ML19296C162

Interim Deficiency Rept Re T Weld Cracking.S&W Performing Metallurgical Analysis of Test Specimens & Graver Engineering Reviewing Voltage Variation & Bead Shape.Rework & Repair or Replacement Planned

ML19296C162
Person / Time
Site:	River Bend
Issue date:	11/12/1979
From:	Weigand J GULF STATES UTILITIES CO.
To:
References
RBG-6881, NUDOCS 8002250312
Download: ML19296C162 (3)

Text

1." EG GULF S7"ETE'S UTELETIES COMPANY c _ e .

est e..cs- e :, a s - - : 2

A ::: r - ? 636 6- :

Novmber 12, 1979 RBG - 6881 File G9.5 Mr. Karl V. Seyfrit, Director Nuclear Regulatory Comission, Region IV 611 Ryan plaza Drive Suite 1000 Arlington, Texas 76011

Dear Mr. Seyfrit:

River Bend Station - Units 1 and 2 Docket Nos. 50-458 and 50-459 On July 12, 1979 Gulf States Utilities subnitted a 30-day status report on the "T" neld 10CFR 50.55(e) reportable deficiency originally identified on June 12, 1979. That status report covered the safety concerns due to a potential defective weld as well as a description of the deficiency. At that date a conmitment uns made to provide an update report within 60 days. On Septenber 12, 1979 Gulf States Utilities sutmitted the required update report cannitting to a final status report on the cause and corrective action within 60 days. To-date the cause of the problen has not been completely confirmed and Gulf States continues to investigate the cause of the cracking through the efforts of both Stone & Webster Engineering Corporation and Graver Energy Systems.

The current status of the investigations involving both the Unit 1 and Unit 2 welds indicates that in both cases the majority of the cracking type deficiencies are located in the root area of the weld. Initial Unit 1 Ur data implied a problem in the interpass areas, however, nere recent examinations indicate the problem to be predominant to the root area although some interpass indications still exist. Unit 2 data does not show evidence of interpass crack-ing.

'1b date, the following UT examinations have been performed.

1) 17 Unit 1 assemblies using the 450 and 700 beam scans.

2) 7 Unit 1 assenblies using the 00 beam scan.

3) 4 Unit 2 assemblies partially UT examined using 450, 700 and 600 tandum bean scans.

8 0 0 2 25 0 3lp

Mr. Karl V. Seyfrit Novenber 12, 1979 Initial evaluations from Graver attribute the cause of the defects to unacceptable bead shape and penetration of the root pass area in the first subnerged are pass caused by major variations in the welding current and variation in electrode position. Two addi-i.ional independent investigations are being performed by Graver and Stone and Webster. These are: -

1) Stone and Webster is perfonning a metallurgical analysis of test specinens obtained flun corner junction assenblies.

2) Graver is perfonning nock-up testing to evaluate the current, voltage variation and bead shape.

Stone and Webster Investication Originally it uns planned to obtain test samples from Unit 1 assemblies. Since Unit 1 and Unit 2 assenblies were welded with the same procedures and appear to have similar deficiencies, it was de-cided to perfonn the metallurgical evaluation on test specinnns from the Unit 2 enbedments.

Stone and Webster has prepared and issued to Graver a course of action required to obtain appropriate test specimens Using the Unit 2 Ur data, sample locations have been identified and Graver has been instructed to cut same. These soecimens have been renoved and shipped to Stone and Webster Boston Laboratories for the metallurgical analysis. Concurrently Stone and Webster is verifying if real de-ficiencies exist in locations other than the root area in the Unit 1 assenblies located at the jobsite.

Graver Investigation Graver has subnitted an initial report on the cause and corrective action for the deficiencies based on neck-up testing. The initial testing showed that cracks were generated when preheating was elimina-ted and voltage variations were induced. Graver has been instructed to perfonn additional nuck-ups following the procedures under which the original welding was perfonned. Graver has recently subnitted a plan to produce four nock-ups in order to thoroughly evaluate the cause and corrective action.

Corrective Action To meet ASME Code acceptance criteria, all deficiencies will be removed and welds will be reworked to specification requirements.

This investigation and potential rework will include all other welds completed by the procedure and welding equignent which led to these existing deficiencies.

Mr. Karl V. Seyfrit Novenber 12, 1979 Tne repair or replacement of the welds will be accomplished at the site after the enbedments have been set in place. Detailed procedures will be developed and approved prior to the rework.

With the above investigations continuing over the next several weeks it is anticipated that a completed investigation and evaluation will permit Gulf States to submit the cause and corrective actions in a detailed, final report to be sutmitted to you on or before February 1, 1980.

Sincerely,

_ ?;, . -

AR J.G. Weigand General Manager -

Nuclear Projects JGW/h1G/ par cc: Nuclear Document Control Project Engineering Files (2)