ML19257A284

Final Deficiency Rept:On 791111,notified Region Re Use of Incorrect Weld Acceptance Criteria on Reactor Pressure Vessel Pedestal Welds.Radiographs of Welds re-evaluated W/Proper Aws Criteria

ML19257A284
Person / Time
Site:	Hartsville
Issue date:	12/06/1979
From:	TENNESSEE VALLEY AUTHORITY
To:
Shared Package
ML19257A283	List: ML19257A282 ML19257A284
References
NCR-HNP-A-073, NCR-HNP-A-73, NUDOCS 8001030555
Download: ML19257A284 (3)

Text

.

ENCLOSURE HARTSVILLE NUCLEAR PLANT - UNIT Al USE OF INCORRECT WELD ACCEPTANCE CRITERIA ON REACTOR PRESSURE VESSEL PEDESTAL WELDS 10CFR50.55(e) REPCRT NO.1 (FINAL)

NCR ENP-A-073 On November 11, 1979, TVA notified NRC-0IE Region II, Inspector R. W.

Wright, of a potentially reportable condition concerning use of incorrect veld acceptance criteria in evaluating radiographs of welds made on the Hartsville Nuclear Plant unit Al reactor pressure vessel (RPV) pedestal.

This condition and corrective actions were described in our letter from

, L. M. Mills to James P. O'Reilly on November 16, 1979, in response to your October 25, 1979, letter, RII: JPO-50-518 and 50-520.

This is the final report en the subject reportable condition.

Description of the teficiency All radiographs of RPV pedestal welds made before August 15, 1979, were evaluated to the acceptance criteria of AWS Dl.1 Structural Welding Code, paragraph 8.15, which refers to welds in new buildings. The radiographs should have been evaluated to the acceptance criteria of AWS Dl.1, paragraph 9.25, which rcfers to welds in new bridees, as required by C. F. Braun Specification 300-05. The acceptance criteria in AWS paragraph 9.25 is slightly more rigid than that of AWS paragraph 8.15.

Evaluation of the RPV welds to AWS Dl.1, paragraph 8.15, rather than paragraph 9.25 resulted in 34 of 629 radiographs with indications which are unacceptable under AWS D1.1, paragraph 9.25, but no radiograph shews an indicatien which is unacceptable under AWS Dl.1, paragraph 8.15.

The unacceptable radiographs under AWS D1.1, paragraph 9.25, fall in two types:

A. Thirty radiographs have aligned indications which do not meet the separation requirements of AWS Figure 9.25.2.1.

B. Four indications exceeding the 1/2-inch maximum allcwable length by AWS paragraph 9.25 but which are less than the 3/4-inch maximum length alloved by paragraph 8.15, 16673(f{f 8001030

$$5

~

~

The indications are tabulated by weld seam below:

Number and Type Indications Seam No. Type A Type B Sector Location 1 1 0 0-1 2 8 2 A-B, B-C, E-F, F-G*, G-H, H-1, M-N, N-0*, P-Q, W-X 3 1 0 19-20 6 2 1 A-B, Q-R*, V-W 7 4 0 A-B, L-M, P-Q, R-S 8 1 0 T-U -

9 3 0 14-15, 17-18, 18-19

~

10 2 0 C-D, D-E 12 1 0 13-14 14 2 0 7-8, 18-19 15 5 0 E-F, G-H, N-0, T-U, U-V 406 0 1 1-2*

Total 30 4

• Type B '

Cause of the Deficiency The deficiency occurred because QC inspectors evaluating welds until August 14, 1979, did so to Quality Control Instruction (QCI) N302, Revision 0, " Radiographic Inspection of AWS Welds," which referenced the new building section of AWS Dl.1 rather than the new bridge section of AWS D1.1 as required by the C. F. Braun specification.

Safety Implications TVA does not believe that the presence of the 34 indications previously listed will jeopardize the RPV pedestal structure during an earthquake or any other anticipated occurrence. We are reporting this condition because C. F. Braun has required repair of weld seam No. 2.

Corrective Actions All radiographs evaluated under AWS Dl.1, paragraph 8.15, have been reevaluated to AWS Dl.1, paragraph 9.25, with results as discussed in the " Description of Deficiency." All radiographs evaluated af ter Augus t 15, 1979, on the RPV pedestal were to the proper criteria of AWS D1.1, paragraph 9.25.

1667 3 d)15

The nonconformnnce report has been reviewed and evaluated by GE/C. F. Braun and the recommended disposition was to accept "as-is" except for seam 2 which was repaired to meet paragraph 9.25 criteria. Seam 2 was rejected due to seversa concentrations of defects (a total of 10 indications) along the length of this seam. Reasons for accepting the other sea =s are:

1. The defects are vidaly scattered in various sea =s with a maximum of five indications in any one seas.

2. The design was based on 42 ksi yield, whereas actual material yield strength varies between 52 and 62 ksi.

3. Brinnell hardness tests made on selected welds during our investigations of your concerns expressed in your October 25, 1979, letter indicate no unacceptable or drastic changes associated with the welding.

4. The AISC-AWS Codes made the extent, type, and application or acceptance criteria for NDE the respcnsibility and discretion of the engineer, CE/C. F. Braun in this case. The base requirement is 10 percent radiography to the new bridge section of Dl.1 with location selecticn by the engineer, whereas in this situation the entire 629 feet of butt welds were radiographically inspected.

Action Taken to Prevent Recurrence QCI N302, Revision 1, distributed on July 19, 1979, and implemented by the Welding Quality control Unit on August 15, 1979, changed the criteria to include evaluation of radiographs of RPV pedestal welds to AWS Dl.1, paragraph 9.25, as specified., Implementation of this revision should prevent recurrence of this problem.

1667 3<tg

_