ML19257A321

Final Rept Re Potential Significant Deficiency Involving Improper Installation & Insp of Embedded Plate Assemblies. Caused by Deficiency in Const & Breakdown in QA Program. QA Program Upgraded & Embed Design Modified

ML19257A321
Person / Time
Site:	Vogtle
Issue date:	11/20/1979
From:	GEORGIA POWER CO.
To:
Shared Package
ML19257A320	List: ML19257A319 ML19257A321
References
NUDOCS 8001030734
Download: ML19257A321 (4)

Text

'

Page 1 of 4 GEORGIA POWER C0f1PANY ALVIN W. V0GTLE NUCLEAR PLANT - UNITS 1 & 2 DOCKET: 50-424 and 50-425 EMBED PLATE ASSEf1BLIES FINAL REPORT November 20, 1979 INITIAL REPORT TO NRC: Verbal report by telephone on November 21, 1978, to Bob Wright (NRC, Region II) by Mr. C. W. Hayes (GPC QA Department)

INTERIM REPORTS TO NRC: 1) Letter W. E. Ehrensperger to J.P. O'Reilly datac December 20, 1978

2) Letter W. E. Ehrensperger to J. P. O'Reilly dated February 22, 1979

3) Letter W. E. Ehrensperger to J. P. O'Reilly dated June 1, 1979

4) Letter Doug Dutton to J. P. O'Reilly dated July 31, 1979

5) Letter Doug Dutton to J. P. O' Reil'y dated October 8, 1979 Descriotion of Occurrence The following description was previously reported in the above referenced Interim Reports. A potential significant deficiency was observed at Plant Vogtle on November 21, 1978 involving the improper installation and inspection of embedded plate assenblies.

An embedded plate assembly consists of a steel plate of specified dimen-sions with hexagonal nuts (" plate nuts") welded to the embedded side, an inspec-tion hole drilled through the plate concentric with the " plate nut", a bolt treaded into and tack welded to the plate nut, and a washer and another nut ("end nut") threaded on to and tack welded to the opposite end of each bolt.

The following deficiencies were noted and previously reported in the interim reports of certain embed assemblies in the auxiliary and control buildings:

1) Some of the bolts for a given embed plate were not fully engaged with the plate nut.

2) Some of the bolts in a given assembly did not have washers.

1668 318 8001080 "7 g3'-l

' Page 2 of 4

3) Some of the bolts did have washers, but the washers were not tack welded to the bolt.

4) Some of the end nuts were not tack welded to the bolt. (Note:

design drawings show a tack weld between bolt and end nut).

5) Some of the bolts were n t tack welded to the plate nut.

6) Some of the bolts may have beenU bent in excess of the design drawings' specified angle of 30 .

7) Some of the bolts could have developed cracks due to excess bending.

8) The plate nut and end nut in each poured embed assembly are standard hexagonal nuts instead of heavy hexagonal nuts as identified on the design drawing.

9) The possibility of cracks in the fillet weld around plate nuts due to pressure applied to bend anchor bolts.

10) Some of the bolts were missing completely on assemblies not yet covered in concrete. (This deficiency was not previously reported in the interim reports).

Analysis of Safety Imolications This deficiency represents a deficiency found in construction which, were it to have remained uncorrected, possibly could have adversely affected the safety of operations of the nuclear power plant. The deficiency also reflected a significant breakdown in a portion of the Quality Assurance Program according to the requirements of 10CFR50, Appendix B.

The deficiency possibly could have resulted in inadequate support of piping, HVAC, or other systems in the auxiliary or control buildings, with poten-tial consequences on the safety of operation of the nuclear power plant through-out the expected lifetime of the plant.

The deficiency caused extensive evaluation, testing, and some repair to embed plates to establish their adequacy to perform their intended safety function.

The load carrying capacity of some embed plates was derated by Engineering and a few plates were abandoned entirely.

Corrective Action A detailed investigation was conducted on site to determine the "as built" condition of fabri::ated and installed embed plate assemblies. The results of this investigation was transmitted to Engineering and Licensing. Engineering per-formed an analysis of the embed assemblies data and recommended the necessary action to bring the embed assemblies in compliance with the design criteria, Based on Engineering's analysis and recommendation, the following action was taken by GPC:

1668 319

Page 3 of '4 4

1) Bolts with thread engagement between 40% and 80% were plug welded using the Engineering approved plug welding procedure.

2) Bolts with less than 40% thread engagement were abandoned.

3) Ultrasonic inspection on each embed plate assembly was conducted to determine:

a) whether or not the bolt was stra'ight; b) whether or not ghe bolt was bent greater than 20 , bolts bent greater than 20 was considered cracked and were abandoned; c) whether or not a bolt had been reincated using the relocation parameter of within 3" from locatiorr on plate as allowed on engineering drawinn; and d) whether or not there were cracks or voids in the fillet weld.

4) To determine the presence of the "end nut" and to verify compliance with design requirements, the following actions were taken:

a) Engineering developed procedures for pull testing of the embed plate assemblies. To assist in qualifying the procedure and gaining experience with the test equipment, Georgia Power Company poured a test block with embed plates containing the various postulated deficiencies including some plates with out any end nuts. These plates were pulled tested and the test data was sub-mitted to Engineering.

b) Engineering modified their procedure prior to testing actual installed embed assemblies to assure that sufficient load was applied to break the bond between the concrete and the bolt, therefore, assuring the presence of the end nuts. A random sample was taken and approximately seventy-five bolts were tested with no failures.

c) Probability analysis and test statistics indicate a high confi-dence le/el regarding the perforrance of the embedded plates for the service intended.

d) All embed plates with only four bolts could not be demonstrated as meeting the design requirenents based on the statistical probability used. All embed plates with only four bolts were pull tested to determine their adequacy.

5) A few embed plates were abandoned.

Based on the Engineering analysis no additional action was required for:

1) Bolts determined to have 80% or greater thread engagement in a standard nut.

2) Use of the standard hex nut instead of the heavy hext nut.

3) Omission of a washer.

1668 320

Page 4 of 4

4) Lack of tack welding of the bolt to the plate nut since the presence of the bolt could be verified through the inspection hole.

As reported in our interim letter of February 22, 1979, the continuous upgrading of our QA Program along with our continuing training program has signi-ficantly improved our field's QC surveillance. Additionally, the embed design has been modified, where practicable, to specify Nelson studs in lieu of the bolt and nut assembly. It is our opinion that adequate corre tive action has been taken.

Summary and Conclusions

1) This deficiency, were it to.have remained uncorrected, could possibly have affected adversely the safety of operations of the nuclear power plant at some time during the expected lifetime of the plant.

2) The deficiency represents a significant breakdown in a portion of the QA Program as outlined in 10CFR50, Appendix B.

3) The deficiency does not represent a significant deficiency in final design. -

4) The deficiency represents a significant deficiency in construction of a component.

5) Based on items 1, 2 & A, the Georgia Power Company concludes that the deficiency comes under the reportino requirements of 10CFR50.55(e).

Records of the foregoing activities are located at the Plant Vogtle con-struction site and are available for your review. Based on the results of our evaluation, Georgia Power Company's embed plates are in conoliance with design requirements and no additional testing is planneci for installed plates.

1668 321