ML20245A105

Safety Evaluation Accepting Util 890515 & 19 Submittals Re Results of Augmented IGSCC Insp & Flaw Evaluation of Feedwater Inlet Nozzle N4-A to Safe End Weld

ML20245A105
Person / Time
Site:	River Bend
Issue date:	06/15/1989
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20245A102	List: ML20245A101 ML20245A105
References
NUDOCS 8906210080
Download: ML20245A105 (2)

Text

.

• I ,, 'o,, UNITED STATES g 'c

,. g NUCLEAR REGULATORY COMMISSION r, .. WASHINGTON, D. C. 20555

% . . . . + )g SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATING TO A FLAW EVALUATION OF A FEEDWATER INLET

~

N0ZZLE (N4-A) TO SAFE END WELD GULF STATES UTILITIES (GSU)

~

RIVER BEND STATION, UNIT 1 DOCKET NO. 50-458

1.0 INTRODUCTION

The staff has reviewed the licensee's submittals dated May 15 and 19, 1989 regarding the results of the augumented intergranular stress corrosion cracking (IGSCC) inspection and their flaw evaluation to support the continued operation of River Bend Station, Unit 1, in its present configuration for one 18-month fuel cycle. During the P,iver Bend 1989 refueling citage, a circumferential flaw was found on feedwater inlet nozzle M4-A to safe end weld. GSU performed an assessment of crack growth rates and fracture mechanics to justify their decision of not repairing the flawed weld during this outa GSU will re-exarnine this flawed weld during the mid-cylce (cycle 3)ge. outage to ensure that the continued operaticn through ' fuel cycle 3 is acceptable.

2.0 EVALUATION 2.1 Inspection The licensee (GSU) reported that a circumferential flaw approximately six inches in length was found in a feedwater inlet nozzle N4-A to safe end weld. The flaw was located in the Inconel 182 butter on the safe end side, with a maximum depth of approximately 0.2 inch (18% through wall) and an average depth of 0.16 inch (14% through wall). l 1

GSU indicated that the results of ultrasonic examinations were not conclusive ,

in discriminating the indication as a IGSCC flaw because the reported branching characteristics displayed by the indication in EBASCO examination was not confirmed by an independent examination using GE's automated Smart UT system. Branching er faceting were typical characteristics of IGSCC indications. GSU also reviewed the original and enhanced version of the construction radiographs of this flawed weld. The radiographs did not show any rejectable indications in the area where UT indications were found. The other three feedwater nozzle to safe end welds were also I inspected and no relevant indications were reported. l

\

i g

P

1 l

2.2 Flaw Evaluation General Electric (GE) performed an assessment of crack growth rate and fracture mechanics for the licensee, assuming the UT indication was an active IGSCC crack. A crack growth rate of 5x10-5 in./hr, bounding all GE's laboratory and Crack Arrest Verification (CAV) test data for Inconel j 182, is used for the crack growth calculation. GE indicated that the crack 4 growth rates in their test data appear to reach a plateau value for applied 1 K (stress intensity) values exceeding 25 ksi*in.g . These tests were performed at a conductivity of about 0.47uS/cm, which was higher than the  ;

conductivity reported at River Bend, which was about 0.25 uS/cm for 90% of the time in 1988. Laboratory testing has shown that crack growth rate decreases at lower conductivity. Therefore, the bounding crack growth rate used for crack growth calculation in River Bend is conservative.

However, the staff has some reservation regarding GE's assumption of a plateau crack growth rate at high K, because GE's data base at high K is -q rather scanty and more. test data, especially at high K, are needed to verify the assumed plateau crack growth behavior.

GE indicated that the water chemistry (oxygen content and dissolved impurities) cod temperature of the flow in the annulus region near the feedwater nozzle to safe end weld are comparable to that in the recircu-lation line. Therefore, the test data for recirculation flow can be applied to calculate growth in the feedwater nozzle to safe end weld.

GE performed a limit load analysis with a structural margin of 3 to ,

determine the allowable flaw sizes as required in ASME Code Section XI, 1 IWB-3640. GE's flaw evaluation has shown that, even using the bounding crack growth rate and the maximum flaw depth, the final flaw size at the end of the next fuel cycle wbuld not exceed the Code allowable (75% '

through wall) in Table IWB-3641-1. Table IWB-3641-1 is for flaws in gas tungsten-arc (GTAW)andgasmetal-arc (GMAW)weldmentsinaustenitic Since Inconel 182 butter was deposited by shielded metal arc (SMAW) piping.

welding process, GE also performed a flaw evaluation based on. ASME Code,Section XI, Table IWB-3641-5. Table IWB-3641-5 is for flaws in SMAW and submerged arc (SAW) weldment, which possesses less toughness than that of GTAW and GMAW weldment. The result of GE's evaluation has shown that the final flaw size after operation of 7000 hours0.081 days <br />1.944 hours <br />0.0116 weeks <br />0.00266 months <br /> (mid-cy:le) would not exceed i the Code allowable in Table IWB-3641-5. i The staff concludes that GE's flaw evaluation is acceptable particularly in consideration of a mid-cycle inspection to verify the extent of crack '

growth in the flawed weld.  :

l

3.0 CONCLUSION

1 Based upon the staff's review of the licensee's submittals, the staff concludes that the flaw examination is acceptable and that River Bend i Station, Unit I can be safely operated for one 18-month fuel cycle in its-present configuration, provided that satisfactory results of the mid-cycle ultrasonic examination of the flawed weld are obtained.

Dated: June 15, 1989 Principal Contributor: W. H. Koo l

._ ____ _ a