ML20090G965

Part 21 & Interim Deficiency Rept Re Broken Capstan Spring Tang in Pacific Scientific Shock Arrestors.Initially Reported on 840119.All Shock Arrestors Returned to Vendor for Insp.Defective Springs Replaced

ML20090G965
Person / Time
Site:	Vogtle
Issue date:	07/03/1984
From:	Foster D GEORGIA POWER CO.
To:	James O'Reilly NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
References
REF-PT21-84-372-000 GN-380, PT21-84-372, PT21-84-372-000, NUDOCS 8407260022
Download: ML20090G965 (8)

Text

_

Georg:a Power Company PRP a ne boro. Geo gia 30630 Telephone 404 Sc4 9961. Ext 3360 0FFICIAL COPY 404 724 8114. Ext 3360 h

D. O. Foster Vice President and Project p ja Power General Manager the soutNvnt p;ystem v

Vogtte Project July 3, 1984 United States Nuclear Regulatory Commission Office of Inspection and Enforcement File: X78G03-M54 Region II - Suite 3100 Log: GN-380 101 Marietta Street Atlanta, Georgia 30303

Reference:

Vogtle Electric Generating Plant-Units 1 and 2, 50-424, 50-425; Pacific-Scientific Shock Arrestors; also GN-354, dated 5/1/84.

Attention: Mr. James P. O'Reilly

- In previous correspondence on this subject, Georgia Power Company stated that the NRC would be informed of the results of Georgia Power Company's evaluation by July 20, 1984. Georgia Power Company has been able to complete its evaluation earlier than previously forecast. Our evaluation has concluded that the metallurgical condition that caused the spring cracking in the Pacific Scientific snubbers could have adversely affected the Vogtle Electric Generating Plant and is therefore reportable to the NRC under the reporting criteria of Part 10 CFR 50.55(e) and Part 10 CFR 21. Since Pacific-Scientific has met with and informed the NRC concerning this subject, Georgia Power Company is reporting this condi-tion under the reporting requirements of Part 10 CFR 50.55(e). A summary of our evaluation is attached for your information.

This response contains no proprietary information and may be placed in the NRC Public Document Room.

Your r y

. O. Foste REF/D0F/tdm xc: U. S. Nuclear Regulatory Commission Document Control Desk Washington, D. C. 20555 R. J. Kelly D. E. Dutton J. A. Bailey L. T. Gucwa R. E. Conway W. F. Sanders 0. Batum M. Malcom G. F. Haad R. H. Pinson G. Bockhold H. H. Gregory W. T. Nickerson P. D. Rice B. M. Guthrie J. T. Beckham D. N. MacLemore R. A. Thomas E. D. Groover J. L. Vota

$ 0 WptBM N $ ig l S

geW/ l

, EVALUATION FOR A SUBSTANTIAL SAFETY HAZARD EVALUATION FOR A SIGNIFICANT DEFICIENCY Pacific-Scientific Shock Arrestors j- Initial and Subsequent' Report:

On January 19, 1984, Mr. C. W. Hayes, Vogtle Quality Assurance Manager, reported a potential significant deficiency to Mr. John Rogge of the USNRC.

1 In our letter GN-354 dated May 1,1984, Georgia Power Company forecast the completion of this evaluation for July 20, 1984.

l Background Information:

. During testing of Pacific-Scientific (P-S) Company's PSA-1 shock arrestors at i

Union Electric's Callaway Station by Daniel International personnel, 4 of 7 snubbers tested revealed a broken capstan spring tang. Pacific-Scientific Company requested the return of the failed springs for an independent -

metallurgical examination. Failed components were returned and were furnished

! to Mettek, 1805 E. Carnegie, Santa Anna, CA 92105 for metallurgical and fractunt analysis. Additionally, springs from the identical lot as those that failed were traced to snubbers located at Kansas Gas and Electric Company, Wolf Creek Station. These were also returned to Pacific-Scientific for test-4 ing and evaluation.

As a result of these tests, some of the spring tangs exhibited micro-crack indications. A metallurgical report by Mettek Material Engineering Technology Laboratories indicated that the spring cracking occurred because of stresses i

induced during spring forming which caused hydrogen cracking during subsequent silver plating. The metallurgical report also explained that although the spring

! fracture face exhibited brittleness at the crack onset, the core of the spring 4

was ductile and spring fracture was simple ductile rupture. This accounted i for the ability of the springs to withstand the full load functional and dynamic 4

load testing conducted and also suggested that the useful life of the springs containing cracks were substantial.

It has been determined that the capstan springs used in the suspect PSA-1 and PSA-3 shock arrestors were manufactured by the backup supplier using a fabrication sequence which did not completely remove the stresses introduced

' into the naterial during cold forming of the tang bends. During subsequent silver plating of the springs, the combination of residual stress in the cold formed material and exposure to liberated hydrogen in the plating bath resulted in hydrogen embrittlement and material cracks. The fabrication sequence included surface nondestructive examination (NDE) after forming of the tangs but not after silver plating. Thus, the hydrogen embrittlement surface cracks were unde- l tected and the capstan springs were installed in the shock arrestors. The part

, numbers for the capstan springs are: 1 i

PSA-1 (1801613)

PSA-3 (1801614) t

I One hundred fifty-one (151) shock arrestors of models PSA-1 and PSA-3 had been received at the Vogtle site. As a result of the failures at the other plants, the units were returned to Pacific-Scientific (P-S) in January 1984.

All units were disassembled, the capstan springs were inspected, defective )

springs were replaced, and all units were reassembled. Those shock arrestors '

with new capstan springs were functionally tested and all units were returned to the site. Tables 1 and 2 identify the shock arrestors which were found to contain defective springs. These units were included in the final engineering evaluation.

Engineering Evaluation:

The design of the Pacific-Scientific shock arrestors is based on converting the linear motion imparted to the pipe by a seismic load to rotational motion within the shock arrestor mechanism. A key component of the design is a capstan spring. This spring transmits a seismic load to the shock arrestor mechanism by being compressed over a shaft. Spring compression occurs by application of a load to two tangs, one at each end of the spring. Snould one tang break, the capstan spring becomes inoperable and the shock arrestor mechanism no longer functions. At two nuclear power generating plants other than Vogtle, some mechanical shock arrestors experienced broken /

cracked tangs on the capstan springs.

Each shock arrestor is designed, fabricated, and shipped to the site in accordance with the requirements of the Code applicable to the pipe it supports, the project specification and the purchase order. During fabrication the individual shock arrestor components are inspected and after assembly each unit is functionally tested. Inspection and testing is performed to ensure conformance to the documents mentioned above.

Shock arrestors that contained defective capstan springs which had to be replaced (tables 1 and 2) were conservatively evaluated on the basis that the deficiency would result in a pipe failure at the pipe support location should a seismic event occur. The following analysis was performed to determine the impact on plant safety:

A. A facility response analysis was conducted to determine if deficient shock arrestors in systems required to place the plant in a safe shutdown condition or mitigate the consequences of an event could result in unacceptable system functional performance and adversely affect plant safety. The analysis conservatively assumed the preexistence of a defective shock arrestor in one train, rendering the train inoperable (due to shock arrestor failure and subsequent failure of the pipe),

concurrent with the most limiting single active failure following the onset of an event (transient or accident condition) which requires a response from that system.

If it was determined that failure of the shock arrestor could result in  :

unacceptable system functional performance, then the deficiency was determined i to be reportable and no further analysis was performed. When failure of )

4 l the shock arrestor did not result in unacceptable system performance, l l then further analyses were performed as indicated in items B, C and D l below, i B. A review for potential flooding was performed to determine whether the l l

l l

existing plant analysis enveloped the effects of piping spool failure at the location of each of the potentially deficientt l' shock arrestors.

C. All lines were reviewed for radioactive content and the potential

! for exceeding offsite exposure guidelines stated in 10 CFR 100 and exposure limits for control room operators in 10 CFR 50, Appendix A,

GDC 19.

D. The analysis included interaction of non-safety related piping with safety related equipment (seismic 2/1).

The results of the engineering evaluation. indicated that the deficient .

shock arrestors noted in Table 3 could have unacceptably compromised system functional performance and adversely affected plant safety had the deficient capstan springs gone undetected.

Conclusion:

l Sin'ce the defective Pacific-Scientific snubbers could have adversely affected the future safe operation of the unit had this defect not been discovered it has been concluded that a reportable condition does exist. The defect in the manufacturing of the spring tangs . represents a significant deficiency and substantial safety hazard. However, since Pacific-Scientific has informed i and met with the NRC concerning this subject, Georgia Power Company, based on

mgulatory guidance in NUREG-0302, Revision 1 and other letters, is reporting this event under the reporting requirements of Part 10 CFR 50.55(e).

Corrective Action:

All shock arrestors which were at the site when Pacific-Scientific notified Georgia Power Company of the potential capstan spring deficiency were returned to Pacific Scientific. Pacific-Scientific disassembled these units and inspected all capstan springs for deficiencies. Those capstan springs without NDE indications are acceptable for service for the design life of the component. Defective capstan springs were replaced and the units re-assembled. Shock arrestors with new capstan springs were functionally tested. All shock arrestors are presently in an acceptable condition.

Pacific-Scientific made revisions to the capstan spring fabrication sequence

, to ensure complete removal of all cold forming stresses in this material after

the tangs are formed; and to confirm, by NDE, that hydrogen embrittlement

, did not occur in the silver plating operation. The former will be accomplished by first forming the tangs followed by age hardening this precipitation hardening material,17-7 PH. The latter will be accomplished by performing surface NDE after pmcipitation hardening of the springs and, a second time, j after silver plating. l l

l i Evaluation of Breakdown of Quality Program ,

i A review of the quality program at Pacific-Scientific was conducted and it was concluded that there was not a breakdown in their quality assurance program.

J 540H-700-6021-2V 26272 '

340H-011-5031-lV 25272 120H-460-1021-lV 05272 900H-550-6021-2V 64272 810H-200-7041-lV 66352 '

400H-720-5021-lV 15052 800H-120-3021-lV 90742 630H-431-2021-lV 80742 940H-151-2021-lV 60742 820H-880-2021-lV 50742 100H-360-3021-lY 40742 620H-600-2021-lV 10742 710H-010-5021-lV 00742 610H-010-5021-lV 89642 030H-401-2021-lV 69642 410H-880-2021-lV 39642 050H-400-2021-lV 30552 320H-600-2021-2V 20552 300H-600-6021-2V 10552 810H-240-4021-lV 28352 120H-400-7041-lV 18352 740H-400-2021-lV 67352 520H-181-2021-lV 57352 810H-270-2021-lV 07352

.oN troppuS .oN laireS l epiP S-P decalpeR sgnirpS natspaC SR0TSERRA KCOHS 3 ASP

)S-P( CIFITNEICS-CIFICAP 2 ELBAT

. a.

TABLE 1 PACIFIC-SCIENTIFIC (P-5)

PSA 1 SECCK ARRESTORS Capstan Sprines Replaced -

P-S pipe Serial No. Succort No. .

20172 V1-1206-006-H020 20173 V2-1205-009-H009 20355 V2-1205-003-H009 21662 V2-1305-081-H002 21663 V2-1305-081-H002 21664 V1-1202-002-H010

~

21665 V2-1204-007-H005 21668 V2-1204-038-H003 21670 V1-1202-072-H005 21672 V1-1407-004-3023 21673 V1-1407-001-H020 21674 V1-1202-134-H003 21677 V2-1205-010-5012 21678 V2-1301-010-H021

• 21680 V1-1205-008-H041 21682 - V1-1208-003-H044 21683 - V1-1205-007-H0 62 21684 V2-1205-007-H046 21900 V1-1217-025-H003 21901 V1-1204-057-H021 22316 V2-1204-192-H002 I

i TABLE 3 Pacific-Scientific Reportable Defective PSA-1 & PSA-3

, Shock Arrestors

. Pipe P-S P-S

. Support No. Serial No. Model No.

}

System 1202 - Nuclear Service Cooling Water

}

Vl-1202-002-H010 21664 PSA-1 Vl-1202-004-H047 25376 PSA-3 Vl-1202-004-H050 25503 PSA-3 i Vl-1202-006-H026 24701 PSA-3 Vl-1202-072-H005 21670 PSA-1 Vl-1202-072-H018 25370 PSA-3 V!-1202-088-H014 24693 PSA-3 Vl-1202-088-H028 24705 PSA-3

! Vl-1202-104-H030 24696 PSA-3

. Vl-1202-134-H036 24708 PSA-3 I Vl-1202-151-H049 24706 PSA-3 Vl-1202-181-H025 25375 PSA-3

Vl-1202-184-H003 21674 PSA-1 l V?.-l202-006-H023 25502 PSA-3 System 1203 - Camponent Cooling Water VI-1203-021-H008 24709 PSA-3 Vl-1203-063-H001 24704 PSA-3 ,

System 1204 - Safety Injection i Vl-1204-042-H018 25382 PSA-3 Vl-1204-057-H021 21901 PSA-1 V2-1204-007-H005 ' 21665 PSA-1

, V2-1204-038-H003 21668 PSA-1 V2-1204-192-H002 22316 PSA-1 System 1205 - Residual Heat Removal VI-1205-007-H062 21683 PSA-1 V1-1205-008-H041 21680 PSA-1 ,

Vl-1250-010-H016 24698 PSA-3 Vl-1250-010-H017 24700 PSA-3 i

Vl-1205-027-H004 25051 PSA-3 i V2-1205-003-H009 20355 PSA-1 i V2-1205-007-H046 21684 PSA-1 V2-1205-009-H009 20173 PSA-1 V2-1205-010-H012 21677 PSA-1 I

System 1206 - Containment Spray

(

i Vl-1206-006-H020 20172 PSA-1 V2-1205-005-H009 27246 PSA-3 V2-1206-006-H003 25501 PSA-3 V2-1206-007-H045 27262 PSA-3

.. _ _ . _ .~ . . _ _ -._ ._ . . _ . _ _ . . _ _ _ , _ . _ . _ . . . _ . _ _ _ _ . _ . _ _ _ _ _ . _ _ . . . . _ .

Table 3 Pacific-Scientific Reportable. Defective PSA-1 and PSA-3 Shock Arrestors (Continued)

Pipe P-S P-S Support No. Serial No. Model No.

System 1208 - Chemical and Volume Control V1-1208-003-H044 21682 PSA-1

, System 1301 - Main Steam V2-1301-010-H021 21678 PSA-1 1

i I

a f

9