ML20214R360

Licensee Contractor and Vendor Inspection Status Report. Quarterly Report,July-September 1986.(White Book)

ML20214R360
Person / Time
Issue date:	11/30/1986
From:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE)
To:
References
NUREG-0040, NUREG-0040-V10-N03, NUREG-40, NUREG-40-V10-N3, NUDOCS 8612050441
Download: ML20214R360 (214)

Text

- - - - - - - - - - -

_\ ,

NUREG-0040 Vol.10, No. 3 LICENSEE CONTRACTOR AND VENDOR INSPECTION STATUS REPORT QUARTERLY REPORT JULY 1986 - SEPTEMBER 1986 UNITED STATES NUCLEAR REGULATORY COMMISSION

f. ,,,,

S8A n8Ai! "'"PDR 0040 R

. . -. . A

7 l

w- ., '1_ _

Available from Superintendent of Documents U.S. Govemment Printing Office Post Office Box 37082 Washington, D.C. 20013-7082 A year's subscription consists of 4 issues for this publication.

Single copies of this publication are available from National Technical Information Service, Springfield, VA 22161

./

l NUREG-0040 l

Vol.10, No. 3 i

LICENSEE CONTRACTOR AND VEND 0R INSPECTION STATUS REPORT i

QUARTERLY REPORT JULY 1986 - SEPTEMBER 1986

$a'te PuI!fshe$$deIIe*r"!*9 Division of Quality Assurance, Vendor, and Technical Training Center Programs Office of Inspection and Enforcement U.S. Nuclear Regulatory Commission Washington, D.C. 20555 s, i

we .

CONTENTS PAGE

1. Preface .............................................. iii

2. Repo r ti ng Fo rma t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

3. NRC Approval Status of QA Program Topical Report Revisions ........................... vii

4. Sample QA Topical Report Revision Approval Letter .................................... ix

5. Inspector Reports .................................... 1

6. Selected Information Noti ces . . . . . . . . . . . . . . . . . . . . . . . . . 161

7. Index ................................................ 207

8. Table of Vendor Inspection Reports Related to Reactor Plants .......................... 209 l

l l

1

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

La PREFACE A fundamental premise of the Nuclear Regulatory Comission's (NRC) nuclear facility licensing and inspection program is that licensees are responsible for the proper construction and safe operation of their nuclear power plants..

The total government-industry system for the inspection of nuclear facilities has been designed to provide for multiple levels of inspection and verification.

Licensees, contractors, and vendors each participate in a quality verification process in accordance with requirements prescribed by, or consistent with, NRC rules and regulations. The NRC inspects to determine whether its requirements are being met by a licensee and his contractors, while the great bulk of the inspection activity is performed by the industry within the framework of ongoing quality verification programs.

In implementing this multilayered approach, a licensee is responsible for developing a detailed quality assurance (QA) plan. This plan includes the QA programs of the licensee's contractors and vendors. The NRC reviews the licensee's and contractor's QA plans to_ determine that implementation of the proposed QA program would be satisfactory and responsive to NRC regulations.

In the case of the principal licensee contractors, such as nuclear steam supply system designers and architect engineering firms, the NRC encourages submittal of a description of corporate-wide QA programs for review and dcceptance by the NRC. Upon acceptance by NRC, described QA programs provide written bases for inspection on a generic basis, rather than with respect to specific consnitments made by a particular licensee. Once accepted by NRC, a corporate QA program of a licensee's contractor will be acceptable for all' license applications that incorporate the program by reference in a Safety Analysis Report (SAR). In such cases, a contractors's QA program will not be reviewed by the NRC as part of the licensing review process, provided that the incorporation in the SAR is without change or modification. However, new or revised regulations, Regulatory Guides, or Standard Review Plans affecting QA program controls may be applied by the NRC to previously accepted QA programs. The status of NRC review of 0A topical reports submitted by the principal contractors is shown in Table 1.

When design and construction activities were high, firms designing nuclear steam supply systems, architect engineering firms designing nuclear power plants, and certain selected major equipment vendors were inspected on a regular basis by NRC to ascertain through direct observation of selected activities whether these design firms and vendors were satisfactorily implementing the accepted QA program. However, with the substantial decline of new plant design activities, the inspection of QA program implementation has been deemphasized. Instead, the t!RC vendor inspection focus has been shifted to vendor activities associated with nuclear plant operation, maintenance, and modifications. Inspection emphasis in now placed on the quality of the vendor products including hardware fabrication, licensee-iii

' v;;ndor interfaces, environmental qualification of equipment, and equipment problems found during operation and corrective action. If nonconformances with NRC requirements and regulations are found, the inspected organization is requested to take appropriate corrective action and to institute preventive measures.to preclude recurrence. If generic implications are identified, NRC assures that affected licensees are expeditiously informed.

In the past, NRC issued confirming letters to the principal contractors to indicate that NRC inspections have confirmed satisfactory implementation of the' accepted QA programs. Licensees and applicants could, at their option, use the letters to fulfill their obligation under 10 CFR'50 Appendix B, Criterion VII, that requires them to perform initial source evaluation audits and subsequent periodic audits to verify 0A program implementation. However, based on-the above described change in nuclear plant design and construction activities, NRC will no longer issue confirming letters to principal contractors since future NRC vendor program inspections will focus on selected areas rather than addressing the implementation of their respective QA programs. Therefore, confirming letters that have already exceeded their three year effective period will not be renewed. Confirming letters issued less than three years ago will remain in effect until the stated effective period expires. Therefore, as the ccnfinning letters expire, licensees and applicants will no longer be allowed to take credit for the NRC acceptance of the implementation of a principal contractor's QA program. Licensees continue to be responsible for the conduct of initial source evaluation audits and subsequent periodic audits to verify QA program implementation. The NRC Division of Quality Assurance, Vendor and Technical Training Center Programs will continue to review revisions to principal contractor QA programs when submitted and, when approved, will list the latest approved revision number and date of the approval letter in i

Table 1 of the next edition ~of the White Book.

i The White Book will continue to be published and will contain copies of all l vandor inspections issued during the calendar quarter specified. The vendor inspection reports list the nuclear facilities to which the results are l applicable thereby informing licensees and vendors of potential problems. In

! addition,- the affected NRC Regional Offices are notified of any significant problem areas that may require special attention.

I i Th2 White Book contains information normally used to establish a " qualified suppliers" list; however, the information contained in this document is not

, adequate nor is it intended to stand by itself as a basis for qualification l of suppliers.

1 This issue of the White Book contains copies of 1&E Information Notices, ccncerning vendor issues, released in the first half of calendar year 1986.

l Each subsequent issue of the White Book will contain those vendor specific j I&E Information Notices released in the quarter covered.

Correspondence with contractors and vendors relative to the inspection data j contained in the White Book is placed in the USNRC Public Document Room, located in Washington, D.C.

iv

ORGAN!ZATION:3 COMPANY, DIVISION CITY, STATE REPORT INSPECTION INSPECTION NO.: Docket / Year / Sequence DATE: ON-SITE HOURS:

CORRESPONDENCE ADDRESS: Corporate Name Division ATTN: Hane/ Title ,

Address City, State Zip Code CPGANIZATIONAL CONTACT: Name/ Title TELEPHONE NUMBER: Telephone Number FUCLEAR INDUSTRY ACTIVITY: Description of type of components, equipmer.t. or cervices supplied.

ASSIGNED INSPECTOR:

Name/ Vendor Program Branch Section Date OTHER INSPECTOR (S): Name/ Vendor Program Franch Section APPROVED BY:

Name/ Chief - Section/ Vendor Program Branch Date l INSPECTIOF PASES AND SCOPE:

A. BASES: Pertain to the inspection criteria that are applicable to the activity being inspected; i.e.,10 CFR Part 21, Appendix B to 10 CFR Fart 50 and Safety Analysis Report or Topical Report commitments.

B. SCOPE: Summarizes the specific areas that were reviewed, ard/or identi-t fies plant systems, equipment or specific components that were inspected.

! For reactive (identified prcblem) inspections, the scope summarizes the j problen. that caused the inspection to be performed.

! PLANT SITE APFLICABILITY: List plant name and docket numbers of licensed facilities for which equipment, services, or records were examined during the inspection.

v

ORGANIZATION: ORGANIZATION CITY, STATE REPORT INSPECTION N0.: RESULTS: PAGE 2 of 2 A. VIOLATIONS: Shown here are any inspection results determined to be in violation of Federal Regulations (such as 10 CFR Part 21) that are applicable to the organization being inspected.

B. NONCONFORMANCES: Shown here are any inspection results determined to be in nonconformance with applicable commitments to NRC requirements.

In addition to identifying the applicable NRC requirements, the specific industry codes and standards, company QA manual sections, or operating procedures which are used to implement these commitments may be referenced.

C. UNRESOLVED ITEMS: Shown here are inspection results about which more information is required in order to determine whether they are acceptable items or whether a violation or nonconformance may exist. Such items will be resolved during subsequent inspections.

D. STATUS OF PREVIOUS INSPECTION FINDINGS: This section is used to identify the status of previously identified violations, items of nonconformance, and/or unresolved items until they are closed by appropriate action.

For all such items, and if closed, include a brief statement concerning action which closed the item. If this section is omitted, all previous inspection findings have been closed.

E. INSPECTION FINDINGS AND OTHER COMMENTS: This section is used to provide significant information concerning the inspection areas identified under

" Inspection Scope." Included are such items as mitigating circumstances concerning a violation or nonconformance, or statements concerning the limitations or depth of inspection (sample size, type of review performed and special circumstances or concerns identified for possible followup).

For reactive inspections, this section will be used to summarize the disposition or status of the condition of event which caused the inspection to be performed.

F. PERSONS CONTACTED: Typed, Name, Title

• present during exit meeting SAMPLE PAGE (EXPLANATIONOFFORMATANDTERMIN0 LOGY) i l

l vi

\

- -- - , ,-~ p- - . - - -

i TABLE 1 NRC APPROVAL STATUS OF QA PROGRAM TOPICAL REPORT REVISIONS DATE OF LATEST NRC PRINCIPAL TOPICAL REPORT REVISION REVISION APPROVAL LETTER Babcock & Wilcox BAW 10096A Revision 4 December 30, 1983 Bechtel BQ-TOP-1 Revision 3A August 28, 1984 Black & Veatch BVTR-1-D Revisicn 0A August 1, 1983 C. F. Braun 21A Amendnent #5 July 16, 1980 Brown & Root B&R-002A Revision 3 April 8, 1980 Burns & Roe B&R0E-COM-1-NP Revision 4A March 14, 1986 Combustion Engineering CENPD-210-A Revision 6 Ebasco Services, Inc. ETR-1001 Revision 12 May 4, 1984 Exxon Nuclear Company XN-NF-1A Revision 8 March 6,1986 Framatome FRA-QP/85 0782 NP Revision 2A March 14, 1986 General Atomic GA-A13010A Amendment #8 October 15, 1984 General Electric Co. NE00-11209-04A Revision 6 July 30, 1986 Gibbs & Hill, Inc. GIBSAR 17-A Amendment 8 February 27, 1985 Gilbert / Commonwealth GAI-TR-106 Revision 3 August 9, 1984 Ralph M. Parsuns P-TOP-QA1 Revision 3A August 26, 1985 Sargent 3 Lundy Engineers SL-TR-1A Revision 6 April 14, 1983 Stone & Webster SWSQAP 1-74A Revision E February 6, 1986 United Engineers &

Constructors UEC-TR-001 Revision 6 September 16, 1982 Westinghouse NTD VCAP-8370/7800 Rev. 10/6A August 29, 1984

• Accepted on 6/27/86 through the provisions of 10 CFR 50.54(a)(3)(iv) vii

1. pa aegjog UNITED STATES ,

g NUCLEAR REGULATORY COMMISSION

[ WASHINGTON, D. C,20655 5 l

(.....,/

SAMPLE QA TOPICAL REPORT REVISION APPROVAL LETTER (ADDRESSEE)

Gentlemen:

Your letter of transmitted Revision of your QA topical report ,

Quality Ae.urance Program." The report describes your QA program for design, I scurement, fabrication, and testing activities.

We have reviewed Revision of the report against the acceptance criteria in Section 17.1 of the NRC's Standard Review Plan for nuclear power plants (NUREG-0800, July 1981). Based on our review and evaluation of Revision ,

we find that the criteria in Appendix B to 10 CFR Part 50 are met. Revision of your QA topical report is, therefore, acceptable, and should be implemented for safety-related applications. Our evaluation is enclosed.

Should regulatory criteria or regulations change such that our conclusions about this topical report are invalidated, we will notify you. You will be given the opportunity to revise and resubmit it should you so desire. Ve note the commitment in the Foreward of the report to keep the NRC informed of changes to the QA program description.

Please incorporate this acceptance letter and its enclosed evaluation into the topical report, identify the report as Revision A (the "A" indicating that it has been found acceptable by the NRC), and transmit a copy to:

Document Control Desk ATTN: Vendor Program Branch U.S. Nuclear Regulatory Commission Washington, D.C. 20555 By way of letters to the same address, please keep us informed of the nuclear units to which this QA topical report applies. Should you have any questions regarding our review or if we can provide assistance, please contact us on (301)492- .

Sincerely, Chief Vendor Program Branch Division of Quality Assurance, Vendor and Technical Training Center Programs Of fice of Inspection and Enforcement ix

3- -. A L.< - Ja. . -a 1

1 i

1 INSPECTORS REPORTS I

i o

I I

i

[

I

ORGANIZATION: AUTOMATIC SWITCH COMPANY FLORHAM PARK, NEW JERSEY PEPORT INSPECTION INSPECTION N0.: 99900369/86-01 DATES: July 9-11, 1986 ON-SITE HOURS: 44 COPRESPONDENCE ADDRESS- Automatic Switch Company ATTN: Fans. R. E. Naumann President 50-56 Hanover Road Florham Park, New Jersey 07932 ORGAhlZATIONAL CONTACT: L. S. Olsen, OA Manager TELEPHONE NUf'.BER: (201) 966-2350 NUCLEAP INCUSTRY ACTIVITY: Manufacturer of solenoid valves and pressure switches.

/1 ASSIGt!FD IllSPECTOR: L Ui < ,[ t u <// t./ C 4 K. R. Naidu, Reactive Inspection Section (RIS) Date OTHER INSPECTOR (S): E. Yachiniak, RI APPROVED BY: 9 4 E. W. Herschoff, Chief RIS, Vendor Program Branch Date INSPECTION CASES AND SCOPE:

A. BASES: Appendix B to 10 CFR Part 50 and 10 CFR Part 21.

B. SCOPE: Obtain additional information on the kits used to refurbish l scram solenoid valves at the Vermont Yankee Nuclear Power Plant and selectively review the implementation of the quality assurance program.

l PLANT SITE APPLICAPILITY: All Eoiling Water Reactors.

1 i .-- - - - -.-..-- -

ORGANIZATION: - AUTOMATIC SWITCH COMPANY FLORHAM PARK, NEW JERSEY REPORT INSPECTION NO.: 99900369/86-01 RESULTS: PAGE 2 of 7 A. VIOLATIONS:

None.

B. NONCONFORMANCES:

None.

C. UNRESOLVED ITEMS:

None.

D. INSPECTION FINDINGS AND OTHER COMMENTS:

1. Background Information On June 14, 1986, Vermont Yankee Nuclear Power Plant reported that one control rod failed to scram and five others hesitated a few seconds before scramming during single rod scram time tests. During the outage that preceded the scram time tests, all of the scram sole-noid pilot valves (SSPV) had been rebuilt with replacement kits manu-factured by Automatic Switch Company (ASCO) and supplied by General Electric Company (GE).

Three types of problems were identified in the SSPVs which operate the i

control rods. In one SSPV, the core spring of the SSPV was separated from the core assembly due to improper assembly techniques. In one valve, on the exhaust side of the SSPV, the diaphram was installed

" backwards due to improper assembly techniques. In four SSPVs, an incorrect core assembly (provided in the kit) was installed.

Subsequent investigation determined that in one additional SSPV, the spring was separated from the core. Also, some solenoid base sub-assemblies were found to be out of round; this condition could prevent free travel of the core assembly.

2. Plant Tour

a. Core Assembly Production Area:

The core assembly is one of the components in the refurbishment parts kit (FV-204-139) that is assembled at ASCO's facility. The assembly consists of a core barrel, spring, needle valve, and various rubber and viton o-rings and disks.

l I

l 2 l

ORGANIZATION: AUTOMATIC SWITCH COMPANY FLORHAM PARK, NEW JERSEY REPORT INSPECTION N0.- 99900369/86-01 RESULTS: PAGE 3 of 7 All these pieces are receipt inspected to applicable MIL standards by QC personnel before being assembled. All core assemblies are currently 100% piece inspected by ASCO QC personnel.

b. Pre-Assembly Inspection Area:

The pre-assembly inspection area is where all nuclear spare parts kits are 100% piece inspected for their respective critical dimensions. PP-467, titled " Scheduling Nuclear Power (NP)

Customer Orders," is the governing QC inspection procedure controlling spare parts intended for installation in nuclear power plants. ASCO implements this procedure if the purchase order for them specifies that the kits are intended for refur-bishing safety-related SSPV in nuclear power plants. Cure date information for rubber 0-rings and disks is also reviewed at this area. Cure dates must conform to the applicable requirements specified on the customers purchase order.

c. Spare Parts Assembly Area:

The quality control inspector's responsibility in the kit assembly area is specified in PP-467. All parts are checked against a bill of material before being inserted into a kit.

After a large group of kits have been assembled (30-50), a single kit is then re-inspected for the correctness of all parts against the kit's bill of material. (All components are then sealed in blister packs.)

d. Solenoid Valve Test Bench Area:

One hundred percent of all solenoid valves manufactured at ASCO, whether nuclear or commercial, are tested for operability and

external leakage. Cycling tests are performed at normal and 85% of normal voltage at the minimum through maximum operating air pressure. A leak test (external) is parformed at the maxi-mum operating air pressure using a soapy bubble solution. Proce-dure TP-8316 and TP-NP8316 are the applicable instructions the test bench personnel use for comercial and nuclear valves, respectively.

i

3. Review of Follow-up Inspections on Kits l

l On June 26, 1986, subsequent to the SSPV malfunctions reported at Vermont Yankee on June 14, 1986, GE brought 200 replacement kits identified as ASCO type 204-139 to Florham Park. These 200 kits had l

1 3

ORGANIZATION: AUTOMATIC SWTTCH COMPANY FLORHAM PARK, NEW JERSEY REPORT INSPECTION NO.- 99900369/26-01 RESULTS: PAGE 4 of 7 been crdered from GE under Vermont Yankee P0 #28565, dated June 25.

1986. Specifics of this P0 required that GE insure that the vendor, ASCO, have an "NRC approved 0A program," that 10 CFR Part 21 was applicable, and that ANSI N45.2.2 level R packing and shipping requirements were imposed. OC personnel informed the NRC inspector that these 200 kits which were originally shipped in tlister pack kits had been unsealed. ASCO manufactured these kits but were unable to determine when they were shipped to GE. Each kit contains 11 individ-ual components such as "0" rings, springs, and screws, and two assem-blies. The two assemblies are the core assembly (ASCO part 65-716-2A) and the solenoid base sub-essembly (ASCO part 44-869-2). These two subassemblies require critical inspections. At GE's request, ASCO inspected the components in the 200 kits and doeurented their findings using "In Process and Final Inspection Requirements" per MP-1-046 to be witnessed or verified by the OC department. ASCO identified no adverse findings in 73 kits. The external diameter of the cores (part #65-717) in 127' kits were under the specified tolerance of 0.500 0.002 at the lip outside diameter. The solenoid base subassembifes (part #44-869-2} in two kits were ebserved to have a slight dent on the outside diameter of the tube halfway between the plug nut and the solenoid base. The scienoid bascs would not accept a 0.506 diameter plug. The unacceptable components were replaced and ASCO supplied 126 kits with 1007. acceptabic components.

One kit was lost.

4. GE Purchase Order Review The NRC inspectors reviewed P0s issued by GE to ASCO from 1984 tc the present only for spare parts kit FV-iO4-139 to better understand the procurement requirerents imposed upon ASCO hy CE.

a) Ihe earliest P0 reviewed was GE PO #334-Ala21 dated 2/14/84 This order required that 3000 spare part kits be produced implementing GE OC plan A-42, revision 4, and QC plan A-106, revision 0. Plan A-42 stated, in part, that the kits should be " identical replacements without change in design or materials." QC plan A-196 contained ouality assurance criteria similar in form to ANSI N45.2. The P0 also required the cure datc information to be identified on the package. ASCO treated this order as commercial because the P0 neither referenced 10 CFR 50 Appendix B nor specifically stated that the kits were safety-related. ASCO performs additional inspection: when the P0 specifies that the kits are intended for installation in nuclear power plants through the implementation of PP-467.

1 i 4

l ORGANIZATION: AUTOMATIC SWITCH COMPANY FLORHAM PARK, NEW JERSEY REPORT INSPECTION NO.: 99900369/86-01 RESULTS: PAGE 5 of 7 b) GE P0 #205-EF848, dated 7/8/85, was an order for 60 spare part ki ts. Thh order was different in that ASCO nuclear procedure PP-467 was implemented. This procedure requires that certain additional QC inspections be performed for nuclear orders. QC ER #96, " Pre-Assembly Inspection of Nuclear Valves, Pressure /

Temperature Switches and Switches", dated 10/4/78, states in detail the sequence of these QC inspections. ASCO implemented these two procedures because revision 3 of the GE P0, dated 11/16/85, changed the safety designation from "non-safety related" to " safety-related."

5. Review of ASCO Purchase Orders to Vendors The inspectors selectively reviewed ASCO's P0s to vendors for the supply of various components to ascertain whether ASCO specified appropriate quality requirements t!brein. Review of the following P0s indicate that ASCO specified appropriate quality requirements to their vendors.

a. PV 69828 dated July 5, 1984 to W. B. Gallagher, Kenilworth, New Jersey, for the supply of 150,000 Buna N "0" rings identi-fied as ASCO part #HV-22-525-5-70, revision EU. ASCO required the vendor to provide a Certificate of Conformance (C of C) and to mark the shipping tags and containers with cure date and compound number in accordance with MIL standard l'i23.

b. PV 71525 dated July 17, 1984 to AM Ludwig Corporation, Parsippany, New Jersey for the supply of 75,000 stainless steel core MXX, ASCO part #HV-65-717, revision AL. ASCO supplied the raw material to the vendor. The vendor was to certify that the ASCO material was used, stating the heat number of the material used, and to identify each container with the heat number.

c. PV 61048 dated February 8, 1984 to Seals Eastern, Inc., Red Bank, New Jersey, for the supply of 150,000 Buna N discs identified as ASCO part GV60-452-014 revision BTR. The P0 required the vendor to mark the shipping tags and containers with cure date and compound number in accordance with MIL ST0 1523.

d. ASCO P0 PV 55108 dated September 13, 1983, to LEE SPRING Company, Brooklyn, New York, for the supply of 50,000 stainless steel i

spring snap on ASCO part # FV-065-774. ASCO required the vendor to provide a C of C.

5 i

i f

ORGANIZATION: AUTOMATIC SWITCH COMPANY FLORHAM PARK, NEW JERSEY r

REPORT INSPECTION NO.: 99900369/86-01 RESULTS: PAGE 6 of 7

6. Review of GE Audits The inspector reviewed the following audits performed by General Electric Company (GE) and determined that GE did not identify any adverse findings.

November 30-31, 1985 GE, San Jose, California May 5, 1985 GE, Wilmington, North Carolina May 14, 1985 GE, Wilmington, North Carolina June 27-29, 1985 GE, San Jose, California

7. Review of ASCO Internal Audits The inspector reviewed six quality assurance audits performed by the ASCO Quality Contial (QC) staff relative to the Production Control Records and purchase orders. The audits were performed during tha periods: March 20, 1986, May 10, 1985, February 24-25, 1985, February 2,1983, January 24, 1983, and April 30, 1982. Records indicate that Production Control Department responded to adverse findings by taking appropriate corrective action which was subse-quently verified by QC. The quality assurance audits satisfy the QA manual requirement in paragraph 6.2.1.1 which requires that the QC department to audit Production Control Records periodically.

8. Results of the Inspection a.

The NRC issued Information Notice 86-78 to alert all licensees and construction permit holders of potential problems in refur-bishment kits (FV-206-139) for ASCO solenoids supplied by GE during 1984-1985, which may not have received 100% QC inspection at ASCO.

b. Two of the three problems which caused improper control rod movement at the Vermont Yankee Nuclear Power Plant were due to improper assembly techniques used during the refurbishment of the scram solenoid valves,

c. The inspectors were able to determine that GE supplied Vermont Yankee refurbishment kits which were part of a 3000 unit procure-ments under GE P0 #336-AL 421, dated February 14, 1986, which did not receive 100% QC inspection.

l 6

ORGANIZATION: AUTOMATIC SWITCH COMPANY FLORHAM PARK, NEW JERSEY REPORT lhSPECTION NO.: 99900369/86-01 RESULTS: PAGE 7 of 7

d. The QA menager informed the inspectors that ASCO developed a quality control ir Dection checklist with the concurrence of GE. This checklist, identified as GE Vendor Print File (VPF) 3061-83-1, is effective for all kits assembled after July 17, 1986. VPF 3061-83-1, specifies dimensional and visual inspec-tions and requires ASCO to perform 100% inspections on all components and subassemblies prior to sealing the kit.

E. EXIT MEETING: ,

The inspectors met with the President and the Assistant QC Manager at the inspection and discussed the scope and findings.

F. PERSONS CONTACTED:

H. R. E. Naumann, President L. S. Olsen, CC Manger A. Sperauskas, Assistant QA Manager E. Picciuti, Assistant Chief Inspector i

l 7

i . _ _ . _ . _ _ _ _ . - _ _ _ _ _ -__ _ - . _ _ _ _ . _ _ - - _ - _ - - _ - _ _ .

ORGANIZATION: BABC0CK & WILC0X LYNCHBURG, VIRGINIA REPORT INSPECTION [NSPECTION N0.: 99900400/86-01 DATES: 6/23-27/86 JN-SITE HOURS: 144 CORRESPONDENCE ADDRESS: Babcock & Wilcox, a. McDermott Company Nuclear Power Division ATTN: C. W. Pryor, Vice President and General Manager Post Office Box 1260 Lynchburg, Virginia 24506-0935 ORGANIZATIONAL CONTACT: T. Stevens, Manager, quality Assurance TELEPHONE NUMBER: (8041385-3138 NUCLEAR INDUSTRY ACTIVITY: Design and engineering services for B&W plants requesting reanalysis and modifications to existing systems, components and structures.

ASSIGNED INSPECTOR: Om e R. P. Correia, Special Projects Inspection

/o 2-4 Date Section (SPIS)

OTHERINSPECTOR(S): P. D. Milano, SPIS X. C. Leu, SPIS D. Ambrosek, Consultant APPROVED BY: <M ' cu e /6 /4 A7ohn W. Craig, Chief, SPIS4 Vendor Program Branch ae INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 21,10 CFR Part 50, Appendix B.

B. SCOPE: (1) Review the status of previous inspection findings, UTTnspect the design and engineering activities performed by B&W for plant modifications.

PLANT SITE APPLICABILITY: Arkansas 1(50-313), Belefonte 1 & 2 (50-438 &

439), Crystal River 3(50-302), Davis-Besse 1 (50-346), Oconee 1, 2, & 3 (50-269,270&287), Rancho Seco 1(50-312), Three Mile Island 1.(50-289),

Washington Nuclear 1(50-460).

9

ORGANIZATION: BABC0CK & WILC0X LiNCHBURG, VIRGINIA REPORT INSPECTION N0.: 99900400/86-01 RESULTS: PAGE 2 ot 10 A. VIOLATIONS:

None.

B. h0NCONFORMANCES:

1. Contrary to Criterion III of 10 CFR 50, Appendix B and B&W Administrative Manual Procedure NPG-0402-01,Section VI, A.7 and A.1, the reported results of reactor power level and DNB power level in calculation No. 32-1158579-00 for PSC 17-83 could not be independently verified. (86-01-01)

2. Contrary to Criterion XVI of 10 CFR 50, Appendix B and B&W Administrative Manual Procedure NPG-1717-02,Section VIII, 1, calculation No. 32-1158579-00 (PSC 17-83) contained recognized code deficiencies, (i.e., inadequate feedwater flow response with best estimate gains and code instabilities during transient evaluations) which were not reported and resolved as required by the procedure. (86-01-02)

3. Contrary to B&W Administrative Manual Procedure, NPG-0903-03, Appendix 1, an overcooling transient is not identified as an analysis performed by computer code Digital Power Train (DPT).

(86-01-03)

4. Contrary to Criterion III of 10 CFR 50, Appendix B and B&W Administrative Manual Procedure NPG-0402-01,Section VI, A.1, calculation No. 32-1163870-00, "Radcal Gamma Thermometer Cable Restraint Stresses" was not demonstrated to be technically accurate and ccmplete, and did not contain clear and concise results.

(86-01-04)

C. UNRESOLVED ITEMS:

None.

D. STATUS OF PREVIOUS INSPECTION FINDINGS:

1. (0 pen) Unresolved Items: PSC 17-83, Overcooling Events at Low Reactor Power ort No. 99900400/86-01 for A reviewSafety Potential of " Unresolved Concern (PSC)Items,"

17-8 Rep /" Overcooling Events at Low Reactor Power" was conducted. These steam generator secondary side transients could produce a relatively high peak power without a reactor trip when initiated from low power levels.

10

ORGANIZATION: BABCOCK & WILC0X LYNCHBURG, VIRGINIA REPORT INSPECTION NO.: 99900400/86-01 RESULTS: PAGE 3 of 10 The review was initiated with a package that contained the PSC close-out notice with the evaluation report and a draft letter to the effected utilities as attachments. The evaluation report stated analyses with computer code Digital Power Train (DPT) showed that core power level did not exceed 70% during the transient. The source of this statement was not referenced. A statement was made that departure from nucleate boiling (DNB) is not expected to occur if the core power level does not exceed 80% full power. The source of this statement was not referenced.

The file for PSC 17-83 was reviewed. This file contained a copy of the letters to the utilities and a letter from R. W. Moore to D. Mars which referenced calculation No. 32-1158579-00. This letter also contained a coment which noted a personal conversation with R. A.

Kochendarfer, Fuels Engineering, the conclusion, of which, was that overfeeding transients should not result in DNB if the core power level does not exceed 80% full power. This conclusion was not supported by adequate documentation or references and was therefore not independently verifiable.

The operating manual for DPT was reviewed. The manual did not include an overcooling transient in its list of system analysis and scoping studies for anticipated transients as required by B&W NPG-0903-03, Appendix 1.

Discussions were held with an engineering supervisor concerning the applicability of the single point kinetics method of analysis to this type of transient. The supervisor could not provide documented direct applicability of DPT to overcooling transients, but did provide as an example, topical reports for the TRAP 2 and CADDS computer codes which showed the single point kinetic method to be acceptable for an over-cooling event.

The NRC inspector reviewed calculation No. 32-1158579-00. The base model was identified, changes that nad been made were recorded, and steady state analysis with modifications to demonstrate stability were provided. Results for the transient cases were included as-well-as a discussion of results as compared to the simulator results.

The graphic results were difficult to verify since the Y-axis was identified only with an acronym.

The calculation's preliminary results had instabilities which were identified as computer code instabilities. A modification was made to the model and the results obtained provided the 70% full reactor r power limit when the transient ended. This was the value stated in l

11 l

ORGANIZATION: BABC0CK & WILC0X LYNCHBURG, VIRGINIA r

REPORT INSPECTION NO.: 99900400/86-01 RESULTS: PAGE 4 of 10 a report to the B&W NSSS utilities. This computer run also had severe instabilities identified on some of the plots. The analyzed transient was terminated by a reactor trip due to these instabilities.

These instabilities were due to the analysis reaching the point where feedwater was being drawn back through the aspirator ports of the steam generator and thus tripping the reactor.

Based on trends indicated on graphs of DPT outputs in the steam generator (SG) downcomer levels and reactor power level, it was not possible to independently conclude that 70% full power was a reasonable or conservative estimate of potential power level which would be reached before feedwater flow would be stopped by a control system as a result of water level reaching the aspirator port of B&W steam generators.

The technical justification for the reported results were discussed.

with the engineering supervisor. He consulted with several individuals and concluded that DPT code instabilities were caused by feedwater being drawn back through the aspirator ports of the steam generators.

There was no documented technical evaluation to justify the conclusions reached from the analysis performed and subsequently transmitted to the B&W utilities.

Ncnconformances 86-01-01, 02, and 03 were identified in this area of the inspection.

2. (Closed)Nonconformance(83-03): This nonconformance involved computer certification files for the CRAFT 2 codes that were reviewed by the supervisor of the originator.

A review of B&W procedure NPG-0403-11, " Technical Document Signatures",

i Rev. 13, dated April 1, 1986 indicated that the revised procedure includes a definition of the meaning and responsibilities related to signatures on technical documents. Unit managers are required to document their decisions that an independent review was properly independent, and to place their initials adjacent to the independent reviewer's signature on technical documents. This nonconformance is considered closed.

In discussions with B&W Quality Assurance (OA) personnel, instances were cited that the requirements in the revised procedure had not been l consistently met and there had been noncompliances with other technical j documents such as specifications, drawing, EIRs, CI/As, and calcula-l tions. B&W issued a memo dated May 14, 1986 to its business unit QA representatives requesting investigations as to the scope of the 12 1 - ._ .

ORGANIZATION: BABC0CK & WILC0X LYNCHBURG, VIRGINIA REPORT INSPECTION NO.: 99900400/86-01 RESULTS: PAGE 5 of 10 problem and needed corrective and preventive measures by June 13, 1986.

Based on the above, a review of the implementation of the B&W proce-dure NPG-403-11 may be the subject of a future inspection.

3. (Closed) Nonconformance (84-03): This nonconformance involved an uncertified computer code (the CORE code) that was used in a safety-related analysis. The analysis has been independently verified and the computer code has been independently certified.

The NRC inspector reviewed B&W QA procedure NPG-902-06, " Computer Program Development and Certification," Rev.12, dated May 1,1985.

The revised procedure establishes the requirements and responsibil-ities for developing a certified computer program that is used to perform safety related calculations. This nonconformance is considered cicsed.

4. (Closed) Nonconformance (84-03): This item involved the description of computer code limitations in computer program manuals.

B&W QA procedure NPG-903-03 " Development of Certified Computer Program User Manuals", Rev. 11 dated March 1, 1985 includes the identification and description of program limitations that affect the validity and/or accuracy of the programs. The subject title has been changed from " Development and Control of Computer Program Manual" to " Development of Certified Computer Program User Manual." This nonconformance is considered closed.

5. (Closed) Nonconformance (84-03): This nonconformance, involved the ASME code indices used in the T3 PIPE computer program for the calcu-lations related to butt welded fittings which use branch connections in the piping analyses and the proper verification of program file calculations.

" Resolution of T3 PIPE Errors" dated January 24, 1986, File No.

2A4/(T3 PIPE 18/2) was reviewed with respect to a general review of l the certification of "T3 PIPE" program and the associated corrective j actions. The NRC inspector observed that the "T3 PIPE" version has j incorporated ASTM code indices. In addition, B&W has reviewed all t 37 identified pipe analysis documents on the Historical Document i List (HDL), and Document Release Notices (DRN's) in the file in the Stress Analysis Unit (SAU), and concluded, that the errors involved have no safety-related effects. This nonconformance is considered closed.

13

ORGANIZATION: BABC0CK & WILC0X LYNCHBURG, VIRGINIA REPORT INSPECTION ,

N0.: 99900400/86-01 RESULTS: PAGE 6 of 10 E. Other Findinos and Comments

1. Oconee Pressurizer Internal Spray Piping Analyses The B&W analysis of the Oconee pressurizer internal spray piping which was performed as a result of the addition of Target Rock Valves and the modified cooldown transient were reviewed. B&W calculation No. 32-1163920-00, " Pressurizer Internal Spray Piping Analysis" dated June 3,1986, analyzed the internal pressurizer spray piping and a pine support for all applicable loads. The calculation indicated that the spray-head and internal pipe meet the stress criteria of the applicable design code. However, the U-bolt, bolting plate, angle hanger and attachment welds exceeded the allowable stress valves for the applicable design code.

B&W's supervising engineer indicated that the calculations contained preliminary results based on very conservative assumptions and B&W would look into the problem further with refined computer modeling boundary conditions and weld analysis. Duke Power Company, the Oconee licensee, had been notified of the problem and was to meet with B&W to discuss the approach to the problem.

B&W stated that they would notify the NRC of the final results as soon as the reanalysis was completed.

2. Limitorque Valve Actuator Weight Discrepancies TVA reported a 10 CFR Part 21 defect in December, 1984 which evolved

when an evaluation of valve and valve actuator seismic analyses for the pipe in the Bellefonte nuclear plant identified actuator weight i discrepancies between pipe stress analyses values and values contained in vendor drawings for Limitorque valve actuators. B&W identified l this potential defect as Potential Safety Concern (PSC) 4-85. At a

meeting with B&W all valve vendors involved were invited to discuss the problem and follow-up actions. Limitorque, Anchor-Darling, Copes-Vulcon and Rockwell manufacturers had supplied valves for Bellefonte. Additional meetings were held between B&W and each valve vendor to discuss specifics of the discrepancies. Further analyses

! showed that Copes-Vulcan valve weights were actually less than calcu-lated values and B&W, Limitorque, and Copes-Vulcan, stated that existing seismic analyses were adequate and no reanalyzes would be required.

l

{

14

l ORGANIZATION: BABC0CK & WILC0X LYNCHBURG, VIRGINIA REPORT INSPECTION NO.: 99900400/86-01 RESULTS: PAGE 7 of 10 PSC 4-85 was still open as of the date of the inspection. A letter to the B&W Owners Group (BWOG) was reviewed by the NRC inspector which indicated that no problem existed for B&W 205 Fuel Assembly (FA) plants: Bellefonte and WPPSS-1. Also, the letter indicated that the same problem may potentially exist for B&W 177 FA plants. However, since 177 FA plants had different requirements on valve qualifications, and different valve vendors than the 205 FA plants, B&W could not assess the impact of the valve actuator weight discrepancies fcr 177 FA plants. B&W stated their belief that licensees of 177 FA. plants should investigate the problem on their own. B&W made a random review of 177 FA plant records and found the same type of weight discrepancies No further evaluation or analyses has been performed by B&W. B&W stated that additional evaluation will not be performed unless requested by the utilities.

B&W personnel noted that while Regulatory Guide 1.48 recommends assur-ance of valve operability under seismic conditions for safety-related valves, 177 FA plants had already been constructed or were in the final stages of construction prior to the issuance of Regulatory Guide 1.48.

3. Radcal Gamma Thermometer - ANO-1 The NRC inspector reviewed calculation 32-1163870-00 "Radcal Gamma Thermometer Cable Restraint Stresses" and two associated drawings Nos. 1163473E-2, " Service Structure Tie Plate Assembly," and 11611344E-2, " Service Structure Tie Plate Details." A comparison of calculation data and results to drawing details was perforced to determine completeness, detail and accuracy of the modification. The following was found as a result of the review:

a. Calculation page numbers 4-6 contained a statement concerning deflections. There was no reference, basis or assumption for this statement, and therefore, it could not be independently verified,

b. The calculation did not include: a stress analysis for the horizontal clamp bars, swing bolts, gusset plates, tie plate, a reference to the weight of cable being restrained, the dead weight of the structure included in the analysis, or clamp bar i bolt torque values. Further, there were no references as to I

why vertical seismic accelerations which were eliminated from the analysis do not influence loadings on the restraint.

l l 15

ORGANIZATION: BABC0CK & WILC0X LYNCHBURG, VIRGINIA REPORT INSPECTION N0.: 99900400/86-01 RESULTS: PAGE 8 of 10 Nonconformance 86-01-04 was found in this area of the inspection.

4. ANO-1 Task Order 1007: Hot Leg Level Taps This task order was prepared in order to design instrument taps for a hot leg level measurement system. The B&W scope of work did not include the actual instrumentation for the system but was limited to the process point taps and two isolation / root values.

The joint configuration for the four penetration taps into each hot leg is to utilize a sleeve insert which will be rolled into each hole machined in the pipe. Prior to rolling, the sleeve would be seal welded at the pipe ID. The installation contractor is required to qualify the rolling procedure to accomplish this configuration and verifying that the sleeve could resist a 5000lb axial pull force. When asked about the basis for this requirement by the NRC inspector, with only a maximum allowed 5% wall thinning of the sleeve, B&W responded that the installation contractor had also expressed some reservation concerning this requirement. The procedure and its qualification test had not yet been conducted.

Additionally, the instrument process line will be inserted into the sleeve and welded which could reduce the rolling efficiency.

A preliminary seismic analysis had been performed which showed that if root valves weighino more than 8 pounds are utilized, the natural frequency for the installation may be less than 33 hertz. The proposed valves for the taps are 1" Kerotest valves which weigh 13 pounds. B&W had not yet completed the final seismic and stress analysis required by the contract. Thus, these records were not available for review.

5. AN0-1 Task Order 460: DHR Pump Seal Cooler Flow This task order provided the analysis to justify a reduction in service water flow for the decay heat removal pump seal cooler to 3 GPM at 130 F. The B&W analysis indicated that the seals would be operating at near process steam temperature due to the reduced flow. Contacts with the pump manufacturer had indicated that this would be allowable but would significantly reduce seal life. Thus, B&W recommended that the DHR pump seals be replaced every 2 years.

16

ORGANIZATION: BABC0CK & WILC0X LYNCHBURG, VIRGINIA REPORT INSPECTION N0.: 99900400/86-01 RESULTS: PAGE 9 of 10

6. Ah0-1 Task Order 944 - Instrument String Calculations This task order was developed to reanalyze the instrument string error for five instruments. In the process of reviewing this task order, one instrument string for High Pressure Injection /

Luw Pressure Injection initiation was evaluated for methodology and input adequacy. This calculation and its results were found to be acceptable.

The B&W reanalysis was found to utilize values for the various types of instrument errors which were different than the values provided by the manufacturer, Bailey Meter Co., on the specification sheets. B&W stated that the new values were determined in a study performed in this area and funded by the Tennessee Valley Authority (TVA). This study (EATF) had determined that, for the instruments tested, most

, had error tolerances lower than the manufacturer had detailed in the specification. The exceptions to this were found for the square root extractors listed on Table 7.1 of this study. Because of these greater inaccuracies, an instrument string utilizing one of the square root extractors was reviewed to verify that the large error values were being utilized.

In addition to the TVA report and manufacturer specification sheets, the setpoint analysis also utilized error values from a Bailey Safety Concern Report (SCR), SCR-001. This SCR dealt with inaccuracies caused by a gain potentiometer in several amplifliers.

While the data from this SCR was utilized in Task Order 944 and several other AN0-1 calculation packages that were reviewed, a similar process or an engineering review was not available to determine whether or not this deficiency would affect other licensed facilities. The B&W program and procedures for review and evaluation of vender-provided safety concerns may be the object of a future inspection.

7. AN0-1 Task Order 1023 - EFIC OTSG Shutdown Bypass Permissive This task order was developed to change the portion of the control circuitry for the Emcrgency Feedwater and Isolation Control system.

This change was necessitated to allow by-passing of the system during plant cooldown af ter a steam generator tube rupture. The original circuit required both steam generators to be less than 750 psig prior to going into by-pass. This prevented system actuation on low steam generator pressure at 600 psig and results in the steam generator with the ruptured tube being isolated and preferentially fed which is contrary to the required actions.

17

ORGANIZATION: BABC0CK & WILC0X LYNCHBURG, VIRGINIA REPORT INSPECTION fl0. : 99900400/86-01 RESULTS: PAGE 10 of 10 This design change was reviewed and found to be satisfactory. In this review, the AN0-1 Technical Specifications, FSAR, and the EFW System Design Description (SDD) were utilized. The EFW minimum flow rate of 500 GPM specified in the Technical Specifications is based on the transient analysis for a loss of main feedwater (LOMF) event.

The transient analysis assumes that the non-safety anticipatory trip is not present. In this case the pressurizer can go water solid.

In the SDD, however, the specified minimum EFW flow rate is 700 gpm per steam generator.

While this higher value is based on the necessity to keep the pres-surizer from going solid in the LOMF event, the two values could lead to confusion when compared to the FSAR. B&W stated that an effort to ensure design data is correctly stated and agreement between documents is underway. However, the SDD for the EFW system had an approval date of June 1986.

8. Recommendation Tracking System Report During discussions with B&W cn providing service information to clients ,

a copy of a newly prepared Recommendation Tracking System Report, dated May 1986, Document I.D. 47-1163743-00, was provided for review.

l The report provides each of B&W owners with recommendations from the Owner's Group along with a status of the responses from each utility.

While briefly reviewing this document, one recommendation, Number TR-027-ADM, was noted which discussed a problem with overconservative gain settings in the Power / Flow Imbalance trip circuit. The gain setting problem had resulted in several reactor trips in 1983. This recommendation also stated that it was an item that had resulted from misadjustment during the alignment procedures.

Since this item had been first recognized in 1983, the NRC inspector inquired whether the information had been provided by an alternate means to the utilities to ensure proper performance by the maintenance personnel. No guidance was provided other than this recommendation and the affect the utility's trip assessment provided to the owner's group af ter each trip.

i l

18

ORGANIZATION: THE BOC GROUP, INCORPORATED MURRAY HILL, NEW PROVIDENCE, NEW JERSEY REPORT INSPECTION INSPECTION N0.: 99901063/86-01 DATES: 09/08-09/86 ON-SITE HOURS: 13 CORRESPONDENCE ADDRESS: The BOC Group, Incorporated ATTN: Mr. Frederick K. Kies Director, Technical Resources 100 Mountain Avenue Murray Hill, New Providence, New Jersey 07974 ORGANIZATIONAL CONTACT: Mr. Frederick K. Kies TELEPHONE NUMBER: (201) 464-8100 NUCLEAR INDUSTRY ACTIVITY: None.

ASSIGNED INSPECTOR: -  %, /d-T f4 J. way, Reactive In ection Section (RIS) Date APPROVED BY: ,

bh, h T 84 g Merschoff, Chi f, RIS, Vendor Program Branch Date INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 50, Appendix B.

B. SCOPE: The purpose of this inspection was to review records pertaining to E-Brite 26-1 metal manufactured by Airco Vacuum Metals (currently The i B0C Group) to determine if any of the material was shipped to nuclear power plants in the 1970s as was stated in an allegation.

PLANT SITE APPLICABILITY: Not identified.

19

ORGANIZATION: THE B0C GROUP, INCORPORATED MURRAY HILL, NEW PROVIDENCE, NEW JERSEY REPORT INSPECTION l NO.: 99901063/86-01 RESULTS: PAGE 2 of 4 '

A. VIOLATIONS:

None.

B. NONCONFORMANCES:

None.

C. UNRESOLVED ITEMS:

None.

D. OTHER FINDINGS AND COMMENTS:

1. Allegation In June 1986, the NRC RIV office was notified of an allegation contained in a June 11, 1986 letter to President Reagan in which the alleger expressed knowledge that defective material was used in nuclear power plants in the United States. In July 1986, the alleger delivered to the NRC RIV office a " Partial Data File Regarding Defective E-Brite 26-1 Materials." The file which consisted of copies of test reports generated by or for Air Reduction Company (AIRCO) was reviewed by the NRC inspector from the Vendor Program Branch (VPB). The reports covered a time period from 1970 through 1976 when AIRC0 Vacuum Metals (AVM), a division of AIRC0, was developing the alloy designated "E-Brite 26-1."

In August 1986, the NRC inspector met with the alleger to gather information pertaining to the allegation involving E-Brite 26-1 material produced by AVM at their Berkely, California facility. The I allegation is that E-Brite 26-1 is a defective material because it

! cannot be welded, and its ductile to brittle transition temperature cannot be controlled thus making its use in nuclear reactors extremely hazardous. The alleger believes ~that E-Brite 26-1 material is in a number of domestic nuclear power plants.

On August 26, 1986, the NRC inspector contacted Mr. Frederick Kies,

! the Director of Technical Resources for The ROC Group located in Murray Hill, New Providence, New Jersey. Mr Kies, who was involved with the development of E-Brite 26-1 at AVM, stated that The B0C Group purchased AIRC0 in 1978. In the 1970s, AVM developed a high purity ferritic stainless steel produced by the electron beam refining process and received a patent on the alloy which was called E-Brite l 26-1. At their Berkley, California facility, AVM melted and cast i

20

ORGANIZATION: THE B0C GROUP, INCORPORATED MURRAY HILL, NEW PROVIDENCE, NEW JERSEY REPORT INSPECTION N0.: 99501063/86-01 RESULTS: PAGE 3 of 4 ingots, and the finished products (e.g., strip, tubing, forgings, etc.) were fabricatea by vendors. In 1978, AVM went out of business and sold the patent rights for E-Brite 26-1 to Allegheny Ludlem.

The records pertaining to the production of E-Brite 26-1 by AVM have been retained at The BCC Group's facility in Murray Hill, New Providence, New Jersey.

At The BOC Group's facility in New Jersey, the NRC inspectcr reviend 6 Monthiy Sales Reports (MSR) and customer lists relating.to E-Brite 26-1 for 1976. The MSR identified invoice no., customer, product, quantity, and price. A product summary on each MSR indicated that the products sold included strip, tube, bar, plate, pipe, fittings, fasteners, and forgings. The customer lists showed that approximately 400 companies had purchased E-Brite 26-1 material from AVM in 1976.

The names of the companies and their locations could not be identified as nuclear customers.

Approximately 2900 shipping records (1970-50, 1971-50, 1972-205, 1973-300, 1974-340, 1975-530, 1976-830, and 1977-600) for E-Brite 26-1 were reviewed. The I:otice of Shipment (NS) document contained information similar to that in a Certificate of Conformance. The NS described the product including referenced standards and heat numbers; and identified the quantity, customer, invoice number, shipper, and the shipment source, date, and destination. In 1970, all the orders reviewed were for slabs, ingots or picces of strip which were shipped to steel companies, research companies, or engineering departments of other companies for development purposes.

The maiority of the orders in 1971 reflected this same product pattern, but the product line began to expand in 1972. None of the orders in 1970,1971, 41d 1972 referenced a standard or specification (i .e. , only "E-Brite 2h-1" was identified). In 1973, orders began to reference " grade XM-27'; under ASTM specifications (e.g., A268 for tubing and A240 for plahe, sheet, and strip).

It was noted that finished items' were shipped to customers from either AVM warehouses in Leetsdale or Glenwillard, Pennsylvania, consignee's (e.g. , R. J. Gallagher,'P. A. Frasse) facilities, or directly from a fabricator's facility. The following are typical items shipped from a fabricator's or consignee's facility on an AVM purchase order:

tubing (Damascus Tube and Teledyne), strip (Universal Cyclops and Superior Steel), plate (G.O. Carlson), fittings (Raymond Brass and i3un Weld Fitting), fasteners (Raymond Brass), forgings (Western Forge & Flange' , pipe (C. A. Roberts, R. J. Galla her and P. A.

Frasse and bar/ rod (Blairsville and Latrobe Steel .

21

I ORGANIZATION: THE B0C GROUP, INCORPORATED MURRAY HILL, NEW PROVIDENCE, NEW JERSFY REPORT INSPECTION NO.: 99901063/86-01 RESULTS: PAGE 4 of 4 ASME Code Cases 1496 and 1505 permitted the use of E-Brite 26-1 products - sheet, strip, and plate (SA-290), bar (SA-479), and seamless and welded tube (SA-268) - in the construction of Section VIII and Section III components, respectively. Both these code cases were approved in 1971. Of the approximate 2800 shipping records reviewed from 1972 through 1977, only 9 orders referenced code case 1496 for ASME Section VIII " Pressure Vessels" requirements.

The Section VIII orders included the following: In October 1974, plate (A-240) was sold to P. A, Frasse and shipped to Sybron Corporation in Elyria, Ohio on invoice no. 4396. Tubing was sold to P. A. Frasse and shipped to Union Carbide in Institute, West Virginia in February 1975. In April 1975, tubing was sent to Camden Alloy Fabricators in Camden, New Jersey, and a flanged / dished head was sent to Dupont in Deepwater, New Jersey on invoice nos. 4898 and 4870, respectively. Invoice nos. 5286 and 5295 covered bar-shipped to Lenape Forge in West Chester, Pennsylvania and forgings shipped to Allied Chemical in Geismar, Louisana, respectively, in September 1975.

Welded tubing (A-268) was shipped on invoice no. 5680 to Union Carbide in Institute, West Virginia in December 1975. In March 1976, a forging was delivered to Allied Chemical and forgings (SA182 and SA240) were sold to P. A. Frasse and delivered to Hampshire Chemical in Nashua, New Hampshire on invuice no. 7049 in November 1976. The 9 customers noted above cannot be identified as nuclear customers.

i None of the orders referenced code case 1505 for ASME Section III

" Nuclear Power Plant Components" requirements. Based upon the i

review of records at The 80C Group's f acility in Murray Hill, New Providence, New Jersey and discussions with Mr. Kies, it was deter-mined that AVM did not supply E-Brite 26-1 material to a nuclear facili ty. Accordin91y, the allegation pertaining to the use of E-Brite 26-1 items in a domestic nuclear power plant cannot be substantiated.

l

2. Persons Contacted Frederick Kies, President i

i i

l i

l

{

22

ORGANIZAT10N: COOPER ENERGY SERVICES GROVE CITY, PENNSYLVANIA REPORT INSPECTION INSPECTION NO.: 99900317/86-01 DATES: 07/28-08/01/86 ON-SITE HOURS: 84 CORRESPONDENCE ADDRESS: Cooper Industries Cooper Energy Services ATTN: Mr. F. B. Stolba, Vice President and General Manager 150 Lincoln Avenue Grove City, Pennsylvania 16127 ORGANIZATIONAL CONTACT: W. H. Allen Lambert, Manager of QA TELEPHONE NUMBER: (412)458-8000 NUCLEAR INDUSTRY ACTIVITY: Original equipment manufacturer of standby diesel generators for nuclear service. Current sales in parts, repair and service only. No nuclear orders for engines.

ASSIGNED INSPECTOR: lb 1 E. H. Trottier, Reactive Inspection Section (RIS)

MkDate OTHER INSPECT 0P(S): W. P. Haass, Program Coordination Section E. Yachii k, RIS APPROVED BY: . n((g E.W.Merschoff, Chief //S,VendorProgramBranch Date INSPECTION BASES AND SCOPE:

A. PASES: 10 CFR Part 21 and 10 CFR Part 50, Appendix B.

B. SCOPE: This inspection was performed in response to two recent and significant Part 21 reports involving Cooper standby diesel generators.

In addition, this inspection sought to verify corrective and preventive actions taken in response to findings of the previous NRC inspection (84-01, March 12-16,1984).

PLANT SITE APPLICABILITY: Pyron 1/2 (50-454, 455); Braidwood 1/2 (50-456/457);

Cooper (50-298); Nine Mile Point 2(50-410); Palo Verde 1/2/3(50-528,529,530);

(continued on next page) -

23 l  :

I

ORGANIZATION: COOPER ENERGY SERVICES l GR0VE CITY, PENNSYLVANIA , !

REPORT INSPECTION NO.: 99900317/86-01 RESULTS: PAGE 2 of 12 PLANT SITE APPLICABILITY (continued) South Texas 1/2(50-498,499); Susquehanna 1/2(50-387,388); Waterford 3 (50-382); Zion 1/2 (50-295, 304).

A. VIOLATIONS:

There were no violations identified as a result of this inspection.

B. NONCONFORMANCES:

Contrary to Standards Numbering and Handling Procedure SA-1, out of date procedures (old revisions) and an out of date procedure index were found in the heat treatment area shop office procedures manual (86-01-01).

C. UNRESOLVED ITEMS:

There were no unresolved items resulting from this inspection.

D. STATUS OF PREVIOUS INSPECTION FINDINGS:

1. Inspection 84-01 (March 12-16,1984)

a. Nonconformance 1 (0 pen): Heat treatment procedures available in the fabrication area were not current revisions. Specifically, procedures HT-17N and HT-18AN were cited as being out of date according to the Standards Manual for Engineering Material Specifications.

The inspector reviewed implementation of the corrective and preventive actions that Cooper submitted in their response to this finding. Neither the corrective action nor the action taken to prevent recurrence has been effective.

In reviewing the revision status of the same two heat treat procedures (HT-17N and HT-18AN) available in the heat treat area shop office, out of date revisions were found and the l procedures index did not reflect the current revision to the

! procedures.

l I

The heat treatment area shop office was found to contain two sets of the subject heat treat procedures. One set was found in the Standards Manual for Engineering Material Specifications, while another set was found in a separate binder containing only the Standards Manual Chapter for heat treating (Chapter 13).

, Both the Manual and separate Chapter 13 binder had separate i

l 24

ORGANIZATION: COOPER ENERGY SERVICES GR0VE CITY, PENNSYLVANIA REP 0Hi INSPECTION N0.: 99900317/86-01 RESULTS: PAGE 3 of 12 procedure revision indexes. In the Standards Manual, procedure HT-17H was current (11/84) in both the index and as found, while procedure HT-18AN showed a revision date of "7/81" in the index and "9/5/75" as found. In the separate binder of heat treatment procedurcs, HT-17N showed a revision date of "11/84" in the index and "2/17/82" as found, while HT-18AN showed a revision date of "2/85" in the index and "7/28/81" as found. In summary, an out of date revision to HT-17N was found in the shop binder; only the index of the shop binder showed the current revision of PT-18AN. Nonconformance 86-01-01 was identified in this area of the inspection.

The preventive action taken by Cooper Energy Services (CES) to prevent recurrence of this nonconformance was found to be inple-mented, but its effectiveness was inadequate, as evidenced by Nonconformance 86-01-01 identified above. The inspector reviewed the standard document distribution list that showed position titles (rather than just names) for the intended distribution of documents such as procedure revisions. This change was made by CES to help ensure that as personnel changes occur in the company, standard documents would reach the required functional area, rather than be sent to a person who may not still be serving in the intended area of responsibility.

t b. Nonconformance 2 (Closed): Quality Control Inspection Plans i

(QCIP) for 11 components on 7 diesel generator units were not signed / stamped or dated by the inspector to indicate various inspection activities had been conducted.

Corrective action and preventive measures for this nonconformance were found to be adequately implemented. Specifically, the subject QCIPs were reviewed and corrected, where possible. A training class was conducted on or about June 15, 1984, to review the requirements of the governing Quality Control Procedure (QCP-10-1).

c. Nonconformance 3 (Closed): Purchase orders were missing for lube oil lines on four nuclear standby diesel engines and connecting rod bearing shells on one engine. These items are classified as critical components.

25

ORGANIZATION: COOPER ENERGY SERV 1CES GROVE CITY, PENNSYLVANIA REPORT INSPECTION NO.: 99900317/86-01 RESULTS: PAGE 4 cf 12 All documents related to this finding were reviewed. The purchase orders for the subject components were found with their respective documentation packages (material test reports). Further, it was determined that purchase orders for critical components are being maintained by CES Quality Assurance personnel.

d. Nonconformance 4 (Closed): Three quality control inspectors were found to have performed nondestructive testing for which they were not qualified; one inspector could not be verified as being qualified to perform such testing (no record of qualification).

The qualification history for these four persons was reviewed.

The three inspectors identified above as not being qualified at the time of performing the nondestructive testing were, in fact, properly and currently qualified to perform such examina-tions. The fourth inspcctor left the company on July 22, 1977.

His official records (including NDE qualification history) were disposed of in 1979, in accordance with the requirements of QCP-10-12, " Quality Program Requirements for Qualification and Certification of NDE Personnel."

e. Nonconformance 5 (Closed): Calibration services were performed by seven companies with no evidence that their calibration program had ever been reviewed and approved by CES. One of these seven companies did not appear on the Approved Vendors List.

The inspectur reviewed records associated with these seven Cdlibration service companies. Each has now been approved by either a review of their Quality Assurance Program, or by demonstrated performance history. All seven companies were found on the current Approved Vendors List,

f. Nonconformance 6 (Closed): Various pieces of measuring and test equipment in use at shop work stations were found overdue for calibration, or their calibration stickers were inadequate or nonexistent. In addition, CES does not investigate the accept-ability of items previously tested or inspected by a piece of METE found to be out of calibration.

The inspector checked several pieces of test equipment in manufacturing areas of the shop. All stickers were legible and all equipment was current in calibration status. An internal memorandum issued August 20, 1986 requires the tool room calibration technician to inform the quality control area 26

ORGANIZATION: COOPER ENERGY SERVICES GR0VE CITY, PENNSYLVANIA REPORT INSPECTION N0.: 99900317/86-01 RESULTS: PAGE 5 of 12 supervisor (responsible for M&TE tool room) when any piece of test equipment is founo out of calibration. The supervisor is responsible for reviewing and investigating all relevant aspects of the out of calibration incident (with particular emphasis on the margin of calibration error and its impact on previous production) and documenting this review in a written report.

The report is submitted to CES Quality Assurance for potential corrective action.

g. Nonconformance 7 (Closed): Some purchase orders for critical components were not signed and dated by QA personnel, while others did not require the supplier to have a QA program.

In discussions with CES personnel, the inspector was advised that purchase requirements are often satisfied by a document that augments the original purchase order. This document is an Engineering Control Specification, having special purchasing requirements for nuclear application and is designated "SGN."

All purchase orders associated with this nonconformance were reviewed and found to have referenced SGNs that require appro-priate vendor qudlity assurance / quality control conditions.

Purchase orders cited by this nonconformance were reviewed and found to have been signed by CES QA representatives as required.

h. Nonconformance 8 (Closed): No evidence was found of documented instructions or procedures that addressed procurement document content and control.

The inspector reviewed Section 5 of the CES Quality Assurance Manual (QAM), Paragraph 5.2 and found that it specifics all requirements for procurement document control, including a review by Quality Assurance for the original procurement docu-ments and any changes thereto.

i. Honconformance 9 (Closed): No evidence was presented to show that three vendors of critical components (Jacket water pumps and crankshaf ts) submitted a copy of their QAM, or of a QAM evaluation checklist.

The inspector reviewed the CES quality control prccedure governing the requirements for vendor selection. The procedure (QCP-10-8) has been revised to allow a venaar's past Performance history to be a basis for selection as an s

27 i

i ORGANIZATION: COOPER ENERGY SERVICES GR0VE CITY, PENNSYLVANIA REPORT INSPECTION N0.: 99900317/86-01 RESULTS: PAGE 6 of 12 approved vendor. This method of approval has been applied to the three vendors who were the subject of this noncon-formance.

J. Nonconformance 10 (Closed): CES had not established measures to preclude the requisition of M&TE calibration nonconformance identified in previous customer audits.

The inspector reviewed the corrective and preventive. measures taken to preclude recurrence of this nonconformance and found them adequate. The computerized tool recall and inventory program is in effect and the listing was found to include the personal calibration equipment of manufacturing personnel.

Tools selected at random by the inspector were found to be within the calibration frequency specified in the master calibration list.

k. Nonconformance 11 (0 pen): The written 10 CFR Part 21 response documenting evaluation of the circumstances surrounding damaged resistance temperature detector wires on a standby diesel engine was not provided within the time specified in QCP-10-14.

In reviewing the status of corrective and preventive measures for this nonconformance, the inspector noted that in at least one instance, no evidence was found to show that CES had implemented their stated preventive action. (Adviseengineer-ing of the 30 day response requirement; followup with engineer-ing if their input is not received in 20 days.) Discussions with CES personnel revealed that in this instance, investigation and determination of the root cause of an engine malfunction took several months. Thus, the efficacy of the 30 day response requirement was questioned, except to satisfy the requirement of the subject CES procedure (QCP-10-14). CES has therefore comitted to revise QCP-10-14 and delete the 30 day requirement for a 10 CFR Part 21 report that often is not technically possible.

1. Nonconformance 12 (Closed): Design changes on several Request for Draf ting Action (RFDA) forms were not verified or checked by an individual other than the person who performed the original design.

28 l l

ORGANIZATION: COOPER ENERGY SERVICES GR0VE CITY, PENNSYLVANIA REPORT INSPECTION N0.: 99900317/86-01 RESULTS: PAGE 7 of 12 Corrective action for this nonconformance was accomplished by a re-review of each of the subject RFDAs. The inspector verified that the re-reviews were conducted and documented. Preventive action was accomplished by a revision to the procedure that controls RFDAs (CES Procedure SA-4).

This procedure revision was reviewed by the inspector and found to be adequate.

m. Nonconformance 13 (Closed): The " Order Affected" block on several RFDAs was not completed as required; several RFDAs that had been approved and implemented, were not signed as being approved.

The inspector reviewed the subject RFDAs and found them to be pruperly signed off. A review of an additional 10 RFDAs did not yield any deficiencies.

n. Nonconformance 14 (Closed): Several nonconforming items /

deviations / failures in CES equipment shipped to nuclear facilities were not evaluated by CES as required by 10 CFR Part 21.

The inspector revieweo all correspondence related to deficiencies in CES-supplied equipment at commercial nuclear pcwer plants.

No evidence was found to show that CES ever failed to conduct a technical evaluation, followed by a 10 CFR Part 21 notifica-tion when appropriate, of any deficiency about which CES had knowledge. It appears that the subject nonconforming items /

deviations / failures were never brought to the attention of CES by the licensee.

o. Nonconformance 15 (Closed): The CES procedure identified as controlling the progress of potential 10 CFR Part 21 noncon-formances did not specifically provide for the review and escalation into the 10 CFR Part 21 review system of identified I problems.

The inspector reviewed the CES QA procedure that governs complicnce to the requirements of 10 CFR Part 21. Paragraph 4.2.1 cf the subject procedure was found to contain a require-l ment that a Material Review Request (MRR) be issued for defi-l ciencies that are potential 10 CFR Part 21 items. The MRR is identified as being a Part 21 item by the words " Reported per i

29

ORGANIZATION: COOPER ENERGY SERVICES GR0VE CITY, PENNSYLVANIA REPORT INSPECTION '

NO.: 99900317/86-01 RESULTS: PAGE U of 12 QCP-10-14" written on it. This alerts all CES employees to the l purpose of the MRR. The inspector noted some inconsistencies in how these MRRs are actually being identified by CES. Some were found to have " Critical" written on them. When brought to the attention of the CES QA staff, it was learned that an internal memorandum was being prepared to correct this, as well as other related 10 CFR Part 21 findings. (seeitemkabove.)

p. Nonconformance 16 (Closed): CES procedures do not require, and records were not prepared to assure compliance with the 2-day reporting requirement found in 10 CFR Part 21.

The inspector examined records (letters to the Nuclear Regulatory Commission (NRC)) that indicated the Vice President and General Manager of CES consistently notified the NRC of a defect / failure within 2 days of learning that the defect / failure is known to have an impact on safety. This 2-day notification is subsequent to the evaluation that establishes the existence of a defect /

failure. Section 4.2.4 of the CES procedure on Part 21 notifi-cation appears to address this matter adequately.

E. OTHER FINDINGS AND COMMENTS:

1. Part 21 Program The major purpose for conducting this inspection was to review CES' program for complying with 10 CFR Part 21. Two recent and signifi-cant engine failures at commercial nuclear power plants were reported to the NRC as being under evaluation by CES. The evaluation process for each failure was reviewed by the inspector. The CES evaluation of each appears to be adequate, as follows:

a. Rocker Arm Failures l

Two of four KSV-20 diesel engines manufactured by Cooper-Bessemer and installed at Commonwealth Edison Company's Byron Station experienced rocker arm failures in November 1985 and February 1986. As a result of the ensuing evaluation, it was initially concluded that the broken rocker arms were a secondary failure that occurred as a l result of the seizure of the valve train crosshead bushing.

l 30 l

l -

I l ORGANIZATION: COOPER ENERGY SERVICES GR0VE CITY, PENNSYLVANIA l

REPORT INSPECTION NO.: 99900317/66-01 RESULTS: PAGE 9 of 12 The latter was thought to be the direct result of some radial misalignment of cylinder liners, which in turn caused cylinder head misalignment. The liner misalignment was corrected, the broken rocker arms were replaced, and the engines were successfully tested.

On March 11, 1986, CES issued a Part 21 report describing the problen, the investigation conducted, and the recommended corrective action. The vendor also issued Service News Bulletin No. 698, " Power Head Assembly," dated March 11, 1986 to identify the problem and present what was thought to be the solution.

On May 5, 1986, another rocker arm failure occurred. Alignment of the cylinder line was rechecked and found to be within toler-ance. In reviewing the entire maintenance history of the engine, a seemingly insignificant fact was uncovered: the utility had a history of ordering an excessive number of push rod retaining rings. Since these retaining rings were also bent in the recent failures, the CES investigation focused on the connection between these two apparently unrelated items. The result was the sclution to the history of seized crosshead bushings and rocker arm failures at Byron Station.

The root cause of the rocker arm failures was a misinterpretation of the vendor's instructions for setting valve tappet to lifter clearance. In setting the tappet clearance, the dial indicator probe was placed on an incorrect surface, giving an improper measurement. More significantly, adjustment must be performed on cylinder pa!rs, rather than on all cylinders with a common engine position. Performing the c?carance check incorrectly resulted in improper location of the push rods, effectively increasing their length, and causing excessive force on the rocker arms. Althoagh the vendor concluded that the problem was not generic, an instruction manual supplement (#72486) has been i prepared for distribution to all affected customers. This is to help assure complete c:arity of the valve clearance procedure.

The necessary corrective action <.as been performed on the diesels at Byron Station. The diesels have been successfully tested, reinspected, and declared operational. They continue to run satisfactorily.

31 I

ORGANIZATION: COOPER ENERGY SERVICES GR0VE CITY, PENNSYLVANIA REPORT INSPECTION N0.: 99900317/86-01 RESULTS: PAGE 10 of 12 l

b. Loose Air Start Valve Seat Inserts On May 5,1986, an attempt to start an engine at Nine Mile-2 was unsuccessful. The engine would not " roll" with starting air. Subsequent investigation revealed that the air start valve seat insert in one cylinder was loose. The loose insert was free to fall out as its air start valve opened, effectively preventing starting air from entering the engine cylinder. The vendor determined that the loose insert was caused by improper correction of the head over-bore for the valve seat insert. The correction involved copper plating the insert, which then deformed as a result of thermal cycling. A second insert was also found to be loose. It appears that the second loose insert was caused by an error in the selection of the insert part number during the manufacturing process. Inserts are specifically matched to the insert bore to assure a proper shrink fit. Selecting an incorrect part did not allow a proper interference fit.

The utility proposed to check the adequacy of all starting air valve inserts on both engines (32 cylinder heads). The check was performed using a specially designed tool and precalculated force to determine insert tightness. No other inserts were found to be loose. Other types of inserts (inlet and exhaust valves) were checked for tightness and found to be acceptable.

The utility performed an audit of CES records regarding air valve inserts and whether or not copper plating was used elsewhere as a corrective measure to assure a proper interference fit. No further errors or copper plating locations were found.

The corrective actions have been completed, the engine tested and declared operational.

CES performed an audit of the records of diesel engines for their other nuclear utility customers. Nine other inserts located in six engines owned by three licensees were identified as having copper plated insert bores. Two of the licensees (Commonwealth Edison Company and Pennsylvania Power and Light Company) have been notified. The third utility (Arizona Public Service) is in the process of being notified,

c. Part 21 Activities CES's Part 21 files were reviewed to determine the adequacy of problem evaluation, reportability evaluation, corrective action, notification of affected customers, and adherence to their procedure. All files appeared to be adequately complete.

. 32 l

ORGANIZATION: COOPER ENERGY SERVICES GR0VE CITY, PENNSYLVANIA REPORT INSPECTION NO.: 99900317/86-01 RESULTS: PAGE 11 of 12 The inspector also reviewed a file of engine problems that were deemed by the vendor to not warrant reporting to the NRC. All items appeared to be adequately categorized in that they were not generic, or not safety related.

The CES's Part 21 file was compared to the listing of Part 21 reports received by the NRC from all sources during the 1985-1986 period. Discrepancies were found in three areas:

- Failure of turbocharger bolts at Palo Verde in September 1985 due to the lack of proper pre-stress to preclude fatigue by vibration was determined to be non-safety related.

Slipping of the fuel rack linkage arm on the governor shaf t was determined to require no corrective action because insufficient assembly torque was used by the maintenance staff.

However, CES will use a larger bolt on future engines.

- Failure of two Cooper diesel engines to sustain an adequate fuel oil prime at Braidwood station was attributed to an architect-engineering installation problem.

Other than the above, which were judged to be acceptable differences, there was complete consistency with NRC records.

The vendor's Part 21 procedure requires all problem identifi-cation and evaluations, potentially reportable under Part 21, to be reported to the QA Manager for purposes of central filing.

Contrary to this requirement, not all nonreportable evaluations were included in the QA Manager's file. Missing from the file are reports or requests from utilities made to other groups within CES (e.g., Engineering Service Group and Field Office).

To remedy this situation, an interoffice memorandum was prepared that requires "...any such communication received by anyone in the C.B. (ESG) organization be copied to the QA Manager regardless of the ultimate disposition." The stated intent of the memo is to ensure QA has a copy of all relevant documents regarding potential 10 CFR Part 21 items. This memo was prepared on July 31, 1986 and sent to all CES's Division Managers.

l 33

ORGANIZATION: COOPER ENERGY SERVICES GR0VE CITY, PENNSYLVANIA REPORT INSPECTION N0.: 99900317/86-01 RESULTS: PAGE 12 of 12

2. Measuring and Test Equipment Control In reviewing the CES program for calibration of measuring and test equipment, the inspector noted some confusion between required calibration frequency, required calibration date and due date marked on calibration stickers attached to individual pieces of test equipment.

In discussions with CES personnel and a review of the governing Quality Control Procedure (QCP-10-15, Section 4.7), it appears that unnecessary confusion is being created by the choice of the date format used to establish when a piece of test equipment is due for calibration, and when it should be next recalled. The QCP specifies, and the master calibration schedule printout calls for, a " year-week" format. For example, a date identified as 86/21 indicates calibration due the twenty-first week of 1986. The difficulty, of course, is that few people can convert that information into a commonly understandable date (i.e., week of May 18). The stated reason for using such a system is that the CES accounting system is based on calendar weeks and the charges for calibration services are billed internally using that format.

The inspector noted that a vernier caliper was marked " Cal Due 7-18-86," but in reviewing the calibration schedule printout, it was last calibrated the 20th week of 1986. Using a quarterly calibration frequency, as required, 17 weeks after the 20th week yields calibration due the 33rd week of 1986 (i.e., week of August 11). When the calibration technician was asked about the oisparity, he produced a conversion card for weeks to dates. When asked why the piece of test equipment was incorrectly dated if a conversion chart was available, the inspector was told the dates are usually estimated by adding 13 weeks to the current date and approxi-mating when that would occur in the usual day / month / year format.

The present system of scheduling calibration due dates for measuring and test equipment at CES is no't well controlled. The inspector l received a verbal commitment from the QA staff that a review and probable change would be forthcoming.

}

34

! ORGANIZATION: CONTROL PRODUCTS CORPORATION GRAFTON, WISCONSIN INSPECTION INSPECTION REPORT N0.: 99901045/86-01 DATES: June 23-26, 1986 ON-SITE HOURS: 50 CORRESPONDENCE ADDRESS: Control Products Corporation ATTN: Mr. R. J. Bertling, President 1000 Hickory Street Grafton, Wisconsin 50324 ORGANIZATIONAL CONTACT: D. J. Schaeffer, QA Manager TELEPHONE NUMBER: (414) 377-0800 NUCLEAR INDUSTRY ACTIVITY: Control Products Corporation manufactures three types of Agastat electrical control relays which are sold and marketed ex-clusively by the Amerace Corporation. The three types of nuclear designation Agastat relays are: magnetic latching type (EML), general purpose (EGP) and timing relays (ETR).

> / /

ASSIGNED INSPECTOR: f4A22  ! 8JB (,

J. K Pe oMno, Reactive Inspection Section (RIS) D te OTHER INSPECTOR (S): T. P. Guilfoil, en National Laboratory APPROVED BY: .

t[e./rt E. W. Merschoff, , RIS, Vendor Program Branch 'Date l

INSPECTION BASES AND SCOPE:

A. BASES: Appendix B to 10 CFR Part 50 and 10 CFR Part Pl.

B. SCOPE: Verify QA program implementation as a follow-up of a recent U.S. Nuclear Regulatory Commission safety system modification inspec-tion at the Dresden nuclear power plant. Additionally, the implemen-tation of the 10 CFR Part 21 program was reviewed.

PLANT SITE APPLICABILITY: All nuclear power stations that use Agastat electrical control relay series EGP, ETR, and EML.

35

ORGANIZATION: CONTROL PRODUCTS CORPORATION GRAFTON, WISCONSIN REPORT INSPECTION NO.: 99901045/86-0E RESULTS: PAGE 2 of 8 A. VIOLATIONS:

None.

B. NONCONFORMANCES:

1. Contrary to Criterion VI, " Document Control," of Appendix B to 10 CFR Part 50, CPC failed to establish adequate measures to control the review and approval of document changes. Out of 28 quality assurance (QA) procedures, 21 were found to have been revised without current revision review and approval. (86-01-01)

2. Contrary to Criterion VII, " Control of Purchased Material, Equipment, and Services," of Appendix B to 10 CFR Part 50, no pre-award survey or periodic vendor survey had been performed for the prccurement of seismically cualified relay base plates from the Overseas CCA company (Chauvin-Avnoux). (86-01-02)

3. Contrary to Criterion XVIII, " Audits," of Appendix B to 10 CFR

' Part 50, CPC failed to establish or implement an audit system to verify compliance with all aspects of its QA prograri. (86-01-03)

C. UNRESOLVED ITEMS:

None.

D. STATUS OF PREVIOUS INSPFCTION FINDINGS:

None.

E. OTHER FINDINGS OR COMMENTS:

1. 10 CFR Part 21

a. The Control Products Corporation (CPC) Procedure QI.23.0, revision dated February 25, 1986, was reviewed for its adequacy to provide compliance with 10 CFR Part 21 reporting requirements.

! In addition, the procedural implementation of the postino

' requirements (10 CFR Part 21.6) were evaluated by inspecting the manufacturing areas.

This review found CPC Procedure QI.23.0 lacking clarity because it did not provide assurance that CPC would notify l

36 l

l l

ORGANIZATION: CONTROL PRODUCTS CORPORATION GRAFTON, WISCONSIN REPORT INSPECTION RESULTS: PAGE 3 of 8 NO.- 99901045/86-01 the Commission in accordance with 10 CFR Part 21.21(b)(1).

CPC's procedure passed the responsibility of reporting to the Amerace Corporation. A procedure revision was being processed by the CPC at the time of the exit meeting. A review of this  !

procedure will be accomplished during a future NRC inspection.

b. The provisions of 10 CFR Part 21 were not imposed on CPC by the Amerace Corporation for the CPC manufacture of nuclear grade Agastat electrical relays. This failure of the Amerace Corporation to impose 10 CFR Part 21 was dis-cussed with Amerace and CPC personnel. Both companys agreed to rectify the oversight immediately. Follow-up of this issue will be performed at the Amerace Corporation offices in the near future.

2. Procedure Revisien Control CPC Procedure A-100 was reviewed and the implementation of procedure control was evaluated by reviewing 28 CPC procedures. Procedure A-100 was adequate with regard to the 10 CFR Part 50, Appendix B requirements. However, the area of revision control was found to need clarification. This area was discussed with CPC personr.el and was under evaluation at the time of the exit meeting.

Of the 20 procedures reviewed, 21 had been revised after the review and approval signature date. Several of these procedures had been revised two or three times following original review and approval dates. The procedures listed below were found to deviate from 10 CFR Part 50, Appendix B requirements (see nonconformance 86-01-01).

Q2.6.1 Inspection Instructions Q2.7.2 Control of Inspection Instructions

, Q2.7.10 Certificate of Compliance and Test Q2.10.1 Process Control Q2.11.1 First Piece Inspection Q2.11.2 Peceiving Inspection

! Q2.13.1 Control of Measuring and Test Equipment Q2.14.1 Handling, Storage and Shipping l 02.15.1 In-process Inspection, Test and Audit I 02.16.1 Control of Non-conforming Material or Items l Q2.16.2 Problem Evaluation 02.16.3 Notification of Customers, Users and/or Agencies 37 l

l

ORGANIZATION: CONTROL PRODUCTS CORPORATION GRAFTON, WISCONSIN REPORT INSPECTION NO.- 99901045/86-01 RESULTS: PAGE 4 of 8 Q2.17.1 Corrective Action Request Q2.18.1 Control and Filing of Records in Quality Assurance 02.18.2 Traceability Records for "E" and "F" Products Q2.19.1 Internal Audits E900 Deviation Request and Authorization E1000 Engineering Stop/ Rescind Order E1100 "C" Sheet-Special Product Information P100 Supplier Selection and Evaluation PS100 Customer Return Procedure

3. Approved Vendors CPC Procedure P-100 relating to supplier selection and evaluation was reviewed and implementation of vendor survey and evaluation requirements were evaluated (see nonconformance 86-01-02).

This review revealed that:

a. Section 5.2.1 of Procedure P-100 requires the QA department to send out QA pre-award survey form EN206 for a potential supplier to complete and send back to CPC. Only two EN206 forms were on file with one EN206 pending return from a vendor out of nine vendors who were on the CPC approved vendors list.

The vendors who had completed EN206 forms were Pennsylvania Pressed Metals, Cosmo Plastics; the pending files were for j Trico Products.

! b. Section 5.2.2 of Procedure P-100 requires the CPC QA department l to complete a supplier evaluation form EN207, sign and date, j and insert in each suppliers file for a vendor status refer-ence, and final supplier selection after internal CPC reviews are complete. Out of nine vendors who were on the approved vendors list, only Pennsylvania Pressed Metals had a form 707 on file.

It should be noted that a review of the components that each vendor supplies indicated that Overseas CCA (Chauvin-Avnoux) is the only vendor that supplies other than commercial grade items to CPC.

l i

1 38

ORGANIZATION: CONTROL PRODUCTS CORPORATION GRAFTON, WISCONSIN REPORT INSPECTION RESULTS: PAGE 5 of 8 NO.: 99901045/86-01

4. Audits The CPC Procedure Q2.19.1, " Internal Quality Audits," was reviewed to evaluate the implementation of planned and periodic audits.

This review sevealed that:

a. The QA Manager had not established an overall plan for periodic audits of CPC's internal functional areas and activities as required by Section 4.1 of CPC Procedure Q2.19.1 and ANSI N45.2.

b. CPC management did not assure that periodic audits were per-formed with established CPC Procedure Q2.I9.1. See p(rocedures nonconformanceas required by Section 86-01-05.)

However, a QA program implementation audit was performed by the CPC OA manager in May of 1985.

5. Quality Assurance Manual CPC's QA Manual was reviewed for compliance with 10 CFR Part 50, Appendix B, and ANSI N45.2-1977. The manuil was in compliance with these documents. The manual had been issued by the QA Manager in June of 1985, was signed by the Vice-President of manufacturing and was approved by the President. Although all revisions to the manual entered since the manual was issued had been properly entered, the method of documenting revisions to the manual sections was confusing and required clarification. This matter was discussed with the OA Manager.

This review resulted in no adverse findings.

6. Receiving Inspections The methods required by QA Procedure A2.11.2, " Receiving Inspection,"

to be followed for the inspection and disposition of production l material received at CPC were reviewed for compliance. Areas examined included:

! a. Procedures for the processing of all incoming production material used in "E" series products before the material is transferred into stores or to the manufacturing floor.

39

)

ORGANIZATION: CONTROL PRODUCTS CORPORATION GRAFTON, WISCONSIN REPORT INSPECTION NO.- 99901045/86-01 RESUL TS: PAGE 6 of 8

b. Review of on file Receiving Reports (EN 24 forms) for com-pleteness and proper distribution.

c. Review of Quality Control Vendor Record Data cards (EN 25 forms) for completeness and proper documentation.

d. Review of First Piece Inspection Records (EN 64 forms) for completeness.

e. Inspection of areas designated for segregating material that has not been inspected or that has been inspected but rejected to prevent its inadvertent use in production.

This review resulted in no advtirse findings.

7. Control of Nonconformine Paterials or Items The methods required by QA Procedure Q2.16.1, " Control of Noncon-forming material or items," to ensure that material cr items which are nonconferraing are properly identified, segregated, docbmented, reviewed and disposed of were reviewed for compliance. Areas examined included:

a. Procedures used to stop work on material found to be noncon-forming,

b. Methods used to physically isolate the nonconferraing material from acceptable material.

c. Review of Defective Material Report (DMR) comittee findings on DMRs associated with material furnished by the nine suppliers

! listed in Section E.3 of this report. The DMR review committee l consisted of representatives from Manufacturing, Engineering.

Purchasing, Production Control, and Quality Assurance.

i

d. Review of Purchasing Department files for DMRs, vendor notifi-l cation and follow-up to obtain vendor response, if required.

l e. Review of incomplete DMPs to ensure interim copies were i

distributed as appropriate.

! f. Review of completed DMR file to ensure completed DMRs were properly closed out.

l This review resulted in no adverse findings.

l 40 1

I ORGANIZATION: CONTROL PRODUCTS CORPORATION GRAFTON, WISCONSIN REPORT INSPECTION NO.- 99901045/86-01 RESULTS: PAGE 7 of 8

8. Problem Evaluations The methods required by 0A Procedure 02.16.2, " Problem Evaluation,"

to be followed when an in-plant problem arises that cannot be handled by a DMR or a problem develops in the field were reviewed.

This review resulted in no adverse findings.

9. EGP Series Power Relay Assembly Technical specifications, process specifications, bills of materials, and floor supervisor copies of routing sheets were reviewed. The actual assembly process required for "E" series GP power relays was observed.

This review and observation resulted in no adverse findings.

10. Test Procedures for EGP Series Power Relay In process test procedures TP-GPR-01, " Test Procedure for GP and TR Series and TP-GPR-02, Dielectric Testing of GP Series Relay Coils,"

were reviewed and the test procedures observed to verify proper testing of the "E" series GP power relay. '

OA Procedure Q2.13.1, " Control of Measuring and Test Equipment," was reviewed and its applicability and use to the test equipment used in testing the "E" series GP power relay was verified.

Calibration procedures, records and periodicity for the test equipment used in testing the "E" series GP power relay were reviewed.

This review and observation resulted in no adverse findings.

11. Production Control Equipment and material used during the production of GP series power relays as well as the production control practices were observed during the inspection. The GP series power relays were being manu-factured in accordance with industry standards.

41

l ORGANIZATION: CONTROL PRODUCTS CORPORATION GRAFTON, WISCONSIN REPORT INSPECTION Hn.- 99901045/86-01 RESULTS: PAGE 8 of 8 F. PERSONS CONTACTED:

*R. J. Bertling President, CPC i

• T. L. Nertec Vice-President, CPC

• D. J. Schaeffer QA Manager, CPC S. Sanfelippo Production Control Manager, CPC R. Hillenbrand Production Supervisor, CPC E. Lesnik Material Manager, CPC

• J. Ferguson QA Manager, Amerace Corporation CAttended exit meeting.

I 42

ORGANIZATION: ENERFAB, INCORPORATED CIfiCINNATI, OHIO REPORT w INSPECTION INSPECTION NO.: 99901064/86-01 DATES: 09/15-18/86 DN-SITE HOURS: 16 CORRESPONDENCE ADDRESS: Enerfab, Incorporated.

ATTN: Mr. Dwaine A. Godfrey President 4955 Spring Grove Avenue Cincinnati, Ohio 45232 ORGANIZATIONAL CONTACT: Mr. Kenneth T. Shinkle TELEPHONE NUMRER: (5131 641-0500 NUCLEAR INDUSTRY ACTIVITY: 1) Containment airlock spare parts for approximately 40 nuclear plants. 2) Nuclear construction activities at Crystal River nuclear plant.

ASSIGNED INSPECTOR: / I?!86 E. Yachiniiak, Reacyve Inspection Section (RIS) Date OTHER INSPECTOR (S):

APPROVED BY: . to/7!rg E. W. Merschoff, Chief, R)5 . Vendor Program Branch Date INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 50, Appendix B; 10 CFR Part 21.

B. SCOPE: As a result of two separate 10 CFR Part 21 notifications to the NRC by Consumers Power, Palisades plant and Indiana and Michigan Electric Company, D.C. Cook nuclear plant, an inspection of the current vendor, Enerfab, Inc., was performed. The most recent notificaticn, by Consumers Power, dealt with the improper design of the pressure equalizing valve (enntinued nn Pano M PLANT SITE APPLICABILITY: Various.

43

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

ORGANIZATION: ENERFAB INC.

CINCINNATI, OHIO REPORT INSPECTION NO.: 99901064/86-01 RESULTS: PAGE 2 of 6 SCOPE: (continued) actuation cam in their containment emergency escape air-lock. The second notification, by Indiana and Michigan Electric Company, dealt with the concern that their containment airlock window glass was not qualified to meet LOCA environmental requirements. Both these items were reviewed to ensure that Enerfab's Part 21 evaluations had been adequately performed. In addition, the Enerfab Quality Assurance program was reviewed in the areas of procurement, receipt inspection, indoctrination and training, internal and external audits, and measuring and test equipment control.

A. VIOLATIONS:

Contrary to Section 21.21, " Notification of failure to comply or existence of a defect," of 10 CFR Part 21, Enerfab failed to adopt appropriate proce-dures to provide for (i) evaluating deviations or (ii) informing the customer (licensee) so that the deviations may be evaluated. (86-01-01)

This is a Severity Level V violation (Supplement VII).

B. NONCONFORMANCES:

1. Contrary to Criterion V, " Instructions, Procedures, and Drawings of Appendix B to 10 CFR Part 50, Enerfab failed to establish adequate procedures and instructions for quantitative or qualitative acceptance criteria for receiving inspection activities and handling, shipping, and storage activities. (86-01-02)

2. Contrary to Criterion VII, " Control of Purchased Material, Equipment, and Services," of Appendix B to 10 CFR Part 50, Enerfab failed to ensure that Quality Techniques, Incorporated was on the approved vendors list for the calibration of gage blocks (standards) used to calibrate Enerfab measuring and test equipment. (86-01-03)

C. UNRESOLVED ITEMS:

On April 16, 1986, the D.C. Cook nuclear plant notified the NRC that pursuant to 10 CFR Part 21 and 10 CFR 50.73(a)(2)(v), a potential problem with their containment airlock observation window glass existed.

Based on test information supplied by Enerfab, a sample of the tempered window glass shattered when exposed to a beta radiation dose of 2.5 megarads over a 10-second period. This beta dose would have been reached in less than one (1) hour during a D.C. Cook design basis loss-of-coolant accident. Because D.C. Cook uses an annealed (untempered) window glass, 44

ORGANIZATION: ENERFAB INC.

CINCINNATI, CHIO l

REPORT INSPECTION N0.: 99901064/86-01 RESULTS: PAGE 3 of 6 there was uncertainty as to the applicability of this test information.

As a precautionary measure, 3/8-inch thick steel ccver plates were installed as beta shields. After extensive review, D.C. Cook reported the above concern to the NRC.

Discussions with Enerfab personnel and ether NRC members were undertaken to determine the significance of the test information. Since the beta dose was delivered in such a short tine, the conclusion was made that the glass shattered from induced thermal stresses, rather than radiation damage.

During the review of infornetion surrounding this concern, the original qualification report for the annealed window glass was located. The minimum test specified parameter for radiation exposure was set at 29.0 megarads from a 1.0 megarad/HR gamma source. Actual exposure was measured at 33.0-36.3 megarads. All four (4) window glass samples were then pressure tested to 60 psig, the required design test conditicn; all passed. At elevated pressures one (1) windcw glass failed at 110 psig and two (2) window glasses survived to 120 psig without breaking.

While it appears that the recent beta radiation test information does not directly affect the qualification of containment airlock window glass, the uncertainty as to whether the effects of beta radiation on the window glass were considered does exist. The original qualification documents do not address the effects of Beta radiation, thus making this an i unresolved item.

D. STATUS OF PREVIOUS INSPECTI6N FINDINGS:

Enerfab now controls all of the original W. J. Woolley (99900390) nuclear containment personnel airlock documentation, and supplies all replacerrent parts and engineering services for Woolley components. This is the first NRC inspection of Enerfab, Incorporated.

E. OTHER FINDINGS AND COMMENTS:

1. Emergency Airlock Equalizing Valve Design Discrepancy On August 8,1986, Consumers Power notified the NRC that as a result of an improper cam design, the coordination between the containment emergency escape airlock doors and their pressure equalizing valves was not correct. Combining the above defect with the arrangement of tre airlock doors actuation mechanism along a common shaf t, the opening of one door through the appropriate handwheel movement caused inadvertent opening of the opposite door's ecualizing valve. This 45

ORGANIZATION: ENERFAB INC.

CINCINNATI, OHI0 l

REPORT INSPECTION N0.: 99901064/86-01 RESULTS: PAGE 4 of 6 configuration would result in a vent path from the inside of contain-ment to the outside environment during any times that one airlock door was open.

After extensive review of original design drawings and other engineer-ing documents, concurrence with Enerfab's Part El design review was established. Based on the available information, it appears that improper cam placement on the common connecting shaf t or improper cam profile design caused early equalizing valve actuation at Palisades.

ll Corrective action recommended by Enerfab for Palisades consisted of the grinding of the cam profile. Depending upon plan specific cam / shaft arrangements, the below list of affected plants may need to follow the above recommended corrective actions.

Oconee 1, 2, 3 50/269, 270, 287 Arkansas 1, 2 50/313, 368 Shoreham 50/322 Shearon Harris 1 50/400

2. Part 21 Procedures During the review of the above two independent utility Part 21 submittals to the NRC, it was found that Enerfab did not have a procedure to provide for the evaluation of deviations as per 10 CFR Part 21.21. The evaluations were, however, found to have been performed adequately.

Violation 86-01-01 was identified in this area.

3. Inadequate Procedures The review of Enerfab's QA program focused on their current business activities in the area of replacement nuclear spare parts. After becoming familiar with their program, it was found that in two l areas, handling, shipping and storage, and receiving inspection, none or inadequate procedures existed.

With many spare parts orders requesting shipping and storage requirements meeting ANSI-N45.2.2, a procedure to assure compliance with the standard should have been available. If it was not. 0A procedure 10-19-002-100, revision 0, dated October 5, 1984, " Material Receiving" does not address the current spare parts activities which 46

ORGANIZATION: ENERFAB INC.

CINCINNATI, OHIO REPORT INSPECTION NO.: 99901064/86-01 RESULTS: PAGE 5 of 6 now constitute most of Enerfab's nuclear business. Specifically, receiving inspection criteria for acceptance of spare parts is not mentioned.

Nonconformance 86-01-0? was identified in this area.

4. Calibration Services While the importance of measuring and test equipment calibration is currently limited to the QC inspector's measuring tools, the calibra-tion of the NBS traceable standards (gage blocks) was not performed by a vendor on the Approved Vendors List. The calibration performed an October 28, 1985 by Quality Techniques, Incorporated was documented appropriately, but did not meet the requirements of Section VII of Appendix B to 10 CFR Part 50.

Nonconformance 86-01-03 was identified in this area.

5. Audits Section 19.0 of the Enerfab QAM covers internal and external audits.

This area was reviewed for conformance to that manual section and no adverse findings were found. Three (3) external audits were performed for companys that did not possess a Quality System Certificate (MM or MS) or an ASME Certificate of Authorization (N-type stamp). Because the Encrfab QAM was issued in November, 1985, the first scheduled internal audit is planned for November, 1986.

6. Measuring and Test Equipment Controls i

The only tools which are kept under the control of the QA department are a 0"-1" micrometer, a 0" "6" dial caliper, and a coating thickness measuring tool. Each had been calibrated to the requirements of l

QA Procedure 50-19-002-070, " Calibration of Measuring and Test l

Equipment."

7. Procurement Document Control A review of the Enerfab procurement process was performed with main emphasis being placed on the activities of the spare parts i

specialist. A sample of six purchase order (P0) documents was taken to verify that the appropriate QA, engineering and material management personnel had reviewed them. No discrepancies were 47

____ O

ORGANIZATION: ENERFAB INC.

CINCINNATI, OHIO 1

REPORT INSPECTION N0.: 99901064/86-01 RESULTS: PAGE 6 of 6 found in this area. It appeared that the majority of the procured items were replacement rubber gaskets and o-rings for personnel air-locks and manways. Whenever quality requirements or 10 CFR Part 21 were specified on the nuclear utility's P0, Enerfab imposed the appropriate sub-vendor quality documents and Part 21 on their P0 to that vendor.

If a commercial item for a safety-related system was to be procured, i a Commercial "off-the-shelf" justification was performed by the Engineering department. This was done for the Corning window glass material, which is commercially available, and has been tested and qualified by Enerfab for nuclear use.

F. PERSONS CONTACTED:

• K. T. Shinkle, Quality Assurance Engineer, Enerfab, Inc.

• R. M. Hulick, Project Manager, Enerfab, Inc.

• D. A. Godfrey, President, Enerfab, Inc.

J. M. Ventura, Senior Engineer / Designer, Enerfab, Inc.

A. E. Davis, Spare Parts Specialist, Enerfab, Inc.

• J. W. Rothel, Senior Vice President, Enerfab, Inc.

• denotes present at exit meeting.

l

! 48 l

l

i

)

l ORGANIZATION: GENERAL ELECTRIC COMPANY SAN JOSE, CALIFORNIA REPORT INSPECTION INSPECTION NO.: 99900403/86-03 DATES: 8/11-12/86 ON-STTF HollR9 17 CORRESPONDENCE ADDRESS: General Electric Company Muclear Energy Business Operating ATTN: Mr. W. H. Bruggeman, Vice President and General Manager 175 Curtner Avenue San Jose, California 95125 ORGANIZATIONAL CONTACT: J. J. Fox, Senior Program Manager TELEPHONE NUMBER: 408-925-6195 NUCLEAR INDUSTRY ACTIVITY: General Electric Company (GE), Nuclear Energy Business Operations (NEB 0) provides spare parts and services to Boiling Water Reactors.

ASSIGNED INSPECTOR: M h4!%

K. R. Naidu, Reactive Inspection Section (RIS) Date OTHER INSPECTOR (S): E. Yachimiak, RIS APPROVED BY: . gg E. W. Merschoff, Chief, K, Vendor Program Branch Date INSPECTION BASES AND SCOPE:

A. BASES: Appendix B to 10 CFR 50 and 10 CFR Part 21.

B. SCOPE

Obtain additional information on kits used to refurbish scram solenoid pilot valves (SSPV) at Vermont Yankee (VY) Nuclear Pcwer Plant and review the 10 CFR Part 21 evaluation of unusual control rod movement on June 14,1986 a t VY.

PLANT SITE APPLICAEILITY: Vermont Yankee (50-271); Limerick (50-352); Pilgrim (50-293); Dresden 2(50-237); Fitzpatrick (50-333), Hatch 1 (50-321); Browns Ferry 1 & 2 (50-250/260); Brunswick (50-324); Duane Arnold (50-331); LaSalle

, (50-373); Nine Mile Point (50-220) and Cooper (50-298).

49

ORGANIZATION: GENERAL ELECTRIC COMPANY SAN JOSE, CALIFORNIA REPORT INSPECTION N0.: 99900403/86-03 RESULTS: PAGE 2 of 7 l

A. VIOLATIONS:

None.

B. NONCONFORMANCE:

Contrary to 10 CFR 50, Appendix B, Criterion V, GE failed to follow Nuclear Service Procedure 30-03 dated August 1, 1985 for the VY purchase order (PO) 2260 supplement I dated August 9, 1984. The VY nuclear power plant required GE to supply 200 safety-related scram solenoid pilot valve (SSPV) refurbishment kits (kits) type FV 204-139. GE supplied VY with 200 non-safety related kits instead, without resolving the discrepancy between the P0 and the supply with VY (86-03-01).

C. UNRESOLVED ITEMS:

Subsequent to the inspection, an internal GE memorandum was made available, only for review, to the NRC inspector at the GE, Bethesda office on September 10, 1986. From this and other related information it appeared that GE, San Jose, had not adec.uately reviewed or evaluated the safety implications of the incorrect performance of VY SSPVs. The statement from -

GE "...the random, small number of occurrences of these discrepancies eliminate any potential safety impact...." does not constitute the basis for an adequate evaluation. The possibility that a number of discrepant SSPVs causing excessively long scram times does exist since five (5) of the VY SSPVs failed to initiate control rod motion until 5 to 7 seconds had elapsed.

Final Safety Analysis Report transient analyses for BWRs usually allow on the order of 200 milliseconds from the initiation of a scram to the start of control of rod movement. Since the inspectors were not provided any supporting evidence to show that the slower scram times would not consti-tute a substantial safety hazard, this item is considered unresolved and open pending receipt of further information.

D. INSPECTION FINDINGS AND OTHER COMMENTS:

1. Background Information On June 14, 1986, during single rod scram time testing, VY nuclear power plant reported that one control rod failed to scram and five others hesitated a few seconds before scramming. The failure of these control rods to function correctly was attributed to the failure of their respective SSPVs. These valves were refurbished by VY personnel with replacement kits that were supplied by GE, San Jose, California and manufactured by Automatic Switch Company (ASCO),

located in Florham Park, New Jersey.

50

ORGANIZATION: GENERAL ELECTRIC COMPANY  !

SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/86-03 RESULTS: PAGE 3 of 7 Three types of problems were identified in the faulty SSPVs. In one SSPV, the core assembly spring was separated from the core assembly.

This was attributed to improper assembly techniques. In another valve, on the exhaust side of the SSPV, the diaphragm was installed backwards. This was again due to improper assembly techniques. In four SSPVs, an incorrect core assembly was provided in the kit. This was attributed to personnel at ASCO selecting the incorrect core assembly which was not detected due to inadequate component QC inspections at ASCO.

Subsequent inspection of the remaining SSPVs found (1) one valve with a deformed core assembly spring and (2) one solenoid base sub-assembly with an out-of-round inside diameter. Both these conditions could have prevented free travel of the core assembly and could have adversely affected the scram performance.

2. Review of GE Purchase Orders The NRC inspectors examined GE P0s to ASCO-for the supply of refur-bishment kits type FV204-139 to ascertain whether they were purchased as safety related items, and whether GE required ASCO to perform appropriate Quality Centrol (QC) inspections. Review of GE P0s during the period of February 1984 to July 1986 indicates that the quality requirements imposed on ASCO increased progressively with time. The following are the details of the review;

a. GE P0 #334-AL421 dated February 14, 1984 to ASCO required the supply of 3000 SSPV FV204-139 type refurbishment kits. GE imposed QC plans A-42 and A-196 respectively. QC plan A-42 did not contain specific inspection requirements.

Similarly, QC plan A-196 described a quality assurance plan with 18 criteria and stated that this plan was applicable to items classified as " safety essential." The term " safety related" was not mentioned in the P0. GE stated that they used the term " safety essential" as the predecessor of the term " safety related." The above mentioned GE P0 referenced neither 10 CFR Part 21 nor 10 CFR Part 50 Appendix B.

b. GE P0 #205-85-F848 dated July 8,1985, required ASCO to supply 60 SSPV FV206-139 type refurbishment kits. Revision 1 to the PO dated August 16, 1985, required ASCO to provide cure dates on the 0-Rings and seals used in the kits. Revision 2 to the P0 dated September 10, 1985 required the cure date to be not more 51

l l

ORGANIZATION: GENERAL ELFCTRIC COMPANY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/86-03 RESULTS: PAGE 4 of 7 than 18 months prior to date of shipment. Revision 3 to the P0 dated November 16, 1985 changed the classification of the kits from non-safety-related to safety related. Revision 4 to the P0 dated November 21, 1985 stated that GE QAR 9 Revision 0 was applicable. The change in classification in Revision 3 to the GE P0 required ASCO to impose on itself, as per ASCO's QA program, ASCO QC procedure PP-467, " Scheduling Nuclear Power Customer Orders." This procedure required ASCO to perform additional QC inspections on individual components contained in the kits because the components are specifically intended for SSPVs installed in nuclear power plants.

c. GE P0 #205-86K032 dated April 14, 1986, required ASCO to supply 300 SSPV FV204-139 type refurbishment kits. This P0 stated that GE QAR 5, Revision 3 was applicable and required a GE representative to inspect the kits at ASCO prior to shipment.

ASCO treated these kits as safety-related and accepted 10 CFR Part 21 reporting requirements. The NRC inspectors reviewed the documents at ASCO for this GE P0 during an inspection conducted July 18-20, 1986 and determined that ASCO implemented their QC procedure PP-467 and stamped the cure dates for the 0-rings and seals on the individual blister packages.

3. Inspections Program by GE for Refurbishment Kits The inspectors discussed the types of inspections performed by GE, before and after July,1986 on SSPV refurbishment kits, which are shipped in blister packs by ASCO to GE and determined the following:

I a. Prior to July,1986, GE personnel verified the following i attributes during receipt of each lot of SSPV refurbishment l kits:

1. Visual verification that the identification on the blister pack is FV204-139,

2. Receipt of a certificate, accompanying the shipment, stating

" Identical replacement without change in design or material per GE QC plan A-42."

3. Document the lot number identifying means of traceability.

4. Compare that the cure date information on the ASCO certif-icate agrees with the information on the individual blister packs.

l

) 52

ORGANIZATION: GENERAL ELECTRIC COMPANY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/86-03 RESULTS: PAGE 5 of 7

b. After July 17, 1986, GE issued P0s to ASCO imposing a pre-kit quality control inspection checklist identified as Vendor Print .

File (VPF) 3061-83-1. VPF #3061-83-1, which was developed with the cooperation of ASCO, requires ASCO to perform 100% inspection on the components in each kit. The plan specifies dimensional and visual inspection requirements. The NRC inspectors deter-mined that this inspection plan contains adequate quantitative acceptance / rejection criteria.

4. Review of the Vermont Yankee Purchase Order Vermont Yankee (VY) issued P0 #22601 dated April 25,1984 to GE for the supply of 100 safety related Pilot Head kits, GE part #236X558-14.

Appendix D,Section III item d of the P0 stated in part, "...the requirements of 10 CFR Part 21 do not apply to replacement parts when the parts are not supplied with IEEE-323 certification." Item 3 of Supplement 1 to this P0 dated August 9,1984, increased the quantity from 100 units to 200 units and revised the identification of the kit as part number FV204-139. Appendix A,Section III of the supplement required a certification to the effect that the materials utilized in the fabrication of replacement are identical to material utilized in the fabrication of the original components. The classification of the kits as safety related remained unaltered.

GE issued two Product Quality Certificates (PQC) dated June 26, 1984 and November 2, 1984 to VY. Each referenced a " delivery to shipment" traceable to the shipping documents and certified that 100 kits met the applicable purchase documents. GE stated that the 200 kits shipped to VY were stored at VY and subsequently used to refurbish the SSPVs which experienced rod movement problems in June 1986.

i

5. Results of the Inspection 1

a. The NRC inspectors determined that GE supplied VY 200 non-safety-related FV204-139 type refurbishment kits instead of safety related kits. Documents reviewed at GE indicate that these kits were from a batch of 3000 kits procured by GE P0 #332-AL421 dated February 14, 1984. The GE Nuclear Service Procedure (NSP) 30-03 in paragraph 3.3.1 states that it is the responsibility of the Personnel of Spare and Renewal Parts to review the customer PO to assure the accuracy of the specifications inclu-ding the appropriate quality assurance requirements. Paragraph 3.3.2 states in part that the discrepancies between GE and the customer shall be resolved prior to formal acceptance of the i

53

ORGANIZATION: GENERAL ELECTRIC COMPANY SAN JOSE, CALIFORNIA l

REPORT INSPECTION NO.: 99900403/86-03 RESULTS: PAGE 6 of 7 customer Purchase Order. Resolution of discrepancies will be reached by GE management, quality assurance and customer and shall be documented by P0 change or other written documentation signed by the customer.

GE stated that they did not obtain a reconciliation from VY to supply 200 non-safety related kits (instead of safety related).

The NRC inspectors informed the GE representatives that failure to follow GE procedure NSP 30-03 to obtain a reconciliation from VY for the variance in the safety classification of the kits is considered a failure to follow procedures and identified this failure as a nonconformance contrary to the requirements of Criterion V of 10 CFR 50 Appendix B (86-03-01).

The NRC inspectors determined that GE supplied the following nuclear power plants with refurbishment kits from the same batch of 3000 received from ASCO.

Plant Quantity Shipped Date of Shipment Limerick 50 06/13/84 Pilgrim 15 06/21/84 Vermont Yankee 100 06/25/84 Pilgrim 8 06/27/84 Dresden 2 50 08/03/84

F1tzpatrick 100 09/05/84 l Vermont Yankee 100 11/05/84 Hatch 1 75 04/08/84 Browns Ferry 100 12/12/84 Brunswick 217 01/14/85 Hatch 1 100 02/28/85 Dresden 2 75 03/01/85 Brunswick 2 03/12/85 Ha tch 100 03/14/85 Duane Arnold 180 03/26/85 LaSalle 150 03/27/85 Nine Mile 1 100 04/02/85 Cooper 150 04/12/85 l Duane Arnolo 150 04/25/85 Browns Ferry 2 140 11/14/85 Mi11 stone 1 150 11/15/85 i

b. NRC issued Information Notice 86-78 alerting all BWR owners of the problems associated with the ASCO SSPV refurbishment kits.

54

ORGANIZATION: GENERAL ELECTRIC COMPANY SAN JOSE, CALIFORNIA l

l REPORT INSPECTION N0.: 99900403/86-03 RESULTS: PAGE 7 of 7

/ c. GE issued SIL No. 441, dated July 17, 1986, recommending corrective action for all BWR owners. This consisted of the following:

1. Continued control rod surveillance per Technical Specifications to detect SSPV performance deterioration,

2. Return of all unused SSPV Replacement Kits FV 204-139 to GE for reinspection,

3. Inspection for correct engagement of the coil spring onto the core assembly for all kits during SSPV refurbishment at the reactor site, and

4. Verification of SSPV operability through scram valve time tests or single rod scram time testing.

E. EXIT INTERVIEW:

The inspectors met with the GE representatives identified in Section F and discussed the scope of the inspection and findings.

F. PERSONS CONTACTED:

• J. J. Fox, QA Operations E. Gibo, Systems Engineer D. Arncid, Manager, Material Logistics W. Dye, Senior Buyer

• J. M. Bricken, QC Engineer

• B. A. Smith, QA Operations C. Lewis, QA Operations R. T. Hill, Licensing R. T. Kern, Electrical Design Engineer R. Waldman, Electrical Design Engineer

• H. P. Williams, Material Services Engineer

• Denotes those persons who were present at the exit interview on August 12, i 1986.

l 55

ORGANIZATION: ITT BARTON CITY OF INDUSTRY, CALIFORNIA REPORT INSPECTION INSPECTION NO.: 99900113/86-01 DATES: 7/7-11/86 ON-SITE HOURS: 44 CORRESPONDENCE ADDRESS: ITT Barton ATTN: Mr. Gerald R. Welt Director, Quality Assurance 900 South Turnbull Canyon Road City of Industry, California 91749 OR'GANIZATIONAL CONTACT: Ms. Jean Dwyer TELEPHONE NUMBER: (818) 961-2547 NUCLEAR INDUSTRY ACTIVITY: 8 to 10%.

ASSIGNED INSPECTOR: 4 /

R. H. Lasky, Equi @ent Qualification Inspection F/h Date Section (EQIS)

OTHERINSPECTOR(S): M. Jacobus, Sa dia National Laboratories

)I APPROVED BY: bWW U Potapovs, Chief, EQIS, \ pndor Program Branch T- 21-tf.

Date INSPECTION BASES AND SCOPE:

A. BASES: Appendix B to 10 CFR Part 50 and 10 CFR Part 21.

B. SCOPE: This inspection consisted of: (1) a technical evaluation of Equipment Qualification (EQ) activities for safety related equipment and (2) verification of implementation of the quality assurance program.

PLANT SITE APPLICABILITY: Plants with ITT Barton differential pressure and pressure electronic transmitters and indicating switches.

I S7

i i

ORGANIZATION: ITT BARTON CITY OF INDUSTRY, CALIFORNIA I

REPORT INSPECTION NO.: 99900113/86-01 RESULTS: PAGE 2 of 5 A. VIOLATIONS:

None.

B. NONCONFORMANCES:

None.

C. UNRESOLVED ITEMS:

None.

D. OTHER FINDIhGS*0R COMMENTS:

1. The inspection team examined ITT Barton's Quality Assurance (QA) manual for compliance to Section VII of Appendix B to 10 CFR Part 50, control of purchased material, equipment, and services. The inspec-tion team then examined selected documents to inspect ITT Barton's compliance to their QA manual. Documents examined were the ITT Barton audit reports of the Nuclear Research Center at Georgia Tech, the Southwest Research Institute and the Westinghouse-Nuclear Technology Division. In the QA manual and audit reports examined, no nonconform-ances or deficiencies were found.

2. The inspection team examined ITT Barton's QA manual for compliance to Section XVIII of Appendix B to 10 CFR 50, Audits. The inspection team then examined selected ITT Barton internal audits to their QA manual. The internal audits examined were receiving inspection, order administration, records, calibration control and program ~ audit. In the QA manual and internal audits examined, no nonconformances or deficiencies were found.

3. Four test programs relating to the Environmental Qualification (EQ) of ITT Barton equipment were examined. These programs were examined for the following:

a. Required test instrumentation with accuracies described, met the requirements of IEEE-STD-323/1974.

b. Equipment interfaces were addressed.

c. Test acceptance criteria was established as described in the test specifications or in the design engineering letters to meet the requirements of IEEE-STD-323/1974.

58

ORGANIZATION: ITT BARTON CITY OF INDUSTRY, CALIFORNIA REPORT INSPECTION NO.: 99900113/86-01 RESULTS: PAGE 3 of 5

d. The same equipment was used for all phases for testing and

, represented a standard production item.

e. Environmental conditions were described (e.g., pressure and temperature profiles, and thermal aging factors consistent with those outlined in the test specifications or test plan),

f. The test results were adequately reduced and evaluated against acceptance criteria described in the test specifications or purchase orders.

g. All prerequisites for the given test as outlined in the test specifications were met.

h. The test equipment included a description of all materials, parts, and subcomponents,

i. Notice of anomaly reports properly documented condition and reports were properly dispositioned. _

j. Appropriate margins were applied as required by test specifications.

1. Differential Pressure, Model 6001 and Pressure, Model 6005 transmitters.

Related EQ test documents (purchase orders, procedures, audits and final test reports) were examined. No noncon-formances or deficiencies were found.

2. Differential pressure and pressure switches, Model 580 series.

The EQ test documents (purchase orders, procedures, audits and final test reports) that were used for the 1983 ITT Barton qualification of Model 580 series were examined.

No nonconformances or deficiencies were found.

ITT Barton has recently ccmpleted additional EQ testing of these switches. During this testing three (3) of the instruments malfunctioned. A 10 CFR Part 21 notification dated April 14, 1986 was written stating that the malfunc-tions were generic in nature and that malfunctions were due 59

ORGANIZATION: ITT BARTON CITY OF INDUSTRY, CALIFORNIA REPORT INSPECTION NO.: 99900113/86-01 RESULTS: PAGE 4 of 5 to the failur:3 of Honeywell snap-acting switches which are a part of the a instruments. ITT Barton revised the 10 CFR Part 21 notificction on April 16, 1986 stating that deflec-tion of the instrument case may have caused the instrument malfunction by affecting the position of the switch actuating mechanism, which caused the switch set point to be shifted.

The inspection team examined the test data from.this recent EQ testing of the Model 580 series instruments and also discussed the test results with ITT Barton. It was con-cluded that the most probable cause of Model 580 series malfunction was the deflection of the instrument case although ITT Barton has nct ruled out the Honeywell switches as contributing to the instrument malfunctions.

ITT Barton plans to retest the Model 580 series instruments upon the completion of redesign of the instrument case.

3. Models 763 and 764, Electronic Transmitters.

Test reports, purchase orders, audits and procedures related te EQ testing of these transmitters were examined by the inspection team.

The inspection team questioned the disposition of one of the Notices of Anomaly. The anomaly was concerned with the erratic behavior of some of the transmit'. rs during the LOCA. The concern was identified in the review of the model 764 transmitter test report. During the testing, several anomalies were experienced with the transmitters, which were attributed to moisture and/or chemical spray permeating into the lead wire gland assembly and causing leakage currents between the two (2) leads. The leakage currents affected the accuracy of the transmitters output signals. ITT Barton dispositioned these anomalies as test methodology based on the observation of water flowing out of the wires outside the test chamber, and the fact that instrument leads would not be exposed to a differential pressure during actut.1 plant installation. The test setup had transmitter lead: penetrating the test chamber creating 60

ORGANIZATION: ITT BARTON CITY OF INDUSTRY, CALIFORNIA REPORT INSPECTION N0.: 99900113/85-01 RESULTS: PAGE 5 of 5 a differential pressure across the leads (LOCA pressure inside chamber and ambient outside of chamber). The inspection team was not provided adequate justification for ITT Barton's disposition of the anomalies as test methodology.

ITT Barton, after the inspection, provided additional test data to resc1ve this concern and justify the test methodology dispositicn of the anomalies. Additional tests performed by ITT Barton removed sections of the lead wires insulation between the transmitter and where the lead wires exited the test chamber. The test chamber was then pressurized and no leakage currents were observed. This test substantiated ITT Barton's position that the test anomalies were due to test methodology.

With the exception of the concern on the test anomalies disposition, the inspection team found no nonconformances or deficiencies related to documentation of E0 of ITT Barton, models 763 and 764, transmitters.

4. ITT Barton Models 352 and 353, Level Measuring Sensors.

The inspection team examined documents related to the EQ testing of the Level Measuring Sensors. Test reports, purchase orders and procedures were examined. No noncon-formances or deficiencies were found.

l t

61

i 0RGANIZATION: ITT HENZE MOBILE, ALABAMA REPORT INSPECTION INSPECTION NO.: 99901060/86-01 DATES: June 23-27, 1986 ON-SITE HOURS: 40 CORRESPONDENCE ADDRESS: ITT Henze ATTN: Mr. Mike Grosso, Vice President and General Manager 2970 Cottage Hill Road Mobile, Alabama 36606 ORGANIZATIONAL CONTACT: Jack Cawoon, Jr.

TELEPHONE NUMBER: (205) 476-7183 NUCLEAR INDUSTRY ACTIVITY: Valve Testing and Repair.

ASSIGNED INSPECTOR: e w) b27 54 J. C. Mrper, Reactive Irispection Section (RIS) &at'e OTHER INSPECTOR (S): None.

APPROVED BY: -

x 9 4 E. W. Merschoff, Chief S, Vendor Program Branch Date 4

INSPECTION BASES AND SCOPE:

l i A. BASES: 10 CFR Part 21 and 10 CFR Part 50 Appendix B.

B. SCOPE: The inspection evaluated the ITT Henze QA program and implementa-tion. Specifically, the inspection addressed Henze's involvement in the reported main steam safety valves with high lif t :et points at North Anna Unit 2.

i PLANT SITE APPLICABILITY: North Anna 50-338, 50-339; Perry 50-440, 50-441.

63 i .

}

ORGANIZATION: ITT HENZE MOBILE, ALABAMA REPORT INSPECTION NO.: 99901060/86-01 RESULTS: PAGE 2 of 9 A. VIOLATIONS:

1. Contrary to Section 21.21 of 10 CFR Part 21, ITT Henze failed to have a procedure pursuant to 10 CFR Part 21. (86-01-01)

2. Contrary to Section 21.31 of 10 CFR Part 21, ITT Henze has never imposed the requirements of 10 CFR Part 21 on applicable subtier vendors. (86-01-02)

B. NONCONFORMANCES:

1. Contrary to Criterion IX of Appendix B to 10 CFR Part 50 and Proce-dure SP-TNG-2-1, Revision 0 of the ITT Henze QA System Procedure, the Welder Qualification Test performed on February 27, 1984 by stamp no. FB was not witnessed by a QC representative. (86-01-03)

2. Contrary to Criterion IX of Appendix B to 10 CFR Part 50 and Proce-dure ITTHS-NDT-2 of the ITT Henze QA System Procedure, Henze did not administer examinations for distinguishing and differentiating con-trast between colors for their NDE inspectors. (86-01-04)

3. Contrary to Criterion XIII of Appendix B to 10 CFR Part 50 and NCA-4134.13 of Section III from the ASME Code, Henze failed to have written procedures or instructions regarding the storage and handling of welding materials intended for use in Nuclear safety related service. (86-01-05)

4. Contrary to Criterion VII of Appendix B to 10 CFR Part 50 and Section l II of the Henze Quality Assurance Manual QAM-40, Henze failed to have procedures to control the changes made to the Qualtiy Assurance Manual, QAM-40. (86-01-06)

5. Contrary to Criterion XVIII of Appendix B to 10 CFR Part 50 and Proce-dure ITTHS-QAV-3, Revision 0 of the Quality Assurance System Proce-j dures, no internal audit was performed on the ITT Henze Nuclear i

t Service Branch in 1984. No review of the February 1985 internal audit was performed by the Nuclear Services Branch Manager. And the l followup and rasponse to a February 1985 internal audit did not occur until late June 1985. (86-01-07)

8. Contrary to Criterion VII of Appendix B to 10 CFR Part 50 and Section NR-1210.17 of the Henze QAM-30, the following companies were employed by Henze and were on the QVL without being subjected to a source evaluation: Mobile Instrument Company in 1986; Welding Engineering l

64

l i

ORGANIZATION: ITT HENZE MOBILE, ALABAMA REPORT INSPECTION N0.: 99901060/86-01 RESULTS: PAGE 3 of 9 l

I Supply in 1986; Nondestructive Technical Services in 1986; Nobert Plating Company in 1986; and the Earle Jorgensen Company in 1984, 1985, and 1986. (86-01-08)

C. UNRESOLVED ITEMS:

None.

D. STATUS OF PREVIOUS INSPECTION FINDINGS:

None.

E. OTHER FINDINGS AND COMMENTS:

1. North Anna Main Steam Safety Relief Valves The NRC Vendor Program Branch inspection at ITT Henze was initiated in response to a VEPC0, North Anna Power Station Unit 2 (Licensee Event Report) LER 86-001-00 dated March 21, 1986. On February 21, 1986, eight of fif teen Unit 2 MSRVs failed to lif t during testing.

The unit was in Mode 3 (Hot Standby) and the licensee was performing periodic testing of the MSRVs in accordance with Section XI of the ASME Boiler and Pressure Vessel Code. The test was conducted using a pneumatic assist test device in addition to the pressure in the main steam lines. The pneumatic assist test device was used in accordance with instructions supplied by the valve manufacturer, Crosby Valve and Gage Company. Following a series of tests (refer to NRC Region II Inspection Report Nos. 50-338/86-15 and 50-339/86-15 for details on the sequence of testing) which determined that there were an excessive number of inoperdble valves, the licensee elected to retest the valves utilizing a steam pressure system. Wyle Laboratories was contracted to perform the testing and refurbishment of the valves.

At Wyle Laboratories, testing was repeated on each valve until three successful successive test results were obtained. Nine of the ten valves needed adjustment, since the valves were out of their setpoint ranges on the high side. The remaining valve was found to be acceptable. Subsequently, VEPC0 sent the remaining five valves to Wyle Laboratories for retesting. Three of those five valves were also set on the high side. All fifteen MSRVs were reset to their proper ranges, refurbished, leak tested, and returned to the licensee for reinsta11ation. In addition, Wyle Laboratorics found that 12 of 15 guide ring setpoints varied from +135 to +48 notches and nozzle

> ring setpoints varied from -21 to -3. The valve manufacturer Crosby 65

ORGANIZATION: ITT HENZE MOBILE, ALABAMA REPORT INSPECTION NO.: 99901060/86-01 RESULTS: PAGE 4 of 9 i

Valve and Gage Co. recommends that guide rings are to be set at +150 notches and the nozzle ring at -20 notches.

Henze North Anna Steam Safety Relief Valve Tests Performed by ITT Henze During September 1984, the licensee contracted the services of ITT Henze to perform valve testing, utilizing the steam pressure method.

Testing and repairs were performcd and completed in September and October of 1984. It was reported by Henze that of the fifteen valves, two were within the setpoint range. The other valves all failed on the low side and were reset by Henze to within the proper setpoint range. Henze increased setpoints on the valves from 24 to 119 psig. The guide and nozzle ring setpoints were all set by Henze to the manufacture specifications of +150 and -20, respectively.

During this inspection, Safety-Relief Valve Reports for the 15 subject valves and the Gauge Calibration Certification used in testing the valves were examined. Review of the Safety-Relief Valve Reports indicated that a VEPC0 representative had witnessed each test. The Safety-Relief Valve Report provides data on the condition as received, work performed as well as a Pretest Report (pressure), test medium and set pressure. According to Henze, 13 out of 15 valves received from VEPC0 were not within ASPE Code tolerance. The safety valves listed below were pre-tested, repaired, and set in the Henze Norfolk plant under the guidance of the Nuclear Branch and VEPC0 0C Rep.

Repaired Sit As Found Leaked Reset at Valve # Pressure (psi) (pretest (psi) at (psi) (psi) 201-C 1085 1093 150 1085 202-A 1095 640 150 1095 203-B 1110 1086 175 1110 204-C 1120 1071 205-A h/A 1120 1135 1112 100 1135 201-A 1085 1046 N/A 1085 202-B 1095 1060 175 1095 203-C 1110 1073 N/A 1110 204-A 1120 1056 150 1120 205-B 1135 1020 200 1135 201-B 1085 1036 100 1085 202-C 1095 1041 350 1095 203-A 1110 1046 100 1110 204-B 1120 1022 200 1120 205-C 1135 1135 1016 1135 '

66

l ORGANIZATION: ITT HENZE MOBILE, ALABAMA REPORT INSPECTION NO.: 99901060/86-01 RESULTS: PAGE 5 of 9 The valve guide and nozzle rings for the valves listed below were all inspected and set to marufacture specifications of +150 notches and -20 notches respectively.

As Found As Reset to Valve # NR GR NR GR 205A -4 +132 -20 150 205C -14 +157 -20 150 204A -3 +148 -20 150 204C -20 +48 -20 150 2038 -20 +50 -20 150 202C -16 +145 -20 150 201A -20 +175 -20 150 ITT Henze has detailed test criteria for testing main steam safety valves. Among some of the written practices are: calibration of service gauges and instrumentation prior to test (pressure gauges are to be calibrated within a tolerance of i 1/4 of 1%); three consecutive tests shall be completed satisfactorily according to the stamped set pressure of each valve; repairs are to be authorized by the vendor; the vendor shall witness each valve test to verify that the setpoint acceptance criteria are met.

The Heise gauge used for testing (model number 710-A, SN S-7-5158) was calibrated and supplied by the Henze Nuclear Branch, used in Norfolk for testing and then returned to the Nuclear Branch for a postfield service calibration check.

The Heise model 710-A, serial number S-7-5158 was calibrated on 5/26/84, 10/9/84 and 11/11/84, traceable to NBS test P7239. There were no adjustments necessary when comparing the initial and adjusted measured values to nominal values. The unit had an accuracy better than 1/4 of 1%. Instrument Control Service performed the calibration, and a certificate of calibration was submitted for each calibration l performed.

From the records reviewed at ITT Henze, it appeared that:

i 1) The Heise gauge used in testing of the MSRVs was calibrated and found to be within the rated accuracy of i 1/4% before and after the testing.

2) All valves were repaired / reset at the original set pressure as witnessed by a VEPC0 representative.

67

ORGANIZATION: ITT HENZE MOBILE, ALABAMA REPORT INSPECTION NO.: 99901060/86-01 RESULTS: PAGE 6 of 9

3) Guide and nozzle ring guides were preset to manufacturers speci-fications of +150 notches and -20 notches respectively.

In general, the valves were repaired and set in accordance with the ITT Henze Service Quality Control Procedures ITTHS-VR-2 and ITTHS-PT-1, which conforms to ASME requirements.

10 CFR Part 21 The ITT Henze shop area was inspected for the posting of Section 206 of the Energy Reorganization Act of 1974, a Part 21 procedure and a Part 21 notice. The Part 21 notice was posted, however, Section 206 and a Part 21 procedure were not. ITT Henze confirmed not having a Part 21 procedure (see violation 86-01-01). From examination of the document packages it was determined that Part 21 was imposed on ITT Henze by their nuclear customers. At the time of the inspection, Henze did not impose the requirements of Part 21 on contractors employed to fill their nuclear orders.

For example, 10 CFR Part 21 was imposed on ITT Henze by CEl, P0 Q-4982-66. To fill the CEI order, ITT Henze contracted and obtained the services of Cabot P0 #110224,1/2/85; Nobert Plating Company P0 #110461; 5/13/85, and ND Technical Services P0 #03062, 11/28/84. ITT Henze did not impose the requirements of 10 CFR Part 21 on any of these contractors (see violation (86-01-02).

Quality Assurance Records Upon review of the ITT Henze QA Manual, there was confusion regarding revision updates. Review of the context of the QA Manual revealed that revisions denoted by a symbolic notation in the left margin of the text. However, those revision changes or symbols were not clearly described on the cover page or in a directive exPl aining the use of the QA Manual. No instructions were given on how manual revision changes are identified. The cover page of the QA Manual was documented as Revision 2, however, some parts of text of the manual were docu-mented with unidentified symbols (Revision 3). Since the QA manual had no directive to inform the reader of updated revisions, references to a prior revision may be misleading. Furthernre, this lack of control of the QA manual revisions may lead to outdated manual revisions being inadvertently used in the field as well as in the field offices. (see nonconformance 86-01-05) 68

OPGANIZATION: ITT HENZE MOBILE, ALABAMA REPORT INSPECTION NO.: 99901060/86-01 RESULTS: PAGE 7 of 9 Special Processes Welder qualifications and welding specifications were evaluated. All of the welder qualifications were satisfactory with the exception of Welder FB. Welder FB qualified according to WPS No. 21, Rev. O, Laboratory-Test Number-2906, conducted by Nondestructive Technical Services. This test was invalid since there was no objective evidence on the WQR that the test was witnessed by an ITT Henze Quality Control Representative (see nonconformance 66-01-03).

The NRC inspector evaluated the welding and welding material storage and handling area. The welding storage area was locked, clean and orderly. However, there were no written procedures or instructions to control the storage and handling of welding materials intended for use in Nuclear safety related service (see nonconformance 86-01-05).

Since there was no procedure, there were loose welding electrodes

, laying around the welding machine, no guidelines for distributing welding materials and no temperature range limits established for handling low hydrogen coated electrodes. There were two welding rod ovens, one was locked in the control room and the other was on the shop floor by the welding machine. It is obvious that the oven in the lock area was for nuclear welding electrodes. However, the use of the oven on the floor was not obvious since it was not marked as " Nuclear" or "Non Nuclear."

No written procedure or instructions addressed the method of welding rod identification within a rod oven, or a storage temperature range for low hydrogen coated electrodes, or identification of the rod ovens as " Nuclear" or "Non Nuclear." At the time of the inspection the E7018 low hydrogen coated electrodes were being stored at 250 F. A temperature of 250 F is the minimum desirable temperature that these electrodes should be stored, in order to ensure minimum moisture absorption.

Eight NDE inspectors qualification records were examined, seven Level II and one Level III. All of the qualification records examined were satisfactory according to ASME, SNT-TC-1A and ITT Henze require-ments with the exception of the requirement for an annual eye examin-ation. Testing for the capability of distinguishing and differen-tiating contrast between colors was not policy at Henze and therefore was not implemented for Henze NDE inspectors on an annual basis (see nonconformance 86-01-04).

I f

69 l

l

ORGANIZATION: ITT HENZE MOBILE, ALABAMA REPORT INSPECTION N0.: 99901060/86-01 RESULTS: PAGE F of 9 Audits Internal Audits Internal audit records were reviewed for the years 1983-86 for audit completeness and quality. All audits were performed by qualified personnel not directly responsible for the work performed. These audits were performed according to the ITT Henze Quality Assurance System Procedure ITT HS-QAV-3, Rev. 0, NR-1210-17, Rev. O.and Quality Assurance Manual-40 Baseline Manual Section XVII-AUD. However, there were a few problems in the implementation of these procedures. There was no evidence that a 1984 internal audit was ever performed (see nonconformance 86-01-07). The nuclear service branch manager was committed to reviewing internal audits by signature approval. No such signature approval was found on a February 13-15, 1985 internal audit of ITT Henze's QA system (see nonconformance 86-01-07). Section XVII-AUD of the ITT Henze QAM-40 and ITTHS-QAV-3 does rot specify a time frame for corrective action of an internal aucit finding. How-ever, Section XVII-AUD does require an immediate resolution to internal audit deficiencies.

Corrective action for the findings of the internal audit of February 13-15, 1985 was not responded to until June 26, 1985. From review of the internal audit the NRC inspector noted that a 30 day response was requested by the auditor performing the internal audit (see nonconformance 86-01-07). Therefore, it can be concluded that the required inmediate resolution referenced in the QAM-40 should have been submitted within 30 days.

External Audits External audits were to be performed according to ITT Henze QAM-30, Section NR-1210.17 and XVII-AUD QAM-40. It is the policy of ITT Henze to perform annual audits of all subcontractors of special processes prior to updating the Nuclear Service Branch QVL. There were several ccmpanies that ITT Henze employed for nuclear power plant l work and had placed on the QVL without being subjected to an annual audit by a Henze quality representative. The Mobile Instrument Company supplied Henze with calibration services and was on the Henze QVL in 1986 without being subjected to the scheduled audit for that year.

The Welding Engineering Supply Company, suppliers of welding electrodes in 1986 was on the QVL and was not subjected to the scheduled audit for that year. Nondestructive Technical Services, suppliers of welder l

l 70

ORGANIZATION: ITT HENZE MOBILE, ALABAMA REPORT INSPECTION N0.: 99901060/86-01 RESULTS: PAGE 9 of 9 Qualification Services in 1986 was not subjected to the scheduled audit for that year. Norbert Plating, suppliers of silver plating on seal rings in 1986 was not subjected to the scheduled audit for that year. Finally, the Earle Jorgensen Company, material suppliers was on the QVL and not subjected to the scheduled audit for the years 1984, 1985 and 1986. (See nonconformance 86-01-08.)

F. PERSONS CONTACTED:*

Name Ti tle Company

1. Clint Rose National Nuclear Coordinator ITT Henze

2. Wayne Prokop Nuclear Branch Manager ITT Henze

3. Mike Grosso General Manager ITT Henze

4. Jack Cahoon, Jr. Technical Director and Quality ITT Henze Manager l *Present at the June 27, 1986 Exit Meeting.

i 71

ORGANIZATION: L. B. F0 STER COMPANY COPHERCE, CALIFORNIA REPORT INSPECTION INSPECTION NO.- 49901061/86-01 DATES; 7/28 79/R6 1N-SITF H0llDRt Q CORRESPONDENCE ADDRESS: L. B. Foster Company ATTN: Mr. David Foltz Associate General Counsel 415 Holiday Drive Pittsburgh, Pennsylvania 15220 ORGANIZATIONAL CONTACT: Ms. Cindy Turner, District Administrator TFIFPHONF NilMRFRt (7191 777 7191 NUCLEAR INDUSTRY ACTIVITY: Supplier of pipe and fittings.

ASSIGNED INSPECTOR: M 8 1T-$(.

ie Inspection Section (RIS) Date J.}.Conway, React [V

. V APPROVED BY: .

If-2 5 W E. W. Merschoff, C , RIS, Vendor Program Branch Date INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 50, Appendix B and 10 CFR Fart 21.

B. SCOPE: This inspection was made as a result of the receipt of an allega-tion pertaining to defective pipe fabricated by L. B. Foster (LBF) and shipped to a nuclear facility.

l l

PLANT SITE APPLICABILITY: Pipe and fittings - Palo Verde (50-528/529/530).

l 73 l

l ~ - - - .___

-s0RGANIZATION: L. B. FOSTER COMPANY COMMERCE, CALIFORNIA

'l REPORT INSPECTION-N0.: 99901061/86-01 RESULTS: PAGE 2 of 4

, A. VIOLATIONS:

, None.

B. NONCONFORMANCES:

None.

C. UNRESOLVED ITEMS:

None.

D. OTHER FINDINGS AND COMMENTS:

1. Allegation In May 1986, the NRC was notified of an allegation made to the Atomic Industrial Forun (AIF) relating to nuclear pipe fabricated by LBF in Hayward, California. In a May 7,1986 letter to the AIF, it was stated that LBF produced "... pipe for atomic energy plants, including

[ supports] for the main and auxiliary cooling pumps." In a July 15, 1986 telephone conversation with the NRC inspector, the alleger indicated that the pipe could be defective due to the falsification of radiographs of the pipe. It was also revealed that the pipe was fabricated in the 1973-1977 time frame and was sent to Bechtel and three nuclear power plants. The alleger could not recall the names of any of the plants, but he thought one of the plants was in California.

In a July 16, 1986 telephone conversation with Mr. G. Stokes, the Area Manager of the Hayward, California office of LBF and an employee of LBF for the past 11 years, it was revealed that the only activity ongoing at the Hayward office was supplying steel pipe, pilings, and railroad rails. The fabrication facility at Hayward was sold to Lincoln Property Company approximately one year ago. LBF never had an ASME N stamp or a Certificate of Authorization (Materials). The majority of pipe fabricated at Hayward was to American Waterworks Association Standards which required some radiographic testing (RT).

RT was performed by service vendors such as Q.C. Services and Pittsburgh Testing. Mr. Stokes indicated that the only material supplied by LBF for a nuclear facility was pipe in the late 1970s for Bechtel (Palo Verde) and steel rail in 1981 to 1983 for J. A. Jones (WNP-2). Recctds for the Bechtel and J. A. Jones material were in the Commerce, California and Federal Way, Washington offices, respectively.

74

ORGANIZATION: L. B. F0 STER COMPANY COMMERCE, CALIFORNIA REPORT INSPECTION NO.: 99901061/86-01 RESULTS: PAGE 3 of 4 The NRC inspector contacted Mr. W. Huddleston, president of Q.C.

Testing in Portland, Oregon. Mr. Huddleston, who owned Q.C. Services in Hayward, California until he sold the company in 1985, indicated that RT was performed for LBF in the 1970s by Q.C. Services, but the items examined were not for a nuclear facility.

Af ter contacting Mr. S. Heisler, Manager of Bechtel's Supplier Quality Department in San Francisco, California, Pechtel researched both their Office Register and Field Register and noted that they ordered material from LBF in the late 1970 time frame. All of the items were "non-safety related" fittings and pipe (f 2 inch) for the Palo Verde nuclear facility.

During a conversation with Mr. J. Hornack on July 28, 1986 and a review of records at LBF's fccility at Commerce, California, it was noted that LBF is currently a marketing organization specializing in rail and track accessories, pipe, piling, foundation construction products, construction equipment and highway products. Mr. Hornack is currently a consultant to LBF but was the Manager of the Los Angeles divisional office in the 1970s. Mr. Hornack indicated that the only material supplied by LBF to a nuclear facility was on a blanket purchase order (P0) in the late 1970s from Bechtel/ Arizona Public Service (APS).

The NRC inspector reviewed P0 No.10407-13-PM-308 " Quality Class R-Non-Nuclear Service Pipe and Fittings 2 inches and smaller," Revision 0, dated Iday 21, 1976, from Bechtel/APS. The P0, including subsequent revisions thru No. 9 dated June 15, 1978 was for approximately 1080 items. All of the items were ordered to a specific ASTM standard which included A-106 (carbon steel pipe); A-312, A-335, and A-376 (stainless steel pipe); A-105 (carbon steel fittings); and A-182 and A-403 (stainless steel fittings). The P0 stated that the items were to be furnished in accordance with Material Requisition (MR) No.13-PM-308. Appendix No. 4Q, an enclosure to the MR, addressed " Supplier Quality Class R Programs," but a copy of this appendix could not be i

located in the LBF records. LBF assigned release numbers (identified l as CPOR) to each item or group of similar items on the P0 and main-l tained individual files for each CPOR. The NRC inspector reviewed approximately 70 CPOR files. The majority of the files reviewed were complete and consisted of a Sales Order write-up, Sales Inquiry (contained quotes from manufacturers / suppliers), Bill of Lading, LBF purchase requisition (PR) to suppliers, certified material test reports l from manufacturers, calculation sheets for price escalation, and a freight bill. LBF prs refercnced the applicable ASTM standard but 1

75

. ORGANIZATION: L. B. F0 STER COMPANY COMMERCE, CALIFORNIA REPORT INSPECTION NO.: 99901061/86-01 RESULTS: PAGE 4 of 4 did not impose any quality requirements (e.g., QA Program) on the suppliers. Suppliers included, among cthers, Capitol Pipe, Tube Sales, Kilsby Tube, Cloris Pipe & Supply, and Gulfalloy. It was noted that all the items were drop shipped at the Palo Verde nuclear site beginning in 1977 and continuing thru January 1983.

Bechtel/APS P0 No. 10407-F-125094-H0 dated September 2, 1980 was also reviewed. This order was for 1800 ft of 1" pipe (ASTM A106) designated as " Quality Class R". The P0 stated, "An approved Vendor Quality Assurance Program in accordance with 13-PM-308 is required for this Procurement." The supplier was Quanex Specialty Tubing, and Astro Pak Specialized Services cleaned and capped the pipe. LBF prs No. 66-2368 and No. 66-2257 to Quanex and Astro Pak, respectively, did not specify a QA program.

It was noted that Bechtel Specification Change Notice No.1855 dated August 31, 1979 to Specification No.13-PM-308 added a requirement for the supplier to have a QA Program. During the conversation with Mr. Hornak, the NRC inspector was told that the Los Angeles Division (LAD) which subsequently became the Commerce, California office never had a QA Program or manual. He also stated that the LAD was never audited / surveyed by Bechtel or APS.

During the review of records at LBF, a type written copy of a " Quality i Assurance Manual," Revision 0, dated September 11, 1979 was found in l the files. Tha document consisted of two pages and addressed QA Personnel Responsibilities, Nonconformances, Procurement, and Quality Surveillance; but the document was not signed by any LBF personnel l (e.g., District Sales Manager).

Based upon the review of records at LBF's facility in Commerce, California and discussions with the individuals noted above, it was i determined that LBF has not supplied any safety-related material to l a nuclear facility. Subsequently, the allegation could not be

! substantiated.

, 2. Personnel Contacted Joe Hornack, Consultant Cindy Turner, District Administrator 76

1 l l

ORGANIZATION: NAMCO CONTROLS, INC.

MENTOR, OHIO REPORT INSPECTION INSPECTION NO.: 99900378/86-01 DATES: 7/21-24/86 ON-SITE HOURS: 48 CORRESPONDENCE ADDRESS: Namco Controls, Inc.

ATTN: Mr. N. E. Swanson President 7567 Tyler Boulevard Mentor, Ohio 44060 ORGANIZATIONAL CONTACT: Mr. Douglas A. Coe, Sales Application Engineer TELEPHONE NUMBER: (216) 946-9900 NUCLEAR INDUSTRY ACTIVITY: Supplies safety-related limit switches and conducts environmental qualification (EQ) testing of its designated limit switches for the commercial nuclear power industry and the military. All of the EQ testing is for commercial nuclear power. Approximately 10 percent of its manufactured products are for commercial nuclear. en-ASSIGNED INSPECTOR: 5NM D.)2/ad /6X/Tff, R. N. Noist,' Eqdipment Qualification Section (EQIS) Date j

OTHERINSPECTOR(S): M. Jacobus, Sandia National Laboratories 1

- [

APPROVED BY: LllMI., b bG-p\p S-NE U. Potapovs, Chief, EQISa Vendor Program Branch Date INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 50, Appendix B and 10 CFR Part 21.

B. SCOPE: This inspection consisted of: (1) Technical evaluation of equip-ment qualification (EQ) test activities for safety-related equipment (2) verification of implementation of corrective action (CA) on the noncon-formance identified in the NRC Inspection Report No. 99900378/63-01,(3) verification of implementation of the quality assurance (QA) program and (continued on next page)

PLANT SITE APPLICABILITY: Various.

77

ORGANIZATION: NAMC0 CONTROLS, INC.

MENTOR, OHIO REPORT INSPECTION N0.: 99900378/86-01 RESULTS: PAGE 2 of 7 SCOPE: (continued) (4) followup on a 10 CFR Part 21 report issued by Vermont Yankee relating to a possible defect of contact block kits supplied by Namco Controls, Inc. (NCI).

A. VIOLATIONS:

None.

B. NONCONFORMANCES:

None.

C. UNRESOLVED ITEMS:

None.

D. STATUS OF PREVIOUS INSPECTION FINDINGS:

(Closed) Nanconformance (83-01, Item B.ll: The QA program was not supported by written policies, procedures, or instructions. Namco has not issued implementing procedures NSP20-0005, -0006, -0007, -0008, and NSP 60-0007.

The NRC lead inspector verified that the material and intent of NSP 20-0005 had been incorporated into other procedures and that procedure NSP20-0008 had been compiled and implemented during a previous inspection documented in Inspection Report No. 99900378/84-01.

The NRC lead inspector during this inspection verified that procedures NSP20-0006 and NSP20-0007 titled " Reproduction and Distribution of Design Documentation" and " Preparation of Test Plans (QTP) and Test Reports (QTR)," respectively, had been compiled and implemented. Procedure NSP60-0007 titied " Inspection Procedure" referenced in procedure NSP40-0003 dated January 1983 titled " Processing Factory Orders and Purchase Orders" has been obsoleted. Information relating to inspection and identification l of material are now compiled in procedures NSP40-0003, Revision A, dated May 1984 titled same as above, NSP60-0010, Revision B, dated March 1984 titled " Identification and Control of Products Dedicated for Use in Nuclear Power Plant Applications" and Sections 8 and 10 of the QA manual titled

" Identification and Control of Paterials, Parts and Components" and

" Inspection," respectively. The NRC inspector verified' implementation

! of these procedures.

1 78

ORGANIZATION: NAMC0 CONTROLS, INC.

MENTOR, OHIO I

REPORT INSPECTION NO.: 99900378/86-01 RESULTS: PAGE 3 of 7 E. OTHER FINDINGS OR COMMENTS:

1. Technical Evaluation - The NRC inspector and Sandia Consultant (NRC i inspection team) conducted a technical evaluation of four completed ,

l programs and one ongoing program for qualification testing of safety-related electrical equipment. The following table summarizes the programs examined, equipment type, plant and types of documents exam-l ined.

Program Equipment Type Plant Documents Examined

1. EA740 Limit Switch with Various Qualification Test Report one inch receptacle and (QTR), Cualification Test connector / cable assembly. Plan (QTP),TestProce-dure (TP), Test Data (TD) and Purchase Orders (P0).

2. EA180 Limit Switch with Various QTR, QTP, TP, TD and P0 one inch receptacle and connector / cable assembly.

3. EA180 Limit Switch Various QTR

4. Half inch receptacle and Various Similarity Analysis Qual-connector / cable assembly ification Package

5. Material analysis of Various Test Data impregnation materials used for qualified limit switches.

The NRC inspection team reviewed the EQ process prescribed in each test report / test plan / procedure and reviewed test results, including the bases for accelerated thermal aging and radiation and verified calculations when applicable.

Four test programs and one ongoing program and related engineering

[ documentation were examined for the following as applicable:

a. Adequate test instrumentation and their accuracies were described and used to meet the requirements of IEEE-STD-323/1974.

b. Equipment interfaces were addressed.

79 1

ORGANIZATION: NAMC0 CONTROLS, INC.

MENTOR, OHIO REPORT INSPECTION NO.: 99900378/86-01 RESULTS: PAGE 4 of 7

c. Test acceptance criteria were established as described in the test specification or in the design engineering documents, such as calculations and engineering letters to meet the requirements of IEEE-STD-323/1974.

d. Same equipment was used during all phases of testing and repre-sented a standard production item.

e. Environmental conditions were established and described (e.g.,

pressure and temperature profiles, and thermal aging factors were consistent with those outlined in the test specification or test plan).

f. Test results were adequately reduced and evaluated against esta-blished acceptance criteria described in customer test specifica-tions or purchase orders.

g. All prerequisites for the given tests as outlined in the test specification had been met.

h. Test equipment included a description of all materials, parts and subcomponents.

i. Notice of Anomaly reports were properly documented.

l

j. Appropriate margins were applied.

k. Functional performance requirements met.

Programs 1 and 2 - EA-180/EA-740 Limit Switches including a receptacle and one inch connector / cable assembly were qualified for inside containment applications in accordance with QTR-130 and QTR-140, respectively.

t l Program 3 - EA-180 Limit Switch was qualified for inside containment l

applications in accordance with QTR 105 as long as a qualified conduit sealing method was employed.

Program 4 - Half-inch receptacle and connector / cable assembly was qualified based on similarity to the series one-inch connectors. The lead wires on the half-inch and one-inch connectors are identical.

The cable for the half-inch connector cable assembly was qualified l by similarity to the same test report as referenced for the one-inch connector cable qualification.

80

ORGANIZATION: NAMC0 CONTROLS, INC.

MENTOR, OHIO REPORT INSPECTION N0.: 99900378/86-01 RESULTS: PAGE 5 of 7 Program 5 - This program was started as a result of an audit conducted by Namco on their niaterial vendor. It was discovered that the vendor was using a different impregnation material than that specified in Namco's process drawing. As a result, Namco tested housings with the new impregnated material to verify that it is functionally equivalent to the required material. In addition, Namco discovered that in some cases the vendor mixed the new and the specified material together prior to impregnating the housing. Currently Namco is performing tests for mixtures of the two materials. The testing includes thermal shock, leak testing, thermal aging, radiation exposure and design basis event testing although not all on the same specimen. The NRC inspection team will review test results during a future inspection.

No nonconformances were identified during the above review.

2. Followup on Two 10 CFR Part 21 Reports One of the 10 CFR Part 21 reports addressed contact block kits that were supplied to Vermont Yankee Nuclear Power Corporation. The NRC inspection team visually inspected contact block kits that were returned to Namco by Vermont Yankee. The contact blocks had large chunks of material broken off at various locations on the sides of the block. The NRC inspector reviewed Namco's Inspection Report (IR) dated May 23, 1983 for that lot of contact blocks which were shipped to Vermont Yankee. The IR showed that no defective contact blocks were shipped from Namco as denoted by acceptance stamp by Namco's quality inspection personnel.

Further discussions with Namco personnel revealed that an inspector from Vermont Yankee performed an audit of the shipping activities relating to replacement contact blocks at the Namco's manufacturing plant at Newton, North Carolina recently. While,there, the Vermont Yankee inspector visually verified that the replacement contact blocks were shipped from the Newton Plant in acceptable condition. However, l the Vermont Yankee inspector discovered at Vermont Yankee's receiving

! inspection af ter opening the packages that some contact blocks had chips / chunks broken off at various locations on the side of the contact block. It was determined by Namco that thermal shock due to j temperature change during the airline flight caused some of the l

contact blocks to chip. Namco placed contact blocks into dry ice

for twenty-four hours at their manufacturing plant and discovered

( that some of the contact blocks exhibited the same condition mentioned l

-bove.

81

ORGANIZATION: NAMC0 CONTROLS, INC.

MENT 0R, OHIO REPORT INSPECTION NO.: 99900378/86-01 RESULTS: PAGE 6 of 7 Namco shipping practices and potential customer notifications will be reviewed during a future inspection at the Newton manufacturing plant.

l The second 10 CFR Part 21 report addressed cracked contact lever arms discovered at Niagara Mohawk's Nine Mile Point in qualified EA-180 i limit switches. The NRC inspection team visually inspected contact I lever arms (old vintage material) sent to Nanico from Nine Mile Point

, and observed the cracks. The riveting process which secures the l contacts to the lever arm is suspected to cause the crackc. Namco l is initiating a program to test the returned contact lever orms in order to assess the significance of these defects. The details of the test programs were not formalized prior to completion of the

! inspection.

Discussions with Namco engineering personnel revealed that the cracking was not limited to the old vintage material, used prior to 1980, but was also observed in new vintage material in use af ter 1980.

l Currently, Namco is conducting a life test on a production contact lever arm (new vintage material) found defective during inspection at Namco's manufacturing plant to determine if the cracks can cause failure of the contact lever arms, and if so, af ter how many normal

! operations.

l A followup inspection will be conducted at the Namco Manufacturing l Plant at Newton, North Carolina to review implementation of their QA program, in areas of process control, in-process and final inspection, and shipping practices.

3. QA Program Implementation Review - The NRC inspection team reviewed five program files containing documentation which supported testing efforts. Documentation included P0's, EQ test plans and procedures, l test reports and test data. The inspection teams review of the above i documentation was to verify continued implementation of the Namco QA program. No nonconformances were noted during this review.

4. QA Manual Review - The NRC team leader reviewed the changes that were l

incorporated into the different sections of both the QA Manual and Namco Standard Practice Manual (implementing procedures) since the last inspection. The NRC team leader determined that the changes were minor and did not change Namco's QA program with respect to the requirements of Appendix B to 10 CFR Part 50.

l 82

ORGANIZATION: NAMC0 CONTROLS, INC.

MENTOR, OHIO REPORT INSPECTION NO.: 99900378/86-01 RESULTS: PAGE 7 of 7

5. Followup on a Licensee Event Report - The NRC team leader followed up on Licensee Event Report (LER) 094 issued by LaSalle County Station Unit 1 on February 7,1985. It stated that the operating lever arms of Namco EA 180 switches could not be sufficiently tightened on the splined switch lever shafts, when the threaded expansion plug was screwed into the end of the shaft. Review of documentation and discussions with Namco revealed that Namco notified each customer to whom they had shipped EA 180 series switches since May 1, 1983 of the potential problem. Also, Namco supplied their customers with a replacement, non-galling expansion plug for each switch, however, LaSalle County Station Unit One was not on the notification list. The NRC team leader determined the limit switch was not procured directly from Namco by LaSalle County Station and was therefore not included in the Namco notification.

l l

l l

! 83

ORGANIZATION: NOVA MACHINE PRODUCTS CORPORATION MIDDLEBURG HEIGHTS, OHIO REPORT INSPECTION INSPECTION NO.: 99901052/86-01 DATES: 7/14-18/86 ON-SITE HOURS: 64 CORRESPONDENCE ADDRESS: NOVA Machine Products Corporation ATTN: Mr. Paul Novosel President Post Office Box 30287 Middleburg Heights, Ohio 44130 ORGANIZATIONAL CONTACT: Jim Fitzwilliam, QA Manager TELEPHONE NUMBER: (216) 267-3200 NUCLEAR INDUSTRY ACTIVITY: Material supplier of fasteners and special order material.

ASSIGNED INSPECTOR:  ! h E. T. Baker, Reactive Inspection Section, (RIS)

M g Dste OTHER INSPECTOR (S): Ronald Dahl, o and Associates APPROVED BY: .

• 7 E. Merschoff, Chieff RIS, Vendor Program Branch ate INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 50, Appendix B; 10 CFR Part 21; and NCA-3800.

! B. SCOPE: This inspection was made as a result of information received from concerning NOVA's use of an unqualified material supplier and questions regarding the applicability and implementation of 10 CFR Part 21.

1 PLANT SITE APPLICABILITY:

l l

10 CFR 2.790 INFORMATION HAS BEEN DELETED 85

ORGANIZATION: NOVA MACHINE PRODUCTS CORPORATION MIDDLEBURG HEIGHTS, OHIO REPORT INSPECTION N0.: 99901052/86-01 RESULTS: PAGE 2 of 7 A. VIOLATIONS:

None.

B. NONCONFORMANCES:

1. Contrary to Criterion II of Appendix B to 10 CFR Part 50 (Appendix B), subparagraph NCA-3864.2 of Section III of the ASMC Boiler and Pressure Vessel Code (the Code), and paragraph 3.2 of NOVA's Quality Assurance Procedure (QAP) NP-RSHS1, a warehouseman employed in that capacity since 3/27/86 had not received the required indoctrination training. (86-01-01) e

2. Contrary to Criterion V of Appendix B and subparagraph NCA-3866.1 of the Code:

a. Neither NOVA's Quality Assurance Manual (QAM) nor QAP NP-11 provided instructions on how many pieces were to be in-spected, what dimensions were required to be checked, or what the acceptance / rejection criteria are for receipt inspection. (86-01-02)

b. Neither NOVA's QAM, QAP NP-12, nor the sequence sheet provide instructions on what dimensions are required to be inspected, what the acceptance / rejection criteria are or a reference to the applicable standard. (86-01-03)

3. Contrary to Criterion VI to Appendix B, subparagraph NCA-3866.2 of the Code and paragraphs 8.0 and 8.2 of NOVA's QAM:

a. On sales order 1515 the QA Manager, during his review, failed to detect that NCA-3800 had been incorrectly listed as a requirement on the sales order. (86-01-04)

b. Neither NOVA's QAM nor their QAPs require that the sequence ,

sheet be approved prior to issuance to the shop. (86-01-05)

c. Neither NOVA's QAM nor their QAPs provide controls over changes to the sequence sheet. (86-01-06)

4. Contrary to Criteria VII and XVIII of Appendix B, subparagraph NCA-3861(a)(3) of the Code, and paragraphs 3.0 and 5.2 of NOVA's QAM:

10 CFR 2.790 INFORMATION HAS BEEN DELETED l

86

ORGANIZATION: NOVA MACHINE PRODUCTS CORPORATION MIDDLEBURG HEIGHTS, OHIO REPORT INSPECTION RESULTS: @ PAGE 3 of 7 N0.: 99901052/86-01

a. NOVA could not provide t. checklist for the 1985 audit of (

(86-01-07)

b. The audit checklist used by NOVA does not cover the posting, procedural, or reporting requirements of 10 CFR Part 21 l when Part 21 is imposed on the subtier supplier by NOVA.

(86-01-08)

5. Contrary to Criterion VIII of Appendix B, subparagraph NCA-3866.6 of the Code and paragraph 6.2 of QAP NP-03, NOVA has r.ot stamped all individual pieces of bar stock 1" in diameter or larger.

(86-01-09)

6. Contrary to Criterinn XV of Appendix B, subparagraph NCA-3867.3 and paragraphs 12.0 and 12.1 of NOVA's QAM:

a. Paragraph 4.1.4.1.1 of QAP NP-11 and paragraph 3.1.1.1 of QAP NP-VP permit NOVA to not complete a nonconformance report on incoming nonconforming material when the nonconformance is minor and can be reworked by NOVA. (86-01-10)

b. Neither Section XII of the QAM nor QAP NP-05 provide controls l for the reworking of nonconforming material. (86-01-11) l

c. Neither the QAM nor the QAPs provide instructions on how nonconforming material returned by a customer will be controlled, documented, or dispositioned. (86-01-12)

7. Contrary to Criterion XVIII of Appendix B and subparagraph NCA-3867.2 of the Code, NOVA's QAM, QAPs, and sequence sheets do not require the recording of the number of pieces inspected or the number of pieces accepted or rejected. (86-01-13)

8. Contrary to Criterion XVI of Appendix B, subparagraph NCA-3879.2 and paragraph 12.2 of NOVA's QAM, the QA Manager had closed nonconformance/ corrective Action Report #26 on 6/2/86 indicating that corrective action had been completed, when in fact as of 7/16/86, corrective action had not been completed. (86-01-14) ,

C. UNRESOLVED ITEMS:

None.

10 CFR 2.790 INFORiiATION HAS BEEN DELETED 87

ORGANIZATION: NOVA MACHINE PRODUCTS CORPORATION MIDDLEBURG HEIGHTS, OHIO REPORT INSPECTION NO.: 99901052/86-01 RESULTS: PAGE 4 of 7 D. STATUS OF PREVIOUS INSPECTION FINDINGS:

None.

E. OTHER FINDINGS OR COMMENTS:

1. Manufacturing Control The NOVA Sales Orders are initiated by the Sales Department and reviewed by the Quality Assurance Manager. The Sales Order requirements are entered on the Manufacturing Sequence Sheet by the Production and Quality Assurance Department. The Manufacturing Sequence Sheet includes such informatior, as: part description, material specification, QA requirements, hold points, inspection requirements and results, testing requirements, special processes required, documentation requirements, material traceability references, design requirements, and indication of final Quality Assurance Review.

The inspectors reviewed 17 sales orders and 22 sequence sheets during the course of the inspection. This resulted in identifying the following nonconformances:

a. On sales order 1515 the QA manager, during his review of the sales order failed to note that NCA-3800 had been incorrectly listed as a requirement on the sales order.

(86-01-04)

b. Neither the QAM nnr the OAPs required that the sequences sheet be reviewed and approved prior to issuance to the shop.

Although notations along with initials and dates on the sales order indicate that the sequence sheets were reviewed and approved at the time the sales order was reviewed and approved, there is no indication on the sequence sheets in the manufactur-ing area indicating they were approved. (86-01-05)

c. Neither the QAM nor the OAPs provide control over changes to the sequence sheet. Additionally, it was noted that sequence sheets in completed files for shipped material contained changes with no indication that the changes had been reviewed and approved at the same level as the original approvals. (86-01-06) 10 CFR 2.790 INFORMATION HAS BEEN DELETED 88

ORGANIZATION: NOVA MACHINE PRODUCTS CORPORATION MIDDLEBURG HEIGHTS, OHIO INSPECTION l REPORT RESULTS: PAGE 5 of 7 N0.: 99901052/86-01

d. The sequence sheets provide instructions for final inspection as to whether 100% or sampling inspection is required and what Acceptable Quality Level (AQL) is to be used. However, the QAM, QAPs, and sequence sheets do not require that the inspector record the number of pieces inspected, the number of pieces accepted, and the number of pieces rejected.

Without this information, a review of the sequence sheet does not indicate whether or not the inspector complied with the sampling plan. (86-01-13)

e. The sequence sheets require a dimensional inspection and reference QAP NP-12. However, neither the QAM nor QAP NP-12 provide instructions on what dimensions are required to be examined, what the acceptance or rejection criteria are, or a reference to the applicable standard (e.g. ANSI B18.1.1, B18.1.2,B18.2.2). (86-01-03) It was noted that the inspector was knowledgeable as to what was required due to his training and experience.

2. Training The inspectors reviewed the training records for the Level III QA inspector, QA Manager / Lead Auditor, and the warehouseman. The training file for the warehouseman did not contain documentation that he had been indoctrinated in NOVA's QAP NP-RSHS-1, although he had been performing in his present capacity for approximately three months. (86-01-01) The other documentation contained in the files, visual acuity test, training attendance records, and training and qualification certifications were found to be acceptable.

3. Receipt Inspection In reviewing the requirements for receipt inspection it was noted that when material is ordered for stock a sequence sheet is not generated. Without a sequence sheet there is no mechanism for providing the inspector with instructions on how many pieces are l

to be inspected, what dimensions are to be inspected, or what the acceptance / rejection criteria are for receipt inspection. (86-01-02)

The extent and results of the receipt inspection actually performed '

were documented on the receiving copy of the NOVA purchase order.

10 CFR 2.790 INFORMATION HAS BEEN DELETED 89 j

ORGANIZATION: NOVA MACHINE PRODUCTS CORPORATION MIDDLEBURG HEIGHTS, OHI0 REPORT INSPECTION N0.: 99901052/86-01 RESULTS: PAGE 6 of 7 1

4. Procurement NOVA's use of a subtier supplier prior to auditing the supplier l and placing them on the Approved Vendor List was one of the reasons for performing the inspection at NOVA. The inspectors reviewed 22 purchase orders placed by NOVA, the audits for the fif teen companies involved, and the associated test reports. For the purchase order files and audit reports reviewed, no instances of NOVA placing purchase orders with subtier suppliers prior to auditing the supplier were noted with the exception of the instance noted by in their audit. In response to the PG&E finding, NOVA hired an external auditor to review their files to determine the extent of their error. The external auditor determined that the error was an isolated incident and based on that finding NOVA concluded that no additional notification of their customers was required under Part 21. The results of the limited review performed by the inspectors tends to support NOVA's conclusion.

However, the review of the audit folders did reveal that NOVA had failed to follow their QAP NP-V5 in that the 19F5 audit of was not documented on NOVA's audit checklist.

The file only contained a single sheet of paper on which appeared the names of the company officials and a brief discussion of the type of work performed. There was no description of what was actually reviewed by the NOVA auditor. (86-01-07)

It was also noted that the audit checklist developed by NOVA for this purpose did not cover the posting, procedural, or reporting requirements of 10 CFR Part 21. (86-01-08)

5. Traceability The inspectors reviewed the shipping, handling, storage, and marking of material in the warehouse to assure that traceability was maintained. Two product forms were selected and examined, 5/16" diameter bolts and 1-7/8"' diameter bar stock, for marking /

tagging, segregation from other material, and retrievability of certified material test reports (CMTR). While all material was found to be identified by either marking or tags and traceable i

to the CMTR, NOVA's QAP NP-03 requires each 1" diameter or greater piece of bar stock to be individually stamped with the appropriate heat number or heat code. Not all bar stock 1" diameter or greater was individually stamped. (86-01-09)

L 10 CFR 2.790 INFORMATION HAS BEEN DELETED 90

I l

ORGANIZATION: HOVA MACHINE PRODUCTS CORPORATION MIDDLEBURG HEIGHTS, OHIO REPORT INSPECTION NO.: 99901052/86-01 RESULTS: PAGE 7 of 7

6. Calibration The inspectors reviewed the QAM and QAPs for controls on measuring and test equipment. Six instruments were selected during a tour ,

of the manufacturing and inspection areas to verify the implemen- {

tation of the requirements. No nonconformances were found in this I area.

7. Nonconforming Material / Corrective Action The inspectors reviewed the QAM and QAPs for controls on nonconform-ing material and requirements for corrective action. In reviewing the implementation of those requirements the inspectors reviewed all of the nonconformance reports generated since the company started in March,1984 and the associated corrective action. The review revealed several procedural inadequacies. If minor nonconformances are discovered on incoming material which NOVA could correct through rework, the procedures NP-Il and NP-V4 permit the QA Manager to not fill out a nonconformance report. This is in conflict with the QAM.

Also, if a nonconformance report is not required then the inspector or QA Manager never gets to the portion of procedures which require that a sequence sheet be used to control rework of nonconforming material (86-01-11). Additionally, if a nonconformance report is not generated, there is no place to document the disposition of the nonconforming material (86-01-10). In practice, NOVA was noting on the blue

" accept" tag that rework was necessary prior to using the material.

Another procedural inadequacy was the lack of instructions in the QAM or QAPs on how material returned t.; a customer would be controlled.

(86-01-12)

The implementation review of the Nonconformance/ Corrective Action Reports revealed that on report #26 the OA Manager had signed the report on 6/2/85, indicating that the corrective action had been completed, when in fact the corrective action had not been completed as of 7/16/86. (86-01-14) 10 CFR 2.790 INFOR!tATION HAS BEEN DELETED 91

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION INSPECTION NO.: 99901047/86-02 DATE: June 21-26, 1986 ON-SITE HOURS: 240 CORRESPONDENCE ADDRESS: Pacific Nuclear Systems, Incorporated Nuclear Packaging, Incorporated ATTN: Mr. David F. Jones Chairman / President 1010 South 336 Street Federp1 Way, Washington 98003 ORGANIZATIONAL CONTACT: Mr. Joe Olivadoti TELEPHONE NUMBER: (206) 874-2235 NUCLEAR INDUSTRY ACTIVITY: Designer and supplier of nuclear waste transpor-tation casks and handling equipment.

ASSIGNED INSPECTOR: .

) 8

4. W. Craig, Chief, Spec \igi Projects Inspection D te Section (SPIS), VendorVrogram Branch C. M. Abbate, SPIS; P. J. Prescott, SPIS; J. White, NRR OTHER INSPECTOR (S -

R. L. ilimberg, SPIS; 5. Clark, Consultant APPROVED BY: u )!cPb dchn W. Craig, Chief, SPIS, V(pdor Program Branch Date INSPECTION PASES AND SCOPE:

A. BASES: 10 CFR Part 71 and 10 CFR Part 21.

B. SCOPE: Review the nondestructive testing (NDT) activities performed on the two Model 125-B shipping casks. The casks will be used to transport nuclear waste from Three Mile Island Unit 2. Thc review included NDT records for visual inspection (VT) of welds, hydrostatic pressure tests, leak tests (LT), radiographic tests (RT), ultrasonic tests (UT), liquid penetrant tests (PT), and the gamma scan of the lead annulus.

PLANT SITE APPLICABILITY: Three File Island Unit 2.

93

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION '

NO.: 99901047/86-02 RESULTS: 6/21-26/86 PAGE 2 of 18 l A. VIOLATIONS:

None.

B. NONCONFORMANCES:

Contrary to 10 CFR 71.119 and Section 3.3.2 of procedure number LT-26,

" Outer Cask Hydrostatic Pressure Test", Revision 0, dated November 6, 1985, LT-26 was not followed by Olympic Northwest Industries (0NI) in that pressure recording charts were not used during the outer cask hydrostatic pressure tests performed on both Model 125-B shipping casks. (86-02-01)

C. UNRESOLVED ITEMS:

None.

D. STATUS OF PREVIOUS INSPFCTION FINDINGS: ,

1. (0 pen) Violation (86-01-01)

Contrary to Section 21.21 of 10 CFR Part 21, NUPAC has not developed and implemented appropriate procedures to provide for evaluating defects; or to assure that a director or responsible officer is informed if a basic component contains a defect or to notify the Commission when information is obtained which reasonably indicates a defect.

This area was not reviewed during this inspection.

2. (0 pen) Violation (86-01-02)

Contrary to Section 21.31 of 10 CFR Part 21, NUPAC issued purchase order (P0) 3104-18, dated January 31, 1984, to (0NI) without specifying that the provisions of 10 CFR Part 21 were applicable.

This area was not reviewed during this inspection.

3. (0 pen) Violation (86-01-03)

Contrary to paragraph 71.103 of Subpart H to 10 CFR Part 71, NUPAC failed to ensure that an adequate QA program was established and executed at a NUPAC subcontractor.

This area was not reviewed during this inspection.

94

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION RESULTS: 6/21-26/86 PAGE 3 of 18 NO.: 99901047/86-02

4. (Closed) Unresolved Item (86-01-04)

NDT documentation required by NRC C of C Number 9200 was not available in the Quality Record files for the Model 125-8 casks. Applicable documentation for RT and PT conducted during fabrication was not reviewed.

Documentation for RT and PT performed during fabrication was reviewed during this inspection. The tests were performed in accor. dance with adequate procedures and performed by qualified personnel. This area is discussed in Sections E.3 and E.7. This item is considered closed.

5. (Closed) Unresolved Item (86-01-05)

RT, PT, VT, hydrostatic, UT, and LT are required by the Safety Analysis Report (SAR) prior to first use of the Model 125-8 casks.

The results of these acceptance tests were not reviewed.

A review of the documentation of the results of the acceptance tests for the two Model 125-B casks determined that the procedures were adequate, the tests were performed by qualified personnel, and the results were acceptable. These items are discussed in Section E. This item is considered closed.

6. (Closed) Unresolved Item (86-01-06)

LT of the Model 125-B casks is required by C of C Number 9200.

The qualification records for the individual who performed the tests were not reviewed.

The records for the individual who performed the leak tests on the Model 125-B casks were reviewed during this inspection. The individual who performed the tests was found to be qualified as discussed in Section E.4. This item is considered closed.

E. INSPECTION FINDINGS AND OTHER COMMENTS:

1. Entrance and Exit Meetings An entrance meeting was conducted on June 21, 1986 at the PNSI/NUPAC office in Feder-1 Way, Washington. The purpose and scope of the inspection were discussed during this meeting. NUPAC is a subsidiary t

l of PNSI and therefore, NUPAC utilizes the PNSI CAM. During the exit meeting conducted on June 26, 1986, the inspection findings and observations were discussed with NUPAC personnel.

95 l

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION N0.- 99901047/86-02 RESULTS: 6/21-26/86 PAGE 4 of 18

2. Nondestructive Testing (NDT)

The NRC issued C of C Number 9200 which approved the design of the NUPAC Model 125-B shipping package (cask) on April 11, 1986. The NDT requirements applicable to this cask design are addressed in the C of C by reference to NUPAC drawings. These tests are also discussed in the Safety Evaluation Report (SER) for the Model 125-B casks.

Chapter 8 of the NUPAC SER for the Model 125-B Fuel Shipping Cask, Revision H, dated February 10, 1986, states that testing.is to be performed in accordance with the requirements delineated on NUPAC Drawing No. X-101-100, sheets 1 through 6, "NUPAC Model 125-B Shipping Cask, Revision H, dated February 10, 1986. Chapter 8 also provides specific steps for LT and additional requirements for shielding integrity tests. The SER states that NDT is to be performed in accordance with the ASME Code,Section III, Subsection NB. This subsection of the ASME Code requires that personnel performing NDT activities are to be qualified in accordance with the guidelines of ASNT SNT-TC-1A.

NDT records for the NDT tests performed on the Model 125-B casks were reviewed during this inspection and are discussed in Sections 3 through 9 below.

3. Radiographic Testing Note 12 of NUPAC Drawing No. X-101-100, Revision H, requires that radiographic testing (RT) be performed in accordance with ASME Code,Section III, Division 1, Subsection NB, Article NB-5000 and Section V, Article 2 for all longitudinal and circumferential seam welds of the outer containment vessel (OCV) inner and outer shells. RT is also required for the centrifugally cast tubes in the inner containment vessel (ICV). Subsection NB-5320 specifies the applicable radiographic acceptance standards. NB 5320 states, in part, that welds which are shown to have any discontinuities, such as any type of crack or zone of incomplete fusion or penetration, are unacceptable, a) Outer Containment Vessel RT of the seam welds on the inner and outer shells of the OCV was performed on July 16-24, 1985, by an employee of the Nooter Corporation who was not a certified Level II examiner in RT in accordance with the guidelines of ASNT SNT-TC-1A since he had not passed the color perception eye examination. All film 96 l

l

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON 1

REPORT INSPECTION N0.: 99901047/86-02 RESULTS: '6/21-26/86 PAGE 5 of 18

)

reviewed by this Nooter employee was subsequently reviewed on July 22-24, 1985, by an employee of ONI who was a certified Level II examiner in RT per ANST SNT-TC-1A.

Nooter Corporation Welding Procedure Specification (WPS) number D-1626, Revision 0, dated May 15, 1985, and Nooter Drawing Number JN-D 68730, Revision 1, dated May 22, 1985, provide requirements for the scam welding of the inner and outer shells of the OCV. The inner shell is fabricated from one . inch thick type 304 stainless steel (SS) plate and the outer shell is fabricated from two-inch thick type 304 SS plate. Both shells consist of two sections. Each section is formed into a tube with a longitudinal weld and then the two sections are joined by a circumferential weld to achieve the 181-inch vessel length. The welds on the one-inch thick inner shell achieve 100% penetration by fusion. The welds on the two-inch thick outer shell are made by back welding a vee weld with shielded metal arc welding (SMAW) in twc passes and submerged arc welding (SAW) in one pass. The root of the vee weld is backgouged to sound metal to achieve complete penetration of the vee weld. The remaining two-thirds of the wall thickness is joined by SAW a U joint to complete the weld. The welding procedure on the outer shell is significant since a review of RT film should confirm that the incomplete penetration in the welds of the outer shell has been eliminated by backgouging.

This is discussed further below in the section c) Results.

Neither RT procedure nor radiographic technique sketches were available for review by the NRC inspectors. Radiographic reports were provided on Nooter Corporation form 108, June 1972 revision, and RT film review by Nooter and ONI were provided on Nooter form 109, June 1972 revision.

The RT of the longitudinal seam welds was performed with an X-Ray source on the outside of the shell and film on the inside diameter of the shell, b) Inner Containment Vessel Wisconsin Centrifugal Incorpnrated (WCI) RT procedure number QCP-188, " Radiographic Procedure ASME Section VIII," Revision E, dated August 3,1984, was reviewed for applicability to the RT of the centrifugally cast tubes in the inner containment vessel of both casks. The procedure was written for ASME Code,Section VIII, and references penetrameter tables NB-5111-1 used for the examination. Also, the procedure is in compliance with 97

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON l

REPORT l INSPECTION '

NO.: 99901047/86-02 RESULTS: 6/21-26/86 PAGE 6 of 18 the ASME Code,Section III, Subsection NB-5000. The American Society for Testing and Materials (ASTM) Standard E-446 was used as a reference for class II casting imperfections.

PT of the centrifugally cast tubes in the ICV was performed on June 24 to July 27, 1985 by an employee of WCI who is a certified Level II examiner in RT in accordance with the guidelines of ASNT SNT-TC-1A.

c) Results (1) The NRC inspectors reviewed 100% of the film covering RT of the longitudinal and circumferential welds on the inner and outer shells of the outer containment vessels on the Model 125-B casks. The 58 radiographs representing cask number 1 met the requirements of the ASME Code,Section III, NB-5300 and Section V, Article 2. The 29 radiographs representing the seam welds on the inner shell of cask number 2 also met these ASME Code requirements.

The 30 radiographs representing the seam welds on the outer shell of cask number 2 met the above ASME requirements except the two radiographs marked R12-(1-4) seam, for stations 1-2 and 2-3. The NRC inspectors located an indication on the film which had not been identified by previous film reviewers. The indication was initially classified as incomplete penetration. The indication had a length of approximately 7/16-inches maximum, a width of approximately 1/16-inches maximum, and a depth estimated at 1/8-inch maximum. The indication was classified as lack of penetration because it was a straight line in the center of the weld and was estimated to be approximately 1/4 to 3/8-inches from the inside diameter of the cuter shell. The indication was determined to be in a location where backgouging (as discussed above) failed to ensure complete penetration of the vee weld. Incomplete penetration is contrary to the requirements of the ASME Code,Section III, Subsection NB-5320 as stated above.

The identification of this weld indication which apparently did not comply with the requirements of C of C Number 9200 was discussed with NUPAC personnel. NUPAC requested that X-Ray, Inc. have another NDT examiner qualified as a Level III in RT review the radiographs for stations 1-2 and 2-3 of the longitudinal weld cf cask 2. The X-Ray, Inc.

examiner stated that he believed that the indication was 98

l ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION i

RESULTS: 6/21-26/86 PAGE 7 of 18 NO.: 99901047/86-02 a slag inclusion. Slag with the dimensions specified above would be acceptable and satisfy the requirements of C of C Number 9200.

Subsequent to the inspection, these radiographs for RIB-(1-4) seam 1-2 and 2-3 were computer enhanced and reviewed by three NDT Level III examiners serving as consultants to the NRC. The consensus of these examiners was that the indication was a slag inclusion in. the weld and acceptable under the ASME Code requirements specified in C of C Number 9200. A summary of this review is attached as an addendum to this report.

(2) WCI fabricated and performed RT for the seven centrifugally cast tubes in each cask. The NRC inspectors reviewed 20%

(180 radiographs) of the 840 radiographs covering RT of the seven centrifuga11y cast tubes in the ICV of each cask.

Based upon the radiographs reviewed and a review of the RT reports prepared by WCI, RT was performed in accordance with the written procedure, QCP-188, the ASME Code,Section V, and the results met the requirements of Subsection NB-5000 (3) Several weaknesses were identified during the review of RT records. These include data sheets documenting the review of radiographs which did not document the imperfections or artifacts present and the initial review of radiographs for the OCV which was performed by an individual who was not qualified to the standards specified in C of C Number 9200.

No items of nonconformance or unresolved items were identified in this area.

4. Fabrication Leak Verification f

a) Procedure Review Sheet 1, NUPAC Drawing No. X-101-100, Revision H, General Note 19 specifies that the inner vessel and outer containment boundaries shall be 1 ak tested (LT) to demonstrate a leak rate nottoexceed1x10~9 atmospheric cubic centimeters per second (atm-cc/sec) per NUPAC leak verification procedure LT-21.

NUPAC procedures LT-21(a), " Inner Vessel Fabrication Verification Leak Test", and LT-21(b), " Outer Cask Fabrication Verification Leak Test", were approved for implementation on November 11, 1985. These procedures provided for the following tests to be 99

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION NO.: 99901047/86-02 RESULTS: 6/21-26/86 PAGE 8 of 18 performed on each of the two packages constructed: 1) ICV -

Structural Integrity, Lid 0-Ring Seal Sealing Integrity, Rupture Disk Sealing Integrity, Vent Port Closure Bolt Sealing Integrity; and 2) OCV - Structural Integrity, Lid 0-Ring Seal Sealing Integrity, Rupture Disk Sealing Integrity, Vent Port Closure Bolt.

The tests were performed with consideration of guidelines provided in Regulatory Guide 7.4, " Leakage Tests on. Pack-ages for Shipment of Radioactive Materials", and ANSI N14.5-1977, "American National Standard for Leakage Tests on Packages for Shipment of Radioactive Materials." Since the specified acceptance criteria for all tests was 1 x 10-7 atm-cc/sec, the Helium Mass S which is sensitive to 5 x 10 gectrometric Envelope atm-cc/sec, Method, was used.

Chicago Bridge and Iron (C8&I) performed the LT on the ICVs for each cask at their facility in Salt Lake City, Utah. These tests took place on November 13, 1985 and December 18, 1985.

NUPAC performed the LT on the OCVs for both casks at the ONI facility in Bremerton, Washington. These tests took place November 27, 1985 and December 17, 1985. NUPAC QA witnessed all leak tests performed on the two casks.

While NUPAC's LT-21 procedures identified the basic methodology for all helium leak tests performed, CB&I supplemented LT-21 with their own procedures, MSTIN, " Helium Leak Testing Procedures-NUPAC Model 125-B Shipping ICV Assembly". This procedure was written in accordance with the test description in the SAR and was approved for use by NUPAC on October 31, 1985.

Both procedures detail equivalent techniques to perform leak testing. Further, while the test procedures LT-21(a) and (b) were reviewed and determined to be adequate, the inspector noted that the actual sequence followed during leak testing deviated frc;n the procedure, but the procedure was not revised to reflect the actual test sequence.

b) Personnel Qualification During the NRC inspection conducted May 5-8, 1986, i.he qualifi-cations of the NUPAC employee who performed LT was identified as an unresolved item.

1 i

100

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION RESULTS: 6/21-26/86 PAGE 9 of 18 NO.: 99901047/86-02 10 CFR 71.119, " Control of special processes" requires that measures be established to assure that special processes such as NDT be controlled and accomplished by qualified personnel using qualified procedures in accordance with applicable codes, standards, specifications, criteria, and other special requirements. NUPAC Procedure QP-14, " Quality Assurance Training" identifies criteria for the certification of personnel in areas pertaining to inspection, examination and testing.

Qualification pursuant to the ASME Code,Section III, Division I, Subsection NB, NB-5000 and Section V was a specific commitment in NUPAC's application relative to the Model 125-B cask for personnel performing liquid penetrant (PT), ultrasonic (UT),

and RT examination. This commitment directly references the qualification guidelines identified in SNT-TC-1A but does not apply to personnel performing LT.

c) Results The inspector reviewed the data associated with the tests and the results indicated that the acceptance criteria of 1 x 10-7 atm-cc/sec was achieved for all tests. In some instances the tests indicated leaking sufficient to require retesting. In these instances Supplier Disposition Requests (SDRs) were written to identify the problem and recommended the corrective action necessary. The inspector also reviewed the calibration data regarding the Mass Spectrometer Leak Detector (MSLO) used for the helium leak tests ar.d verified that the MSLD was calibrated immediately before and after each test.

A review of the qualifications of the NUPAC employee who perform-ed the LT was performed including his training, experience, and ability. Interviews with the individual who performed and witnessed the tests and with his supervisor were conducted. The inspectors concluded that the individual was qualified and capable of performing LT and had an excellent working knowledge of LT based upon experience and training.

Based upon the material reviewed, the fabrication leak tests of all vessels used for the assembly of both Model 125-B shipping casks were performed by qualified personnel and in Tes accordance with written procedures and requirements.results indicated atm-cc/sec was achieved in all cases.

101

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION NO.: 99901047/86-02 RESULTS: 6/21-26/86 PAGE 10 of 18 No items of nonconformance or unresolved items were identified in this area.

5. Visual Examination a) Personnel Qualification General Note 9 on General Arrangement Drawing No. X-101-100, Revision H, requires that all welds be visually inspected in accordance with Section 8.15.1, " Quality of Welds", of the American Welding Society (AWS) Structural Welding Code for Steel, D1.1-85. VT requirements are also contained in NUPAC's

" Procedure for Visual Inspection, Weldments and Adjacent Materials", VT-01, Revision 0, dated February 19, 1985. Visual inspection (VT) of welds consists of examination of the weld preparation, observation of in-process welding, and the exam-ination of the dimensional accuracy and quality of the final weld.

Section 3.1 of VT-01 requires that NUPAC assure that all person-nel performing VT are qualified and trained. Section 4.1 of VT-01 requires that all personnel conducting VT be certified to the requirements of Section IV of AWS QC-1, " Qualification and Certification of Welding Inspectors" or be approved spe~cifically for VT by the Corporate Quality Director.

CB&I and ONI personnel performed the VT of the welds on the two Model 125-B shipping casks. The shop travelers used during the fabrication of the casks identify the steps in which VT was performed and include an area for the sign-off of the inspection indicating that the VT was performed and the results obtained.

Eight employees at CB&I were certified as CB&I Level II VT examiners and one employee was certified as a CB&I Level I VT examiner. The person at ONI performing VT was also qualified to be a VT examiner. The records for each of the VT examiners were reviewed with respect to eye examination results, training received in the area of VT, and experience in VT. The individuals performing VT were found to be qualified to the requirements outlined in NUPAC procedure VT-01.

l 1

102 l

ORGANIZATION: NUCLEAR PACKAGING INCORFORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION NO.: 99901047/86-02 RESULTS: 6/21-26/86 PAGE 11 of 18 b) Procedure Review Section 8.15.1 of AWS D1.1-85 states that welds shall be accept-able by visual inspection if various criteria are met. These criteria include no cracking, complete fusion and penetration, and undercut and underrun within acceptable dimensions. VT-01 provides the basic methods and requirements for visual examin-ation of weldments and adjacent materials. The procedure was reviewed and was adequate to perform an acceptable visual examination of welds and adjacent materials. The procedure outlines the responsibility and acceptance criteria to be used.

The procedure also identifies the two methods which can be used to perform VT and the steps to be followed during visual inspection.

Shop travelers used during the fabrication were reviewed with respect to VT. VT was performed and signed off by qualified examiners from CB&I and ONI. In addition, a NUPAC employee observed VT as well as other fabrication activities at CB&I and ONI.

This NUPAC employee was certified as an AWS Welding Inspector per the requirements of Section IV of AWS QC-1. NUPAC quality planners (shop travelers) were also reviewed during this inspection, and it was found that the NUPAC VT examiner had accepted, stamped, and dated the visual inspection performed on the welds.

No items of nonconformance or unresolved items were identified in this area.

6. Hydrostatic Pressure Tests a) Procedure Review Section 71.85(b) of 10 CFR Part 71 requires, in part, that the containment system be tested at an internal pressure at least fifty percent higher than the maximum normal operating pressure.

Note 8 on NUPAC Drawing No. X-101-130, Revision B, dated June 21,1985, " Assembly ICV NUPAC Model 125-B Shipping Cask",

103 l

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION N0.: 99901047/86-02 RESULTS: 6/21-26/86 PAGE 12 of 18 requires that the inner vessel containment shall be subjected to an internal test pressure equal to 190 pounds per square inch (psi) and that testing be performed in accordance with the NUPAC fabrication verification pressure test procedure, LT-25. Note 8 on Drawing No. X-101-140, Revision A, dated June 25, 1985, " Assembly Cask NUPAC Model 125-B Shipping Cask,"

requires that the outer cask containment boundaries be subjected to an internal test pressure equal to 190 psi and that the testing be in accordance with the NUPAC fabrication verification pressure test procedure, LT-26.

The hydrostatic pressure test is performed by filling the cask with water and pressurizing it to 190 psi. The pressure is observed for ten minutes and the test is acceptable if the pressure drop does not exceed five psi over the ten minute test period.

CB&I performed the inner vessel containment hydrostatic pressure tests while ONI performed the outer vessel containment hydro-static pressure tests.

b) Inner Vessel Containment LT-25, " Inner Vessel Hydrostatic Pressure Test", Revision 0, dated November 6, 1985, and SHTP-1, " Shop Hydrostatic Test Prncedure", Revision 1, dated August 8, 1985, were used to perform the ICV hydrostatic pressure tests on both Model 125-B Shipping Casks. LT-25 outlines the scope, referenced documents, technical requirements, test conditions, and acceptance criteria, but does not specify test media temperature or pressure gage requirements. SHTP-1 is a CB&I procedure which was used in addition to LT-25 to perform the ICV hydro-static pressure tests. SHTP-1 is a detailed procedure which defines the test step-by-step and includes safety considerations, test equipment layouts, test media temperature, and pressure gage requirements.

The equipment used during the test was calibrated prior to the tests. The two test result recording charts for the casks indicated that the pressure was held at the required 190 psi for ten minutes with no drop in pressure.

l

{

104 l

l I

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED

' FEDERAL WAY, WASHINGTON REPORT INSPECTION NO.- 99901047/86-02 RESULTS: 6/21-26/86 PAGE 13 of 18 c) Outer Vessel Containment LT-26, " Outer Cask Hydrostatic Pressure Test", Revision 0, dated November 6, 1985, was the procedure used by ONI to perform the outer vessel pressure tests and delineates the requirements for performing the hydrostatic pressure test on the outer vessel containment. The procedure includes the referenced documents, technical requirements, test conditions, and acceptance criteria, but temperature of test media and pressure gage calibration requirements were not specified.

' Calibration data was reviewed and it was noted that the equipment used during the tests was calitrated prior to use. Section 3.3.2 of LT-26 requires that recording charts be used, but no pressure recording charts were used during the outer vessel containment pressure tests.

The test results for these tests were transmitted from ONI to NUPAC by two letters dated November 20, 1985 and December 11, 1985, stating that the tests were performed in accordance with LT-26 and that the rEsults were acceptable. A NUPAC QA employee was present during the tests and NUPAC accepted the results.

Nonconformance 86-02-01 was identified in this area.

7. Liquid Penetrant Testing The requirements for liauid penetrant testing (PT) are contained in Note 10, HUPAC Drawing No. X-101-100, Sheet 1, Revision H, which specifies that welds identified on the drawings for the NUPAC Model 125-B casks are to be liquid penetrant inspected on root and final pass in accordance with ASME Code,Section III, Division I, Subsection NB, Article NB-5000 and Section V, Article 6.

Additionally, the SER for these casks states that the drawings shall identify the weld joints to be nondestructively examined, the method used, and the code or standard for the examination procedure.

The NRC inspector reviewed the applicable PT procedures, personnel qualifications and test results for two vendors who performed PT on l the Model 125-B shipping package for NUPAC. CB&I Procedure PT SWI, end ONI Procedures QI 12.1 and QI-12.7 were reviewed.

105

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION N0.- 99901047/86-02 RESULTS: 6/21-26/86 PAGE 14 of 18 A representative sample of PT reports written by CB&I and ONI for '

the Model 125-B shipping package were reviewed. The documentation reviewed for the PT reports indicated that the PT results were sa tisfactory. Additionally, CB&I shop travelers were reviewed.

These travelers identified the steps during cask fabrication which specified PT examination and the individuals who performed the PT.

Based upon this review, PT was performed as specified in the travelers.

A sample of NDT travelers and NDT reports which were reviewed for PT were written in accordance with the applicable test procedures. It was also noted that the personnel records maintained by NUPAC on their sub-tier vendors for PT were current and found to be in accordance with the applicable procedures.

No items of nonconformance or unresolved items were identified in this area.

8. Gamma Scan The description of the Model 125-B cask contained in C of C Number 9200 includes the 3.88-inch thick lead annulus. The SER for the-Model 125-B cask also discusses the lead shielding and corresponding radiation readings for the casks. Specifically, the radial shielding of the Model 125-B cask consists of 3.88-inches of lead sandwiched by a 2-inch outer cask steel cylinder and a 1-inch inner cask steel cylinder proceeded by a 1-inch steel inner vessel shell.

Section 8.0, " Acceptance Tests and Maintenance Program," of the SAR for the casks states that the lead shielding integrity shall be confirmed via garma scanning. The SAR also discusses the method utilized to derive maximum acceptable dose values. The acceptable values were determined by NUPAC based solely upon calculations; representative test blocks were not fabricated and used to determine acceptable ganea scan readings. The acceptance values were based upon calculations which assumed up to a 10% loss of shielding.

a) Procedure The gamma scan of the Model 125-B casks was performed for NUPAC by X-Ray, Inc. Purchase Order (P0) 3206-IT, dated July 30, 1985, contained the specifications for this test.

Included in the P0 were the requirements that the gamma scan be performed in accordance with NUPAC Procedure GS-001, 106

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION NO.: 99901047/86-02 RESULTS: 6/21-26/86 PAGE 15 of 18 Revision 4, that the scan be performed at Metalex Products, Ltd.,

and that the gamma scan be performed under the presence and direction of NUPAC QA personnel.

NUPAC Procedure GS-001, Revision 4, dated February 2, 1983, outlines the minimum requirements for gamma scan of lead lined transportation casks. Addendum No. I to GS-001, Revision 4, dated September 9,1985, states that no reading indicating a loss of shielding in excess of 5% from the normal design lead thickness will be acceptable.

b) Results X-Ray, Inc. reports numbered 9255, dated October 2, 1985, and 9279, dated October 24, 1985, document the gamma scan test results for cask 1 and 2. These reports state that the gamma scan was performed utilizing GS-001, Revision 4, Addendum No.1, and identify the source and survey instruments used. Charts of gamma readings (grids as specified in GS-001) were attached to the reports.

• During the gamma scan, readings which exceeded a 5% decrease in lead shield thickness values were identified and Quality Discrepancy Reports (QDRs) were written (QDR Nos. 243 and 263).

Since the readings did not exceed the 10% decrease in lead shield thickness acceptance values contained in the SAR, NUPAC deter-mined that the observed gamma readings were acceptable.

SDR No. 674 dated September 11, 1985 was also reviewed. This SDR identified a reduction in the lead annulus thickness frcm the value of 4.00 to 3.75-inches specified in NUPAC Drawing No.

X-101-120 to an actual value of approximately 3.482-inches to 3.62-inches on cask 1 and actual value of approximately 3.454-inches to 3.489-inches on cask 2. These values were the result of dimensional tolerances of the 2-inch and 1-inch shells and due to the 2-inch shell being 0.100-inch over gage. NUPAC i

determined that based upon the following items the casks were acceptable for use: (1) the minimum acceptable defect free lead thickness was 3.488-inches (based upon gamma scan acceptance criteria), and (2) the additional thickness of steel provides a lead equivalent shielding value.

The inspectors reviewed the radiation readings recorded on the grids during the gamma scan for each cask. No readings were identified which exceed the acceptance values contained in the SAR.

107

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION i NO.: 99901047/86-02 RESULTS: 6/21-26/86 PAGE 16 of 18 )

No items of nonconformance or unresolved items were identified )

in this area. '

9. Ultrasonic Examination NUPAC Drawing No. X-101-100, sheet 1, Revision H, Note 11 states that ultrasonic inspection (UT) shall be performed on indicated welds and fcrgings in accordance with ASME Code,Section III, Division 1, Subsection NB, Article NB-5000 and Section V, Article 5. During this inspection the inspectors reviewed the applicable NUPAC drawings to identify the welds and forgings which required UT. Additionally, the qualifications of the personnel who performed the UT, the test procedures used and the test results were reviewed.

Gulf Coast Machine and Supply Company (Gulfco) performed UT on various forgings used in the Model 125-B casks, including cask lid (Gulfco Report No. UT-3212), cask base (Gulfco Report No. UT-3219), and outer cask collar (Gulfco Report No. 3218 and 3221). The UT reports from Gulfco certified that personnel performing UT examinations had been qualified to SNT-TC-1A and that the forgings identified on the UT reports had been inspected and found acceptable. Additionally, the Gulfco UT procedure UT-388 was reviewed to determine whether or not the procedures were adequate. ASME Code Section V, Article 5, Ultrasonic Examination General Requirements was the basis for the procedural review.

Based upon the UT documents reviewed, Gulfco UT activities were performed by qualified personnel utilizing adeouate procedures, and the UT test results were acceptable. However, the UT data generated during the tests (transducer recordings) was not available to be reviewed.

CB&I performed UT on welds made during Model 125-B cask assembly.

CB&I shop travelers identify the operations / steps performed during assembly of the casks. These travelers were audited to determine whether UT had been performed on the appropriate welds and which individuals had performed the UT. CB&I UT reports and CB&I UT procedure UT-1-PSX were also reviewed.

Based upon an audit of CB&I travelers and CB&I Record Drawings and NUPAC Drawing No. X-101-100, Revision H, UT was performed upon the appropriate welds. Further, based upon a review of CB&I UT examination procedures, UT reports and CB&I NDT personnel qualification records, UT testing was performed by qualified personnel utilizing adequate procedures. While the UT test results were acceptable, CB&I UT reports were not i

108 f

1 I

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION NO.: 99901047/86-02 RESULTS: 6/21-26/86 PAGE 17 of 18 completely filled out and the data generated during the UT tests (transducer recordings) was not available for review.

UT activities at CB&I were also reviewed by NUPAC personnel during cask fabrication.

NUPAC QA Instructions require a NUPAC inspector to verify certain activities at vendors performing work for NUPAC. For example, OA Instruction IT-56, Body Weldment, ICV requires a NUPAC inspector to verify activities such as UT of welds and identification of welds to ensure that testing is performed in accordance with requirements of the C of C. IT-56 was performed by a NUPAC inspector during fabrication of both Model 125-B casks.

No items of nonconformance or unresolved items were identified in this area.

F. PERSONS CONTACTED:

• C. Temus, Director of Regulatory Affairs, NUPAC

• J. R. 011vadoti, QA Manager, NUPAC

• G. R. Hayes, EG&G Quality Engineer, EG&G Idaho, Inc.

K. Hanna, QC Engineer, NUPAC D. Schmoker, Vice President of Engineering, NUPAC

• Attended Exit Meeting l

l f 109 i

ORGANIZATION: NUCLEAR PACKAGING INCORPORATED FEDERAL WAY, WASHINGTON REPORT INSPECTION NO.: 99901047/86-02 RESULTS: 6/21-26/86 PAGE 18 of 18 Addendum to Inspection Report No. 99901047/86-02 Subsequent to the NRC inspection conducted June 21-26, 1986, at the NUPAC facility in Federal Way, Washington, information concerning the indication on the radiographs classified as incomplete penetration was reviewed. The NRC inspectors had located an indication on film marked R1B-(1-4), film 2-3 which they interpreted to be incomplete or lack of penetration.

fncomplete penetration does not meet the requirements of the ASME Code,Section III, Subsection NB, Article NB-5320 which specifies that incomplete penetration is unacceptable. The linear length of the indication was approximately 7/16 or 0.438-inches. Article NB-5320 (b)(2) specifies that any elongated indication which has a length greater than 1/3 t is a discontinuity that is unacceptable.

The outer shell of the outer cask longitudinal seam weld has a thickness of 2 1/8-inches. Since t is the thickness of the thinner portion of the weld,1/3 t equals 1/3 of 2 1/8-inches which equals 0.708-inches. Therefore, the 0.438-inch elongated indication is acceptable under NB-5320(b)(2) if the indication was determined to be a slag inclusion.

On July 15, 1986, Messrs. Sam Wenk of Port St. Lucie, Florida and Ed Martindale of EMAR Enterprises who are certified to be qualified to a NDT Level III in RT in accordance with the guidelines of ASNT SNT-TC-1A, reviewed the RT film marked RIB-(1-4) seam, Film # 2-3. The NRC inspectors who performed the inspection were also present during this review and discussion of the indication and additional new information provided by a computer enchancment of the radiograph.

In addition to the initial 0.438-inch indication, a second 0.125 inch indication was identified approximately 0.250-inches from the initial Indication toward station number 3.

The review concluded that the indications exhibited width without sharpness on each end, and even density across the indication which suggests volume.

These characteristics of the indications and the location of the indications suggest volumetric slag inclusions. The consensus of the three RT Level III interpreters, and the NRC inspectors was that the 0.438-inch and 0.125-inch long linear indications are classified as slag which are acceptable to the requirements of the ASME Code,Section III, Subsection NB, Article NB-5320.

110

ORGANIZATION: SOLID STATE CONTRCtS. INCORPORATED COLUMBUS, OHIO REPORT INSPECTION INSPECTION NO.- 99900276/86-01 DATES: 7/28-30/86 ON-SITE HOURS: 22 CORRESPONDENCE ADDRESS: Solid State Controls, Inc.

l 875 Dearborn Drive ATTN: Mr. R. F. Cassidy President Columbus, Ohio 43216 ORGANIZATIONAL CONTACT: W. G. Hampton TELEPHONE NUMBER: (614) 846-7500 NUCLEAR INDUSTRY ACTIVITY: Manufactures battery chargers, isolation trans-formers and uninterruptable power supplies. Less than 1% of the apparatus is intended for installation in nuclear power plants.

ASSIGNED INSPECTOR: - -

(,v 2 K. R. NaYdu,' Reactive Inspection Section (RIS) Date OTHERINSPECTOR(S):

APPROVED BY: -

f7 c E. W. Merschoff, Chi RIS, Vendor Program Branch Date INSPECTION BASES AND SCOPE:

1 A. BASES: 10 CFR Part 21 and Appendix B to 10 CFR Part 50.

B. SCOPE: Review implementation of the quality assurance program in selected areas. Observe inverters manufactured for the Surry Nuclear Power Plant and review related records.

PLANT SITE APPLICABILITY: Surry 1 & 2 (50-280; 50-281).

111 i

ORGANIZATION: SOLID STATE CONTROLS, INC.

COLUMBUS, OHIO l

REPORT INSPECTION NO.: 99900276/86-01 RESULTS: PAGE 2 of 7 A. VIOLATIONS:

None.

B. NONCONFORMANCE:

None.

C. UNRESOLVED ITEMS:

None.

C. ACTION TAKEN ON PREVIOUS INSPECTION FINDINGS:

1. (Closed) Nonconformance - 99900276/82-01 Item A. The nonconformance identified that contrary to Criterion V of 10 CFR 50, Appendix B, Solid State Controls Incorporated (SCI) did not maintain records to assure compliance with their 10 CFR Part 21 procedure concerning failures of inverters at St. Lucie and Watts Bar nuclear power plants.

SCI developed Procedure SCI-QA-20, Revision 0, dated January 11, 1983 to investigate equipment failures in nuclear facilities. A failure report form is required to be filled out for each reported failure.

The inspector reviewed fifty-five " Failure Reports" written in 1983.

Five of these reports pertained to normal wear and tear problems in inverters installed in nuclear power plants. The SCI Quality Assurance (QA) manager stated that to increase the efficiency of the reporting system, work done on service calls are documented in Field Service Reports (FSR) which are subsequently reviewed by QA personnel to determine whether the problems are generic in nature. The inspector reviewed twenty-three FSRs written in 1986. Four of these reports pertained to inverters installed in nuclear power plants and the problems identified were not generic in nature. The NRC inspector informed the SCI QA manager that documenting the service calls in FSRs instead of Failure Reports was not in accordance with their QA manual requirements. Prior to the conclusion of the inspection, the QA manager revised procedure SCI-QA-20 to specify the use of FSRs instead of Failure Reports to document failures in inverters installed in nuclear power plants.

2. (Closed) Nonconformance 99900276/82-01 Item B. The nonconformance identified that contrary to Criterion V of 10 CFR 50, Appendix B and SCI Quality Control Instruction V Revision 2 dated December 2,1980, the project engineer used a new form SF-105A (instead of 105) to docu-112

! ORGANIZATION: SOLID STATE CONTROLS, INC.

COLUMBUS, OHIO 1

REPORT INSPECTION N0.- 99900276/86-01 RESULTS: PAGE 3 of 7 ment the evaluation of all customer and internal specifications. The SF-105A form eliminated the Drafting Supervisor from the Design Peview committee. The project engineer failed to submit the review packages to the QA manager for his review.

SCI in letter dated February 4, 1983 to the NRC stated that the engineering department was reorganized. The reorganization eliminated the project engineer and drafting supervisor jobs and created the production engineering supervisor. Consequently, SCI rev.ised the procedure to insure that organizational changes are reviewed by the SCI QA manager for possible impact on the quality program. A routing sheet which accompanies the documents requires the reviewers to sign and date the routing sheet to indicate their concurrence. The latest revision of the SF-105 form is E. During the current inspection the inspector reviewed the SF-105-E form which was completed for the Virginia Electric Power Company's purchase order for eight inverters dnd determined that SCI was in compliance with their QA manual require-ments.

E. INSPECTION FINDINGS AND OTHER COMMENTS:

1. Action taken on Capacitor Problems Cn August 28, 1984, Louisiana Power t; Light Company, owner of Waterford-3 nuclear power plant reported significant construction Deficiency No. 116 in accordance with the requirements of 10 C/R 50.55(e). The report identified that capacitors with identification Code CDEKBXK1056PI, Style 020138 with date code 8139 installed in static uninterruptable power supply (SUPS) units manufactured by SCI failed. SCI informed the NRC in a letter, dated February 6, 1985, that the failure rates are higher than normal for some types of ,

capacitors installed in their equipment during the period May 1981 through June 1983.

SCI determined that capacitors manufactured by Electric Utilities Company (EUC) or Cornell-Dubilier (CDE) with date codes 8117 through 8326 installed in their SUPSs may have higher than normal failure rates. In letters dated February 4,1985, SCI informed all their nuclear power plant customers of the problem and suggested that they inspect the capacitors in their SUPSs to verify the date codes. SCI furnisned the telephone number to call if they had problems. SCI informed the inspector that they did not receive any replies to their 113 i

ORGANIZATION: SOLID STATE CONTROLS, INC.

COLUMBUS, OHIO REPORT INSPECTION NO.- 99900276/86-01 RESULTS: PAGE 4 of 7 letter from their customers. Review of the Field Service Reports indicated that there were no high failure rates in 1985 and 1986. The normal service life of capacitors is about 5-6 years and SCI recommends replacement of capacitors periodically to maintain the integrity of the SUPSs.

2. Review of Purchase Orders During the inspection, SCI was manufacturing SUPSs for Virginia Electric Power Corporation (VEPC0), for installation at the Surry nuclear power station. VEPC0 purchase order (P0) ET-16288 dated December 9, 1985 to supply eight SUPSs consisting of inverters, rectifier / chargers, regulating line conditioners, static switches and manual bypass switches. The requirements of 10 CFR Part 21 and 10 CFR Part 50 Appendix B were applicable. Certificates of Conformance were required. Specification NUS-2061 Revision 1 dated November 27, 1985 specified the technical requirements.

SCI internal job order (J0) for this P0 is 27,135 JC dated June 5, 1986. The J0 identified that this equipment is class 1E intended for installation in a nuclear power plant. Additionally, the J0 was '

properly stamped " CONTROL JOB CLASS 1E."

A Design Review Committee Report, Form SF-105E, was prepared as required after the JO was reviewed by the Design Engineer, Test Supervisor, Product Design Supervisor, and QA Manager.

3. Review of SCI Procurement Activities The inspector reviewed the SCI procurement activities by examining the evaluation of suppliers, and SCI P0s issued to vendors.

a. SCI evaluates their suppliers and qualified suppliers are listed on an Approved Vendor List consistent with their QA manual requirements. The inspector reviewed SCI Supplier Quality Assurance System Evaluation Surveys performed on three Vendors and determined them to be acceptable.

b. The inspector reviewed SCI P0s to five vendors for the supply of various components. SCI stamped all the P0s to indicate that 10 CFR Part 21 was applicable.

114 I

l

ORGANIZATION: SOLID STATE CONTROLS, INC.

COLUMBUS, OHIO REPORT INSPECTION RESULTS: PAGE 5 of 7 NO.: 99900276/86-01 I

c. The inspector informed the SCI QA manager that SCI issued P0s to industrial distributors for safety-related equipment who may not inform their principals of the 10 CFR Part 21 requirements.

The SCI QA manager informed the inspector that he will contact the distributors of components to stress the importance of the 10 CFR Part 21 statement on the P0s issued to them, so that the manufacturers are aware that the equipment manufactured by them is being installed in nuclear power plants.

4. Review of Receipt Inspection Activities SCI QC inspectors perform 100% inspections on all material received if the quantity is fifty or less. For larger quantities, inspections are performed based on a sampling plan to meet MIL-Q-105D require-ments. A QC inspector affixes an " Accepted" stamp on the device af ter he inspects it and determines it acceptable and validates the stamp with his signature and date. If the QC inspector determines that a component is unacceptable, he affixes a " Rejected" tag to the component, stores the component on a shelf which is exclusively reserved for rejected material, and sends copies of the P0 to the QA manager identifying the unacceptable attributes. The NRC inspector discussed with a QC inspector the inspections he performed to accept circuit breakers and capacitors and determined the following:

a. The QC inspector stated that he uses Circuit Breaker Inspection Procedure SCI-QA-15-3 Revision 1 dated March 11, 1983 to inspect circuit breakers for the following attributes,

a. Physical damage

b. Name plate data to match purchase order requirements

c. Verify the operation of the breaker with an ohmmeter across the terminals

d. Verify the operation of the auxiliary switch, shunt trip attachment or any other device specified The QC inspector stated that if he determines that the circuit breaker is acceptable, he affixes an " Accepted" stamp to the device and validates it with his stamp and date. If he deter-mines the breaker unacceptable, he affixes a " Rejected" tag and sends copies of the purchase order to the QA manager to obtain disposition.

115 l

ORGANI7ATION: SOLID STATE CONTROLS, INC.

COLUMBUS, OHI0 REPORT INSPECTION NO.- 99900276/86-01 RESL'LTS: PAGE 6 of 7

b. The QC inspector stated that he inspects cylindrical capacitors according to SCI drawings 010099 to 010119. These drawings list the part numbers of various sizes of capacitors and indicate their respective characteristics such as manufacturer, minimum capacitance, minimum voltage, diameter and length. The QC inspector stated that for capacitors received in qdantities less than 50, he performs 100% inspections. He stated that he utilizes a sampling plan with Acceptance Quality Levels meeting MIL-Q-105D to inspect more than 50 capacitors.

c. The NRC inspector observed that all unacceptable material was stored on a shelf. The NRC inspector reviewed several " Rejected" tags and determined that sufficient information was provided on the tag to identify the discrepancy.

d. SCI employs two Quality Assurance Auditors. The NRC inspector.

reviewed the qualifications of both the auditors and determined them acceptable.

5. Review of Tests Performed on SUPSs SCI performs tests on standard SUPSs completed utilizing procedure SCI-QA-5 Revision 6, dated May 23, 1985. The test procedure is adequate and lists all the standard tests. The procedure was modified to meet additional requirements specified in the VEPC0 specification. The inspector reviewed the final test report for SUPS identified as 271350401 shipped to VEPCO and determined that the test results r'et the VEPC0 specification requirements.

6. Review of Calibration Activities The NRC inspector randomly selected the serial numbers from crimping tools and measuring equipment on the assembly floor and reviewed their calibration records. The inspector determined that the calibra-tien of all the equipment was current. The calibration records indicated that standards traceable to the National Bureau of Standards were used to calibrate the SCI equipment.

7. Results of Inspection The results of the inspection are as follows:

a. The failure of capacitors installed in SCI SUPS were limited to units manufactured between 1981-1983. Subsequent failures have not been reported.

I 116 l

l t

ORGANIZATION: SOLID STATE CONTROLS, TNC.

COLUMBUS, OHIO REPORT INSPECTION RESULTS: PAGE 7 of 7 NO.: 99900276/86-01

b. The SCI QA manager revised procedure SCI-QA-20 to specify the use of Field Service Reports to trend the failure of various components prior to the conclusion of the inspection.

E. EXIT INTERVIEW:

The inspector met with the QA manager and the Quality Assurance Auditors, and discussed the scope of the inspection and the findings.

F. PERSONS CONTACTED:

W. G. Hampton, Manager Quality Assurance R. J. McClung, Quality Assurance Auditor D. Dellinger, Project Engineer l

l 117

r ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REPORT INSPECTION INSPECTION NO.: 99901058/86-01 DATE: 5/19-23/86 ON-SITE HOURS: 205 CORRESPONDENCE ADDRESS: Southern Company Services ATTN: Doug Dutton Vice President Post Office Box 2625 Birmingham, Alabama 35202 i

ORGANIZATIONAL CONTACT: Doug Dutton, Vice President TELEPHONE NUMBER: (205) 870-6011 NUCLEAR INDUSTRY ACTIVITY: Design and engineering services for operating plants and plants under construction within the Southern Company Organization.

I ASSIGNED INSPECTOR: [ww /0 2-8 6 R. P. Correid, Special Projects Inspection Section Date OTHERINSPECTOR(S): P. D. Milano, SPIS K. C. Leu, SPIS S. V. Athavale. Quality Assurance Branch T. Del Gaizo, Consultant fiD. Golden,(onsultant

APPROVED BY:

• " , I ( d$

/Juhn W. Craig, Chief. SPIS, Mendor Program Branch ate INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 21,10 CFR Part 50 B. SCOPE: The inspection consisted of an evaluation of design and engineering activities performed for the E. I. Hatch plants and J. M. Farley plants.

PLANT SITE APPLICABILITY: E. I. Hatch plants (50-321, 366) and J. M. Farley (50-348,364) l l 119

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REPORT INSPECTION N0.: 99901058/86-01 RESULTS: PAGE 2 of 18 i

A. VIOLATIONS:

No violations were identified during the inspection.

B. NONCONFORMANCES:

1. Contrary to 10 CFR Part 50, Appendix B, Criterion III, and SCS procedures 10502.7-5 and 220100.7-5 for identification of structures, systems and components, procedures for calculation and domestic 10502.4-4,4-5(Hatch)and 220100.4-4, 4-5 drawing) preparation,(Farley , do not adequately provide for the ide classifications of equipment / materials. Equipment / material safety classifications are shown on equipment lists only and not on applica-ble drawings, calculations and other supporting documentation.

(86-01-01)

2. Contrary to 10 CFR 50, Appendix B, Criterion III, and SCS procedures 10502.4-4 (Hatch) and 220100.4-4 (Farley) calculations SNE-86-002, Rev. 0., 85082-MP, Rev. 3 and SNC-85-098, Rev. O, do not consistently contain required information such as the calculation number, page and sheet numbering listed on each page, the signatures of the preparer and reviewer, and applicable codes and standards ;and sources of input data. (86-01-02)

3. Contrary to 10 CFR 50, Appendix B, Criterion III and IEEE-344-1975, Section 4, the SCS safety evaluation for modification package DCP-85-165 and calculation SNC-85-95 did not include a seismic qualification for the fuse and fuse holder. (86-01-03)

4. Contrary to 10 CFR 50, Appendix B, Criterion III and IEEE-279-1971, Section 3.9, SCS modification packages for plant Farley 82-0-1346, 83-2-2387 and 84-0-2642 did not include an analysis to establish margins for replacement instrument settings although the new instruments characteristics (accuracy, drift, minimum resolution, response time, etc.) are different from the instruments which were replaced. (86-01-04)

5. Contrary to 10 CFR 50, Appendix 2. Criterion III, SCS precedures 10502 and 220100 Section 4.4, and ANSI N45-2.11, Section 4.2, supporting documentation for modification packages PCR 82-1-1306, DCR 85-051, PCR 82-0-1345, DCR 79-134-1, DCR 83-218, DCR 83-262 120

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA PEPORT INSPECTION RESULTS: PAGE 3 of 18 N0.: 99901058/86-01 and DCR 85-215 did not contain sufficient documentation such that verification of the adequacy of the design could be completed without recourse to the originator. Also, field changes were not subjected to the same design control measures commensurate with those applied to the original design. (86-01-05)

6. Contrary to Criterion III of 10 CFR 50, Appendix B, and Section 4.2 of ANSI N45.2.11, Hatch I cable tray calculations for DCR 83-236, SNC 85-083, Cable Tray Support RB-087-A-2 did not provide a reference list, did not specify the analytical methods and did not contain sufficient detail to permit understanding the contents. Farley 1 calculation No. SC-84-1-3052-001, " Missile Door Operator Bracket,"

did not separate criteria from assumptions; and data which was critical to the analysis used in calculating the clearance to the New Fuel Bridge Crane was obtained verbally without written confir-mation. (86-01-06)

C. UNRESOLVED ITEM No unresolved items were identified during this inspection.

D. STATUS OF PREVIOUS FINDINGS None. This is the first NRC Vendor Program Branch inspection of the Southern Company Services. \

E. INSPECTION FINDINGS AND OTHER COMMENTS:

Southern Company Services Overview Southern Company Services (SCS) is one of six companies which comprises the Southern Company. The other five entities are Alabama Power Company, Mississippi Power Company, Georgia Power Company, Gulf Power Company and Southern Electric International. Nuclear units in the Southern Electric System are the Edwin 1. Hatch Nuc1 car Plants, the Joseph M. Farley Nuclear Plants and the Vogtle Electric Generating Plants.

SCS' engineering scope of work includes: site evaluation and selection; specification, bid evaluation and procurement of major equipment; construc- I tion permits and operating licenses; and design and engineering services for both operating plants and plants under construction. During this inspection, only the design and engineering activities for operating plants were reviewed. Current engineerin projects is approximately fifty percent (g manpower 50%) of total dedicated to nu SCS' engineering personnel. The scope of this inspection involved only Hatch and Farley plant activities involving Design Change Requests (DCRs).

121

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REPORT INSPECTION NO.: 99901058/86-01 RESULTS: PAGE 4 of 18 SCS responsibilities for the Hatch project included licensing and engi-neering/ procurement activities for both construction and post operational phases of the project. During the construction of the Hatch plants, SCS was primarily responsible for the balance-of-plant (B0P) design and certain safety-related system design (e.g. Control Building, Diesel Generator Building, Intake Structure and systems within these buildings). SCS was also responsible for administering the Bechtel (A-E) and General Electric (GE) (NSSS) contracts. Currently, SCS is the responsible A-E for engineer-ing and procurement.

The SCS Hatch Project Nuclear Plant Support organization provides managers, utility engineering and discipline directors as well as an on-site staff to support requests from Georgia Power Company. Requests include enoineer-ing assistance, design changes, as-built notice and drawing maintenance for both permanent and temporary on-site support. DCRs nay include design drawings, material specifications, safety evaluations and other supporting documentation. The work package may also include environmental and fire protection evaluations, functional test procedures, craf t work instructions dnd technical specification revisions. In the current DCR system all requests generated from Georgia Power Company are forwarded directly to SCS. Upon receipt of the request, SCS determines if the in-house staff can support the man-power, time and/or expertise required.

During construction phases of the Farley project, SCS responsibilities included licersing, engineering, and procurement. The engineering responsibility during construction was primarily the BOP work for Unit 1.

The Unit 2 B0P responsibilities were transferred to Bechtel. Current SCS' Farley activity includes engineering support for both units primarily for the B0P and other areas as assigned by Alabama Power Company. Mcdifica-tions for the Farley units assigned to SCS by Alabama Power Company may be in the form of a Production Changes Request (PCR) or an Engineering Support Request (ES). These requests are either assigned directly to SCS or Bechtel by Alabama Power Company. Apr oximately 57% of Alabama Power Company's PCR's and ES's are assigned to SCS. PCR's assigned to SCS are processed by the appropriate disciplines nd packaged as Production Change Notices (PCN) and returned to Alabama Power Company for implementation.

PCN packages contain the same type of supporting documentation as the DCR response packages produced by the SCS' Nuclear Plant Support - Hatch.

Upon completion of the modifications, Alabama Power Company develops a Work Completion Notice (WCN) which is sent tc SCS Nuclear Plant Support -

Fa rley. Final documents are then prepared by SCS for official drawing transmittal back to Alabama Power Company.

122

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REFORT INSPECTION NO.: 99901058/86-01 RESULTS: PAGE 5 of 18 SCS Quality Assurance programs for the Hatch and Farley project support groups provides quality assurance guidance to the engineering staff and vendor shop quality surveillances and audits. Responsibilities include evaluation of SCS engineering and procurement procedures for compliance with Safety Analysis Report (SAR) commitments, NRC regulations and guides, industry codes and standards as well as SCS corporate policy. The SCS Quality Assurance (QA) program is structured to comply with ANSI N45.2, Quality Assurance Program Requirements for Nuclear Facilities. Included in the QA program are committments to comply with referenced NRC regu-lations, codes, standtrds, and guides identified in applicable plant SARs.

Yearly audits of the SGS Hatch and Farley engineering work activities are performed by SCS QA personnel to assure that applicable elements of the SCS QA program have been effectively developed, documented and implemented.

10 CFR Part 21 Evaluations The NRC inspector reviewed the Plant Hatch Operational Support Policy and Procedures for the identification, evaluation, and reporting of significant deficiencies. Section 2.4.2 of the procedures manual states the Nuclear Safety and Licensing section (NSL) of the Nuclear Safety and Fuel (NSF) organization is responsible for evaluation of significant deficiencies.

In an interview with SCS NSL staff engineers, the NRC inspector was briefed cn SCS policy and procedures for the identification, evaluation and report-ing of deficiencies as applicable to Plant Hatch. Potential deficiencies reportable under 10 CFR Part 21 are received from Georgia Power Company (GPC) via work requests with pertinent information. All work requests received by NSL are logged and tracked by both the manager and engineer responsible for the evaluation. Monthly status reports are written by the engineering staff to report progress on evaluations to SCS managers and GPC.

Three work requests involving potential 10 CFR Part 21 reportable deficien-cies from GPC were reviewed by the NRC inspector. The following is a brief j

description of each problem and followup evaluations by SCS NSF staff.

1. File HII-A.22.44, " Errors in Engineering Computer Program," Hatch Units 1 and 2. GE reported to GPC that potential nonconservative results using engineering computer program RVRIZ02 may occur if GE guidelines (ref. NEDE) were not followed when using the program.

4 i

123

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REPORT INSPECTION N0.: 99901058/86-01 RESULTS: PAGE 6 of 18 Bechtel, the A-E for the Hatch plants, was contacted and subsequently performed an evaluation of the computer program problem. Bechtel i

concluded that they had used the program as GE had recommended and, therefore, found no problems. SCS had not used the program and, therefore, considered the problem not applicable to Hatch and not reportable under 10 CFR Part 21. The NRC inspector determined that applicable NSF procedures were followed and the item was properly considered closed out.

2. File HII-A.22.28 " Pipe Whip Restraints," Hatch Unit 2. This defi-ciency, discovered by Bechtel on 6/24/83, involved pipe whip restraints which were not designed nor installed on the control rod drive (CRD),

Reactor Water Clean-Up (RWCU), Reactor Core Isolation Cooling (RCIC) and the Auxiliary Steam (AS) systems. The systems were isolated while the required restraints were designed, installed and/or scheduled for installation. SCS conducted an audit of the Bechtel Hatch- .

project's quality assurance program implementation and concluded that the missing whip restraints incident was an isolated event. Bechtel subsequently changed their QA procedures. Followup documentation was completed by SCS personnel and the deficiency evaluation closed out 11/2/83.

3. File HII-A.27.40 "RHR Service Water Pumps." Originally, GPC had requested SCS to review planned modifications to the RHR Service Water (SW) Pumps. SCS contracted Bechtel to perform the RHR-SW pump modifications analysis. Bechtel contacted Johnson Pumps, the original pump vendor, to obtain the required design information.

The modification involved relocation of the pumps support / restraint system to a location out of a potential flood area.

Johnson Pumps, contracted Mcdonald Engineering to perform a stress analysis for the proposed pump modifications. Mcdonald's analysis required that the flanges and their bolts be upgraded to higher strength capabilities (i.e. a thicker flange and higher strength bol ts) . Johnson did not inform 'Bechtel of the design change require-ments. At a later date, Johnson Pump discovered the deficiency and reported it to SCS. SCS inturn initiated the corrective action.

i l

I An SCS audit of Johnson Pump revealed that Johnson Pump had received the design change requirements from Mcdonald Engineering and failed to follow procedures which would have prevented the deficiency. SCS notified GPC of the RHR-SW pump problems. GPC, inturn, instructed SCS to evaluate and initiate corrective action. SCS completed all l procedural documentation and closed out the report on 2/19/85.

l i

l 124

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REPORT INSPECTION NO.: 99901058/86-01 RESULTS: PAGE 7 of 18 Based upon the sample of 10 CFR Part 21 evaluations reviewed, the NRC inspector determined that the NSL engineering staff had followed applicable procedures and had maintained completed, orderly files which were accessible, clear and concise. No items of noncompliance or unresolved items were identified in this area.

Quality Assurance Audit Report Review The NRC inspectors reviewed the " Quality Assurance Technical Audit Report on the SCS Hatch Nuclear Operational Support Group, GPC, EWO 31412Z, 3148ZZ" dated 12/5/85. Two DCR's, which were part of the QA audit evalu-ation, 83-30 and 83-262 were requested by the NRC inspectors for review of these design modification packages.

DCR 83-80 was a non-safety related modification package issued to GPC to increase the size of a steam trap to accommodate drain flow to provide sufficient capacity to operate without using the bypass. Six Field Change Requests (FCR) were prepared as part of this modification package. There were no calculations or any other supporting documentation which addressed piping pressure rating, over-pressure protection and instrument set-points.

FCRs accompanying various checklists to describe possible impacts of the changes on 10 CFR Part 50 Appendix P, Equipment Environmental Qualifications (EEQ) and 10 CFR Part 50.59 safety evaluations did not contain any techni-cal justifications of the approved changes.

Modific6 tion package 83-262, involved the installation of spray water piping and mixing cones to moisture separator reheater drain tanks. In both cases, the audits were found to have been of appropriate scope and definition. The audit findings were documented and deficiencies subse-quently corrected.

Noncnnformance 86-01-05 was identified in this area of the inspection.

10 CFR Part 50.59 Procedure Review

! SCS procedures and interpretations of the requirements of 10 CFR Part 50.59,

! " Changes, tests, and experiments" were reviewed. Appendix D of the SCS l

procedure manuals " Plant Hatch Operational Support Policy and Procedures,"

i parts D-5 and D-6, and " Plant Farley Operational Support Policy and l Prucedures," parts D-8 and D-9, describe the requirements in determining whether safety-related,10 CFR Part 50.59 or unreviewed safety question l

l I

125 l

I l

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA i l

l REPORT INSPECTION N0.: 99901058/86-01 RESULTS: PAGE 8 cf 18 applicability are involved in a design change. Both the Hatch and Farley procedures incorporate a checklist used to make these determinations. In the case of Hatch modifications, all Nuclear Safety Evaluation Checklists are reviewed and signed by a Nuclear Safety and Fuels engineer including those changes permitted in 10 CFR 50.59. In the case of Farley, only those evaluations which required further evaluation as a result of 10 CFR Part 50.59 are reviewed and signed by Nuclear Safety and Fuels engineers.

Written detailed evaluations are only required when one or more of the answers to the questions on the checklist are "YES." In cases where check-lists contained only "N0" responses, a written evaluation as to why these answers were "N0" was not provided.

Changes to the Hatch and Farley procedures manual covering the design input to a modification, design verification, and safety evaluation were reviewed by the NRC inspection team. The Hatch procedure revisions had been imple-mented. The revised safety evaluation checklist requires a review in all cases by Nuclear Safety and Fuels staff engineers and supporting analysis by them to be provided and maintained for each modification. These analyses are maintained in separate files from the modification package.

Drawing and Calculation Procedures Review The NRC team reviewed several electrical, structural and mechanical modifi-cation packages for Hatch and Farley. The team had ger. oral concerns with drawings and calculations which were part of the modification packages not identifying the quality classification (i.e., safety-related, non-safety-related) of the item (s) being modified. SCS' procedures 10502.7-5 (Hatch)and 220100.7-5 (Farley), " Nuclear Quality Classification System for Design, Equipment and Materials" require that the quality classifi-cation code be placed on applicable documents in accordance with the corresponding documents preparation procedure. Procedures for the preparation and approval of domestic drawings (SCS procedures 10502.4-5 (Hatch)and 220100.4-5 (Farley)) require the quality classification of the equipment / material and its Q or non-Q icentity be placed on the drawing if the drawing is a bill of material. Procedures for the preparation and review of design calculations (SCS procedures 10502.4-4 (Hatch) and 220100.4-4 (Farley) do not provide a method of identification of the system, structure or basic component's quality classification [or its Q or non-Q identity].

Nonconformance 86-01-01 was identified in this area of the inspection.

126

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REPORT INSPECTION RESULTS: PAGE 9 of 18 I N0.: 99901058/86-01 l

i Electrical Modification Package Review The NRC team reviewed three electrical modification packages for Hatch Unit 1 and one modification package for Farley Unit I nuclear stations.

These packages were reviewed for (1) consistency with the original design basis requirements, (2) technical adequacy of the chosen design approach, (3) conformance with the applicable regulatory criteria and FSAR commit-ments, and (4) completeness of design details.

The findings identified weaknesses in the system for design control and independent verification of the completed design. The system in place does not effectively address the proper application of safety evaluation reviews for modifications es required by 10 CFR Part 50.59.

1. PCR 82-1-1306: Service Water Dilution Flow Loop This Farley Unit 1 PCR was prepared to relocate service water dilution loop flow switch, FS-580. The modification was necessitated by a technical specification item which required the switch to be operated while observing the operation of steam generator blowdown valve, RC-00238. Additionally, in the state-ment of work to be completed, a change to the 24 ydc power supply for the instrument loop, including this flow switch, was included because it was operating near the limit of its capability. Thus, it was being replaced by a separate 48 vdc supply.

In the original design, the flow switch energized an auxiliary relay which closed contacts in the steam generator blowdown processing system. PCR 82-1-1306 changed the circuit such that the flow switch is now the line contact for the steam generator blowdown processing circuit allowing the elimination of the auxiliary relay.

This PCR also deleted the function of the flow switch in the waste processing system. This portiori of the change was not addressed in the statement of work for PCR 82-1-1306. The original design for l

l the waste gas discharge valve solenoid had a contact from the vent l stack radiation monitor which shut the valve on a radiation alarm.

In a September 1977 design change, the auxiliary relay contact from the dilution loop flow switch was added in parallel to the radiation monitor contacts altering the circuit design. Thus, a vent stack radiation monitor alarm would not cause the waste gas discharge valve i

to shut unless the dilution loop flow switch auxiliary relay contacts were also open. The design change made in 1977 also failed to address l the addition of a signal from the steam generator blowdown processing I

circuit to the waste gas discharge valve logic, t

127 l

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REPORT INSPECTION NO.: 99501058/86-01 RESULTS: PAGE 10 of 18 Although PCR 82-1-1306 deleted the contact in the waste gas system, the original design criteria for this circuit was violated by the September 1977 modification. As a result of questions asked during this inspection, SCS personnel and Bechtel personnel reviewed drawings and made an initial determination that PCR 82-1-1306 and the September 1977 change had been completed. Subsequently, plant personnel at Farley Unit I found that while the relays had been installed, the electrical termination changes had not been made.

The normal practice for PCR's is that following receipt, Farley Unit I completes the change and returns the as-built information to SCS. There was an apparent failure to ensure that a change had been properly completed in acccrdance with PCR 82-1-1306 and the September 1977 modification.

Questions concerning the failure to install the modifications and the as-found installation of this equipment which is inconsistent with the as-built drawings reviewed during this inspection will be forwarded to the NRC Region II office for information and appropriate action.

For the portion of the design change which described the 24 vdc power supply, the capability of the instruments to operate with a 48 vdc supply was reviewed. The SCS personnel stated that the instruments had dual voltage input terminals. However, this information was not included in the documentation, and would not be available for the design reviewer / checker.

Nonconformance 86-01-05 was identified in this area of the inspection.

2. DCR 85-051: Jumper Cell 36 in 250 volt dc Station Service Battery This Hatch Unit 2 design change request was provided to SCS for review. The development of the information for the change and its justification was done by Bechtel.

The documentation package provided during the inspection included only a technical and licensing evaluation of the proposed modifi-ca tion. The documentation package did not include information for other areas which would be necessary for a complete review and verification of the change request. While the technical justi-fication stated that the jumpering of a battery cell (cell 36) would not reduce the battery voltage under load below technical specification limits, it did not specify any service test to verify the load capacity requirements. Further, the DCR did not 128

l L ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMIPGHAM, ALABAMA REPCRT INSPECTION N0.: 99901058/86-01 RESULTS: PAGE 11 of 18 address any changes to the battery charger equalizer or float voltage settings nor did it discuss the selection of a new pilot cell since cell 36 was a pilot cell. Finally, the package did not contain the specification requirements for the jumper cables.

In discussion with SCS personnel at the plant to obtain information on these concerns, the inspectors were told that the DCR was not implemented. Instead, the battery had been completely replaced.

The DCR, however, had not been cancelled as of the date of the inspection.

Nonconformance 85-01-05 was identified in this area of the inspection.

3. PCR 82-0-1345: Lube Oil Temperature Control for 1C and 2C Diesels This Farley PCR modified the temperature sensing and control for the IC and 2C emergency diesel " keep warm" lube oil systsm. Since the original design had the heater temperature switches on the discharge line in close proximity to the heater, the heater was subjected to excessive cycling. The PCR provided for a resistance temperature detector (RTD), transmitter, and controller circuit '

with the RTD located in the lube oil sump.

The design thange was reviewed and several areas were identified in which the documentation was not complete. Since the new control system was composed of a loop of instruments versus the original switch, a failure modes and effects analysis (FMEA) should have been perforned to assure that no decrease in reliability would result from the modification. Also, since this system would now be performing the low lube oil temperature ali.rm function, the consequences of an open RTD failure should have been addressed.

Because of the changes to the low temperature alarm circuit, the procedures and setpoint calculation sheet were requested for the I

instrument. It was found that procedures are not available and a revised setpoint calculation had not been performed. SCS felt that t

the design change provided a system of better accuracy such that a change to the lube oil low temperature alarm was not necessary.

In a letter dated July 31, 1985, the diesel manufacturer, Colt, provided a response to Alabama Power Company's request for review of the proposed modification. Colt responded that the r.ew design 129

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REPORT INSPECTION NO.: 99901058/86-01 RESULTS: PAGE 12 of 18 was more complicated than necessary and provided an alternate design.

These comments were not accepted by SCS and no documentation of the evaluation of Colt's response was available.

Nonconformance 86-01-05 was identified in this area of the inspection.

4. DCR 85-054: RPS Scram Relay Replacement This Hatch Unit 2 design change request was developed to replace the original GE Model 105 relay contactors with a new Model 305 relay. Since the original units are not available, the vendor was providing the new equipment as an acceptable and improved alternate.

The inspectors reviewed the new relay's ability to meet the seismic requirements for each location in the control panel. The Model 105 relays were tested seismically to functional failure. However, the Model 305 relays were tested to a lower seismic input. The documen-tation package contained a statement that it was " felt" that the new relays are acceptable seismically. Further documentation was requested and reviewed which showed that the calculation met the seismic criteria.

5. DCR 85-218: AC breakers Overcurrent Trip This Hatch modif1 cation package for replacing the existing EC-2A type s overcurrent trips with " micro versa" trip units for safety related AK series breakers, and contained the related calculations SNE-86-002, Rev.0, 85082-MP, Rev. 3, and SNC-85-098, Rev 0. These calculations were part of the final modification package. The team noted that the calculations did not have proper page and sheet numbers. Some pages did not have a calculation number, page number, or signatures of the preparer and reviewer. Calculations did not have references to the applicable codes and standards, and sources of input data were not identified. Based upon these findir.cs, these documents had not been independently verified as required by SCS procedures.

Nonconformance 86-01-02 was identified in this area of the inspection.

6. DCR 85-165: Seismic Qualifications of Fuses and Box This Hatch DCP added 30 amp fuses to the shunt connected dc ammeter circuit of the 125 volts dc station distributier, system. The team reviewed the safety evaluation for this modification which states the 130

) .

ORGANIZATION: SOUTHERN COMPANY SERVICES -,

BIRMINGHAM, ALABAMA f REPORT INSPECTION RESULTS:

' 'PAGE 13 of 18 N0.: 99901058/86-01 ,

equipment being added is not required [to function af ter a' seismic event. Based on this safety evaluation, the' civil department calcu-lation SNC-85-95 dated 12/19/85 concluded that seismic qualification of the fuse and the fuse holder was not,reqtilred. ,

?

Although the fuse and fuse hold (r will not be required to function-df ter a seismic event, the nonqualified fuses and fuseholders were installed in a class 1E panel and may become dislodged during a seismic event and act as a missile and damage other class.1E components and wiring inside the class 1E panel Yr may fail such that a fault in the 1E circuit is generated. :The team reviewed the analysis of this system by the licensee for 10 CFR Part 50 Appendix R requirements and noted that a fault developed in the circuit can lead to unavailability of class 1E,125/250 v.dc switchgears 1A and IB.

Nonconformance 86-01-03 was identified in't.his area of the inspection.

7. pCN-S-85-1: Replacement Of The Level Switch On The Reactor Makeup

' Water Tank e This PCN involved the replacement of an existirg instrument in the safety system with a new instrument or a string of instruments consisting of a primary sensor, signal conditioner and bi-stables.

These new instruments were set to the settings of the replaced instruments. The applicability of the old instrument settings to the new instrument was not documented. This analysis is required since the characteristics (accuracy, drift, response time, etc.)

of the new instruments are different than the old ones. The team was informed by SCS personnel that the individual instruments settings were provided by the vendor? of the instruments. SCS does not have a procedure by which an analysis of a system may be performed as a result of instrumentation changes.

Nonconformance 86-01-04 was identified in this area of the inspection.

8. DCR 84-80: Fire Wrapping of Cables The team reviewed DCR-84-80 Rev. O issued for the addition of fire wrapping on cables on elevation 130. The team found that the fire wrap used was TSI, Thermo-lag type and was rated for one hour.

10 CFR Part 50 Appendix R, Section III G-2-C requires that the fire area where one hour fire wrap is used shall have fire detectors and an automatic fire suppression system. The team was informed that all the fire areas were not equipped with the fire detection and protection equipment but exemptions were granted by the NRC via a 131

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA l l

REPORT INSPECTION NO.: 99901058/86-01 RESULTS: PAGE 14 of 18 letter, dated April 18, 1984. The team reviewed two calculations for derating of the cable current capacities due to use of fire wrap and found that the correct derating factor was used. The calculations did not address other derating factors and factors for temperature corrections. However, the team was informed that complete ampacity calculations were performed under a different calculation number which was reviewed and found to be acceptable.

Seismic Modification Package Review Modification packages for seismic modifications at Farley 1, 2 and Hatch 1, 2 were reviewed.

The inspector audited a selected sampling of modification packages at Farley 1, 2 and Hatch 1, 2 to determine whether the applicable regulatory, technical, and QA requirements were included. These documents are listed below:

Farley Production Change Request (PCR), PCR 84-2-3048(2113PB)

Farley Missile Door Operator Bracket Mod., SC-84-1-3052-001 Farley Unit 2 PCR, PCR-84-2-2544 UP (2115)

Hatch Field Deviation Requests, DCR 86-165 Hatch Support Documents, DCR-83-236 Hatch 1 Reactor Bldg Support Documentation (4 Volumes), DCR-85-144 Hatch 1 Reactor Bldg Support Documentation (4 Volumes), DCR-85-117 Hatch 1, Appendix R Reactor Bldg Wrap Torus, DCR-83-236, SNC-84-083 Hatch 1, Appendix R Reactor Bldg. EL.87 and EL.130 Re: Qualification of cable tray and conducts, Books No. I and No. 4 Summary of DCR-85-117 Cable tray and Conduct Support Modification Hatch Field Deviation Request, DCR-86-148 Hatch Field Deviation Request, DCR-86-114 The following are findings with respect to Hatch 1: (DCR 236, SNC 85-083, cable tray support RB-087-A-2, Book I of 8)

a. References used in the analysis and calculations were not specified,

b. Methods used in obtaining the seismic loads were not specified.

c. Analysis and calculation results were not in sufficient detail to facilitate the verification of results.

132

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REPORT INSPECTION RESULTS: PAGE 15 of 18 NO.: 99901058/86-01 The following are findings with respect to Farley 1: (Cal. No.

SC-84-1-3052-001, " Missile Door Operator Bracket Modification")

a. Criteria and assumptions were not separated;

b. A dimension used in calculating the clearance to the New Fuel Bridge Crane was obtained verbally without written confi rmation.

Nonconformance 86-01-06 was identified in this area of the inspection.

The NRC inspector also reviewed procurement procedures and checked appli-cable 10 CFR Part 21 implementation. One document related to Farley 2 purchase order SS85-1859 (PCR 84-2-3048) concerning modification of the Missile Door Bracket (dated 9/12/85 from SCS to Overly Manufacturing Company) did not specify the requirements of 10 CFR Part 21 in the P0.

After a discussion with SCS personel involved with this P0, a document containing the Part 21 specification was produced from another source.

Mechanical Systems Modification Reviews The NRC team reviewed several mechanical modification packages for Hatch Units 1 and 2. The packages were reviewed for (1) consistency with the original design basis requirements, (2) technical adequacy of the design approach, (3) confortaance with applicable regulatory criteria and FSAR conunitments, and (4) completeness of design details.

In' design packages compiled in the late 1970's and early 1980's, a number of concerns were identified, largely related to documentation. Speci fi-cally, the number and type of calculations or analysis needed to support the calculations themselves did not always clearly identify the source of input, assumptions for later verification, and acceptance criteria. More

! recent packages reviewed (e.g.86-055) indicate that many of the prior problems have been addressed and corrected. Draft revisions of project l

procedures on design input and desigri verification were reviewed and dedress the documentation and quality of the packages.

l 1.. DCR 79-134-1: Modification to Containment Penetrations X-26 and X-220 1

This package consisted of the design sketches and a number of field change requests. It did not contain: 10 CFR Part 50.59 review, a safety evaluation, a design input sheet, a list of installation instruments required, or purchase specifications.

Nonconformance 86-01-05 were identified in this area of the inspection.

l 133

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA G

REPORT INSPECTION N0.: 99901058/86-01 RESULTS: PAGE 16 of 18

2. DCR 83-218: Motor Cooling Modifications to PSW and PHR SW Pumps Calculation SNH-83-004 (determination of PSW flow to RHR SW pump motors with failed sightglasses downstream of PSW motor coolers) did not demonstrate minimum cooling flow still existed following the piping change. Similarly, calculations were not performed to demon-strate the effects of removing relief valves from the PHR SW piping.

Where engineering judgements were used in lieu of calculations, the judgements were not documented. The calculation did not . state the acceptance criteria and therefore there was no way of determining whether or not the results were acceptable. Similarly, calculation assumptions #1 and f7 were not justified. The calculation did not contain necessary information to

• 11t independent review.

Nonconformance 86-01-05 was identified in this area of the inspection.

3. DCR 83-262: Modifications to Heater Drain Tank Piping The package only contained FCRs accompanied by various checklists used to describe possible impacts of the modification on 10 CFR Part 50, Appendix R. Electrical Equipment Qualification (EEQ), and Safety Evaluations. There was no indication of a technical justifi-cation of the approved field changes.

Nonconformance 86-01-05 was identified in this area of the inspection.

4. DCR 85-215 and 86-063: Deletion of Head Spray, Plant Hatch Units 1 and 2 The head spray system on boiling water reactors is intended to cool down the reactor pressure vessel (RVP) head upon plant shutdown.

For plants where cooling the RPV head is critical, head spray is beneficial. But for plants for which RPV head cooling time is not critical, such as Hatch, it provides no benefit (ref. GE memorandum MDE-109-1284).

Design packages for this modification were reviewed. GE recommended removal of head spray as part of the Hatch Performance Improvement Program. An analysis performed by GE for removal of head spray provided the justification, a safety evaluation and alternate means for accomplishing removal. SCS developed the sketches to remove the head spray piping between the RPV nozzle flange and a flange at elevation 202 ft., 4 inches downstream of valve 511-F019, and for the installation of blind flanges. Additionally, pipe support E11-RHR-402 was found to require the addition of a north-south directional res train t. A sketch was provided for this purpose.

134

ORGANIZATION: SOUTHERN COMPANY SERVICES I BIRMlNGHAM, ALABAMA REPORT INSPECTION RESULTS: PAGE 17 of 18 NO.: 99901058/86-01 Technicai specification changes were not made to allow locking and racking cut the breakers for valves E11-F002 and E11-F023 as described in the modification. Upon discovery of this problem, GPC was notified l i

! by SCS not to perform this part of the modification. A hand calcula-l tion was performed by SCS to determine if pipe support E11-RHR-402 was  ;

adequate to withstand loadings with water in the piping. Previous 1 analyses assumed the pipe would be empty. This calculaiton was found to be unverifiable because the following items were not clearly stated: calculation number, design criteria, assumptions, calculation methodology, input to the calculation, references, or conclusions.

SCS provided a second calculation on the head spray piping analysis.

This calculation was performed during the inspection in response to discussions on the first calculation.

The second calculation, SM04036-060, dated May 22, 1986, was reviewed.

The seismic loading on the pipe supports after removal of portions of the head spray piping by DCR-85-215 was calculated using ccmputer code PIPESD. The calculation was in a format consistent with SCS procedures The calculation appeared to be technically correct.

A pipe support stress calculation in support of an FCR generated as part of the head spray deletion modification on Plant Hatch Unit 1 (DCR 85-215) was not independently verifiable. This calculation was done to confirm that the bracing added to pipe support E11-RHR-402 under FCR 85-215-3 would be adequate. This calculation was found to be unverifiable because the calculation did not clearly state:

design criteria, assumptions made, input to the calculation, calcu-lation methodology, refert.nces, or conclusions.

Prior to the end of the inspection, SCS engineers completed a computer calculation for the seismic response of the head spray piping as modified by DCR 85-215 and associated FCR's. This calculation was found to be independently verifiable and appeared to be technically correct.

f Nonconformance 86-01-05 was identified in this area of the inspection.

135

I t.

ORGANIZATION: SOUTHERN COMPANY SERVICES BIRMINGHAM, ALABAMA REPORT INSPECTION N0.: 99901058/86-01 RESULTS: PAGE 18 of 18

5. DCR 86-055: Hydrogen Water Chemistry Test, Plant Hatch Unit 1 The addition of hydrogen to the reactor coolant system in BWR's to scavenge oxygen may result in reduced stress corrosion cracking in recirculation piping. The purpose of this test is to verify the efficiency of hydrogen analysis in the water chemistry control system in the plant and to determine the additional shielding requirements due to the increased nitrogen 16 carryover.

The design package for the test was reviewed. Since this design package did not include any design calculations by SCS and relied on work by GE, it was reviewed only for its format as compared to earlier design packages. This design package showed a higher level of organi-zation, including logs and checklists for the contents of each section.

The inspector concluded that the method of organization in this design package is an improvement over older design packages which were reviewed during the inspection.

i 136

ORGANIZATION: UNISTRUT CORPORATION WAYNE, MICHIGAN INSPECTION INSPECTION REPORT ON-SITE HOURS 36 j NO.- 99900162/86-01 DATES: 6/2-6/86 CORRESPONDENCE ADDRESS: Unistrut Corporation ATTN: Mr. Timothy B. Conduit  ;

i Senior Vice President '

35005 Michigan Avenue West Wayne, Michigan 48184 j ORGANIZATIONAL CONTACT: Bruce McClure, Material Superintendent  !

TFIFDMANF NUMRFQ* (119) 791 anan NUCLEAR INDUSTRY ACTIVITY: Manufacture of fasteners and channels.

ASSIGNED INSPECTOR: u A nzf/fy4 J. C. th$per, ReactivefInspection Section (RIS) Date OTHERINSPECTOR(S):

f'[es FC APPROVED BY: .

E. W. Merschoff, Chie S, Vendor Program Branch Date INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 21 and 10 CFR Part 50 Appendix B.

B. SCOPE: The inspection was made as a result of 10 CFR Part 50.55 (e) reports from Byron and Braidwood of defective spot welds on P1004A Unistrut material used in the fabrication of cable tray and conduit hangers. In addition, the implementation of the Unistrut QA program was evaluated.

PLANT SITE APPLICABILITY: Braidwood 50-456, 50-457; Byron 50-454, 50-455; Watts Bar 50-390, 50-391; Comanche Peak 50-445, 50-446.

l 137

ORGANIZATION: UNISTRUT CORPORATION WAYNE, MICHIGAN REPORT INSPECTION NO.- 99900362/86-01 RESULTS: PAGE 2 of 9 A. VIOLATIONS:

1. Contrary to Section 21.6 of 10 CFR Part 21, the Unistrut Corporation did not post the requirements of 10 CFR Part 21, Section 206 or a Part 21 procedure. (86-01-01)

2. Contrary to Section 21.21 of 10 CFR Part 21, the Unistrut Part 21 procedure, (" Processing of Reportable Events") does not adequately describe time limits for notification of a known defect or identify (86-01-02) to the Commission.

3. Contrary to Section 21.31 of 10 CFR Part 21, the Unistrut Corporation failed to impose 10 CFR Part 21 on purchase orders to vendors of calibration, material suppliers (mechanical properties and chemical certification) and suppliers of heat treating services. (86-01-03)

B. NONCONFORMANCES:

1. Contrary to criterion V of Appendix B to 10 CFR Part 50 and paragraph 2.0 of the QA procedure 20-01-003 Revision 2, 48 ft of red ticketed rejected material # HTHR075A00 was not separated from usable material and placed in the MRB designated area. (86-01-04)

2. Contrary to Criterion V of Appendix B to 10 CFR Part 50 and Section XIX paragraph 19.6 of the QA manual, external audits were not perform on at least eight vendors of calibration, heat treating and material suppliers in the years between 1984-1986. (86-01-05)

3. Contrary to paragraph 19.3 various internal audits scheduled in 1986, 1985, and 1984 were not completed and/or done according to schedule.

Furthermore, implementation of corrective actions were not being completed in a timely fashion. (86-01-06)

4. Contrary to criterion V of Appendix B to 10 CFR Part 50, Section III of QA procedure 20-01-031, and Section 8 of ASNT SNT-TC-1A, there were no records of annual eye examinations for a Level II and a Level III NDE inspector. Two Level II NDE inspectors and one Lev J III NDE inspector had no record of having an annual review of cert ; cation, the former in 1985 and the latter never. (86-01-07)

5. Contrary to Criterion V of Appendix B to 10 CFR Part 50, and Section III of the QA procedure 20-01-024 Revision 2, documentation did not exist to show that a new inspector was given his first four weeks 138

ORGANIZATION: UNISTRUT CORPORATION >

WAYNE, MICHIGAN

! REPORT INSPECTION NO.- 99900362/86-01 RESULTS: PAGE 3 of 9 written progress report. The personnel qualifications files for two inspectors were incomplete and nonexistent for one inspector.

(86-01-08)

6. Contrary to Criterion V of Appendix B to 10 CFR Part 50, CA procedure 20-01-030, ASNT SNT-TC-1A Section 2, four lead auditors had no evi-dence of having their audit communication skills signed and dated on their record of lead auditor qualifications. Four lead auditors had no record of audit training. Two lead auditors had no objective evidence of audit participation. (86-01-09)

C. UNRESOLVED ITEMS:

None.

D. STATUS OF PREVIOUS INSPECTION FIDNINGS:

None.

E. OTHER FINDINGS OR COMMENTS:

1. Inspection Issues The VPB inspection of the Unistrut Corporation was made as a result of 10 CFR Part 50.55(e) reports from Commonwealth Edison's Braidwood Unit 2 (March 6,1986) and Byron Unit 2 (March 12,1986) which reported defective resistance welds for cable tray hanger material.

The suspect material, P1004A, consised of two single P1000 channels spaced back-to-back and joined by two side plates with four rows of spot welds at three inches on center.

The defective resistance welded material was identified at Braidwood Unit 2, during the installation 'of cable trays in containment by the electrical contractor G. K. Newberg . The cable tray hangers were prefabricated by Systems Control Corporation (SCC) using the P1004A Unistrut material.

Unistrut made a visit to the Braidwood site and confirmed that the P1004A material in question was their product. Commonwealth Edison reviewed the control numbers of the discrepant material and concluded that the majority of the material was manufactured in late 1978 or 139

l ORGANIZATION: UNISTRUT CORPORATION WAYNE, MICHIGAN REPORT INSPECTION NO.- 99900362/86-01 RESULTS: PAGE 4 of 9 in early 1979. It was concluded that 580 feet of rejected P1004A material was inadvertently shipped to SCC by Unistrut and used for cable tray hangers for the Byron /Braidwood project, and primarily used at Braidwood Unit 2.

During April 1978, the Pennsylvania Power and Light Company

issued a 10 CFR Part 50.55(e) regarding inadequate spot welding l of Unistrut material used in field fabricated supports for safety related electrical raceways, HVAC ducts and instrumentation panels.

As a result, the NRC performed an audit on Unistrut in March 1979.

According to the NRC inspector that performed this audit, " Prior to the detection of failed spot welds at the job site, the Unistrut material had been purchased as standard off-the-shelf items, with no formal Quality Assurance /QC program involved. Testing was performed by QC on a random basis, however, this was very limited in scope....It was also determined that monitoring of weld machines and welding operators was virtually non-existent...The assessment of the failed spot welds brought about the creation of a formal Quality Control (QC) program, with definitive procedures for welding, testing, and calibration of welding equipment...."

The Unistrut Corporation provided the inspector with a draf t of a report dealing with the circumstances surrounding the defective P1004A material supplied to the Byron and Braidwood project. Ba-sically the report identified late 1978 and/or early 1979 as being the time frame in which the material was originally welded. The material was rejected and scheduled for rework. Unistrut concluded that a partial rework (rewelding) of at least some of the material was initiated at that time, but before the rework was completed and a QC inspection performed, the material was inadvertently

! shipped to a customer. During a Commonwealth Edison walk down of the P1004A mateiral at Braidwood and Byron Units 1 & 2, discrepant resistant welds were noted. As a result of the Unistrut investi-gation, it was concludeo that control numbers with the prefix 7 were not of concern since the frequency rate of failed welds for these control numbers were extremely low and inherent for the process.

Unistrut contends that the discrepant material was produced in the late 1978 and early 1979. At that time Unistrut P1004A was being produced on three different resistant welding machines, the single heat spot welder #356, the multihead spot welder #6901, and single head spot welder #6940. Machines #6901 and #6940 were new equipment 140

ORGANIZATION: UNISTRUT CORPORATION WAYNE. MICHIGAN REPORT INSPECTION RESULTS: PAGE 5 of 9 NO.- 99900362/86-01 acquired in 1978 and 1979. Ur.istrut believes that during the prove-out stage for these new machines, some rejected material may have been produced. Unistrut contends that a partial rework (re-welding) of at least some of the material was initiated. However, before the rework was completed, 580 ft of discrepant material was in-advertently shipped to Commonwealth Edison.

The NRC inspector concludes that technical problems existed with resistance welding machines and techniques at Unistrut at the time that the subject material was being welded. These problems were due to the new spot welding machines #6901 and #6940 being in the prove-out stage. Unistrut's QC recognized the defective resistance welds in P1004A control numbers 8N-7791, 8N-7792, 8E-7806, 8E-7828, 8E-7829, and 8N-7830 and rejected the material.

During reworking (re welding) to upgrade the rejected material, a break down in the quality program occurred. This may have been attributed to the fact that the formal QA program had just been implemented (around mid 1978) prior to the welding of the subject material. Therefore, it is conceivable that during the rework the material may have been inadvertently shipped prior to the completion and final QC acceptance by personnel unknowledgeable about quality procedures. The material that was resistance welded during this time frame is probably suspect. However, Unistrut, identified only three shipments of P1004A material that were made to SCC during that time frame (1978-1981). From these quality records, Unistrut was able to show that only one lot of 580 feet of rejected material could not be accounted for. It was reported by Unistrut that at the time in question, the red tag from the 580 feet of rejected material was located in the plant but the material was missing.

This appears to be an isolated instance.

2. 10 CFR Part 21 From examination of the document' packages it was determined that Part 21 was imposed on P0s to Unistrut by their Nuclear customers, Baldwin Associates, Bechtel, and Brown and Root. The requirements of Part 21 were not imposed on the contractors employed to fill these nuclear orders by Unistrut. The following vendors of calibra-tion services were issued P0s by Unistrut: Detroit Testing Machine Co. (Tensile Machine Calibration), A. A. Jansson (Gage Block Calibra-tion),

Dearborn Gage Company (Gage Block Calibration),

Mitutoyo MFG Co. LTD (Gage Block Calibration). These P0s did not specify Part 21 141

ORGANIZATION: UNISTRUT CORPORATION WAYNE, MICHIGAN REPORT INSPECTION NO.: 99900362/86-01 RESULTS: PAGE 6 of 9 as applicable to their calibration vendors. Likewise, P0s issued to material suppliers, Stateline Steel (P0-85781/6/84 - P0-09389, 9/17/84 with certification of mechanical and chemical properties) and Triad Petal (P0-P05523 5/15/81 - P0-P08588, 9/25/84) did not specify Part 21 as an applicable requirement and the Unistrut P0-0899 May 17, 1984 to General Fasteners for heat treated studs ASTM A193-83 B7 (Certification of mechanical and chemical properties were supplied),

did not specify Part 21 as an applicable precurement document (See Violation 86-01-03)

The Unistrut Part 21 procedure " Processing of Reportable Events" was reviewed. It was determined that the procedure did not make provisions i for a timely oral (two days) or written (5 days) response following the l the receipt of information regarding a reportable item. (See Violation 86-01-02).

3. Nonconforming Material The NRC inspector evaluated Unistrut's control of defective material.

Defective materials were tagged with a hard red copy of a rejection ticket and put into areas designated as MRB designated areas. The NRC inspector noted that 48 ft of material #HTHR075A00 had a hard red rejection ticket attached, however, it was not separated frem the usable material. (86-01-04) According to Unistrut there was not enough space in the MRB designated area to store the 48 ft of material, as a result it was not in the MRB designated area.

4. Calibration of Measurino and Test Equipment l The NRC inspector checked the up-to-date calibration of measuring and test equipment against the Unistrut procedure 20-01-026 Rev. 2 which .

described calibration frequencies. The Tensile Machine Model NST, J Serial #1007 had current calibration traceable to NBS 737.04/231572.

Micrometer gage block Set Serial #743612 had current calibration traceable to NBS 738/230272-83,'the Hardness Tester Serial #6718 had a current certificate of calibration traceable to NBS. One micrometer,

1. 15 and three Calipers (#16, #18, #19) from the floor area were checked for current calibration against the calibration index. All of the instruments checked were found to have current up-to-date calibration.

l 142

ORGANIZATION: UNISTRUT CORPORATION WAYNE, MICHIGAN REPORT INSPECTION NO.- 99900362/86-01 RESULTS: PAGE 7 of 9

5. Audits Internal Audits Internal audit records were examined for the years 1983-86. These audits were performed according to an authorized audit check list and in accordance with the Internal Audit procedure 20-01-027. However, there were problems found with Unistrut's ability to maintain their established audit schedule. In May,1984 internal audits were scheduled for sections X, XI, and XII of the QA Manual. The scheduled May 1984 audit was not performed until July 1984. The internal audit scheduled for August 1984 of QA Manual sections XIII, XIV, and XV, was not performed until October of that year.

In March 1985, QA Manual sections II, VIII, and IX were to be audited, however no audit was performed. QA Manual sections X, XI, and XII .

were to be audited in June 1985, however no audit was performed. QA Manual sections XIII, XIV, and XV were to be audited in September 1985 and section XIX in December 1985, again no audits were performed.

The internal audit performed an January 7,1986 of the QA Manual, identified some deficient areas in Sections III, VIII and XVII. As of the time of the inspection, implementation of corrective action had not been taken. Unistrut has comitted to implementing corrective action of deficient areas of their audited program within 30 days of the internal audit report. Audits of QA Manual sections IV, V, VI, and VII were scheduled during the month of April 1986. However, as of the date of the inspection, the April 1986 audit had not been initiated. (See 86-01-05 and 86-01-06)

External Audits External Audits were to be performed according to the Unistrut proce-dure 20-01-007 and performed on an annual basis (not to exceed a 12 month period by more than 30 days). Procedure 20-01-007 describes the responsibility of the Purchasing Department in locating suppliers of materials, the QA Manager's responsibility in source inspections, and requirements for companies being placed and remaining on the Approved Vendor List (AVL). It was noted by the NRC inspector that Unistrut has a provision in procedure 20-01-007 that states, in part, "...In the event the material, equipment on services required are nuclear related and would amount to or exceed ten thousand ($10,000.00) annual volume of business, the manager of Quality Assurance, will arrange for and conduct a source inspection and audit prior to the issuance of a 143

ORGANIZATION: UNISTRUT CORPORATION WAYNE, MICHIGAN REPORT INSPECTION NO.: 99900362/86-01 RESULTS: PAGE 8 of 9 purchase order...." The philosophy of this provision is not in accordence with the requirements of procurement outlined in ASME section III and/or 10 CFR 50 Appendix B. The Quality Assurance of Nuclear Safety related products by source inspection is to be 1 independent of cost considerations.

The AVL for 1979-86 was reviewed and found to contain vendors used by l Unistrut to fill their nuclear safety related orders. However, several vendors were on the AVL without benefit of an audit by Unistrut. The following companies were not audited during 1985 and 1986: A. A. Jansson, General Fasteners, Detroit Testing Machine Company, Rouge Steel Company, Traid Metals, Stateline Steel Company and United Testing Systems. The following companies were not audited during 1984: A. A. Jansson, Rouge Steel Company, National Testing and Research Laboratory. (See Nonconformance 86-01-05)

6. Training / Qualification Training / Qualification records were reviewed for nine quality inspec-tors (QC and NDE) and eleven welders. Two level II inspectors (in 1985) and one level III NDE (PT) inspector had no record of having an annual evaluation by a certifying authority. The level III inspec-tor had signed off as being the certifying authority for the lower level inspector. However, records indicate that the Level III inspector had never been subjected to an evaluation by a certifying

authority. There was no objective evidence of an annual eye examina-tion for one level II and one level III NDE inspector as required by section 8 of ASNT SNT-TC-1A. (See nonconformance 86-01-07). All

( other NDE inspectors were recertified on a three year basis as a result of continuing satisfactory performance and ccntinuing compliance with Part 8 and 9 of ASNT-TC-IA.

Unistrut requires new employees involved in Quality Assurance / Quality Control to be trained according to training procedure 20-01-024 Rev 2.

, As part of this requirement each employee must spend the first four l weeks of their employment in an on the job training program. The QA l manager or the supervisor of QA is to review and document, on a weekly basis, the employee's progress during the first four weeks. Once the new employee's 90 day probationary and training period is completed satisfactorily, they will be removed from the probationary classifi-cation. A file of personnel qualifications will be maintained with specific information regarding proper certification of qualifications, levels of capabilities and records of past performance, history, and 144 i

l

ORGANIZATION: UNISTRUT CORPORATION WAYNE, MICHIGAN REPORT INSPECTION NO.- 99900362/86-01 RESULTS: PAGE 9 of 9 evaluations every two years or as required. Upon review of the Unistrut training records, it was determined that a number of personnel qualification files were incomplete. For example, one inspector had no record of having a file of personnel qualifications.

Two inspectors had files of personnel qualifications however they were incompleted and not up-to-date. One inspector had no record of ever having a written progress report during the first four weeks of employment by the QA manager or the supervisor of QC. (See Noncon- l formance 86-01-08) l Unistrut procedure 20-01-030, Lead Auditor Qualifications, states that the lead auditor shall have qualifications similar or consistent with N45.2.23. N45.2.23 has specific requirements for completion and docu-mentation of competence in the area of communication skills, training, and auditor participation. The NRC inspector reviewed the records of four lead auditors. Of these auditors four had no documented evidence of competence in communication skills. Four lead auditors had no objective evidence of ever taking an audit training course. Two lead auditors had no objective evidence of fulfilling their audit partici- l pation requirement. (See Nenconformance 86-01-09)

Eleven welder qualifications were reviewed for welders since 1978. Of the welder qualifications examined, all were qualified according to AWS D1.1, GMAW in the 3G position. In addition, the welders were all trained and indoctrinated on the Unistrut Quality program.

F. PERSONS CONTACTED:

Timothy Condit Manufactured Products Unistrut

• Bruce McClure Senior Vice President Unistrut Materials Superintendent Unistrut

• Attended Exit Meeting 145

ORGANIZATION: WESTERN PIPING AND ENGINEERING SAN FRANCISCO, CALIFORNIA REPORT INSPECTION INSPECTION NO.: 99900302/86-01 DATE: 8/11-12/86 ON-SITE HOURS: 11 CORRESPONDENCE ADDRESS: Western Piping and Engineering ATTN: Mr. K. A. Friedman, President 1485 Yosemite Avenue San Francisco, California 94124 ORGANIZATIONAL CONTACT: Mr. G. Pappas, Quality Assurance Manager TELEPHONE NUMBER: (415) 822-6464 NUCLEAR INDUSTRY ACTIVITY: Design and engineering of vessels, appurtenances, component supports, piping subassemblies, and material supplier of ferrous forgings, plates and welding materials.

ASSIGNED INSPECTOR: n c9i!/[

4. P. Correia, Special ProjQts Injpection Section Date (SPIS) r APPROVED BY:

{otin W. Craig, Chief, SPIS, Vende Program Branch ate INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 21, and 10 CFR Part 50, Appendix B.

B. SCOPE: The inspection consisted of an examination of quality assurance and engineering records related to allegations concerning the certifi-cation and manufacturing of pipe clamps supplied by Western Piping and Engineering to various U.S. nuclear plants, a foreign plant, and other U.S. customers.

PLANT SITE APPLICABILITY: River Bend (50-458); Perry (50-440); Comanche Peak (50-445); Peach Bottom (50-277); Seabrook (50-443); and Cofrentes (Spain).

147

ORGANIZATION: WESTERN PIPING AND ENGINEERING SAN FRANCISCO, CALIFORNIA REPORT INSPECTION NO.: 99900302/66-01 RESULTS: PAGE 2 of 4 A. VIOLATIONS:

1 There were no violations identified during this inspection.

B. NONCONFORMANCES:

There were no nonconformances identified during this inspection.

C. UNRESOLVED ITEMS:

No unresolved items were identified during this inspection.

D. STATUS OF PREVIOUS INSPECTION FINDINGS:

There were no open items from previous inspections.

E. OTHER FINDINGS OR COMMENTS:

The inspection at Western Piping and Engineering (WPE) was performed in response to allegations concerning the certifications of the engineer who performed either the design calculations or the certification of such calculations for clamps manufactured by WPE. The allegation was that the engineer was not a registered Professional Engineer (PE) in the State of Cali fornia. Also an allegation concerning the use of illicit materials in the clamps for the River Bend Nuclear Plant was addressed.

1. Pipe Clamp Certification Activities The NRC inspector reviewec the WPE files of the engineer who performed design calculations and/or certification of such calculations. The files examined were of a recent WPE QA audit (dated 6/21/86) in which a resume of the engineer was included as-well-as an audit question-naire which followed the guidelines for demonstrating PE qualifica-tions established by Appendix C of the ASME Code,Section III, as required by Section 2 of ANSI /ASME N626.3 " Qualifications and Duties of Personnel Engaged in ASME Boiler and Pressure Vessel Code,Section III, Divisions 1 and 2, Certifying Activities." In particular, paragraph 2.2 of ANSI /ASME N626.3 requires, in part, that personnel engaged in ASME certifying activities be a registered Professional Engineer in at least one state of the United States or Province of Canada with specified years of experience in certifying activities as delineated in paragraphs 2.3 through 2.6 of the aforementioned standard.

148

ORGANIZATION: WESTERN PIPING AND ENGINEERING SAN FRANCISCO, CALIFORNIA REPORT INSPECTION NO.: 99900302/86-01 RESULTS: PAGE 3 of 4 Also, maintenance of current knowledge of Code requirements and continued professional development in his or her speciality field through various means is required. The standard also requires in part, that the Owner, Designer, or N-Certificate Holder, as applicable, must review the qualification of the PE at least once every three years to assure that his/her qualifications have been maintained with a continuing record of all such activity included in the qualification records of the PE. The records demonstrated that the engineer identified in the allegation was at the time period in question, and is currently, a registered Professional Engineer in the states of Pennsylvania and New York but not in California.

The allegation specified that the following nuclear power plants have had clamps designed and manufactured by WPE and were certified by personnel not having a PE registration in the state of California:

River Bend, Perry, Peach Bottom, Comanche Peak, Seabrook and Cofrentes (a Spanish nuclear plant). The NRC inspector examined the WPE record files for each of thest nuclear power plants except Seabrook. WPE has no record of having designed or manufactured a clamp which was to be used in the Seabrook facility. All purchase orders and technical specifications for these nuclear plants indicated that the clamp assemblies, as a minimum, be designed, in accordance to the require-ments of ASME Boiler and Pressure Vessel Code,Section III, Subsection NF, Class 1. In addition, the record files for purchasers of WPE clamps by the Paul Munroe Hydraulics Company, which did not specify the facility in which they were to be installed, also required the same ASME design certifications.

The file examined during the NRC inspection included all requirements set forth in the ANSI /ASME standard referenced above.

10 CFR Part 2, Appendix C, Section VIII states that NRC inspections of vendors are conducted to determine whether they are meeting their contractual obligations to licensees. There were no requirements identified in any of the examined record files which indicated that the design engincer certifying the clamps procured from WPE

! be a registered PE in the state of California. This allegatior. was not substantiatea and no nonconformances found during this part of the inspection, since the certifying engineer did meet the ASME requirements as required by procurement specifications.

l 149

l ORGANIZATION: WESTERN PIPING AND ENGINEERING SAN FRANCISCO, CALIFORNIA REPORT INSPECTION N0.: 99900302/86-01 RESULTS: PAGE 4 of 4

2. Materials in Pipe Clamps supplied for River Bend The allegation that pipe clamps supplied for the River Bend Nuclear Power Plant contained improper materials was reviewed. All WPE clamps designed and manufactured for the River Bend facility were purchased by General Electric (GE) (ref: GE P0 No. 205 AM 674). GE require-ments for certifications, data sheets, drawings, codes and standards were examined and found in WPE QA records including GE QA certifica-tion that their requirements were met. All materials, their respec- _

tive certifications for compliance to P0 specifications, both non-destructive examinations and destructive testing requirements, chemical analyses and results were reviewed. All WPE nonconformance reports (NCR) written during the design and manufacturing of the River Bend clamps were reviewed. Of the twelve total NCR's written, ten were dispositioned with appropriated approvals for acceptance or rejection. Two NCR's were voided upon discovery that the suspected nonconformance was not valid. All nonconformances disposition "use-as-is" were justified by either being within code requirements or included an engineering analyses with results being acceptable without compromising code requirements. Based upon the documents reviewed, pipe clamps supplied by WPE were manufactured in accordance with the utility's purchase specifications.

The allegation was not substantiated and no items of nonconformance were identified.

F. PERSONS CONTACTED K. A. Friedman, President, WPE M. Wright, Project Manager, WPE G. Pappas, Quality Assurance Manager, WPE 150

l ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION BLAIRSVILLE, PENNSYLVANIA l

l REPORT INSPECTION INSPECTION

! N0.: 99900005/86-01 DATE: 8/18-20/86 ON-SITE HOURS: 62 i

CORRESPONDENCE ADDRESS: Westinghouse Electric Corporation Nuclear Fuel Division ATTN: Dr. Richard Slember General Manager Post Office Box 355 Pittsburgh, Pennsylvania 15230 ORGANIZATIONAL CONTACT: Ron Cost TELEPHONE NUMBER: (412) 374-2359 NUCLEAR INDUSTRY ACTIVITY: Nuclear fuel tubing supplier for Westinghouse.

rh , ,

ASSIGNED INSPECTOR: t .bn I 7 Sb t

f(M. Abbate, Special Projecy Ins @ction

,SPIS) 0 Section P. J. Prescott, SPIS OTHERINSPECT0Rp{5): D. . L nn, S , Consultant APPROVED BY: (

fohnW.Craig, Chief,SPIS,VqorProgramBranch hh vate L

INSPECTION BASES AND SCOPE:

l A. BASES: 10 CFR Part 21, 10 CFR Part 50.

B. SCOPE: Observe fuel tube manufacturing activities, review ultrasonic l testing (UT) activities, review QA program implementation and follow-up

! Westinghouse corrective action on previous NRC inspection firidings.

PLANT SITE APPLICABILITY: Nuclear power facilities using Westinghouse fuel.

l 151 i

l

ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION BLAIRSVILLE, PENNSYLVANIA REPORT INSPECTION N0.: 99900005/86-01 RESULTS: PAGE 2 of 9 A. VIOLATIONS:

None.

B. NONCONFORMANCES:

1. Contrary to Section 17.1.2 of WCAP-8370/7800, Revision 10A/6A, dated August 1984 and Section 6.12 of Specialty Metals Plant (SMP) Product Assurance Department Administrative Procedure PA-003, Revision C, the standard used in ultrasonic testing (UT) to validate tube ovality is not serialized and maintained unc'er the equipment calibration control system nor are the dimensions of the standard traceable to National Bureau of Standards (NBS) instrumentation. "

2. Contrary to Section 17.1.5 of WCAP-8370/7800, Revision 10A/6A, dated August 1984 and Manufacturing Orders (MO) 200 and 206, both dated February 25, 1986, the data for vertical linearity, horizontal linearity and calibrated attenuation was not recorded on the cali-bration data reports, dated February 24, 1986 and March 27,1986, for two UT flow detectors (serial numbers 33 and 40).

3. Contrary to Section 17.1.6 of WCAP 2370/7800, Revision 10A/6A, dated August 1984, procedures QS-231, QS-249, QS-261, and QS-262 were not revised when QS-118 was superseded by PA-103.

C. UNRESOLVED ITEMS:

Paragraph 4.5.3 of Material Specification NFD 31008, " Seamless Zircaloy -

4 Tubing," Revision 28, dated May 16, 1986, requires, in part, that tube ovality not exceed 0.0013-inch Total Indicator Reading (TIR). Paragraph 6.4 of Procedure QC-301, " Final Inspection-Ultrasonic Dimensional Setup and Calibration," Revision J, outlines the method used to certify that this dimensional requirement is met. While QC-301 is accurate, it is unclear whether the actual requirement of TIR is being met. This item is discussed below in Section E.2.

D. STATUS OF PREVIOUS INSPECTION FINDINGS:

1. (Closed) Nonconformance (82-02, Item B.1)

Contrary to Section 5 of the Topical Report and Material Specification NFD 31008, paragraph 3.3.1 and Table I, there was no evidence of the submittal to and approval by the purchaser of the procedures for the outside surface finish process, chemical composition testing, and tensile testing.

152

ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION BLAIRSVILLE, PENNSYLVANIA REPORT INSPECTION NO.: 99900005/86-01 RESULTS: PAGE 3 of 9 During the inspection, Material Specification NFD 31008, " Seamless l Zircaloy-4 Tubing," Revision 28, dated May 16, 1986 and the list of approved documents were reviewed. The inspector verified that the purchaser had reviewed and approved the procedures for the outside surface finish process, inside surface finish process, chemical composition testing and tensile testing. This item is considered closed.

2. (Closed) Nonconformance (82-02, Item B.2)

Contrary to Section 5 of the Topical Report and the Quality Procedure, QC-300, paragraph 2.1, manufacturing at the time of this inspection did not have the latest revision of Specification NFD 31008 listed on the QCF-3003 form in reference to Order No.

548H20313.

During this inspection, form QCF-3003s, " Customer and SMP Inspection Requirements Fuel & WABA Type Tubing Lots," for ten months were reviewed to verify that the correct revision of Specification NFD 31008 was specified. Based upon the documents reviewed, this item is considered closed.

3. (Closed) Nonconformance (82-02, Item B.3)

Contrary to Section 5 of the Topical Report and Specification NFD 31008, paragraph 3.2, the identity of some material with respect to ingot melt number and lot number was not maintained at all stages of manufacture as evidenced by:

a. Two tubes of lot no. F73 2266 were found to have been mixed with lot no. F73 2257 which is from a different ingot melt number and heat treat lot number. These two tubes had been reworked on traveler card G 13448, and

b. In addition to the above, another tube in lot no. F73 2257 could not be accounted for by comparing inspection records with the actual piece count of the lot.

During this inspection, the accountability (traceability) system implemented by the SMP was reviewed. The ingot and lot numbers are recorded on the follow sheets (travelers) at the beginning of the fabrication process. The lot numbers are also etched on the end of each tube and are maintained there until the tube is cut to length.

From that point on, accountability is maintained solely by the follow sheet. If a tube has to be reworked, a separate follow sheet is 153 l

ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION BLAIRSVILLE, PENNSYLVANIA REPORT INSPECTION N0,: 99900005/86-01 RESULTS: PAGE 4 of 9 generated from the original by a computer. As a final accountability check, the total number of tubes in a lot which are ultrasonically tested must equal the sum of the tubes from that lot which are boxed and the tubes from that lot which were scrapped. These numbers are recorded on form QCF-3004-2, " Lot Piece Accountability Log Sheet," and form QCF-3005-1, " Scrap Accountability Log Sheet." If all the tubes are accounted for the lot is released.

During the inspection the inspector noted that the tubes, including rework tubes, were being properly identified with follow sheets and the etched lot numbers were maintained on the tubes and accounta-bility was maintained. Based upon these observations and reviews, this item is considered closed.

4. (Closed) Nonconformance (82-02, Item B.4)

Contrary to Section 5 of the Topical Report and the QA Program Manual, paragraph 5.2, the quality procedures did not include the practice and application of "T tags" which are used to make engineering dispositions of nonconforming material.

During this inspection, the inspector reviewed Procedure QC-318, "Dispositioning of Fuel and WABA Tubing After UT Dimensional and Flaw Inspection," Revision F, dated August 15, 1986, which outlines the use cf T tags, and observed the implementation of the procedure and T tags on the shop floor. Based on the review and observation, this item is considered closed.

5. (Closed) Nonconformance (82-02, Item B.5)

Contrary to Section 5 of the Topical Report and the Quality Procedure, QC-103, paragraph 3.2.1, form QCF-1030 information has not been supplied by inspectors to their supervisors, although above normal reject rates have occurred.

During this inspection Procedure BI-108, " Corrective Action System,"

Revision A, dated February 20, 1986, was reviewed. The procedure requires form BI-1080, "Above Normal Reject / Scrap Investigation,"

to be filled out when an above normal reject rate is observed by an operator. This form then goes to the Quality Engineer for dispositioning and the Process Engineer who determines the cause of the problem. The inspectcr reviewed several instances where the procedure was followed ard form BIF-1080 was used by the operators. This item is considered closed.

154 l

l

ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION BLAIRSVILLE, PENNSYLVANIA REPORT INSPECTION N0.: 99900005/86-01 RESULTS: PAGE 5 of 9 E. INSPECTION FINDINGS AND OTHER COMMENTS:

1. Entrance and Exit Meetings An entrance meeting was conducted on August 18, 1986 at the p Westinghouse Specialty Metals Plant (SMP) located in Blairsville, Pennsylvania. The purpose and scope of the inspection were discussed at that time. The SMP Quality Assurance (QA) program is outlined in Topical Report WCAP 8370/7800, "Westinghous'e Water Reactor Divisions Quality Assurance Plan," Revision 10A/6A, dated August 1984. The topical report, which was reviewed and approved by the NRC, describes the QA program of the Westinghouse Corporation.

During the exit meeting on August 20, 1986, the inspection findings and observations were summarized.

2. Ultrasonic Testing Areas examined during the inspection pertaining to the ultrasonic testing (UT) of tubes included equipment calibration, training and qualification of personnel, procedures and observation of testing.

A verification of calibration and certification of UT equipment was conducted on seven pieces of equipment, including standards used for calibration. During this review, it was noted that the standard used to validate tube ovality was not serialized and maintained in the equipment calibration control system nor were the dimensions of the standard confirmed to any known NBS traceable instrument. It was also noted that the data for vertical linearity, horizontal linearity, and calibrated attenuation was not recorded on the calibration data reports, dated February 24, 1986 and March 27, 1986, for two UT flow detectors (serial numbers 33 and 40).

Nonconformance Items 86-01-01 and. 86-01-02 were identified in this area.

Training and qualification records for three operators and two supervisors were examined. In the instances reviewed, the training, testing, qualifications and maintenance of qualifications met the requirements outlined in ASNT SNT-TC-1A, " Training and Qualification of Nondestructive Test Personnel."

155 l

l

ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION BLAIRSVILLE, PENNSYLVANIA REPORT INSPECTION N0.: 99900005/86-01 RESULTS: PAGE 6 of 9 The inspector also observed a UT operator calibrating the equipment and performing UT on tubes in lot no. P41-8726. This was accomplished by using Procedure QC-301, " Final Inspection - Ultrasonic Dimensional Setup and Calibration," Revision J, and Procedure QC-320, " Final Inspection - Operation of Ultrasonic Tester," Revision F. The inspector found that the operator followed these procedures and had a good working knowledge of the procedures and equipment being used.

Paragraph 4.5.3 of NFD 31008, " Seamless Zircaloy-4 Tubing," Revision 28, dated May 16, 1986, requires, in part, that ovality not exceed 0.0013-inch Total Indicator Reading (TIR). The setup to determine tube ovality by UT is outlined in Procedure QC-301 and is conducted in the following manner: A tube sample of the same size and material composition (referred to as a dynamic reference standard) is squeezed in a vice and measured with the ' aid of a laser micro-meter to determine true ovality. The dynamic reference standard is then run on the UT machine at predetermined speeds to obtain the UT ovality reading. The two readings are then compared to assure the UT equipment is performing accurately.

Technically, this procedure as used to verify tube ovality is accurate.

However, it is unclear whether the actual requirement of TIR is being met since TIR implies that the reading is taken in one plane, 360' around the circumference rather than obtaining an average as done in QC-301. An average is obtained due to the fact that the tube is moving as the reading is being taken instead of being stationary.

Unresolved Item 86-01-01 was identified in this area.

3. Procurement The NRC inspector examined several procurement records to determine whether the Westinghouse procedures and their implementation met the requirements of 10 CFR 21 and the specified on the Westinghouse pur applicable chase quality order form 200.requirements These require-as ments are imposed on the subvendors supplying Westinghouse with the raw material used in the manufacturing process. A selective sample of purchase orders and Certificates of Compliance for the raw material procured from subvendors by Westinghouse were examined.

l These documents were in compliance with the requirements of 10 CFR Part 21 and quality requirements. The subvendors used for these purchases appeared on the latest revision of the approved vendors list.

Required documentation was in place and the Westinghouse QA procedures for the use of such documentation had been implmented as required.

1 i

156

ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION BLAIRSVILLE, PENNSYLVANIA I REPORT INSPECTION NO.: 99900005/86-01 RESULTS: PAGE 7 of 9 l

4. Equipment Calibration Procedures QC-600, " Gage Control System," Revision F, dated June 20, 1986, QC-601, " Calibrations-Micrometer," Revision C, dated January 6, 1986, and associated calibration records for the equipment utilized during the in-process inspection of the material being manufactured at the SMP were examined by the NRC inspector.

The calibration records examined were found to be current'and in compliance with QC-600. The associated logs and files were also found to be accurate and in compliance with the applicable quality control procedures.

5. Material Rejection Notice The NRC inspector reviewed Material Rejection Notice (MRN) No.

348244, for compliance with WCAP-8370/7800 and Purchasing Department Procedure PD-251, " Purchasing of Controlled Materials and Services," dated April 2, 1986. The MRN is a form used by Westinghouse to document discrepancies noted upon the receipt of raw material supplied by subvendors. The form includes a description of the discrepancy noted and a disposition which requires a review by the Westinghouse Engineering Department.

During the review of MRN No. 348244 by the NRC inspector, it was i noted that on May 15, 1986 a subvendor (Western Zirconium) shipped  !

defective raw material, unknowingly, to Westinghouse who received the material, along with material certifications, on May 19, 1986.

Westinghouse did not generate MRN No. 348244 until they were notified of the defective material by Western Zirconium on May 23, 1986. In an interview with a Westinghouse Quality Engineer, it was stated that Western Zirconium could not have notified Westinghouse prior to the shipment because the actual material qualification documentation had not been completed by the subvendor. Other MRN's reviewed during the inspection by the NRC inspector were found to be in compliance with the applicable procedures and no'nonconformances or violations were noted. l l 6. Manufacturing Activities i

A sample of Process Engineering (PE) procedures which define fabrica-tion activities were reviewed including:

- PE-202, "Zircaloy - ID Blasting," Rev. K, April 25, 1986; )

- PE-208, "Zircaloy - S:C Grit Blasting," Rev. D, April 25, 1986; 157

ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION BLAIRSVILLE, PENNSYLVANIA REPORT INSPECTION NO.: 99900005/86-01 RESULTS: PAGE 8 of 9 PE-216, "Zircaloy Stress Relief Annealing of Final Size Zirc-4 Tubing," Rev. D, July 7, 1986; PE-218, "Zircaloy-Vacuum Annealing," Rev. I, June 11, 1986; and PE-250, "Zircaloy-Swaging Thimble Tube," Rev. H. July 9, 1985.

Following this review, activities controlled by the procedures were observed and the procedures were discussed with personnel at their work stations. Plant personnel were knowledgeable and demonstrated a good working knowledge of PE's. Work areas were clean and proce-dures at the work stations were complete and current.

A sample of Quality Services Standard Operating Procedures (QS) which control testing in the laboratory were reviewed including:

l QS-231D, " Determination of Dimensions on Non-Destructive Test l Standards," June 18, 1984; QS-261C, " Chloride Ion Determination By Solid State Electrode Method," June 13, 1984; QS-249B, " Determination of ASTM Grain Size of Inconel 600 Final Size Tubing," June 28, 1984; and QS-262A, " Room Temperature Bend Testing of Inconel and Titanium

Tubing on Satec Tester," June 8, 1984.

Each of these procedures contained a reference to a procedure which had been deleted, QS-118. QS-118 had been replaced by Product j Administrative Procedure PA-103, " Quality Services Laboratory Sample and Data Handling," Rev. B, May 5, 1986. PA-103 was initially issued on July 7, 1985.

These procedures were being reviewed for deletion at the time of the inspection and were deleted prior to the conclusion of the inspection.

Additionally, QS-503, " Determination of Hydride Orientation of Tubing," August 13, 1986 was reviewed and discussed with laboratory personnel. The laboratory personnel observed while preparing samples for analysis were knowledgeable and proficient.

Nonconformance Item 86-01-03 was identified in this area.

l l

l l

l l

158

1 ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION i BLAIRSVILLE, PENNSYLVANIA I

l REPORT INSPECTION NO.: 99900005/86-01 RESULTS: PAGE 9 of 9 F. PERSONS CONTACTED:

• G. S. Pail, Manager, Product Assurance

• R. Carmody, Manager, Manufacturing

• J. B. Narayan, Fellow Engineer, Product Assurance

• W. A. Long, Lead Supervisor of Audit

• R. Cost, Manager, Quality Assurance, Nuclear Fuel Division

• J. E. Czesnakowski, Manager, Process Engineering

• R. D. Petrosky, Manager, Programs and Document Control

• G. A. Ruby, Quhlity Assurance Engineer

• G. D. Federowicz, Controller

• C. R. Mitchell, Manager, Human Resources

• T. M. Sanders, Manager, Advanced Process Development

• H. B. Werner, Manager, Laboratory

• J. R. Bellas Unit Manager of Inspection

• J. Yurigan, Laboratory Technician

• L. A. Winter, Quality Engineer

• D. L. Himes, Document Control Coordinator

• M. Nance, Document Control Coordinator

• K. Trosell, Manager, Materials

• R. S. Wengewics, Senior NDE Engineer P. Moore, Senior Technician J. Sabo, Senior Quality Engineer B. Patz, Level I UT Operator R. J. Joschak, Quality Assurance Engineer D. Mazurek, Laboratory Technician J. Wilkinson, Laboratory Technician l

159

s k

SELECTED If1FORftATION NOTICES 1

i f

i 1

. l SSINS Ns.: 6835 IN 86-56 UNITED STATES NUCLEAR REGULATORY COMISSION OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, D.C. 20555 July 10, 1986 IE INFORMATION NOTICE NO. 86-56: RELIABILITY OF MAIN STEAM SAFETY VALVES Addressees:

All pressurized-water reactor facilities holding an operating license or construction permit.  :

Purpose:

This information notice (IN) is provided as additional notification of NRC's concern for the reliability of spring-actuated main steam safety valves follow-

-ing reports of multiple failures during testing and problems during power operations and scram recovery. IN 86-05, " Main Steam Safety Valve Test Fail-ures and Ring Setting Adjustments," previously addressed the problem of inade-quate flow capacity of these valves.

It is expected that recipients will review the information for applicability to their facilities and consider actions, if appropriate, to preclude similar problems from occurring at their facilities. However, suggestions contained in this notice do not constitute NRC requirements; therefore, no specific action or written response is required.

Description of Circumstances:

While researching IN 86-05, the following problems with main steam safety valves (MSSVs) that had occurred during testing, power operations, or scram recovery were tabulated from the licensee event report files.

MAIN STEAM SAFETY VALVE PROBLEMS (01/01/81-03/01/86)

NUMBER NUMBER TESTING POWER POST-SCRAM PROBLEM VALVES PLANTS (EVENTS) (EVENTS) (EVENTS) 13 6 5 0 1 FAILURE TO OPEN FAILURE TO RECLOSE 15 9 2 1 8 6 4 2 3 1 SPURIOUS OPENING 7 4 1 2 0 LEAKING SET POINT DRIFT HIGH 44 10 11 0 1 LOW 97 14 18 3 3 i

i UNSPECIFIED 75 11 14 0 0 l

l 161

-e ..-, __ ww--- w g---

IN 86-56 July 10, 1936 Page 2 of 3 i i

A number of reports have been received concerning events involving MSSVs. Four '

of the more significant reports are summarized in this and the following para-graphs.

At North Anna 2, 8 of 15 valves would not lift at the maximum pressure available to the testing device at the site (1147 to*1161 psig). All 15 valves were subsequently sent to Wyle Laboratories for further testing. The as-found sstpoints varied from 1105 to 1223 psig compared with the specified 1085 i 11 to 1135 i 11 psig setpoints. All valves were refurbished and setpoints were j adjusted to be within technical specification limits (LER 50/339-86/001).

Salen 2 reported one valve leaking during heatup following a refueling outage.

After the spindle nut was backed away from the' forked lever, the valve reseated, but later it lifted and did not immediately reseat. It was gagged shut.

Subsequently two other valves prematurely lifted. They too were gagged shut.

Later, when a fourth valve lifted, it was declared inoperable. All 20 MSSVs 4

were then tested and reset to the appropriate setpoints. The licensee was unable to determine the cause for the setpoint drift (LER 50/311-85-007).

Ocenee 2 reported that 2 MSSVs had failed to reseat promptly. They reseated at 900 psig instead of 1010 psig following a transient that included a reactor scram (LER 50/270-85/006).

On October 19, 1985, the 16 MSSVs of Calvert Cliffs Unit 2 were tested to check

! and, if required, to adjust the relief pressure setpoints. Unit 2 contains 2 stc:aa generators, each with 8 MSSVs on a steam header. Eleven of the 16 valves 4

were determined to be out of specification, with the as-found setpoints between 22 to 69 psi higher than their nominal setpoints, compared with the technical sp2cification requirement of

• 10 psi. The Licensee Event Report is included in this notice as. Attachment 1 to give an example of the probless found during testing, and because it is an especially thorough treatment of corrective actions

-thtt may be of value to other facilities.

Discussion:

i Ths safety significance of failure of the MSSVs to open on a PWR is a potential for over pressurizing the secondary system with a possibility of a loss of its prassure boundary integrity. Failure to reclose has led to overcooling tran-sients and lower-than-normal water levels in the steam generator. Spurious i opsning, usually at power, has led to reactor scrams. Loung valves tend to htve more problems than properly functioning valves. 54tptint drift-low may cause spurious opening of the valves and may interd. synn listically with a steam generator tube rupture to cause relief throL9h tu 4 Jlted steam genera-ter in the case where the MSSVs on the faulted steaun generator have icwer-than-required setpoints. Setpoint drift-high can cause secondary pres-sure to rise above the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code-specified system pressure limit.

162 i

IN 86-56 i July 10, 1986 l Page 3 of 3 No specific action or written response is required by this information notice.

If you have any questions about this matter, please contact the Regional Administrator of the appropriate regional office or this office.

$L ' -

)rdan,51 rector ard L.

Division Emergency Preparedness and En eering Response Office of Inspection and Enforcement Technical

Contact:

Mary S. Wegner, IE (301)492-4511 l Attachments:

1. LER 50/312-85/011

2. Recently Issued IE Information Notices l

153

r .

i m,,e,a .==

, ua .ne =T se===.

e.es . -

UCENSEE EVENT REPORT ILER) '" ' " " * * '

• u T man n.

Calvert Cliffs, Unit 2 e , e.es. . . i.

was . , , , , , , , , , , 3 ,1 i8 i lo, 0,5 Main Steam Safety Valve Setpoints Out of Specification '

even ears a see ammen . as , en e. - -ie..,mve .

A- esT Team Team M Mf gWWe# usuTu ea? Te4A 8aewvT hasse eggs 47 ansmen.

IM s iste re r ei i l

i 1[0 1l 9 85 8l 5 Q 1l1 0l1 1l 2 1l9 8l5

,,,,,,,, n= - e sue wee um , v. ,,e me eme w a e, e . .,-e e isie seiei i i

=ess = 3 nw sues. ---

=,

~

en sema. sum.n "~

l T oi ci 0

==n= Z ==== { eumme. as.

( ,

==wn== 1 auses.

==== e

======w Z  :::le.,w, .yg

==

n sus--. --_-

Restienten _

R NInsges Entsmed unmen n= un no

- febe no ammeen j z

emme

P. M. Knoetgen - Engineer - It0 '

30li 2,60 i i 4 i 869 i emmun ese use e.g. u- -- esema = ,== , n.

e848e ee- ^^ u gM" '

"'w""n"n

, e eause everam soumeneet "*"W',**

,y,, 7,,,,8,8,68

(-f(I X SB i i RV i Di 2:4,3 Y h , , , , , , , hkh[

I I I ! I I I 46 ase., _ . .se I l l t i i t hk esame ear Team 7 7a n eeiec = - .we "30 **""*

-a a ... , , ,

-2_-. _

no r

On 19 October 1985 Unit 2 was shutdown in MODE 3 in preparation for a refueling outage. A Surveillance Test Procedure (STP) was performed to determine Main Steam Safety Valve (MSSV) Setpoints on No. 21 and 22 Steam Generators. Eleven of sixteen MSSVs were determined to be out of specification, and Technical Specification action statement 3/4 7.1.la was entered. The MSSVs were reset to within specifications by 2130,19 October 1985. At the conclusion of testing, the plant was placed in cold shutdown for refueling.

l Long term corrective action included full flow testing of two MSSVs. The testing device used in the STP was tested. All sixteen MSSVs were inspected and overhauled. Procedural enhancements were incorporated into the STP to improve reliability. During reactor coolant system heatup on 7 December,1985, all 16 MSSV setpoints were. tested and adjusted as required. Twelve hours later an independent check to verify a sample of four valves was performed satisfactorily.

Additionally, four more Unit 2 MSSVs will be checked after four months of operation, and all Unit 1 MSSVs will be checked during the next outage.

The causes of this event appear to be more attributable to measuring technique than to material condition.

Attachment 1 IN 36-56 164 July 10, 1986 S.

,, t: ' :' 5 4

1 l

l

. ,a . us. - ...m r . - i

'**

• oves case .a nes-ne. !

,! UCENSEE EVENT REPORT ILER) TEXT CONTINUATION l umu w=

., _ ... . . . . _ . _ , . j i.. i lat z : r c..t; I Calvert Cliffs, Unit 2 I o is lo lo t o l 3:1 : 8 8i 5 h 0 :110l l 0:1 l0 2 op 0 l5 rarr. . e-a . m On 19 October 1985, with Unit 2 reactor in MODE 3 in preparation for a refueling outage, Surveillance Test Procedure (STP) M-3-2, " Main Steam Safety Valves (SB-RV)" was performed on No. 21 and 22 Steam Generator (S/0) Main Steam Safety Valves (MSSV) to check and, if required, adjust the relief pressure setpoints. Unit 2 contains two Steam Generators (S/G), each with its own Main Steam (SB) header. Each steam header is served by eight Dresser Industries type C-3707RA maxiflow safety valves (RV). The STP was commenced at approximately 1530,19 October, with reactor coolant (AB) temperature at 5000F, Eleven of the sixteen scfety valves were determined to be out of specification, and TS action statement 3.7.1.la was entered. 'Ihe required lift setpoints and their as found condition are listed below:

Required Setpoint As Found Valve (110 psig) Setpoint RV-3993 985- . 1015 RV-3995 995 ~1035 RV-3999 1035 1057 RV-4000 985 1037 RV-400* 985 1040 RV-4002 995 1059 RV-4003 995 1047 RV-4004 1015 1070

.RV-4005 1015 1054 RV-4006 1035 1104 RV-4007 1035 1106 The immediate corrective action, as required by STP M-3-2, was to reset the safety valve setpoints to within their required tolerance bands. By approximately 2120, 19 October, the eleven relief valves had been reset. The setpoints of the valves were determined from 3 " simmers" of each valve using a Dresser Industries 1566 hydroset (a hydraulic testing device). After completion of testing, plant cooldown was continued, entering MODE 4 at 0200,20 October, and MODE 5 at 1200,20 October.

Attachment 1 IN 36-56 July 10. 1986 Page 2 of 5 165

1

• * . u - . . . .% ,,e , . . .

l UCENSEE EVENT REPORT it.ER) TEXT CONTINUATION .a== eves o=e na me e.

sums. am e. .

seesu t y ..aes .g.

secast mussesa ial Lee suunnesa ses l pads a3

.... -::rac;t n c.7. :: .

Calvert Cliffs, Unit 2 oistoloioi 3:1;885 i ._lI 0 :1l i .l 0i 1[' 0 :op3 0 g 5 wr a, . --- .

.e .m 1

The following additional corrective actions were taken:

1. The hydroset used to perform STP-M-3-2 was sent to the vendor for testing. The hydroset was determined to have internal leakage. Test 4

results showed the hydroset reading high but in specification for the range used during the MSSV STP. Because of the internalleakage, the hydroset could not be tested over its full range of operation.

2. All sixteen Unit 2 Main Steam Safety Valves (MSSV) were disassembled for inspection and overhaul.

a. Nine of sixteen valves indicated varying degrees of vertical wear marks on the sliding surfaces of the disk holder and guide .. The wearing appeared primarily along the bottom one Inch of the disk holder and along the top and bottom one to two inches of the guide. The manufacturer has stated that vertical abrasions are normal wear, and will cause valve hangup only if gmss galling or spalling is experienced. Only one MSSV had high wear, but not sufficient to cause valve malfunction.

b. Based on the manufacturer's recommendation, the disk guides were machined to allow more clearance for thermal expansion. The specification used for manufacture of the guides was increased by the vendor by 0.005 inches in December 1983. Baltimore Gas and Electric Co. was not

. notified of this change since it affected valves with set pressures above 1250 psig. Twelve of sixteen guides were at or slightly below the original minimum inside diameter of the guide. However, disk holder to guide clearances were within specifications, and would allow normal opening and closing,

c. Four MSSVs had a heavy, black, crusty substance on the guide and disk holder. 'lhe substance was analyzed by X-ray ,

analyses, and was determined to be the remains of Fryquel, i the hydraulic fluid used in the Main Steam Isolation Valves (SB-ISV) (MSIV). Fryquel from a leak in the MSIV hydraulic I system in late July 1985 had apparently entered the MSSV j through the valve cover vent. The vendor has stated that  ;

Fryquel would not likely effeet opening or closing of MSSVs.

d. Thirteen of sixteen MSSVs had bent spindles

, However, all dimensions were less than ,

half the minimum required to have any major effect on ve'fe performance.  !

i e

Attachment 1 IN 86-56 166 July 10, 1986

.e .

q "M" ' ^ # "

l

'- u , e e.

gg,awm. .

.a.=evts e.e .e. sies me.

UCINSEE EVENT REPORT ILER) TEXT CONTINUATION

- sma am esies seems, - - -- e, ,,,,,,,,,,,,,, , , , , , ,

..=wv .a.e si

• ' .e. i.:f*! M t r Clm Calvert Cliffs, Unit 2 o is io io io i 31 l i 8 Si 5l l01:1l l0 i 1 0 i 4 op 0 l5

3. Two valves, RV 3992 and R" 3993, were tested under full flow 1 conditions at Wyle Laboratories to determines Lift setpoint and blowdowns Hydroset to lift setpoint correlation; and Dermal effects on ]

lift setpoint.

a. The lift setpoint of the valves was determined using full flow ,

tests, and a curve developed which relates the setpoint

, determined from using the hydroset with the actual lift setpoint of the valve. Dis curve will be incorporated into STP-M-3-2.

b. De MS8Vs were determined to be thermally stable after four ,

hours of haatup. A heatup transient test was performed to i

determine the effect of thermal non4quilibrium on lift setpoint. No effect was noted.

c. Valve settings recommended by the manufacturer yielded j

5.4% to 7.8% blowdown.

I

4. Several procedural enhancements have been incorporated into the surveillanace test procedure for MSSVs. They includes

a. A requirement to maintain reactor coolant system (AB) temperature at 530*P i S*P for four hours prior to testing, to more accurately reflect normal operating conditions.

b. Use of preesiste monitoring equipment for the hydroset which is easier to use and less subject to operator interpretation.

c. A calculatiosi of expected hydroset pressure at lift setpoint.

Dis assists the operator by allowing him to approach the lift setpoint slowly and avoid overshoot.

l'

d. A requirement that three lift setpoints should be within 10 pelg of each other to provide repeatable, valid data.

e. Use of the lift setpoint-hydroset correlation curve developed from the full-flow testing at Wyle Laboratories to determine

( the actuallift setpoint measured.

f. A requirement to watch for hydroset pressure decrease, as l

well as to listen for an audible simmer, to determine when the MSSV lifts, as recommended by the hydroset vendor.

Attachment 1 IN 86-56 July 10,1936 Page 4 of 5

167 j

.e

"' u -- -.uvuren,r LICENSEE EVENT REPQRT (LER) TEXT CONTINUATION an==ovco owe ao mes-me.

i e.u em

. un ... osaev wa na e u. u= a= l e is

.u. inat=t r i =::

Calvert Cliffs, Unit 2 i i

,wr.

oistoloIoi 31;8 85 1 __

0 1l1l l0:1 05 3

op 0l5

.-- - e=au.m h

g. A requirement to rwhack the calibration of gauges after testing has been completed to ensure valid pressure measurement.

h. Mandatory QC coverage for all steps of the STP.

5. During the reactor coolant system heatup on 7 December,1985, all sixteen MSSVs were tested and adjusted as required. Approximately twelve hours later, four MSSVs were independently tested by a different technician to verify the previous results and were found satisfactory.

' 6. During the first outage following four months of operation, the setpoints of four Unit 2 MSSVs will be verified.

7. During the next Unit 1 outage of sufficient duration, all Unit 1 MSSV setpoints will be checked. If any valves are adjusted, they will be re-checked during the first outage following three months of operation.

'the safety consequences of this event were analyzed by the NSSS vendor,

' Combustion Engineering. 'three analyses crediting MSSV operation were considered. The results of a Small Break Loss of Coolant Accident (SBLOCA) would be no worse than the current SBLOCA analysis results. The peak secondary pressure was determined not to exceed the upset limit during a Loss of Load or Loss of Load to One Steam Generator transient, and the as-found MSSV setpoints would not be expected to have any adverse effect on the radiation release consequences of either of,these events.

Although material deficiencies wer* found during MSSV inspection, the apparent setpoint changes are not explained by the as-found material condition of the valves. The cause of this event appears to be attributed more to measuring technique than material condition.

Based on the evaluation of Unit 2 MSSV material condition, it is felt that Unit 1 MSSVs will open, provide full capacity, and reseat as designed.

Similar events have occurred on Unit 1 (LERs 85-03,79-16) and Unit 2 (LERs 84-04, 82-48).

t

'the contact for further discussion of this event is Peter M. Knoetgen,(301) 260-  !

4869.  !

i Attachment 1 II86-56 July 10,1986 Page 5 of 5 16d 2,'" "'"

i SSINS No.: 6835 IN 86-57 UNITED STATES NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT'

' WASHINGTON, D.C. 20555 July 11, 1986 IE INFORMATION NOTICE NO. 86-57: OPERATING PROBLEMS WITH SOLEN 0ID OPERATED VALVES AT NUCLEAR POWER PLANTS

. Addressees:

A11' nuclear power reactor facilities holding an operating license or a construction permit.

Purpose:

This notice is to advise recipients of a series of valve failures that have occurred recently at several nuclear power plants. It is expected that recipi-ents will review the events discussed below for applicability to their facili-ties and consider actions, if app.ropriate, to preclude similar valve failures occurring at their facilities. However, suggestions contained in this notice

do not~ constitute NRC requirements; therefore, no sp'ecific action or written i response is required.

Description of Circumstances:

The NRC has received reports from licensees of operating nuclear power plants involving failures of certain valves that are actuated by solenoid operated i valves (SOVs) to operate properly. These failures have adversely affected the 1 intended functions of the main steam isolation system, pressure relief and '

fluid control systems. Attachment 1 to this information notice describes the i

failure events and the corrective actions taken.

Discussion: ,

In most of the cases described in Attachment 1, the cause for triggering the event was attributed to a malfunctioning SOV that served as a pilot valve. This in turn resulted in the malfunction of the associated main valve. The failures of the SOVs can be traced to the following different causes: (1) potentially l high-temperature ambient conditions are not being continuously monitored in areas where SOVs are installed and operating in an energized state, (2) hydrocarbon l

contaminants, probably because backup air systems (e.g., plant service or shop air systems) are being used periodically and are not designed to " oil-free" specifications as required for Class IE servi.:e, (3) chloride contaminants j causing open circuits in coils of the SOVs, possibly as a result of questionable I handling, packaging, and storage procedures, (4) an active replacement parts program associated with the elastomers and other short-lived subcomponents used in SOVs has not been adequately maintained, and (5) lubricants have been used excessively during maintenance. ASCO provides installation and maintenance 159

IN 86-57 July 11, 1986 Page 2 of 2 sheets with all its valves and rebuild kits. For additional information ASCO should be contacted.

Because of the recurring S0V failures discussed above, NRC's evaluation of the problem is continuing. Depending on the results of the evaluation, specific actions may be requested.

No specific action or written response is required by this information notice.

If you have any questions about this matter, please contact the Regional Administrator of the appropriate regional office or this office.

1 ft V dward L. rdan, Director l Division Emergency Preparedness  !

and Eng neering Response Office of Inspection and Enforcement Technical Contacts: Vincent D. Thomas, IE (301)492-4755 George A. Schnebli, Region II (404)331-4875 Attachments:

1. Examples of Solenoid-Operated Valve Failures at Operating Nuclear Power Plants

2. List of Recently Issued IE Information Notices

! l l

i 170

t .

Atttchment 1 IN 86-57 July 11, 1986

- Page 1 of 4 EXAMPLES OF SOLEN 0ID-OPERATED VALVE FAILURES

< AT OPERATING NUCLEAR POWER PLANTS Brunswick Station l 1. Main Steam Isolation Valve (MSIV) Solenoid Failures On September 27, 1985 at Brunswick Unit 2, during.the performance of a periodic test to demonstrate operability of the MSIVs, three out of eight isolation valves failed to fast close as designed. The fast-close test was required before returning Unit 2 to full power operation after the plant had been placed in cold shutdown on September 26, 1985. Two of the three affected valves were installed as inboard and outboard MSIVs in the same main steam line, which would be a significant safety problem in the event of a failure of that steam line.

The licensee's initial investigation isolated the cause for the MSIV failures to the dual SOVs that serve as pilot valves that supply operating air to the MSIV operators to open or close the MSIVs. The faulty SOVs were identified as Automatic Switch Company (ASCO) Model NPL8323A36E. A more detailed review of the problems determined that the causes for i

failure were attributed to valve disc-to-seat sticking of the SOV and portions of the elastomer disc material plugging the SOV exhaust port.

These failures prevented closing the associated MSIV. Ethylene propylene j (EP) was the elastomer substance used for seals and valve disc material in

this model 50V.

The licensee's failure analysis of the SOVs included technical assessments of the problems from the valve manufacturer (ASCO), the supplier of the EP material (Minnesota Rubber), and Carolina Power and Light's (CP&L's) research center (Harris Energy and Environmental Center, Raleigh, North Carolina). The findings resulting from this joint effort indicated that the SOV failures could have been caused by a combination of hydrocarbon contamination of the air system and high ambient temperature conditions, causing degradation of the EP valve seating and seal material.

The ASCO Model NPL8323A36E SOVs were installed in Brunswick Unit 1 in t

June 1983 and in Unit 2 in August 1984 to meet the Environmental Qualifi-cation (EQ) Program requirements. The Unit 1 SOVs were subsequently replaced during the 1985 outage when modifications were being made to the l MSIVs. The new SOVs (NP8323A36V) were identical to the old ones except l the valves contained Viton seats and seal materials in lieu of EP.

Additionally, the information provided from ASCO shows the following:

a. Ethylene propylene is resistant to higher levels of radiation (200 megarads) than Viton. However, EP absorbs hydrocarbons that can cause swelling and loss of mechanical properties. It is unsuitable l in applications where the air system is not designed to " oil-free" j specifications.

171

L Attachment 1 IN 86-57 July 11, 1986 Page 2 of 4

b. Viton has superior high-temperature performance when compared to EP and is impervious to hydrocarbons. Its' major aisadvantage is that it is less resistant to radiation than EP by a factor of ten. ASCO

. recommends Viton for applications that are not oil-free and where radiation levels do not exceed 20 megarads.

On the basis of a licensee review of the Brunswick Station maintenance history, which showed the performance of Viton to be satisfactory in ASCO valves, and the available literature and industry experience, the licensee replaced all Unit 2 dual solenoid valves with valves having Viton seats and seals. Because Viton has a 20 megarad limit, the licensee plans to replace these elastomers every 3.3 years to meet environmental qualifica-tion requirements for the MSIV application.

After replacing the faulty valves with valves having Viton disc and seal material, the licensee experienced several SOV failures resulting from open circuits of the de coils on Unit 2. (Brunswick Station employs ASCO NP8323A36V valves that use one ac coil and one dc coil in applications using the. subject dual solenoid valve.)

! On October 5, 1985, the de coils of two MSIVs failed during the perfor-mance of post-maintenance testing of the MSIVs. Investigation into the i

failures indicated an open circuit in the de coils. The coils were replaced and the v,alves subsequently retested satisfactorily.

On October 15, 1985, an unplanned closure of an MSIV occurred while Unit 2 was operating at 99 percent full power. Closure of the MSIV occurred when the ac solenoid coil portion of the MSIV associated SOV was de-energized in accordance with a periodic test procedure. It was not known then that there was an open circuit in the associated de solenoid coil portion of

( the dual SOV. Consequently, when the ac coil was de-energized, closure of i

the MSIV resulted. The failed de coil was replaced and then retested l

satisfactorily.

Investigation into the failures of the de coil by the licensee determined that the failures appeared to be separation of the very fine coil wire at the junction point where it connects to the much larger field lead. This connection point is a soldered connection that is then taped and lacquered.

Further analysis of the coils (two failed de coils plus five spares from storage) by the CP&L Research Center indicated the separation might be corrosion induced by chloride contaminants. To date, the licensee and ASCO are unable to determine the source of the chloride. However, followup investigation by the NRC revealed that ASCO had previously experienced similar dc coil open circuit anomalies after a surface ship-ment of SOVs overseas to Japan. At that time, ASCO believed that the salt water ambient conditions during shipping may have been the source of the i

chlorine-induced failures. ASCO recommends specific handling, packaging,

and storage conditions for spare parts and valves at facilities.

172

. _ _ - . _ _ - _ _ - _ _ - _ - - - - - - - - - - - - - - - - - - -' ~ - ' " - ~ ' ' -- ~

Attachment 1 IN 86-57 July 11, 1986 Page 3 of 4 1

4

The licensee initiated a temporary surveillance program to monitor opera-bility of the solenoid coils on October 16, 1985. A modification was performed to install a voltage dropping resistor in the individual coil circuits so that they can be monitored directly from cabinets in the

' control room. This allows continuity of the coil circuitry to be verified by measuring a voltage drop across the resistor. According to the licensee, until the cause for failure can be determined, plans are to check the coil circuitry for continuity on a daily basis.

4

2. Scram Discharge Solenoid Valve Failure 4 In November 1985, Carolina Power and Light's Brunswick facility experi-enced problems with several scram discharge SOVs. The problems were idsntified during periodic surveillance testing to determine the single rod insertion times and resulted in several rods with slow insertion times. Initial troubleshooting isolated the problem to the S0Vs in the scram discharge line for two of the control rods, which were subsequently replaced and tested satisfactorily.

The licensee disassembled the failed SOVs, which were manufactured by ASCO 1

(Model HV-90-405-2A), for. failure analysis. When the valves were disas-i sembled, it was noted that copious amounts of silicone lubricant had been applied by the licensee to all gaskets, seals, and diaphrages internal to the valves during previous routine maintenance. The licensee believes-that the excessive amount of lubricant may have blocked some of the

i. valves' internal passages or caused sticking of the diaphrages, thereby contributing to the slow insertion times. The technical manual for the subject valves states that body passage gaskets should be lubricated with moderate amounts of Dow Corning's Valve Seal Silicone Lubricant or an equivalent high grade silicone grease.

The licensee conducted successful scram tests on all other rods. A l periodic retest of 10 percent of the control rods every 120 days as required by the Technical Specifications provides sufficient assurance that this problem does not exist in other SOVs. In addition, the licensee j stated that maintenance procedures and practices would be reviewed and

! modified, as required, to prevent the application of excessive amounts of lubricant during repair or overhaul of components.

Haddam Neck Nuclear Power Plant On September 10, 1985, the Haddam Neck Nuclear Power Plant was operating at 100 percent power when one of the six SOVs in the auxiliary feedwater system (AFW) failed to change state when de-energized. This failure was detected during the performance of a preventive maintenance procedure developed to periodically I cycle each of the six SOVs to prevent a sticking problem similar to SOV fail-ures previously experienced on November 2, 1984. In that earlier event, two ,

feedwater bypass valves failed to open automatically and the cause was deter- I mined to be sticking SOVs. The faulty SOV was ASCO Model NP8320A-185E and the licensee has been unable to determine the cause of the malfunction. The 173

Atttchment 1 IN 86-57 July 11, 1986 Page 4 of 4 licensee's plans are to periodically cycle the SOVs until they are either replaced with an upgraded model or the specific cause of the existing sticking problem is determined and corrected.

Millstone Nuclear Power Station Unit 1 On December 24, 1985, while performing a control rod scram time test at Mill-stone Unit 1, three control rods failed to insert during the performance of single rod scram time testing. In all cases, the control rod was immediately inserted and electrically disabled.

Investigation-into the failures revealed that in the first case the cause for failure of one sticking SOV was attributed to deterioration of the BUNA-N valve disc material within the valve. According to the licensee, this type of i

failure had been identified by General Electric in their Service Information Letter No. 128, Revision 1, dated March 2, 1984.

The licensee's investigation of the other two control rod drop failures failed to reveal the causes for failure other than a misalignment problem of one SOV's internals, which prevented proper movement. However, in each case, the SOVs were disassembled, overhauled, ratested satisfactorily, and returned to service.

Grand Gulf Nuclear Station, Unit 1 Another failure of sticking SOVs occurred at Grand Gulf Unit 1 on February 10, 1985, and was the subject of IE Information Notice No. 85-17, entitled "Possi-

, ble Sticking of ASCO Solenoid Valves."

l l

1 l

174

SSINS No.: 6835 IN 86-61 UNITED STATES NUCLEAR REGULATORY COWilSSION OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, D.C. 20555 July 28, 1986 IE INFORMATION NOTICE NO. 86-61: FAILURE OF AUXILIARY FEEDWATER MANUAL ISOLATION VALVE Addressees:

All licensees of nuclear power plants and holders of construction permits.

Purpose:

This notice is provided to alert licensees to a failure of a manual isolation valve as the result of a lack of maintenance on the valve during the operational life of the plant.

It is suggested that recipients review this information for applicability and consider actions, as appropriate, to preclude this and similar problems from occurring at their facilities. However, suggestions contained in this notice do not constitute NRC requirements; therefore, no specific action or written response is required.

Description of Circumstances:

Following the loss of integrated control system (ICS) power at the Rancho Seco l plant on December 26, 1985, the plant tripped and an overcooling transient 1 occurred. Details of this event are documented in the NRC incident investiga- i tion team's report, " Loss of Integrated Control System Power and Overcooling l Transient at Rancho Seco on December 26, 1985," NUREG-1195, February 1986.

When power was lost to the ICS, the plant responded as designed; the auxiliary feedwater (AFW) ICS flow control valves, as well as other valves, went to the 50 percent open position. The auxiliary feedwater flow was excessive. When the local manual attempt to close the flow control valve to the "A" once-through steam generator (OTSG) was unsuccessful, the operator attempted to close the manual isolation valve. This isolation valve could not be moved, even when a valve wrench was used.

The NRC incident investigation team (IIT) found that the failure of the auxiliary feedwater manual isolation valve was the result of a lack of any maintenance on this valve during the operational life of the plant, about 10-12 years. The lack of a preventive maintenance program resulted in the valve being inadequately lubricated, which caused the valve to seize.

175

. .. -- , . . - . _ _ = -. -= - -- -

(

IN 86-61 July 28, 1986

, Page 2 of 3 Discussion:

i The AFW manual isolation valve is a locked-open valve located in the AFW

discharge header to the "A" OTSG.

i The. licensee, Sacramento Muncipal Utility District (SMUD), considers that the entire AFW system, which would include this

! manual isolation valve, is safety-related. However, it appears that this valve was only intended to be used to-isolate the AFW (ICS) flow control valve for maintenance. The valve is a 6-inch, ANSI Class 900-lb, pressure seal gate manu-factured by Velan Engineering. _It is categorized as an ASME " Category E" valve

} (i.e., it is normally locked open to fulfill its function).Section XI of the ASME Boiler and Pressure Vessel Code (1974 Edition) requires no regular testing-of Category E valves. The positions of the valves are merely recorded to verify that each valve is locked or sealed in its correct position. The ASME Code no longer includes a Category E.

The Velan instruction manual provides the following guidance regarding

maintenance and operation of the valve

Lubrication of the stem threads and other working components should be

' performed frequently and at least every 6 months. A lubrication schedule recommends stem thread lubrication whenever the threads appear dry and greasing of the yoke sleeve bearings concurrently with stem thread lubrication.

Valves that are not operated frequently and may remain open or closed for

long periods of time should be worked, even if only partially, about once a month.

Proper lubrication of the stem and sleeve can reduce required operating torque by 7 to 30 percent.

] A caution also is provided not to use valve wrenches on the handwheels.

i A review of the maintenance history of the "A" manual isolation valve indicated that no maintenance (preventive or corrective) had been performed on the valve during the operational life of the plant (i.e., since 1974). The licensee had no program for preventive maintenance of manual isolation valves in the plant.

Two similar valves had failed previously, which prevented movement of the valve from the open position. The discharge isolation valve from the "A" AFW pump

failed on November 20, 1979 and the AFW manual isolation valve to~the "B" 0TSG  !

1 failed on February 20, 1980. In both cases, the yoke bearings were found seized and had to be replaced. i f During the December 26, 1985 incident, it was necessary to isolate AFW flow to the OTSG to terminate the overcooling. When the flow control valve could not be closed, the operator tried to close the manual isolation valve.

Because of the lack of maintenance, the isolation valve could not be closed i when it was needed. To address this problem, the SMUD has identified about l 100 manual isolation valves with functions similar to the AFW manual isolation valve, that will now be included in their preventive maintenance program.

176 u_ __ _ . - . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . . _ . _ _ _ _ _ _ _ . _ _ _ _ . . - _ , , _

IN 86-61 Julu 28, 1986 Page 3 of 3 Additional discussion on the AFW manual isolation valve is included in Section 5.3 of NUREG-1195.

No specific action or written response is required by this information notice.

If you have any questions regarding this matter, please contact the Regional Administrator of the appropriate NRC regional office or this office.

-- J _

ard Jordan, Director Divisi of Emerger.cy Preparedness and gineering Recponse Office of Inspection and Enforcement Technical

Contact:

R. Wright, NRR (301) 492-8900 H. Bailey, IE (301) 492-9006

Attachment:

List of Recently Issued IE Information Notices l

l l

l 177

SSINS NO: 6835 IN 86-62 UNITED STATES NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, D.C. 20555 July 31, 1986 IE INFORMATION NOTICE NO. 86-62: POTENTIAL PROBLEMS IN WESTINGHOUSE MOLDED CASE CIRCUIT BREAKERS EQUIPPED WITH A SHUNT TRIP Addressees:

All nuclear power reactor facilities holding an operating license or a con-struction permit.

Purpose:

This notice is to alert recipients to a potentially significant problem involv-ing the failure of shunt trip coils in Westinghouse molded case circuit break-ers (breakers) type LBB 22250 MW. Breaker types LB 22250, HLB 22250, and DA 22250 have similar operating mechanisms which can affect the operation of the shunt trip coils. However, to date no failures of the shunt trip coil have been reported for these breakers. It is expected that recipients will review this information for applicability to their facilities and consider actions, if appropriate, to preclude a similar problem from occurring at their facilities.

However, suggestions contained in this information notice do not constitute NRC requirements; therefore no specific action or written response is required.

Description of 'ircumstances:

C On February 6, 1986, an open occurred or was discovered in the shunt trip coil (STC) circuitry which could have resulted in the failure of the associated breaker to trip open on an abnormal voltage condition at the Peach Bottom Reactor. The breaker is identified as an LBB 22250 MW type molded case circuit .

breaker manufactured by Westinghouse. The breaker is installed in a transfer panel to protect the 120-V 60 Hertz power supply to the reactor protection system (RPS) from undervoltage, overvoltage and underfrequency conditions.

In RPS applications, loss of the STC prevents the breakers from tripping automatically on abnormal voltage conditions. Tripping of this breaker on abnormal voltage conditions is essential because it interrupts a potentially damaging abnormal voltage supply to the RPS relays, scram solenoids, and other safety-related electronic devices. However, failures of the STC does not, by itself, prevent a reactor scram. Also, the breaker is equipped with a magnetic overcurrent trip device that protects the circuits in the event of a fault.

The STC in series with a contact is activated through a toggle linkage by the breaker's moving main contact arms. The failure of the STC occurred when the contact did not open when the breaker tripped resulting in the overheating of 179 .

_ _ _ _ _ _ . - - - - - __ _ . - _ _ - - =- .- -- -

IN 86-62 A July 31, 1986 Page 2 of 3 the STC and ultimately in an open circuit in the shunt trip circuit. If the breaker had been subjected to an abnormal voltage condition after reset the open shunt trip circuit would have prevented breaker trip.

j On January 16 and 24, 1986 the same types of problems were identified in similar breakers at the Peach Botton Unit 3 plant. On November 9, 1984, Limerick 1 rsported an identical failure.

It is essential that the STC remain operable to assure the circuit breaker trips in the event of an abnormal voltage condition.

Administrative measures have been established to periodically test and verify the operability of the STC at the Peach Bottom and Limerick nuclear power plants.

Discussion:

The failures described above have occurred in LBB 22250 MW type breakers with a STC.

Westinghouse supplied the breakers to ASCO Electrical Products, Incorpo-rated who installed them in safety-related RPS power supply monitoring panels raquired for boiling water reactors (BWRs). LB 22250, HLB 22250, and DA 22250 i type breakers, which have similar operating mechanisms to LBB 22250, may have i

been supplied to other manufacturers for use in safety-related applications.

The performance of the breaker is affected only when used with a STC. The STC is energized by the closure of a normally open contact which is actuated through a toggle linkage by the breaker's moving main contact arms. Westinghouse stated that the moving contact arms may be impeded from being fully displaced i

i to the "open" position by excessive material in the rivet which holds the handle post to the operating mechanism.

The 2 pole molded case circuit breaker is equipped with an A contact that is operated by toggle operating links controlled by the breaker operating handle.

The A contact is an auxiliary switch which is open when the breaker is in the open or tripped position and is closed when the breaker is closed ("0N" posi-tion). This A contact is in series with the 125-V de STC. When the breaker is closed, the auxiliary A contact is closed and the STC is ready to receive a signal to trip the breaker if any one of the RPS power monitoring relays senses an abnormal voltage condition. The STC is not rated for continuous duty and will overheat and be damaged if subjected to full voltage for more than,a few l seconds.

The STC can be made to fail, after the breaker has tripped by either of the following actions or conditions:

1. If the operating handle is pushed to the "Oh" position without resetting the breaker, when a trip signal is present. This will not close the

! breaker, but will close the A contact, thus continuously energizing the STC.

I 2. If the previously described interference between the operating handle and

" the toggle operating links exist the act of resetting the breaker and, before clearing the trip signal, attempting to close the breaker. The i

i toggle operating links may jam in the "0N" position, unless the operating handle is pushed toward the "0FF" or " RESET" position, again causing the l STC to be continuously energized.

1 i

Id0 i

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

IN 86-62 July 31, 1986 l

Page 3 of 3 The breaker will trip even if the operating handle is forcibly kept in the "0N" position when~a trip signal is received or present if the STC is operable. However, because the operating handle is in the "0N" position, the contact will not open to disconnect the power supply to the STC.

Westinghouse recommends the following tests to ascertain the operability of the breaker.

1. Perform a continuity check on the STC after each breaker operation via the two leads exiting the breaker.

2. -Operability of the STC protection can be determined by completing the following tests:

a. Manually close the breaker and continue to hold the operating handle in the "0N" position.

b. Apply the shunt trip rated voltage through the two leads exiting the i .

breaker until the breaker trips or one second elapses.

c. While continuing to hold the handle in the "0N" position, perform a continuity check of the shunt trip circuit. Continuity indicates a malfunction in the contact. An open circuit indicates that the contact has performed its intended function.

Westinghouse revised the circuit breaker mechanism assembly drawing to specify the removal of excessive rivet material and prevent recurrence of this problem.

No specific action or written response is required by this notice. If you have any questions regarding this notice, please contact the Regional Administrator of the appropriate NRC regional office or this office.

2 '

Divisip).'ofJordan, Emergency Director Preparedness and gineering Response Office of Inspection and Enforcement Technical Contacts: K. R. Naidu, IE (301) 492-4179 James C. Stewart, IE (301) 492-9061

Attachment:

List of Recently Issued IE Information Notices l

181

(_.____ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ - - - - _ _ . _ - _ _ . _

SSINS No.: 6835 IN 86-65 UNITED STATES NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND. ENFORCEMENT WASHINGTON, D.C. ~20555 l

August 14, 1986 I

IE INFORMATION NOTICE NO. 85-65: MALFUNCTIONS OF ITT BARTON MODEL 580 SERIES SWITCHES DURING REQUALIFICATION TESTING Addressees:

All nuclear power reactor facilities holding an operating license or a construction permit.

Purpose:

i _This notice is provided to alert recipients to a potentially significant safety problem pertaining to malfunctions of ITT~Barton Model 580 Series indicating 4

switches under accident conditions. It is expected that recipients will review the information for applicability to their facilities and consider actions, if appropriate, to preclude malfunctions occurring at their facilities. However, suggestions contained in this notice do not constitute NRC requirements; i therefore, no specific action or written response is required.

Description of Problem:

On April 14, 1986, NRC was notified by ITT'Barton Instruments Company of the defect in accordance with 10 CFR Part 21. According to the report, during requalification testing of ITT Barton Model 580 Series indicating switches a under loss-of-coolant accident conditions, switch malfunctions occurred when the chamber temperature was raised to 340*F. The symptoms of the malfunctions were:

- In three of the five instruments tested, one or both of the switches failed to operate (no change in switch state) when input pre m re to the instrument was varied.

l-

- Switch set-point shift in excess of the allowable 10 percent occurred on two of the instruments.

When the chamber temperature was lowered to room ambient, all of the switches

, changed state when input pressure was varied; however, the switch set-point

, shifts which occurred at temperature were permanent.

The initial report indicated that the failure was due to a specific component ,

used, a Honeywell Snap-Acting Switch, part number 11SM 403. A subsequent

~

report indicated that, as a result of additional testing, it was determin'ed

! that there is also a deflection of the instrument case which-affects the position of the snap switch and the setpoint. Furthermore, the deflection can result in failure of case seal integrity.

183 1

IN 86-65 August 14, 1986 Page 2 of 2 ITT Barton has indicated that they are evaluating case / seal redesigns and a realistic date for completion of ratesting would be April 15, 1987.

In previous tests, the instruments performed successfully when subjected to 220*F after aging, irradiation, and seismic testing.

This information notice is being issued to ensure that all nuclear power plant end-users have been notified of this problem.

A contact at ITT Barton Instruments Company knowledgeable of this prc!1em is:

John P. Doyon Manager, Sales and Service 3 (818) 961-2547 extension 456 No specific action or written response is required by this information notice.

If you have any questions about this matter, please contact the Regional Administrator of the appropriate regional office or this office.

orden Directo'r Divisio f Emergency Preparedness and E ineering Response Office of Inspection and Enforcement Technical Contacts: Eric Weiss, IE (301) 492-9005 Pentti Koutaniemi, IE (301) 492-9428 Attachments: List of Recently Issued IE Information Notices i

l 1

184

SSINS No.: 6835

, IN 86-66

-UNITED STATES NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT

-WASHINGTON, D.C. 20555' August 15, 1986, IE INFORMATION NOTICE NO. 86-66: POTENTIAL FOR FAILURE OF REPLACEMENT AC COILS SUPPLIED BY THE WESTINGHOUSE ELECTRIC CORPORATION FOR USE IN CLASS 1E MOTOR STARTERS AND CONTACTORS Addressees:

All nuclear power reactor facilities holding an operating license or a construction permit.

Purpose:

This notice is to alert licensees that certain Westinghouse Electric Corporation ac coils have shown a higher failure rate than previously experienced.- These

-coils are similar to Class 1E coils manufactured and supplied by Westinghouse as replacement parts for use in Class 1E motor starters and contactors.

i It is expected that the licensees will review this information for applicability to their facilities and consider actions, if appropriate, to preclude the use of the defective coils. However, suggestions contained in this information notice do not. constitute NRC requirements; therefore, no specific action or written notice is required.

Description of Circumstances:

On June 19, 1986 Westinghouse Water Reactor Division (WRD) submitted a 10 CFR Part 21 report to the NRC indicating a higher-than-normal failure rate for ac coils. The report stated that most of the failures had been encountered during the initial hours of energization. These coils were manufactured at the Westing-house Control Division (WCD) facility in Coamo, Puerto Rico, between June 1, 1984 and December 31, 1985 and are provided as replacement parts for use in motor starters and contactors.

Discussion:

During early June 1986, Westinghouse concluded that the Class 1E coils used in certain motor starters and contactors in nuclear plants could be subject to failure. This conclusion was based on information from various non-nuclear customers about failures of similar coils. However, it should be noted that up to this time no nuclear power plant Class 1E coil failures had been reported to WRD.

185

IN 86-66 August 15, 1986 Page 2 of 2 Discussions between Westinghouse divisions have determined that the non-Class 1E coils manufactured between June 1, 1984 and December 31, 1985 experienced a signi-ficant increase in failure rate compared to prior history. These reported coil failures occurred primarily less than 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> to several hours after the coils had been continuously energized. To correct the problem, WCD made a number of manufacturing changes that included revised materials and processes. Westinghouse ,

tests performed on newly manufactured coils have indicated that the modified coils are not subject to the same failures as those coils manufactured during June 1, 1984 through December 31, 1985. Westinghouse states, "Any coil manufactured since the beginning of 1986 is free of the potential for such failures."

WRD has determined that any coil which has been successfully energized contin-ususly (not cumulative) for 5 days or more is expected to perform satisfactorily and need not be replaced. Attached is a list of the plants with potentially dGfective coils that have been notified by WRD, followed by the WRD list of date codes of potentially defective coils.

For facilities not on the list and using or stocking any WCD coils with one of the listed date codes, it is suggested that the Westinghouse Water Reactor Division be contacted for information on the need for corrective action.

4 No specific action or written response is required by this information notice.

If you have any questions about this matter, please contact the Regional l Administrator of the appropriate NRC regional office or this office.

4

. Ed rdL.I) n M ctor Division of Emergency Preparedness and gineering Response Office of Inspection and Enforcement Technical

Contact:

L. B. Parker, IE (301) 492-7190 Attachments:

1. List of Facilities and Date Codes

2. List of Recently Issued IE Information Notices 186

SSINS NO. 6835 IN 86-71 UNITED STATES NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, D.C. 20555 August 19, 1986 l

! IE INFORMATION NOTICE NO. 86-71: RECENT IDENTIFIED PROBLEMS WITH LIMITORQUE MOTOR OPERATORS Addressees:

All nuclear power reactor facilities holding an operating license or a construction permit.

Purpose:

l This notice is provided to alert recipients of two potential problems discovered with Limitorque motor operators. It is expected that recipients will review this information for applicability to their facilities and consider actions, if appropriate, to preclude a similar problem from occurring at their facilities.

However, suggestions contained in this notice do not constitute NRC requirements; therefore, no specific action or written response is required.

Past Related Documents:

IE Information Notice 86-03 " Potential Deficiencies in Environmental Qualification of Limitorque Motor Valve Operator Wiring," January 14, 1986.

Description of Circumstances - Burn Damage to Internal Wiring On November 8, 1985 Georgia Power Company submitted a preliminary report to the NRC indicating that it had. discovered burn damage to internal wiring in several Limitorque motor operators installed in their Vogtle Unit 1 Power Plant.

Evidence suggested the burn damage had been caused by electric heater elements installed in the limit switch compartment for storage purposes within certain types of Limitorque motor operators.

On March 20, 1986 Georgia Power Company submitted a final report to the NRC which suggested that the burn damage was a generic problem applicable to all Limitorque motor operators. This assumption was based on a sampling inspection of 104 Limitorque motor operators installed in Vogtle Unit 1. Forty-six of the motor operators examined were Limitorque type SMB-000, and six of these were found to have burnt internal wiring. Out of the 58 operators other than type SMB-000 which were inspected, 5 were found to contain wires deemed susceptible to damage because of their close proximity to the heater elements (less than inch).

137

IN 86-71 August 19, 1986 Page 2 of 3 On May 20, 1986 an NRC inspector, along with Limitorque, Georgia Power, and Bechtel personnel, performed a random inspection on four type SMB-000 operators installed in Vogtle Unit 2. Three of the four operators were found to contain

' burnt internal wiring. One type SMB-00 operator also was inspected and was found to contain wiring susceptible to damage.

Discussion:

T.he wiring in question is Limitorque installed internal wiring located in the ep:rator limit switch compartment. The wiring is being burnt as a result of its close proximity to, or contact with, the installed limit switch compartment  !

olsctric heater element or heater bracket. The wiring is not properly routed End is not restrained from contacting the heater or heater bracket. Although the heater is not a seismically or environmentally qualified part and is intended for use only during storage, its use has been shown to cause serious degradation of snvironmentally qualified internal wiring.

! Th2 burnt wiring has been discovered only in Limitorque type SMB-000 motor 4

cp;rators that contain previously energized heaters, although any Limitorque j cperator that contains a previously energized heater could possibly exhibit a siallar problem.

] 'D7scription of Circumstances - Cracked Limit Switch Rotors Several licensees have subreitted reports to the NRC concerning a problem with cracked limit switch rotors on Limitorque motor operators installed inside and cutside of containment. The limit switches are used for control of the motor l cpsrator and also provide indication of valve position in the control room.

i The cracks ha've been found on white melamine limit switch rotors. Most of these l cracks were found in the area where the limit switch rntors are pinned to the i

pinion shafts. Some cracks have been found to extend halfway through the melamine rotors, weakening them to the extent that they are easily broken.

I Discussion:

In its letter dated February 21, 1984 to the Westinghouse Electric Corporation,

Electro Mechanical Division, Cheswick, Pennsylvania, Limitorque recommended

! ,that any white limit switch components which are found with cracks should be i

replaced with components manufactured with a later design, brown colored material.

l Violations of 10 CFR Part 21 have been issued to Limitorque for failure to report

and for failure to evaluate defects discovered in their motor operators. Licensees

! are reminded of their responsibility to ensure that procurement documents include 1

a contractual requirement that the provisions of 10 CFR Part 21 apply (when applicable) and that Criterion VII of 10 CFR Part 50 Appendix B requires that purchased material, equipment, and services conform to the requirements of the i procurement documents.

3 138 l

IN 86-71 August 19, 1986 Page 3 of 3 No specific action or written response is required by this information notice.

If you have any questions about this matter, please contact the Regional Admin-istrator df the appropriate NRC regional office or this office.

.- / /-t

/

Edward L.

)ordan, cW Director tb' n Division . Emergency Preparedness and Eng neering Response Office of Inspection and Enforcement Technical

Contact:

Jeffrey Jacobson, IE (301) 492-8845

Attachment:

List of Recently Issued IE Information Notices i

4 Id')

SSINS No.: 6835' IN 86-72 UNITED STATES NUCLEAR REGULATORY COMISSION 0FFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, D.C. 20555 August 19, 1986 IE INFORMATION NOTICE N0. 86-72: FAILURE 17 7 PH STAINLESS STEEL SPRINGS IN VALCOR VALVES DUE TO HYDROGEN DBRITTLEMENT 4 Addressees:

All nuclear power reactor facilities holding an operating license or a construction permit. ,

Purpose:

This notice is provided to inform recipients of a potentially significant safety 4 problem that could result from the failure of springs in solenoid globa valves manufactured by Valcor Engineering Corporation. According to the valve manu-i facturer these valve springs may fail when exposed to high temperature reactor j coolant containing hydrogen.

l It is expected that recipients will review the infomation for' applicability 1

to their facilities and consider action, as appropriate, to preclude a similar problem from occurring at their facility. However, suggestions contained'in i this information notice do not constitute NRC requirements; .tberefore, no spe-a cific action or written response is required.

Description of Circumstances:

Difficulties were experienced with the operability of two solenoid-operated

globe valves (Model V526-6190A, p/n 454660001) in the charging system at the i Fort Calhoun Station, Unit 1 in August 1985. When ' shot, the valves could not be reopened without securing all charging pumps. Du.ing a refueling out' age in January 1986, the two valves were disassembled and examined to determine the cause of the valve malfunction. It was found that disc guide assembly springs in both valves had undergone complete and catastrophic failure. The springs, i

which initially had 25 coils, were found in sections of od y 1-2 coils. Metal-10rgical analysis of the failed springs attributed the probable cause of failure was due to hydrogen embrittlement. The spring is made of 17-7 PH stainless steel.

Discussion with the valve manufacturer, Valcor Engineering Corporation, revealed that during 1982-83 one failure occurred at Prairie Island Nuclear Generating Station and two failures occurred at North Anna Nuclear Generating Station.

These spring failures were also attributed to hydrogen embrittlement.

191

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

IN 86-72 August 19, 1986 Page 2 of 2 Discussion:

4 Based on analysis and evaluation of the earlier spring failures, the valve j

manufacturer issued a letter in 1983 to affected licensees requesting informa-

• tien on valve application and advised customers of a potential spring problem.

Elgiloy springs were offered on receipt of information confirming use with r: actor chemistry fluid or specific customer request. With the occurrence of the third similar event, the valve manufacturer is planning to issue a second 1ctter to affected licensees conservatively recommending that valves with spring  ;

mat rial of 17-7 PH stainless steel used in borated water or reactor chemistry wat:r be closely monitored and evaluated for any change in normal operation such

' as increased seat leakage or an increase in the time required to change position.

Th se conditions could be attributed to broken springs. The manufacturer has concluded that hydrogen embrittlement of stainless steel springs is a complex function of high temperature, water chemistry, water flow condition, and time of exp:sure to the service condition. Therefore, all such springs made of 17-7 PH stainless steel used in Valcor valves in nuclear power plant may be susceptible to replacement.

this failure mode under these conditions and should be considered for i The above described events are an indication of potential licensee / vendor int rface problem. Based on the information received by the NRC, the vendor was not completely informed via the purchase specifications regarding the service condition to which the valve would be exposed. Further, all users of Valcor

' valves were not notified of the initial problem through either oversight by the vendor or as a result of the valves being supplied through an intermediate source.

j No specific action or written response is required by this information notice, If you have any questions about this matter, please contact the Regional Administrator of the appropriate regional office or this office.

i' dwar.d Jordan, Director Divifio of Emergency Preparedness j anch ngineering Response

Office of Inspection and Enforcement Technical

Contact:

L. D. Vaughan, IE

! (301) 492-8811

)

l

Attachment:

List of Recently Issued IE Information Notices l

i I

4 f 192

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

SSINS No.: 6835 e IN 86-73 L UNITED STATES NUCLEAR REGULATORY COMISSION OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, D.C. 20555 August 20, 1986 IE INFORMATION NOTICE NO. 86-73: RECENT EMERGENCY DIESEL GENERATOR PROBLEMS Addressees:

All nuclear power reactor facilities holding an operating license or a construction permit.

Purpose:

This notice is to alert addressets to vibration-induced fuel line wear and of a deficiency-in the design of the field flash circuitry on nuclear plant emer-gency diesel generators. Recipients are expected to review the information for applicability to their facilities and consider actions, if appropriate, to pre-clude similar problems occurring at their facilities. However, suggestions contained in this information notice do not constitute NRC requirements; there-fore, no specific action or written response is required.

Description of Circumstances:

Nine Nile Point Unit 2 While conducting diesel generator testing in early May 1986, it was discovered that diesel fuel lines had experienced extensive wear and fuel leaks in the l area of the clamps that mount the fuel lines to the diesel engine. The diesels  !

are Cooper-Bessemer model KSV-16-T, 600 rpm, 4 stroke,16 cylinder units with low total operating hours.

Fuel line damage was caused by vibration from the diesel engine and fuel system l pulsation induced by rapid, repeated cycling of a fuel systea relief valve.

This valve relieves from the low pressure fuel system via a cooler to the fuel day tank to control low pressure fuel system pressure. The manufacturer proposes to correct the problem by inserting plastic sleeves between the fuel lina and its hold down clamps and installing a dashpot on the relief valve to dampen its operation. 1 Watts Bar Units 1 and 2 In April 1986 a deficiency was identified which affects all five standby diesel generators (DGs) at Watts Bar Nuclear Plant and could prevent the DGs from developing a voltage output when required in an emergency. The affected DGs are tandem 16-645 E4 units supplied by Morrison-Knudson Co. The normal shutdown cycle of the DG includes a 10-minute cooldown run at about 450 rpm. If during l 193 l

IN 86-73 August 20, 1986 Page 2 of 2 this idle period an emergency start signal were received, the DG would accelerate to the normal 900 rps operating speed, but the generator field would not be flashed and an output voltage therefore would not be developed.

This problem has also been 6etermined to exist on the HPCS DG at Grand Gulf Nuclear Station. This unit was supplied by General Motors Corporation.

The root cause of this deficiency has been found to be a design error by the manufacturer. During a normal or emergency start, as the DG accelerates past 475 rpm, logic is completed to flash the generator field. When output voltage has built up, the field flash circuitry is automatically disabled. The logic design is such that engine speed must go below 200 rps to re-enable the field flash circuitry, thus no field flash will occur if an emergency start signal is received during the 450 rps cooldown period. Field flash would be needed under these circumstances because the self-excitation path is interrupted early in the shutdown sequence.

The corrective action proposed by the DG manufacturer is to modify the control circuitry to eliminate the speed dependence of field flash reset.

No specific action or written response is required by this information notice.

If you have any questions about this matter, please contact the Regional Administrator of the appropriate regional office or this office.

$1 dward . Jordan, Director Division of Emergency Preparedness and L insering Response Office of Inspection and Enforcement Technical

Contact:

Kevin Wolley, IE (301) 492-8373

Attachment:

List of Recently Issued IE Information Notices 194

SSINS No.: 6835 IN 86-77 UNITED STATES NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, D.C. 20555 August 28, 1986 IE INFORMATION NOTICE NO. 86-77: COMPUTER PROGRAM ERROR REPORT HANDLING Addressees:

All nuclear power reactor facilities holding a construction permit or an operating license and nuclear fuel manufacturing facilities.

Purpose:

This notice is to alert addressees that errors are being identified in computer programs used during safety-related design activities. These design activities, including facility modifications and reload calculations, may be invalidated by errors found in computer programs used to support safety-related design calcula-tions. While these errors are contained in error reports prepared by computer service bureaus, licensees using the program as a basis for safety-related activities may not be aware that a significant number of errors are being iden-tified. It is expected that recipients will review the infonnation for applica-bility to their quality assurance programs and consider actions, if appropriate, to preclude similar problems from occurring at their facilities. However, suggestions contained in this information notice do not constitute NRC require-ments; therefore, no specific action or written response is required.

Description of the Circumstances:

During reviews of the implementation of computer program error report handling procedures at various architect engineering companies (A/Es), nuclear steam supply system companies (NSSS), and nuclear fuel manufacturing facilities, the NRC has learned that there are a significant number of errors being found in computer programs used for safety-related design. Further, users (Licensees, A/E's) may not be implementing appropriate measures to ensure that these errors do not invalidate safety-related calculations already completed, in progress, or to be conducted at a future date.

The computer program errors have usually been found by individual program users and reported to the computer service bureaus which subsequently report these errors to all customers using the program, provided that the requirements of 10 CFR 21 are specified in contracts between the service bureaus and affected customers.

195

IN 86-77 August 28, 1986 Page 2 of 3 Discussion:

A computer program is a basic component as defined in 10 CFR 21 when used in safety-related design activities. In addition, control measures are required to prevent the use of incorrect or defective material, parts, and components as discussed in Criterion VIII of Appendix B to 10 CFR 50. Similarly, measures are required to ensure that conditions adverse to quality, such as deficiencies and nonconfonnances, are promptly identified and corrected as discussed in Criterion XVI, Appendix B,10 CFR 50.

Utilities holding a CP or OL have the primary responsibility to ensure that computer coda errors are adequately reviewed and their impact on past and present safety-related design calculations are evaluated.

As an example, a recent 10 CFR 21 report to the NRC identified an error in the Rayleigh frequency calculation contained in the GT STRUDL computer code which resulted in numerous safety-related systems having to be reanalyzed. The error was found during perfonnance of an analysis check on a previously completed Duct Support and was related to the method used by the program to select seismic design loads. The code, marketed by the Control Data Corporation (Service Bureau) and technically supported by its author, Georgia Institute of Technology, is used throughout the nuclear industry, primarily in the analysis and design of pipe supports and general building structural design. A thorough review and evaluation of affected designs, in addition to applicable vendors that may have used GT STRUDL in safety-related applications, is currently being performed by several CP holders. One CP holder recently reported that this error affected 960 calculations. However, program users and Service Bureau subscribers such as Licensees, A/E's and NSSS organizations who did not specify the requirements of 10 CFR 21 in their contract with the service bureau may not be aware of this error.

Another example involved the discovery by a nuclear fuels manufacturer of an input error in a loss-of-coolant-accident (LOCA) code used to calculate the effects of fuel rod heatup. This computer code error was identified during perfonnance of a fuel reload analysis. The error resulted in the value of fuel rod decay heat generation being too low, causing the calculated peak cladding temperature (PCT) to be unconservative. Corrected calculations showed that the value of PCT exceeded 2200 F, resulting in several licerisees having to reduce power to comply with the provisions of 10 CFR 50.46.

Several documents are available which may be useful to licensees when reviewing computer code controls used by vendors engaged in safety-related activities.

NUREG-0040, " Licensee Contractor and Vendor Inspection Status Report," published quarterly by the NRC provides a detailed, technical and programmatic review of organizations engaged in supplying safety-related equipment or services to licensed facilities. This NUREG discusses important plant safety elements, including computer code usage, maintenance, and error report handling, for firms such as NSSSs, AEs, and nuclear fuel suppliers. Documents that also may be 196

IN 86-77 August 28, 1986 Page 3 of 3 useful include IE Information Notices Nos. 85-52, " Errors In Dose Assessment Computer Codes and Reporting Requirements Under 10 CFR Part 21" (July 10,1985) and 83-31. " Error in the ADLPIPE Computer Program" (May 19,1983). These two notices focus primarily on computer code errors and 10 CFR 21 reporting respon-sibility.

No specific action or written response is required by this information notice.

If you have any questions regarding this matter, please contact the Regional Administrator of the appropriate NRC regional office or this office.

a ctor Divisi of Emergency Preparedness and ngineering Response Office of Inspection and Enforcement Technical

Contact:

R. Pettis, Jr. , IE (301) 492-9039

Attachment:

List of Recently Issued IE Information Notices 197

SSINS No.: 6835 IN 86-78 UNITED STATES NUCLEAR REGULATORY C0lWISSION OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, D.C. 20555 September 2, 1986 f IE INFORMATION NOTICE NO. 86-78: SCRAM SOLENOID PILOT VALVE (SSPV) REBUILD KIT PROBLEMS Addressees:

All boiling water reactor facilities holding an operating license or a construction permit.

Purpose:

This notice is to alert recipients of a potential problem with kits used to refurbish the scram solenoid pilot valves. Recipients are expected to review the information for applicability to their facilities and consider actions, if appropriate, to preclude similar problems occurring at their facilities.

However, suggestions contained in this information notice do not constitute NRC requirements; therefore, no specific action or written response is required.

Description of Circumstances:

On June 14, 1986, Vermont Yankee nuclear power plant reported that one control rod failed to scram and five others hesitated a few seconds before scramming during a single-rod scram time test. During the outage that preceded the scram time test, all of the scram solenoid pilot valves had been rebuilt with replacement kits supplied by General Electric Company (GE). The reactor had been taken critical to perform shutdown margin tests using the in-sequence critical method before the testing that identified the problem with the scram solenoid pilot valves.

Three types of problems were identified in the scram solenoid valves (SSPVs) which operated six control rods. In the one SSPV associated with the failure to scram, the core spring of the SSPV was separated from the core assembly. On another SSPV, the diaphragm was installed backwards on the exhaust side of the solenoid valve. On the remaining four SSPVs, an incorrect core assembly, provided with the kits, was installed in the valve. The latter two types of problems were associated with delayed scram initiation. Subsequent inspection of the remaining SSPVs revealed two other types of discrepancies. These were (1) out-of-round inside diameter of the solenoid base subassembly and (2) a deformed spring. Although these two discrepancies did not cause abnormal scram performance in this case, they could have had an adverse effect on scram performance.

199

IN 86-78 August , 1986 Page 2 of 2 Discussion:

These problems would likely delay but not prevent rod insertion during normal operation because backup scram valves would depressurize the air header and cause the control rods to insert.

The defective rebuild kits are used at BWR-2s, 3s, and most 4s and 5s. Vermont Yankee used replacement kits (ASCO type 204-139) to refurbish the scram solenoids in the Hydraulic Control Units (HCU). GE purchased 3000 of these replacement kits from Automatic Switch Company (ASCO), the manufacturer of the solenoid valves. GE purchased these kits as non-safety-related items and sold them as nuclear grade. Each kit contains 11 components, of which two are assemblies. The two assemblies are the core assembly (ASCO part 65-716-2A) and the solenoid base sub-assembly (ASCO part 44-869-23).

On June 26, 1986, subsequent to the event at Vermont Yankee, GE returned to ASCO 200 replacement kits (ASCO type 204-139) from their stock and req'uested ASCO to perform critical inspections. ASCO inspected the two assemblies in each kit and rejected 127 core assemblies and two solenoid base assemblies for out of tolerance conditions. The rejected parts were replaced with acceptable assemblies.

Although the extent of the distribution of the parts kits used to rebuild the scram solenoid valves is not known at this time, preliminary information suggests that these kits may be in wide distribution. GE issued Rapid Information Communication Services Information Letter (RICSIL) No. 008 on June 27, 1986 and SIL No. 441 on July 17, 1986 regarding these problems with the rebuild kits. GE has advised all affected utilities to return spare rebuild kits for reinspection.

No specific action or written response is required by this information notice.

If you have questions about this matter, please contact the Regional Adminis-trator of the appropriate NRC regional office or this office.

gwa/ Jordan, rd

,1 Director -

Divisi of Emergency Preparedness and Engineering Response Office of Inspection and Enforcement Technical Contacts: Eric Weiss, IE (301) 492-9005 K. R. Naidu, IE (301) 492-4179

Attachment:

List of Recently Issued IE Information Notices 200

SSINS No.: 6835 IN 86-81 UNITED STATES NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, D.C. 20555 September 15, 1986 IE INFCRMATION NOTICE NO. 86-81: BROKEN INNER-EXTERNAL CLOSURE SPRINGS ON ATWOOD & MORRILL MAIN STEAM ISOLATION VALVES Addressees:

All nuclear power reactor facilities holding an operating license or a construc-tion permit.

Purpose:

This notice is provided to inform recipients of a potentially significant safety problem that could result from broken inner-external closure springs on main steam isolation valves (MSIVs) manufactured by Atwood & Morrill Co., Inc. The springs that failed were manufactured by Duer Spring and Manufacturing Co.

Quench cracks, which apparently developed during the' manufacturing process, caused the springs to fail.

It is expected that the recipients review the information for applicability to their facilities and cor. sider actions, if appropriate, to preclude a similar problem from occurring at their facilities. However, suggestions contained in this information notice do not constitute NRC requirements; therefore, no specific action or written response is required.

Description of Circumstances:

During an inspection of MSIVs following low pressure seat leakage tests at Enrico Fermi Unit 2 in May 1986, the licensee observed that four inner-external closure springs were broken into two-to-seven pieces. The licensee analyzed two of the broken springs and determined the failure to be quench cracking caused by the heat treatment process during manufacturing. There are eight MSIVs with eight inner-external and eight outer-external closure springs per valve. All outer-external closure springs were found intact and undamaged. All 64 external springs were from the same heat number (#8067703). The external closure-spring assemblies (inner and outer) perform a safety function to close the MSIVs.

Discussion:

Springs from Duer Spring and Manufacturing Co. were supplied to PWR and BWR nuclear power plants on MSIVs manufactured by Atwood & Morrill. The external closing springs that failed at Fermi 2 were the inner springs around the valve yoke rods. The following effects on valve operation could occur with one or b

9 more springs failing:

i 201

l IN 86-81 September 12, 1986 Page 2 of 2

1. Slower closing time.with steam in the normal forward flow direction. For BWRs, General Electric estimates the closure time would increase from the Technical Specification requirements of 3 to 5 seconds but will remain under 10 seconds.

2. Increased difficulty in meeting low pressure seat leakage test requirements in BWR units.

3. Valve closure with steam in the reverse flow direction in PWR units could be adversely affected since the springs provide the only external closing assistance. Internal closing assistance is provided by the use of a pilot poppet design in these valves.

The spring manufacturer also performed laboratory metallographic examination of the failed springs and verified that the failure was the result of quench cracking. Atwood & Morrill and Duer Spring have recommended that at the first available opportunity all the external closing springs on all MSIVs be cleaned and magnetic particle testing be performed. Duer has provided an inspection procedure.

Atwood & Morrill is issuing a letter to all affected customers concerning this event which recommends that the above actions be taken. The Atwood & Morrill list of affected units is attached. In addition, General Electric has issued a service information letter, SIL No. 442 dated July 18, 1986, to owners of affected BWRs recommending a visual inspection and, in some cases, load tests.

The NRC has reviewed the two procedures and has determined that either recommendation is adequate for BWRs. For PWRs, only the Atwood & Morrill recommendations are appropriate. However, consideration should be given to visual inspections of_the springs following any cycling of the valve until completion of the Atwood & Morrill recommended inspection.

No specific action or written response is required by this information notice.

If you have any questions about this matter, please contact the Regional Admini-strator of the appropriate regional office or this office.

.f

dward .

L - - . - -

Jordan, Director Divisi of Emergency Preparedness and Engineering Response Office of Inspection and Enforcer. writ Technical

Contact:

J. C. Stone, IE (301) 492-9044 Attachments:

1. List of affected plants as identified by Atwood & Morrill

2. List of Recently Issued IE Information Notices 202

Attachment 1 IN 86-81 September 15, 1986 Page 1 of 3 ATWOOD & MORRILL CO., INC.

Main Steam Isolation Valves Furnished With External Closing Springs Manufactured by Duer Spring & Manuf. Co.

PWR UNITS g Utility Station Quantity and Valve Size i

-l Duke Power Co. Catawba 1 & 2 8 34-inch )

Duke Power Co. McGuire 1 & 2 8 32-inch Duke Power Co. Perkins 1, 2, & 3 12 32-inch Units Cancelled Duke Power Co. Cherokee 1, 2, & 3 12 32-inch Units Cancelled -

Houston Lighting South Texas 8 32-inch

& Power Project 1 & 2 South Carolina Virgil Summer 1 3 32-inch Elec & Gas TVA Bellefonte 1 & 2 8 32-inch TVA Sequoyah 1 & 2 8 32-inch TVA Watts Bar 1 & 2 8 32-inch 203

Attrchment 1 IN 86-81 September 15, 1986 Page 2 of 3 ATWOOD & MORRILL CO., INC.

Main Steam Isolation Valves Furnished With External Closing Springs Manufactured by Duer Spring & Manuf. Co.

BWR UNITS Utility Station Quantiy & Valve Size Boston Edison Pilgrim 1 8 20-inch Cleveland Elec. Perry 1 & 2 16 26-inch Illuminating ~

Detroit Edison Enrico Fermi 2 8 26-inch Georgia Power Hatch 1 8 24-inch Gulf States Util. River Bend 1 8 24-inch Houston Lighting Allens Creek 1 8 24-inch Unit cancelled

& Power Illinois Power Clinton 1 8 24-inch Jersey Central Oyster Creek 1 4 24-inch Power & Light Mississippi Power Grand Gulf 1 & 2 16 28-inch

& Light Niagara Mohawk Nine Mile Point 1 2 24-inch Power Corp.

Northern States Monticello 8 18-inch Power Co.

Pennsylvania Susquehanna 1 & 2 16 26-inch

. Power & Light Philadelphia Limerick 1 & 2 16 26-inch Electric Philadelphia Peach Botton 2 & 3 16 26-inch Electric Public Service Hope Creek 1 & 2 16 26-inch Unit 2 cancelled Elec & Gas 204

( _ _ _ _ _ _ _ _ _ _ - I

. - . . . ~ - ;

I . . . .

Attachment 1 IN 86-81 September 15, 1986 Page 3 of 3 ATWOOD & MORRILL CO., INC.

Main Steam Isolation Valves Furnished With External Closing Springs Manufactured by Duer Spring & Manuf. Co. )

BWR UNITS Utility Station Quantiv & Valve Size TVA Browns Ferry 1, 2, 24 26-inch

&3 TVA Hartsville A1, A2, 32 26-inch Units B1 & B2 B1 & B2 Cancelled ,

TVA Phipps Bend 1 & 2 16 26-inch Units cancelled 205

{

INDEX FACILITY REPORT HUNER PAGE Automatic Switch Company Florham Park, New Jersey 99900369/86-01 1 Babcock & Wilcox Lynchburg, Virginia 99900400/86-01 9 The BOC Group, Incorporated Murray Hill, New Providence, N.J. 99901063/86-01 19 Cooper Energy Services Grove City, Pennsylvania 99900317/86-01 23 Control Products Corporation Grafton, Wisconsin 99901045/86-01 35 Enerfab, Incorporated Cincinnati, Ohio 99901064/86-01 43 General Electric Company San Jose, California 99900403/86-01 49 ITT Barton City of Industry, California 99900113/86-01 57 ITT Henze Mobile, Alabama 99901060/86-01 63 L. B. Foster Company Commerce, California 99901061/86-01 73 NAMCO Controls, Incorporated Mento, Ohio 99900378/86-01 77 NOVA Machine Products Corporation Middleburg Heights, Ohio 99901052/86-01 85 Nuclear Packaging Incorporated Federal Way, Washington 99901047/86-02 93 Solid State Controls, Incorporated Columbus. Ohio 99900276/86-01 111 Southern Conpany Services Birmingham, Alabama 99901058/86-01 119 Unistrut Corporation Wayne, Michigan 99900362/86-01 137 207

r 1

1 INDEX (centinued)

FACILITY - R" PORT NUMBER PAGE Western Piping and Engineering San Francisco, California 99900302/86-01 147 Westinghouse Electric Corporation Blairsville, Pennsylvania 99900005/86-01 151 INFORMATION NOTICE # SUBJECT 86-56 RELIABILITY OF MAIN STEAM SAFETY 161 VALVES 86-57 OPERATING PROBLEMS UITH SOLEN 0ID 169 OPERATED VALVES AT NUCLEAR POWER PLANTS 86-61 FAILURE OF AUXILIARY FEEDWATER 175 MANUAL ISOLATION VALVE 86-62 POTENTIAL PROBLEMS IN WESTINGHOUSE 179 M0LDED CASE CIRCUIT BREAKERS EQUIPPED WITH A SHUNT TRIP 86-65 MALFUNCTIONS OF ITT BARTON MODEL 183 580 SERIES SWITCHES DURING REQUALIFICATION TESTING 86-66 POTENTIAL FOR FAILURE OF REPLACE- 185 MENT AC COILS SUPPLIED BY THE WESTINGHOUSE ELECTRIC CORPORATION FOR USE IN CLASS 1E MOTOR STARTERS AND CONTACTORS 86-71 RECENT IDENTIFIED PROBLEMS WITH 187 LIMITORQUE F10 TOR OPERATORS 86-72 FAILURE 17-7 PH STAINLESS STEEL 191 SPRINGS IN VALCOR VALVES DUE TO HYDROGEN Er1BRITTLEMENT 86-73 RECENT EMERGENCY DIESEL GENERATOR 193 PROBLEMS 86-77 COMPUTER PROGRAM ERROR REPORT 195 HANDLING 36-78 SCRA!!S0LEN0IDPIL0TVALVE(SSPV) 199 REBUILD KIT PROBLEMS 86-81 BROKEN INNER-EXTERNAL CLOSURE 201 SPRINGS ON ATWOOD & MORRILL MAIN STEAM ISOLATION VALVES 208 L _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ - _ ____ _

l VE00R INSPECTION REPDRTS RELATED TO REACTOR PLANTS

+ i A . B  : B B B i B ! C l C i C D D : D D l

• i R i E  : R R R : i 0 i 0  : R A l 1 : R : U !

e i k i L i A 0 i U ! R M i 0  : Y : V A E : A e PLANTS N : L 1. W N : 0 A  : P  ! S : I : B 5 : h

+ l 5 : E  : D N : 5 : N N  : E i T S : L! D E:

• 1.A : F  : W 5 l W  ! C l R l A i 0 E :  :

• ! 5 l 0  ! 0 I 1 l H l l L' B : i N : A e N : 0 : F : C L : E i l E : C : R:

• ! 1 i T ! D E K 2 :  : R S A : i ! N :

• i L E : R  : P 1 5 : N : 2= 0 :

• l 2 : l 1 : R 1 :  : E : V ! E Y : E L:

• ! l 1 & Y : L A l E : 0 3 : D :

• l & 2 l l 2 : K l lR 1 : N :  :

VENDORS

• l 2 11 l l l l l l l l l

• l l l l 2 !  : 1 l  ! 3 l l 1 : i

: 6 :  : &  :  : &  :  :

l l l 3 :  :  : 2 :  :  : l 2 :  :

AUTOMATIC SWITCH COMFANY : ALL BOILING WATER REACTORS :  : l l l l l l 1 l l l l l L l l l l l l l.....:  : :_ .:.....: _:.....:  : _...:.. .:  :.  : .:.

THE BOC GROUP, INC. NOT IDENTIFIED  :  :  :  !  !  :  :  :  :

l l l l l l l l l l l l l l BABC0CK & WILC01 313 l I I  :  : l l- l 1 : I l l l l NUCLEAR POWER DIVI:10N :  :  :  : i  :  !  !  : ,

:  :: -;  :-  ; _--: - .....; ... _ _..:. j COGFER INDUSTRIES  :  : I  : I I  :  :  : l COOPER ENERSY SERVICES :  : l l l l l l l l l l CONTROL FRCDUCTS CORP.  ! ALL huCLEAft POWER STATIONS THAT USE A6ASTAT  : l  :

i ELECTRICAL CONTROL RELAY SERIES E6P, ETR, & EML  : l l l l GEhERAL ELECTRIC CD.  :  : 250 324 :  : I :  :  : 237 : I SANJOSE,CA  :  :  : 260 l l l  ! l l l l ITT BARTON  : FLANTS WITH ITT BARTON DIFFERENTIAL PRESSURE AND  :  :  :

PRESSURE ELECTR0NIC TRANSMITTERS AND INDICATING SWITCHES l l l l

-1 .:  :  :.. .:.....;.....: ....; i. ..; :_....:.

ITT HENZE  :  :  :  :  :  : l l l l 1 l l  !  !  !  ! l l l l l l l l l L.B. FOSTER  : l l l l l l l l l l l l l 4 1'

X-APPLIES TO ALL PLANTS DOCKETNO.- APPLIES ONLY TO THE IDENTIFIED UNIT 209 I

I

. . _ . . . . . . _ . . _ _ _ - . . . _ _ . _ m-.

l' VENDOR INSPECTION REPORTS RELATED TO REACTOR PLANTS

• ! A i Bt B t B l B l BiC I C C l D i D ! D l D e i R l EI R I R ! R ! Y 0 ! 0 ! R i A l 1 R l Ui e  ! K I L i A 1 0 1 U l R I M 0: Y l V i A I E : A i e PLANTS t N I Li I I N ! N I O I A I P l S I I ! B i S I N I e

! S I EI D 1 N i S I N I N I E: T l 5 : L: D i E e i A ! F i N l S I N I i C l R I A 1 0 1 E  !  !

e i S I OI OI I I ! ! I H I L B !  ! N i A I e  ! I N l 0 tF ! C I & I E I i i E : C l 1 R I e i 1 i i ! D 1 E i K : 2: I R I S I A 1 1 N i e & I EI i R i l i P i 1 S N I 2 1 0 t a 1 2 I I 1 i R I 1 1 I E i i V i E : Y l & i L i e i i 1 : & I Y I & I  ! A l l E i i 0 3  : D 1 e i  ! & I 2 1  ! 2 l l K 1 R I i 1 N I l l VENDORS e l I 2 I i 1 : I  ! i l 1 I  !

e  ! I i 1 2 i i l 1 l l 3 ! I 1 1 I I e i  ! I l& I &I i  ! i & I i  !

e i I i  ! 3I I I 2 !  !  !  ! 2: I  !

... ____. g. g. g  ; g. g g g. . ... . g .g.  :

NAMCO  ! VARIOUS I I I  !  !  !  !  !  ! I I I I  !  ! I I I I I i  !  !  !  !  !

.. . .... . .p. .g .. . p. . p p. . i p .g.. .:  : _g p. .g.. p NUCLEAR PACKAGING INC. I i i  !  !  !  !  !  !  !  !  ! 1  !

I I I I I I 1 i I  :  : l-

. . .. .g.. .p....; p. .g .g .g -

.p.... _i p.... .

NOVA MACHIE PRODUCTS I I I I I I I  ! I I I 1 CORPORATION I I I i l i  ! I I i  !  !  !  !

. . _ . p. . p. . p g ... p.... p. . p. . p.  : . .g.. .g p __p. .p SOLIO STATE CONTROLS, INCL  ! I i l l I l l l l  ! I i l i l l I 1 i i t  ! I I

...... ... p_ .g .g p.. p .g ___ p.. g.. . .. p. _g p.. p SOUTHERN COMPANY SERVICES  !  !  ! I i l I I  !  !  !  :

1 I I I I I I I  :  :  ! l

.. .. .. .g ... p. . p .g . p. . p. .g . p_ .; _ . p. .p. .p. .p UNISTRUT CORPORATION I i l I l l l I I l l  : i i l I i  !  !  : i l l l l l l

_ _ _ . p.... p. p .g . p.... p.... p.... p _g . . p. .:_ p. . ; p E STINGHOUSE ELECTRIC 1 NUCLEAR PONER FACILITIES USING 1 l l l  ! i  !

NUCLEAR FUEL DIVISION  ! E STIN6 HOUSE FUEL I  ! I i  ! 1 i

. . p. . p. . p. . p . g p. .p. .p....g ....! -g p....p. .p K STERN PIPIN6 AND I i  ! i I i 445 l i I  !

EN61EERIN6  : 1 l l 1 l l l 1 1  : I I

. . ... . _ .... . p.. p ..g . p. ,.. p . p . p....;_. . . _i  : - : _... p.._. p I-APPLIES TO ALL PLANTS DOCKETNO.- APPLIES ONLY TO THE IDENIlFIED UN11 210 L_ ____ _ _ _ _ - _ - . _ - - - _ - - - - - - - - - - - - - - - .

VENDOR INSPECTION REPORTS RELATED TO REACTOR PLANTS

• l F H J l L l L N : N 0 : P P P P R

• ! A ! A ! A : I I l 0 : C i A : E : E: I A e R T l A : S : M ! N : R l 0 : L i A R l L :l~ N  :

• PLANTS l L l C l l A E E: T i N l 0 l C l R 6 : C l

• E : H : F- L R H : E l l H : Y ! R H:

• ! Y : 1I : L 1 : M l l E : V l l 1 0 :

• i 1 : T E : C l I l A l l E  : B I l M l

• ! 1 l & : Z K : L N l 1 R  : 0 : & :  : 5 :

• l k i 2 : P C  : E : N : 2 : D T 2 1  : E'

• l 2 ! l A l 0 1 l l A k l E  : T l l &  : C l

• l l l T ! N l & P 3 :  : 0 t l 2 l 0 :

• l l l R N : 2 : T l 1 : 1 : M  :  :

• l I ! T l l l L  : 2 l l l l 1 :

VENDORS

• l l l C l Y : 1 l 2 :  ! k l 2 l l l

• l l l K !  ! ! I k l l l 3 : &  :  !

• l l l l 6 : 2 : l l 3 : l i

• l l  : 2 :  :  :  :  :  :  :  :  :  :

AUTOMATIC SWITCH COMPANY : ALL BOILING WATER REACTORS  :  :  : l l  !

!  !  :  :  :  :  :  : l l  :

THE BOC 6ROUP, INC. I NOT IDENTIFIED l  :  :  :  :  :  : l l l l l l l l l l l l l l l l

. _.- -g.....: -

BABCDCK L WILCDI l l l l l l I  : l l I :

NUCLEAR POWER DIVISION l l l l  !  !  :  :  ! 1-COOPER INDUSTRIES  :  :  :  :  : 410 :  :  : I : l l l l COOPER ENERGY SERVICES l l l l l l l l l l l l l l CONTROL PRODUCTS CORP. ALL NUCLEAR POWER STAT 10NS THAT USE A6ASTAT  :  :  :  :  :

ELECTRICAL CONTROL RELAY SERIES E6P, ETR, & EML  : i  :

GENERAL ELECTRIC CD. l l 321 : I 373 352 220 l  :  : 293 :

SAN JOSE, CA  :  :  !  :  : l l l l l l l l ITT BARTON PLANTS WITH ITT BARTON DIFFERENTIAL PRESSURE AND .

FRESSURE ELECTRONIC TRANSMITTERS MD INDICATIN6 SWITCHES l l l IIT HENIE  :  :  : l l l 1 :  :  : I  :  :

!  !  !  !  :  :  ! l l l l l

-; -l  :-  ;- -: ..: .... - l  :._...:.

L.B. FDSTER  : l l l l l l l l 1 :  :  :  :  :

I-APFLIES TO ALL PLAhTS DOCKETNO.- ADFLIES ONLY TO THE IDENTIFIED UNIT 211

VENDOR INSPECTION REPORTS RELATED TO REACTOR PLANTS

• l F l H J l L i L N I N i 0 : P P l P P l R

• l A I A : A : I 1 I O C 1 A ! E EI I I A  :

e  ! R l T : A I S! M l N : R 0 : L l A I R : L I N

• PLANTS  ! L ! C  ! A : E : E : T l N : 0: C l R 6 I C e E l H : F : L ! R : H i E : l H Y ! R H e i Y : l 1 : L 1 i M E : V I  : 1 1 0  :

• l l 1 : T l E: C l 1 : A lE B 1 : M i i e 1 l 6 l Z I K L ! N : 1 : R ! 0 6 l l S I e 2 i P l C l l E I N : 2'! D l T

& 2 : 1 : E

• 1 2! l A 0i 1 l l A & ! E: T I & C i e  : N l 6 : P T  : 3 l l 0 i  ! 2 1 0

• l l l R ! N : 2 : T I 1 i  ! ! : M i i e l I i I T l l  ! & l 12 l l l 1 :

VENDORS

• l 1  : C l Y 1 2 :  !& 2 i l l  !

e  !  ! l K 1 : & !  : 3 l & i  :

• 1  :  : &I 2 l l 3 l l l l t l  ! 2 l  :  : I l l l i NANCO l VARIOUS :  ! i  !  ! 1 I i  !  !

l l l l l l l l l l 1 l  :

NUCLEAR PACKA6ING INC. I i  :  :  :  ! i l l l

: l l l l l l i  :  : l

- .....:-  :.. .g ....:.....:  : ...:.

NOVA MACHINE PRODUCTS l l l l  : 410 l l l l l l l l CORPORATION i  !  !  ! l l l  !  !  !  !  !

SOLID STATE CONTROLS, INC: i  : l l l 1 l l 1 i l l l l 1 l l l l i  :

SOUTHERN COMPANY SERVICES: I ! ! l l l l l l l l i  !

l  ! l l 1 l l l l l l l l l

...:..... - -; -: .: _: ....: ...g- -: ....:_  :.....:.....:.

UNISTRUT CORPORATION l l l  !  ! l l l l l l l l l l l l l l l l l l 1  :  :  :

NESilN6 HOUSE ELECTRIC l NUCLEAR PONER FACILITIES USING l l  !  !  !  !  !  !

NUCLEAR FUEL DIVISION I NESTINGHOUSE FUEL l l l l  ! l l  !

NESTERN PIPIN6 AND  ! l  !  :  :  :  :  :  : 277 440 :  : I EN61NEERIN6  :  : i l  !  ! l l l  ! l l l I-APPLIES TO ALL PLANTS DOCKETNO.- AFPLIES ONLY TO THE IDENTIFIED UNIT 212 t _ - _ _ _ - - - - . -

VENDOR INSPECTION REPORIS RELATED TO REACTOR PLANTS e l R ! S : S : S : S  : T l V V W W l W ! Z

• l 1 : E l 0 : U U l H l E: 0 I A A l N : I

• l V A  : R S  : R R : 6 : T i T P 0 e PLANTS  ! E : B : T l R l 0  : E l M i T l E : T i  ! N :

• l R R : E ! Y : U l E : 0 I L l R : S : W

• l l 0 : I :  : E l N ! E l F:  : P 1

• i B 0 A 1 : H M i T ! 0 : B l P : E l e  : E l K S : k : A 1 : 1 : R A ! S : 2 e N :  : 2 : N i L : Y : & D : R S:  !

5  ! D : 1 : P N l E : A ! 2 :  :  : i e  ! & R: A N l l 3 l 1 1

• l 1 : 2 ! 0 l l 1 K :  : & 2 l l

• l 6 :  : J l l 1 i S : E!  !  ! 2 : E :  :

VENDORS

• l 2 :  :  : 6 l E: l 3 :  :

• l l l 1 :  : 2 l 1 :  : l l

• 1  : & l l 6 :  :  :  :  :  :

• l l l 2 l l l 2 :  !  !  !  !  !

AUTOMATIC SWITCH C0itPANY : ALL BOILING WATER REACTORS : i l l l l l l l l l l l l l  !  ! l l l ---- !  !  ! ---- l l --- l l l l l THE BOC SROUP, INC. NOT IDENTIFIED l l  ! l l l l l l l l l  !  :  : l l l l l l B1BC0CK & NILCOI l l l l l 289 l l l l 460 l l NUCLEAR POWER DIVISION : l l l l l l l l l l  !

COOPER INDUSTRIES  :  : I l l I :  :  : I  :  : I l COOPER ENER6Y SERVICES :  :  :  ! l l l l l l l l l .:  : .:  : .;  :  :. -

CONTROL PRODUCTS CORP. l ALL NUCLEAR POWER STATIONS THAT USE A6ASTAT l l l i ELECTRICAL CONTROL RELAY SERIES ESP, ETR, & EML  :  :

SENERAL ELECTRIC CD.  :  :  :  :  : I l l l l l l SANJOSE,CA  :  :  :  : I  ! l l l  !  !

-l  :  :  :.....;  : .:  : ,___;  ;  :;  ;

ITT BARTON l PLANTS WITH ITT BARTON DIFFERENTIAL PRESSURE AND  : l l FRESSURE ELECTPONIC TRANSMITTERS AND INDICATING SWITCHES :  :

ITT HENIE  : l l l l l  !  !  !  : l l l l l l l  :  : l l l  !

L.B. FOSTER l l l l l l l l l l t  !

l l l l l l l l l l l l l

...... l- -:  : . ___ : .;  : .; ...: .:. .; .;  ;

I-APPLIES TO ALL PLAhTS DOCKETNO.- APPLIES ONLY TO THE IDENTIFIED UNIT 213

1.

t.

VENDOR INSPECTION REPORTS RELATED TD REACTOR PLANTS

• l R: S : S : S : 5 : T V V N N ! N : Z

• l 1 : E l'0 U !U l H : E : 0 : A A l N I

• l V A 'R : S : R R! 6 : T T P 0 :

• PLANTS- l E : B T R l 0 : E : N : T E : T l l N

• l R : R : E ! Y : U l E' 0 : L i R S : N :

• l l 0 l I E : N :E : F : P 1

• l B l 0 : A 1 l H : M i T l l 0 : B P 6 :

• l E: K 5: & : A l 1 'l R A S : 2 :

a l N l l l 2 ! N : L Y : & ! D R S :

• i D l 1 : P N l E : A 2  : l l l

• l l E : R : A N: 3 1 l 1 :

• i ! l 2 0  : I : K l l & 2 :

• l k l l J l 1 : S l E :  :  : 2 : &

WNDMS *  ! 2 l  !  : & E l l l l 3 l l

• l l l 1 :  : 2 1 1  :  :  :  :  :

• ! i & l 6 :  : l l l l

• : 1 i 2 l l l 2 :  :  : l l l l NAMCO l VARIOUS : l l l l  :  :  :  :  :

:  :  ! l l l l l l  !  !

NUCLEAR PACKAGING INC.  !  !  !  ! 320 :  :  :  :  :  :  :

:  : l l l l l l l  :

NOVA MACHINE PRODUCTS  !  !  !  :  :  : I I :  :  :

CORPORATION  :  :  !  :  :  :  :  :  ! I SOLID STATE CONTROLS, INCl l l l I  :  :  ! l l l l l l l l l l l l l l l l l l SOUTHERN COMPANY SERVICES:  :  :  :  :  : l l l  !  !  !

:  :  : l l l l  !  :

UNISTRUT CORPORATION  : l l l l l l l l l I : l l l l l l l l l l l l l  :

.;  :.- -:. ___ -l NESTIN6 HOUSE ELECTRIC l NUCLEAR PONER FACILITIES USIN6 l l l l l  !  !

NUCLEAR FUEL DIVISION  : NESTIN6 HOUSE FUEL l l l l

!  !  !  ! l l --- l ----- l l l ---- l l l NESTERN PIPING AND  : 458 443 l l  ! l l l l l l l ENGINEERING l l l l l l l l l l I-APPLIES TO ALL PLANTS DOCKETWO.- APPLIES ONLY TO THE IDENTIFIED UNIT 214 C __ _ ----------.---------------------------- _ - - - - - - . . _ .

U S. NUCLt 4 4 REGULATORv COedMISSiON 1 mtPOMT NuMSE A #Asegaea ey TsDC, ear ve# Ne , es ears senC fores 335 NU G

'fi'- BIBLIOGRAPHIC DATA SHEET g uf\s itauCTIO,. ON T-E afvt.sf 3 LE AVE SL ANK 2 Tif AND SuSTIT LE Lic nsee Contractor and Vendor Inspection Status Report Qu rterly Report -- July 1986-September 1986 ft DATE REPORT COMPLETED vtan upT

.. Auf Oais. Septerhber 1986

/ 6 DATE REPORT iS5UED j MONT-j vfAa November 1986 y PEnFORMING ORGAN 12 lON teAME AND M AILING ADDRESS fiacSse le Cests 8 PWOJECT/Y ASE/ WORE UNIT NUMBER Division of Q lity Assurance, Vendor and Technical /

Training Cen r Programs p riN Oa oa ANT Nuana Office of Inspe ion and Enforcement <

l U.S. Nuclear Regu atory Commission Washington, DC 20 5 .

10. SPONSOmeNG ORGAN 12 Af EON NAVE A MA: LING ADORES $ ttacweele cases ita YvPE OF MEPOHT Quarterly Same as 7. above. , , E ,o o co, E , f o ,,,,,,,,, ,,,,

12 svPPLEMENTAnv NOTES u Auf a ACT ,m e. . , ,

A This periodical covers the res its of nspections performed by the NRC's Vendor Program Branch that have eengdistributedtotheinspected organizations during the period om' July 1986 thru September 1986. Also, included in this issue are the res its of certain inspections performed prior to July 1986 that were not 'n luded in previous issues of NUREG-0040.

14 DOCUwf NT ANALvlis - e mEvWomO5'DESCmiPTOR$ 15 AV AILa $s LIT Y ST ATEWENT Unlimited

'6 SECum,Tv CLAssi7tCAtlON f Tme peser

. loENTi.if as oPEN Noto TE o. Unclassified

< r.e ,

t( e,"

\ Unclassified K 13 NVU9tm 06 PAGES

\

r is PaicE

• U.S.GGvCR W hi P914Tth Orr 1CE s 19 96- 181 68 3 t a00 20

I UNITED STATES sncm couamcuss aan NUCLEAR REGULATORY COMMISSION m5Gjst,"cE5 '"

WASHINGTON, D.C. 20555

, wjgg, OFFICIAL BUSINESS PENALTY FOR PRIVATE USE,8300 120555078877 1 1ANINV US NRC ADM-DIV 0F PUB SVCS POLICY & PUB MGT BR-PDR NUREG W-501 WASHINGTON OC 20555 I

__.