~~~~J~~very other refueling outage *. In accordance with the guidelimC

~\\88'.~0-~#-~Ms weld now becomes IGSCC Cat_egory F weld which is

.requ.tredtf nS'~ct:tii~-1.every refUe 1 i ng outage.

• ... ~1'.:.

~\\.. -~?i

*~i5:i!l./~lt(f\\..~~~ r::.9}::.,~r;~!.;:r.

The licensee performed crack growth calculations for fatigue and IGSCC to ensure the final crack size *at the end of next fuel cycle would not *exceed the ASHE Code allowable. The licensee used PC-CRACK program to calculate the crack growth.

The calculation of fatigue crack growth was based on the normal/upset design basis transient events. The thermal transient events of five startup and shutdown cooling, with one restart of an idle recirculation loop during each startup and $hutdown event, were postulated.for the evaluation period. The fatigue crack growth rates used in the cal~ulatiQns were based on Figure C-3210-1 of Appendix C in ASHE Code,Section XI.

The calculated fatigue crack growth resulting from these postulated thermal

• transient events was very small, less that 0.00003 inch in each weld.

~

-. '.1 I

.-~ :.._"--.~.'.:: --~~~**~~ ;~~~:~:*:*

• ~

• f

E;~~2-~

'. '. '* ~ *.. '

The licens~(atculated; the critical flaw sizes using the methodologies and the acceptance;,c~1ter.h*provided in IWB-3640 and Appendix C of ASME Code;Section XI.

The acceptance criteria were for flaws in materials fabricated by submerged arc welding (SAW). A safety factor of 2.77 was applied to the bounding load combination of pressure, weight, thermal and seismic loads, and a Z factor of 1.612 was used in the calculations. The nonnal\\upset condition was detenained to be the limiting condition. The critical flaw size for weld PD1A-D14 was calculated to be 0.735 inch in depth and 48 inches in length, and that for weld PS2-TEE/202-48, it was 0.716 inch in depth and 44 inches in length.

The licensee used PC-CRACK program.to calculate the IGSCC crack growth.

The through~wall distribution of the residual stresses and the crack growth rate equation as delineated in NUREG-0313 was used in the calculations. The

. applied loads consisted of design basis loads of normal operating p*re~-s~ec;:..

weight and thermal load*.

The results of the licensee's crack growth.. :~

calculations had shown that it would take at least 35000 hours and 4*J~ours:...

to reach the critical flaw sizes of welds PD1A-D14 and PS2-TEE/202-4B1:.\\f,,.~<~',.~

respectively. The allowed crack growth periods have a substantial margin: Q.V9r*

the operating hours in the next fuel cycle (approximately 13000 hours}::f:\\:'.' :I,;.*

Furthermore, for weld PDlA-014, the reported crack growth during the l~$.~~ fUJk.

cycle was only 0.1 inch in depth.

The sta,ff has reviewed the l 1censee-'~;,flaii~

evaluation and finds the results to be acceptable to support the safe operation of Dresden, Unit 2, for the next fue~ cycle without repairing the subject flawed welds.

The staff performed an independent bounding crack growth calculation for IGSCC in the subject welds using a conservative crack growth model of a 360° circ1111ferential surface flaw.

The influence function used in the staff's calculation was based on that rec0111Dended in NUREG-0313.

The results of the staff's calculation showed that the final crack size at the end of the next fuel cycle (13,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />) will not exceed the Code allowable.

3.2 *cONCLUSIQN Based on* a. "'-"~."~f... t.h:!tl 1censee' s submittal and the staff's independent crack -.. _ 1 ** ;. *~,*~~lt\\l~,the,staff concludes that' Dresden, Unit _2, can be operated;~,.. :...

.,,.f;~~zf"!;}j cycle (approximately 13,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />) without repa.trh*:,~..

'* '-~~*~~.i_ind PS2-TEE/202-48; beca.use the structural integrity of tM" s *

-:We4J.welds will be maintained *

.-... * *:. *::~~ *

• }~~~-l~~'.~~~~.~At~~~;~;t:~:~~**,

. * ~ *

• In accorctUee:: ifl,tfi":,:Tss:y the-subject fl awed we 1 ds are required to be re-inspected every refueling outage. Continued plant operation beyond the next fuel cycle should be supported by results of re-inspection.

.*.* **::t:-..

~..

• i - ~... '-'..4-...:::**

~o~~~~

-* *-~<.,

~

4.0

• BUERDfiU~~*.-.~<~.. :.

~-:~ * *-!

Letter, e; RYbak (C~d),' to USNRC Document Control Desk, *oresden Nuclear Power Station Unit 2 Recirculation System and Reactor Head Flaw Indication Evaluations,* October 16, 1995.

USNRC, NUREG-0313, Revision 2, *Technical Report on Material Selection and Processing Gu1de11nes for BWR Coolant Pressure Boundary P1p1ng,* January 1988.

GE Report SASR 90-33, ORF 137-0010, *structural Flaw EvaluatioR of Cracks 1n the Top Head at Quad C1t1es Nuclear Power Station *unit 2,* April 1990~

GE Report SASR 90-77, *structural.Fl aw Evaluation of Head Flange Weld and Upper Shelf Weld at Dresden Nuclear Power Station Unit 2,

• October 1990 *.. t

~-v*.., :;..:.i,* *-., "? :.,: -..,._.*

USNRC, Generic Letter 92-01, Revision 1, *Reactor Vessel Structural

'-~-~.,.::(' :*.:' *. ~ * --~-

i;;::;:: :;~ ";i~: L~!~; ~:~~::!.~~ 1m~.°f cracks Handbook,* Del Resea111~,f ;~~ ~

'J ~ ~

.e

• D. Farrar 2..

cycle without repa1r1ng the subject flawed welds.

In accordance with the Technical Specifications, the subject welds PDIA-Dl4 and PS2-TEE/202-4B are required t9 be re-inspected every refuelingoutage. Continued plant operation beyond the next fuel cycle shpuld be supported by the results.of *the piping re-inspection.

{)'"-' >:.:*:;

Di stri but ion:

~:,

Docket File

• PUBLIC PDIII-2 r/f' J. Roe (JWR)

R. Capra

,.... :C. Moore*

DOCUMENT To receive OFFICE NAME DATE.

J. Stang OGC, 015818 ACRS, T2E26 P. Hii and, RII I*

DR93862:LTR JSTANG

. 02.

/96

• c* = Copy without enclosures *e* = Copy with enclosures "N" = No copy

~ D:PDIII-2

~

RCAPRA 11.JY 02/15 /96 OFFICIAL RECORD COPY